Redox-active N-heterocyclic carbenes: Design, synthesis, and evaluation of their electronic properties

Article abstract of DOI:10.1021/om900698x

To investigate effects of redox-active functional groups on the coordination chemistry and electronic properties of N-heterocyclic carbenes (NHCs), we prepared a series of complexes comprising 1,3- diferrocenylimidazolylidene and -benzimidazolylidene (1 a

Full text of DOI:10.1021/om900698x

Organometallics 2009, 28, 6695–6706 6695
DOI: 10.1021/om900698x
Redox-Active N-Heterocyclic Carbenes: Design, Synthesis, and
Evaluation of Their Electronic Properties
Evelyn L. Rosen, C. Daniel Varnado Jr., Andrew G. Tennyson, Dimitri M. Khramov,
Justin W. Kamplain, Daphne H. Sung, Philip T. Cresswell, 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 August 6, 2009
To investigate effects of redox-active functional groups on the coordination chemistry and
electronic properties of N-heterocyclic carbenes (NHCs), we prepared a series of complexes
comprising 1,3-diferrocenylimidazolylidene and -benzimidazolylidene (1 and 2, respectively),
1-ferrocenyl-3-methyl- and 1,3-diphenyl-5-ferrocenylbenzimidazolylidene (3 and 4, respectively),
N,N0-diisobutyldiaminocarbene[3]ferrocenophane (FcDAC), and 1,3-dimesitylnaphthoquinoimida-
zolylidene (NqMes) ligands and coordinated [Ir(COD)Cl] (COD=1,5-cyclooctadiene), [Ir(CO)2Cl],
and [M(CO)5] (M=Cr, Mo, W) units. The coordination chemistry of the aforementioned NHCs was
investigated by X-ray crystallography, and their electronic properties were studied by NMR and IR
spectroscopy, as well as electrochemistry. No significant variation in νCO was observed among metal
carbonyl complexes supported by 2-4 and FcDAC, indicating that the number (one vs two) of redox-
active groups, the location (N atom vs backbone) of the redox-active group, and carbene ring
identities (strained six-membered, nonaromatic vs five-membered, heteroaromatic) did not have a
significant effect on ligand electron-donating ability. Because the shifts in νCO upon oxidation of 1-3
and FcDAC were similar in magnitude but opposite in sign to NqMes, we conclude that the
enhancement or attenuation of ligand donating is primarily Coulombic in origin (i.e., due to the
molecule acquiring a positive or negative charge).
Introduction
with redox-active ligands are shown in Figure 1.2,3,8,9 It is
important to note that these complexes, as well as many
others, contain neutral ligands that are capable of adopting
cationic states as well as those that transition to anionic
states. Additionally, there are ligand classes where multiple
oxidation states can be accessed.10-13
Oxidation or reduction of a substitutionally inert ligand
can tune the electronic properties of a complex without
changing the oxidation state of the metal or steric para-
meters.14-16 These effects can be quantified via ligation to a
transition metal-carbonyl complex followed by measure-
ment of the carbonyl stretching frequency (νCO) as a function
of ligand oxidation state. For example, the complexes shown
Transition metal-based complexes containing redox-
active ligands have been used in diverse areas,1 including
catalysis,2,3 sensing,4,5 and optical materials.6,7 The appeal of
ligands with redox-active functional groups stems from their
abilities to change the electronic properties of a metal with-
out the need for further synthetic modifications. In many
cases, these ligands exhibit reversible redox processes and
therefore enable “switchable” control of the electronic prop-
erties of metals by chemical redox agents or bulk elec-
trolysis. Examples of previously reported metal complexes
*Corresponding author. E-mail: bielawski@cm.utexas.edu.
(1) Allgeier, A. M.; Mirkin, C. A. Angew. Chem., Int. Ed. 1998, 37,
894–908.
(2) Gregson, C. K. A.; Gibson, V. C.; Long, N. J.; Marshall, E. L.;
Oxford, P. J.; White, A. J. P. J. Am. Chem. Soc. 2006, 128, 7410–7411.
(3) Lorkovic, I. M.; Duff, R. R., Jr.; Wrighton, M. S. J. Am. Chem.
Soc. 1995, 117, 3617–3618.
(9) Ringenberg, M. R.; Kokatam, S. L.; Heiden, Z. M.; Rauchfuss,
T. B. J. Am. Chem. Soc. 2008, 130, 788–789.
(10) Boyer, J. L.; Cundari, T. R.; DeYonker, N. J.; Rauchfuss, T. B.;
Wilson, S. R. Inorg. Chem. 2009, 48, 638–645.
(11) Zanello, P.; Corsini, M. Coord. Chem. Rev. 2006, 250, 2000–
2022.
(4) Sibert, J. W.; Forshee, P. B.; Lynch, V. Inorg. Chem. 2005, 44,
8602–8609.
(5) Beer, P. D. Chem. Soc. Rev. 1989, 18, 409–450.
(6) Liu, Z.; Yasseri, A. A.; Lindsey, J. S.; Bocian, D. F. Science 2003,
302, 1543–1545.
(7) Mori, H.; Miyoshi, E. J. Theor. Comput. Chem. 2005, 4, 333–344.
(8) McNitt, K. A.; Parimal, K.; Share, A. I.; Fahrenbach, A. C.;
Witlicki, E. H.; Pink, M.; Bediako, D. K.; Plaisier, C. L.; Le, N.;
Heeringa, L. P.; Griend, D. A. V.; Flood, A. H. J. Am. Chem. Soc.
2009, 131, 1305–1313.
(12) Ward, M. D.; McCleverty, J. A. J. Chem. Soc., Dalton Trans.
2002, 275–288.
(13) Downs, H. H.; Buchanan, R. M.; Pierpont, C. G. Inorg. Chem.
1978, 18, 1736–1740.
(14) Higuchi, M.; Ikeda, I.; Hirao, T. J. Org. Chem. 1997, 62, 1072–
1078.
(15) Yeung, L. K.; Kim, J. E.; Chung, Y. K.; Rieger, P. H.; Sweigart,
D. A. Organometallics 1996, 15, 3891–3897.
(16) Lorkovic, I. M.; Wrighton, M. S.; Davis, W. M. J. Am. Chem.
Soc. 1994, 116, 6220–6228.
r
2009 American Chemical Society
Published on Web 11/10/2009
pubs.acs.org/Organometallics

