Indirectly connected bis(N-Heterocyclic Carbene) bimetallic complexes: Dependence of metal-metal electronic coupling on linker geometry

Article abstract of DOI:10.1021/om9004109

Reaction of l,l',3,3'-tetra(tert-amyl)benzobis(imidazolylidene) (1) with 2 equiv of FcN₃ or FcNCS afforded bisadducts [(FcN₃) ₂(1)] (2) or [(FcNCS)₂(1)] (3), respectively (Fc = ferrocene). To the best of our kno

Full text of DOI:10.1021/om9004109

5142 Organometallics 2009, 28, 5142–5147
DOI: 10.1021/om9004109
Indirectly Connected Bis(N-Heterocyclic Carbene) Bimetallic Complexes:
Dependence of Metal-Metal Electronic Coupling on Linker Geometry
Andrew G. Tennyson, Dimitri M. Khramov, C. Daniel Varnado Jr., Philip T. Creswell,
Justin W. Kamplain, Vincent M. Lynch, and Christopher W. Bielawski*
Department of Chemistry and Biochemistry, The University of Texas at Austin, 1 University Station A5300,
Austin, Texas 78712
Received May 18, 2009
Reaction of 1,10,3,30-tetra(tert-amyl)benzobis(imidazolylidene) (1) with 2 equiv of FcN3 or FcNCS
afforded bisadducts [(FcN3)2(1)] (2) or [(FcNCS)2(1)] (3), respectively (Fc = ferrocene). To the best
of our knowledge, these represent the first examples of complexes comprising metals indirectly
connected to the carbene atoms of N-heterocyclic carbenes (NHCs) via their ligand sets. Cyclic and
differential pulse voltammetry indicated that bis(NHC) 1 facilitated significant electronic coupling
between ferrocene centers in 2 (ΔE = 140 mV), but not in 3. We believe the different degrees of
electronic interaction are due to geometric factors: the triazene linker in 2 is nearly coplanar with the
bis(NHC) scaffold, whereas the isothiocyanate linker is orthogonal, as determined by X-ray
crystallography. Employing this “indirect connection” strategy should enable tuning of metal-
metal interactions by simple alteration the organic linker between NHC and MLn fragments rather
than complete redesign thereof. Given that NHC-reactive azide or isothiocyanate groups can be
incorporated into both organic and inorganic compounds, this approach is envisioned to facilitate
access to otherwise inaccessible catalysts and materials.
Introduction
in emissive materials,13,14 to antibiotics15,16 and chemothera-
peutics.17 Variants featuring two or more NHC units have
recently emerged18 and show promise in asymmetric19-21
and tandem catalysis,22,23 and in other applications. An im-
portant subtype of bis(NHC) comprises two NHC moieties
connected via a π-conjugated linker, such as benzobis-
(imidazolylidene) (I),24,25 triazolyldiylidene (II),26 bitriazolyli-
dene (III),27 pyridylbis(imidazolylidene) (IV),28-30 and
Arduengo’s preparation of a stable N-heterocyclic carbene
(NHC) was a landmark achievement in modern synthetic
chemistry.1 Since then, organometallic complexes supported
by NHCs have filled many roles, ranging from catalysts for
polymerization and organic transformations,2-12 to dopants
*Corresponding author. E-mail: bielawski@cm.utexas.edu.
(1) Arduengo III, A. J.; Harlow, R. L.; Kline, M. J. Am. Chem. Soc.
1991, 113, 361–363.
(17) Barnard, P. J.; Berners-Price, S. J. Coord. Chem. Rev. 2007, 251,
1889–1902.
(2) Marion, N.; Nolan, S. P. Acc. Chem. Res. 2008, 41, 1440–1449.
(3) Praetorius, J. M.; Crudden, C. M. Dalton Trans. 2008, 4079–4094.
(18) Hahn, F. E.; Jahnke, M. C. Angew. Chem., Int. Ed. 2008, 47,
3122–3172.
ꢀ
ꢀ
(4) Dı
´
ez-Gonzalez, S.; Nolan, S. P. Acc. Chem. Res. 2008, 41, 349–
(19) Lowry, R. J.; Veige, M. K.; Clement, O.; Abboud, K. A.;
358.
Ghiviriga, I.; Veige, A. S. Organometallics 2008, 27, 5184–5195.
(20) Gade, L. H.; Bellemin-Laponnaz, S. Top. Organomet. Chem.
2007, 21, 117–157.
€
ꢀ
(5) Wurtz, S.; Glorius, F. Acc. Chem. Res. 2008, 41, 1523–1533.
(6) Gomez, F. J.; Kamber, N. E.; Deschamps, N. M.; Cole, A. P.;
Wender, P. A.; Waymouth, R. M. Organometallics 2007, 26, 4541–4545.
(7) Lappert, M. F. J. Organomet. Chem. 2005, 690, 5467–5473.
