Structural and spectroscopic characterization of 17- and 18-electron piano-stool complexes of chromium. Thermochemical analyses of weak Cr-H bonds

Article abstract of DOI:10.1021/ic302460y

The 17-electron radical CpCr(CO)₂(IMe)^? (IMe = 1,3-dimethylimidazol-2-ylidene) was synthesized by the reaction of IMe with [CpCr(CO)₃]₂, and characterized by single crystal X-ray diffraction and by electron paramagnetic resonance (EPR), IR, and variable temperature ¹H NMR spectroscopy. The metal-centered radical is monomeric under all conditions and exhibits Curie paramagnetic behavior in solution. An electrochemically reversible reduction to 18-electron CpCr(CO) ₂(IMe)^- takes place at E_1/2 = -1.89(1) V vs Cp₂Fe^+?/0 in MeCN, and was accomplished chemically with KC₈ in tetrahydrofuran (THF). The salts K⁺(18-crown- 6)[CpCr(CO)₂(IMe)]^-·¹/₂THF and K⁺[CpCr(CO)₂(IMe)]^-·³/ ₄THF were crystallographically characterized. Monomeric ion pairs are found in the former, whereas the latter has a polymeric structure because of a network of K·O_(CO) interactions. Protonation of K ⁺(18-crown-6)[CpCr(CO)₂(IMe)] ^-·¹/₂THF gives the hydride CpCr(CO) ₂(IMe)H, which could not be isolated, but was characterized in solution; a pK_a of 27.2(4) was determined in MeCN. A thermochemical analysis provides the Cr-H bond dissociation free energy (BDFE) for CpCr(CO)₂(IMe)H in MeCN solution as 47.3(6) kcal mol^-1. This value is exceptionally low for a transition metal hydride, and implies that the reaction 2 [Cr-H] → 2 [Cr^?] + H₂ is exergonic (ΔG = -9.0(8) kcal mol^-1). This analysis explains the experimental observation that generated solutions of the hydride produce CpCr(CO)₂(IMe)^? (typically on the time scale of days). By contrast, CpCr(CO)₂(PCy₃)H has a higher Cr-H BDFE (52.9(4) kcal mol^-1), is more stable with respect to H ₂ loss, and is isolable.

Full text of DOI:10.1021/ic302460y

Article
pubs.acs.org/IC
Structural and Spectroscopic Characterization of 17- and 18-Electron
Piano-Stool Complexes of Chromium. Thermochemical Analyses of
Weak Cr−H Bonds
Edwin F. van der Eide,† Monte L. Helm,† Eric D. Walter,‡ and R. Morris Bullock*,†
‡
†Chemical and Materials Sciences Division and Environmental and Molecular Sciences Laboratory, Pacific Northwest National
Laboratory, P.O. Box 999, Richland, Washington 99352, United States
S
* Supporting Information
ABSTRACT: The 17-electron radical CpCr(CO)2(IMe)•
(IMe = 1,3-dimethylimidazol-2-ylidene) was synthesized by
the reaction of IMe with [CpCr(CO)3]2, and characterized by
single crystal X-ray diffraction and by electron paramagnetic
resonance (EPR), IR, and variable temperature 1H NMR
spectroscopy. The metal-centered radical is monomeric under
all conditions and exhibits Curie paramagnetic behavior in
solution. An electrochemically reversible reduction to 18-
electron CpCr(CO)2(IMe)− takes place at E1/2 = −1.89(1) V
vs Cp2Fe+•/0 in MeCN, and was accomplished chemically with
KC8 in tetrahydrofuran (THF). The salts K+(18-crown-
6)[CpCr(CO)2 (IMe)]− ·1 /2 THF and K+ [CpCr-
(CO)2(IMe)]−·3/4THF were crystallographically character-
ized. Monomeric ion pairs are found in the former, whereas the latter has a polymeric structure because of a network of
K···O(CO) interactions. Protonation of K+(18-crown-6)[CpCr(CO)2(IMe)]−·1/2THF gives the hydride CpCr(CO)2(IMe)H,
which could not be isolated, but was characterized in solution; a pKa of 27.2(4) was determined in MeCN. A thermochemical
analysis provides the Cr−H bond dissociation free energy (BDFE) for CpCr(CO)2(IMe)H in MeCN solution as 47.3(6) kcal
mol−1. This value is exceptionally low for a transition metal hydride, and implies that the reaction 2 [Cr−H] → 2 [Cr•] + H2 is
exergonic (ΔG = −9.0(8) kcal mol−1). This analysis explains the experimental observation that generated solutions of the
hydride produce CpCr(CO)2(IMe)• (typically on the time scale of days). By contrast, CpCr(CO)2(PCy3)H has a higher Cr−H
BDFE (52.9(4) kcal mol−1), is more stable with respect to H2 loss, and is isolable.
We have recently shown that N-heterocyclic carbenes
(NHCs) significantly stabilize 17-electron tungsten radicals
CpW(CO)2(L)•.10−12 In accordance with the increased
stability of CpW(CO)2(NHC)• with respect to the parent
INTRODUCTION
■
Metal-centered 17-electron radicals have received much
attention for several decades.1 Such radicals are usually very
reactive. Typical reactivity patterns are dimerization,2 binding
of a donor ligand to form 19-electron adducts3 (of importance
to substitution4 and disproportionation5 reactions), and
halogen atom abstraction2 from halogenated hydrocarbons.
Metal-centered radicals have also been observed6 to abstract
hydrogen atoms from hydrocarbons to give 18-electron
hydrides; the reverse reaction, hydrogen atom transfer from
transition metal hydrides to hydrocarbyl radicals, has been
studied7 in detail. Hydrogen atom transfer to and from organic
compounds, mediated by transition metal hydrides and radicals,
have found application in chain transfer processes in radical
polymerizations and in reductive cyclization reactions.8,9 In this
regard, determination of the thermodynamic properties of M-H
bonds, either calorimetrically or using thermochemical cycles, is
of interest. In 18-electron metal hydrides, M-H homolytic bond
dissociation enthalpies (BDEs) as high as 82 kcal mol−1
(CpOs(CO)2H)6 and as low as 55 kcal mol−1 (HV-
(CO)4(Ph2P(CH2)4PPh2))9 have been measured.
•
CpW(CO)3 , the W-H bonds in CpW(CO)2(NHC)H were
found11,12 to be 5−8 kcal mol−1 weaker than in CpW(CO)3H.
Because chromium hydrides generally have weaker M-H bonds
than tungsten hydrides (e.g.: BDE = 62 kcal mol−1 for
CpCr(CO)3H13,14 vs 73 kcal mol−1 for CpW(CO)3H13), we
reasoned that CpCr(CO)2(NHC)H should have a particularly
weak Cr−H bond. Here we report a structural, electrochemical,
and spectroscopic study of the 17-electron radical CpCr-
(CO)2(IMe)• (IMe = 1,3-dimethylimidazol-2-ylidene) and the
electron-rich anion CpCr(CO)2(IMe)− and the hydride
CpCr(CO)2(IMe)H derived from it. An analogous series of
tricyclohexylphosphine complexes is also described. We show
that CpCr(CO)2(IMe)H (which is not isolable in pure form)
has a Cr−H bond that is 9 kcal mol−1 weaker than in the parent
CpCr(CO)3H, and even weaker than that of HV(CO)4(Ph2P-
Received: November 9, 2012
Published: January 23, 2013
© 2013 American Chemical Society
1591
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603

Inorganic Chemistry
Article
Scheme 1
(CH2)4PPh2),9 whereas Cr−H bond weakening is less dramatic
in CpCr(CO)2(PCy3)H.
RESULTS
■
Synthesis and Structure of CpCr(CO)2(IMe)•. CpCr-
(CO)2(IMe)• was synthesized (Scheme 1) by the addition of a
freshly generated THF/Et2O solution of 1,3-dimethylimidazol-
2-ylidene15 (IMe) to a dark green suspension of [CpCr(CO)3]2
in Et2O at 20 °C, resulting in the evolution of a gas
(presumably CO) and a color change from dark green to
light orange. Precipitation of the product by addition of hexane,
followed by sublimation at <1 mTorr at 100−110 °C, afforded
analytically pure CpCr(CO)2(IMe)• in 80% yield as an
amorphous, dull orange-brown powder.
Figure 1. Two views of one of the four crystallographically
independent molecules of CpCr(CO)2(IMe)• (50% probability
ellipsoids, hydrogens omitted). Selected bond lengths (Å): Cr1−
Cpcentroid, 1.873(3); Cr1−C3, 2.053(6); Cr1−C1, 1.822(6); Cr1−C2,
1.804(6); C1−O1, 1.166(7); C2−O2, 1.192(8). Selected bond angles
(deg): C1−Cr1−C2, 78.2(3); C1−Cr1−C3, 96.6(3); C2−Cr1−C3,
95.7(3).
Orange single crystals of CpCr(CO)2(IMe)• were grown by
diffusion of hexane into a fluorobenzene solution, and were
analyzed by X-ray diffraction. The complex crystallizes in the
centrosymmetric space group P1, and the asymmetric unit
̅
contains four crystallographically independent molecules. Since
the independent molecules have nearly identical metrical
parameters (see below and in the Supporting Information,
Table S1), only one molecule is displayed in Figure 1. The
shortest separation between any pair of chromium atoms is
6.3952(14) Å, confirming the monomeric nature of the radicals.
In the nearly Cs-symmetric molecules, the carbene orientation
is such that the approximate mirror plane bisects the carbene
ligand. We quantify this orientation using the parameter θ,
which we previously12 defined as the absolute value of the
dihedral angle between the planes Cpcentroid−Cr−C(carbene) and
Cr−C(carbene)−(N)2. For the molecule shown in Figure 1, θ is
89.6°, and it falls in the range 85−88° for the other three
molecules. The three legs of the piano-stool geometry are
irregularly positioned around Cr: the C(CO)−Cr−C(CO) angle of
78.2(3)° is significantly smaller than the C(carbene)−Cr−C(CO)
angles of 96.6(3) and 95.7(3)°. (In the four independent
molecules, angles C(CO)−Cr−C(CO) span the range 76.1(3) to
78.6(3)°, and the angles C(carbene)−Cr−C(CO) span the range
95.1(3) to 100.6(3)°.)
in hexane, but dissolves readily in aromatic and in polar
solvents, giving yellow, air-sensitive solutions.16 The IR
spectrum in tetrahydrofuran (THF) (Table 1) shows carbonyl
stretching bands (νCO
̃
) at 1899 and 1778 cm−1. Two bands are
always observed, regardless of the solvent or the concentration.
We also recorded the near-infrared (NIR) spectrum, based on
Atwood and Geiger’s observation17 that 17-electron piano-stool
complexes undergo low-energy electronic transitions involving
the singly occupied molecular orbital (SOMO). As shown in
Figure 2, CpCr(CO)2(IMe)• displays a weak absorption at λmax
= 1455 nm.
The 1H NMR spectrum of CpCr(CO)2(IMe)• displays
paramagnetically broadened and shifted resonances. In toluene-
d8 (500 MHz, 295 K) these are found at 30.7 (≈ 2H, Δν1/2
≈
4500 Hz), 12.9 (≈ 5H, Δν1/2 = 690 Hz) and 3.1 ppm (≈ 6H,
Δν1/2 = 240 Hz); similar shifts and linewidths are observed in
CD3CN solvent. The temperature dependence of the observed
1H chemical shifts (δobs) was investigated in toluene-d8 over the
temperature range −30 to +70 °C (Figure 3). Equation 1
shows the expected temperature dependence of δobs for a Curie
paramagnet, which CpCr(CO)2(IMe)• obeys. The term CMT−1
is the hyperfine shift, and δdia is the (hypothetical) diamagnetic
Spectroscopic and Electrochemical Characterization
of CpCr(CO)2(IMe)•. CpCr(CO)2(IMe)• is sparingly soluble
1592
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603