6696 Organometallics, Vol. 28, No. 23, 2009
Rosen et al.
class of ligands for a broad range of transition metals.33-43
One reason for this attention is their similar coordination
chemistry to phosphines.38,44 However, due to their strong
electron-donating abilities and unique steric parameters,
NHCs often afford complexes that are relatively stable
toward ligand displacement and show significantly improved
catalyst activities.45,46 Furthermore, NHCs can be synthe-
sized from readily available starting materials47 using ex-
tensive metal complexation methodology via free NHCs48 or
transmetalation via Ag-NHC intermediates.49 Despite
these advantages, surprisingly few redox-active NHCs and
metal complexes thereof have been reported in the literature
(see Figure 3 for examples).50-59 Bildstein and co-workers
reported60,61 the first NHC containing N-ferrocenyl groups
(A and B), but the ability of these redox-active moieties to
modulate electronic properties was not explored in detail.
Although a [W(CO)5] complex supported by 1-ferrocenyl-
3-methylbenzimidazolylidene has been prepared, its electro-
nic properties were not studied as a function of the ferrocene
oxidation state. A variety of NHC-supported complexes
bearing N-ferrocenyl substituents have been reported, but
Figure 1. Examples of metal complexes supported by redox-
active ligands.
in Figure 2 display shifts in carbonyl stretching energy
(ΔνCO) of up to 35 cm-1 1,17-19
Although correlation be-
.
tween ΔνCO and the characteristics of transition metal com-
plexes is not well-understood,20,21 ligands with a ΔνCO value
in the mid to upper end of this range often result in a
measurable influence. For example, a Re carbonyl complex
containing a 1,10-bis(diphenylphosphanyl)cobaltocene li-
gand shows ΔνCO values from -11 to -17 cm-1 11
As a
.
ꢀ
(33) Dıez-Gonzalez, S.; Nolan, S. P. Top. Organomet. Chem. 2007, 21,
47–82.
(34) Kantchev, E. A. B.; O’Brien, C. J.; Organ, M. G. Angew. Chem.,
result, nucleophilic attack on the carbonyl groups was
enhanced 200-fold upon oxidation of the metal center.
Despite their unique properties, many classes of redox-
active ligands suffer from a number of practical and funda-
mental limitations. A large majority of redox-active ligands
are either bi- or multidentate; monodentate analogues ap-
pear to be much less common.18,22,23 Although multidentate
ligands can afford stable complexes, the range of possible
geometries is often restricted. Other limitations include
difficulties associated with metal coordination and/or an
inability to impart significant electronic changes at a metal
center upon changing the oxidation state of its redox-active
ligand. Furthermore, some complexes exhibit irreversible
electrochemical behavior, inhibiting widespread use in sen-
sing and catalysis.24,25 The development of new classes of
redox-active ligands could overcome these limitations.
Over the past 50 years,26-31 N-heterocyclic carbenes
(NHCs)32 and related compounds have become a versatile
Int. Ed. 2007, 46, 2768–2813.
(35) Marion, N.; Navarro, O.; Mei, J.; Stevens, E. D.; Scott, N. M.;
Nolan, S. P. J. Am. Chem. Soc. 2006, 128, 4101–4111.
(36) Navarro, O.; Marion, N.; Oonishi, Y.; Kelly, R. A., III; Nolan,
S. P. J. Org. Chem. 2006, 71, 685–692.
(37) Cavell, K. J.; McGuinness, D. S. Coord. Chem. Rev. 2004, 248,
671–681.
(38) Peris, E.; Crabtree, R. H. Coord. Chem. Rev. 2004, 248, 2239–
2246.
(39) Herrmann, W. A. Angew. Chem., Int. Ed. 2002, 41, 1290–1309.
(40) Hillier, A. C.; Grasa, G. A.; Viciu, M. S.; Lee, H. M.; Yang, C.;
Nolan, S. P. J. Organomet. Chem. 2002, 653, 69–82.
(41) Jørgensen, M.; Lee, S.; Liu, X.; Wolkowski, J. P.; Hartwig, J. F.
J. Am. Chem. Soc. 2002, 124, 12557–12565.
(42) Gibby, J. L. J. E.; Farnworth, M. V.; Tekavec, T. N. J. Am.
Chem. Soc. 2002, 124, 15188–15189.
(43) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18–29.
(44) Herrmann, W. A.; Mihalios, D.; Ofele, K.; Kiprof, P.; Belmed-
jahed, F. Chem. Ber. Recl. 1992, 125, 1795–1799.
(45) Scott, N. M.; Clavier, H.; Mahjoor, P.; Stevens, E. D.; Nolan,
S. P. Organometallics 2008, 27, 3181–3186.
(46) Jafarpour, L.; Stevens, E. D.; Nolan, S. P. J. Organomet. Chem.
(17) Yang, K.; Bott, S. G.; Richmond, M. G. Organometallics 1995,
14, 2387–2394.
(18) Miller, T. M.; Ahmed, K. J.; Wrighton, M. S. Inorg. Chem. 1989,
28, 2347–2355.
(19) Kotz, J. C.; Nivert, C. L.; Lieber, J. M.; Reed, R. C.
2000, 606, 49–54.
(47) Peris, E. Top. Organomet. Chem. 2007, 21, 83–116.
(48) Herrmann, W. A.; Kocher, K. Angew. Chem., Int. Ed. 1997, 36,
2163–2187.
(49) Wang, H. M. J.; Lin, I. J. B. Organometallics 1998, 17, 972–975.
(50) Arduengo, A. J., III; Iconaru, L. I. Dalton Trans. 2009, 6903–
6914.
(51) Arduengo, A. J., III; Tapu, D.; Marshall, W. J. J. Am. Chem.
Soc. 2005, 127, 16400–16401.
€
J. Organomet. Chem. 1975, 91, 87–95.
(20) Allgeier, A. M.; Slone, C. S.; Mirkin, C. A.; Liable-Sands, L. M.;
Yap, G. P. A.; Rheingold, A. L. J. Am. Chem. Soc. 1997, 119, 550–559.
(21) Sassano, C. A.; Mirkin, C. A. J. Am. Chem. Soc. 1995, 117,
11379–11380.
(22) Blackmore, K. J.; Lal, N.; Ziller, J. W.; Heyduk, A. F. Eur. J.
Inorg. Chem. 2009, 735–743.
(52) Arduengo, A. J., III; Bannenberg, T. P.; Tapu, D.; Marshall,
W. J. Chem. Lett. 2005, 34, 1010–1011.
(53) Arduengo, A. J., III; Tapu, D.; Marshall, W. J. Angew. Chem.,
(23) Shu, C.-F.; Wrighton, M. S. Inorg. Chem. 1988, 27, 4326–4329.
(24) Guerro, M.; Pham, N. H.; Massue, J.; Bellec, N.; Lorcy, D.
Tetrahedron 2008, 64, 5285–5290.
(25) Traill, P. R.; Bond, A. M.; Wedd, A. G. Inorg. Chem. 1994, 33,
5754–5760.
(26) Bourissou, D.; Guerret, O.; Gabbaı, F. P.; Bertrand, G. Chem.
Int. Ed. 2005, 44, 7240–7244.
(54) Arduengo, A. J., III; Tapu, D.; Marshall, W. J. Angew. Chem.
2005, 117, 7406–7410.
(55) Arduengo, A. J., III; Bannenberga, T. P.; Tapu, D.; Marshall,
W. J. Tetrahedron Lett. 2005, 46, 6847–6850.
(56) Labande, A.; Daran, J.-C.; Manoury, E.; Poli, R. Eur. J. Inorg.
Chem. 2007, 1205–1209.
Rev. 2000, 100, 39–91.
(27) Cardin, D. J.; Cetinkaya, B.; Cetinkaya, E.; Lappert, M. F.
J. Chem. Soc., Dalton Trans. 1973, 514–522.
(57) Bertogg, A.; Camponovo, F.; Togni, A. Eur. J. Inorg. Chem.
2005, 347–356.
€
(28) Wanzlick, H.-W.; Schonherr, H.-J. Angew. Chem., Int. Ed. Engl.
(58) Gischig, S.; Togni, A. Organometallics 2004, 23, 2479–2487.
(59) Seo, H.; Kim, B. Y.; Lee, J. H.; Park, H.-J.; Son, S. U.; Chung,
Y. K. Organometallics 2003, 22, 4783–4791.
1968, 7, 141–142.
€
(29) Ofele, K. J. Organomet. Chem. 1968, 12, P42–P43.
€
(30) Winberg, H. E.; Coffman, D. D. J. Am. Chem. Soc. 1965, 87,
2776–2777.
(31) Breslow, R. J. Am. Chem. Soc. 1957, 79, 1762–1763.
(32) Arduengo, A. J., III; Harlow, R. L.; Kline, M. J. Am. Chem. Soc.
1991, 113, 361–363.
(60) Bildstein, B.; Malaun, M.; Kopacka, H.; Wurst, K.; Mitterbock,
M.; Ongania, K.-H.; Opromolla, G.; Zanello, P. Organometallics 1999,
18, 4325–4336.
(61) Bildstein, B.; Malaun, M.; Kopacka, H.; Ongania, K.-H.; Wurst,
K. J. Organomet. Chem. 1999, 572, 177–187.

