Intramolecular optical electron transfer in mixed-valent dinuclear iron-ruthenium complexes featuring a 1,4-diethynylaryl spacer

Article abstract of DOI:10.1021/om700584m

The ground-state electronic structure and the lowest-lying excited states of the cationic mixed-valent dinuclear complexes [(η²-dppe) (η⁵-C₅Me₅)Fe[C≡C-1,4-(C ₆H₄)C≡C]Ru(η²-dppe) ₂(X)][PF₆] (X = Cl, 2; X = C≡C(4-C₆H ₄NO₂), 5) are discussed, with particular emphasis on the photoinduced intramolecular electron transfer between the ruthenium and iron centers. The location and intensities of the low-lying absorptions exhibited in the near-infrared (near-IR) range by these heterodinuclear mixed-valent (MV) complexes correlate with predictions based on the Hush model, strongly suggesting that they correspond to intervalence charge-transfer (IVCT) bands.

Full text of DOI:10.1021/om700584m

Organometallics 2008, 27, 1063–1072
1063
Intramolecular Optical Electron Transfer in Mixed-Valent
Dinuclear Iron-Ruthenium Complexes Featuring a
1,4-Diethynylaryl Spacer
Nicolas Gauthier,† Céline Olivier,† Stéphane Rigaut,*,† Daniel Touchard,†
Thierry Roisnel,†,‡ Mark G. Humphrey,*,§ and Frédéric Paul*,†,‡
UniVersité de Rennes 1, Sciences Chimiques de Rennes, Campus de Beaulieu,
35042 Rennes Cedex, France, CNRS (UMR 6226), UniVersité de Rennes 1, Sciences Chimiques de Rennes,
Campus de Beaulieu, 35042 Rennes Cedex, France, and Department of Chemistry, Australian National
UniVersity, Canberra, Australian Capital Territory 0200, Australia
ReceiVed June 14, 2007
The ground-state electronic structure and the lowest-lying excited states of the cationic mixed-valent
dinuclear complexes [(η2-dppe)(η5-C5Me5)Fe[Ct C-1,4-(C6H4)Ct C]Ru(η2-dppe)2(X)][PF6] (X ) Cl, 2;
X ) Ct C(4-C6H4NO2), 5) are discussed, with particular emphasis on the photoinduced intramolecular
electron transfer between the ruthenium and iron centers. The location and intensities of the low-lying
absorptions exhibited in the near-infrared (near-IR) range by these heterodinuclear mixed-valent (MV)
complexes correlate with predictions based on the Hush model, strongly suggesting that they correspond
to intervalence charge-transfer (IVCT) bands.
2
2+ in Chart 1) designed for switching the third-order nonlinear
Introduction
optical (NLO) activity between three redox states.9 The suc-
cessful use of this redox family for such a purpose rests on the
kinetic stability and on the widely differing optical properties
of the various redox states.10,11 In order to gain a better
understanding of the molecular origin of the NLO properties
of 2/2[PF6]/2[PF6]2, we have started to investigate more closely
the lowest-lying excited states of each redox congener. In
contrast to what might have been anticipated from previous
studies on related complexes,10,12–15 our investigations on the
cationic complex 2[PF6] strongly suggest that the low-lying
intense absorption at 1124 nm corresponds to a Ru-Fe
intervalence charge-transfer (IVCT) band. This result is of
particular interest in the context of recent investigations of
related symmetric homodinuclear ruthenium MV complexes
such as 3[PF6] and 4[PF6], where the nature of the most intense
low-lying absorptions has also been discussed.13,14,16 With the
MV complex 3[PF6], the classical Hush model was unable to
rationalize the spectral features of the low-energy absorption
near 1500 nm that was believed to be an IVCT band.13,16 With
In the field of mixed-valent (MV) complexes, dinuclear
compounds featuring a carbon-rich bridge directly linked to the
redox centers by metal–carbon bonds have attracted particular