(8) Ketz, B. E.; Cole, A. P.; Waymouth, R. M. Organometallics 2004,
23, 2835–2837.
(9) Peris, E.; Crabtree, R. H. Coord. Chem. Rev. 2004, 248, 2239–
2246.
(10) Bourissou, D.; Guerret, O.; Gabbaı, F. P.; Bertrand, G. Chem.
(21) Mata, J. A.; Poyatos, M.; Peris, E. Coord. Chem. Rev. 2007, 251,
841–859.
(22) Poyatos, M.; Mata, J. A.; Peris, E. Chem. Rev. 2009, ASAP
article, DOI: 10.1021/cr800501s.
(23) Boeda, F.; Nolan, S. P. Annu. Rep. Prog. Chem., Sect. B 2008,
104, 184–210.
(24) Khramov, D. M.; Boydston, A. J.; Bielawski, C. W. Angew.
Rev. 2000, 100, 39–91.
(11) Arduengo III, A. J. Acc. Chem. Res. 1999, 32, 913–921.
(12) Herrmann, W. A.; Kocher, K. Angew. Chem., Int. Ed. 1997, 36,
2163–2187.
Chem., Int. Ed. 2006, 45, 6186–6189.
(25) Boydston, A. J.; Williams, K. A.; Bielawski, C. W. J. Am. Chem.
Soc. 2005, 127, 12496–12497.
€
ꢀ
(26) Mas-Marza, E.; Mata, J. A.; Peris, E. Angew. Chem., Int. Ed.
(13) Schildknecht, C.; Lennartz, C.; Kahle, K.; Molt, O.; Fuchs, E.;
Moonen, N.; Rudolph, J.; Wagenblast, G.; Amrhein, P.; Proelss, J.;
Muenster, I. Proc. SPIE 2007, 6655, 665502.
(14) Sajoto, T.; Djurovich, P. I.; Tamayo, A.; Yousufuddin, M.; Bau,
R.; Thompson, M. E.; Holmes, R. J.; Forrest, S. R. Inorg. Chem. 2005,
44, 7992–8003.
2007, 46, 3729–3731.
(27) Poyatos, M.; McNamara, W.; Incarvito, C.; Clot, E.; Peris, E.;
Crabtree, R. H. Organometallics 2008, 27, 2128–2136.
(28) dit Dominique, F. J.-B.; Gornitzka, H.; Sournia-Saquet, A.;
Hemmert, C. Dalton Trans. 2009, 340–352.
ꢀ
(29) Raynal, M.; Cazin, C. S. J.; Vallee, C.; Olivier-Bourbigou, H.;
(15) Kascatan-Nebioglu, A.; Panzner, M. J.; Tessier, C. A.; Cannon,
C. L.; Youngs, W. J. Coord. Chem. Rev. 2007, 251, 884–895.
(16) Garrison, J. C.; Youngs, W. J. Chem. Rev. 2005, 105, 3978–4008.
Braunstein, P. Chem. Commun. 2008, 3983–3985.
(30) Bedford, R. B.; Betham, M.; Bruce, D. W.; Danopoulos, A. A.;
Frost, R. M.; Hird, M. J. Org. Chem. 2006, 71, 1104–1110.
r
pubs.acs.org/Organometallics
Published on Web 07/28/2009
2009 American Chemical Society

Article
Organometallics, Vol. 28, No. 17, 2009 5143
pyridazinylbis(imidazolylidene) (V),31,32 which enables elec-
tronic communication from one carbene-bound species to
another (Figure 1). Bis(NHC) I, in particular, has been used
to prepare phosphorescent bimetallic complexes,33 lumines-
cent organic34,35 and conducting organometallic polymers,25,36
polymeric catalysts,37 and self-assembled materials.38
The utility of ditopic NHCs ultimately depends on how
well they facilitate electronic coupling between multiple
ligated metals.39 Electrochemical interactions across brid-
ging ligands can be qualitatively analyzed40 by measuring
the separation in metal-based redox potentials (ΔE) and
subsequently classified according to the Robin-Day
system.41 Previously, we33 and others42 have shown that
bimetallic complexes supported by I can be assigned to class
I or II, depending on the identity of their MLn units. For
MLn = [Ir(COD)Cl] (COD = 1,5-cyclooctadiene)33 or
[Fe(Cp)(CO)I] (Cp = cyclopentadienyl),42 modest interac-
tions (ΔE = 42-80 mV) were observed, identifying these
complexes as class II. However, no interaction is measurable
(ΔE ≈ 0) for MLn = [Ir(ppy)2] (ppy = 2-phenylpyridyl)33 or
[Ru(p-cymene)X2] (X = Br or Cl),42 revealing these com-
plexes as class I systems.
Because the structures of the aforementioned bis(NHC)-
supported bimetallic complexes are nearly identical, the
various degrees in metal-metal electronic coupling likely
arise from differences in how well the HOMO/LUMO
energies of the MLn fragments match those of the NHCs.