Article
Organometallics, Vol. 28, No. 23, 2009 6697
Figure 2. Selected examples of redox-switchable ligands. The absolute values of the change in metal-carbonyl stretching frequencies
(ΔνCO) upon oxidation of the neutral ligands shown are indicated. R = Ph; M = Cr, Mo, W; Fc = ferrocene.
Figure 3. Examples of ferrocene-functionalized NHC-based ligands (n = 0, 1, or 2. X = donor group).
these efforts have primarily focused on either chiral modi-
fication for asymmetric catalysis or capitalizing on the steric
parameters of ferrocene rather than its electrochemical prop-
erties (e.g., C),56,58,59 with few exceptions (D in Figure 3,
FcDAC in Figure 4).62,63
We recently introduced two classes of NHCs containing
complementary redox-active moieties. As shown in Figure 4,
FcDAC is an N,N0-disubstituted diaminocarbene[3]ferro-
cenophane, whereas NqMes features a naphthoquinone
annulated to 1,3-dimesitylimidazolylidene. The former can
undergo oxidation (ferrocene T ferrocenium), whereas the
latter can undergo reduction (quinone T semiquinone).
Expanding the role of NHCs beyond simple phosphine
analogues, we showed that both of these ligands can adopt
multiple oxidation states and alter the electron density of
coordinated metals.63-65
Herein we report efforts to further explore how the
incorporation of redox-active functionalities into NHCs
influences their coordination chemistry and electronic prop-
erties. As summarized in Figures 4 and 5, we sought to
investigate the effects of various structural characteristics:
(I) imidazolylidene vs benzimidazolylidene frameworks (e.g.,
1 vs 2); (II) incorporating one vs two ferrocene units (e.g., 2 vs
3); (III) attaching a ferrocene unit to an NHC nitrogen atom
vs to the backbone of a benzimidazolylidene (e.g., 3 vs 4);
(IV) aromatic vs nonaromatic cyclic NHCs (e.g., 2 vs
FcDAC); and (V) oxidizable vs reducible redox-active func-
tionalities (e.g., 2 vs NqMes). Fortunately, methodology has
already been developed by us and others for preparing metal
complexes of 1,57 FcDAC,63 and NqMes.64 To access the
remaining complexes, we improved the overall syntheses for
260 and 361 and developed a route to 4. Given that the
spectroscopic, structural, and electrochemical properties of
NHC-supported complexes of [Ir(COD)Cl] (COD = 1,5-
cyclooctadiene) have been extensively explored,66-70 analo-
gous complexes containing 1-4, FcDAC, and NqMes were
studied. Conversion of these complexes to [Ir(CO)2Cl] ana-
logues was accomplished via treatment with CO(g), enabling
more direct measurement of the metal electron density and
ligand donicity via IR spectroscopy.66,67 We also pursued
[M(CO)5] (M = Cr, Mo, W) complexes supported by 1-4,
FcDAC, and NqMes, given that analogous complexes sup-
ported by NHCs have been studied extensively by X-ray
crystallography and IR spectroscopy.61,71-85 Overall, we
found that the NHC coordination chemistry and donating
ability are most strongly influenced by the NHC backbone
€
(68) (a) Leuthausser, S.; Schwarz, D.; Plenio, H. Chem.;Eur. J.
€
2007, 13, 7195–7203. (b) Vorfalt, T.; Leuthausser, S.; Plenio, H. Angew.
Chem., Int. Ed. 2009, 48, 5191–5194.
(69) Altenhoff, G.; Goddard, R.; Lehmann, C. W.; Glorius, F. J. Am.
Chem. Soc. 2004, 126, 15195–15201.
(70) Chianese, A. R.; Li, X.; Janzen, M. C.; Faller, J. W.; Crabtree,
R. H. Organometallics 2003, 22, 1663–1667.
(71) Kim, S.; Choi, S. Y.; Lee, Y. T.; Park, K. H.; Sitzmann, H.;
Chung, Y. K. J. Organomet. Chem. 2007, 692, 5390–5394.
€
(72) Frey, G. D.; Ofele, K.; Krist, H. G.; Herdtweck, E.; Herrmann,
W. A. Inorg. Chim. Acta 2006, 359, 2622–2634.
(73) Nonnenmacher, M.; Kunz, D.; Rominger, F.; Oeser, T.
J. Organomet. Chem. 2005, 690, 5647–5653.
(74) Sakurai, H.; Sugitani, K.; Moriuchi, T.; Hirao, T. J. Organomet.
Chem. 2005, 690, 1750–1755.
(75) Hahn, F. E.; Langenhahn, V.; Meier, N.; Lugger, T.; Fehlhammer,
€
W. P. Chem.;Eur. J. 2003, 9, 704–712.
€
(76) Tafipolsky, M.; Scherer, W.; Ofele, K.; Artus, G.; Pedersen, B.;
Herrmann, W. A.; McGrady, G. S. J. Am. Chem. Soc. 2002, 124, 5865–5880.
(77) Bolm, C.; Kesselgruber, M.; Raabe, G. Organometallics 2001,
21, 707–710.
(78) Hahn, F. E.; Langenhahn, V.; Pape, T. Chem. Commun. 2005,
5390–5392.
€
(79) Hahn, F. E.; Paas, M.; Van, D. L.; Frohlich, R. Chem.;Eur. J.
2005, 11, 5080–5085.
(80) Krahulic, K. E.; Enright, G. D.; Parvez, M.; Roesler, R. J. Am.
(62) Varnado, C. D., Jr.; Lynch, V. M.; Bielawski, C. W. Dalton
Trans. 2009, 7253-7261.
(63) Khramov, D. M.; Rosen, E. L.; Lynch, V. M.; Bielawski, C. W.
Angew. Chem., Int. Ed. 2008, 47, 2267–2270.
(64) Sanderson, M. D.; Kamplain, J. W.; Bielawski, C. W. J. Am.
Chem. Soc. 2005, 127, 4142–4143.
(81) Hahn, F. E.; Plumed, C. G.; Munder, M.; Lugger, T. Chem.;
Eur. J. 2004, 10, 6285–6293.
(82) Bildstein, B.; Malaun, M.; Kopacka, H.; Ongania, K.-H.; Wurst,
K. J. Organomet. Chem. 1998, 552, 45–61.
€
€
Chem. Soc. 2006, 128, 16514–16515.
(65) Siemeling, U.; Farber, C.; Bruhn, C. Chem. Commun. 2009, 98–
100.
(66) Wolf, S.; Plenio, H. J. Organomet. Chem. 2009, 694, 1487–1492.
(67) Kelly, R. A., III; Clavier, H.; Giudice, S.; Scott, N. M.; Stevens,
E. D.; Bordner, J.; Samardjiev, I.; Hoff, C. D.; Cavallo, L.; Nolan, S. P.
Organometallics 2008, 27, 202–210.
(83) Ku, R.-Z.; Huang, J.-C.; Cho, J.-Y.; Kiang, F.-M.; Reddy,
K. R.; Chen, Y.-C.; Lee, K.-J.; Lee, J.-H.; Lee, G.-H.; Peng, S.-M.;
Liu, S.-T. Organometallics 1999, 18, 2145–2154.
(84) Herrmann, W. A.; Goossen, L. J.; Artus, G. R. J.; Kocher, C.
Organometallics 1997, 16, 2472–2477.
(85) Herrmann, W. A.; Kocher, C.; Goossen, L. J.; Artus, G. R. J.
€
€
Chem.;Eur. J. 1996, 2, 1627–1636.