attention as a potentially interesting class of compounds for
various applications in the field of molecular-scale electronics.1–3
The classical Hush treatment used to rationalize the electron
transfer of purely inorganic MV complexes has been success-
fully extended to symmetrical representatives of this new class
of carbon-rich organometallic MV complexes such as 1[PF6],4–6
but this simple model has seldom been applied to unsymmetrical
representatives.7,8
Some of us recently reported a new redox family of
heterometallic iron/ruthenium aryl acetylide complexes (2/2+/
† Sciences Chimiques de Rennes, Université de Rennes 1.
‡ CNRS (UMR 6226), Université de Rennes 1.
§ Australian National University.
(1) (a) Kim, B.; Beebe, J. M.; Olivier, C.; Rigaut, S.; Touchard, D.;
Kushmerick, J. G.; Zhu, X.-Y.; Frisbie, C. D. J. Phys. Chem. C 2007, 111,
7521–7526. (b) Rigaut, S.; Touchard, D.; Dixneuf, P. H. In Redox Systems
Under Nano-space Control; Hirao, T., Ed.; Springer: Heidelberg, Germany,
2006. (c) Low, P. J. Dalton Trans. 2005, 2821–2824. (d) Long, N. J.;
Williams, C. K. Angew. Chem., Int. Ed. Engl. 2003, 42, 2586–2617. (e)
Paul, F.; Lapinte, C. In Unusual Structures and Physical Properties in
Organometallic Chemistry; Gielen, M., Willem, R., Wrackmeyer, B., Eds.;
Wiley: San Francisco, 2002; pp 219–295.
(9) Samoc, M.; Gauthier, N.; Cifuentes, M. P.; Paul, F.; Lapinte, C.;
Humphrey, M. G. Angew. Chem., Int. Ed. 2006, 45, 7376–7379.
(10) Powell, C. E.; Cifuentes, M. P.; Morrall, J. P.; Stranger, R.;
Humphrey, M. G.; Samoc, M.; Luther-Davies, B.; Heath, G. A. J. Am. Chem.
Soc. 2003, 125, 602–610.
(2) Ceccon, A.; Santi, S.; Orian, L.; Bisello, A. Coord. Chem. ReV. 2004,
(11) Cifuentes, M. P.; Powell, C. E.; Morrall, J. P. L.; McDonagh, A. M.;
Lucas, N. T.; Humphrey, M. G.; Samoc, M.; Houbrechts, S.; Asselberghs,
I.; Clays, K.; Persoons, A.; Isoshima, T. J. Am. Chem. Soc. 2006, 128,
10819–10832.
248, 683–724.
(3) Paul, F.; Lapinte, C. Coord. Chem. ReV. 1998, 178/180, 427–505.
(4) Ibn Ghazala, S.; Paul, F.; Toupet, L.; Roisnel, T.; Hapiot, P.; Lapinte,
C. J. Am. Chem. Soc. 2006, 128, 2463–2476.
(12) Paul, F.; Toupet, L.; Thépot, J.-Y.; Costuas, K.; Halet, J.-F.; Lapinte,
C. Organometallics 2005, 24, 5464–5478.
(5) Crutchley, R. J. AdV. Inorg. Chem. 1994, 41, 273–325.
(6) (a) Launay, J.-P.; Coudret, C. In Electron Transfer in Chemistry;
Balzani, V., de Silva, A. P., Eds.; Wiley-VCH: Weinheim, Germany, 2000;
Vol. 5, pp 3–47. (b) Launay, J.-P. Chem. Soc. ReV. 2001, 30, 386–397.
(7) (a) Coat, F.; Guillevic, M.-A.; Toupet, L.; Paul, F.; Lapinte, C.
Organometallics 1997, 16, 5988–5998. (b) Paul, F.; Meyer, W.; Jiao, H.;
Toupet, L.; Gladysz, J. A.; Lapinte, C. J. Am. Chem. Soc. 2000, 122, 9405–
9414.
(13) Klein, A.; Lavastre, O.; Fiedler, J. Organometallics 2006, 25, 635–
643.
(14) Maurer, J.; Sarkar, B.; Schwederski, B.; Kaim, W.; Winter, R. F.;
Zalis, S. Organometallics 2006, 25, 3701–3712.
(15) Rigaut, S.; Olivier, C.; Costuas, K.; Choua, S.; Fadhel, O.; Massue,
J.; Turek, P.; Saillard, J.-Y.; Dixneuf, P. H.; Touchard, D. J. Am. Chem.
Soc. 2006, 128, 5859–5876.
(8) Bruce, M. I.; Costuas, K.; Davin, T.; Ellis, B. J.; Halet, J.-F.; Lapinte,
C.; Low, P. J.; Smith, M. E.; Skelton, B. W.; Toupet, L.; White, A. H.
Organometallics 2005, 24, 3864–3881.
(16) Colbert, M. C. C.; Lewis, J.; Long, N. J.; Raithby, P. R.; Younus,
M.; White, A. J. P.; Williams, D. J.; Payne, N. J.; Yellowlees, L.; Beljonne,
D.; Chawdhury, N.; Friend, R. H. Organometallics 1998, 17, 3034–3043.
10.1021/om700584m CCC: $40.75
2008 American Chemical Society
Publication on Web 02/27/2008

