Quinobis(imidazolylidene): Synthesis and study of an electron-configurable bis(N-heterocyclic carbene) and Its bimetallic complexes

Article abstract of DOI:10.1002/chem.200901883

Reaction of bromanil with N,N′;W-dimesitylformamidine followed by deprotonation with NaN(SiMe₃)₂ afforded 1,1′,3,3′-tetramesitylquinobis(imidazolylidene) (1), a bis(N-heterocyclic carbene) (NHC) with two NHC moieties connected by a redox active p-quinone residue, in 72 % yield of isolated compound. Bimetallic complexes of 1 were prepared by coupling to FcN₃ (2) or FcNCS (3; Fc = ferrocenyl) or coordination to [M(cod)Cl] (4a or 4b, where M = Rh or Ir, respectively; cod = 1,5-cyclooctadiene). Treatment of 4a and 4b with excess CO(g) afforded the corresponding [M(CO)₂Cl] complexes 5 a and 5 b, respectively. Analysis of 2-5 by NMR spectroscopy and X-ray diffraction indicated that the electron-deficient quinone did not significantly affect the inherent spectral properties or coordination chemistry of the flanking imidazolylidene units, as compared to analogous NHCs. Infrared spectroscopy and cyclic voltammetry revealed that decreasing the electron density at ML _n, afforded an increase in the stretching energy and a decrease in the reduction potential of the quinone, indicative of metal-quinone electronic interaction. Differential pulse voltammetry and chronoamperometry of the metal-centered oxidations in 2-4 revealed two single, one-electron peaks. Thus, the metal atoms bound to 1 are oxidized at indistinguishable potentials and do not appear electronically coupled. However, the metal-quinone interaction was used to increase the electron density at coordinated metal atoms. Infrared spectroelectrochemistry revealed that the average v_CO values for 5 a and 5 b decreased by 14 and 15 cm^-1, respectively, upon reduction of the quinone embedded within 1. These shifts correspond to 10 and 12 cm ^-1 decreases in the Tolman electronic parameter of this ditopic ligand.

Full text of DOI:10.1002/chem.200901883

DOI: 10.1002/chem.200901883
Quinobis(imidazolylidene): Synthesis and Study of an Electron-Configurable
Bis(N-Heterocyclic Carbene) and Its Bimetallic Complexes
Andrew G. Tennyson, Robert J. Ono, Todd W. Hudnall, Dimitri M. Khramov,
Joyce A. V. Er, Justin W. Kamplain, Vincent M. Lynch, Jonathan L. Sessler,* and
Christopher W. Bielawski*[a]
Abstract: Reaction of bromanil with
N,N’-dimesitylformamidine followed by
X-ray diffraction indicated that the
electron-deficient quinone did not sig-
nificantly affect the inherent spectral
properties or coordination chemistry of
the flanking imidazolylidene units, as
compared to analogous NHCs. Infrared
spectroscopy and cyclic voltammetry
revealed that decreasing the electron
density at MLn afforded an increase in
the stretching energy and a decrease in
the reduction potential of the quinone,
indicative of metal–quinone electronic
interaction. Differential pulse voltam-
metry and chronoamperometry of the
metal-centered oxidations in 2–4 re-
vealed two single, one-electron peaks.
Thus, the metal atoms bound to 1 are
oxidized at indistinguishable potentials
and do not appear electronically cou-
pled. However, the metal–quinone in-
teraction was used to increase the elec-
tron density at coordinated metal
atoms. Infrared spectroelectrochemis-
try revealed that the average nCO
values for 5a and 5b decreased by 14
and 15 cmꢀ1, respectively, upon reduc-
tion of the quinone embedded within
1. These shifts correspond to 10 and
12 cmꢀ1 decreases in the Tolman elec-
tronic parameter of this ditopic ligand.
deprotonation with NaN
forded 1,1’,3,3’-tetramesitylquinobis-
ACHTUNGTRENNUNG(imidazolylidene) (1), a bis(N-hetero-
ACHTUNGTERN(NUNG SiMe3)2 af-
cyclic carbene) (NHC) with two NHC
moieties connected by a redox active
p-quinone residue, in 72% yield of iso-
lated compound. Bimetallic complexes
of 1 were prepared by coupling to FcN3
(2) or FcNCS (3; Fc=ferrocenyl) or
coordination to [MACHTUNGTRENNUNG(cod)Cl] (4a or 4b,
where M=Rh or Ir, respectively;
cod=1,5-cyclooctadiene). Treatment of
4a and 4b with excess CO(g) afforded
the corresponding [M(CO)2Cl] com-
plexes 5a and 5b, respectively. Analy-
sis of 2–5 by NMR spectroscopy and
Keywords: bimetallic complexes ·
carbene ligands · electrochemistry ·
ligand design · quinones
Introduction
Variants featuring two NHC moieties linked by a p-conju-
gated system, such as benzobis(imidazolylidene) (BBI),[20,21]
triazolyldiylidene (A),[22] bitriazolylidene (B),[23] pyridylbis-
From enantioselective synthesis[1] to the production of dihy-
drogen,[2] bimetallic catalysts have been employed in a varie-
ty of applications. Organometallic systems bearing N-hetero-
cyclic carbene (NHC) ligands are useful catalysts for func-
tional-group transformations and polymerization reac-
tions,[3–13] and complexes featuring multitopic NHCs have re-
cently emerged[14] as promising candidates for asymmetric
and tandem catalysis,[15–17] as well as other applications.[18,19]
AHCTUNGTRENNUNG
(imidazolylidene) (C),[24–26] and pyridazinylbis(imidazolyl-
AHCTUNGTRENNUNG
idene) (D)[27,28] have found numerous uses as ditopic ligands
(Scheme 1). For example, BBIs have yielded phosphorescent
complexes,[29] polymeric catalysts,[30] emissive and conducting
polymers,[21,31–33] and self-assembled materials.[34,35] The p-
systems linking the NHC units in BBI and A–D enable
metal–metal interactions that can potentially modify the ex-
isting characteristics of the individual metal atoms (e.g., cat-
alytic activity or phosphorescence) or create materials that
exhibit new physical properties or functions altogether.
In nature, the reactivity of a bimetallic active site is often
controlled through alteration of the stereoelectronic envi-
ronment, such as ligand rearrangement or protonation/de-
protonation.[36–40] A synthetic bimetallic complex lacks the
secondary coordination sphere that enzymes employ to en-
capsulate and stabilize metal–ligand geometries, and there-
[a] Dr. A. G. Tennyson, R. J. Ono, Dr. T. W. Hudnall,
Dr. D. M. Khramov, J. A. V. Er, Dr. J. W. Kamplain, Dr. V. M. Lynch,
Prof. J. L. Sessler, Prof. C. W. Bielawski
Department of Chemistry & Biochemistry
The University of Texas at Austin
Austin, TX 78712 (USA)
E-mail: sessler@cm.utexas.edu
bielawski@cm.utexas.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200901883.
304
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 304 – 315

FULL PAPER
ously showed that redox active groups can be built into
NHCs, whereby redox changes thereof can be used to tune
the electronic properties of coordinated metal atoms.[57,58]
Initial efforts focused on the monotopic NqMes (Scheme 3),
a
napthoquinone-annulated 1,3-dimesitylimidazolylidene,
which enabled fundamental studies of metal–NHC interac-
tions[58–60] and new redox-switchable catalysts.[61] Because the
BBI scaffold has been successfully employed in a range of
applications,[21,29,31–33,35] we envisioned that a variant incorpo-
rating an electroactive group could potentially enable
redox-switchable control over the intrinsic properties of
these materials. We reasoned that a ditopic analogue of
NqMes could be obtained by replacing the benzo linker in
BBI with a p-quinone moiety by similar synthetic methods,
thus affording an electron-configurable bis-NHC that com-
bines the electronic tunability of NqMes with the structural
versatility of the BBI scaffold.
Scheme 1. Ditopic NHCs with p-conjugated linkers. For BBI, R=bulky
alkyl or aryl.
fore analogous tuning in small molecules is difficult to ach-
ieve, particularly in a reversible manner. Electron-configura-
ble ligands (i.e., those that undergo a reversible redox
change to alter their electron-donating/withdrawing charac-
ter) are attractive alternatives, given that electrochemical re-
versibility can be easily evaluated by cyclic voltammetry
(CV).[41] Furthermore, many chemical redox agents with a
wide range of potentials are known and often commercially
available.[42] Complexes featuring electron-configurable li-
gands have yielded invaluable contributions to fields that
range from fundamental studies of metal–ligand electronic
interactions to practical redox-switchable catalysis.[43–49] Bi-
metallic variants comprising redox-active ditopic linkers
have also been reported (Scheme 2),[50–55] such as tetrathia-
fulvene (E), p-quinonediphosphine (F), and quinodiimine
(G). However, many of these scaffolds are highly specialized
and difficult to modify; thus incorporating electron donat-
ing/withdrawing or solubilizing substituents into these sys-
tems can be synthetically challenging.
Scheme 3. Structures of electron-configurable NHCs.
Herein we report the synthesis of 1,1’,3,3’-tetramesitylqui-
nobis(imidazolylidene) (1, Scheme 3) and investigations into
the structural and electronic properties of bimetallic com-
plexes thereof. We employed NHC coupling chemistry[62,63]
to affix ferrocene (Fc) groups,[64] given that Fc displays well-
defined electrochemical behavior. Additionally, we synthe-
sized coordination complexes of 1 featuring [MACHTNUTRGNE(UNG cod)Cl] and
[M(CO)2Cl] (M=Rh or Ir, cod=1,5-cyclooctadiene) units.
To explore how metal coordination affects quinobis(imid-
AHCTUNGTREGaNNUN zolylidene) (QBI) electron density, the quinone stretching
energies and reduction potentials in these complexes were
analyzed by IR spectroscopy and electrochemistry. Similarly,
metal oxidation potentials and carbonyl stretching frequen-
cies were measured to ascertain the electron-donating ability
of 1 to coordinated MLn units. Subsequent spectroelectro-
chemical IR analyses of quinone/carbonyl stretching ener-
gies upon metal oxidation and ligand reduction, respectively,
provided insight into the electronic interactions[65] between
the quinone in 1 and coordinated metal atoms. We found
that 1 exhibits the coordination chemistry of a BBI, but in-
creases electron density at bound MLn units upon reduction
of the quinone moiety.
Scheme 2. Examples of redox-active ditopic ligands.
Unlike the aforementioned redox active ditopic ligands,
BBIs and other ditopic NHCs can be synthesized and readi-
ly modified to exhibit a range of solubilities and stereoelec-
tronic properties.[56] Inspired by these advantages, we previ-
Results and Discussion
Condensation of bromoanil with four equivalents of N,N’-di-
mesitylformamidine in CH3CN afforded bis-azolium [1H2]
Chem. Eur. J. 2010, 16, 304 – 315
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
305

