Silylation of iron-bound carbon monoxide affords a terminal Fe carbyne

Article abstract of DOI:10.1021/ja109678y

A series of monocarbonyl iron complexes in the formal oxidation states 0, +1, and +2 are accessible when supported by a tetradentate tris(phosphino)silyl ligand (SiP^iPr₃ = [Si(o-C₆H ₄PiPr₂)₃

Full text of DOI:10.1021/ja109678y

ARTICLE
pubs.acs.org/JACS
Silylation of Iron-Bound Carbon Monoxide Affords a
Terminal Fe Carbyne
Yunho Lee and Jonas C. Peters*
Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, United States
S Supporting Information
b
ABSTRACT: A series of monocarbonyl iron complexes in the
formal oxidation states 0, þ1, and þ2 are accessible when
supported by a tetradentate tris(phosphino)silyl ligand (SiPiPr
3
= [Si(o-C6H4PiPr2)3]-). X-ray diffraction (XRD) studies of
these carbonyl complexes establish little geometrical change
about the iron center as a function of oxidation state. It is
possible to functionalize the terminal CO ligand of the most
reduced carbonyl adduct by addition of SiMe3þ to afford a well-
defined iron carbyne species, (SiPiPr3)FetC—OSiMe3. Single-crystal XRD data of this iron carbyne derivative reveal an unusually
short FetC—OSiMe3 bond distance (1.671(2) Å) and a substantially elongated C-O distance (1.278(3) Å), consistent with Fe-
C carbyne character. The overall trigonal bipyramidal geometry of (SiPiPr3)FetC—OSiMe3 compares well with that of the
corresponding carbonyls, (SiPiPr3)Fe(CO)-, (SiPiPr3)Fe(CO), and (SiPiPr3)Fe(CO)þ. Details regarding the electronic structure of
the carbyne complex have been explored via the collection of comparative M€ossbauer data for all of the complexes featured and also
via DFT calculations. In sum, these data point to a strongly π-accepting Fischer-type carbyne ligand that confers stability to a low-
valent iron(0) rather than high-valent iron(IV) center.
’ INTRODUCTION
In this paper we establish the utility of the (SiPiPr3)Fe
scaffold in supporting a series of isostructural CO complexes
that span the formal oxidation states Fe0-CO, FeI-CO,
and FeII-CO. Silylation of the terminal Fe0-CO species
affords access to a well-defined and structurally unusual
Fischer-type carbyne complex,8,9 (SiPiPr3)Fe(COSiR3), with a
distinctly short Fe-C bond distance (ca. 1.67 Å). Fischer and co-
workers have reported the singular previous example of a
structurally characterized carbyne complex of iron.10 Because
the mode by which the featured carbyne is generated can, in
principle, be regarded as a net four-electron reduction with
concomitant partial O-atom transfer, it is of interest to consider
the relative states of oxidation of the structurally similar iron CO
and carbyne complexes described herein. Comparative
M€ossbauer and DFT data are presented to shed light on this
issue.
There has been a surge of interest in the chemistry of low-
valent, redox-active complexes of the mid-to-late first-row transi-
tion metals in recent years.1,2 This interest has been in part
motivated by a desire to generate species that feature uncommon
metal-to-ligand multiply bonded species such as oxos, imides,
and nitrides via partial or complete group transfer. The study of
such species will help to expose the respective roles they might
play in small-molecule transformations that include, for example,
nitrogen reduction, C-H oxygenations and aminations, and
aziridinations.3,4 With these goals in mind, our group recently
introduced “(SiPR )Fe” (where SiPR3 = [Si(o-C6H4PPh2)3]- or
3
[Si(o-C6H4PiPr2)3]- for R = Ph and Pr, respectively) as a
i
versatile low-valent iron scaffold that can accommodate a range
of ligands in an axial fifth site trans to the silyl anchor. A salient
example concerns the ligand N2, where terminally bonded
(SiPiPr3)Fe-N2 complexes can be accessed in three formal iron
oxidation states (0, þ1, þ2), and reductive functionalization of
the terminally bonded N2 ligand is possible.4,5 It was of interest to
us to explore whether a parallel manifold of chemistry might also
be accessible for carbon monoxide. A conceptual link between
the chemical reduction of CO and N2 is well appreciated within
the organometallic community.6 The very recent discovery that
VFe-nitrogenase enzymes can catalyze CO reduction to higher
order carbon chains when the N2 substrate is replaced by CO
lends new motivation to comparative studies along this line of
inquiry.7
’ RESULTS AND DISCUSSION
Monovalent, yellow (SiPiPr3)Fe(CO) (1) is accessible by
ligand substitution (>90% yield) from the previously reported
N2 adduct species (SiPiPr3)Fe(N2)4 by CO(g) (eq 1). An X-band
EPR spectrum of 1 at 77 K in 2-MeTHF/THF (1:9) reveals a
nearly axial signal (see Supporting Information), in accord with
solution magnetic data (μeff = 2.2 μB in C6D6 at 22 ꢀC) for an S =
1/2 species. Complexes in which a terminal carbonyl and a silyl
Received: October 27, 2010
Published: March 04, 2011
r
2011 American Chemical Society
4438
dx.doi.org/10.1021/ja109678y J. Am. Chem. Soc. 2011, 133, 4438–4446
|

Journal of the American Chemical Society
ARTICLE
Figure 1. Cyclic voltammogram of (SiPiPr3)Fe(CO) (1); scan rate =
100 mV/s. FeII/I couple at -0.68 V and FeI/0 couple at -1.93 V vs Fc/
Fcþ were observed in THF with 0.3 M tetra-n-butylammonium hexa-
fluorophosphate as an electrolyte.
ligand are trans disposed, as they are in 1 (—SiFeC = 178.7(2)ꢀ),
are uncommon.11,12 The cyclic voltammetry of 1 (Figure 1) is
interesting in that it reveals two reversible one-electron redox
couples, consistent with the electrochemical generation of
the Fe-COþ cation at -0.68 V and its corresponding anion
Fe-CO- at -1.9 V (versus Fc/Fcþ). We were thus encouraged
to pursue the chemical generation of both the anion and the
cation of 1.
ðSiPiPr3ÞFe-N2 þ CO f ðSiPiPr3ÞFe - CO þ N2
ð1Þ
Chemical reduction of a yellow solution of 1 effects a dramatic
color change to an inky red, and subsequent workup affords the
diamagnetic ion-paired species (SiPiPr3)Fe(CONa(THF)3) (2)
in approximately quantitative yield (Scheme 1). As is to be
expected, the ν(CO) at 1717 cm-1 (KBr pellet; 13CO = 1676
cm-1) is substantially lower than that of neutral 1 (1850 cm-1
;
13CO = 1806 cm-1) owing to an increase in π-back-donation.
Addition of two equiv of 12-crown-4 to 2 fully encapsulates the
Naþ cation to afford the terminally bonded, anionic CO adduct,
{Na(12-C-4)2}{(SiPiPr3)Fe(CO)} (3) (KBr pellet; ν(CO) =
1757 cm ;
-1 13CO = 1713 cm-1). The X-ray structures of 1 and 2
are shown in Figure 2. Neutral 1 is isostructural to its corre-
sponding N2 derivative (see Figure 2a and Table 1).4 Ion-paired
2 reveals a crystallographically imposed three-fold axis and hence
an essentially perfect trigonal bipyramidal (TBP) structure type
(τ = 1.0).13 The Fe-C bond distance contracts appreciably to
1.732(3) Å in 2 compared with that for neutral 1 (1.769(2) Å,
Table 1), and there is modest elongation of the CO bond
(1.188(3) Å in 2 vs 1.169(2) Å in 1).