Inorganic Chemistry
Article
a
Table 1. IR Data (νCO
̃
, in cm−1)
entry
complex
MeCN
THF
1
2
3
4
5
6
7
8
9
CpCr(CO)2(IMe)•
1893
1902
1737
1737
1762
1900
1910
1977
1980
1767
1777
1655
1656
1681
1810
1841
1908
1908
1899
1909
1725
1739
1767
1904
1917
1778
CpCr(CO)2(PCy3)•
1787
1644
1653
1681
1819
1851
K+[CpCr(CO)2(IMe)]−·3/4THF
K+(18-crown-6)[CpCr(CO)2(IMe)]−·1/2THF
K+(18-crown-6)[CpCr(CO)2(PCy3)]−
CpCr(CO)2(IMe)H
CpCr(CO)2(PCy3)H
[CpCr(CO)2(IMe)(MeCN)]+[B(C6F5)4]−
[CpCr(CO)2(PCy3)(MeCN)]+[B(C6F5)4]−
a
Spectra are provided in the Supporting Information
Figure 2. NIR spectra of CpCr(CO)2(IMe)• (blue trace) and
CpCr(CO)2(PCy3)• (black trace), recorded in hexafluorobenzene.
Figure 4. EPR (ν = 9.4 GHz) spectra of CpCr(CO)2(IMe)• in
toluene. The lower lines in each pair are obtained by simulation (290
K: giso = 2.028, Aiso(53Cr) = 35 MHz; 80 K: g1 = 2.068, g2 = 2.018, g3 =
1.995, A1(53Cr) = 65 MHz, A2(53Cr) = 30 MHz, A3(53Cr) = 10 MHz).
See Supporting Information for simulation details.
1
agreement with the small (<1 MHz) AH determined by H
Figure 3. Temperature dependence of the observed 1H chemical shifts
of CpCr(CO)2(IMe)• (500 MHz, toluene-d8) from −30 to +70 °C.
NMR spectroscopy. The spectra were simulated for a low spin
(S = /2) system; the employed spin-Hamiltonian parameters
1
are provided in the caption of Figure 4.
Cyclic voltammetry of CpCr(CO)2(IMe)• in MeCN reveals
an electrochemically reversible (ia/ic ≈ 0.97 at υ = 0.1 V s−1,
Figure 5) reduction at E1/2 = −1.89(1) V vs Cp2Fe+•/0, with a
65 mV peak-to-peak separation (cf. 66 mV for Cp2Co+/• when
cobaltocene is added as a reference); we assign this wave to the
couple CpCr(CO)2(IMe)•/−. A quasi-reversible (ic/ia ≈ 0.66 at
υ = 0.1 V s−1) oxidation with a peak-to-peak separation of 95
mV occurs at E1/2 = −0.61(1) V vs Cp2Fe+•/0. The cathodic
wave is broader than the anodic wave, and broadens further as
the scan rate is increased. Since separate experiments (see
below) show that the acetonitrile adduct CpCr(CO)2(IMe)-
(MeCN)+ is produced upon oxidation, the quasi-reversible
process is assigned to the couple CpCr(CO)2(IMe)(MeCN)+/
CpCr(CO)2(IMe)•.
shift obtained at T−1 = 0. The electron−proton coupling
constants (AH, in Hz) can be obtained from the slopes CM by
the relationship in Equation 2,18 which gives AH = 0.94, 0.26,
and 0.07 MHz for the carbene CH, Cp, and carbene CH3
protons, respectively. Additionally, y-intercepts of 5.3(3),
6.0(1), and 1.3(1) ppm are realistic diamagnetic shifts expected
for these types of protons.
δobs = CMT−1 + δ
(1)
dia
3γHkB
AH = C
M 2πg μBS(S + 1)
(2)
e
Electron paramagnetic resonance (EPR) spectra of CpCr-
(CO)2(IMe)• were recorded at 9.4 GHz in toluene, in both the
fluid and the frozen state (Figure 4). Hyperfine coupling (A) to
53Cr (I = 3/2, 9.6% natural abundance) is observed in both
Reactivity of CpCr(CO)2(IMe)•. We treated CpCr-
(CO)2(IMe)• with 1000 psi H2 in CD3CN in a polyether
ether ketone (PEEK) NMR tube,19 and recorded H NMR
1
1
spectra, but coupling to H nuclei is not observed, which is in
spectra at 1 h intervals, to check for the possible formation of
1593
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603

Inorganic Chemistry
Article
Figure 5. Background-corrected cyclic voltammograms of CpCr-
−
(CO)2(IMe)• (υ = 0.1 V s−1, 0.1 M nBu4N+PF6 in MeCN).
Horizontal arrows indicate initial scan directions (no Faradaic current
passing at the start of the scan). The dotted voltammogram is
vertically offset for clarity.
18-electron hydride CpCr(CO)2(IMe)H (Scheme 1). How-
ever, no observable amounts of the hydride formed even after
16 h. (We have generated CpCr(CO)2(IMe)H in solution by
protonation of CpCr(CO)2(IMe)−, as described in a later
section.) Further, treatment of CpCr(CO)2(IMe)• with the
radical trap20 2,6-di-tert-butyl-1,4-benzoquinone did not result
in adduct formation, as the IR spectrum remained unchanged.
In contrast, we found12 earlier that the tungsten radical
CpW(CO)2(IMe)• does form a stable adduct.
Figure 6. Two crystallographically independent molecules of K+(18-
crown-6)[CpCr(CO)2(IMe)]−·1/2THF (50% probability ellipsoids;
hydrogens and THF solvent omitted). Selected metrical data are
provided in Table 2.
Chemical redox reactions proceeded in accordance with the
electrochemical observations. Oxidation of CpCr(CO)2(IMe)•
with Ph3C+B(C6F5)4− (E°(Ph3C+) ≈ −0.1 V vs Cp2Fe+•/0
)
21 in
We have also determined the crystal structure of K+[CpCr-
(CO)2(IMe)]− in the absence of the crown ether. A saturated
THF solution of K+[CpCr(CO)2(IMe)]− (no 18-crown-6
added), allowed to concentrate slowly over several weeks,
deposited several orange single crystals. Analysis by X-ray
diffraction showed them to be of composition K+[CpCr-
(CO)2(IMe)]−·3/4THF, crystallized in the space group P21,
with an asymmetric unit containing four formula units of
K+[CpCr(CO)2(IMe)]− and three molecules of THF. Two
THF molecules are simply solvents of crystallization, and the
remaining tetrameric conglomerate {K+ [CpCr-
(CO)2(IMe)]−}4(THF) is the building block for somewhat
flattened pillars that run parallel to the crystallographic a axis
(Figure 7 and Supporting Information, Figure S1). Cyclo-
pentadienide, carbene, and potassium-coordinating THF
ligands form the periphery of the pillars, while the carbonyl
ligands point toward the centrally located potassium cations,
forming a network of K···O(CO) interactions. Nineteen unique
K···O(CO) interactions occur (average d(K···O(CO)) = 2.77 Å;
range = 2.59−2.97 Å); only one in four potassium cations is
terminally ligated by a THF (d(K···O) = 2.745(2) Å).
Parameters pertaining to these and other interactions are
tabulated in detail in the Supporting Information.
MeCN generated a species with increased νCO values (Table 1),
̃
which we assign to the cation cis-CpCr(CO)2(IMe)(MeCN)+.
Coordination of MeCN is crucial for the stabilization of the
cation; complete decomposition took place within 5 min when
the oxidation of CpCr(CO)2(IMe)• was performed in
fluorobenzene. Even in neat MeCN, the stability of the cation
is rather limited; it stays intact for short periods of time
(minutes), but complete decomposition of cis-CpCr-
(CO)2(IMe)(MeCN)+ occurs within an hour. We have not
undertaken further efforts to isolate it. On the other hand,
reduction of CpCr(CO)2(IMe)• with KC8 in THF proceeded
smoothly and cleanly to give an orange solution of K+[CpCr-
(CO)2(IMe)]− (Scheme 1). This ion pair was thoroughly
characterized, as described below.
Structures of K+(18-crown-6)[CpCr(CO)2(IMe)]− and
K+[CpCr(CO)2(IMe)]−. Addition of 18-crown-6 to a filtered
THF solution of K+[CpCr(CO)2(IMe)]− and subsequent
precipitation with hexane gave K+(18-crown-6)[CpCr-
(CO)2(IMe)]−·1/2THF as fine, orange needles in 87% yield.
Single crystals were grown from a supersaturated THF solution
at room temperature, giving orange blocks suitable for X-ray
diffraction. The complex crystallizes in the non-centrosym-
metric space group Pca21, with two crystallographically
independent contact ion pairs and one THF molecule per
asymmetric unit. In both ion pairs (Figure 6), all of the six
crown oxygens bind to the potassium cation. The range of
K···O(crown) distances is 2.83−3.10 Å (average: 2.94 Å) for K1,
and 2.83−3.00 Å (average: 2.91 Å) for K2, and the potassiums
are displaced significantly (0.90 Å for K1; 0.84 Å for K2) away
from the mean planes formed by the crown oxygens, toward
the carbonyls on the Cr. Whereas both Cr1-bound carbonyls
further interact22 with K1, only one Cr2-bound carbonyl
undergoes such an interaction with K2 (see also Table 2).
The four individual anionic Cr fragments, together with the
nearest potassium cations, are displayed in Figure 8. In addition
to the K···O interactions, the carbene ligands on Cr1 and Cr2
interact with K1 and K3 in κ1 and η3 fashions, respectively
(K1···N2, 3.301(3); K3···N3, 3.276(3); K3···N4, 3.339(3);
K3···C15, 3.073(3) Å). Whereas both carbonyls on Cr3 and
Cr4 pinch a potassium cation (K4 and K1, respectively), there
is no such pinching for the Cr1 and Cr2 carbonyls.
NMR Characterization of K+(18-crown-6)[CpCr-
(CO)2(IMe)]−·1/2THF. Generation and Characterization
of CpCr(CO)2(IMe)H. The 1H NMR spectrum of K+(18-
1594
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603