C. W. Bielawski et al.
[Br]2 in excellent yield (91%, Scheme 4). While other bases
could be used to trap the HBr generated by this double-Mi-
chael addition, the best yields were obtained by using an
pound 1 was also prepared from [1H2]ACTHUNGTERN[NUG OTf]2 in comparable
yields. The diagnostic 13C signal for the 2,2’-positions in 1
was significantly downfield (d=232.6 ppm, [D8]THF,
Table 1), but in accord with
those of other NHCs (d=
205.9–244.5 ppm)[20,70,76–79] and
similar to that of NqMes (d=
231.7 ppm, C6D6).[58] The struc-
ture of 1 was confirmed by X-
ray analysis (Figure 1). The N-
C-N angle observed in
1
(102.3(3)8, Table 1) was nearly
identical to that of NqMes
(102.1(6)8) and slightly more
acute than those of known
BBIs (R=Ad or tBu, 104.4–
Scheme 4. Synthesis of 1. i) 4 equiv N,N’-dimesitylformamidine, CH3CN, 1108C, 24 h. ii) Excess MeOTf,
CH2Cl2, RT, 2 h. iii) 2 equiv NaNACHTNUTRGNEUNG(SiMe3)2, THF, RT, 30 min.
excess of formamidine as the sacrificial base.[66,67] Although
[1H2][Br]2 served as the synthetic precursor for 1, poor solu-
bility even in polar organic solvents, such as CH3CN or
DMSO, precluded detailed electrochemical analyses (vide
infra). To enhance organic solubility of the bis-azolium,
anion exchange[68] was effected by treating [1H2][Br]2 with
MeOTf (OTf=CF3SO3), affording [1H2]ACTHNUTRGNEUNG[OTf]2 (eliminating
volatile CH3Br) in high yield (94%). The 13C NMR signals
for the quinone carbon atoms and those in the 2,2’-positions
in [1H2][Br]2 (d=163.7 and 141.2 ppm, respectively) and
[1H2]ACHTUNGTRENNUNG[OTf]2 (d=164.9 and 143.7 ppm, respectively; Table 1)
Table 1. Selected spectroscopic and structural data.
(C2,2ꢁ) [ppm] (C4,4ꢁ) [ppm] nCO [cmꢀ1
[d]
d
G
d
E
]
N-C-N [8][e]
Figure 1. ORTEP showing 50% probability thermal ellipsoids and select-
ed atom labels for 1. Hydrogen atoms and solvent molecules have been
[a]
[1H2]
E
143.7
232.6
150.5
141.2
201.0
195.3
193.0
188.7
164.9
168.0
164.9
171.1
164.5
164.9
164.1
164.1
1710
1676
1662
1692
1680
1678
1688
1686
110.2(3)
102.3(3)
105.7(5)
108.0(4)
104.11(16)
104.14(17)
1[b]
ꢀ
omitted for clarity. Selected bond lengths [ꢂ] and angles [8]: N1 C1,
2[c]
ꢀ
ꢀ
ꢀ
ꢀ
1.375(4), N2 C1, 1.378(4), N2 C3, 1.386(4), O1 C4, 1.218(4), C2 C3,
3[c]
ꢀ
1.367(4), C2 C4, 1.476(4); N1-C1-N2, 102.3(3).
4a[c]
4b[c]
5a[c]
5b[c]
[f]
–
–
104.88).[20] Presumably, the acute N-C-N angle observed in 1
reflects the smaller steric bulk of mesityl versus tertiary
alkyl groups.[80] Given the similar N-C-N angles in 1 and 1,3-
dimesitylimidazolylidene (IMes) of 102.3(3) and 101.4(3)8,
respectively,[81] the QBI could be viewed as two IMes frag-
ments annulated by p-quinone.
[f]
[a] 13C NMR measurements performed in CD3CN. [b] 13C NMR meas-
urements performed in [D8]THF. [c] 13C NMR measurements performed
in CDCl3. [d] Quinone stretching energies nCO measured in KBr matrices.
[e] This metric parameter is sensitive to the electronic environment at
the QBI 2,2’-positions. [f] Not determined.
To explore the coordination chemistry and electronic
properties of 1, we envisioned using electrochemical tech-
niques to probe the metal-to-ligand and ligand-to-metal
electronic interactions. Thus, we sought to affix redox active
MLn units to the redox active QBI by capitalizing on the re-
activity of NHCs toward electrophiles.[62,63] Given that ferro-
cene (FcH) can be functionalized[64] with NHC-reactive moi-
eties and exhibits well-defined electrochemical behavior, we
pursued derivatives bearing groups that would facilitate cou-
pling to 1. Previously, we attached Fc-based electrochemical
handles to BBI (R=tert-amyl) by using NHC–azide[82] and
NHC–isothiocyanate[62] coupling chemistry to afford adducts
II and III, respectively (Scheme 5).[83] Electrochemical anal-
agreed with those of napthoquinone-annulated imidazolium
chloride [NqMesH][Cl] (d=173.4 and 141.8 ppm, respec-
tively).[58] Structural assignment of [1H2]
ACTHNUGTRNE[UNG OTf]2 was con-
firmed by X-ray diffraction (see Figure S1 in the Supporting
Information). This complex has an N1-C2-N2 angle of
110.2(3)8 (Table 1),[69] which compares well to that of
[NqMesH][Cl] (108.07(14)8)[58] and analogous values ob-
served in other benzimidazolium-derived compounds
(110.1–111.78).[20,70–76]
Deprotonation of [1H2][Br]2 with NaN
ACHTUNGTRENN(NUG SiMe3)2 afforded
quinobis(imidazolylidene) 1 in good yield (79%). Com-
306
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 304 – 315

Quinobis(imidazolylidene)
FULL PAPER
Scheme 5. Previously reported BBI adducts of FcN3 (II) and FcNCS
(III).
ysis of II revealed electronic coupling between the two
metal centers, whereby oxidation of one ferrocene unit shift-
ed that of the other to a higher potential (DE1/2 =0.14 V).
Conversely, III only displayed one redox couple, indicating
that the iron centers are oxidized at indistinguishable[84] po-
tentials and do not interact. The lack of electronic coupling
between metal atoms in III was attributed to the NCS link-
age being orthogonal to the BBI, thus interrupting conjuga-
tion.
Figure 2. ORTEP showing 50% probability thermal ellipsoids and select-
ed atom labels for 2. Hydrogen atoms and solvent molecules have been
ꢀ
omitted for clarity. Selected bond lengths [ꢂ] and angles [8]: N1 C1
ꢀ
ꢀ
1.381(7), N2 C1 1.392(7), N3 C1 1.316(7); N1-C1-N2 105.7(5), N3-N4-
N5 110.5(5), N2-C1-N3-N4 17.3(9), N4-N5-C23-C27 165.4(8).
To prepare QBI-supported analogues of II and III and
enable similar electrochemical investigations, we combined
1 with 2 equiv of FcN3 or FcNCS to obtain the desired ad-
ducts [(FcN3)2(1)] (2) or [(FcNCS)2(1)] (3) in modest yields
(54 and 51%, respectively, Scheme 6). Consistent with the
loss of carbenoid character, the 13C NMR signal for the 2,2’-
positions of these compounds had shifted significantly from
d=232.6 ppm in 1 ([D8]THF) to 150.5 and 141.8 ppm in 2
and 3 (CDCl3), respectively (Table 1). Structural confirma-
tion of 2 and 3 was achieved by crystallographic analysis
(see Figures 2 and 3, respectively).
and Figure 3, as well as Table 1), slightly more acute than
those found in II and III, reflecting the different N-C-N geo-
metries induced by aryl versus bulky N-alkyl substituents.
Because the N-C-N bond angle in 3 is greater than that in 2
and comparable to that of [1H2]ACHTNUGTRENUNG[OTf]2, we conclude that the
QBI core in 3 bears significant positive charge and thus has
zwitterionic character, consistent with III.[83] Whereas the
triazene linkers in 2 are nearly coplanar with the QBI core
(N2-C1-N3-N4 17.3(9)8), the isothiocyanate groups in 3 are
nearly perpendicular (N1-C1-C5-S1 88.7(6)8), that is, geome-
tries similar to those observed in II and III (31.3(6) and
92.6(5)8, respectively).[83] Presumably, the triazene units are
more coplanar with the QBI in 2 than the BBI in II due to
diminished steric crowding of the N-mesityl substituents.
Furthermore, the internal angles for triazene and isothiocya-
To investigate how 1 interacts with FcN3 and FcNCS rela-
tive to a BBI, we compared the metric parameters of 2 and
3 to BBI-supported analogues II and III. The N-C-N angles
in 2 and 3 were found to be 105.7(5) and 108.0(4)8 (Figure 2
nate linkages in
2
and
3
(110.5(5) and 136.5(4)8, respec-
tively) are nearly identical to
those observed in II and III
(110.3(3) and 135.3(3)8, respec-
tively). Because 1 forms ad-
ducts with FcN3 and FcNCS
that are comparable to their
BBI-supported analogues, we
conclude that the reactivity of
the imidazolylidene moieties in
1
is preserved despite their
being linked by a redox-active
quinone moiety.
After verifying that 1 engag-
ed in NHC–electrophile cou-
pling chemistry, we sought to
explore
chemistry of this bis-NHC by
affixing [M(cod)Cl] units (M=
the
coordination
AHCTUNGTRENNUNG
Rh or Ir) to 1, chosen for their
well-studied spectroscopic and
structural properties in other
Scheme 6. Syntheses of complexes 2–5. i) 2 equiv FcN3, THF, 16 h, 54% yield. ii) 2 equiv FcNCS, toluene, 16 h,
51% yield. iii) 1 equiv [M
ACHTUNGTERN(NUNG cod)Cl]2, THF, 16 h (M=Rh or Ir for 4a and 4b in 69 and 64% yield, respectively).
iv) 1 atm CO(g), CH2Cl2, 30 min, 91 and 93% yield for 5a and 5b, respectively.
Chem. Eur. J. 2010, 16, 304 – 315
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
307