Derivatization of the CO ligand via addition of trimethylsilyl
trifluoromethanesulfonate to a frozen THF solution of 2 affords
the featured carbyne complex (SiPiPr3)Fe(COSiMe3) (4) as a
red solid. The 1H NMR spectrum of diamagnetic 4 features a new
resonance at 0.34 ppm for the -SiMe3 group. The 31P NMR
spectrum of 4 shows a singlet at 104.6 ppm, which splits into a
doublet in a 13C-enriched sample (2JCP = 16.3 Hz, Figure 3a)
owing to coupling to the -COSiMe3 C-atom. A quartet at
Figure 2. Displacement ellipsoid (50%) representations of (a)
(SiPiPr3)Fe(CO) (1), (b) (SiPiPr3)Fe(CONa(THF)3) (2), (c)
(SiPiPr3)Fe(COSiMe3) (4), and (d) {(SiPiPr3)Fe(CO)}{B(ArF)4}
(5). Hydrogen atoms and molecules of co-crystallization are omitted
for clarity.
2
250.3 ppm (q, JCP = 16.4 Hz, Figure 3b) in the 13C NMR
spectrum demonstrates 13C coupling to three equivalent phos-
phorus atoms to cement this assignment. This resonance shifts
downfield by ∼10 ppm from that of its corresponding precursor
complex 2. Two 29Si NMR signals were detected for 4 at 94.5 (q,
2JSiP = 40 Hz) and 20.3 (s) ppm, corresponding to the Si atoms in
the SiPiPr and TMS units, respectively. These peaks reveal
3
2
29Si-13C coupling in a 13C-enriched sample: 94.3 (dq, JSiP
=
4439
dx.doi.org/10.1021/ja109678y |J. Am. Chem. Soc. 2011, 133, 4438–4446

Journal of the American Chemical Society
ARTICLE
Table 1. Physical Parameters for the Species 1, 2, 4, and 5
{FeCO}{B(ArF)4}d (5)
Fe-CO (1)
Fe{CONa(THF)3} (2)
Fe(COSiMe3)e (4)
a
b
ν(CO) (cm-1
ν(CO) (cm-1
C-O (Å)
Fe-C (Å)
Fe-Si (Å)
C-O-X (ꢀ)
Si-Fe-C (ꢀ)
τ
)
)
1943
1852
1711
-
1959
1850
1717
-
1.102(4)
1.842(3)
2.3245(7)
-
1.169(2)
1.769(2)
2.2942(4)
-
1.188(3)
1.732(3)
2.2586(8)
180.000(1)
180.000(1)
1
1.278(3)
1.671(2)
2.2973(6)
155.0(2)
177.90(7)
0.98
178.69(9)
0.99
178.04(5)
0.95
spin-statec
2.7 μB, S = 1
2.0 μB, S = 1/2
diamagnetic
diamagnetic
a In THF. b In KBr pellet. c Evans method in THF-d8. d B(ArF)4 = B(3,5-(CF3)2-C6H3)4. e Contains 1.
Scheme 1
organic alkynyl ether molecules with a CtC-O-X-type link-
age.16 Stang has, for example, structurally determined that the
acetylenic-oxygen C(sp)-O bond distance in -CtC-O-
SO2R-type derivatives is far shorter (ca. 1.33 Å) than in
corresponding but saturated alkyl C(sp3)-O species, where
the distance is ca. 1.46 Å.16a DFT data for 4 also support the
multiple bond character of the Fe-C bond (vide infra).17
Because the generation of the featured carbyne can, in
principle, be regarded as a net four-electron CO reduction with
concomitant partial O-atom transfer to Si, it is of interest to
consider the relative states of oxidation of the structurally related
Fe-CO and carbyne complexes under present consideration.
The electronic structure of carbyne 4 is perhaps most interesting
in this context. If one assigns an oxidation state to its iron center
Figure 3. Multinuclear NMR data of (SiPiPr3)Fe(COSiMe3) (4):
(a) 31P NMR spectrum of (SiPiPr3)Fe(13COSiMe3) (4-13CO), (b)
13C NMR spectrum of (SiPiPr3)Fe(13COSiMe3) (4-13CO), and (c)
29Si NMR spectra of (SiPiPr3)Fe(COSiMe3) (4, top) and (SiPiPr3)Fe-
(13COSiMe3) (4-13CO, bottom). All spectra were measured in C6D6 at
room temperature.
by removal of all ligands in their closed-shell configurations (i.e,
-
{SiPiPr
}
and {COSiMe3}3-),18 an Fe(4þ) d4 assignment is
2
2
40 Hz, JSiC = 21 Hz) and 20.0 (d, JSiC = 7 Hz) ppm.
Representative NMR data are collected in Figure 3.
3
suggested. This scenario draws analogy between 4 and several
The solid-state structure of 4 (Figure 2c) establishes the five-
coordinate TBP (τ = 0.91) geometry at the iron center with an
η1-bound “COSiMe3” ligand. The Fe-C bond distance is
unusually short (1.671(2) Å), and the C-O distance is sub-
stantially elongated (1.278(3) Å) in comparison to 1 and 2,
consistent with Fe-C carbyne character. Indeed, the Fe-C
distance in 4 is appreciably shorter than that of Fischer’s original
iron aminomethylidyne complex (1.734(6) Å).10a The Si-O
distance is 1.674(2) Å, consistent with Si-O distances of
structurally related RMe2Si-O-CtN and other terminal silox-
ycarbyne metal complexes,14,15 and the —C-O-Si angle is
155.0(2)ꢀ. The 1.278(3) Å C-O bond distance can be com-
pared to C-O single bonds in the few structurally characterized
three-fold symmetric L3FetNx species that have been charac-
terized, where Nx = NR2- or N3- 3c,19
However, one can
.
alternatively regard the carbyne ligand as a {COSiMe3}þ
closed-shell cation, akin to the {CONa}þ ligand of 2, with one
σ-donor and two π-acceptor orbitals at carbon. This formulation
is more suggestive of “Fischer-type” carbyne character, where the
polarization is inverted and the iron center back-donates to
carbon. In this limit, the iron center is better described as low-
valent d8 iron(0). A final scenario worthy of mention is one
where the tris(phosphino)silyl ligand is also regarded as a
cation, {SiPiPr3}þ. Appreciating the high degree of covalent
bonding in 4, and hence the limited utility of assigning the
number of “d” electrons present, we conducted comparative
4440
dx.doi.org/10.1021/ja109678y |J. Am. Chem. Soc. 2011, 133, 4438–4446

Journal of the American Chemical Society
ARTICLE
Figure 5. HOMO - HOMO-4 of (SiPiPr3)Fe(COSiMe3) (4) derived
from the single-point DFT calculations with corresponding Mulliken pop-
ulations (see Experimental Section); energies in cm-1. Lobal represen-
tations correspond to the orbitals indicated by the number with 0.07
isocontours.
Figure 4. Zero-field M€ossbauer spectra of (SiPiPr3)Fe(CO) (1),
(SiPiPr3)Fe(CONa(THF)3) (2), {Na(12-C-4)2}{(SiPiPr3)Fe(CO)}
(3), (SiPiPr3)Fe(COSiMe3) (4), and {(SiPiPr3)Fe(CO)}{B(ArF)4}
(5) recorded at 77 K.
indistinguishable from that of 2, supporting the {C-O-Na}þ/
{C-O-SiMe3}þ analogy drawn above. This fact appears to rule
out the plausibility of the Fe(IV) d4 configuration. It is interest-
ing, therefore, to compare the isomer shifts of 2, 3, and 4 to those
that have been reported for other five-coordinate, d8 TBP
complexes featuring a high degree of covalency due to combina-
tions of CO and phosphine ligands. Such data have been
tabulated in several studies for d8 [Fe(CO)4(L)] species, where
L for example is CO, PR3, and P(OR)3.21,24-26 The isomer shifts
of such d8 TBP [Fe(CO)4(L)] structure types generally fall within
a range of δ -0.09 (e.g., Fe(CO)5) to -0.18 (e.g., [Fe(CO)4-
(SiCl3)][Bu4N]27). While the isomer shift values for 2, 3, and 4
are somewhat more positive, they are in a reasonable range for
the d8 configuration given slight geometric variability, in addition
to the presence of a (SiPiPr3)Fe fragment that is chemically
distinct from “Fe(CO)4”. Moreover, comparison of neutral 1
with neutral 4 shows a substantial decrease in the isomer shift
value as the SiMe3 group is added to the CO O-atom, again
consistent with formal reduction of the iron center from d7 to d8.