Inorganic Chemistry
Article
Table 2. Selected Bond Lengths (Å) and (Dihedral) Angles (deg) for the K+[CpCr(CO)2(IMe)]− Salts
K(18-crown-6)+
[CpCr(CO)2(IMe)]−·1/2THF
K+[CpCr(CO)2(IMe)]−·3/4THF
Cr1
Cr2
Cr1
Cr2
Cr3
Cr4
Cr−Cpcentroid
Cr−C(carbene)
Cr−C(CO)
1.863(1)
2.045(2)
1.867(1)
2.049(2)
1.862(2)
2.036(3)
1.862(2)
2.055(3)
1.871(2)
2.032(3)
1.876(2)
2.045(3)
a
1.779(2),
1.777(-
2)
1.778(2),
1.779(2)
1.784(3),
1.757(-
3)
1.768(3),
1.777(-
3)
1.772(3),
1.776(-
3)
1.774(3),
1.774(-
3)
a
C−O
1.195(2),
1.197(-
2)
1.200(2),
1.209(2)
1.201(4),
1.227(-
4)
1.216(4),
1.202(-
4)
1.203(4),
1.207(-
4)
1.198(4),
1.205(-
4)
a
b
c
c
c
c
K···O(CO)
3.003(1),
2.971(-
1)
3.859(2) ,
2.852(1)
2.80
2.78
2.78
2.73
d
d
K···O(crown or THF)
C(CO)−Cr−C(CO)
θ
2.94
2.91
2.745(2)
81.86(7)
84.7
89.41(8)
87.1
92.23(14)
73.9
91.76(15)
80.5
84.06(14)
88.6
86.75(14)
83.5
a
b
When two values are given, the first value applies to the CO ligand with the lowest label number (see Figures 6 and 8). Not considered a
c
d
significant K···O interaction. Average of six, five, or four values (see Figure 8); average of all 19 values is 2.77 Å. Average of six values.
Figure 8. Structures of the four CpCr(CO)2(IMe)− fragments and the
nearest K+ ions (50% probability ellipsoids; hydrogens omitted) in
K+[CpCr(CO)2(IMe)]−·3/4THF. Atoms labeled with superscript ‘i’ or
‘ii’ are related to their nonsuperscripted counterparts by unit
translation along the crystallographic a axis. Selected metrical data
Figure 7. Perspective view of the crystal structure of K+[CpCr-
are provided in Table 2.
(CO)2(IMe)]−·3/4THF along the crystallographic a axis.
crown-6)[CpCr(CO)2(IMe)]−·1/2THF in CD3CN displays
broad resonances at 7.0 (≈ 2H, Δν1/2 = 200 Hz), 4.09 (≈
5H, Δν1/2 = 35 Hz), and 3.86 ppm (≈ 6H, Δν1/2 = 7 Hz) for
the protons belonging to the anion. The addition23 of about 1
mg of KC8 resulted in a drastic sharpening and slight shifting of
these signals, now appearing as singlets at 6.84 (2H), 4.00
(5H), and 3.84 ppm (6H).
produced in about 95% yield. Although the hydride was
generated cleanly, it is not thermodynamically stable; IR signals
due to CpCr(CO)2(IMe)• always became noticeable after 10−
60 min, and grew in intensity at the expense of the signals due
to CpCr(CO)2(IMe)H. A 45 mM solution of CpCr-
(CO)2(IMe)H in CD3CN was found by 1H NMR spectroscopy
to be 75% decomposed after 1 day at room temperature.
CpCr(CO)2(IMe)• was identified as the major organometallic
product, and a weak singlet at 4.57 ppm was observed for H2.
The same observations were made when a solution of
CpCr(CO)2(IMe)H was maintained in the dark. We were
unable to isolate samples of CpCr(CO)2(IMe)H because of its
instability, which seems to increase at higher concentrations.
For example, removal of volatiles from a freshly generated
Et2O/THF solution of CpCr(CO)2(IMe)H gave a residue
which, by IR spectroscopy, consisted of less than 20% of the
hydride, the remainder being the 17-electron radical. Thus,
attempts to isolate pure CpCr(CO)2(IMe)H were fruitless.
Reaction of an orange CD3CN solution of K+(18-crown-
6)[CpCr(CO)2(IMe)]−·1/2THF with the acid [H-DBU]+BF4
−
(pKaMeCN = 24.3, DBU = 1,8-diazabicyclo[5.4.0]undec-7-ene)24
or with NH4 PF6− (pKaMeCN = 16.5)25 resulted in a light yellow
+
1
solution, giving new H NMR resonances at 7.09 (2H), 4.63
(5H), 3.78 (6H), and −5.33 ppm (1H), which we assigned to
the hydride CpCr(CO)2(IMe)H (Scheme 1). Although the
hydride is drawn in Scheme 1 with cis geometry, we cannot
draw any conclusions on the geometry based on the available
data.26 Integration of these resonances against the “internal
standard” 18-crown-6 (24H) showed that the hydride is
1595
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603

Inorganic Chemistry
Article
Scheme 2
The equilibrium constant for the protonation reaction of
pKaMeCN of conjugate acid = 26.9824); the equilibrium was
approached from both directions.
K+(18-crown-6)[CpCr(CO)2(IMe)]−·1/2THF (10 mM [Cr])
Orange crystals of CpCr(CO)2(PCy3)• and light yellow
crystals of cis-CpCr(CO)2(PCy3)H, suitable for X-ray dif-
fraction, were grown from supersaturated MeCN solutions. The
molecular structures are shown in Figure 9, and selected
−
was determined in CD3CN with the acid [HtBuP1(pyrr)]+BF4
(pKa
= 28.4, tBuP1(pyrr) = (tert-butylimino)tris(1-
MeCN
pyrrolidinyl)phosphorane).24,27 1H NMR and IR spectroscopy
suggested that the protonation equilibrium is established
rapidly. Two 1H NMR equilibrium measurements with different
acid concentrations gave the same protonation equilibrium
constant (K = 0.06); thus, the pKaMeCN of the hydride was
determined to be 27.2. The unavailability of clean samples of
CpCr(CO)2(IMe)H prevented us from approaching the
equilibrium from the opposite direction. Therefore, we
conservatively assign a relatively large uncertainty of 0.4 in
the quoted pKaMeCN of CpCr(CO)2(IMe)H.
CpCr(CO)2(PCy3)•, [CpCr(CO)2(PCy3)]−, and CpCr-
(CO)2(PCy3)H. We also synthesized (Scheme 2) an analogous
series of complexes containing the tricyclohexylphosphine
ligand, one of the more electron-donating and bulky28
phosphines. The reaction of [CpCr(CO)3]2 with PCy3 in
MeCN/Et2O at 20 °C proceeded slowly, but cleanly provided
CpCr(CO)2(PCy3)• in about 65% yield as a yellow/orange
crystalline precipitate over the course of several days. The room
temperature EPR spectrum in toluene displays a broad doublet
at giso = 2.035 (Aiso(31P) = 94 MHz), which sharpens upon
cooling. Like the IMe analogue, CpCr(CO)2(PCy3)• exhibits
Curie paramagnetic behavior in solution. (Relevant spectra and
figures can be found in the Supporting Information.)
Electrochemically, CpCr(CO)2(PCy3)• is reversibly reduced
to CpCr(CO)2(PCy3)− at E1/2 = −1.58(1) V vs Cp2Fe+•/0 in
MeCN. Treatment of the radical with KC8 and 18-crown-6
provided K+(18-crown-6)[CpCr(CO)2(PCy3)]− in 90% yield
as a yellow powder. Protonation with NH4 PF6 in MeCN
cleanly provided CpCr(CO)2(PCy3)H. Because of its low
solubility in MeCN, it precipitates from the reaction mixture,
and was isolated as light yellow crystals in 80% yield. In the 1H
NMR spectrum in CD3CN, the signal due to the hydride is a
doublet (2JHP = 80 Hz) at −6.22 ppm; the large H-P coupling
constant suggests that the cis isomer is the predominant
isomer.29 The pKaMeCN of CpCr(CO)2(PCy3)H was deter-
mined to be 26.1(3), using the phosphazene base (tert-
butylimino)tris(dimethylamino)phosphorane) (tBuP1(dma),
Figure 9. Molecular structures of CpCr(CO)2(PCy3)• (top) and
CpCr(CO)2(PCy3)H (50% probability ellipsoids; C-bonded hydro-
gens omitted).
+
−
metrical parameters are provided in Table 3. In cis-CpCr-
(CO)2(PCy3)H, the hydride was located and refined, although
it should be kept in mind that its position is not obtained with
accuracy in an X-ray diffraction experiment.30 Nevertheless, the
presence of an additional ligand (as compared to the radical) is
also clearly revealed by the C1−Cr1−P1 and C2−Cr1−P1
1596
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603