For example, the interaction observed between [Ir(COD)Cl]
fragments and not [Ir(ppy)2] could be due to the fact that the
former comprise Ir(I) centers, whose d-orbitals are higher in
energy than those for Ir(III). One obvious drawback to this
feature is that it requires judicious selection of the metal
centers for maximum interaction. Thus, we sought to elim-
inate the dependence on MLn identity and simultaneously
enable tuning of metal-metal interactions in bis(NHC)-
based systems.
Figure 1. Ditopic NHCs linked by a π-conjugated system.
Figure 2. Two different configurations of bis(NHC)-supported
bimetallic complexes.
the bis(NHC) or MLn fragments and therefore their iden-
tities. Because most applications require electronic interac-
tion between specific MLn units, changing these components
may not always be possible. An alternative approach would
be to modify one of the ligands to react with an NHC,
whereby the resulting complex would comprise metals
indirectly connected to the carbene atom of a bis(NHC) via
short, conjugated linkers (-E-, Figure 2). Appropriately
designed linkers would facilitate metal-NHC and metal-
metal electronic coupling, despite increasing their spatial
separation. Tuning of the electronic interactions between
NHC-ligated metals could then be effected by changing the
identity of the linkers while preserving the identities of both
the bis(NHC) and MLn components.
In addition to their ubiquitous organometallic chemistry,
NHCs can react with a variety of electrophilic functional
groups.43-47 We have found that azides couple with NHCs
to afford acyclic triazenes that exhibit excellent electronic
communication across the N3 linker.48,49 Isothiocyanates are
isoelectronic with azides and have a similarly rich chemistry
with NHCs.43 Because ferrocene can be readily monofunc-
tionalized50 and is an excellent electrochemical handle,
we pursued azido- and isothiocyanatoferrocene as suitable
MLn fragments for preparing indirect bimetallic complexes
supported by bis(NHC)s.
To date, NHC-supported bimetallic complexes have fea-
tured bonds directly between the carbene functionalities and
the metal centers (Figure 2). Such complexes are typically
prepared via coordination of the MLn units to a free bis-
(NHC) or transmetalation. Tuning the electronic coupling
across this construct necessitates modifying the electronics of
(31) Liu, B.; Chen, W.; Jin, S. Organometallics 2007, 26, 3660–3667.
(32) Scheele, U. J.; Dechert, S.; Meyer, F. Inorg. Chim. Acta 2006,
359, 4891–4900.
(33) Tennyson, A. G.; Rosen, E. L.; Collins, M. S.; Lynch, V. M.;
Bielawski, C. W. Inorg. Chem. 2009, 48, 6924-6933.
(34) Coady, D. J.; Norris, B. C.; Khramov, D. M.; Tennyson, A. G.;
Bielawski, C. W. Angew. Chem., Int. Ed. 2009, 48, 5187–5190.
(35) Tennyson, A. G.; Kamplain, J. W.; Bielawski, C. W. Chem.
Commun. 2009, 2124–2126.
(36) Boydston, A. J.; Rice, J. D.; Sanderson, M. D.; Dykhno, O. L.;
Bielawski, C. W. Organometallics 2006, 25, 6087–6098.
(43) Delaude, L. Eur. J. Inorg. Chem. 2009, 1681–1699.
(44) Xue, J.; Yang, Y.; Huang, X. Synlett 2007, 1533–1536.
(45) Liu, M.-F.; Wang, B.; Cheng, Y. Chem. Commun. 2006, 1215–
1217.
(46) Nair, V.; Bindu, S.; Sreekumar, V. Angew. Chem., Int. Ed. 2004,
43, 5130–5135.
(37) Karimi, B.; Akhavan, P. F. Chem. Commun. 2009, 3750–3752.
(38) Hahn, F. E.; Radloff, C.; Pape, T.; Hepp, A. Organometallics
2008, 27, 6408–6410.
(39) The terms “coupling” and “interaction” were used to describe
how a change at one metal center influences a similar change at the other.
In the context of this article, electrochemical methods were employed to
study the dependence of one ferrocene oxidation process on the other in
bimetallic complexes supported by a bis(NHC) and used to gain insight
into the factors affecting metal-metal interactions in such systems.
(40) For factors that complicate quantitative analysis by this method,
(47) For earlier examples of using 2-arylthiocarbamoyl imidazoli-
nium salts in 1,3-dipolar cycloadditions, see: (a) Winberg, H. E.;
Coffman, D. D. J. Am. Chem. Soc. 1965, 87, 2776–2777. (b) Regitz, M.;
€
Schlosser, W.; Weber, B.; Liedhegener, A. Liebigs Ann. Chem. 1971, 748, 1–
19.
(48) Khramov, D. M.; Bielawski, C. W. J. Org. Chem. 2007, 72, 9407–
9417.
ꢁ
see: Barriere, F.; Geiger, W. E. J. Am. Chem. Soc. 2006, 128, 3980–3989.