As a final point of comparison, we have recently reported
M€ossbauer data for an isostructural and isoelectronic series of
M€ossbauer measurements on 1-4. We also prepared the cationic,
blue, and S = 1 Fe(II) carbonyl, {(SiPiPr3)Fe(CO)}{B(ArF)4}
(5) (μeff = 2.69 μB, benzene, ν(CO) = 1959 cm-1 (13CO =
1915 cm-1), KBr pellet), by oxidation of 1 with H(OEt2)2-
[B(3,5-(CF3)2-C6H3)4] for additional comparison (Scheme 1).
Complex 5 is an unusual example of a terminal carbonyl adduct
that populates a triplet ground state.20
Figure 4 overlays the M€ossbauer spectra for 1-5. Although
the isomer shift can often be identified with a particular Fe
valence state, strong covalency can compress the range of isomer
shifts and thereby not give a definitive indication of the iron
valence. In addition, the local geometry of an iron center impacts
both isomer shift and quadropole splitting parameters.21-23 It is
therefore most instructive to internally compare the present set
of data, where the geometry at iron is highly conserved and only
the axial ligand and overall molecular charge is altered. As can be
gleaned from Figure 4, the isomer shifts of the various com-
pounds are positive and compressed over a rather narrow range.
This fact intimates that much of the redox/charge-transfer
chemistry occurs at the terminal CO/COSiMe3 ligands in this
series of complexes, presumably via relative π-back-bonding
contributions from iron that are compensated by its σ-accepting
contribution. Comparison of the isomer shifts of the terminally
bonded carbonyl adducts Fe-CO- 3, Fe-CO 1, and Fe-COþ
5 expectedly shows that the isomer shift δ for these three
complexes shifts positive (δ þ0.09 (3), þ0.21 (1), þ0.31 (5))
as the overall molecular charge, and hence the relative state of
iron oxidation, shifts positive from formal Fe(0) to Fe(I) to
Fe(II). The trend reflects a decrease in s-electron density at Fe as
electrons are successively removed from the system. The carbyne
complex 4 affords an isomer shift (δ þ0.06) that is almost
Fe-N2 and Fe-N2SiMe3 species supported by the (SiPiPr
)
3
ligand.5 The complex (SiPiPr3)Fe(N2SiMe3) is isoelectronic with
carbyne 4 and was likewise formulated as d8. Its isomer shift is
appreciably positive (δ þ0.19) relative to 4, presumably reflect-
ing the higher electronegativity of nitrogen relative to carbon,
and hence greater covalency in the Fe-C interaction relative to
the Fe-N interaction. While the isomer shifts of the series Fe-
N2-, Fe-N2, Fe-N2þ, and Fe(N2SiMe3) follow the same trend
observed for Fe-CO-, Fe-CO, Fe-COþ, and Fe(COSiMe3),
they are all shifted positive and fall within the range from þ0.18
(Fe(N2SiMe3)) to þ0.53 (Fe-N2þ). The very similar
4441
dx.doi.org/10.1021/ja109678y |J. Am. Chem. Soc. 2011, 133, 4438–4446

Journal of the American Chemical Society
ARTICLE
Table 2. Selected Bond Indices and Bond Orbital Occupancies from Natural Bond Orbital Analysis
{FeN2}{NaTHF3}
FeN2TMS
{FeCO}{NaTHF3}
FeCOTMS
Wiberg Bond Index
Fe-L
1.1051
2.3702
0.0856
0.7294
1.4849
1.8371
0.7609
0.6768
1.4845
1.8569
0.6766a
0.5236
0.7808
L-L0
1.6177
0.0384
0.7115
L0-Na or Si
Fe-Si
Bond Orbital Occupancy
Fe-L
L-L0
1.93622 (20.43% Fe, 79.57% N)
1.94325 (21.03% Fe, 78.97% N)
1.79698 (53.11% Fe, 46.89% N)
1.79025 (49.38% Fe, 50.62% N)
1.98931 (47.80% N, 52.20% N)
1.85804 (34.37% Fe, 65.63% C)
1.88259 (35.86% Fe, 64.14% C)
1.73379 (67.84% Fe, 32.16% C)
1.73122 (69.19% Fe, 30.81% C)
1.98781 (22.88% C, 77.12% O)
1.99372 (53.73% N, 46.27% N)
1.99372 (53.73% N, 46.27% N)
1.99065 (48.96% N, 51.04% N)
1.61953 (54.43% Fe, 45.57% Si)
1.81913 (27.88% Fe, 72.12% P)
1.81921 (27.89% Fe, 72.11% P)
1.81920 (27.89% Fe, 72.11% P)
1.83706
1.99089 (27.53% C, 72.47% O)
Fe-Si
Fe-P
1.63573 (56.44% Fe, 43.56% Si)
1.84746 (31.78% Fe, 68.22% P)
1.84705 (31.52% Fe, 68.48% P)
1.85138 (31.21% Fe, 68.79% P)
1.88808
1.61800 (55.48% Fe, 44.52% Si)
1.80671 (30.59% Fe, 69.41% P)
1.80666 (30.59% Fe, 69.41% P)
1.80667 (30.59% Fe, 69.41% P)
1.84666
1.64338 (59.08% Fe, 40.92% Si)
1.84533 (35.09% Fe, 64.91% P)
1.84533 (35.15% Fe, 64.85% P)
1.84533 (35.35% Fe, 64.65% P)
1.88179
Fe (LP)
1.83698
1.88710
1.84652
1.88382
1.64537
1.67127
1.64532
1.67118
a Wiberg bond indices of free CO and N2 reveal 2.1834 and 3.0493, respectively. The bond indices for CtC and C-O bonds in propynyl tosylate11a are
2.6761 and 1.0649, respectively. Geometry optimization and NBO analysis for propynyl tosylate were run on the Gaussian03 suite of programs with the
B3LYP level of theory with the 6-311G** basis set for all atoms.
quadrupole splitting values for carbyne 4 (EQ = 1.11 mm/s) and
(SiPiPr3)Fe(N2SiMe3) (EQ = 1.26 mm/s) further establish their
close electronic structure relationship.
A Wiberg analysis of (SiPiPr3)Fe(N2Na(THF)3) provides an
N-N Wiberg index of 2.37, compared with 3.05 for free N2.
Hence, substantial π-bond character remains in (SiPiPr3)Fe-
(N2Na(THF)3). Replacement of Na(THF)3þ by SiMe3þ, as in
(SiPiPr3)Fe(N2SiMe3), attenuates the π-bonding (N-N Wiberg
index =1.84), but not nearly to the same extent as occurs for the
C-O linkage in 4. Consistent with the X-ray data collected here
and reported elsewhere,5 the axial Fe-Si bond distance remains
within a rather narrow range, regardless of the relative iron
oxidation state and the axially bound ligand for both the N2 and
CO series. The largest discrepancy within the CO series concerns
cation 5, where the Fe-Si bond distance is 2.3245(7) Å, relative
to the distance of 2.2586(8) Å in anion 2. This difference may in
part reflect less σ-donation from Fe to Si as the Fe center is
oxidized. However, the triplet spin state in 5 relative to the singlet
spin state in 2 may be the dominant factor. Consistent with the
latter idea, the average of the Fe-P distances for triplet 5 (2.39
Å) is vastly greater than those for singlet 2 (2.19 Å). Indeed, the
large difference in these Fe-P values underscores that the
difference in Fe-Si bond distances is rather small between 2
and 5. As should therefore be expected, the Fe-Si Wiberg bond
indices for Fe(N2Na(THF)3), Fe(CONa(THF)3), Fe(N2SiMe3),
and Fe(COSiMe3) supported by (SiPiPr3) also fall within a
relatively narrow range, between 0.68 and 0.78, reflecting con-
sistent bond character across the series.