C. W. Bielawski et al.
Figure 3. ORTEP showing 50% probability thermal ellipsoids and select-
ed atom labels for 3. Hydrogen atoms and solvent molecules have been
ꢀ
omitted for clarity. Selected bond lengths [ꢂ] and angles [8]: N1 C1
ꢀ
ꢀ
ꢀ
1.360(7), N2 C1 1.351(6), S1 C5 1.706(6), C1 C5 1.486(7); N1-C1-N2
108.0(4), S1-C5-N3 136.5(4), N1-C1-C5-S1 88.7(6), C25-C24-N3-C5
16.5(10).
Figure 4. ORTEP showing 50% probability thermal ellipsoids and select-
ed atom labels for 4a. Hydrogen atoms and solvent molecules have been
ꢀ
omitted for clarity. Selected bond lengths [ꢂ] and angles [8]: N1 C1
ꢀ
ꢀ
ꢀ
ꢀ
1.376(3), N2 C1 1.376(3), C1 Rh1 2.033(2), Rh1 C5 2.224(2), Rh1 C6
ꢀ
ꢀ
2.198(2), Rh1 C7 2.127(2), Rh1 C8 2.103(2); N1-C1-N2 104.11(16), N1-
NHC-supported complexes.[85–89] Furthermore, [M
ACTHNUTRGNEU(GN cod)Cl]
C1-Rh1-Cl1 101.88(17). The cod bite angle is 86.58.
units are easily converted to [M(CO)2Cl], enabling investi-
gation of the electron density at the metal center by IR
spectroscopic analysis of the CO stretching frequencies.[85,86]
201.0 and 195.3 ppm (in CDCl3) in 4a and 4b to d=193.0
and 188.7 ppm (CDCl3) in 5a and 5b (Table 1), respectively,
upfield shifts consistent with those of analogous NqMes-sup-
ported complexes.[58] Although single crystals of 5a were ob-
tained (Figure 5), diffraction-quality crystals of 5b remained
Reaction of 1 with [M
ACHTUNGTNER(NUNG cod)Cl]2 in THF afforded bimetallic
complexes [{M(cod)Cl}2(1)] (4a, M=Rh; 4b, M=Ir) in
ACHTUNGTRENNUNG
good yields (69 and 64% yield, respectively; Scheme 6). The
13C NMR signal for the 2,2’-positions in 4a found at d=
201.0 ppm (JRhC =53.1 Hz in CDCl3, Table 1) was upfield rel-
ative to 1 (d=232.6 ppm in [D8]THF), but consistent with
ꢀ
elusive. The Rh1 C1 distance in 5a (2.067(4) ꢂ) is within
the range of values reported for other NHC-supported
those of other [Rh
ACHTUNGTRENNUNG(cod)Cl] complexes supported by NHCs,
[Rh(CO)2Cl] complexes (2.06–2.09 ꢂ).[95,96] Furthermore,
ꢀ
including NqMes
(d=180.6–225.8 ppm;
JRhC =41–
the rhodium–carbonyl and C O bond lengths measured cis
(1.787(7) and 1.065(9) ꢂ, respectively) and trans (1.890(5)
and 1.113(8) ꢂ, respectively) to the NHCs in 5a compare
76 Hz).[57–59,70,85,90–94] Similarly, the 13C signal for the 2,2’-posi-
tions in 4b (d=195.3 ppm; CDCl3) was in good agreement
with those reported for other NHC-supported [IrACTHNUTRGNEUG(N cod)Cl]
complexes (d=179.6–208.2 ppm).[22,70,87,93,94]
Crystals of 4a suitable for X-ray diffraction enabled struc-
tural confirmation (Figure 4). A structure of the Ir congener
suitable for publication could not be obtained due to twin-
ꢀ
ning. The Rh1 C1 distance of 2.033(2) ꢂ in 4a is within the
range of rhodium–NHC bond lengths observed in analogous
BBI complexes (2.003–2.073 ꢂ)[20] and monometallic NHC-
supported variants (2.00–2.09 ꢂ).[22,57,59,90,92,94–96] In addition,
the average cis and trans rhodium–olefin distances in 4a
(2.115 and 2.211 ꢂ, respectively) are comparable to those
observed in the BBI-based variants (2.128–2.190 and 2.110–
2.207 ꢂ for cis and trans positions, respectively) and consis-
tent with the Rh–cod bond lengths in monometallic NHC-
supported analogues (2.09–2.23 ꢂ).[22,57,59,90,92,94–96] Further-
more, the bite angle of the cod ligand in 4a (86.58) is within
the range of values for NHC-supported [RhACTHNUTRGENU(GN cod)Cl] com-
Figure 5. ORTEP showing 50% probability thermal ellipsoids and select-
ed atom labels for 5a. Hydrogen atoms and solvent molecules have been
plexes (85–898).
Independently treating 4a and 4b with excess CO(g) af-
forded [{Rh(CO)2Cl}2(1)] (5a) and [{Ir(CO)2Cl}2(1)] (5b),
respectively, in near-quantitative yields (Scheme 6). The
13C NMR signals for the 2,2’-positions changed from d=
ꢀ
omitted for clarity. Selected bond lengths [ꢂ] and angles [8]: N1 C1
ꢀ
ꢀ
ꢀ
ꢀ
1.354(11), N2 C1 1.372(11), O3 C3 1.115(9), O4 C4 1.067(10), C1 Rh1
ꢀ
ꢀ
2.063(5), Rh1 C3 1.889(6), Rh1 C4 1.776(8); N1-C1-N2 104.14(17), N1-
C1-Rh1-Cl1 92.5(6).
308
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 304 – 315

Quinobis(imidazolylidene)
FULL PAPER
ꢀ
well to the range of Rh CO (1.84–1.88 ꢂ and 1.90–1.91 ꢂ
Complexes 5a and 5b also displayed IR-active metal–car-
bonyl stretching modes, which allow measurement of the
electron density at the coordinated metal centers. Because
NHCs are strong s donors, a CO ligand trans to the NHC
will have a higher stretching frequency than a CO ligand
with cis orientation.[85] The solid-state CO stretching bands
trans and cis to the NHC units in 5a occurred at 2089 and
2005 cmꢀ1, respectively, and were within the range observed
for other NHC-supported [Rh(CO)2Cl] complexes (trans:
ꢀ
for the cis and trans positions, respectively) and C O
(1.065(9) and 1.113(8) ꢂ, respectively) distances observed in
similar complexes supported by NHCs.[95,96] Given that the
structural features of [RhACHTNUTRGNE(NUG cod)Cl] or [Rh(CO)2Cl] fragments
coordinated to 1 agree well with those bound by other
NHCs, we conclude that the Rh–NHC interactions are com-
parable. Collectively, the spectral and structural similarities
between 4 and 5 as well as other NHC-supported [M-
(cod)Cl] and [M(CO)2Cl] complexes indicate that the qui-
2057–2093 cmꢀ1,
cis:
1984–2006 cmꢀ1),
including
none linker does not perturb the fundamental coordination
chemistry of the two imidazolylidene units.
NqMes.[57,58,95,96] Similarly, the nCO values in KBr for conge-
ner 5b at 2075 and 1989 cmꢀ1 (trans and cis, respectively)
were consistent with NHC-supported [Ir(CO)2Cl] complexes
(trans: 2055–2072 cmꢀ1, cis: 1971–1989 cmꢀ1), but at higher-
than-average energies.[22,86,88,101]
After studying the structural and spectroscopic features of
the MLn units bound to 1, we sought to measure the elec-
tron density within the QBI scaffold via IR spectroscopic
analysis of the quinone carbonyl functional groups. As the
system becomes more electron deficient, the nCO (quinone)
value should increase (e.g., cyanil>chloranil>p-quinone>
duroquinone).[97–99] Of all the compounds investigated,
[1H2]-
A more quantitative measure of the electron donating
ability of a ligand is its Tolman electronic parameter (TEP),
which can be calculated from the stretching energies of
metal carbonyl complexes it supports.[102] Work pioneered
by Crabtree et al. and expanded by Nolan et al. and Plenio
et al. allows determination of the TEP for NHC-supported
[Ir(CO)2Cl] complexes: TEP=0.847ꢃnav (Ir)+336 cmꢀ1,
where nav (Ir) is the average nCO value of the Ir-bound cis
and trans carbonyl ligands.[86,87,89] Recently, Plenio and Wolf
developed an equation for converting nav (Rh) values for
NHC-supported [Rh(CO)2Cl] complexes onto the same
energy scale as [Ir(CO)2Cl] complexes, whereby nav (Ir)=
0.8695ꢃnav (Rh)+250.7 cmꢀ1, enabling comparison of the
TEP values of these two congeners.[85] For complexes 5a
and 5b, the average solid-state CO stretching energies nav
are 2047 and 2032 cmꢀ1, respectively, with no significant dif-
ference between the solid- and solution-state measurements
(Table 2). Using nav (5a)=2047 cmꢀ1, the Ir-scaled value was
[Br]2 and [1H2]ACHTUNGTRENNUNG[OTf]2 displayed the highest values for nCO in
KBr at 1708 and 1710 cmꢀ1 (Table 1), respectively, consistent
with these molecules being the most electron deficient due
to their dicationic nature. Deprotonation to form 1 resulted
in a substantial shift in nCO to lower energy (1676 cmꢀ1 in
KBr), reflecting the greater electron density within the neu-
tral bis-carbene, and it agrees well with the value observed
in NqMes (1671 cmꢀ1 in KBr).[58] Whereas the solid-state
quinone stretching frequency for triazene 2 (1662 cmꢀ1) was
slightly lower in energy than that of 1, the nCO value for iso-
thiocyanate adduct 3 (1692 cmꢀ1) was significantly higher,
that is, FcN3 donates electron density to 1, and FcNCS is ef-
fectively electron-withdrawing. Consistent with its proposed
zwitterionic formulation, the nCO (quinone) for 3 is closest in
energy to that observed for [1H2]ACTHNUTRGEN[UNG OTf]2.
Table 2. Carbonyl stretching frequencies and Tolman electronic parame-
ters.
Although adducts 2 and 3 exhibit significant disparity in
quinone stretching energies, bimetallic complexes 4 and 5,
featuring metal atoms bound directly to the NHC moieties,
Solid-state[a]
TEP
[cmꢀ1
Solution-state[b]
nav TEP
[cmꢀ1 [cmꢀ1
Reduced[c]
TEP
[cmꢀ1
]
nav
[cmꢀ1
nav
[cmꢀ1
display a narrower range of nCO.AHCTUNGERN(NUG quinone) values (1678–
G
]
E
]
R
]
G
]
N
]
U
1688 cmꢀ1; see Table 1). For 4a and 5a, the nCO (quinone)
5a
5b
2047
2032
2055.9
2057.1
2048
2032
2056.6
2057.1
2034
2018
2046
2045
values of 1680 and 1688 cmꢀ1 in KBr, respectively, reflect
the weaker electron-withdrawing character of [RhACTHNUGRTNEUNG(cod)Cl]
[a] Average values for metal-carbonyl stretching modes (nav) and Tolman
electronic parameters (TEP) obtained in KBr matrices. [b] Average
values for metal-carbonyl stretching modes (nav) and Tolman electronic
parameters (TEP) obtained in CH2Cl2 solution. [c] Values for one-elec-
tron reduced complexes of 5a and 5b obtained by IR spectroelectro-
chemistry in CH2Cl2 containing 0.1m nBu4NPF6 (see Table 4 and accom-
panying discussion).
compared to [Rh(CO)2Cl] and are in good agreement with
the values observed for the NqMes-supported analogues
(1670 and 1680 cmꢀ1 in KBr, respectively).[58] Interestingly,
the nCO (quinone) for congeners 4b and 5b occurred at
slightly lower energies in the solid-state (1678 and
1686 cmꢀ1, respectively), consistent with the better p-back-
bonding ability of Ir relative to Rh (a consequence of the
slightly lower electronegativity and larger ionic radius of the
former).[100] Because the electron density of the quinone in 1
is influenced by the coordinated MLn units, we conclude
that the metal and quinone are electronically coupled, be-
havior exhibited in monometallic analogues supported by
NqMes.[58] By extension, the electron density of MLn frag-
ments coordinated to 1 should be sensitive to changes in the
quinone (e.g., quinone reduction; vide infra).
calculated to be 2030.6 cmꢀ1. With nav values for both 5a
and 5b on the same scale, their respective TEP values of
2055.9 and 2057.1 cmꢀ1 were found to be nearly identical,
that is,
1 donates electron density equally well to
[Rh(CO)2Cl] and [Ir(CO)2Cl] fragments. Furthermore, the
TEP values for 5a and 5b are within the range observed
in NHC-supported [Rh(CO)2Cl] and [Ir(CO)2Cl] com-
plexes (2046.7–2057.3 cmꢀ1), closely resembling that of
Chem. Eur. J. 2010, 16, 304 – 315
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
309