(49) Khramov, D. M.; Bielawski, C. W. Chem. Commun. 2005, 4958–
(41) Zanello, P. Inorganic Electrochemistry: Theory, Practice and
Application; The Royal Society of Chemistry: Cambridge, 2003.
(42) Mercs, L.; Neels, A.; Albrecht, M. Dalton Trans. 2008, 5570–
5576.
4960.
€
(50) Bildstein, B.; Malaun, M.; Kopacka, H.; Wurst, K.; Mitterbock,
M.; Ongania, K.-H.; Opromolla, G.; Zanello, P. Organometallics 1999,
18, 4325–4336.

5144 Organometallics, Vol. 28, No. 17, 2009
Tennyson et al.
Experimental Section
was removed under reduced pressure to afford a light tan
powder. These solids were extracted with 4 ꢀ 1 mL portions
of toluene and then filtered through a 0.20 μm PTFE filter to
remove LiCl and any unreacted starting material. Removal of
solvent in vacuo afforded 74 mg (0.17 mmol, 94% yield) of the
desired product as a light tan powder. Spectral data were
consistent with reported literature values.24 Epa (PhCN) =
+0.37 V (irreversible).
Materials and Methods. [1,10,3,30-Tetra(tert-amyl)benzobis-
(imidazolium)][Cl]2, [1H2][Cl]2, was prepared from 1,2,4,5-tetra-
(tert-amylamino)benzene
as
previously
described.24
Dichloromethane (CH2Cl2) and toluene were distilled from
CaH2. Tetrahydrofuran was distilled from Na/benzophenone.
Solvents were degassed by three consecutive freeze-pump-
thaw cycles. All other reagents were purchased from Aldrich or
1-Bromoferrocene. To a solution of 1,10-dibromoferrocene53
(2.06 g, 6.0 mmol) in 30 mL of THF at -35 °C was dropwise
added n-BuLi (2.2 M in THF, 2.7 mL, 6.0 mmol), maintaining
the internal reaction temperature below -30 °C. After stirring
the reaction for 1 h at -35 °C, water (3.0 mL, 0.17 mmol) was
added dropwise, and the reaction was allowed to warm to room
temperature over the course of 1 h. This mixture was extracted
with 10 mL of Et2O, and the organic layer was then collected,
washed 3 ꢀ 50 mL of water, and dried with Na2SO4. Removal of
solvent under reduced pressure afforded 1.5 g (5.7 mmol, 95%
yield) of the desired product, which was contaminated with 10%
Fc. This material was suitable for subsequent reactions and used
without further purification. Spectral data were consistent with
reported literature values.50
Acros and used without further purification. H and 13C{1H}
1
NMR spectra were recorded using a Varian 400 or 500 MHz
spectrometer. Chemical shifts δ (in ppm) are referenced to
tetramethylsilane using the protio solvent as an internal stan-
dard. For 1H NMR: CDCl3, 7.24 ppm; CD2Cl2, 5.32 ppm;
C6D6, 7.15 ppm. For 13C NMR: CDCl3, 77.0 ppm; CD2Cl2, 53.8
ppm; C6D6, 128.0 ppm. Coupling constants are expressed in
hertz (Hz). High-resolution mass spectra (HRMS) were ob-
tained with a VG analytical ZAB2-E instrument (ESI or CI).
Unless specified otherwise, all compound syntheses and manip-
ulations were carried out under a nitrogen atmosphere using
standard Schlenk or drybox techniques.
Electrochemistry. Electrochemical experiments were con-
ducted on CH Instruments electrochemical workstations
(series 660C and 700B) using a gastight, three-electrode cell
under an atmosphere of dry nitrogen. The cell was equipped
with platinum working and counter electrodes, as well as a silver
wire quasi-reference electrode. Unless specified otherwise, mea-
surements were performed on 1.0 mM solutions of analyte in dry
CH2Cl2 or PhCN with 0.1 M [tetra-n-butylammonium][PF6] as
the electrolyte and decamethylferrocene (Fc*) as the internal
standard. Differential pulse voltammetry measurements were
performed using 50 mV pulse amplitudes and 2 mV data
intervals. Chronoamperometry experiments were performed
using a 25 μm diameter Au ultramicroelectrode as the working
electrode, enabling independent determination of D0 and n by
plotting i(t)/iss vs t-1/2 and using the Cottrell equation.51 Data
deconvolution and fitting were performed using the Origin
8.0 software package. All potentials listed in the text were
determined by cyclic voltammetry at 100 mV s-1 scan rates
and referenced to a saturated calomel electrode (SCE) by
shifting (Fc*)0/+ to -0.057 V (CH2Cl2) or -0.073 V (PhCN).52
1-Azidoferrocene (FcN3). To a mixture of 1-bromoferrocene
(17 g, 90% purity, 58 mmol) and CuCl (7.4 g, 75 mmol) in
400 mL of degassed ethanol was added a solution of NaN3
(8.2 g, 0.13 mol) in 40 mL of H2O, and this suspension was then
allowed to stir at room temperature. After 24 h, the reaction was
filtered through Celite. The filtrate volume was then reduced by
80%, and 1 L of H2O was added. This orange-brown suspension
was then extracted with Et2O (3 ꢀ 200 mL) and the combined
organic fractions were washed with water (2 ꢀ 300 mL). After
drying with Na2SO4, the solvent was removed in vacuo to afford
9.0 g (40 mmol, 69% yield) of the desired product as an orange
solid. The material was found to retain the 10% Fc from
the starting material as an inseparable impurity. 1H NMR
(400 MHz, C6D6): δ 4.10-3.95 (br m, 7H from product and
3H from Fc), 3.65 (t, J = 1.8, 2H). 13C NMR (75 MHz, C6D6): δ
69.3, 68.2, 65.5, 60.9. E1/2 (FeII/III) = +0.59 V (quasi-re-
versible). Spectral data were consistent with reported values.54
1-Isothiocyanatoferrocene (FcNCS). 1-Aminoferrocene50
(0.28 g, 1.4 mmol) and thiocarbonylbisimidazole (0.27 g, tech.