6698 Organometallics, Vol. 28, No. 23, 2009
Rosen et al.
electrochemical information available.66-70 Additionally,
the [Ir(COD)Cl] unit subsequently can be converted to
[Ir(CO)2Cl] upon reaction with CO(g), enabling further
interrogation of the complex’s carbene ligand via IR spectro-
scopic analysis.66,67 Two routes were employed for the
preparation of the desired [Ir(COD)Cl] complexes supported
by 1-4, FcDAC, and NqMes, depending on the stability of
the free NHC. For NHCs that dimerize or decompose in free
form (i.e., 1, 3, and 4), deprotonation of the azolium was
achieved with Ag2O followed by transmetalation49 to
[Ir(COD)(μ-Cl)]2 (route A, Scheme 2, 99-100% overall
yield). Alternatively, the free NHCs 2, FcDAC, and NqMes
were successfully generated upon treatment with KOtBu or
NaHMDS and then coordinated48 to iridium (route B,
56-100% overall yield). For (1-4)a and 6a, the 13C NMR
chemical shifts for the 2-position were observed from δ =
182.2 to 194.6 ppm (Table S5), within the range observed for
other NHC-supported [Ir(COD)Cl] complexes (δ = 179.6-
208.2 ppm).68,90-93 Similarly, the analogous signal observed
in the 13C NMR spectrum of 5a (δ = 213.2 ppm) was
comparable to that observed in its rhodium congener (δ =
225.8 ppm),63 but substantially different from (1-4)a and 6a.
Because the carbene nucleus is strongly influenced by the
ring system comprising it, apparent by the markedly distinct
13C NMR chemical shift of 5a vs (1-4)a and 6a, we expected
that the structural features of 5a would be similarly unique.
Single crystals of (1-3)a and (5,6)a were obtained
and analyzed by X-ray diffraction to obtain their respective
structural parameters and enable comparison to crys-
tallographically characterized analogues.94 The iridium-
NHC distances observed in these complexes (2.022(10) A
for 1a, 2.020(5) A for 2a, 2.030(7) A for 3a, 2.068(3) A
Figure 4. Complementary redox-active carbenes (R=isobutyl,
Ph, or 2-Ad;63,65 Mes= 2,4,6-trimethylphenyl).
Figure 5. Redox-active NHC-based ligands studied herein.
(i.e., imidazolylidene vs benzimidazolylidene vs nonaro-
matic). We also discovered that the tunability of these
redox-active ligands is primarily determined by Coulombic
factors: addition of a positive charge reduces NHC donating
ability, whereas a negative charge enhances it. Surprisingly,
these effects are largely independent of the location of the
redox-active functionality as long as it is in reasonably close
proximity to the carbene.
Results and Discussion
Synthesis of N-Heterocyclic Carbene Precursors. Palla-
dium-catalyzed coupling of aminoferrocene60 with 1,2-di-
bromobenzene (Scheme 1A), followed by cyclization with
triethylorthoformate and HCl(aq), afforded 1,3-diferroce-
nylbenzimidazolium chloride, [2H][Cl], in excellent overall
yield (91%). Alternatively, aminoferrocene underwent nu-
cleophilic aromatic substitution with 2-fluoronitrobenzene
(Scheme 1B), which, following subsequent formylative cycli-
zation and alkylation, produced 1-ferrocenyl-3-methylbenz-
imidazolium iodide, [3H][I], in 85% overall yield. To access
[4H][Cl], we prepared 1,2-dichloro-4-diazoniumbenzene tet-
rafluoroborate86 as a suitable precursor. Reaction of this salt
with ferrocene under acidic conditions afforded 1,2-di-
chloro-4-ferrocenylbenzene (Scheme 1C), which was then
subjected to Pd-catalyzed aryl amination and cyclization to
give 1,3-diphenyl-5-ferrocenylbenzimidazolium chloride,
[4H][Cl], in 20% overall yield after three steps. The spectral
properties for [2H][Cl] and [3H][I] obtained by these modi-
fied procedures were identical to literature values.60,61 Simi-
for 5a, 2.033(5)
A for 6a) were consistent with
other NHC-supported [Ir(COD)Cl] complexes (1.99 to
2.091 A).67,68,95-100 In these complexes, the Ir-COD bond
distances trans to the NHC range from 2.134 to 2.227 A,
whereas the distances for the Ir-COD bonds in the cis
position range from 2.081 to 2.155 A. The corresponding
lengths in (1-3)a and 6a (2.189(4)-2.191(7) A) agreed well
with these structural features. Only 5a appeared to deviate
significantly from the other [Ir(COD)Cl] complexes, judging
by 13C NMR shifts and N-C-N angles (121.9(3)° for 5a vs
102.8(8)-105.3(4)° for (1-3)a and 6a). In contrast, com-
plexes (1-4)a and 6a shared highly conserved spectros-
copic and structural features. We conclude that the distinct
(90) Jeletic, M. S.; Jan, M. T.; Ghiviriga, I.; Abboud, K. A.; Veige,
A. S. Dalton Trans. 2009, 2764–2776.
(91) Peng, H. M.; Webster, R. D.; Li, X. Organometallics 2008, 27,
4484–4493.
ꢀ
(92) Zanardi, A.; Corberan, R.; Mata, J. A.; Peris, E. Organometallics
1
larly, the H and 13C NMR chemical shifts for the 2-posi-
2008, 27, 3570–3576.
tion in [4H][Cl] (δ = 10.47 and 142.2 ppm, respectively, in
DMSO-d6) were consistent with analogous compounds re-
ported in the literature.87-89
Synthesis of Iridium Complexes. With the precursors to
NHCs 1-4, FcDAC, and NqMes in hand, we pursued their
respective [Ir(COD)Cl] (COD = 1,5-cyclooctadiene) com-
plexes, given the abundance of spectroscopic, structural, and
ꢀ
(93) Mas-Marza, E.; Mata, J. A.; Peris, E. Angew. Chem., Int. Ed.
2007, 46, 3729–3731.
(94) See Supporting Information.
(95) Kownacki, I.; Kubicki, M.; Szubert, K.; Marciniec, B. J. Orga-
nomet. Chem. 2008, 693, 321–328.
(96) Chen, D.; Banphavichit, V.; Reibenspies, J.; Burgess, K. Orga-
nometallics 2007, 26, 855–859.
(97) Iglesias, M.; Beetstra, D. J.; Stasch, A.; Horton, P. N.; Hurst-
house, M. B.; Coles, S. J.; Cavell, K. J.; Dervisi, A.; Fallis, I. A.
Organometallics 2007, 26, 4800–4809.
(86) Doyle, M. P.; Bryker, W. J. J. Org. Chem. 1979, 44, 1572–1574.
(87) Vu, P. D.; Boydston, A. J.; Bielawski, C. W. Green Chem. 2007, 9,
1158–1159.
(88) Khramov, D. M.; Bielawski, C. W. J. Org. Chem. 2007, 72, 9407–
9417.
(89) Khramov, D. M.; Boydston, A. J.; Bielawski, C. W. Angew.
(98) Herrmann, W. A.; Baskakov, D.; Herdtweck, E.; Hoffmann,
S. D.; Bunlaksananusorn, T.; Rampf, F.; Rodefeld, L. Organometallics
2006, 25, 2449–2456.
ꢀ
(99) Viciano, M.; Poyatos, M.; Sanau, M.; Peris, E.; Rossin, A.;
ꢀ
Ujaque, G.; Lledos, A. Organometallics 2006, 25, 1120–1134.
ꢀ
(100) Vicent, C.; Viciano, M.; Mas-Marza, E.; Sanau, M.; Peris, E.
ꢀ
Chem., Int. Ed. 2006, 45, 6186–6189.
Organometallics 2006, 25, 3713–3720.

Article
Organometallics, Vol. 28, No. 23, 2009 6699
Scheme 1. Syntheses of Carbene Precursorsa
a iPr HCl = 1,3-bis(2,6-diisopropylphenyl)imidazolium chloride
3
has been previously reported,63 whose 13C chemical shift
(δ=212.8 ppm in CDCl3) was comparable to that observed
in 5b (δ = 202.4 ppm in CDCl3).94
Scheme 2. Syntheses of (1-6)a and (1-6)b
Interestingly, the 13C NMR signals for the 2-position in
(1-6)b (180.7-202.4 ppm) were upfield of their respective
signals found in (1-6)a (182.2-213.2 ppm),94 indicating
greater shielding of the carbene nuclei in the [Ir(CO)2Cl] vs
[Ir(COD)Cl] complexes. Replacing COD with more elec-
tron-withdrawing, π-acidic carbonyl ligands should decrease
the overall metal electron density. As such, a coordinated
NHC should donate more electron density to the metal. As
the donation increases, the metal-NHC interaction will
shift from a simple metal-carbene σ interaction to one
that features more multiple-bond character. This pheno-
menom effectively results in an increase in the shielding of
the carbene nucleus and an upfield shift in the 13C NMR
signal.
Crystal structures were obtained for (1-3)b and (5,6)b,
thus enabling comparison to other [Ir(CO)2Cl] complexes
supported by NHCs (see Figures 6-10).94 The iridium-NHC
bond lengths in (1-3)b and (5,6)b (2.121(3) A for 1b, 2.080(4)
A for 2b, 2.071(3) A for 3b, 2.121(3) for 5b, 2.071(4) A for
6b)94 agreed well with the values in analogous compounds
(2.065-2.122 A).67,69,97 Additionally, the metal-carbonyl
distances in (1-3)b and (5,6)b (trans: 1.877(4)-1.900(5) A;
cis: 1.827(4)-1.888(4) A) were comparable to values ob-
served in related [Ir(CO)2Cl] complexes (1.854-1.915 and
1.86-1.912 A for the trans and cis positions relative to the
NHC, respectively).67,69,97 In general, the metal-carbon
bonds trans to the NHC will be longer than those cis due
to the strong trans effect of NHCs. Whereas the N1-C1-N2
angles in (1-3)b and 6b varied minimally (105.5(3)-
106.1(3)°), the corresponding value in 5b was substantially
more obtuse (122.4(2)°). Similarly, the chemical shifts for
(1-4)b and 6b were conserved, with the signal for 5b
significantly downfield (202.4 ppm for 5b vs 180.7-186.9
ppm for (1-4)b and 6b).94
Route A. For 1: (i) 0.5 equiv of Ag2O, 1,2-dichloroethane, 15 h, 94%;
(ii) CH2Cl2, 6 h, 99%. For 3: (i) CH2Cl2, 16 h, 91%; (ii) CH2Cl2, 12 h,
99%. For 4: (i) CH2Cl2, 2 h, 100%; (ii) THF, 7 h, 60 °C, 100%. Route B.
For 2: (iþii) KOtBu, THF, 12 h, 60 °C, 100%. For FcDAC: (i) Na-
HMDS, toluene, 20 min; (ii) 2 h, 71%. For NqMes: (i) NaHMDS, THF,
20 min; (ii) 12 h, 56%. Complexes (1-6)a were obtained by reaction with
0.5 equiv [Ir(COD)Cl]2. The carbonylation reactions were performed by
purging with CO(g) (see Experimental Section for details). Unless
specified otherwise, all reactions were performed at ambient tempera-
ture. L=1-4, FcDAC (5), or NqMes (6).
coordination chemistry of 5a compared to (1-4)a and 6a
reflects the fundamental structural differences between the
six-membered, nonaromatic, cyclic FcDAC and five-mem-
bered, heteroaromatic 1-4 and NqMes.
To gain additional insight into the electronic properties
and donicity of 1-4, FcDAC, and NqMes via IR spectro-
scopic analysis, we prepared their metal carbonyl complexes.
The [Ir(CO)2Cl] complexes (1-6)b were obtained in excellent
yields (88-100%) by bubbling 1 atm of CO(g) through
CH2Cl2 solutions of the respective [Ir(COD)Cl] complexes
(1-6)a (Scheme 2). Complexes (1-4)b and 6b exhibited a
range of values (δ=180.7-186.9 ppm, CDCl3),94 consistent
with known NHC-supported [Ir(CO)2Cl] complexes (δ =
178.0-202.3 ppm).67 A [Rh(CO)2Cl] complex of FcDAC