On the basis of these experimental data, we favor a simplistic
limiting assignment of iron(0) d8 for 4, but DFT analysis of this
complex forces us to further consider the iron(2-) d10 scenario.
As illustrated by Figure 5, the five highest filled orbitals of 4 all
have significant iron d-orbital character. For example, the
HOMO is ca. 47% Fe and 17% Ccarbyne in character.28 The
HOMO-2 orbital in 4, of approximate a1 symmetry, is strongly
mixed between Fe 3dz2/4s and (SiPiPr3) Si 5pz (ca. 38% Fe and
28% Si) and hence cannot be reliably assigned to either iron or
the (SiPiPr3) ligand. We therefore add the caveat to our d8
formulation for 4 that a d10 configuration also has some merit.
To additionally probe this issue, a natural bond orbital (NBO)
analysis29 of 4 and 2 was performed. This analysis points to
several features worth mentioning. The Wiberg bond index
between two connected atoms provides a measure of their bond
order. Given the {C-O-Na}þ/{C-O-SiMe3}þ analogy to
which we drew attention above, it was of interest to compare this
index for the C-O bond in the two species. In 4, a Wiberg C-O
bond index of 0.68 is calculated, compared with 1.62 for 2
(Table 2). These values can be compared with that of free CO,
for which the Wiberg bond index has been calculated as 2.18.
These values indicate that the π-bonding of CO has been
obliterated in 4, whereas substantial π-bond character remains
in the C-O linkage of 2. Accordingly, the Fe-C Wiberg bond
index of 4 increases (1.86) in comparison to that of 2 (1.48) to
compensate, implying stronger Fe-C π-bond character in 4. As
for the M€ossbauer data, it is interesting to compare these indices
with the analogous data calculated for the isostructural and
isoelectronic species where N2 is substituted for CO (Table 2).5
The NBO analysis of 4 identifies an orbital that is σ-bonding
between Fe and the Si atom of the (SiPiPr3) ligand. The bond
orbital occupancy calculated distributes the Fe-Si σ-bond elec-
tron density between Fe and Si in ca. 60/40 ratio, respectively.
For comparison, the electron pairs that are π-bonding between
Fe and the {COSiMe3}þ ligand are also gleaned from the NBO
analysis, where the π-electron density is ca. 70/30 weighted on
4442
dx.doi.org/10.1021/ja109678y |J. Am. Chem. Soc. 2011, 133, 4438–4446

Journal of the American Chemical Society
ARTICLE
the Fe, and the σ-bond electron density is ca. 35/65 weighted on
the C-atom. How one uses these data to aid the assignment of a
d-electron configuration is not clear-cut. For example, two almost
pure lone pairs reside on iron in 4 (Table 2), and, as noted above,
two additional electron pairs are π-back-bonding to the
{COSiMe3}þ but reside heavily on iron. Excluding the latter
two electron pairs leads to a d4 assignment, but such back-
bonding electron pairs are typically accounted for in d-electron
counts, and doing so here leads to a d8 assignment that is better
reflected in the M€ossbauer data, which is most consistent with a
low-valent iron center. But the a1 symmetry orbital that is
energetically within the d-orbital manifold and is σ-bonding
between Fe and Si also resides largely on iron. Hence, it is
somewhat arbitrary to exclude this final pair of electrons from the
d-electron count. Including them instead provides a d10 config-
uration. Hence, the NBO analysis adds a degree of merit to the
d10 configuration, where the mixed frontier orbital of a1 sym-
metry resides largely on iron, as is also true of the two π-bonds
between iron and carbon. A similar argument can be made for the
diazenido species (SiPiPr3)Fe(N2SiMe3). The placement of a
filled a1 orbital within the frontier set of orbitals, which arises
from a strong orbital mixing scenario, therefore confuses a
straightforward d-electron assignment for carbyne 4. In this
regard, the bonding situation finds analogy to that well appre-
ciated in metal nitrosyl complexes and more recently metalla-
boratranes.30,31 In the limit of the latter extreme, the iron center
acts as a σ-donor to a cationic and Lewis acidic tetravalent silicon
center.32 In sum, the d8 configuration for 4, while not perfect,
seems to fit best the collective data at hand and finds context by
considering its electronic structure as similar to the large family of
d8 Fe(CO)xL5-x structure types. This configuration also respects
Pauling’s principle of electroneutrality.33
Frenking and co-workers have studied extensively by theore-
tical means five-coordinate iron complexes that feature Fe-C
multiple bonds.34 Such species include, for example, (CO)4FeC
with an axial bound C ligand and I(CO)3Fe(CH) with axially
bound I and CH ligands. Both species are isoelectronic with 4.
NBO and Wiberg bond index analyses for these species point to
bonding themes consistent with our own description of 4. The
key issue to highlight concerns the nature of the Fe-C bond. For
both (CO)4FeC and I(CO)3Fe(CH), there is strong Fe-C
multiple bonding, and as for the case of 4, there is a σ-donor
orbital best described as CfFe and two π-bonds best described
as FefC. Hence, even though ligands such as “C” and “CH” are
commonly considered “Schrock-type” in high-valent metal
complexes,35,36 when matched with low-valent iron centers they
are better formulated as “Fischer-type”. It is this property, and
hence reversal of polarization of the π-bonding in relation to a
Schrock-type alkylidyne, that renders such ligands compatible
with high d-count, low-valent metals. Were the axially bound C
or CR ligand to be strongly π-donating, there would be a strong
destabilization of filled d-orbitals.37 While such destabilization
leads to a favorable electronic structure for pseudotetrahedral
L3FetNx species,19 it perturbs favorable π-bonding in TBP
structure types. For the Fe(C), Fe(CR), and Fe(COSiMe3)
species under present discussion, however, there is net stabiliza-
tion via π-back-bonding. Hence, when interacting with an L4Fe
or XL3Fe fragment, both CH and COSiMe3 are best formulated
as “Fischer-type” ligands, and they give very similar Fe-C
Wiberg indices (Wiberg index = 1.73 for I(CO)3Fe(CH) vs
1.86 for 4). A similar overall formulation is true of the Fe-C
interaction in (CO)4FeC, though differences in the orbital
makeup and net polarization of the Fe-C bonds arise due to
rehybridization at carbon. Indeed, given Frenking’s conclusions
and the experimental data presented here for 4, it is reasonable to
hypothesize likely electronic stability for a hypothetical complex
that is isoelectronic with 4, namely “{(SiPiPr3)Fe(C)}- ”, featur-
ing a terminally bonded C-ligand.38 Efforts are underway to
explore the feasibility of such a species in our laboratories.
’ EXPERIMENTAL SECTION
General Considerations. All manipulations were carried out
using standard Schlenk or glovebox techniques under a N2 atmo-
sphere. Unless otherwise noted, solvents were deoxygenated and
dried by thoroughly sparging with Ar or N2 gas followed by passage
through an activated alumina column in the solvent purification
system by SG Water, USA LLC. Non-halogenated solvents were
tested with a standard purple solution of sodium benzophenone ketyl
in tetrahydrofuran (THF) in order to confirm effective oxygen and
moisture removal. HSiPiPr3,4 (SiPiPr3)Fe(CH3),5 (SiPiPr3)Fe(N2),4,5 and
H(OEt2)2[B(3,5-C6H3(CF3)2)4]39 were prepared according to litera-
ture procedures. All reagents were purchased from commercial
vendors and used without further purification unless otherwise stated.