C. W. Bielawski et al.
4,5-dichloro-1,3-bis(2,6-diisopropylphenyl)imidazolylidene
(2055.1 cmꢀ1).[85,86]
To probe the variation of electron density of 1 in com-
plexes thereof and to gain insight into how this redox-active
ligand affects the electronics of coordinated MLn units, we
performed a series of electrochemical measurements on bis-
azolium [1H2]ACHTUNGTRENNUNG
[OTf]2 and complexes 2–5 (Table 3).[69] Cyclic
Table 3. Electrochemical properties.[a]
Metal oxidation
Ligand reduction
E1/2
[V]
DE
[V]
n
E1/2
[V]
DE
[V]
n
+A
Figure 6. Experimental (A) and simulated (B) EPR spectra of 1 CHTUNGTRENNUNG[CoCp2]
in THF.
[b]
[1H2]
E
–
–
–
0.31
0.12
0.18
0.092
0.088
1.05
0.91
0.94
0.94
1
0.93
0.90
0.92
2[c]
0.44
0.46
1.00
0.12
0.11
0.10
1.93
1.96
1.90
ꢀ0.48
0.083
ꢀ0.40
3[c]
4a[c]
lated with the nCO (quinone) values. All complexes exhibited
only one ligand reduction within the solvent window, and
CA revealed these to be one-electron processes.[69] Quinone
reduction in 2 and 3 occurred at ꢀ0.48 and +0.083 V
(Table 3), the highest and lowest energies of all the com-
plexes analyzed, respectively (Table 3). Complexes 4a and
4b displayed ligand reductions at ꢀ0.40 and ꢀ0.42 V, and
the QBI0/1ꢀ couples for 5a and 5b were measured to be
ꢀ0.31 and ꢀ0.34 V, respectively. Because the quinone reduc-
tion for 3 occurred at the lowest energy of the complexes
studied, we believe its QBI core is the most electron-defi-
4b[c]
5a[c]
5b[c]
0.97
–
–
0.10
–
–
1.94
–
–
ꢀ0.42
ꢀ0.31
ꢀ0.34
0.089
0.13
0.10
[a] DE=EpaꢀEpc. n determined by chronoamperometry. [b] Measure-
ments performed with 1 mm analyte and 0.1m nBu4NPF6 in CH3CN.
[c] Measurements performed with 1 mm analyte and 0.1m nBu4NPF6 in
CH2Cl2.
voltammetry (CV) of [1H2]ACHTNUGTRNEUNG[OTf]2 in CH3CN displayed two
quasireversible peaks at +0.31 V and ꢀ0.66 V versus
SCE,[103] assigned as the first[104] and second quinone reduc-
tions, respectively. Chronoamperometry (CA) confirmed
that each peak corresponded to a one-electron process.[69]
Because the second reduction occurs at ꢀ0.66 V in the dicat-
cient and has zwitterionic character. The ACTHNGUTERNNUG
QBI0/1ꢀ couple ob-
served in 5 is less negative than the analogous couples ob-
served in 4 presumably due to the greater electron-with-
drawing ability of [M(CO)2Cl] versus [MACTHNUTRGNE(NUG cod)Cl]. Further-
more, the reductions in complexes 4b and 5b occurred at
slightly higher energies than in their Rh congeners, reflect-
ing the slightly diminished orbital overlap of Ir with 1. Over-
all, the QBI reduction potentials agree well with the qui-
none stretching frequencies observed for complexes 2–5, for
which increasing nCO (quinone) correlated to lower QBI0/1ꢀ
energies (3>5a>5b>4a>4b>2).
ion of [1H2]ACHTUNGTRENNUNG[OTf]2, compounds comprising neutral 1 should
exhibit only the first reduction within the solvent window,
thus simplifying electrochemical analysis. Furthermore, the
potential of the QBI0/1ꢀ couple should serve as a diagnostic
tool for the variation of ligand electron density in complexes
2–5: values approaching +0.31 V would suggest significant
bis-azolium and, therefore, zwitterionic character.
After analyzing the QBI0/1ꢀ potentials in 2–5, we com-
pared the MLn oxidation potentials to those of BBI-support-
ed analogues to determine the electron-donating character
of 1 and any metal–metal electronic coupling. Complexes 2–
5 displayed only one metal-centered peak corresponding to
two indistinguishable one-electron oxidations (for 2 and 3,
FeII/III; for 4 and 5, MI/II; Table 3), as judged by differential
pulse voltammetry (DPV) and CA.[69] Ferrocene oxidation
in 2 and 3 occurred at +0.44 and +0.46 V (Table 3), respec-
tively, lower than for free FcN3 and FcNCS (+0.59 and
+0.72 V, respectively).[83] Because the FeII/III couple in
FcNCS shifts more upon adduct formation than that of
FcN3, 1 donates more electron density to the FcNCS units,
an assessment supported by the reduction potentials for 2
and 3 (ꢀ0.48 and +0.083 V, respectively). Interestingly, the
Because we were interested in the effects of ligand reduc-
tion on coordinated MLn fragments in complexes 2–5, we
sought to study the nature of the reduced ligand (i.e., 1ꢀ).
Although all attempts to characterize the electrochemical
properties of 1 by CV resulted in rapid decomposition, reac-
tion of 1 with a stoichiometric quantity of CoCp2 in THF af-
forded a compound that exhibited a sharp, isotropic EPR
signal at g=2.008 (Figure 6A).[105] This value is consistent
with those of other semiquinone radical anionic spe-
cies[106,107] and the signal for 1ꢀ matched a simulated isotrop-
ic spectrum centered at g=2.008 (Figure 6B). The absence
of hyperfine coupling indicated that the spin density was lo-
calized primarily on the quinone moiety and leads us to sug-
gest that reduction of the QBI linker in complexes 2–5
could increase the overall electron-donating ability of 1
without significantly altering the electronic topology of the
NHC sites.
[MACTHNUGRTENUNG(cod)Cl] oxidations in 4a and 4b at +1.01 and +0.97 V,
respectively, were similar to those of BBI-based Rh and Ir
complexes.[29] No metal-based oxidations for complex 5a or
5b were observed within the solvent window, a result that is
consistent with the electrochemical behavior of other known
Complexes 2–5 exhibited QBI0/1ꢀ potentials that were in-
dicative of the extent of metal–ligand interaction and corre-
310
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 304 – 315

Quinobis(imidazolylidene)
FULL PAPER
NHC-supported [M(CO)2Cl] complexes (M=Rh and
Ir).[60,85,87] On the basis of these findings, as well as the IR
spectroscopic analyses of 5a and 5b, we conclude that 1 do-
nates electron density in a similar manner to BBI or IMes.
This finding suggests retention of imidazolylidene coordina-
tion chemistry.
and reactivity on the electron density of the metal center,
any catalytic, electronic, or physical properties by extension.
We employed IR spectroelectrochemistry to measure the
shift in nCO (quinone) for 2–4 (in CH2Cl2 containing 0.1m
nBu4NPF6) upon oxidation of the metal centers (Table 4).
Having shown that the electronic properties of 1 and its
bimetallic complexes are consistent with those of other
NHC-supported analogues, we sought to determine how
changing the QBI oxidation state would affect the electron
density at coordinated metal atoms. To study the extent of
these metal–ligand interactions in 2–5, we investigated the
electronic impact of MLn oxidation on the ditopic ligand
and QBI reduction on the coordinated metal centers, re-
spectively, by means of IR spectroelectrochemistry. We hy-
pothesized that oxidation of the coordinated metals in 2–4
would render them more electron deficient and thus with-
draw more electron density from 1, affording an increase in
the nCO (quinone) value (for 2 and 3, see Scheme 7A; for 4a
and 4b, see Scheme 7B). Alternatively, reducing the QBI in
5a or 5b should render the ligand more electron rich and
therefore more donating, resulting in greater electron densi-
ty within the [M(CO)2Cl] fragments and a decrease in the
Table 4. IR Spectroelectrochemical properties.[a]
Metal oxidation[b]
Ligand reduction[c]
nCO
ox
red
nCO
nCO
DnCO
nCO
Dnav
2
3
4a
4b
1662
1694
1681
1678
1681
1704
1689
1687
+19
+10
+8
5a
5b
2088, 2008
2074, 1990
2074, 1994
2060, 1976
ꢀ14
ꢀ15
+9
[a] All values in cmꢀ1. Measurements were performed on 10 mm analyte
red
in CH2Cl2 with 0.1m nBu4NPF6. [b] nCO and nCO correspond to the qui-
none stretching energies in the neutral and oxidized complexes, respec-
ox
red
tively; DnCO =nCO ꢀnCO. [c] nCO and nCO correspond to the metal-
bound carbonyl stretching energies in the neutral and reduced complexes,
respectively; Dnav is the average of the two DnCO values (DnCO
=
red
nCO ꢀnCO).
Oxidizing the ferrocene units in 2 afforded a 19 cmꢀ1 in-
crease in the quinone stretching energy, but shifts of
ꢁ10 cmꢀ1 were observed in 3 and 4. Presumably, the smaller
DnCO (quinone) for 3 versus 2 reflects the linker geometry:
direct metal–quinone interaction is possible across the co-
planar triazene units in 2, but
ꢀ
average nCO (M CO) values nav (see Scheme 7C). This type
of transformation would ultimately represent control over
the TEP and, given the dependence of spectral properties
not the orthogonal NCS moiety
in 3. Because the DnCO (qui-
none) values for 3 and 4 were
similar despite different struc-
tural and electronic features, we
believe they are derived from
the buildup of positive charge
on the molecule. Thus, the DnCO
(quinone)ꢂ9 cmꢀ1 in these sys-
tems arises mainly from the
Coulombic impact of metal oxi-
dation on ligand electron densi-
ty, whereby direct metal–qui-
none coupling was relatively
minor. However, the DnCO (qui-
none) value of 19 cmꢀ1 ob-
served in 2 is consistent with a
relatively significant metal–
ligand electronic interaction. As
observed in a BBI (R=tert-
amyl) system, we believe the
linkage geometry between the
bis-NHC and coordinated metal
atoms governs the metal–ligand
interaction. Specifically, we pro-
pose that orthogonality pre-
Scheme 7. Oxidation of A) Fc (X=N3 and NCS for 2 and 3, respectively) and B) [MACTHNUTRGNE(UNG cod)Cl] units (M=Rh
cludes effective overlap (III
and 3 and 4), while coplanarity
facilitates enhanced it (II and
2).
and Ir for 4a and 4b, respectively) decreases the electron density within the QBI, thus increasing the quinone
bond strength and nCO (quinone). Conversely, reduction of the QBI (C) increases the electron density at the
coordinated metal atoms (M=Rh and Ir for 5a and 5b, respectively), causing a decrease in the carbonyl bond
order and nCO. R=mesityl.
Chem. Eur. J. 2010, 16, 304 – 315
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
311