Inorganic Chemistry
Article
reported in 1968.37 Fortier, Macartney, Baird and co-workers
have established that reactions of [CpCr(CO)3]2 with PR3
Table 3. Bond Lengths (Å) and Angles (deg) for
CpCr(CO)2(PCy3)• and CpCr(CO)2(PCy3)H
involve the monomer CpCr(CO)3 (which is in equilibrium38
•
CpCr(CO)2(PCy3)•
CpCr(CO)2(PCy3)H
with the dimer), and that the substitution proceeds by an
associative mechanism.34 We assume that this mechanism is
also operative in the synthesis of the radicals reported here. On
the basis of the timescales during which the reactions take place
(seconds for IMe, days for PCy3), we conclude that IMe is the
superior nucleophile. IMe is also the better electron donor, as
Cr1−Cpcentroid
Cr1−P1
Cr1−C1, Cr1−C2
C1−O1, C2−O2
Cr1−H1
1.854(1)
1.852(1)
2.3760(6)
1.827(2), 1.838(2)
1.158(3), 1.159(3)
NA
2.3315(4)
1.823(2), 1.819(2)
1.163(2), 1.165(2)
1.53(2)30
judged by νCO
values that are ≈ 10 cm−1 lower for
̃
C1−Cr1−C2
C1−Cr1−P1
C2−Cr1−P1
79.64(9)
82.85(7)
CpCr(CO)2(IMe)• than for CpCr(CO)2(PCy3)• (Table 1,
entries 1 and 2).
94.40(7)
84.62(5)
97.67(7)
107.77(5)
The monomeric nature of both radicals was expected, and
unequivocally established by the crystal structure analyses. For
CpCr(CO)2(IMe)•, the four statistically identical Cr−C(carbene)
bond lengths of 2.05 Å are at the short end of the range of
2.04−2.25 Å found for other crystallographically character-
ized39 N-heterocyclic carbene complexes of chromium.
However, they are still > 0.2 Å longer than the Cr−C(CO)
bonds (1.80−1.82 Å), and can be regarded as Cr−C single
bonds, as would be expected for the singlet carbenes.
Compared to the carbene complex, the phosphorus atom in
CpCr(CO)2(PCy3)• is placed at a significantly larger distance
from chromium (d = 2.3760(6) Å): this Cr−P distance is also
0.031(1) Å longer than in CpCr(CO)2(PPh3)•.35 In both
structures, key features are the small (76−80°) CCO−Cr−CCO
angles in comparison to much larger (94−101°) C(carbene)−Cr−
C(CO) or P−Cr−C(CO) angles. Such an irregular placement of
the three legs of the piano-stool has precedent in related
phosphine-ligated radicals,35,36,40 and shows that the effectively
angles, which become significantly different in going from the
radical to the hydride.
Thermochemistry: Cr−H BDFE and Thermodynamics
of H2 Loss. Equations 3 and 4 demonstrate how the M-H
homolytic bond dissociation free energy (BDFE) and the BDE
(enthalpy) can be derived from the thermodynamic acidities of
hydrides and the formal potentials of M•/− electrochemical
couples in acetonitrile solvent.31,32 Because the M•/− wave is
fully reversible (Figure 5), we assume that the formal potential
(E°) is the same as the half-wave potential (E1/2). The results
for this thermochemical analysis for CpCr(CO)2(L)H are
detailed in Scheme 3. Additionally, using the homolytic bond
dissocation free energy of H2 (103.6 kcal mol−1),31 the
thermodynamic stability of the hydrides with respect to loss
of dihydrogen, in the overall reaction 2 Cr−H → 2 Cr• + H2,
was evaluated.33
2
Cs-symmetric radicals have A″ electronic ground states. The
·
BDFE = ΔGH (MH)
orientation of the IMe and PCy3 ligands is such that steric
repulsion with the CpCr(CO)2 fragment is minimized, which is
accomplished in different ways because of the differing
rotational symmetries of these ligands (2- and 3-fold,
respectively). Thus, in CpCr(CO)2(IMe)• the approximate
mirror plane bisects the carbene ligand, an orientation that we
previously found10 by DFT calculations for the heavier
congener CpW(CO)2(IMe)•. In CpCr(CO)2(PCy3)•, the
pseudo-3-fold CpCr(CO)2 fragment is staggered with respect
to the PCy3 ligand.
= 1.37 pKa(MH) + 23.06 E°(M•/−) + 53.6
(3)
·
BDE = ΔHH(MH)
= 1.37 pKa(MH) + 23.06 E°(M•/−) + 59.5
(4)
DISCUSSION
■
17-Electron Chromium Complexes. Whereas CpCr-
(CO)2(PCy3)• is a bulky addition to a family of well-
characterized34−36 complexes CpCr(CO)2(PR3)•, carbene
analogues CpCr(CO)2(NHC)• were heretofore not known.
This is perhaps surprising, considering that isoelectronic (yet
cationic) complexes Cr(CO)4(NHC)2+• were reported in 1983
The EPR spectra of the radicals are in most aspects similar to
the spectra for CpCr(CO)2(PPh3)• and related low-spin, d5
piano-stool complexes;36,41 frozen solution spectra are rhombic,
and one of the principal g values is smaller than the free-
electron value (2.0023). The radicals are largely Cr-centered as
expected, but for CpCr(CO)2(IMe)• it is further corroborated
̈
by Hofmann, Ofele and co-workers, and considering that
Cr(CO)5(IMe), the first transition metal NHC complex, was
a
Scheme 3. Thermochemical Data for CpCr(CO)2(L)H
a
MeCN solvent, T = 293 K.
1597
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603

Inorganic Chemistry
Article
by the observation of hyperfine coupling to 53Cr. The g-
anisotropy (gmax − gmin) is significantly smaller for the carbene
complex (0.073) than for the phosphine-ligated CpCr-
(CO)2(PCy3)• (0.096) and CpCr(CO)2(PPh3)• (0.113).36 It
is noteworthy that room temperature EPR signals for
CpCr(CO)2(IMe)• and CpCr(CO)2(PCy3)• are so easily
observed, considering that efficient electron spin−lattice
relaxation can cause extreme broadening (often to the point
of unobservability) of room temperature signals in related Cr
piano-stool radicals. We think that the NIR spectra provide at
least a qualitative explanation. Atwood and Geiger reported17
that 17-electron piano-stool radicals have a weak (ε ≈ 100 M−1
cm−1) electronic transition in the near- to mid-IR region, which
anion; similar observations were made for CpW-
(CO)2(IMes)−.10
As expected, the crystal structures of both K+[CpCr-
(CO)2(IMe)]− salts, with and without the crown ether, show
contact ion pairing through K···O(CO) interactions. Because
both structures contain more than one (two and four,
respectively) crystallographically independent ion pairs in the
asymmetric unit, the relative structural flexibility of these
interactions became apparent. In K+(18-crown-6)[CpCr-
(CO)2(IMe)]−·1/2THF, for example, the potassium in one
ion pair is pinched by two Cr-bound carbonyls, while in the
other it is ligated by only one Cr-bound carbonyl. Precedent for
these respective situations is found in the structures of K+(18-
10
2
is probably best thought of as an interconversion between A″
crown-6)[CpW(CO)2(IMes)]−
and K+(18-crown-6)-
and A′ electronic states. Whereas λmax for this transition is at
[(C5H4CH2CH2PPh2)M(CO)3]− (M = Cr, Mo, W).44 The
potassium cation accommodates this reduction in the number
of bonded donors by binding more strongly to the remaining
donors (see Table 2). The pinching vs nonpinching
interactions are probably the cause of the significantly different
CCO−Cr−CCO angles.
2
17
•
2500 nm for (C5Ph5)Cr(CO)3 , tailing into the mid-IR, we
find it at 1455 nm for CpCr(CO)2(IMe)• and at 1660 nm for
CpCr(CO)2(PCy3)• (Figure 2), that is, at considerably higher
energy. Therefore, it is possible that the interconversion
2
2
between A″ and A′ electronic states, and hence the electron
spin−lattice relaxation efficiency, is drastically reduced in the
radicals reported here.
The structure of crown-free K+[CpCr(CO)2(IMe)]−·3/4THF
shows that the anion-based ligands (largely the carbonyls)
alone almost succeed in filling the coordination spheres of the
potassium cationsjust one in twenty K···O interactions is due
to a THF ligand, and two-thirds of the THF molecules present
are lattice solvents that do not interact with potassium at all. In
The Curie paramagnetic behavior that we find by variable
temperature 1H NMR spectroscopy for CpCr(CO)2(IMe)• and
CpCr(CO)2(PCy3)• was previously also found for CpCr-
(CO)3• and (C5Me5)Cr(CO)3• in an elegant study by Wayland
and co-workers.42 The latter two radicals are involved in
temperature-dependent equilibria with the diamagnetic dimers,
which complicates the analysis, although the authors success-
fully accounted for the dimerization. The strictly monomeric
nature of our radicals makes the analyses straightforward;
furthermore, they provide information on multiple types of
fact, there are crystal structures of K+ salts of related M(0)
45
anions (for example, K+[CpFe(CO)2]−
and K+[CpMo-
(CO)3]− 46) that are entirely free of additional Lewis bases. On
average, the K···O(CO) interactions are stronger in the crown-
free (d(K···O(CO)
) = 2.77 Å) than in the crown-containing
av
(d(K···O(CO))av = 2.94 Å) material. This is probably related to
the differing number of O ligands surrounding K+ in both
structures: seven and eight in the latter, and a maximum of six
in the former. Further, the Cr−C and C−O bond lengths
(Table 2) may suggest that there is more metal-to-carbonyl
backbonding in the crown-free (d(Cr−C)av = 1.773 Å, d(C-
1
protons instead of one. The hyperfine shifts in the H NMR
spectra appear to be predominantly caused by Fermi contact
coupling of the nuclei with the electron, dipolar (pseudocon-
tact) interactions being less important. For example, were the
latter dominant, we would have expected for CpCr-
(CO)2(IMe)• that the carbene CH3 signal would be shifted
to a greater extent than the carbene CH signal, because of the
different Cr···H separations (d(Cr···H3C) = 3.1−4.5 Å;
d(Cr···HC) = 5.1 Å); however, the opposite is observed
(Figure 3).
O)av = 1.207 Å) than in the crown-containing (d(Cr−C)av
=
1.778 Å, d(C-O)av = 1.200 Å) structure, in line with the
expectation47 that stronger K···O(CO) interactions should
encourage backbonding. However, this point has to be made
with caution: although the values are averages of four or eight
individual bond lengths, it is hard to establish how significant
the differences are.
Electron-Rich Anions CpCr(CO)2(IMe)− and CpCr-
(CO)2(PCy3)−. The difference in electron-richness of the IMe
and PCy3 complexes, already substantial in the neutral radicals,
seems to be even larger in the anions. Comparison between
The IR data give qualitative information about the nature of
the K+[CpCr(CO)2(IMe)]− salts in solution. We make the
assumption that K+(18-crown-6)[CpCr(CO)2(IMe)]−·1/2THF
exists largely as solvent-separated ion pairs in MeCN, a polar
and high-dielectric (εr = 37.5) solvent. In THF (εr = 7.5), the
low-energy band appears at lower energy than in MeCN, a
behavior that contrasts that of the neutral species CpCr-
(CO)2(IMe)• and CpCr(CO)2(IMe)H, in which the bands
shift to higher energy upon going from MeCN to THF (Table
1). As Darensbourg has shown that the interaction of a
carbonyl oxygen with an alkali metal cation will decrease its
entries 4 and 5 of Table 1 shows that νCO values differ by about
̃
25 cm−1 in the anions, where the difference is about 10 cm−1 in
the radicals. Also, while CpCr(CO)2(PCy3)− is quite reducing
(E° = −1.58(1) V vs Cp2Fe+•/0, 70 mV more negative than the
PEt3 derivative43), it is easily surpassed by CpCr(CO)2(IMe)−
(−1.89(1) V vs Cp2Fe+•/0). The latter redox potential is also
about 0.35 V more negative than that of the tungsten derivative
CpW(CO)2(IMe)•/− (−1.54(2) V vs Cp2Fe+•/0);12 differences
of about 0.3 V between Cr and Mo/W congeners are
13
documented for the parent anions CpM(CO)3
.
̃
νCO,
47 we interpret our observation as a sign that K+(18-crown-
•/−
The highly reducing nature of the anions is also the
(indirect) cause of their broadened NMR spectra. Dissolution
of the salts in MeCN adventitiously generates small amounts
(ca. 1−2%, observed in IR spectrum) of the radical, with which
the anion is in rapid electron-transfer exchange. The hypothesis
is confirmed by the observed sharpening of the signals when
KC8 is added, reducing any radical that is present back to the
6)[CpCr(CO)2(IMe)]−·1/2THF remains somewhat ion-paired
in solution. Crown-free K+[CpCr(CO)2(IMe)]− may be
undergoing some ion-pairing even in MeCN, as its νCO bands
̃
are somewhat broader (yet hardly shifted) than in the crown-
containing derivative. In THF, both bands are now shifted to
lower energy, which suggests that the absence of the crown
ether causes both carbonyls to interact with K+ in THF. In light
1598
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603