Elemental analyses were performed by Midwest Microlab, LLC,
Indianapolis, IN. Deuterated solvents were purchased from Cam-
bridge Isotope Laboratories, Inc., degassed, and dried over activated
3-Å molecular sieves prior to use.
X-ray Crystallography Procedures. X-ray diffraction studies were
carried out at the MIT Department of Chemistry X-ray Diffraction
Facility on a Bruker Three-Circle Platform diffractometer, equipped
with a CCD detector. Data were collected at 100 K using Mo KR (λ =
0.71073 Å) or Cu KR (λ = 1.54178 Å) radiation and solved using
SHELX v. 6.14.40 X-ray-quality crystals were grown as described in
below. The crystals were mounted on a glass fiber or nylon loop with
Paratone N oil. Structures were determined using direct methods with
standard Fourier techniques using the Bruker AXS software package.
Spectroscopic Measurements. Varian Mercury-300, Varian Inova-
1
500, and Bruker AVANCE-400 spectrometers were used to record H,
13C, 29Si, and 31P NMR spectra at ambient temperature unless otherwise
1
indicated. H and 13C chemical shifts were referenced to the residual
solvent peaks. 29Si chemical shifts were referenced to external tetramethyl-
silane (TMS) (δ = 0 ppm). 31P chemical shifts were referenced to external
85% phosphoric acid (δ = 0 ppm). Solution magnetic moments were
determined by the method of Evans.41 Optical spectroscopy measure-
ments were taken on a Cary 50 UV/vis spectrophotometer using a 1-cm
two-window quartz cell sealed with a standard closed cap purchased from
Starna Cells, Inc. (catalog no. 1-Q-10-GL14-C). Infrared spectra were
recorded on a BioRad FTS 3000 EXCALIBUR series FT-IR spectrometer
using a solution IR cell or KBr pellet. X-band EPR spectroscopy
measurements were taken on a Bruker EMS spectrometer. The low-
temperature spectra were collected using a liquid nitrogen coldfinger
dewar, and collected spectra were simulated using the W95EPR
program.42 57Fe M€ossbauer spectra were measured with a constant
acceleration spectrometer (SEE Co., Minneapolis, MN). Isomer shifts
are quoted relative to Fe metal at room temperature. Data were analyzed
and simulated with WMOSS software (Web Research Corp., Edina, MN).
Electrochemistry. Electrochemical measurements were carried out
in a glovebox under a dinitrogen atmosphere in a one-compartment cell
using a CH Instruments 600B electrochemical analyzer. A glassy carbon
electrode was used as the working electrode, and platinum wire was used
as the auxiliary electrode. The reference electrode was Ag/AgNO3 in
THF. The ferrocene couple Fcþ/Fc was used as an external reference.
Solutions (THF) of electrolyte (0.3 M tetra-n-butylammonium
hexafluorophosphate) and analyte were also prepared under an inert
atmosphere.
4443
dx.doi.org/10.1021/ja109678y |J. Am. Chem. Soc. 2011, 133, 4438–4446

Journal of the American Chemical Society
ARTICLE
DFT Calculations. Single-point calculations and NBO43 analysis on
2 and 4 were run on the Gaussian0344 suite of programs with the
RB3LYP45 level of theory with the 6-31þþG** basis set for all atoms.
Geometries for 2 and 4 were obtained from XRD data. To crudely assign
the relative contribution of atomic orbitals to the HOMO, we calculated
the Mulliken atomic spin density of a hypothetical {4}þ species.
Mulliken atomic spin densities of the corresponding virtual cationic
species of 2 and 4 were derived from the original single-point calcula-
tions of each singlet species by assigning 2 and 4 as doublet species
without further optimization.
{Na(12-C-4)2}{(SiPiPr3)Fe(CO)} (3). To a dark red solution of 2
(16 mg, 0.017 mmol) in THF (10 mL) was added a solution of 12-
crown-4 (6 mg, 0.034 mmol) in THF (1 mL). The reaction mixture was
stirred for 30 min at room temperature. The resulting darker solution
was filtered through Celite. Volatiles were removed under vacuum to
give a dark violet powder. The solid was collected on a medium porosity
glass-frit and washed with pentane (3 mLꢀ3). {Na(12-C-4)2}
{(SiPiPr3)Fe(CO)} (3, 18 mg, 0.017 mmol, 98%) was obtained as a
dark crystalline solid from recrystallization with THF and pentane at
room temperature. 1H NMR (THF-d8, ppm): 8.0 (bs, 3H), 7.1 (d, J =
6.9 Hz, 3H), 6.9 (m, 6H), 3.6 (32H, 12-C-4), 2.0 (bs, 6H), 0.9 (bs, 18H),
0.5 (bs, 18H). 31P NMR (THF-d8, ppm; measured at -70 ꢀC): 94.4.
Anal. Calcd for C53H86FeNaO9P3Si: C, 59.65; H, 8.12. Found: C, 59.72;
H, 8.00. IR (νCO, cm-1): 1757 (KBr pellet), 1779 (THF solution).
{Na(12-C-4)2}{(SiPiPr3)Fe(13CO)} (3-13CO). The same proce-
dure used for 2-13CO was conducted. Spectroscopic features in the 1H
NMR were identical to those for 30. 31P NMR (THF-d8, ppm; measured
at -70 ꢀC): 94.4. 13C NMR (THF-d8, ppm, measured at -70 ꢀC):
239.7 (s). IR (ν13CO, cm-1): 1713 (KBr).
(SiPiPr3)Fe(CO) (1). (SiPiPr3)Fe(N2) (0.055 g, 0.080 mmol) was
dissolved in THF (20 mL) in a 50 mL sealable Schlenk tube. The
reaction tube was taken out from the drybox and then connected to a
Schlenk line and charged with CO gas (1 atm) after freeze-pump-
thaw degassing three times. The reaction mixture was stirred overnight
at room temperature (3 h at 60 ꢀC). After all volatiles were removed by
vacuum, the resulting product was dissolved in benzene (10 mL) and
filtered through Celite. Product (SiPiPr3)Fe(CO) (1, 0.050 g, 0.072
mmol, 91%) was obtained as a yellow powder after removing volatiles by
vacuum, washing with additional portions of pentane, and drying under
(SiPiPr3)Fe(COSiMe3) (4). To a frozen solution of 2 (80 mg, 0.086
mmol) in THF (50 mL) was added trimethylsilyl triflate (20 mg, 0.090
mmol). The reaction mixture was slowly warmed to room temperature
overnight. All volatiles were removed by vacuum, and the resulting red
oil was dissolved in pentane. The red solution was filtered through
Celite, and volatiles were removed by vacuum. The resulting product,
(SiPiPr3)Fe(COSiMe3) (4, 65 mg, 0.085 mmol, 99%),46 was obtained as
a red powder after drying under vacuum. 1H NMR (C6D6, ppm): 8.14
(d, J = 7.2 Hz, 3H), 7.28 (d, J = 7.5 Hz, 3H), 7.22 (t, J = 6.9 Hz, 3H), 7.06
(t, J = 7.5 Hz, 3H), 2.40 (bs, 6H), 1.04 (m, 18H), 0.58 (bs, 18H), 0.34 (s,
9H). 31P NMR (C6D6, ppm): 104.6 (s). 29Si NMR (C6D6, ppm): 94.5
1
vacuum. H NMR (C6D6, ppm): 11.8, 7.2, 6.8, 4.6, 1.6, -3.8. Yellow
X-ray-quality crystals were grown by evaporation of a benzene/pentane
solution of 1: TBP (τ = 0.97). UV-vis (THF, nm {cm-1 M-1}): 260
{9400}, 313 {sh; ∼4700}, 375 {sh; 3100}, 990 {280}. μeff (C6D6, Evans
method,41 20 ꢀC): 2.0 μB. IR (νCO, cm-1): 1848 (KBr pellet), 1852
(THF solution). Anal. Calcd for C37H54FeOP3Si: C, 64.25; H, 7.87.