6700 Organometallics, Vol. 28, No. 23, 2009
Rosen et al.
Figure 8. ORTEP diagram showing 50% probability thermal
ellipsoids and selected atom labels for 3b. Hydrogen atoms have
been omitted for clarity. Selected bond lengths (A) and angles
(deg): Ir-C2, 1.877(4); Ir-C3, 1.827(4); C2-O2, 1.141(5);
C3-O3, 1.142(5).
Figure 6. ORTEP diagram showing 50% probability thermal
ellipsoids and selected atom labels for 1b. Hydrogen atoms have
been omitted for clarity. Selected bond lengths (A) and angles
(deg): Ir-C1, 2.089(6); Ir-C2, 1.892(8); Ir-C3, 1.859(11); C2-
O2, 1.143(10); C3-O3, 1.091(12); N1-C1-N2, 105.9(5).
Figure 9. ORTEP diagram showing 50% probability thermal
ellipsoids and selected atom labels for 5b. Hydrogen atoms have
been omitted for clarity. Selected bond lengths (A) and angles
(deg): Ir-Cl, 2.3632(8); Ir-C1, 2.121(3); Ir-C2, 1.891(3);
Ir-C3, 1.888(4); C2-O2, 1.136(4); C3-O3, 1.024(4); N1-
C1-N2, 122.4(2); N1-C1-Ir-Cl, 78.71(18).
Figure 7. ORTEP diagram showing 50% probability thermal
ellipsoids and selected atom labels for 2b. Hydrogen atoms have
been omitted for clarity. Selected bond lengths (A) and angles
(deg): Ir-C1, 2.080(4); Ir-C2, 1.894(4); Ir-C3, 1.870(8);
C2-O2, 1.127(5); C3-O3, 1.131(8); N1-C1-N2, 106.1(3).
Synthesis of Group 6 Complexes. To gain additional insight
into the electron-donating ability of these NHCs comprising
redox-active moieties, we prepared a series of [M(CO)5]
complexes72 supported by 1-4, FcDAC, and NqMes. Photo-
lysis of the homoleptic carbonyl complexes [M(CO)6] (M =
Cr, Mo, W) in THF afforded [M(CO)5(THF)],101 where the
coordinated THF was readily displaced with free NHC
(Scheme 3). Unfortunately, the FcDAC complexes were
very unstable and only the tungsten congener 5e was ob-
tained. In contrast, 2c-e and 6c-e were purified via column
chromatography and were found to be bench stable for days.
The range of 13C NMR chemical shifts for the 2-positions of
Figure 10. ORTEP diagram showing 50% probability thermal
ellipsoids and selected atom labels for 6b. Hydrogen atoms have
been omitted for clarity. Selected bond lengths (A) and angles
(deg): Ir-C1, 2.071(4); Ir-C2, 1.900(5); Ir-C3, 1.843(5);
C2-O2, 1.120(5); C3-O3, 1.117(5); N1-C1-N2, 105.5(3).
(101) Schubert, U.; Friedrich, P.; Orama, O. J. Organomet. Chem.
1978, 144, 175–179.

Article
Organometallics, Vol. 28, No. 23, 2009 6701
Scheme 3. Synthesis (top) and Structures (bottom) of Various
Group 6 Complexesa
a For these complexes, L = 2, FcDAC, or NqMes.
Figure 11. ORTEP diagram showing 50% probability thermal
ellipsoids and selected atom labels for 2e. Hydrogen atoms have
been omitted for clarity. Selected bond lengths (A) and angles
(deg): W-C1, 2.299(4); W-C2, 1.978(4); W-C3, 2.034(3); W-
C4, 2.029(3); W-C5, 2.034(3); W-C6, 2.029(3); O2-C2,
1.164(5); O3-C3, 1.143(4); O4-C4, 1.157(4); O5-C5,
1.143(4); O6-C6, 1.157(4); N1-C1-N2, 103.9(3).
the NHCs in 2c-e, 5e, and 6c-e (201.4-222.5 ppm,
CDCl3)94 was consistent with those observed in previously
reported group 6 [M(CO)5] complexes bearing NHC ligands
(188.3-226.1 ppm).61,71-85
Single crystals of the [M(CO)5] complexes supported by 2
and NqMes were obtained and subjected to X-ray diffrac-
tion.94 The Cr-NHC distances in 2c and 6c (2.158(3) and
2.125(3) A, respectively)94 were consistent with those ob-
served in NHC-supported [Cr(CO)5] complexes (2.098-
2.155 A).71-77 Additionally, the trans (1.852(4) and
1.857(3) A) and cis (1.895 and 1.894 A) chromium-carbonyl
bond lengths fell within the range of values observed in
analogous complexes (1.840-1.868 and 1.888-1.901 A for
trans and cis, respectively). Although no related structures of
[Mo(CO)5] complexes are known, the metric parameters of
2d and 6d (Mo-CNHC=2.328(2) and 2.257(5) A; Mo-Ctrans
=
1.975(3) and 1.999(7) A; Mo-Ccis = 2.049 and 2.035 A;
respectively)94 are similar to those of their tungsten congeners
2e (W-CNHC =2.299(4) A; W-Ctrans=1.978(4) A; W-Ccis=
2.032 A; see Figure 11) and 6e (W-CNHC = 2.259(5) A;
W-Ctrans =1.993(5) A; W-Ccis =2.039 A; see Figure 12).
A variety of complexes featuring an NHC coordinated to
[W(CO)5] have been structurally characterized,61,72,73,75,78-85
whose range of values for W-CNHC (2.242-2.296 A),
W-Ctrans (1.935-2.010 A), and W-Ccis (2.016-2.045 A)
encompass those exhibited by 2e and 6e. Overall, the 13C
NMR chemical shifts and structural features for 2c-e and
6c-e do not vary substantially beyond their different atomic
radii of Cr vs Mo and W. Surprisingly, the key structural and
NMR spectroscopic features of the aforementioned NHC-
supported [M(CO)5] complexes appear to be independent of
the number or nature of the redox-active functionalities
present on the NHC scaffold (e.g., two ferrocene units in 2
vs one naphthoquinone moiety in NqMes).
Figure 12. ORTEP diagram showing 50% probability thermal
ellipsoids and selected atom labels for 6e. Hydrogen atoms and
solvent molecules have been omitted for clarity. Selected bond
lengths (A) and angles (deg): W-C1, 2.259(5); W-C2, 1.993(5);
W-C3, 2.048(5); W-C4, 2.036(6); W-C5, 2.029(6); W-C6,
2.042(6); O2-C2, 1.153(6); O3-C3, 1.128(6); O4-C4, 1.147(7);
O5-C5, 1.149(7); O6-C6, 1.143(6); N1-C1-N2, 103.8(4).
are highly sensitive to the features of the cyclic system that
comprises it. Compared to the imidazolylidene-derived 1-4
and NqMes, we believe the steric effects of the N-mesityl
substituents and the six-membered ring in FcDAC cause the
carbene hybridization to adopt relatively greater sp-charac-
ter, resulting in longer bonds to coordinated metals.102,103
Although the 13C NMR and crystallographic analyses of
the [Ir(CO)2Cl] and [M(CO)5] complexes of the imidazolyli-
dene-based NHCs 1-4 and NqMes displayed highly con-
served features, the Ir-CNHC distance and N-C-N angle in
5b were significantly longer and more obtuse, respectively,
than those observed in (1-4)b or 6b. On the basis of these
results, we conclude that the intrinsic electronic properties of
the carbene nuclei in the aforementioned complexes (as
determined by their respective 13C NMR chemical shifts)
as well as their coordination chemistry (as measured by their
respective N1-C1-N2 angles and M-Ccarbene distances)
(102) Arduengo, A. J., III; Dias, H. V. R.; Dixon, D. A.; Harlow,
R. L.; Klooster, W. T.; Koetzle, T. F. J. Am. Chem. Soc. 1994, 116, 6812–
6822.
(103) Arduengo, A. J., III; Dias, H. V. R.; Harlow, R. L.; Kline, M.
J. Am. Chem. Soc. 1992, 114, 5530–5534.