C. W. Bielawski et al.
Because oxidizing the MLn units in 2–4 diminishes the
QBI electron density and thus increases nCO (quinone), we
reasoned that reducing the quinone moiety in 5a and 5b
would afford more electron rich [M(CO)2Cl] units and
lower the observed average carbonyl stretching energies nav.
Indeed, reduction of 1 decreased nav for 5a and 5b by 14
and 15 cmꢀ1, respectively (see Figure 7 and Table 4), similar
The quinone stretching energies and reduction potentials
in 2–5 correlate with the relative electron-donating/with-
drawing abilities of the MLn fragments, whereby more elec-
tron deficient metals increase the nCO (quinone) value and
the QBI reduction potential (3>5a>5b>4a>4b>2).
Overall, these results indicate that the quinone in 1 was
electronically perturbed upon MLn binding. Further infor-
mation on this interaction was obtained by IR spectroelec-
trochemistry, in which oxidizing the metal centers in 2–4 in-
creased nCO (quinone) in accordance with the extent of
NHC–MLn overlap. Conversely, this interaction was also
used to affect the electronic properties of coordinated
metals, whereby reducing the QBI moiety increased the
electron density on the metal center. Ligand reduction in 5a
and 5b decreased the average nCO values by 14 and 15 cmꢀ1,
and corresponding TEPs by 10 and 12 cmꢀ1, respectively.
The similarity of these shifts is consistent with the observa-
tion that 1 donates equal electron density to [Rh(CO)2Cl]
and [Ir(CO)2Cl].
Collectively, these results suggest that the metal–quinone
electronic interaction in bimetallic complexes of 1 could be
used as a practical means to increase the electron density at
the MLn units through quinone reduction. Because the imi-
dazolylidene units in 1 exhibit similar structure/reactivity to
those found in BBIs and analogous NHCs, despite the
redox-active quinone linker, we believe 1 could be used to
prepare derivatives with electroactive properties. Where
BBIs have facilitated access to phosphorescent bimetallic
complexes,[29] polymeric catalysts,[30] emissive and conductive
polymers,[21,31–33] as well as self-assembled materials,[34,35] sub-
stitution with a redox-active bis-NHC like 1 could afford an-
alogues whose intrinsic properties can be modulated by
changing the ligand oxidation state.
Figure 7. Normalized difference IR spectra at 60 s intervals showing the
shift in nCO upon reduction (Eapp =ꢀ1.0 V) of A) 5a!5aꢀ and B) 5b!
5bꢀ in CH2Cl2 containing 10 mm analyte and 0.1m nBu4NPF6.
to NqMes.[60] These nav shifts are interpreted as indicating
ꢀ
C O bonds weakened by increased electron density on the
metal center, and reflect the greater electron donating abili-
ty of anionic 1ꢀ versus neutral 1. The TEP values for 5aꢀ
and 5bꢀ of 2046 and 2045 cmꢀ1 confirm enhanced ligand-do-
nating ability upon reduction (Table 2), that is, 1ꢀ is among
the most electron donating NHCs (2046.7–2057.3 cmꢀ1).[85,86]
Given that the spectroscopic and structural features of 5a
and 5b are similar to those of other NHC-supported
[M(CO)2Cl] complexes, we conclude that 1 retains the coor-
dination chemistry of the BBI scaffold. Thus, 1 could be
used as a structural replacement for BBI that enables tuning
of the electron density at coordinated metal centers through
changes in quinone redox state and facilitate redox-switcha-
ble control over the intrinsic properties of bis-NHC-support-
ed complexes.
Experimental Section
Materials and methods: Compounds 1–5 were prepared under an N2 at-
mosphere by using dry-box or Schlenk techniques; all other syntheses
were performed under ambient conditions. Azidoferrocene (FcN3) was
prepared as previously described.[83] Dichloromethane and toluene were
distilled from CaH2. THF was distilled from Na/benzophenone. Solvents
were degassed by three consecutive freeze–pump–thaw cycles. All other
reagents were used without further purification. 1H, 13C{1H}, and 19F{1H}
NMR spectra were recorded on a Varian 300, 400, or 500 MHz spectrom-
eter. Chemical shifts d (in ppm) for 1H and 13C NMR are referenced to
tetramethylsilane by using the residual solvent as an internal standard.
For 1H NMR: CDCl3, d=7.24 ppm; CD2Cl2, d=5.32 ppm; CD3CN, d=
1.94 ppm; C6D6, d=7.15 ppm; [D6]DMSO, d=2.49 ppm; [D8]THF, d=
3.58 ppm. For 13C NMR: CDCl3, d=77.0 ppm; CD2Cl2, d=53.8 ppm;
CD3CN, d=118.69 ppm; C6D6, d=128.0 ppm; [D6]DMSO, d=39.5 ppm;
[D8]THF, d=67.6 ppm. Chemical shifts for 19F NMR are referenced to
CFCl3 (0.00 ppm) as an external standard. Coupling constants are ex-
pressed in hertz. FTIR spectra were recorded on a Perkin-Elmer Spec-
trum BX system. For the [M(CO)2Cl] complexes, we attribute the nCO
bands to the carbonyl ligands coordinated cis and trans to their respective
NHC.[108] High-resolution mass spectra (HRMS) were obtained with a
VG analytical ZAB2-E instrument (ESI or CI). EPR spectra were ac-
quired on a Bruker continuous-wave X-band EPR spectrometer and si-
mulated by using the integrated Simfonia software package. Elemental
Conclusions
We have prepared quinobis(imidazolylidene) 1, an electron-
configurable bis-NHC in which two imidazolylidene units
are linked by a redox active p-quinone in excellent yield
after two steps. Treating 1 with FcN3, FcNCS, and [M-
ACHTUNGTRENNUNG(cod)Cl] afforded complexes 2, 3, 4a (M=Rh), or 4b (M=
Ir), respectively. Complexes 4a and 4b were converted to
their [M(CO)2Cl] derivatives 5a and 5b upon reaction with
excess CO(g). Because the NMR spectroscopic and structur-
al features of these bimetallic species were similar to those
of BBI-supported analogues, we believe that the coordina-
tion chemistry exhibited by the imidazolyidene units in BBI
is retained in 1, despite annulation to an electron-withdraw-
ing quinone linker.
312
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 304 – 315