90%, 1.5 mmol) were dissolved in 10 mL of CH2Cl2, and the
resulting mixture was allowed to stir at room temperature. After
4 h, the reaction was concentrated to 5 mL under reduced
pressure, and the resulting residue was then purified by column
chromatography (Al2O3, 9:1 hexanes/EtOAc, Rf = 0.3) to
afford 0.24 g (1.0 mmol, 71% yield) of the desired product as
an orange solid after removal of the residual solvent. 1H NMR
(300 MHz, C6D6): δ 3.93 (t, J = 1.8, 2H), 3.89 (s, 5H), 3.53 (t,
J = 1.8, 2H). 13C NMR (75 MHz, C6D6): δ 132.8, 85.4, 70.1,
66.2, 65.9. E1/2 (FeII/III) = +0.72 V (quasi-reversible). HRMS
calcd for C11H10NSFe [M+]: 243.9883. Found: 243.9887. Spec-
tral data were consistent with reported values.55,56
1,2,4,5-Tetra(tert-amylamino)benzene.
To
[1,3-bis(2,6-
diisopropylphenyl)imidazolium][Cl] (90 mg, 0.21 mmol), Pd-
(OAc)2 (30 mg, 0.13 mmol), and NaOtBu (35 mg, 0.36 mmol)
was added 4 mL of toluene, and the resulting solution was then
stirred at room temperature to generate an active coupling
catalyst. After 10 min, this clear orange solution was added to
a mixture of 1,2,4,5-tetrabromobenzene (1.95 g, 4.95 mmol),
tert-amyl amine (1.82 g, 20.9 mmol), and NaOtBu (2.33 g,
24.2 mmol) in 40 mL of toluene, and the reaction was then
heated to 110 °C for 16 h, resulting in the gradual formation of a
dark brown solution with a beige suspensate. This mixture was
allowed to cool to room temperature and then filtered through
Celite under a cone of nitrogen. The filtrate was concentrated to
dryness under reduced pressure, and the solids were then taken
up in a minimal amount of hexanes, filtered through Celite, and
dried in vacuo to afford 1.5 g (3.6 mmol, 73% yield) of the
desired product as a golden powder. Spectral data were con-
sistent with reported literature values.24
[(FcN3)2(1)] (2). To a solution of FcN3 (290 mg, 1.3 mmol) in
2 mL of THF was added a solution of 1 (250 mg, 0.57 mmol) in
3 mL of THF, and the reaction was allowed to stir at room
temperature, resulting in the gradual formation of a cloudy,
dark red solution. After 16 h, the solvent was removed under
reduced pressure. The resulting dark red residue was then
washed with 4 ꢀ 3 mL portions of Et2O and then dried in vacuo
to afford 320 mg (0.36 mmol, 63% yield) of the desired product
1,10,3,30-Tetra(tert-amyl)benzobis(imidazolylidene) (1). To a
suspension of [1H2][Cl]2 (94 mg, 0.18 mmol) in 2 mL of THF
was slowly added lithium diisopropylamide (0.388 M, 0.46 mL,
0.18 mmol). The reaction was then stirred at room tempe-
rature, which resulted in a gradual change from a light beige
suspension to a dark brown solution. After 15 min, the solvent
(53) Dong, T.-Y.; Lai, L.-L. J. Organomet. Chem. 1996, 509, 131–134.
€
(54) Koster, S. D.; Dittrich, J.; Gasser, G.; Husken, N.; Castaneda, I.
C. H.; Jios, J. L.; Vedova, C. O. D.; Metzler-Nolte, N. Organometallics
€
~
ꢀ
2008, 27, 6326–6332.