Inorganic Chemistry
Article
of the polymeric nature of crown-free K+[CpCr(CO)2(IMe)]−
in the solid state, as determined crystallographically, it is
possible that ion pair aggregation occurs in solution.
in CpCr(CO)2(IMe)H is an important factor in causing its very
low Cr−H BDFE and its comparatively high acidity. Although
PCy3 is a bulky ligand, its 3-fold rotational symmetry allows it
to gear quite effectively with the CpCr(CO)2H fragment
(Figure 9); such gearing is less optimal for the IMe ligand with
its 2-fold rotational symmetry. For example, we characterized
the stable cis-CpW(CO)2(IMe)H by X-ray diffraction, and
found that one N-methyl is rather close to the hydride
(d(WH···CH3) = 2.43(4) Å).12 An increased crowding can be
expected for the chromium analogue, because M−C(carbene), M−
Hydrides with Weak Cr−H Bonds. The Cr−H bond
properties of the hydrides described herein can be compared
with those of the parent CpCr(CO)3H, since its Cr−H BDE
has been reliably determined to be 61−62 kcal mol−1 by
calorimetry.14 This bond strength is also obtained by
thermochemical analysis: its pKaMeCN (13.3)48 and E° for the
13
•/−
couple CpCr(CO)3
(−0.688 V vs Cp2Fe+•/0
)
provide,
using Equations 3 and 4, a BDFE value of 56.0 kcal mol−1 and a
BDE value of 61.9 kcal mol−1. Hoff and co-workers found that
the Cr−H BDEs of several phosphine/phosphite derivatives
CpCr(CO)2(PR3)H (R = Ph, 59.8; Et, 59.9; OMe, 62.7 kcal
mol−1) are not very different from that of CpCr(CO)3H.14 Our
determined Cr−H BDE for CpCr(CO)2(IMe)H (53.2(6) kcal
mol−1) is not only significantly lower than that of the
aforementioned hydrides, it is even lower than the 55−58
kcal mol−1 that Norton and co-workers9 found for a series of
vanadium hydrides HV(CO)4(Ph2P(CH2)nPPh2) (n = 1−4).
Thus, the Cr−H bond in CpCr(CO)2(IMe)H appears to be
the weakest M-H bond observed among organometallic
hydrides. On the other hand, the Cr−H bond in CpCr-
(CO)2(PCy3)H is only about 3 kcal mol−1 weaker than the
Cr−H bond in CpCr(CO)3H, and about 1 kcal mol−1 weaker
than that in CpCr(CO)2(PPh3)H.
C
(Cp), and M−C(CO) distances are about 0.15 Å shorter for Cr
than for W. Although we have not been able to address the
issue of cis/trans isomerism in CpCr(CO)2(IMe)H, either
isomer can be expected to have close interactions of an N-
methyl group with another basal ligand, as illustrated in Figure
10.52
Figure 10. Steric interactions of N-CH3 groups with other basal
ligands in CpCr(CO)2(IMe)H.
Inspection of Scheme 3 reveals that the 5.6(7) kcal mol−1
higher BDFE for CpCr(CO)2(PCy3)H as compared to
CpCr(CO)2(IMe)H is mainly caused by the large difference
in formal potentials of the [Cr]•/− couples. The pKaMeCN of the
hydrides (as expected, both higher than the 21.8 determined for
CpCr(CO)2(PPh3)H)49 differ by only 1.1(5) units, and are
remarkably similar considering our recent observation that
substitution of PMe3 for either IMe or IMes (1,3-
dimesitylimidazol-2-ylidene) increases the pKa of related
tungsten hydrides by about 5 units.11,12 Admittedly, this is
not a direct comparison, and we should assume that
substitution of PMe3 for PCy3 itself will somewhat lower the
thermodynamic acidity of metal hydrides, taking into account
that PCy3 is a better donor28 than PMe3.
The experimentally observed (in)stabilities of our hydrides
are in agreement with the thermochemical analysis (Scheme 3).
Thus, CpCr(CO)2(IMe)H should not be stable with respect to
formation of the 17-electron radical and H2. Indeed, we can
only characterize CpCr(CO)2(IMe)H in solution. We find that
it slowly decomposes to form the radical, and attempts to
isolate the hydride have been unsuccessful. Since the same
result is obtained when the sample is shielded from light, we
conclude that the decomposition is a thermal process. On the
other hand, CpCr(CO)2(PCy3)H was relatively easily charac-
terized and obtained in analytically pure form. The non-
observation of CpCr(CO)2(IMe)H in the reaction of CpCr-
(CO)2(IMe)• with excess H2 (1000 psi) in principle also agrees
with the thermochemical analysis. However, we do not know
the kinetics of the hydrogenation reaction, and that experiment
on its own would have been inconclusive with regard to the
thermochemistry.
CONCLUSIONS
■
A full synthetic and spectroscopic and characterization of
chromium piano-stool chromium hydrides CpCr(CO)2(L)H
(L = PCy3, IMe), as well as derived anions and 17-electron
radicals, has been reported. Thermochemical analyses correctly
predict the stability of these hydrides with respect to H atom
loss in the form of dihydrogen. CpCr(CO)2(IMe)H is unstable
in this regard, because its Cr−H BDFE is only 47.3(6) kcal
mol−1, nearly 6 kcal mol−1 lower than that of CpCr-
(CO)2(PCy3)H. The comparison suggests the intriguing
possibility that while the free ligand IMe is much smaller and
more nucleophilic than PCy3, the carbene exerts a greater steric
pressure on the other ligands than the phosphine does, once
incorporated in the coordination sphere.
EXPERIMENTAL SECTION
■
General Procedures. All manipulations were carried out under N2
using standard vacuum line, Schlenk, and inert-atmosphere glovebox
techniques. Acetonitrile, diethyl ether, hexanes, fluorobenzene,
tetrahydrofuran, and toluene were purified by passage through neutral
alumina, using an Innovative Technology, Inc., Pure Solv solvent
purification system. Hexafluorobenzene (>99%, Aldrich) was stirred
over P2O5 and vacuum transferred. Deuterated solvents (Cambridge
Isotope Laboratories, 99.5% D or greater) were dried as follows:
toluene-d8 was vacuum transferred from sodium-potassium alloy;
CD3CN was stirred over P2O5 and then vacuum distilled through a
glass wool plug. Tetrabutylammonium hexafluorophosphate (anhy-
drous, >99%, Fluka), ammonium hexafluorophosphate (>98%, Fluka),
t
t
DBU (>99%, Fluka), Schwesinger bases BuP1(pyrr) and BuP1(dma)
(Fluka), ferrocene (Aldrich), and cobaltocene (Strem) were used as
received. Potassium hydride was obtained as a 30 wt % suspension in
mineral oil; in the glovebox, the mineral oil was washed away with
hexanes, and the KH was dried under vacuum. KC8 was obtained by
heating potassium and graphite (1:8 mol ratio) to 200 °C. Acids [H-
Why is the Cr−H bond in CpCr(CO)2(IMe)H so weak?
Sirsch, McGrady and co-workers have demonstrated that even
the hydride CpCr(CO)3H has a sterically crowded basal ligand
set,50 while Skagestad and Tilset have noted that extremely
bulky ligands may increase thermodynamic acidities of
hydrides.51 Therefore, we think that extreme steric crowding
−
−
DBU]+BF4 , [HtBuP1(dma)]+OTf−, and [HtBuP1(pyrr)]+BF4 were
prepared by protonation of the conjugate bases with ethereal HBF4 (or
1599
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603