Quinobis(imidazolylidene)
FULL PAPER
analyses were performed by Columbia Analytical Labs (Tucson, AZ) or
Atlantic Microlabs (Norcross, GA).
13C NMR (75 MHz, CDCl3): d=164.9, 150.5, 138.7, 134.5, 131.9, 129.6,
125.0, 104.1, 69.2, 67.4, 63.5, 21.2, 18.1 ppm; IR (KBr): n˜ =1662 cmꢀ1 (C=
O, quinone); electrochemistry: E1/2 =+0.44 V (FeII/III, DE=0.12 V, quasir-
eversible, D0 =8.2ꢃ10ꢀ10 m2 sꢀ1, n=1.93), ꢀ0.48 V (QBI0/ꢀ, DE=0.18 V,
Electrochemistry: Electrochemical experiments were conducted on CH
Instruments Electrochemical Workstations (series 660D and 700B) using
a gas-tight, three-electrode cell under an atmosphere of dry nitrogen.
Cells were equipped with gold working, tungsten counter-, and silver qua-
sireference electrodes. Measurements were performed in dry CH2Cl2 with
0.1m nBu4NPF6 electrolyte and decamethylferrocene (Fc*) internal stan-
dard. Differential pulse voltammetry measurements were performed with
50 mV pulse amplitudes and 2 mV data intervals. Chronoamperometry
experiments were performed with a 25 mm-diameter Au ultramicroelec-
trode as working electrode to enable independent determination of D0
and n by plotting i(t)/iss versus tꢀ1/2 and using the Cottrell equation. Data
deconvolution and fitting were performed with the Origin 8.0 software
package. All reported potentials were determined at 100 mVsꢀ1 scan rate
and referenced to saturated calomel electrode (SCE) by shifting (Fc*)0/+
to ꢀ0.057 V (CH2Cl2) or ꢀ0.044 V (CH3CN).[109]
quasireversible, D0 =8.5ꢃ10ꢀ10 m2 sꢀ1
,
n=0.91); HRMS calcd for
C64H63N10O2Fe2 [M+]: 1115.3839; found: 1115.3834; elemental analysis
(%) calcd for C64H62N10O2Fe2: C 68.94, H 5.61, N 12.56; found: C 69.26,
H 5.48, N 12.69.
AHCTUNGTREG[NUNN (FcNCS)2(1)] (3): A solution of FcNCS (20 mg, 82 mmol) in toluene
(2 mL) was added to a solution of 1 (27 mg, 41 mmol) in toluene (2 mL),
and the resulting mixture was stirred at room temperature, resulting in
the gradual formation of a cloudy, dark brown-red solution. After 16 h,
the solvent was removed under reduced pressure and the remaining resi-
due was washed with Et2O (4ꢃ3 mL). Removal of the residual solvent
under vacuum afforded the desired product (24 mg (21 mmol, 51% yield)
as a dark brown solid. 1H NMR (400 MHz, CDCl3): d=6.93 (brs, 8H),
4.83 (brs, 4H), 3.97 (brs, 4H), 3.75 (s, 10H), 2.33 (s, 24H), 2.23 ppm (s,
12H); 13C NMR (100 MHz, CD2Cl2): d=171.1, 141.2, 135.44, 135.37,
130.0, 129.8, 129.5, 128.8, 126.3, 68.6, 68.4, 65.8, 65.7, 65.5, 65.3, 64.7, 60.3,
31.9, 30.0, 29.7, 29.3, 22.7, 21.2, 21.0, 20.8, 18.7, 14.2, 14.1 ppm; IR (KBr):
1,1’,3,3’-Tetramesitylquinobis(imidazolium) dibromide ([1H2][Br]2): Bro-
manil (500 mg, 1.18 mmol) and N,N’-dimesitylformamidine[66] (1.30 g,
4.63 mmol) were dissolved in CH3CN (25 mL) and the solution heated to
1108C, resulting in gradual formation of a red precipitate. After 24 h, the
mixture was cooled to room temperature. The precipitate was collected
by vacuum filtration, washed with excess CH3CN and Et2O, and dried
under vacuum to afford the desired product (880 mg, 1.07 mmol; 91%
n˜ =1692 cmꢀ1 (C=O, quinone); electrochemistry: E1/2 =+0.46 V (FeII/III
DE=0.11 V, quasireversible, D0 =1.3ꢃ10ꢀ9 m2 sꢀ1
n=1.96), +0.083 V
(QBI0/ꢀ
n=0.94);
,
,
,
DE=0.092 V, quasireversible, D0 =1.3ꢃ10ꢀ9 m2 sꢀ1
,
HRMS calcd for C66H64N6O2S2Fe2 [M+]: 1148.32255; found: 1148.3223;
elemental analysis (%) calcd for C66.25H62.5Cl0.5N6O2S2Fe2 (3·0.25CH2Cl2):
C 68.11, H 5.39, N 7.19; found: C 67.99, H 5.56, N 7.18.
1
yield) as a dark orange solid. M.p. 2968C (decomp); H NMR (500 MHz,
[D6]DMSO): d=10.17 (s, 2H), 7.14 (s, 8H), 2.31 (s, 12H), 2.19 (S, 24H);
13C NMR (100 MHz, [D6]DMSO): d=163.7, 141.2, 134.7, 130.7, 129.3,
128.8, 20.6, 17.8; IR (KBr): n˜ =1708 cmꢀ1 (C=O, quinone); HRMS calcd
for C44H46BrN4O2 [MꢀBr]+: 741.2804; found: 741.2811; elemental analy-
sis (%) calcd for C44H48Br2N4O3 ([1H2][Br]2·H2O): C 62.86, H 5.75, N
6.66; found: C 62.96, H 5.61, N 6.58.
[{Rh
ACHTUNGTENR(NGNU cod)Cl}2(1)] (4a): Bis-NHC 1 (70 mg, 0.10 mmol) and [{Rh-
AHCTUNGTREG(NNNU cod)Cl}2] (51 mg, 0.10 mmol) were dissolved in THF (10 mL). The re-
sulting brown solution was stirred for 12 h at room temperature, during
which the color gradually changed to dark red-brown. The residual sol-
vent was then removed under reduced pressure and the resulting dark
red-brown solid was washed with Et2O to remove any unconsumed [{Rh-
1,1’,3,3’-Tetramesitylquinobis(imidazolium) bis-triflate ([1H2]ACTHNUTRGEN[NUG OTf]2):
After [1H2][Br]2 (100 mg, 122 mmol) was suspended in CH2Cl2 (5 mL),
ACHTUNGRTENNUN(G cod)Cl}2]. Subsequent purification by chromatography (Al2O3, first 3/1
MeOTf (0.10 mL, 0.91 mmol) was added, which resulted in the formation
hexanes/EtOAc to remove impurities then CH2Cl2) afforded the desired
product (80 mg, 69 mmol; 69% yield) as a dark brown powder. 1H NMR
(400 MHz, CDCl3): d=7.03 (s, 4H), 6.95 (s, 4H), 4.52 (brs, 4H), 3.26
(brs, 4H), 2.35 (s, 12H) overlaps 2.32 (d, J=9.6, 12H), 1.90 (d, J=8.0,
12H), 1.76–1.61 (br m, 8H), 1.60–1.46 ppm (br m, 8H); 13C NMR
(125 MHz, CDCl3): d=201.0 (d, J=53.1), 164.5, 139.38, 139.36, 136.9,
136.6, 133.6, 133.5, 133.2, 131.3, 131.2, 130.3, 130.1, 128.4, 128.2, 98.44,
98.39, 68.6, 68.5, 68.4, 32.51, 32.46, 28.03, 28.01, 21.2, 19.9, 19.8, 18.3 ppm;
of
a yellow suspensate. After 2 h, the precipitate was collected by
vacuum filtration, washed with excess CH2Cl2 and Et2O, and then dried
under vacuum to afford the desired product (111 mg, 0.115 mmol; 94%
yield) as a yellow solid. M.p. 3468C (decomp); 1H NMR (400 MHz,
CD3CN): d=9.30 (s, 2H), 7.17 (s, 8H), 2.37 (s, 12H), 2.15 ppm (s, 24H);
13C NMR (75 MHz, CD3CN): d=164.9, 145.7, 143.7, 135.4, 131.8, 130.8,
129.2, 21.1, 17.8 ppm; 19F NMR (376 MHz, CD3CN): d=ꢀ79.3 ppm; IR
(KBr): n˜ =1710 cmꢀ1 (C=O, quinone); electrochemistry: E1/2 =+0.31 V
IR (KBr): n˜ =1680 cmꢀ1 (C=O, quinone); electrochemistry:
E1/2 =+
(first reduction, E=0.12 V, quasireversible, D0 =8.0ꢃ10ꢀ10 m2 sꢀ1, n=
1.00 V (RhI/II, DE=0.10 V, quasireversible, D0 =1.1ꢃ10ꢀ9 m2 sꢀ1, n=1.90),
ꢀ0.40 V (QBI0/ꢀ, DE=0.088 V, quasireversible, D0 =1.1ꢃ10ꢀ9 m2 sꢀ1, n=
0.94); HRMS calcd for C60H68Cl2N4O2Rh2 [Mꢀ]: 1152.2829; found:
1152.2835; elemental analysis (%) calcd for C60H68N4O2Rh2: C 62.45, H
5.94, N 4.86; found: C 62.20, H 5.70, N 4.70.
~
1.05); HRMS calcd for C45H46F3N4O5S [MꢀOTf]+: 811.3148; found:
811.3141; elemental analysis (%) calcd for C46H46F6N4O8S2: C 57.49, H
4.82, N 5.83; found: C 57.38, H 4.72, N 5.90.
1,1’,3,3’-Tetramesitylquinobis(imidazolylidene) (1): [1H2][Br]2 (500 mg,
0.61 mmol) and NaNACHTUNGTRENNUNG(SiMe3)2 (217 mg, 1.19 mmol) were dissolved in
[{IrACHTNURGTNEN(UG cod)Cl}2(1)] (4b): Bis-NHC 1 (35 mg, 53 mmol) and [{IrACHTUNGTRENNUNG(cod)Cl}2]
THF (30 mL) and the mixture was stirred at room temperature. After
30 min, this green mixture was filtered through a 0.4 mm PTFE filter and
then concentrated under vacuum to afford the desired product (320 mg,
0.48 mmol; 79% yield) as a dark green solid. This compound must be
stored in the cold and under an inert atmosphere to prevent decomposi-
(36 mg, 54 mmol) were dissolved in THF (5 mL). The resulting brown-red
mixture was stirred for 16 h at room temperature, during which the color
gradually changed to dark brown. The residual solvent was then removed
under reduced pressure and the resulting dark brown solid was washed
with Et2O to remove any unconsumed [{IrACTHUNRTGNEUNG(cod)Cl}2]. Subsequent purifi-
cation by chromatography (Al2O3, first 3/1 hexanes/EtOAc to remove im-
1
tion. M.p. 828C (decomp); H NMR (500 MHz, C6D6): d=6.92 (2 s, 8H),
2.09 (s, 24H), 2.03 ppm (s, 12H); 13C NMR (100 MHz, [D8]THF): d=
232.6, 168.0, 138.5, 137.0, 134.9, 132.8, 129.6, 21.0, 18.0 ppm; IR (KBr):
n˜ =1676 cmꢀ1 (C=O, quinone); elemental analysis (%) calcd for
C44H48N4O4 [1·2H2O]: C 75.83, H 6.94, N 8.04; found: C 76.32, H 6.88, N
7.84.
purities then CH2Cl2) afforded the desired product (45 mg, 34 mmol;
1
64% yield) as a dark brown-black powder. H NMR (300 MHz, CD2Cl2):
d=7.03 (s, 4H), 7.00 (s, 4H), 4.15 (brs, 4H), 3.04 (brs, 4H), 2.38 (s,
12H), 2.25 (d, J=4.8, 12H), 1.99 (d, J=3.0, 12H), 1.57 ppm (brs, 8H);
13C NMR (125 MHz, CDCl3). d=195.3, 164.9, 139.28, 136.5, 136.2, 133.5,
133.32, 133.28, 131.04, 131.00, 130.3, 130.14, 130.07, 129.9, 129.7, 128.5,
128.3, 128.2, 128.0, 86.2, 86.1, 52.7, 52.5, 52.3, 52.1, 33.35, 33.33, 28.5, 21.4,
21.2, 21.0, 20.0, 19.8, 19.7, 19.6, 18.7, 18.42, 18.40, 18.2 ppm; IR (KBr):
ACHTUNGTRENNUNG[(FcN3)2(1)] (2): A solution of FcN3 (71 mg, 0.28 mmol) in THF (2 mL)
was added to a solution of 1 (93 mg, 0.14 mmol) in THF (3 mL), and the
resulting mixture stirred at room temperature, resulting in gradual forma-
tion of a cloudy, dark brown-red solution. After 16 h, the solvent was re-
moved under reduced pressure and the remaining residue was washed
with Et2O (4ꢃ3 mL). Removal of the residual solvent under vacuum af-
forded the desired product (85 mg, 76 mmol; 54% yield) as a brick red
solid. 1H NMR (400 MHz, CDCl3): d=6.94 (s, 8H), 4.03 (t, J=2.0, 4H),
3.97 (s, 8H), 3.91 (t, J=2.0, 4H), 2.31 (s, 12H), 2.13 ppm (s, 24H);
n˜ =1678 cmꢀ1 (C=O, quinone); electrochemistry:
E
,
1/2 =+0.97 V (IrI/II
n=1.94), ꢀ0.42 V
(QBI0/ꢀ DE=0.089 V, quasireversible, D0 =9.9ꢃ10ꢀ10 m2 sꢀ1
, , n=0.93);
,
DE=0.10 V, quasireversible, D0 =9.8ꢃ10ꢀ10 m2 sꢀ1
HRMS calcd for C60H68ClN4O2Ir2 [M+]: 1297.42838; found: 1297.4271; el-
emental analysis (%) calcd for C60H68Cl2N4O2Ir2: C 54.08, H 5.14, N 4.20;
found: C 53.63, H 5.01, N 4.01.
Chem. Eur. J. 2010, 16, 304 – 315
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
313