(55) Knox, G. R.; Pauson, P. L.; Willison, D.; Solcaniova, E.; Toma,
ꢂꢀ
ꢀ
(51) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Funda-
mentals and Applications, 2nd ed.; John Wiley & Sons, Inc.: Hoboken, 2001.
(52) Noviandri, I.; Brown, K. N.; Fleming, D. S.; Gulyas, P. T.; Lay,
P. A.; Masters, A. F.; Phillips, L. J. Phys. Chem. B 1999, 103, 6713–6722.
S. Organometallics 1990, 9, 301–306.
(56) Silva, M. E. N. P. R. A.; Pombeiro, A. J. L.; da Silva, J. J. R. F.;
Herrmann, R.; Deus, N.; Bozak, R. E. J. Organomet. Chem. 1994, 480,
81–90.

Article
as
Organometallics, Vol. 28, No. 17, 2009 5145
a
brick red solid. 1H NMR (400 MHz, CD2Cl2):
δ 7.70 (s, 2H), 4.68 (t, J = 1.8, 4H), 4.28 (t, J = 1.8, 4H), 4.19
(s, 10H), 2.24 (q, J = 7.2, 8H), 1.80 (s, 24H), 0.76 (t, J = 7.2,
12H). 13C NMR (100 MHz, CD2Cl2): δ 161.0, 127.9, 106.9, 98.9,
69.2, 67.2, 64.5, 63.0, 34.2, 29.2, 8.3. HRMS calcd for
C48H64N10Fe2 [M+]: 892.4014. Found: 892.4008. E1/2 (FeII/III):
+0.52 V (quasi-reversible), +0.66 V (quasi-reversible). CCDC:
729959.57
[(FcNCS)2(1)] (3). To a solution of FcNCS (110 mg, 0.43
mmol) in 2 mL of THF was added a solution of 1 (95 mg, 0.22
mmol) in 2 mL of THF, and the reaction mixture was allowed to
stir at room temperature, resulting in the gradual formation of a
light orange suspension. After 16 h, the solvent was removed
under reduced pressure. The resulting light orange residue was
then washed with 4 ꢀ 3 mL portions of Et2O and dried in vacuo
to afford 180 mg (0.195 mmol, 89% yield) of the desired product
as a light orange solid. 1H NMR (500 MHz, CDCl3): δ 8.31 (s,
2H), 5.07 (m, 4H), 4.24-4.22 (m, 14H), 2.40-2.06 (m, 32H),
0.94 (t, J = 7.6, 12H). 13C NMR (100 MHz, CDCl3): δ 166.8,
153.5, 128.2, 104.3, 103.0, 69.4, 68.4, 65.9, 65.5, 33.9, 28.0, 27.0,
9.0. HRMS calcd for C50H66N6Fe2 [M+]: 926.3495. Found:
926.3397. E1/2 (FeII/III): +0.32 V (quasi-reversible). D0: (8.0 (
0.3) ꢀ 10-6 cm2 s-1, n = 2.1 ( 0.1. CCDC: 729958.57
Figure 3. ORTEP diagram showing 50% probability thermal
ellipsoids and selected atom labels for 2. Hydrogen atoms have
been omitted for clarity. Selected bond lengths (A) and angles
(deg): N1-C1, 1.381(4); N1-C3, 1.408(4); N2-C1, 1.372(5);
N2-C2, 1.410(4); N3-C1, 1.326(5); N3-N4, 1.356(5); N4-N5,
1.269(4); N5-C15, 1.401(6); C2-C3, 1.398(5); C2-C4,
1.376(5); C1-N3-N4, 112.5(3); N1-C1-N2, 109.7(3); N4-
N5-C15, 112.1(4); N3-N4-N5, 110.3(3); N1-C1-N3-N4,
31.3(6); N3-N4-N5-C15, 173.0(5); N4-N5-C15-C16,
174.8(6). The Fe-centroid distances for the functionalized
and unfunctionalized Cp rings are 1.632 and 1.637 A, respec-
tively.
Results and Discussion
Addition of 124 to 2 equiv of FcN3 or FcNCS (Fc =
ferrocene) in THF afforded the adduct [(FcN3)2(1)] (2) or
[(FcNCS)2(1)] (3) in 63% or 89% yield, respectively
13
(Scheme 1). The diagnostic C NMR signals for the 2,20-
positions shifted from δ = 230.2 ppm (C6D6) in 1 to
161.8 ppm (CD2Cl2) in 2 and 166.8 ppm (CDCl3) in 3.