Found: C, 64.18; H, 8.35.
(SiPiPr3)Fe(13CO) (1-13CO). (SiPiPr3)Fe(N2) (0.100 g, 0.145 mmol)
was dissolved in benzene (20 mL) in a 50 mL sealable Schlenk tube. The
reaction tube was taken out from the drybox and then connected to a
Schlenk line and charged with 13CO gas (1 atm) after freeze-pump-thaw
degassing three times. The reaction mixture was then stirred overnight at
60 ꢀC. The resulting yellow solution was filtered through Celite, and all
volatiles were removed by vacuum. Product (SiPiPr3)Fe(13CO) (1-13CO,
0.100 g, 0.144 mmol, 99%) was obtained as a yellow powder after washing
with additional portions of pentane and drying under vacuum. Spectro-
2
(q, JSiP = 40 Hz), 20.3 (s). Red X-ray-quality crystals were grown by
evaporation of a pentane solution of 4 at -35 ꢀC: TBP (τ = 0.98). UV-
vis (THF, nm {cm-1 M-1}): 260 {13 500}, 424 {4100}, 990 {70}. Anal.
Calcd for C40H63FeOP3Si2: C, 62.81; H, 8.30. Found: C, 62.46; H, 7.70.
(SiPiPr3)Fe(13COSiMe3) (4-13CO). The same procedure used for
1
2-13CO was conducted. Spectroscopic features in the H NMR were
1
scopic features in the H NMR were identical to those for 1-13CO. IR
identical to those for 4. 31P NMR (C6D6, ppm): 104.5 (d, 2JCP = 16.3 Hz).
13C NMR (THF-d8, ppm; measured at room temperature): 250.3 (q,
2JCP = 16.4 Hz). 29Si NMR (C6D6, ppm; measured for 10 h at room
temperature):94.3 (dq, 2JSiP =40Hz, 2JSiC =21Hz), 20.0(d, 2JSiC =7Hz).
(SiPiPr3)Fe(COSiiPr3). To a frozen solution of 2 (38 mg, 0.041
mmol) in THF (20 mL) was added trimethylsilyl triflate (13 mg, 0.041
mmol). The reaction mixture was slowly warmed to room temperature
overnight. All volatiles were removed by vacuum, and the resulting red
oil was dissolved in pentane. The red solution was filtered through
Celite, and volatiles were removed by vacuum. The resulting product,
(SiPiPr3)Fe(COSiiPr3) (35 mg, 0.040 mmol, 98%),46 was obtained as a
(ν13CO, cm-1): 1806 (KBr pellet), 1808 (THF solution).
(SiPiPr3)Fe{CONa(THF)3} (2). A dark green solution of sodium
naphthalide was prepared by stirring a naphthalene solution (18 mg,
0.14 mmol) in THF (10 mL) over excess sodium metal (18 mg, 0.78
mmol) for 3 h at room temperature. The resulting sodium naphthalide
solution was filtered away from any remaining sodium and added
dropwise to the red solution of 1 (51 mg, 0.073 mmol) in THF (10
mL) at -35 ꢀC, causing the color change to dark red over a period of
several minutes. After stirring overnight at room temperature, the
reaction mixture was filtered through Celite. Volatiles were removed
under vacuum to give a dark red powder. The solid was collected on a
medium-porosity glass frit and washed with pentane (5 mL ꢀ 2),
benzene/pentane (1/5, 6 mL), and pentane (5 mL ꢀ 2). The resulting
product, (SiPiPr3)Fe{CONa(THF)3} (2, 67 mg, 0.072 mmol, 99%), was
1
red powder after drying under vacuum. H NMR (C6D6, ppm): 8.17
(d, J = 6.9 Hz, 3H), 7.34 (d, J = 7.2 Hz, 3H), 7.64 (t, J = 7.5 Hz, 3H), 7.09
(t, J = 7.5 Hz, 3H), 2.47 (bs, 6H), 1.48 (sep, J = 7.2 Hz, 3H), 1.22 (d, J =
7.2 Hz, 18H), 1.10 (m, 18H), 0.67 (bs, 18H). 31P NMR (C6D6, ppm):
1
2
obtained as a dark red powder after drying under vacuum. H NMR
102.6 (s). 29Si NMR (C6D6, ppm): 94.7 (q, JSiP = 42 Hz), 20.6 (s).
(THF-d8, ppm): 8.3 (bs, 3H), 7.2 (d, J = 7.5 Hz, 3H), 6.9 (t, J = 6.9 Hz,
3H), 6.7 (bs, 3H), 3.6 (THF), 1.8 (THF), 1.0 (bs, 18H), 0.60 (bs, 18H).
31P NMR (THF-d8, ppm): 94.4 (s, at -70 ꢀC). Dark red X-ray-quality
crystals were grown by diffusion of pentane vapors into a THF solution
of 2: TBP (τ = 1.0). UV-vis (THF, nm {cm-1 M-1}): 260 {13 000},
330 {sh}, 424 {3700}, ∼990 {120}. IR (νCO, cm-1): 1717 (KBr pellet),
1711 (THF solution).
UV-vis (THF, nm {cm-1 M-1}): 260 {13 000}, 435 {4100}, 990 {68}.
{(SiPiPr3)Fe(CO)}{B(3,5-(CF3)2-C6H3)4} (5). (SiPiPr3)Fe(CO)
(1, 50 mg, 0.072 mmol) was dissolved in C6H6 (10 mL) and frozen at
-35 ꢀC. After H(OEt2)2[B(3,5-(CF3)2-C6H3)4] (73 mg, 0.072 mmol)
was added to the frozen solution, the mixture was vigorously stirred for
30 min at room temperature, resulting in precipitation of a blue solid.
The precipitate was collected on a medium-porosity glass frit and
washed with benzene (10 mL ꢀ 3) and pentane (10 mL ꢀ 3). The
resulting powder, {(SiPiPr3)Fe(CO)}{B(3,5-(CF3)2-C6H3)4} (5, 80 mg,
0.051 mmol, 72%), was dried under vacuum for a short time. 1H NMR
(THF-d8, ppm): 15.2, 8.2, 7.6, 6.9, 4.6, -2.9, -6.2. Blue X-ray-quality
crystals were grown by diffusion of pentane vapors into a THF/benzene
(SiPiPr3)Fe{13CONa(THF)3} (2-13CO). The same procedure
used for 1-13CO was conducted. Spectroscopic features in the 1H
NMR were identical to those for 2. 31P NMR (THF-d8, ppm; measured
at -70 ꢀC): 94.44. 13C NMR (THF-d8, ppm, measured at -70 ꢀC):
239.6 (s). IR (ν13CO, cm-1): 1676 (KBr).
4444
dx.doi.org/10.1021/ja109678y |J. Am. Chem. Soc. 2011, 133, 4438–4446

Journal of the American Chemical Society
ARTICLE
(1/1) solution of 5: TBP (τ = 0.99). UV-vis (THF, nm {cm-1 M-1}):
380 {sh, 980}, 470 {150}, 585 {140}, 700 {sh, 45}, 845 {20}, ∼1,100
{54}. μeff (THF-d8, Evans method,41 20 ꢀC): 2.7 μB. Anal. Calcd for
C69H66BF24FeOP3Si: C, 53.30; H, 4.28. Found: C, 53.05; H, 4.20. IR
(νCO, cm-1): 1959 (KBr pellet), 1943 (THF solution). H2(g) generated
from the reaction was detected by GC analysis.