Inorganic Chemistry
Article
HOTf) in Et2O. The resulting white precipitates were purified by
precipitation from THF or MeCN by addition of Et2O, and washed
with Et2O. Tricyclohexylphosphine (Strem) was sublimed at 90 °C at
<1 mTorr onto a water-cooled finger. [CpCr(CO)3]2,53 and 1,3-
dimethylimidazolium iodide54 were prepared as described in the
literature. Elemental analyses were performed by Atlantic Microlab
(Norcross, GA). Low %C analyses were obtained for the highly air-
sensitive and easily oxidizable anionic Cr0 complexes, an issue we also
encountered for related tungsten complexes.12 Uncertainties in the
measured pKa and E° values, as well as in the derived thermochemical
energy data, are provided at the 2σ confidence interval.
eventually resulted in solidification, to give an orange-brown powder
suspended in a yellow solution. After cooling to −35 °C overnight, the
powder was filtered, washed with hexane (2 × 5 mL), and dried under
vacuum. Sublimation at 100−110 °C at <1 mTorr onto a water-cooled
finger gave the analytically pure product as an orange/brown powder.
1
Yield: 0.277 g (1.03 mmol, 79%). H NMR (CD3CN, 500 MHz, 295
K): δ ≈ 31.1 (br, fwhm ≈ 4500 Hz, 2H, =CH), 13.1 (br, fwhm = 690
Hz, 5H, Cp), 4.9 (br, fwhm = 210 Hz, 6H, N-CH3). 1H NMR
(toluene-d8, 500 MHz, 295 K): δ ≈ 30.7 (br, fwhm ≈ 4500 Hz, 2H,
=CH), 12.9 (br, fwhm = 690 Hz, 5H, Cp), 3.1 (br, fwhm = 240 Hz,
6H, N-CH3). Anal. Calcd. for C12H13N2O2Cr: C, 53.53; H, 4.87; N,
10.40. Found: C, 53.33; H, 4.92; N, 10.53. Single crystals of
CpCr(CO)2(IMe)• were grown by diffusion of hexane into a
fluorobenzene solution of it at room temperature. An oily precipitate
was initially obtained, from which orange rods grew over the course of
several days.
Instrumentation. Electrochemical measurements were performed
using a CH Instruments potentiostat equipped with a standard three-
electrode cell consisting of a 4 mL disposable glass vial fitted with a
polyethylene cap having openings sized to closely accept each
electrode. For each experiment, the cell was assembled and used
within the glovebox, with electrodes connected to the potentiostat via
RF-shielded cables fed through the glovebox wall. The working
electrode (1 mm PEEK-encased glassy carbon, Cypress Systems
EE040) was polished using alumina (BAS CF-1050, dried at 150 °C
under vacuum) suspended in acetonitrile, and then rinsed with neat
acetonitrile. A glassy carbon rod (Structure Probe, Inc.) was used as
the counterelectrode, and a silver wire suspended in a solution of 0.1
M nBu4N+PF6− in acetonitrile and separated from the analyte solution
by a porous Teflon tip (CH Instruments 112) was used as the
pseudoreference electrode. Potentials are reported vs the Cp2Fe+•/0
couple, and were determined versus cobaltocene (E° = −1.33 V vs
Cp2Fe+•/0). NMR experiments were carried out using Varian 300 or
Synthesis of CpCr(CO)2(PCy3)•. A solution of PCy3 (0.170 g,
0.606 mmol) in Et2O (1 mL) was added to a dark green suspension of
[CpCr(CO)3]2 (0.110 g, 0.273 mmol, corresponding to 0.547 mmol
•
monomeric CpCr(CO)3 ) in MeCN (3 mL). The suspension was
stirred at room temperature for two days to afford the product as a
yellow/orange precipitate, which was washed with MeCN (3 × 3 mL)
and dried under vacuum to give a yellow/orange powder. Yield: 0.160
1
g (0.353 mmol, 65%). H NMR (toluene-d8, 500 MHz, 295 K): δ ≈
12.4 (br, fwhm = 670 Hz, 5H, Cp), 5.7 (br, fwhm = 225 Hz, 6H, Cy
CH2), 4.4 (br, fwhm ≈ 350 Hz, 6H, Cy CH2), 3.0 (br, 12H, CH2),
1.63 (br, fwhm = 30 Hz, 3H, Cy CH2), 0.86 (br, fwhm = 55 Hz, 3H,
Cy CH2), −7.6 (br, fwhm ≈ 1500 Hz, 3H, Cy CH). No resonance
observed in 31P{1H} NMR spectrum, and 13C NMR spectrum was not
recorded. A crystalline sample of higher analytical purity was obtained
by dissolving 20 mg of the material in 0.6 mL of boiling MeCN in an
NMR tube. The solution was left to stand at room temperature for
several hours, affording about 15 mg of orange crystals, some of which
were suitable for X-ray diffraction. Anal. Calcd. for C25H38O2PCr: C,
66.21; H, 8.45. Found: C, 65.52; H, 8.22 (powder). Found: C, 65.91;
H, 8.48 (crystals).
1
500 MHz spectrometers. H and 13C NMR spectra were referenced
1
relative to NMR solvent (protic residual for H) peaks. 31P NMR
spectra were referenced with respect to a referenced 1H NMR
spectrum with the use of Varian’s mref command. EPR spectra were
recorded at or below 1 mM (making sure that further dilution did not
result in a change in signal shape) in liquid or frozen toluene solutions,
using a Bruker Elexsys X-band EPR spectrometer equipped with a
helium-cooled cryostat; g values were derived from the field/frequency
ratios, and simulations were performed with EasySpin.55 Solution IR
spectra (data are provided in Table 1) were recorded using a Nicolet
iS10 FTIR spectrometer with demountable sealed liquid CaF2 cells
(International Crystal Laboratories). Solid-state IR spectra were
measured as nujol mulls between CaF2 plates. Vis-NIR spectra were
recorded on a Agilent Cary 5000 UV−vis-NIR spectrometer, using
quartz cuvettes with 10 mm path length.
Synthesis of K+(18-crown-6)[CpCr(CO)2(IMe)]−·0.5 THF.
CpCr(CO)2(IMe)• (0.071 g, 0.26 mmol) and an excess of KC8
(0.072 g, ≈ 0.5 mmol K) were stirred in 2 mL of THF, giving an
orange solution (containing suspended graphite/KC8 particles). After
30 min, the solution was filtered and added to a solution of 18-crown-6
(0.086 g, 0.33 mmol) in 1 mL of THF. Hexane (2 mL) was added,
resulting in a slight cloudiness of the orange solution. The inside of the
vial was scratched with a spatula, immediately inducing the
precipitation of the title compound as thin, orange plates. The crystals
were washed with Et2O (3 × 2 mL) and dried under vacuum. Yield:
Synthesis of CpCr(CO)2(IMe)•. 1,3-Dimethylimidazolium iodide
(0.426 g, 1.90 mmol) and KH (0.206 g, 5.1 mmol) were placed in a 20
mL vial, and the suspension was stirred in THF (3 mL). Slow
evolution of a gas (presumably H2) was observed. After 4 h,
effervescene had stopped, and Et2O (10 mL) was added to ensure
nearly complete precipitation of the KI. The solution of the generated
1,3-dimethylimidazol-2-ylidene was filtered and added, in portions of
about 0.5 mL, to a dark green suspension of [CpCr(CO)3]2 (0.261 g,
1
0.140 g (0.23 mmol, 87%). H NMR (CD3CN, 500 MHz, 293 K): δ
6.84 (s, 2H, =CH), 4.00 (s, 5H, Cp), 3.84 (s, 6H, N-CH3), 3.64 (m,
2H, THF), 3.57 (s, 24H, 18-crown-6), 1.80 (m, 2H, THF). 13C{1H}
NMR (CD3CN, 125 MHz, 293 K): δ 254.5 (s, Cr-CO), 223.3 (s, Cr-
CN2), 121.1 (s, =CH), 81.6 (s, Cp), 70.8 (s, 18-crown-6), 68.2 (s,
THF), 39.2 (s, N-CH3), 26.2 (s, THF). Anal. Calcd. for
C26H41CrKN2O8.5: C, 51.30; H, 6.79; N, 4.60. Found: C, 49.78; H,
6.78; N, 4.54. Single crystals of K+(18-crown-6)[CpCr-
(CO)2(IMe)]−·1/2THF were grown by dissolving 12 mg of the
material in 0.5 mL of boiling THF in an NMR tube. The solution was
left to stand at room temperature for several hours, affording orange
blocks.
•
0.649 mmol, corresponding to 1.30 mmol monomeric CpCr(CO)3 )
in Et2O (2 mL). Addition of portions of the IMe solution each time
resulted in effervescence and a temporary color change from green to
orange/yellow; the green color then reappeared because of dissolution
of remaining starting material. When the green color did not reappear
(a sign that the starting material was consumed), the addition of the
IMe solution was stopped. The resulting solution, containing an
unidentified fluffy precipitate, was filtered into 3 mL of hexane,
affording an orange, cloudy solution. Under reduced pressure, the
volume was reduced to about 5 mL. (The crude solution of
CpCr(CO)2(IMe)• should not be evaporated to dryness; the oily
residue suddenly turns black and becomes intractable. The extent and
cause of this apparent decomposition was not investigated further, but
it is possible that excess free IMe is involved. Once CpCr(CO)2(IMe)•
has solidified, it is thermally stable and readily handled.) Hexane (5
mL) was added, and the volume was again reduced to about 5 mL.
This procedure of hexane addition and concentration was repeated
twice to remove most of the Et2O and THF. Most of the product had
precipitated as an orange, viscous liquid. Vigorous stirring and scraping
Synthesis of K+[CpCr(CO)2(IMe)]− (crown-free). A solution of
K+[CpCr(CO)2(IMe)]− was generated as described above. Addition
of several volumes of hexane resulted in the precipitation of a yellow/
orange powder. The powder was washed with Et2O (3 × 4 mL) and
dried, and K+[CpCr(CO)2(IMe)]− could be isolated in about 90%
yield. The powder is very fluffy and is not convenient to handle. It was
analyzed by IR spectroscopy, but not by elemental analysis. Single
crystals of K+[CpCr(CO)2(IMe)]−·3/4THF were grown by dissolving
20 mg of the material in 0.5 mL of hot THF in an NMR tube (not all
material dissolved). The solution was allowed to slowly concentrate
over several weeks in the glovebox, affording orange blocks.
1600
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603

Inorganic Chemistry
Article
Table 4. Crystallographic Data
complex
CpCr(CO)2(IMe)• K+(18-crown-6)[CpCr(CO)2(IMe)]−·
K+[CpCr(CO)2(IMe)]−·
CpCr(CO)2(PCy3)•
C25H38O2PCr
453.52
CpCr(CO)2(PCy3)H
C25H39O2PCr
454.53
1/2THF
3/4THF
empirical
formula
C12H13N2O2Cr
269.24
C26H41KN2O8.5Cr
C15H19KN2O2.75Cr
362.42
formula weight
608.71
(g mol−1
)
space group
P1 (No. 2)
Pca21 (No. 29)
a: 16.4326(5)
P21 (No. 4)
a: 7.5173(2)
P21/n (No. 14)
a: 14.5523(6)
P1 (No. 2)
̅
̅
unit cell lengths a: 11.4850(5)
(Å)
a: 9.7528(5)
b: 14.6464(7)
c: 15.3720(7)
b: 22.3721(6)
c: 16.2384(5)
α: 90
b: 25.0725(6)
c: 17.4342(4)
α: 90
b: 10.0674(5)
c: 15.7122(6)
α: 90
b: 11.2597(7)
c: 11.5526(7)
α: 74.059(3)
unit cell angles
(deg)
α: 70.022(2)
β: 87.357(2)
γ: 87.104(2)
2425.93(19)
8 (4)
β: 90
β: 98.219(2)
γ: 90
β: 94.732(3)
γ: 90
β: 73.138(3)
γ: 77.070(2)
1153.00(12)
2 (1)
γ: 90
volume (Å3)
Z (Z′)
5969.8(3)
8 (2)
3252.20(14)
8 (4)
2294.05(17)
4 (1)
calcd. density
1.474
1.355
1.480
1.313
1.309
(g cm−3
)
μ (Mo Kα,
mm−1
0.933
0.572
0.970
0.587
0.584
)
a
b
R1 (I > 2σ(I))
0.0869
0.2426
0.0340
0.0791
0.0389
0.0949
0.0460
0.1213
0.0387
0.1253
wR2 (all data)
a
b
R1 = (∑||Fo| − |Fc||)/∑|Fo|. wR2 = [∑w(F2o − F2c)2/∑w(Fo2)2]1/2
.
Synthesis of K+(18-crown-6)[CpCr(CO)2(PCy3)]−. CpCr-
Synthesis of CpCr(CO)2(PCy3)H. A solution of K+(18-crown-
6)[CpCr(CO)2(PCy3)]− (0.101 g, 0.133 mmol) in MeCN (2 mL)
(CO)2(PCy3)• (0.101 g, 0.223 mmol) was dissolved in THF (2
mL), and an excess of KC8 (0.050 g, ≈ 0.4 mmol K) was added. The
suspension was stirred for 30 min, filtered, and added to a solution of
18-crown-6 (0.070 g, 0.26 mmol) in 0.5 mL of THF. Because some of
the K+[CpCr(CO)2(PCy3)]− had precipitated, the black filter cake was
washed with MeCN (ca. 5 mL total) until the initially yellow washings
were nearly colorless. The resulting yellow MeCN/THF solution was
evaporated to dryness, leaving a yellow powder, which was washed
with hexane (4 × 5 mL) and dried under vacuum. Yield: 0.150 g
(0.198 mmol, 89%). 1H NMR (CD3CN, 500 MHz, 298 K): δ 4.06 (d,
3JHP = 1.4 Hz, 5H, Cp), 3.58 (s, 24H, 18-crown-6), 2.02 (app d, 6H,
Cy CH2), 1.72 (m, 6H, Cy CH2), 1.67 (m, 3H, Cy CH, overlapped),
1.64 (m, 3H, Cy CH2, overlapped), 1.35 (app q, 6H, Cy CH2), 1.19
(m, 6H, Cy CH2, overlapped), 1.18 (m, 3H, Cy CH2, overlapped).
13C{1H} NMR (CD3CN, 125 MHz, 298 K): δ 256.2 (d, 2JCP = 18 Hz,
+
was added to a solution of NH4 PF6− (0.027 g, 0.17 mmol) in MeCN
(1 mL). The color changed from orange to light yellow, and the
product started to precipitate as a yellow crystalline material soon after
combination of the reagents. The mixture was kept at −35 °C for 2h,
after which the crystals were collected, washed with cold MeCN (4 × 2
mL) and dried under vacuum. Yield: 0.049 g (0.11 mmol, ≈ 80%). 1H
NMR (CD3CN, 500 MHz, 298 K): δ 4.73 (d, 3JHP = 0.8 Hz, 5H, Cp),
1.91 (m, 6H, Cy CH2), 1.88 (m, 3H, Cy CH, overlapped), 1.82 (m,
6H, Cy CH2), 1.69 (m, 3H, Cy CH2), 1.40−1.19 (several m, 15H, Cy
2
CH2), −6.22 (d, JHP = 80 Hz, 1H, Cr-H). 13C{1H} NMR (CD3CN,
125 MHz, 298 K): δ 86.1 (s, Cp), 40.7 (d, 1JCP = 18 Hz, Cy CH), 30.9
(d, JCP = 1.5 Hz, Cy CH2), 28.5 (d, JCP = 10 Hz, Cy CH2), 27.2 (d, 4JCP
= 1 Hz, Cy CH2). Because of low solubility, carbonyl carbons were not
observed, even with 10,000 transients. 31P{1H} NMR (CD3CN, 202
MHz, 298 K): δ 91.2 (s). Anal. Calcd. for C25H39O2PCr: C, 66.06; H,
8.65. Found: C, 65.78; H, 8.47. Single crystals of CpCr(CO)2(PCy3)H
were grown by dissolving 9 mg of the material in 0.5 mL of boiling
MeCN in an NMR tube. The solution was kept at 25 °C overnight,
affording light yellow blocks.
1
Cr-CO), 80.6 (s, Cp), 70.9 (s, 18-crown-6), 41.5 (d, JCP = 9 Hz, Cy
CH), 31.2 (d, JCP = 1 Hz, Cy CH2), 29.1 (d, JCP = 9 Hz, Cy CH2), 28.0
(d, 4JCP = 1 Hz, Cy CH2). 31P{1H} NMR (CD3CN, 202 MHz, 298 K):
δ 104.4 (s). Anal. Calcd. for C37H62KO8PCr: C, 58.71; H, 8.26. Found:
C, 55.97; H, 7.96.
Crystallography. For all reported structures, a 10× microscope
was used to identify suitable crystals of the same habit. Each crystal
was coated in Paratone, affixed to a Nylon loop, and placed under
streaming nitrogen (100 K for the IMe complexes, 145 K for the PCy3
complexes) in a Bruker KAPPA APEX II CCD diffractometer with
0.71073 Å Mo Kα radiation. The space groups were determined on
the basis of systematic absences and intensity statistics. The structures
were solved by direct methods and refined by full-matrix least-squares
on F2. Anisotropic displacement parameters were determined for all
nonhydrogen atoms. Hydrogen atoms were placed at idealized
positions and refined with fixed isotropic displacement parameters,
with the exception of the Cr-bonded hydrogen in CpCr(CO)2(PCy3)-
H, which was isotropically refined.
Generation and Spectroscopic Characterization of CpCr-
(CO)2(IMe)H. An orange solution of K+(18-crown-6)[CpCr-
(CO)2(IMe)]−·0.5 THF (16 mg, 26 μmol) in 0.3 mL of CD3CN
−
was combined with a colorless solution of [H-DBU]+BF4 (9 mg, 38
μmol, about 1.5 equiv) in 0.3 mL of CD3CN, affording a light yellow
solution. 1H NMR (CD3CN, 500 MHz, 293 K): δ 7.09 (s, 2H, =CH),
4.63 (s, 5H, Cp), 3.78 (s, 6H, N-CH3), −5.33 (s, 1H, Cr-H). 13C{1H}
NMR (CD3CN, 125 MHz, 293 K): δ 244.0 (s, Cr-CO), 198.8 (s, Cr-
CN2), 124.4 (s, =CH), 86.9 (s, Cp), 40.0 (br, N-CH3).
Attempted Synthesis of CpCr(CO)2(IMe)H. K+(18-crown-6)-
+
[CpCr(CO)2(IMe)]−·0.5 THF (51 mg, 84 μmol) and NH4 PF6− (22
mg, 0.13 mmol) were suspended in a mixture of Et2O (2.5 mL) and
THF (0.5 mL). The mixture was stirred for about 15 min, until the
orange crystalline starting material had reacted away, and a yellow
solution of CpCr(CO)2(IMe)H (containing a white precipitate) was
obtained. The solution was filtered, and the volatiles were removed
under vacuum. An orange, sticky residue was obtained; scraping
resulted in solidification. IR analysis revealed that > 80% of [Cr] was in
the form of CpCr(CO)2(IMe)•. On the basis of that result, attempts
to purify this material were not undertaken.
The structure of CpCr(CO)2(IMe)• is of only moderate quality
(wR2 = 0.24), and several significant peaks of residual electron density
(ca. 2−3 e Å−3) were located within 1−2 Å of some of the Cr centers.
These peaks were chemically meaningless. It is plausible that there is
positional disorder in some of the molecules, but we have not been
able to model this disorder. Refinement of K+(18-crown-6)[CpCr-
(CO)2(IMe)]−·1/2THF (Pca21) gave a Flack parameter56 of 0.226(9)
(based on 9227 Friedel pairs, 95% of data), which we interpreted as an
1601
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603