Whereas the 4,40-protons in 2 resonated at nearly the same
frequency as in 1 (δ = 7.70 vs 7.89 ppm in C6D6 and CD2Cl2,
respectively), the same signals in 3 were significantly down-
field (δ = 8.31 ppm, CDCl3), suggestive of greater positive
charge within the benzobis(imidazole) core. For example,
the analogous protons in [1H2][Cl]2, featuring a formally
dicationic benzobis(imidazolium), are observed at δ = 8.55
ppm (DMSO-d6).24
Figure 4. ORTEP diagram showing 50% probability thermal
ellipsoids and selected atom labels for 3. Hydrogen atoms have
been omitted for clarity. Selected bond lengths (A) and angles
(deg): N1-C1, 1.361(5); N1-C2, 1.406(5); N2-C1, 1.355(5);
N2-C3, 1.407(5); N3-C15, 1.282(5); N3-C16, 1.420(5); S1-
C15, 1.716(5); C1-C15, 1.517(5); C2-C3, 1.405(5); C2-C4,
1.394(5); C3-C4, 1.372(5); N1-C1-N2, 110.4(3); C1-C15-
N3, 116.3(4); C15-N3-C16, 118.6(4); S1-C15-N3, 135.3(3);
N1-C1-C15-N3, 92.6(5); C17-C16-N3-C15, 126.5(6). The
Fe-centroid distances for the functionalized and unfunctiona-
lized Cp rings are 1.645 and 1.642 A, respectively.
Scheme 1. Syntheses of Adducts 2 and 3a
and subsequent analysis by X-ray diffraction confirmed their
structures (see Figures 3 and 4). The bis(NHC) cores of 2 or 3
did not show significant deviation from metric parameters
observed in other benzobis(imidazolylidene)s.24 However,
the structural features of the FcN3 units changed signifi-
cantly upon incorporation into 2. Most notably, the N3-
N4-N5 angle of 110.3(3)° was substantially different from
the near-linear value of 172.94(12)° observed in free FcN3.58
Although the N4-N5 and N5-C15 bond lengths of 1.269(4)
and 1.401(6) A in 2 differed only slightly from the corre-
sponding values in free FcN3 (1.2388(12) and 1.4326(13) A,
respectively),58 the N3-N4 distance was markedly elongated
to 1.356(5) A from 1.1325(14) A, indicative of a significant
decrease in bond order. Additionally, the N4-N5-C15-
C16 torsion angle of 174.8(6)° revealed greater triazene-Cp
coplanarity than in free FcN3 (157.82(16)°). Whereas the
a (i) 2 equiv of FcN3 (63% yield) or (ii) 2 equiv of FcNCS (89% yield).
Single crystals of 2 and 3 were grown independently by
vapor diffusion of hexanes into saturated CH2Cl2 solutions,
(57) X-ray crystal structure data were collected for 2 and 3 (CCDC
729959 and 729958, respectively) and deposited with the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ,
UK.
(58) Walla, P.; Arion, V. B.; Brinker, U. H. J. Org. Chem. 2006, 71,
3274–3277.

5146 Organometallics, Vol. 28, No. 17, 2009
Tennyson et al.
14 or imidazole moieties featuring significantly shorter
iron-iron distances (<9 A).50,60
In contrast to 2, only one quasi-reversible oxidation was
observed in the CV of 3 in CH2Cl2 at +0.32 V (Figure 5C).
This potential is significantly lower than that in free FcNCS
(+0.72 V), suggesting that the iron centers in 3 are more
electron rich. This observation supports the zwitterionic
formulation of 3, wherein the electron density of the FcNCS
is increased, and complements the 1H NMR data described
above. Closer analysis of the oxidation of 3 by DPV and
subsequent deconvolution confirmed the presence of only
one peak (Figure 5D). Chronoamperometry enabled deter-
mination of the diffusion coefficient D0 = (8.0 ( 0.3) ꢀ 10-6
cm2 s-1 and number of electrons n = 2.1 ( 0.1 associated
with this oxidation process. Thus, the two ferrocene-based
oxidations in 3 are occurring at indistinguishable poten-
tials.61
The absence of electronic coupling between the ferrocene
centers in 3 is believed to be due to the near-orthogonality of
the isothiocyanate linkers. In contrast, the triazene linkers in
2 are rotated only 31.3(6)° from the bis(NHC) plane, result-
ing in better overlap between the NHC moieties and the Cp
rings and, as a result, the iron centers. Consequently, the
electron density at the metals in 2 will be more sensitive to
variations in the electron density within the connecting bis-
(NHC) unit than those in 3. To support this notion, the
oxidation of the first ferrocene unit in 2 was found to increase
the energy of the second oxidation by 140 mV, but the
oxidation of one ferrocene unit in 3 did not influence the
other, presumably due to the interruption of conjugation.
Figure 5. (A) CV of 2 (100 mV s-1 scan rate). (B) DPV of the
oxidation processes in 2 (), 50 mV amplitude, 2 mV interval)
showing deconvolution into two peaks (gray lines) and the fitted
peak (black line). (C) CV of 3 (100 mV s-1 scan rate). (D) DPV of
the oxidation process in 3 (), 50 mV amplitude, 2 mV interval)
showing deconvolution into one peak (black line). All measure-
ments were performed in CH2Cl2 solution containing 1 mM
analyte and 0.1 M [Bu4N][PF6] under an inert atmosphere.