6702 Organometallics, Vol. 28, No. 23, 2009
Rosen et al.
Table 1. Carbonyl Stretching Energies for Complexes (1-6)b,
2c-e, and 6c-ea
Figure 13. Representative NHCs with TEP values similar to
1-4, FcDAC, and NqMes.
the five-membered aromatic and strained, six-membered
nonaromatic systems. Whereas the cyclic nature of FcDAC
has a significant impact on structural features, it does not
appear to greatly affect its ligand donating ability.
To gain more insight into the donating abilities of the
aforementioned NHCs, efforts turned toward evaluating
their Tolman electronic parameters (TEPs), which can be
derived from the metal-carbonyl stretching energies.104,105
For [Ir(CO)2Cl] complexes, Nolan enhanced an equation
developed by Crabtree for determining the TEP from the
[Ir(CO)2Cl]
[M(CO)5]
νCO
νav
TEP
νCO
1b
2b
3b
4b
5b
6b
2058, 1982
2064, 1984
2068, 1986
2068, 1985
2062, 1982
2072, 1988
2020.0
2024.0
2027.0
2026.5
2022.0
2030.0
2046.9
2050.3
2052.9
2052.4
2048.6
2055.4
2c
2d
2e
6cb
6d
6e
2052, 1927, 1887*
2052, 1926, 1889*
2059, 1919, 1883*
2056, 1975*, 1933, 1678
2063, 1980*, 1933, 1675
2062, 1976*, 1928, 1676
observed νav, whereby TEP = 0.847 ꢀ νav þ 336 cm-1 67,70
.
The TEPs calculated for 1-4, FcDAC, and NqMes
(2046.9-2055.4 cm-1) were consistent with the range ob-
served in other NHC-supported [Ir(CO)2Cl] complexes
(2049.5-2057.3 cm-1; see Figure 13 for representative ex-
amples).67 For comparison, the TEP for FcDAC compares
well to that for IAd (2048.6 vs 2049.5 cm-1, respectively) and
NqMes to IPrCl (2055.4 vs 2055.1 cm-1, respectively).
Similarly, there are other reported NHCs that exhibit
comparable TEPs to the 2046.9 cm-1 measured for 1 (IAd,
2049.5 cm-1), 2050.3 cm-1 for 2 (ItBu, 2050.1 cm-1), and
2052.4 and 2052.9 cm-1 for 4 and 3 (SIPr, 2052.2 cm-1).
When viewed in the context of previously reported NHCs
(i.e., not including 2-4 or 6), 1 and FcDAC were among the
most electron-donating, NqMes was among the least, and
2-4 were intermediate in donicity.
Determination of ligand TEP values from group 6
[M(CO)5] carbonyl stretching energies has not yet been
detailed in the literature. Nonetheless, examination of the
carbonyl stretching energies in 2c-e, 5e, and 6c-e in com-
parison to previously reported analogues should enable
qualitative evaluation of the donating abilities of 1-4,
FcDAC, and NqMes. Prior to this examination, a discussion
regarding the number and symmetry of IR-active carbonyl
stretching modes is helpful. In an idealized geometry, the C4v
[M(CO)5(NHC)] scaffold has two sets of symmetry-inequi-
valent CO groups in the equatorial (1) and axial (2) posi-
tions.106 For COax, the vibrational mode has A1 symmetry
and is IR and Raman active.107 For COeq, the irreducible
representation for the CO stretching modes is A1, B1, and E,
where B1 is not IR active. To distinguish between the
equatorial and axial carbonyls, we will use superscripted
(1) and (2) after the relevant irreducible representation.
Because NHCs have a strong σ-donating effect, the CO
trans to the NHC in the axial position will have the weakest
a MLn = [Ir(CO)2Cl] for (1-6)b, [Cr(CO)5] for 2c and 6c, [Mo(CO)5]
for 2d and 6d, and [W(CO)5] for 2e and 6e. For the [M(CO)5] complexes,
the A1(2) mode is highest in energy and the A1(1) mode is denoted with an
asterisk (*). Measurements performed in CH2Cl2. b Measurements
performed in CHCl3.
Infrared Spectroscopy. Metal-bound carbonyls are useful
spectroscopic handles for measuring the electron density at
ligated metal centers. Increasing the electron density on a
metal will increase its π-back-bonding ability, thus reducing
the C-O bond order and stretching frequency (νCO). For
example, a more donating NHC will increase the electron
density at the coordinated metal and thus lower the carbonyl
stretching energy, allowing measurement of the ligand doni-
city. Many [Ir(CO)2Cl] complexes supported by NHCs have
been prepared for this reason; therefore we sought to deter-
mine and compare the donating abilities of 1-4, FcDAC,
and NqMes to known NHCs.67-70
Complexes (1-6)b exhibited a range of trans (2058-
2072 cm-1) and cis (1982-1988 cm-1) carbonyl stretching
energies (see Table 1) consistent with those observed in
known NHC-supported [Ir(CO)2Cl] complexes (trans:
2055-2072 cm-1, cis: 1971-1989 cm-1).67,69,93,97 Some
remarkable trends become apparent upon examination
of the average values exhibited by (1-6)b (νav = 2020.0-
2030.0 cm-1). The naphthoquinone-annulated NqMes was
less electron donating than the other imidazolylidene-based
NHCs (1-4), suggesting that the quinone moiety decreased
the donating ability of its fused carbene. Imidazolylidene 1
was more donating than benzimidazolylidenes 2-4, further
evidence that annulation decreases the electron density at the
carbene. Given the narrow range of νav values for 2-4, we
conclude that the electron-donating ability of the benzimi-
dazolylidene scaffold does not significantly vary with the
number of redox-active groups present in the ligand or their
position relative to the carbene atom. Because the carbonyl
stretching energies for 1b and 5b are similar, no measurable
alteration in carbene electron density is observable between
(104) Tolman, C. A. Chem. Rev. 1977, 77, 313–348.
€
(105) Strohmeier, W.; Muller, F.-J. Chem. Ber. 1967, 100, 2812–2821.
(106) Cotton, F. A.; Kraihanzel, C. S. J. Am. Chem. Soc. 1962, 84,
4432–4438.
(107) Cotton, F. A. Chemical Applications of Group Theory, 3rd ed.;
John Wiley & Sons: New York, 1990.

Article
Organometallics, Vol. 28, No. 23, 2009 6703
Figure 14. Representative cyclic voltammograms (100 mV s-1 scan rate) of Fc* in CH2Cl2 with 0.1 M [Bu4N][PF6] and 1 mM (A) 2a,
(B) 6a, (C) 2b, and (D) 6b. Features are labeled according to metal oxidation (FeII/III or IrI/II), quinone reduction (Q0/-), or
decamethylferrocene internal standard (Fc*).
Table 2. Electrochemical Properties of (1-6)a and (1-6)ba
Table 3. Electrochemical Properties of 2c-e and 6c-ea
E1/2 (V)
E1/2 (V)
2c
2d
2e
0.67
0.72
0.69
6c
6d
6e
0.98, -0.65
1.15, -0.68
1.15, -0.62
a MLn = [Cr(CO)5] for 2c and 6c, [Mo(CO)5] for 2d and 6d, and
[W(CO)5] for 2e and 6e. Measurements performed in CH2Cl2 containing
0.1 M [Bu4N][PF6] at 100 mV s-1 scan rate.
(1)
cantly more intense that the A1 mode).108-110 For the
[M(CO)5] (M = Cr, Mo, W) complexes supported by 2 and
[Ir(COD)Cl]
E1/2 (V)
[Ir(CO)2Cl]
E1/2 (V)
(2)
NqMes, the A1 stretching energies (2052-2063 cm-1
,
ΔE (mV)b
Table 1) coincided with the range of values observed in other
NHC-supported analogues (2048-2064.1 cm-1).61,71-85
The A1(1) modes for the pentacarbonyls supported by 2 (2c,
1887 cm-1; 2d, 1889 cm-1; 2e, 1883 cm-1) and NqMes (6c,
1975 cm-1; 6d, 1980 cm-1; 6e, 1976 cm-1) were consistent
with other reported values (1882-1980 cm-1). Similarly, the
CO stretching energies for the E(1) modes in 2c-e and 6c-e
(1919-1933 cm-1) agreed well with those observed in other
NHC-supported [M(CO)5] complexes (1900-1966 cm-1).
These results suggest that either an oxidizable or a reducible
functional group could be incorporated into an NHC with-
out significantly altering its fundamental electron-donating
ability to [M(CO)5] fragments comprising group 6 metals.
1a
2a
3a
4a
5a
6a
0.72, 0.58
0.75, 0.62
0.62
0.53
1.02, 0.76
0.95, -0.66
1b
2b
3b
4b
5b
6b
0.78, 0.68 (sh)
0.79, 0.70 (sh)
0.71
60, 100 (80)
40, 80 (60)
90
40
180
120
0.57
0.94c
-0.54c
a MLn = [Ir(COD)Cl] for (1-6)a and [Ir(CO)2Cl] for (1-6)b. Mea-
surements were performed in CH2Cl2 containing 0.1 M [Bu4N][PF6] at
100 mV s-1 scan rate. b ΔE = E1/2(IrCO) - E1/2(IrCOD). Where two
oxidations are present, the averaged value is presented in parentheses.
c The iridium-centered oxidation could not be observed within the
solvent window.
(2)
π-back-bonding interaction with the metal; thus the A1
mode will be highest in energy.106 However, the relative
(1)
ordering of the A1 þ E(1) modes varies depending on the
€
(108) Herrmann, W. A.; Roesky, P. W.; Elison, M.; Artus, G.; Ofele,
(1)
K. Organometallics 1995, 14, 1085–1086.
nature of the complex, where the A1 mode can vary in
€
(109) Ofele, K.; Herrmann, W. A.; Mihalios, D.; Elison, M.; Herdtweck,
energy greatly or be underneath the E(1) band. Correct
identification can be achieved by comparing the relative
intensities of the bands (in general, the E(1) mode is signifi-
E.; Scherer, W.; Mink, J. J. Organomet. Chem. 1993, 459, 177–184.
€
(110) Ofele, K.; Roos, E.; Herberhold, M. Z. Naturforsch. B 1976, 31,
1070–1077.