Inorganic Chemistry
Article
indication for inversion twinning. Indeed, inclusion of twinning in the
model refined to a fraction of 0.23(1) for the inverted structure. The
refinement of K+[CpCr(CO)2(IMe)]−·3/4THF (P21) gave a Flack
parameter of 0.004(12) (based on 7135 Friedel pairs, 94% of data),
which we interpret as an indication that the correct absolute structure
was determined and that inclusion of inversion twinning was not
required.
The following is a list of programs used: data reductions, SAINT-
Plus version 6.63;57 absorption correction, SADABS;58 structural
solutions, SHELXS-97;59 structural refinement, SHELXL-97;60
graphics, Ortep-3 (version 2.02)61 for Windows. Solution and
refinement were done in the program OLEX2.62 Crystallographic
data are listed in Table 4; electronic files (CIF format) are provided in
the Supporting Information, and have also been deposited at the
Cambridge Crystallographic Data Centre (CCDC 910168−910172).
(8) (a) Hartung, J.; Pulling, M. E.; Smith, D. M.; Yang, D. X.;
Norton, J. R. Tetrahedron 2008, 64, 11822−11830. (b) Choi, J.; Tang,
L. H.; Norton, J. R. J. Am. Chem. Soc. 2007, 129, 234−240. (c) Tang,
L. H.; Norton, J. R. Macromolecules 2006, 39, 8236−8240. (d) Tang, L.
H.; Norton, J. R. Macromolecules 2006, 39, 8229−8235. (e) Tang, L.
H.; Norton, J. R. Macromolecules 2004, 37, 241−243. (f) Tang, L. H.;
Norton, J. R.; Edwards, J. C. Macromolecules 2003, 36, 9716−9720.
(g) O’Connor, J. M.; Friese, S. J. Organometallics 2008, 27, 4280−
4281.
(9) Choi, J.; Pulling, M. E.; Smith, D. M.; Norton, J. R. J. Am. Chem.
Soc. 2008, 130, 4250−4252.
(10) Roberts, J. A. S.; Franz, J. A.; van der Eide, E. F.; Walter, E. D.;
Petersen, J. L.; DuBois, D. L.; Bullock, R. M. J. Am. Chem. Soc. 2011,
133, 14593−14603.
(11) Roberts, J. A. S.; Appel, A. M.; DuBois, D. L.; Bullock, R. M. J.
Am. Chem. Soc. 2011, 133, 14604−14613.
ASSOCIATED CONTENT
(12) van der Eide, E. F.; Liu, T. B.; Camaioni, D. M.; Walter, E. D.;
Bullock, R. M. Organometallics 2012, 31, 1775−1789.
■
S
* Supporting Information
(13) Tilset, M.; Parker, V. D. J. Am. Chem. Soc. 1989, 111, 6711−
6717; as modified in J. Am. Chem. Soc. 1990, 112, 2843.
(14) Kiss, G.; Zhang, K.; Mukerjee, S. L.; Hoff, C. D.; Roper, G. C. J.
Am. Chem. Soc. 1990, 112, 5657−5658.
Additional details (tables, figure) on the reported crystal
structures; 1H NMR, IR, EPR, CV characterization data
(figures); EPR spectra of CpCr(CO)2(PCy3)• and simulation
details; crystallographic data (CIF format). This material is
available free of charge via the Internet at http://pubs.acs.org.
(15) It is convenient to generate this carbene in solution (by the
reaction of 1,3-dimethylimidazolium iodide with potassium hydride);
the neat material has been reported to be much less stable than
solutions of it. Arduengo, A. J., III; Dias, H. V. R.; Harlow, R. L.;
Kline, M. J. Am. Chem. Soc. 1992, 114, 5530−5534.
AUTHOR INFORMATION
■
(16) Air exposure of CpCr(CO)2(IMe)• solutions gives black,
insoluble precipitates, which were not further characterized.
(17) Atwood, C. G.; Geiger, W. E. J. Am. Chem. Soc. 1994, 116,
10849−10850.
Corresponding Author
*E-mail: morris.bullock@pnnl.gov.
Notes
The authors declare no competing financial interest.
(18) Bertini, I.; Luchinat, C.; Parigi, G. Solution NMR of
Paramagnetic Molecules. In Current Methods in Inorganic Chemistry;
Elsevier: New York, 2001; Vol. 2, p 33.
(19) (a) Wallen, S. L.; Schoenbachler, L. K.; Dawson, E. D.;
Blatchford, M. A. Anal. Chem. 2000, 72, 4230−4234. (b) Yonker, C.
R.; Linehan, J. C. J. Organomet. Chem. 2002, 650, 249−257.
(20) (a) Hanaya, M.; Iwaizumi, M. J. Organomet. Chem. 1992, 435,
337−345. (b) Hanaya, M.; Iwaizumi, M. Organometallics 1989, 8,
672−676.
(21) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877−910.
(22) We classify the K···OC interaction as side-on when ∠(K−O−
C) ≤ 90° (an arbitrary cutoff). In those cases, dotted lines between
potassium and the carbonyl carbons are drawn.
ACKNOWLEDGMENTS
■
We thank the U.S. Department of Energy, Office of Science,
Office of Basic Energy Sciences, Division of Chemical Sciences,
Geosciences and Biosciences for support. M.L.H. carried out
the crystallographic studies and was supported as part of the
Center for Molecular Electrocatalysis, an Energy Frontier
Research Center funded by the U.S. Department of Energy,
Office of Science, Office of Basic Energy Sciences. We thank
Dr. John C. Linehan for assistance with the high-pressure NMR
experiments, and Dr. Birgit Schwenzer for assistance with
recording the NIR spectra. The EPR studies were performed at
the William R. Wiley Environmental Molecular Sciences
Laboratory (EMSL), a national scientific user facility sponsored
by the Department of Energy’s Office of Biological and
Environmental Research located at PNNL. Pacific Northwest
National Laboratory is operated by Battelle for the U.S.
Department of Energy.
(23) (a) The use of MeCN solvent for reductions using KC8 may
seem unusual (THF is more commonly used), but it has precedent
(ref 23b). However, it is known that MeCN can be deprotonated by
KC8 (ref 23c), and we do not know whether the reduction of the
radical occurs directly by KC8 or by the cyanomethyl anion. (b) Mock,
M. T.; Potter, R. G.; O’Hagan, M. J.; Camaioni, D. M.; Dougherty, W.
G.; Kassel, W. S.; DuBois, D. L. Inorg. Chem. 2011, 50, 11914−11928.
(c) Savoia, D.; Trombini, C.; Umani-Ronchi, A. Tetrahedron Lett.
1977, 18, 653−656.
REFERENCES
(24) Kaljurand, I.; Kutt, A.; Soovali, L.; Rodima, T.; Maemets, V.;
̈
̈
̈
■
Leito, I.; Koppel, I. A. J. Org. Chem. 2005, 70, 1019−1028.
(1) Baird, M. C. Chem. Rev. 1988, 88, 1217−1227.
(2) Scott, S. L.; Espenson, J. H.; Zhu, Z. J. Am. Chem. Soc. 1993, 115,
1789−1797.
(3) Tyler, D. R. Acc. Chem. Res. 1991, 24, 325−331.
(4) Cahoon, J. F.; Kling, M. F.; Sawyer, K. R.; Frei, H.; Harris, C. B. J.
Am. Chem. Soc. 2006, 128, 3152−3153.
(25) Coetzee, J. F.; Padmanabhan, G. R. J. Am. Chem. Soc. 1965, 87,
5005−5010.
(26) We found (ref 12) that CpW(CO)2(IMe)H exists in solution as
a mixture of cis and trans isomers; they interconvert slow enough to be
distinguished by 1H NMR spectroscopy. Cis/trans isomerism may
1
occur for CpCr(CO)2(IMe)H as well, but only one set of H NMR
(5) (a) Cahoon, J. F.; Kling, M. F.; Sawyer, K. R.; Andersen, L. K.;
Harris, C. B. J. Mol. Struct. 2008, 890, 328−338. (b) Cahoon, J. F.;
Kling, M. F.; Schmatz, S.; Harris, C. B. J. Am. Chem. Soc. 2005, 127,
12555−12565.
(6) Zhang, J.; Grills, D. C.; Huang, K. W.; Fujita, E.; Bullock, R. M. J.
Am. Chem. Soc. 2005, 127, 15684−15685.
(7) (a) Eisenberg, D. C.; Lawrie, C. J. C.; Moody, A. E.; Norton, J. R.
J. Am. Chem. Soc. 1991, 113, 4888−4895. (b) Eisenberg, D. C.;
Norton, J. R. Isr. J. Chem. 1991, 31, 55−66.
signals is observed.
(27) Schwesinger, R.; Willaredt, J.; Schlemper, H.; Keller, M.;
Schmitt, D.; Fritz, H. Chem. Ber. 1994, 127, 2435−2454.
(28) Tolman, C. A. Chem. Rev. 1977, 77, 313−348.
(29) Cheng, T.-Y.; Brunschwig, B. S.; Bullock, R. M. J. Am. Chem. Soc.
1998, 120, 13121−13137.
(30) (a) Crabtree, R. H. The Organometallic Chemistry of the
Transition Metals, 5th ed.; Wiley: New York, 2009; p 76. (b) Bau, R.;
1602
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603