C. W. Bielawski et al.
[{Rh(CO)2Cl}2(1)] (5a): Complex 4a (25 mg, 22 mmol) was dissolved in
CH2Cl2 (5 mL) and sparged with CO(g) at room temperature until the
solvent had evaporated. The residue was washed with hexanes (4ꢃ5 mL)
under a static CO atmosphere, then dried under vacuum to afford the de-
sired product (21 mg, 20 mmol; 91% yield) as a light orange powder.
1H NMR (300 MHz, CDCl3): d=6.97 (s, 8H), 2.33 (s, 12H), 2.11 (s,
24H); 13C NMR (75 MHz, CDCl3): d=193.0 (d, J=45.8), 184.0 (d, J=
54.7), 182.0 (d, J=72.7), 164.1, 140.2, 134.1, 132.5, 131.2, 129.8, 21.3, 18.6;
[17] J. A. Mata, M. Poyatos, E. Peris, Coord. Chem. Rev. 2007, 251,
841–859.
[18] M. Poyatos, J. A. Mata, E. Peris, Chem. Rev. 2009, 109, 3677–3707.
[19] F. Boeda, S. P. Nolan, Annu. Rep. Prog. Chem. Sect. B 2008, 104,
184–210.
[20] D. M. Khramov, A. J. Boydston, C. W. Bielawski, Angew. Chem.
2006, 118, 6332–6335; Angew. Chem. Int. Ed. 2006, 45, 6186–6189.
[21] A. J. Boydston, K. A. Williams, C. W. Bielawski, J. Am. Chem. Soc.
2005, 127, 12496–12497.
IR (KBr): n=2089 (IrCO, trans), 2005 (IrCO, cis), 1688 cmꢀ1(C=O, qui-
~
~
none). IR (CH2Cl2, NaCl): n=2089 (IrCO, trans), 2005 (IrCO, cis),
[22] E. Mas-Marzꢅ, J. A. Mata, E. Peris, Angew. Chem. 2007, 119,
3803–3805; Angew. Chem. Int. Ed. 2007, 46, 3729–3731.
[23] M. Poyatos, W. McNamara, C. Incarvito, E. Clot, E. Peris, R. H.
Crabtree, Organometallics 2008, 27, 2128–2136.
[24] F. J.-B. dit Dominique, H. Gornitzka, A. Sournia-Saquet, C. Hem-
mert, Dalton Trans. 2009, 340–352.
1686 cmꢀ1 (C=O, quinone); electrochemistry:
E
1/2 =ꢀ0.31 V (QBI0/ꢀ
,
DE=0.13 V, quasireversible, D0 =9.0ꢃ10ꢀ10 m2 sꢀ1
,
n=0.90); HRMS
calcd for C48H44N4O6Rh2Cl2 [Mꢀ]: 1048.0748; found: 1048.0754; elemental
analysis (%) calcd for C48H44Cl2N4O6Rh2: C 54.93, H 4.23, N 5.34; found:
C 54.70, H 4.20, N 4.90.
[25] M. Raynal, C. S. J. Cazin, C. Vallꢉe, H. Olivier-Bourbigou, P.
Braunstein, Chem. Commun. 2008, 3983–3985.
[{Ir(CO)2Cl}2(1)] (5b): Complex 4b (20 mg, 15 mmol) was dissolved in
CH2Cl2 (5 mL) and sparged with CO(g) at room temperature until the
solvent had evaporated. The residue was washed hexanes (4ꢃ5 mL)
under a static CO atmosphere, then dried under vacuum to afford the de-
sired product (17 mg, 14 mmol; 93% yield) as a light orange powder.
1H NMR (300 MHz, CDCl3): d=6.96 (s, 8H), 2.33 (s, 12H), 2.11 ppm (s,
24H); 13C NMR (75 MHz, CDCl3): d=188.7, 178.8, 167.7, 164.1, 140.4,
134.0, 132.2, 131.0, 129.8, 21.3, 18.6 ppm; IR (KBr): n˜ =2075 (IrCO,
trans), 1989 (IrCO, cis), 1686 cmꢀ1 (C=O, quinone); IR (CH2Cl2, NaCl):
n˜ =2074 (IrCO, trans), 1990 (IrCO, cis), 1688 cmꢀ1 (C=O, quinone); elec-
trochemistry: ꢀ0.34 V (QBI0/ꢀ, DE=0.10 V, quasireversible, D0 =9.3ꢃ
[26] R. B. Bedford, M. Betham, D. W. Bruce, A. A. Danopoulos, R. M.
Frost, M. Hird, J. Org. Chem. 2006, 71, 1104–1110.
[27] B. Liu, W. Chen, S. Jin, Organometallics 2007, 26, 3660–3667.
[28] U. J. Scheele, S. Dechert, F. Meyer, Inorg. Chim. Acta 2006, 359,
4891–4900.
[29] A. G. Tennyson, E. L. Rosen, M. S. Collins, V. M. Lynch, C. W. Bie-
lawski, Inorg. Chem. 2009, 48, 6924–6933.
[30] B. Karimi, P. F. Akhavan, Chem. Commun. 2009, 3750–3752.
[31] D. J. Coady, B. C. Norris, D. M. Khramov, A. G. Tennyson, C. W.
Bielawski, Angew. Chem. Int. Ed. 2009, 121, 5289–5292; Angew.
Chem. Int. Ed. 2009, 48, 5187–5190.
10ꢀ10 m2 sꢀ1
,
n=0.92); HRMS calcd for C48H44N4O6Ir2Cl [M+]:
1193.22024; found: 1193.2188; elemental analysis (%) calcd for
C48H44Cl2N4O6Ir2: C 46.94, H 3.61, N 4.56; found: C 46.80, H 3.30, N
4.50.
[32] A. G. Tennyson, J. W. Kamplain, C. W. Bielawski, Chem. Commun.
2009, 2124–2126.
[33] A. J. Boydston, J. D. Rice, M. D. Sanderson, O. L. Dykhno, C. W.
Bielawski, Organometallics 2006, 25, 6087–6098.
[34] C. Radloff, F. E. Hahn, T. Pape, R. Frçhlich, Dalton Trans. 2009,
7215–7222.
[35] F. E. Hahn, C. Radloff, T. Pape, A. Hepp, Organometallics 2008,
27, 6408–6410.
[36] J. K. Lassila, D. Herschlag, Biochemistry 2008, 47, 12853–12859.
[37] J. G. Zalatan, I. Catrina, R. Mitchell, P. K. Grzyska, P. J. OꢁBrien,
D. Herschlag, A. C. Hengge, J. Am. Chem. Soc. 2007, 129, 9789–
9798.
[38] M. H. Sazinsky, S. J. Lippard, Acc. Chem. Res. 2006, 39, 558–566.
[39] J. S. Valentine, P. A. Doucette, S. Z. Potter, Annu. Rev. Biochem.
2005, 74, 563–593.
Acknowledgements
We are grateful to the NSF (CHE-0645563 and CHE-0749571 for C.W.B.
and J.L.S., respectively), the ONR (N00014-08-1-0729), the Arnold and
Mabel Beckman Foundation, the Sloan Foundation, and the Robert A.
Welch Foundation (F-1621 and F-1018 to C.W.B and J.L.S, respectively)
for their generous financial support. Funding for the EPR spectrometer
was supplied by the NSF (CHE-0639239). C.W.B. is a Cottrell Scholar of
the Research Corporation.
[40] J. S. Valentine, P. J. Hart, Proc. Natl. Acad. Sci. USA 2003, 100,
3617–3622.
[41] A. J. Bard, L. R. Faulkner, Electrochemical Methods: Fundamentals
and Applications, 2nd ed., Wiley, New York, 2001.
[1] C. Schneider, Angew. Chem. 2009, 121, 2116–2118; Angew. Chem.
Int. Ed. 2009, 48, 2082–2084.
[2] A. J. Esswein, A. S. Veige, D. G. Nocera, J. Am. Chem. Soc. 2005,
127, 16641–16651.
[3] N. Marion, S. P. Nolan, Acc. Chem. Res. 2008, 41, 1440–1449.
[4] J. M. Praetorius, C. M. Crudden, Dalton Trans. 2008, 4079–4094.
[5] S. Dꢄez-Gonzꢅlez, S. P. Nolan, Acc. Chem. Res. 2008, 41, 349–358.
[6] S. Wꢆrtz, F. Glorius, Acc. Chem. Res. 2008, 41, 1523–1533.
[7] F. J. Gꢇmez, N. E. Kamber, N. M. Deschamps, A. P. Cole, P. A.
Wender, R. M. Waymouth, Organometallics 2007, 26, 4541–4545.
[8] M. F. Lappert, J. Organomet. Chem. 2005, 690, 5467–5473.
[9] B. E. Ketz, A. P. Cole, R. M. Waymouth, Organometallics 2004, 23,
2835–2837.
[42] N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877–910.
[43] C. K. A. Gregson, V. C. Gibson, N. J. Long, E. L. Marshall, P. J.
Oxford, A. J. P. White, J. Am. Chem. Soc. 2006, 128, 7410–7411.
[44] M. Sꢆßner, H. Plenio, Angew. Chem. 2005, 117, 7045–7048;
Angew. Chem. Int. Ed. 2005, 44, 6885–6888.
[45] H. Plenio, C. Aberle, Chem. Eur. J. 2001, 7, 4438–4446.
[46] H. Plenio, C. Aberle, Y. A. Shihadeh, J. M. Lloris, R. Martꢄnez-
MꢅÇez, T. Pardo, J. Soto, Chem. Eur. J. 2001, 7, 2848–2861.
[47] H. Plenio, C. Aberle, Angew. Chem. 1998, 110, 1467–1470; Angew.
Chem. Int. Ed. 1998, 37, 1397–1399.
[48] A. M. Allgeier, C. A. Mirkin, Angew. Chem. 1998, 110, 936–952;
Angew. Chem. Int. Ed. 1998, 37, 894–908.
[49] I. M. Lorkovic, R. R. Duff, Jr., M. S. Wrighton, J. Am. Chem. Soc.
1995, 117, 3617–3618.
[50] B. Baeza, L. Casarrubios, P. Ramꢄirez-Lopez, M. Gꢇmez-Gallego,
M. A. Sierra, Organometallics 2009, 28, 956–959.
[10] E. Peris, R. H. Crabtree, Coord. Chem. Rev. 2004, 248, 2239–2246.
[11] D. Bourissou, O. Guerret, F. P. Gabbaꢈ, G. Bertrand, Chem. Rev.
2000, 100, 39–91.
[12] A. J. Arduengo III, Acc. Chem. Res. 1999, 32, 913–921.
[13] W. A. Herrmann, K. Kçcher, Angew. Chem. 1997, 119, 2156–2182;
Angew. Chem. Int. Ed. Engl. 1997, 36, 2162–2187.
[51] G. Gachot, P. Pellon, T. Roisnel, D. Lorcy, J. Organomet. Chem.
2009, 694, 2531–2535.
[14] F. E. Hahn, M. C. Jahnke, Angew. Chem. 2008, 120, 3166–3216;
Angew. Chem. Int. Ed. 2008, 47, 3122–3172.
[52] F. Ba, N. Cabon, F. R.-L. Guen, P. L. Poul, N. L. Poul, Y. L. Mest,
S. Golhen, B. Caro, Organometallics 2008, 27, 6396–6399.
[53] S. L. Caldwell, J. B. Gilroy, R. Jain, E. Crawford, B. O. Patrick,
R. G. Hicks, Can. J. Chem. 2008, 86, 976–981.
[15] R. J. Lowry, M. K. Veige, O. Clꢉment, K. A. Abboud, I. Ghiviriga,
A. S. Veige, Organometallics 2008, 27, 5184–5195.
[16] L. H. Gade, S. Bellemin-Laponnaz, Top Organomet. Chem. 2007,
21, 117–157.
314
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 304 – 315