6704 Organometallics, Vol. 28, No. 23, 2009
Rosen et al.
Figure 15. Normalized IR difference spectra at 60 s intervals showing the shift in metal carbonyl stretching energies upon oxidation
(Eapp=þ1.2 V) of 2b (A) and 2e (C) or reduction (Eapp=-1.2 V) of 6b (B) and 6e (D) in CH2Cl2 containing 10 mM analyte and 0.1 M
[Bu4N][PF6].
Electrochemistry. Complexes 1a and 2a, comprising N,N0-
diferrocenyl NHCs, exhibited two sets of quasi-reversible
peaks (1a, þ0.58 and þ0.72 V; 2a, þ0.62 and þ0.75 V; see
Figures S10, and 14A, respectively, as well as Table 2) in
CH2Cl2,94 which were attributed to the oxidation of their
first and second ferrocene units, respectively. The potential
separation between the two couples of 140 mV for 1a and
130 mV for 2a is consistent with other reported diferrocenyl-
functionalized NHCs.57,60,61,82 As expected, complexes 3a
and 4a, supported by monoferrocenyl-functionalized NHCs,
exhibited only one ferrocene-based oxidation at þ0.62 and
þ0.53 V, respectively.94 None of the iridium-based oxida-
tions could be observed within the solvent window for
(1-4)a.
In contrast, 5a and 6a exhibit quasi-reversible redox
processes at þ1.02 and þ0.95 V (see Figures S13 and 14B,
as well as Table 2)94 that were attributed to IrI/II couples,
values that are consistent with other NHC-supported
[Ir(COD)Cl] complexes.111,112 A reduction feature was also
observed in 6a at -0.66 V, similar to the quinone reduction
observed in the previously reported [Rh(COD)Cl] analo-
gue.64 The ferrocene oxidation in 5a occurs at þ0.76 V,
higher than the first oxidations in (1-4)a, reflecting the
influence of the strained six-membered ring and the inability
of the orthogonal nitrogen lone pairs to donate electron
density into the Cp rings to which they are linked. Addition-
ally, the 90 mV difference between Fc0/þ potentials in 3a vs
4a revealed that an N-bound ferrocene was more electron
deficient than a C-bound ferrocene. However, because the
Fc0/þ couple in 3a occurs at an identical potential to the
first oxidation observed in 2a (þ0.62 V), we conclude that
the overall electron density does not depend on the num-
ber of ferrocene units within the NHCs. Furthermore,
the small separation between ferrocene oxidations in 1a
and 2a (<40 mV) suggests that the imidazolylidene-
and benzimidazolylidene-based NHC scaffolds exert simi-
lar influences on the electronic environment at the iron
centers.
The FeII/III couples in (1-4)b exhibited anodic shifts of
40-100 mV relative to those observed in (1-4)a, reflecting
the greater electron-withdrawing character of [Ir(CO)2Cl]
compared to [Ir(COD)Cl] and agreeing with the 13C NMR
results. Similarly, the ferrocene oxidation in 5b was observed
at þ0.94 V, a potential 180 mV higher than that observed
for the analogous oxidation in 5a. A greater shift for 5 than
1-4 is expected, given that the ferrocene unit in FcDAC
is linked to the NHC at two positions instead of one.
The quinone reduction in 6b at -0.54 V was 120 mV higher
than in 6a, consistent with the reduced electron density
at the metal.64 Given the large shift in potentials for the
NHC-based oxidations in 5b and 6b, it was not surpri-
sing that the IrI/II couples were shifted beyond the sol-
vent window and could not be observed. These latter
results were consistent with the electrochemical properties
of related NHC-supported [M(CO)2Cl] complexes (M=Rh
and Ir).66,68,111
As summarized in Table 3, a relatively narrow range of
M0/þ oxidation potentials was observed for the [M(CO)5]
complexes supported by 2 (Cr: þ0.67 V; Mo: þ0.72 V; W:
þ0.69 V).94 These values indicate that the electronic inter-
actions between 2 and the [M(CO)5] fragments are nearly
indistinguishable, presumably due to equivalent me-
tal-NHC interactions. Overall, the range of metal-centered
oxidation potentials observed in the [M(CO)5] complexes
supported by 2 and NqMes (from þ0.67 to þ1.15 V) were
consistent with values measured in analogous complexes
supported by other NHCs (þ0.43 to þ1.2 V).61,82 However,
the M0/þ couple in 6c (þ0.98 V) occurs at a potential
substantially lower than its Mo and W congeners (þ1.15 V
for 6d and 6e, respectively), in contrast to 2.94 In gene-
ral, as the NHC and metal orbital energies converge,
(111) Tennyson, A. G.; Ono, R. J.; Hudnall, T. W.; Khramov, D. M.;
Er, J. A. V.; Kamplain, J. W.; Lynch, V. M.; Sessler, J. L.; Bielawski,
C. W. Chem.-Eur. J. 2009, DOI: 10.1002/chem.200901883.
(112) Tennyson, A. G.; Rosen, E. L.; Collins, M. S.; Lynch, V. M.;
Bielawski, C. W. Inorg. Chem. 2009, 48, 6924–6933.