J. Am. Chem. Soc. 2005, 127, 17196–17197. (c) Sadique, A. R.; Brennessel,
W. W.; Holland, P. L. Inorg. Chem. 2008, 47, 784–786. (d) Benson, E. E.;
Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M. Chem. Soc. Rev. 2009, 38,
89–99.
(4) (a) Mankad, N. P.; Whited, M. T.; Peters, J. C. Angew. Chem., Int.
Ed. 2007, 46, 5768–5771. (b) Whited, M. T.; Mankad, N. P.; Lee, Y.;
Oblad, P. F.; Peters, J. C. Inorg. Chem. 2009, 48, 2507–2517.
(5) Lee, Y; Mankad, N. P.; Peters, J. C. Nature Chem. 2010, 2, 558–
565.
(6) Cummins, C. C. Chem. Commun. 1998, 1777–1786.
(7) Lee, C. L.; Hu, Y.; Ribbe, M. W. Science 2010, 642.
(8) Fischer, E. O.; Kreis, G.; Kreiter, C. G.; M€uller, J.; Huttner, G.;
Lorenz, H. Angew. Chem., Int. Ed. Engl. 1973, 12, 564.
(9) Transition Metal Carbyne Complexes; Kreissel, F. R., Ed.; Kluwer
Academic Publishers: Dordrecht, The Netherlands, 1993.
(10) (a) Fischer, E. O.; Schneider, J.; Neugebauer, D. Angew. Chem.,
Int. Ed. Engl. 1984, 23, 820–821. A related iron complex that was
spectroscopically but not structurally characterized has also been
described:(b) Anderson, S.; Hill, A. F. J. Organomet. Chem. 1990, 394,
C24–C26. (c) Anderson, S.; Hill, A. F.; Ng, Y. T. Organometallics 2000,
19, 15–21.
(11) (a) Joslin, F. L.; Stobart, S. R. J. Chem. Soc., Chem. Commun.
1989, 504–505. (b) Hendriksen, D. E.; Oswald, A. A.; Ansell, G. B.; Leta,
S.; Kastrup, R. V. Organometallics 1989, 8, 1153–1157.
(12) For examples of Fe(I)-CO species, see:(a) Kisko, J. L.; Hascall,
T.; Parkin, G. J. Am. Chem. Soc. 1998, 120, 10561–10562. (b) MacBeth,
D. E.; Harkins, S. B.; Peters, J. C. Can. J. Chem. 2005, 83, 332–340.
(c) Mock, M. T.; Popescu, C. V.; Yap, G. P. A.; Dougherty, W. G.; Riordan,
C. G. Inorg. Chem. 2008, 47, 1889–1891.
{(SiPiPr3)Fe(13CO)}{B(3,5-(CF3)2-C6H3)4} (5-13CO). The same
procedure used for 1-13CO was conducted. Spectroscopic features in the
1H NMR were identical to those for 5. IR (ν13CO, cm-1): 1915 (KBr).
{(SiPiPr3)Fe(Cl)}{B(3,5-(CF3)2-C6H3)4}. (SiPiPr3)Fe(CH3)
(21
mg, 0.031 mmol) was dissolved in CH2Cl2 (5 mL) and cooled to -
35 ꢀC. After H(OEt2)2[B(3,5-(CF3)2-C6H3)4] (30 mg, 0.030 mmol) was
added to the red solution, the mixture was vigorously stirred for 30 min at
room temperature, resulting in a dark green solution. After volatiles were
removed by vacuum, the resulting powder was collected on a medium-
porosity glass frit and washed with benzene (10 mL ꢀ 3) and pentane (10
mL ꢀ 3). The resulting product, {(SiPiPr3)Fe(Cl)}{B(3,5-(CF3)2-C6H3)4}
1
(30 mg, 0.019 mmol, 65%), was then dried under vacuum. H NMR
(CD2Cl2, ppm): 11.26, 8.27, 7.69, 7.54, 6.85, -13.04. Dark red X-ray-quality
crystals were grown by diffusion of pentane vapors into a CH2Cl2 solution of
{(SiPiPr3)Fe(Cl)}{B(3,5-(CF3)2-C6H3)4}: TBP (τ = 1.0). UV-vis (THF,
nm {cm-1 M-1}): 387 {4500}, 430 {sh, 3300}, 512 {930}, 596 {670}, 841
{470}. μeff (CD2Cl2, Evans method,41 20 ꢀC): 3.9 μB. Anal. Calcd for
C68H66BClF24FeP3Si: C, 52.28; H, 4.26. Found: C, 52.31; H, 4.39.
’ ASSOCIATED CONTENT
S
(13) Addison, A. W.; Rao, T. N.; Reedijk, J.; Vanrijn, J.; Verschoor,
G. C. J. Chem. Soc., Dalton Trans. 1984, 1349–1356.
Supporting Information. Characterization data for 1-5;
b
crystallographic data for 1, 2, 4, and 5 (CIF); details of DFT
calculations; complete ref 44. This material is available free of
charge via the Internet at http://pubs.acs.org.
(14) (a) Ayoko, G. A.; Eaborn, C.; Hitchcock, P. B. Organometallics
1995, 14, 4960–4961. (b) Al-Juaid, S. S.; Al-Nasr, A. A.; Ayoko, G. A.;
Eaborn, C.; Hitchcock, P. J. Organomet. Chem. 1995, 488, 155–160.
(15) (a) Vrtis, R. N.; Rao, C. P.; Warner, S.; Lippard, S. J. J. Am. Chem.
Soc. 1988, 110, 2669–2670. (b) Vrtis, R. N.; Liu, S.; Rao, S. P.; Bott, S. G.;
Lippard, S. J. Organometallics 1991, 10, 275–285. (c) Protasiewicz, J. D.;
Masschelein, A.; Lippard, S. J. J. Am. Chem. Soc. 1993, 115, 808–810.
(d) Sakaba, H.; Yoshida, M.; Kabuto, C.; Kabuto, K. J. Am. Chem. Soc.
2005, 127, 7276–7277.
’ AUTHOR INFORMATION
Corresponding Author
jpeters@caltech.edu
(16) (a) Stang, P. J.; Crittell, C. M.; Arif, A. M.; Karni, M.; Apeloig, Y.
J. Am. Chem. Soc. 1991, 113, 7461–7470. (b) Gleiter, R.; Werz, D. B.;
Rominger, F.; Zhutov, E.; Zefirov, N. S.; Proskurnina, M. V. Eur. J. Org.
Chem. 2007, 5834–5839.
(17) A Wiberg bond index analysis suggests strong multiple bond
character in the Fe-C linkage (bond index = 1.85) and single bond
character at the C-O linkage (bond index = 0.68) of 4. See Supporting
Information for details.
’ ACKNOWLEDGMENT
We acknowledge the NSF (CHE-0750234) for financial
support. Dr. Neal P. Mankad provided assistance with XRD
analyses, and Prof. Theodore A. Betley provided access to a
M€ossbauer spectrometer at Harvard University. Prof. Michael T.
Green and Dr. Marc-Etienne Moret are acknowledged for
insightful discussions concerning theoretical data. We also want
to express gratitude to a reviewer for suggestions that improved
the reported X-ray data.
(18) Parkin, G. In Comprehensive Organometallic Chemistry; Crab-
tree, R. H., Mingos, D. M. P., Eds.; Elsevier: Amsterdam, 2007; Vol. 1, pp
1-57.
(19) (a) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2003, 125,
10782–10783. (b) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2004, 126,
6252–6254. (c) 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. (d) Thomas, C. M.; Mankad, N. P.;
Peters, J. C. J. Am. Chem. Soc. 2006, 128, 4956–4957. (e) Vogel, C.;
Heinemann, F. W.; Sutter, J.; Anthon, C.; Meyer, K. Angew. Chem., Int.