Inorganic Chemistry
Article
Teller, R. G.; Kirtley, S. W.; Koetzle, T. F. Acc. Chem. Res. 1979, 12,
176−183.
MeCN may be explained in the same way, as steric pressure could
facilitate the (reversible) dissociation of MeCN. Even if only small
equilibrium quantities of 16-electron CpCr(CO)2(IMe)+ are present
in MeCN solution, this would provide a pathway for decomposition
and for the electrochemical reduction of CpCr(CO)2(IMe)(MeCN)+
to CpCr(CO)2(IMe)• by a CE mechanism. Little is known about
related cations. However, the tricarbonyl CpCr(CO)3(MeCN)+ has
been isolated as the PF6− salt (ref 52b), its electrochemistry in MeCN
is known (ref 52c), and it has been reported to be unstable in CH2Cl2
(ref 52d). (b) Herberhold, M.; Ott, J.; Haumaier, L. Chem. Ber. 1985,
118, 3143−3150. (c) Tilset, M. Inorg. Chem. 1994, 33, 3121−3126.
(d) Ryan, O. B.; Tilset, M.; Parker, V. D. J. Am. Chem. Soc. 1990, 112,
2618−2626.
(31) Wayner, D. D. M.; Parker, V. D. Acc. Chem. Res. 1993, 26, 287−
294.
(32) Ellis, W. W.; Raebiger, J. W.; Curtis, C. J.; Bruno, J. W.; DuBois,
D. L. J. Am. Chem. Soc. 2004, 126, 2738−2743.
(33) The hydricity (ΔGH-) of CpCr(CO)2(IMe)H (59.2(9) kcal
mol−1) was also determined, using the equation ΔGH-(MH) = 1.37
pKa(MH) + 23.06 [E°(M•/−) + E°(M+/•)] + 79.6. Note that we use
E°(MS+/M•) instead of E°(M+/•), so that coordination of MeCN to
the 16-electron cation M+ (to form MS+) is taken into account. We
conservatively assign a relatively large uncertainty of
30 mV to
E°(MS+/M•); the wave is quasi-reversible, and E° may slightly deviate
from E1/2. The hydricity of CpCr(CO)2(IMe)H is similar to the 57(1)
kcal mol−1 (ref 11) for the tungsten complex CpW(CO)2(IMes)H
(IMes = 1,3-dimesitylimidazol-2-ylidene), also taking the formation of
MS+ into account.
(53) Birdwhistell, R.; Hackett, P.; Manning, A. R. J. Organomet. Chem.
1978, 157, 239−241.
(54) Benac, B. L.; Burgess, E. M.; Arduengo, A. J., III Org. Synth.
1986, 64, 92−95.
(55) Stoll, S.; Schweiger, A. J. Magn. Reson. 2006, 178, 42−55.
(56) (a) Flack, H. D.; Bernardinelli, G. Acta Crystallogr., Sect. A 1999,
55, 908−915. (b) Flack, H. D.; Bernardinelli, G. J. Appl. Crystallogr.
2000, 33, 1143−1148.
(34) Watkins, W. C.; Hensel, K.; Fortier, S.; Macartney, D. H.; Baird,
M. C.; McLain, S. J. Organometallics 1992, 11, 2418−2424.
(35) Cooley, N. A.; Watson, K. A.; Fortier, S.; Baird, M. C.
Organometallics 1986, 5, 2563−2565.
(57) SAINT, v. 6.63; Bruker AXS Inc.: Madison, WI.
(36) Fortier, S.; Baird, M. C.; Preston, K. F.; Morton, J. R.; Ziegler,
T.; Jaeger, T. J.; Watkins, W. C.; MacNeil, J. H.; Watson, K. A.;
Hensel, K.; Le Page, Y.; Charland, J.-P.; Williams, A. J. J. Am. Chem.
Soc. 1991, 113, 542−551.
(58) Sheldrick, G. M. SADABS Program for Absorption Correction of
Area Detector Frames; Bruker AXS.: Madison, WI.
(59) Sheldrick, G. M. SHELXS-97 Program for Crystal Structure
Solution; Institut fur Anorganische Chemie der Universitat Gottingen:
̈
̈
̈
̈
(37) (a) Ackermann, K.; Hofmann, P.; Kohler, F. H.; Kratzer, H.;
̈
Gottingen, Germany, 1997.
̈
̈
Krist, H.; Ofele, K.; Schmidt, H. R. Z. Naturforsch. B 1983, 38, 1313−
(60) Sheldrick, G. M. SHELXL-97 Program for Crystal Structure
Refinement; Institut fur Anorganische Chemie der Universitat
̈
1324. (b) Ofele, K. J. Organomet. Chem. 1968, 12, P42−P43.
(c) Tafipolsky, M.; Scherer, W.; Ofele, K.; Artus, G.; Pedersen, B.;
̈
̈
̈
̈
Gottingen: Gottingen, Germany, 1997.
(61) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
̈
̈
Herrmann, W. A.; McGrady, G. S. J. Am. Chem. Soc. 2002, 124, 5865−
5880.
(38) (a) Adams, R. D.; Collins, D. E.; Cotton, F. A. J. Am. Chem. Soc.
1974, 96, 749−754. (b) McLain, S. J. J. Am. Chem. Soc. 1988, 110,
643−644.
(62) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.;
Puschmann, H. J. Appl. Crystallogr. 2009, 42, 339−341.
(39) (a) A CSD search (version 5.33, August 2012) for 5-membered
NHC complexes of chromium gave 37 hits. (b) Cr−C(carbene) bond
lengths in other Cr-IMe complexes: Cr(CO)4(IMe)2, 2.131(3) Å (ref
37a.); [Cr(CO)4(IMe)2]+•[PF6]−, 2.113(4) and 2.119(4) Å (ref 37a.);
Cr(CO)5(IMe), 2.138(2) and 2.145(2) Å (two independent
molecules, ref 37c.). (c) For a monodentate NHC, the heretofore
shortest Cr−C(carbene) bond length is 2.075(4) Å, found in
(benzimidazol-2-ylidene)Cr(CO)5. See: Hahn, F. E.; Langenhahn,
V.; Meier, N.; Lugger, T.; Fehlhammer, W. P. Chem.Eur. J. 2003, 9,
̈
704−712.
(40) Hammack, D. J.; Dillard, M. M.; Castellani, M. P.; Rheingold, A.
L.; Rieger, A. L.; Rieger, P. H. Organometallics 1996, 15, 4791−4797.
(41) Rieger, P. H. Coord. Chem. Rev. 1994, 135/136, 203−286.
(42) Woska, D. C.; Ni, Y. P.; Wayland, B. B. Inorg. Chem. 1999, 38,
4135−4138.
(43) Tilset, M. J. Am. Chem. Soc. 1992, 114, 2740−2741.
(44) Fischer, P. J.; Heerboth, A. P.; Herm, Z. R.; Kucera, B. E.
Organometallics 2007, 26, 6669−6673.
(45) Hey-Hawkins, E.; von Schnering, H. G. Z. Naturforsch. B 1991,
46, 621−624.
(46) Braunschweig, H.; Radacki, K.; Kraft, K.; Stellwag, S. Z.
Naturforsch. B 2010, 65, 1073−1076.
(47) (a) Darensbourg, M. Y. Prog. Inorg. Chem. 1985, 33, 221−274.
(b) Darensbourg, M. Y.; Jimenez, P.; Sackett, J. R.; Hanckel, J. M.;
Kump, R. L. J. Am. Chem. Soc. 1982, 104, 1521−1530.
(48) Moore, E. J.; Sullivan, J. M.; Norton, J. R. J. Am. Chem. Soc.
1986, 108, 2257−2263.
(49) Parker, V. D.; Handoo, K. L.; Roness, F.; Tilset, M. J. Am. Chem.
Soc. 1991, 113, 7493−7498.
(50) Burchell, R. P. L.; Sirsch, P.; Decken, A.; McGrady, G. S. Dalton
Trans. 2009, 5851−5857.
(51) Skagestad, V.; Tilset, M. J. Am. Chem. Soc. 1993, 115, 5077−
5083.
(52) (a) The low stability of CpCr(CO)2(IMe)(MeCN)+ and the
quasi-reversibile electrochemical oxidation of CpCr(CO)2(IMe)• in
1603
dx.doi.org/10.1021/ic302460y | Inorg. Chem. 2013, 52, 1591−1603