Article
Organometallics, Vol. 28, No. 23, 2009 6705
Table 4. Spectroelectrochemical Resultsa
latter case. Because the NHC ligand is conserved in these
complexes, we conclude that the observed variation is due
to the difference in d-orbital energy between the mid-
transition group 6 metals and the late-transition group 9
iridium, as judged by the relative electron affinities of
these metals (Cr: 64.3 kJ mol-1; Mo: 71.9 kJ mol-1; W: 78.6
kJ mol-1; Ir: 151 kJ mol-1).113-115
Spectroelectrochemistry. A powerful method for determin-
ing the electronic influence of a redox-active substituent
within an NHC at a coordinated metal is IR spectro-
electrochemistry.17-19 Oxidation of the ferrocene units in
1-3 and FcDAC should decrease the donating abilities of the
NHCs, thus lowering the electron density at the coordinated
metal carbonyls and result in an increased νav. Alternatively,
reducing the quinone in complexes supported by NqMes
should increase the carbene’s electron-donating ability, af-
fording a more electron rich [Ir(CO)2Cl] or [M(CO)5] frag-
ment with a concomitant decrease in carbonyl stretching
energies. To explore the relationship between oxidation
state of the redox-active functionality and donicity of their
respective NHCs, we sought to measure the shift in the
average νCO of [Ir(CO)2Cl] and [M(CO)5] complexes upon
oxidation of the ferrocene units in 1-3 and reduction
of NqMes.
[Ir(CO)2Cl]
νav TEP*
[M(CO)5]
νCO Δν A1
(1)
νCO
*
*
Δνav ΔTEP
*
1b 2072, 2035.0 2059.6 þ15
þ12.7 2c 2058,
1980†,
1944
þ93
þ91
þ93
-93
-102
-98
1998
2b 2076, 2037.0 2061.3 þ13
þ11.0 2d 2058,
1980†,
1944
1998
Surprisingly, oxidation of (1-3)b and 5b resulted in
nearly identical shifts in νav (1b: þ15 cm-1; 2b: þ13 cm-1
;
3b: þ13 cm-1; 5b: þ13 cm-1; for 2b, see Figure 15A; for the
others, see Figures S24-S26; key features are summarized in
Table 4)94 that corresponded to ΔTEP values ranging from
þ11.0 to þ12.7 cm-1. These results suggest that the scaffold
structure (i.e., imidazolylidene vs benzimidazolylidene vs
nonaromatic) or the number of ferrocene units (i.e., one vs
two) does not obfuscate the redox tunability of NHC doni-
city when a redox-active group is directly connected via an
N substituent. Closely paralleling the ferrocene oxidation
3b 2080, 2040.0 2063.9 þ13
þ11.0 2e 2064,
1976†,
1936
2000
4bb
-
-
-
-
-
6c 2048,
1916,
1882†
5b 2074, 2035.0 2059.6 þ13
þ11.0 6d 2058,
1920,
1996
1878†
measurements, reduction of 6b afforded a Δνav of -12.5 cm-1
,
6b 2060, 2017.5 2044.8 -12.5 -10.6 6e 2056,
resulting in a decrease of the TEP by 10.6 cm-1 (see
Figure 15B). Because the values of Δνav and ΔTEP for 6b are
nearly identical in magnitude but opposite in sign to those
observed for (1-3)b and 5b, we conclude that the redox
tunability of these ligands does not strongly depend on the
specific chemical identity of the redox-active functional group.
Rather, the overall charge imparted to the molecule upon redox
change appears to be the origin of the ligand’s enhanced or
attenuated donating ability.
1975
1914,
1878†
a MLn = [Ir(CO)2Cl] for (1-6)b, [Cr(CO)5] for 2c and 6c, [Mo(CO)5]
for 2d and 6d, and [W(CO)5] for 2e and 6e. Measurements were
performed in CH2Cl2 containing 0.1 M [Bu4N][PF6] under the condi-
tions specified in Figure 15. b This compound was found to decompose
under the bulk electrolysis described above, thus its spectroelectrochem-
ical properties could not be determined. Values with an asterisk (*)
correspond to the in situ oxidized or reduced complex. The A1(1) mode is
(1)
denoted with a cross (†). Values of Δνav, ΔTEP and Δν A1 were
obtained by subtracting the values observed for the neutral complexes
from those for the oxidized/reduced complexes.
Similar shifts were observed in the A1(1) carbonyl stretching
modes for the [M(CO)5] complexes of 2 and NqMes, albeit
with dramatically greater magnitude. Oxidation of 2c-e
(1)
increased the ν A1 energies 91-93 cm-1 (for 2e, see
the extent of their interaction should increase. For a poorly
matched combination, the NHC should “experience” the
influence from a generic nþ ion (which will not depend on
the metal identity). Conversely, well-matched combinations
should exhibit more significant interactions and greater
dependence on metal identity and associated properties
(e.g., electronegativity, electron affinity, ionization potential).
Given that the metal oxidation potentials for 6d,e are higher
than 6c and that Mo/W d-orbital energies are lower than Cr,
we conclude that NqMes is a better energy match with
[Cr(CO)5], whereby the greater interaction increases electron
donation to the metal and thus reduces its oxidation potential.
Interestingly, the quinone reductions in 6c-e occurred
at higher energies (from -0.62 to -0.68 V) than the
corresponding [Ir(CO)2Cl] complex 6b (-0.54 V), con-
sistent with a greater NHC-metal interaction in the
Figure 15C; for 2c,d, see Figures S27-S28; also see
Table 4),94 consistent with reduced metal electron density
and consistent with shifts observed in complexes supported by
ferrocene-functionalized phosphines.17-19 Conversely, the Δν
A1(1) values for 6c-e ranged from -93 to -102 cm-1 (for 6e,
see Figure 15D; for 6c,d, see Figures S29-S30),94 indicative of
enhanced donicity upon reduction of NqMes. As observed with
(1)
the [Ir(CO)2Cl] complexes, the magnitude of Δν A1 was
nearly identical for 2 vs NqMes, albeit with opposite signs
(positive for oxidation, negative for reduction).
(113) Allen, L. C. J. Am. Chem. Soc. 1989, 111, 9003–9114.
(114) Allred, A. L. J. Inorg. Nucl. Chem. 1961, 17, 215–221.
(115) Pauling, L. The Nature of the Chemical Bond, 3rd ed.; Cornell
University Press: Ithaca, 1960.

6706 Organometallics, Vol. 28, No. 23, 2009
Rosen et al.
Overall, the modulation of the ligand donating abilities of
1-3, FcDAC, and NqMes upon electrochemical switching of
the redox-active units (by oxidation or reduction) was largely
independent of their molecular characteristics, contrary to
expectations. No significant dependence on the presence
(FcDAC vs 1-3 and NqMes) or extent of an aromatic system
(1 vs 2) was observed. Oxidation of one or two of the
ferrocene units (2 vs 3) afforded the same change in NHC
electron-donating ability, indicating that a second oxidation
had marginal impact beyond the first. Furthermore, the
magnitude of the enhanced donicity observed for NqMes
upon reduction matched the attenuation observed for the
ferrocene-functionalized NHCs upon oxidation. Collec-
tively, these results suggest that the changes in electron-
donating ability of 1-3, FcDAC, and NqMes are largely
due to Coulombic effects, where removal or addition of an
electron alters the overall molecular charge.
number of ferrocene units nor the identity of the aromatic
backbone strongly influenced the electron density within the
NHC. In contrast, the Fc0/þ couples in 4a,b and 5a,b were
distinct from those observed in 1-3, evidence that both
the attachment point of the redox-active group (relative to
the carbene nucleus) and the ring comprising the NHC
affected the electronic environment at the iron centers in
the respective complexes.
Spectroelectrochemical IR analyses revealed a narrow range
of Δνav values for (1-3)b and 5b (þ13 to þ15 cm-1), corres-
ponding to increases in TEPs from 11.0 to 12.7 cm-1, upon
oxidation of the ferrocene units. Reduction of the quinone
in 6b, however, decreased its νav by 12.5 cm-1, indicating
enhanced donation by NqMes (ΔTEP = -10.6 cm-1). Inter-
estingly, the shiftfor NqMes was nearly equal in magnitude but
opposite in sign to the shifts observed for 1-3 and FcDAC.
Because the changes in ligand donating abilities did not depend
significantly on the NHC characteristics, we surmise that the
observed trends primarily reflect Coulombic effects, whereby
addition of a positive or negative charge to the ligand alters its
TEP by roughly þ14 or -11 cm-1, respectively.
Conclusions
In sum, we have developed families of [Ir(COD)Cl],
[Ir(CO)2Cl], and [M(CO)5] (M=Cr, Mo, W) complexes sup-
ported by NHCs comprising redox-active ferrocene (1-4 and
FcDAC) and quinone (NqMes) functionalities. Although the
13C NMR spectroscopic and structural features of these com-
plexes were consistent with previously reported analogues,
those comprising FcDAC were notably distinct from the other
redox-active NHCs studied (1-4 or NqMes), presumably
reflecting the greater sp-hybridization of the carbene nucleus
in the former.
Measurement of the electron density at the metal, and thus
the donating ability of the NHCs, via IR spectroscopic
analysis of carbonyl stretching energies revealed different
behavior than the X-ray diffraction or 13C NMR spectro-
scopic results. Complexes 1b and 6b exhibited the lowest and
highest νav values, respectively, demonstrating that the NHC
backbone (imidazolylidene vs benzimidazolylidene vs
naphthoquinone) had a significant effect on the donating
ability of the respective carbenes. However, the average
carbonyl stretching energies for complexes (2-5)b were
similar (differences of no more than 3 cm-1 were observed),
suggesting the benzimidazolylidene scaffold could be func-
tionalized with one or two redox-active moieties at the N
atoms or backbone without significantly perturbing its
donicity. Relative to previously reported NHCs, 1 and
FcDAC were among the most electron-donating, NqMes
was comparable with the least, and 2-4 were intermediate.
Because electrochemical analyses revealed similar ferrocene
oxidation potentials for the [Ir(COD)Cl] and [Ir(CO)2Cl]
complexes supported by 1-3, we conclude that neither the
On the basis of these results, NHCs with electrochemically
tunable electronic properties could be obtained via incorpor-
ating a redox-active moiety by the most straightforward
synthetic route available, without requiring extensive ligand
design. For an NHC that can be electrochemically toggled to
a less donating state, it need only feature a functionality that
(1) is in close proximity to the carbene and (2) endows the
molecule with a positive charge upon oxidation. Conversely,
an analogous NHC bearing a redox-active group that ac-
quires a negative charge upon reduction could function as a
ligand that can be redox-switched to a more donating state.
Ultimately, we believe our findings will simplify the rational
design of NHCs for use in redox-switchable applications that
employ electrochemical control to attain both enhanced and
attenuated electron-donating states.
Acknowledgment. We are grateful to the United States
Army Research Office (W911NF-05-1-0430 and W911-
NF-06-1-0147), the National Science Foundation (CHE-
0645563), and the Robert A. Welch Foundation (F-1621)
for financial support.
Supporting Information Available: Synthetic and experimen-
tal details; tables of crystallographic, spectroscopic, and struc-
tural data (Tables S1-S7); ORTEP diagrams of (1-3)a, (5,6)a,
2c,d, and 6c,d (Figures S1-S9); CVs of 1a, (3-5)a, 1b, (3-5)b,
2c-e, and 6c-e (Figures S10-S23); spectroelectrochemical
1
spectra for 1b, 3b, 5b, 2c,d, and 6c,d (Figures S24-S30); H
and 13C NMR spectra; CIFs. This material is available free of
charge via the Internet at http://pubs.acs.org.