Potentials have been adjusted to SCE.
triazene linkers are nearly coplanar with the bis(NHC) unit
in 2 (N1-C1-N3-N4 = 31.3(6)°), the isothiocyanate moi-
eties in 3 are nearly perpendicular (N1-C1-C15-N3 =
92.6(5)°). Similarly, the C15-N3-C16-C17 angle of
126.5(6)° in 3 is much more acute than the analogous linker
in 2 (N4-N5-C15-C16 = 174.3(6)°). Collectively, these
results suggest that electronic coupling between metals
across the triazene linkers in 2 should be feasible, whereas
the perpendicular isothiocyanate linkers in 3 might interrupt
conjugation and preclude any such interactions.
Having elucidated the structures of 2 and 3, we sought to
investigate the electronic interaction between the ferrocene
centers therein via electroanalytical methods. Cyclic voltam-
metry (CV) of 2 in CH2Cl2 revealed two overlapping quasi-
reversible oxidations at +0.52 and +0.66 V vs SCE
(Figure 5A).59 Because these values are similar to the oxida-
tion potential for FcN3 (+0.59 V),56 we conclude that the
electronic environment at the iron centers is not significantly
perturbed upon adduct formation. Differential pulse vol-
tammetry (DPV) of these oxidations clarified the two over-
lapping peaks, and the associated ΔE of these two redox
processes was determined to be 140 mV (Figure 5B). Decon-
volution of the DPV spectrum revealed that the two under-
lying peaks have nearly identical areas, and thus each peak
corresponds to a one-electron oxidation.
Summary and Conclusions
Adapting NHC-azide48,49 and -isothiocyanate43 coupling
chemistries for reaction with bis(NHC) 1 afforded adducts 2
and 3, respectively. To the best of our knowledge, these are
the first examples of complexes featuring metals indirectly
connected to the carbene atoms of NHCs. Electrochemical
analysis of 2 reveals significant electronic coupling (ΔE =
140 V) between ferrocene units (Fe Fe > 16.5 A), but 3
3 3 3
displays no measurable metal-metal interaction. We sur-
mise that the geometry of the linker (i.e., coplanar with vs
perpendicular to the NHC moieties) determines the degree of
this interaction, whereby increasing coplanarity improves
the metal-metal electronic coupling across the bis(NHC)
scaffold. These geometry-dependent results build on pre-
vious observations that the match between HOMO/LUMO
energies of the MLn fragment and the bis(NHC) apparently
influences the extent of interaction.33,42 This novel “indirect
connection” strategy should enable tuning of metal-metal
interactions via alteration of the organic linkers between
NHC and MLn fragments while preserving their identities.
Given the abundance of NHC-reactive functional groups
and relative ease of incorporating them into various sub-
strates,43,46 tuning of the electronic coupling between metals
across bis(NHC) scaffolds could be effected by simply
using different linkers (e.g., replacing isothiocyanate with
azide). This approach is envisioned to facilitate access to new
The degree of metal-metal interaction in 2 is impressive
given that the metals are indirectly connected to the carbene
atoms of 1 via triazene linkers and are separated by greater
than 16.5 A. Other bimetallic complexes supported by bis-
(NHC) I exhibit much weaker interactions (ΔE = 42-80
mV) despite being having direct metal-NHC connections and
metal-metal distances of less than 11 A.42 Comparable
ferrocene-ferrocene interactions (ΔE = 90-126 mV) to 2
have been previously observed, albeit when linked by group
(60) Jones, S. C.; Barlow, S.; O’Hare, D. Chem.;Eur. J. 2005, 11,
4473–4481.
(61) By indistinguishable, we do not mean identical, merely that the
component processes cannot be resolved. Even for two identical, non-
interacting redox processes, the statistical separation is 36 mV due to
entropic factors. See ref 51, p 246.
(59) All potentials are reported relative to SCE.

Article
Organometallics, Vol. 28, No. 17, 2009 5147
bimetallic catalysts and organometallic materials, and allow
optimization of existing systems.
would also like to thank our reviewers for their insightful
comments regarding systems featuring multiple redox pro-
cesses and the associated electrochemical considerations.
Acknowledgment. We are grateful to the National
Science Foundation (CHE-0645563), the Office of Naval
Research (N00014-08-1-0729), the Robert A. Welch Foun-
dation (F-1621), and the Arnold and Mabel Beckman
Foundation for their generous financial support. C.W.B.
is a Cottrell Scholar of Research Corporation. We
Supporting Information Available: 1H and 13C NMR spectra
of FcN3, FcNCS, and compounds 2 and 3, electrochemical data
for FcN3, FcNCS, 1 and 3, and X-ray crystallographic data,
including tables and CIFs, for 2 and 3. This material is available
free of charge via the Internet at http://pubs.acs.org.