Quinobis(imidazolylidene)
FULL PAPER
[54] D. Guo, J. K. McCusker, Inorg. Chem. 2007, 46, 3257–3274.
[55] T. Hirao, Coord. Chem. Rev. 2002, 226, 81–91.
[56] A. J. Boydston, C. W. Bielawski, Dalton Trans. 2006, 4073–4077.
[57] D. M. Khramov, E. L. Rosen, V. M. Lynch, C. W. Bielawski,
Angew. Chem. 2008, 120, 2299–2302; Angew. Chem. Int. Ed. 2008,
47, 2267–2270.
[84] By indistinguishable, we do not mean identical, merely that the
component processes cannot be resolved: for two identical, nonin-
teracting redox processes, the statistical separation is 36 mV due to
entropic factors.
[85] S. Wolf, H. Plenio, J. Organomet. Chem. 2009, 694, 1487–1492.
[86] R. A. Kelly III, H. Clavier, S. Giudice, N. M. Scott, E. D. Stevens, J.
Bordner, I. Samardjiev, C. D. Hoff, L. Cavallo, S. P. Nolan, Organo-
metallics 2008, 27, 202–210.
[58] M. D. Sanderson, J. W. Kamplain, C. W. Bielawski, J. Am. Chem.
Soc. 2006, 128, 16514–16515.
[59] D. M. Khramov, V. M. Lynch, C. W. Bielawski, Organometallics
2007, 26, 6042–6049.
[60] E. R. Rosen, C. D. Varnado, A. G. Tennyson, D. M. Khramov, J. W.
Kamplain, D. H. Sung, P. J. Creswell, V. M. Lynch, C. W. Bielawski,
Organometallics, DOI: 10.1021/om900698x.
[87] a) S. Leuthꢊußer, D. Schwarz, H. Plenio, Chem. Eur. J. 2007, 13,
7195–7203; b) T. Vorfalt, S. Leuthꢊußer, H. Plenio, Angew. Chem.
2009, 121, 5293–5296; Angew. Chem. Int. Ed. 2009, 48, 5191–5194.
[88] G. Altenhoff, R. Goddard, C. W. Lehmann, F. Glorius, J. Am.
Chem. Soc. 2004, 126, 15195–15201.
[61] D. H. Sung, J. W. Kamplain, M. D. Sanderson, C. W. Bielawski, un-
published results.
[89] A. R. Chianese, X. Li, M. C. Janzen, J. W. Faller, R. H. Crabtree,
Organometallics 2003, 22, 1663–1667.
[62] L. Delaude, Eur. J. Inorg. Chem. 2009, 1681–1699.
[63] V. Nair, S. Bindu, V. Sreekumar, Angew. Chem. 2004, 116, 5240–
5245; Angew. Chem. Int. Ed. 2004, 43, 5130–5135.
[64] B. Bildstein, M. Malaun, H. Kopacka, K. Wurst, M. Mitterbçck,
K.-H. Ongania, G. Opromolla, P. Zanello, Organometallics 1999,
18, 4325–4336.
[65] “Coupling” and “interaction” are used to describe how changing in
the electronic properties of a metal center influences those of the
ligand (and vice versa).
[66] K. M. Kuhn, R. H. Grubbs, Org. Lett. 2008, 10, 2075–2077.
[67] Substituting chloranil for bromanil afforded the corresponding bis-
azolium chloride salt, albeit in slightly lower yield (87%). All at-
tempts to generate the bis-NHC from this precursor have been un-
successful.
[90] M. Iglesias, D. J. Beetstra, B. Kariuki, K. J. Cavell, A. Dervisi, I. A.
Fallis, Eur. J. Inorg. Chem. 2009, 1913–1919.
[91] F. Ullah, M. K. Kindermann, P. G. Jones, J. Heinicke, Organometal-
lics 2009, 28, 2441–2449.
[92] A. Bittermann, P. Hꢊrter, E. Herdtweck, S. D. Hoffmann, W. A.
Herrmann, J. Organomet. Chem. 2008, 693, 2079–2090.
[93] H. M. Peng, R. D. Webster, X. Li, Organometallics 2008, 27, 4484–
4493.
[94] A. Zanardi, R. Corberꢅn, J. A. Mata, E. Peris, Organometallics
2008, 27, 3570–3576.
[95] S. Burling, M. F. Mahon, S. P. Reade, M. K. Whittlesey, Organome-
tallics 2006, 25, 3761–3767.
[96] P. Bazinet, T.-G. Ong, J. S. OꢁBrien, N. Lavoie, E. Bell, G. P. A.
Yap, I. Korobkov, D. S. Richeson, Organometallics 2007, 26, 2885–
2895.
[68] P. D. Vu, A. J. Boydston, C. W. Bielawski, Green Chem. 2007, 9,
1158–1159.
[97] P. R. Campodꢇnico, A. Aizman, R. Contreras, Chem. Phys. Lett.
2009, 471, 168–173.
[69] Supporting Information available: 1H and 13C NMR spectra; syn-
theses of FcNCS and [1H2][Cl]2, ORTEP of [1H2]
crystallographic details, data tables, and CIF files for [1H2]
1–3, 4a, and 5a; electrochemical figures (CV, DPV, CA) for [1H2]-
[OTf]2 and 2–5; spectroelectrochemical spectra for 2–4.
ACHTUNGTRENNUNG
[98] C. Vazquez, J. C. Calabrese, D. A. Dixon, J. S. Miller, J. Org. Chem.
1993, 58, 65–81.
[99] E. C. M. Chen, W. E. Wentworth, J. Chem. Phys. 1975, 63, 3183–
3191.
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
[70] M. S. Jeletic, M. T. Jan, I. Ghiviriga, K. A. Abboud, A. S. Veige,
Dalton Trans. 2009, 2764–2776.
[71] A. J. Boydston, C. S. Pecinovsky, S. T. Chao, C. W. Bielawski, J.
Am. Chem. Soc. 2007, 129, 14550–14551.
[72] H. V. Huynh, Y. Han, J. H. H. Ho, G. K. Tan, Organometallics
2006, 25, 3267–3274.
[73] S. Saravanakumar, A. I. Oprea, M. K. Kindermann, P. G. Jones, J.
Heinicke, Chem. Eur. J. 2006, 12, 3143–3154.
[100] F. A. Cotton, G. Wilkinson, C. A. Murillo, M. Bochmann, Ad-
vanced Inorganic Chemistry, 6th ed., Wiley, New York, 1999.
[101] M. Iglesias, D. J. Beetstra, A. Stasch, P. N. Horton, M. B. Hurst-
house, S. J. Coles, K. J. Cavell, A. Dervisi, I. A. Fallis, Organometal-
lics 2007, 26, 4800–4809.
[102] C. A. Tolman, Chem. Rev. 1977, 77, 313–348.
[103] All potentials are reported relative to saturated calomel electrode
(SCE).
[74] F. E. Hahn, M. C. Jahnke, V. Gomez-Benitez, D. Morales-Morales,
T. Pape, Organometallics 2005, 24, 6458–6463.
[75] H. V. Huynh, J. H. H. Ho, T. C. Neo, L. L. Koh, J. Organomet.
Chem. 2005, 690, 3854–3860.
[76] A. A. Danopoulos, S. Winston, T. Gelbrich, M. B. Hursthouse, R. P.
Tooze, Chem. Commun. 2002, 482–483.
[77] J. A. Wright, A. A. Danopoulos, W. B. Motherwell, R. J. Carroll, S.
Ellwood, J. Saßmannshausen, Eur. J. Inorg. Chem. 2006, 4857–
4865.
[78] X. Hu, I. Castro-Rodriguez, K. Meyer, Organometallics 2003, 22,
3016–3018.
[104] Qualitative evidence that the higher potential peak is a ligand-
based reduction came upon addition of the Fc* reference, where-
upon a deep blue color developed, indicative of Fc* oxidation.
[105] This chemical reduction appears quasireversible: 1ꢀ can be gener-
ated and isolated, then treated with FcBF4 to afford an NMR spec-
trum featuring a signal at d=6.92 ppm, characteristic of neutral
QBI 1.
[106] Y. Song, G. R. Buettner, S. Parkin, B. A. Wagner, L. W. Robertson,
H.-J. Lehmler, J. Org. Chem. 2008, 73, 8296–8304.
[107] H. Wu, D. Zhang, G. Zhang, D. Zhu, J. Org. Chem. 2008, 73, 4271–
4274.
[79] F. E. Hahn, L. Wittenbecher, R. Boese, D. Blꢊser, Chem. Eur. J.
1999, 5, 1931–1935.
[80] D. M. Khramov, A. J. Boydston, C. W. Bielawski, Org. Lett. 2006, 8,
1831–1834.
[81] A. J. Arduengo III, H. V. R. Dias, R. L. Harlow, M. Kline, J. Am.
Chem. Soc. 1992, 114, 5530–5534.
[108] Because no symmetry operation exchanges the cis and trans car-
bonyl ligands on one metal center, they will not be vibrationally
coupled and will therefore exhibit independent stretching modes.
[109] I. Noviandri, K. N. Brown, D. S. Fleming, P. T. Gulyas, P. A. Lay,
A. F. Masters, L. Phillips, J. Phys. Chem. B 1999, 103, 6713–6722.
[82] D. M. Khramov, C. W. Bielawski, J. Org. Chem. 2007, 72, 9407–
9417.
[83] A. G. Tennyson, D. M. Khramov, C. D. Varnado, Jr., P. T. Creswell,
J. W. Kamplain, V. M. Lynch, C. W. Bielawski, Organometallics
2009, 28, 5142–5147.
Received: July 7, 2009
Published online: November 27, 2009
Chem. Eur. J. 2010, 16, 304 – 315
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
315