Ed. 2008, 47, 2681–2684. (f) Scepaniak, J. J.; Young, J. A.; Bontchev,
R. P.; Smith, J. M. Angew. Chem., Int. Ed. 2009, 48, 3158–3160.
(20) Detrich, J. L.; Konecny, R.; Vetter, W. M.; Doren, D.; Rhein-
gold, A. L.; Theopold, K. H. J. Am. Chem. Soc. 1996, 118, 1703–1712.
(21) (a) Gibb, T. C. Principles of M€ossbauer Spectroscopy; Wiley: New
York, 1976. (b) Bancroft, G. M. M€ossbauer Spectroscopy—An Introduc-
tion for Inorganic and Geochemists; McGraw-Hill: London, 1971.
(22) (a) Clark, M. G.; Cullen, W. R.; Garrod, R. E. B.; Maddock,
A. G.; Sams, J. R. Inorg. Chem. 1973, 12, 1045. (b) Pribula, C. G.; Brown,
T. L.; Munck, E. J. Am. Chem. Soc. 1974, 96, 4149.
’ REFERENCES
(1) (a) Mindiola, D. J.; Hillhouse, G. L. J. Am. Chem. Soc. 2001, 123,
4623–4624. (b) Jenkins, D. M.; Betley, T. A.; Peters, J. C. J. Am. Chem.
Soc. 2002, 124, 11238–11239. (c) Brown, S. D.; Betley, T. A.; Peters,
J. C. J. Am. Chem. Soc. 2003, 125, 322–323. (d) Dai, X. L.; Warren, T. H.
J. Am. Chem. Soc. 2004, 126, 10085–10094. (e) Ni, C. B.; Fettinger, J. C.;
Long, G. J.; Brynda, M.; Power, P. P. Chem. Commun. 2008, 6045–6047.
(f) Shay, D. T.; Yap, G. P. A.; Zakharov, L. N.; Rheingold, A. L.;
Theopold, K. H. Angew. Chem., Int. Ed. 2005, 44, 1508–1510.
(g) Ingleson, M. J.; Pink, M.; Fan, H.; Caulton, K. G. J. Am. Chem.
Soc. 2008, 130, 4262–4267.
(2) (a) Kieber-Emmons, M. T.; Riordan, C. G. Acc. Chem. Res. 2007,
40, 618–625. (b) Kitiachvili, K. D.; Mindiola, D. J.; Hillhouse, G. H.
J. Am. Chem. Soc. 2004, 126, 10554–10555.
(3) (a) Lu, C. C.; Saouma, C. T.; Day, M. W.; Peters, J. C. J. Am.
Chem. Soc. 2007, 129, 4–5. (b) Laitar, D. S.; M€uller, P.; Sadighi, J. P.
4445
dx.doi.org/10.1021/ja109678y |J. Am. Chem. Soc. 2011, 133, 4438–4446

Journal of the American Chemical Society
ARTICLE
(23) Sosinsky, B. A.; Norem, N.; Shong, R. G. Inorg. Chem. 1982, 21,
4229–4233.
(24) Collins, R. L.; Pettit, R. J. Chem. Phys. 1963, 39, 3433.
(25) Sosinsky, B. A.; Norem, N.; Shong, R. G. Inorg. Chem. 1982, 21,
4229–4233.
(26) Dias, G. H. M.; Morigaki, M. K. Polyhedron 1992, 11, 1629–
1636.
(27) Schmidt, G.; Balk, H. J. J. Organomet. Chem. 1974, 80, 257.
(28) To crudely estimate the relative contribution of atomic orbitals
to the HOMO, we calculated the Mulliken atomic spin density of the
hypothetical {4}þ.
(29) (a) Feed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988,
899.(b) Weinhold, F.; Landis, C. R. Valency and Bonding: A Natural Bond
Ordital Donor-Acceptor Perspective; Cambridge University Press: New
York, 2005.
(30) (a) Enemark, J. H.; Feltham, R. D. Coord. Chem. Rev. 1974, 13,
339–406.(b) Richter-Addo, G. B.; Legzdin, P. Metal Nitrosyls; Oxford
University Press: Oxford, UK, 1992.
(31) (a) Hill, A. F. Organometallics 2006, 25, 4741–4743. (b) Parkin,
G. Organometallics 2006, 25, 4744–4747. (c) Figueroa, J. S.; Melnick,
J. G.; Parkin, G. Inorg. Chem. 2006, 45, 7056–7058. (d) Sircoglou, M.;
Bontemps, S.; Bouhadir, G.; Saffon, N.; Miqueu, K.; Gu, W.; Mercy, M.;
Chen, C.-H.; Foxman, B. M.; Maron, L.; Ozerov, O. V.; Bourissou, D.
J. Am. Chem. Soc. 2008, 130, 16729–16738. (e) Kuzu, I.; Krummenacher,
I.; Meyer, J.; Armbruster, F.; Breher, F. Dalton Trans. 2008, 5836–5865.
(f) Braunschweig, H.; Dewhurst, R. D.; Schneider, A. Chem. Rev. 2010,
110, 3924–3957.
(32) (a) Gualco, P.; Lin, T.-P.; Sircoglou, M.; Mercy, M.; Ladeira, S.;
Bouhadir, G.; Pꢀerez, L. M.; Amgoune, A.; Maron, L.; Gabbaï, F.;
Bourissou, D. Angew. Chem., Int. Ed. 2009, 48, 9892–9895. (b) Corey,
J. Y.; Braddock-Wilking, J. Chem. Rev. 1999, 99, 175–292.
(33) Pauling, L. J. Phys. Chem. 1952, 56, 361–365.
(34) (a) Chen, Y.; Petz, W.; Frenking, G. Organometallics 2000, 19,
2698–2706. (b) Krapp, A.; Pandey, K. K.; Frenking, G. J. Am. Chem. Soc.
2007, 129, 7596–7610.
(35) Schrock, R. R. J. Am. Chem. Soc. 1974, 96, 6796–6797.
(36) McLain, S. J.; Wood, C. D.; Messerle, L. W.; Schrock, R. R.;
Hollander, F. J.; Youngs, W. J.; Churchill, M. R. J. Am. Chem. Soc. 1978,
100, 5962–5964.
(37) Mankad, N. P.; M€uller, P.; Peters, J. C. J. Am. Chem. Soc. 2010,
132, 4083–4085.
(38) (a) Peters, J. C.; Odom, A. L.; Cummins, C. C. J. Chem. Soc.,
Chem. Commun. 1997, 1995. (b) Carlson, R. G.; Gile, M. A.; Heppert,
J. A.; Mason, M. H.; Powell, D. R.; Velde, D. V.; Vilain, J. M. J. Am. Chem.
Soc. 2002, 124, 1580–1581.
(39) (a) Yakelis, N. A.; Bergman, R. G. Organometallics 2005, 24,
3579–3581. (b) Brookhart, M.; Grant, B.; Volpe, A. F., Jr. Organome-
tallics 1992, 11, 3920–3922.
(40) Sheldrick, G. M. Acta Crystallogr. A 2008, 64, 112–122.
(41) (a) Evans, D. F. J. Chem. Soc. 1959, 2003–2005. (b) Sur, S. K.
J. Magn. Reson. 1989, 82, 169–173.
(42) Neese, F. QCPE Bull. 1995, 15, 5.
(43) (a) Feed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988,
899.(b) Weinhold, F.; Landis, C. R. Valency and Bonding: A Natural Bond
Orbital Donor-Acceptor Perspective; Cambridge University Press: New
York, 2005.
(44) Frisch, M. J.; et al. Gaussian 03, Revision C.02; Gaussian, Inc.:
Wallingford, CT, 2004.
(45) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Lee, C.;
Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(46) A trace amount of 1 can be seen in the 1H NMR spectrum.
4446
dx.doi.org/10.1021/ja109678y |J. Am. Chem. Soc. 2011, 133, 4438–4446