1064 Organometallics, Vol. 27, No. 6, 2008
Chart 1. Selected Dinuclear MV Complexes and Redox Interconversions among 2, 2+, and 22+
Gauthier et al.
the MV complex 4[PF6], Winter and co-workers have proposed
that the low-energy absorption is dominated by contributions
from ligand-centered (LC) π-π* transitions rather than metal-
to-metal charge-transfer transitions.14 The current belief with
these dinuclear carbon-rich MV complexes containing electron-
rich ruthenium centers is that the low-lying intense transitions
should not a priori be envisioned as metal-to-metal or interva-
lence charge-transfer (MMCT or IVCT) transitions.17 More
generally, the appropriateness of using the Hush model with
such MV complexes exhibiting a sizable delocalization of the
unpaired electron in the ground state (GS) has been questioned.8,13
We report in this contribution the synthesis of 2[PF6] and of a
MV analogue containing a 4-nitrophenyl alkynyl ligand in place
of the chloride ligand (5[PF6]), as well as the good match that
we have observed between experimental data for 2[PF6] and
5[PF6] and theoretical predictions based on the simple two-level
model initially developed by Hush18 and subsequently imple-
mented by others.19–21
excess of p-nitrophenylacetylene under similar conditions. Both
compounds were obtained in fair yields and were fully
characterized by LSIMS, elemental analysis, cyclic voltammetry,
and the usual spectroscopy (see the Experimental Section).
1
The H and 31P NMR spectra of 2 and 5 confirm the purity
of the isolated samples and are diagnostic of iron- (near 100
ppm) and ruthenium-bound (50–55 ppm region) dppe ligands
on acetylide metal centers;24,25 as expected, these 31P NMR
signals are in a 1:2 ratio. One intense absorption is observed in
the infrared spectra in the νCt
region near 2055 cm-1
,
C
characteristic of Ru(II) and Fe(II) acetylides. These broad
absorptions correspond to the overlap of the two metal alkynyl
stretches expected for the bridging ligand in 2 and 5. These
were previously reported at 2053 cm-1 for 8-H,24 at 2067 cm-1
for 9-H,25 and at 2052 cm-1 for 10 (Chart 2).25 Notably, no
additional metal alkynyl stretch is clearly detected for 5, in spite
of the presence of the 4-nitrophenylacetylide ligand on ruthe-
nium, but a slight shoulder is apparent at ca. 2070 cm-1. A
corresponding feature appears as a very small peak at 2074 cm-1
for 10.26
Results
Synthesis and Characterization of the Heterodinuclear
MV Complexes 2[PF6] and 5[PF6]. The MV complexes 2[PF6]
and 5[PF6] were isolated from their neutral precursors by
chemical oxidation using 1 equiv of ferrocenium hexafluoro-
phosphate (Scheme 2). The large separation between the redox
potentials corresponding to the mono- and dioxidized states
(g0.65 V; see Table 1) permitted their clean isolation as red-
brown solids.3,4 The monooxidized compounds 2[PF6] and
5[PF6] exhibit the same cyclic voltammograms as their neutral
parents, but infrared spectroscopy clearly reveals their oxidized
nature. Thus, two νCt C stretches are now observed for each of
Synthesis and Characterization of the Heterodinuclear
Complexes 2 and 5. The dinuclear compound 2 was obtained
from reaction between the functional mononuclear Fe(II) alkynyl
complex 6,22 containing a terminal ethyne functionality, and
the five-coordinate [cis-Cl(dppe)2Ru][OTf] (dppe ) 1,2-bis-
(diphenylphosphino)ethane) cation (7[OTf])23 (Scheme 1). The
analogue 5 was isolated from 2 after reaction with a 4-fold
(17) Olivier, C.; Choua, S.; Turek, P.; Touchard, D.; Rigaut, S. Chem.
Commun. 2007, 3100–3102.
these compounds, one of them being located below 1950 cm-1
,
(18) (a) Hush, N. S. Prog. Inorg. Chem. 1967, 8, 357–389. (b) Hush,
N. S Prog. Inorg. Chem. 1967, 8, 391–444.
(19) Brunschwig, B. S.; Creutz, C.; Sutin, N. Chem. Soc. ReV. 2002,
31, 168–184.
(24) Denis, R.; Toupet, L.; Paul, F.; Lapinte, C. Organometallics 2000,
19, 4240–4251.
(20) Nelsen, S. F. Chem. Eur. J. 2000, 4, 581–588.
(21) Demadis, K. D.; Hartshorn, C. M.; Meyer, T. J. Chem. ReV. 2001,
101, 2655–2685.
(25) Touchard, D.; Haquette, P.; Guesmi, S.; Le Pichon, L.; Daridor,
A.; Toupet, L.; Dixneuf, P. H. Organometallics 1997, 16, 3640–3648.
(26) Younus, M.; Long, N. J.; Raithby, P. R.; Lewis, J.; Page, N. A.;
White, A. J. P.; Williams, D. J.; Colbert, M. C. B.; Hodge, A. J.; Khan,
M. S.; Parker, D. G. J. Organomet. Chem. 1999, 578, 198–209.
(22) Courmarcel, J.; Le Gland, G.; Toupet, L.; Paul, F.; Lapinte, C. J.
Organomet. Chem. 2003, 670, 108–122.
(23) Polam, J. R.; Porter, L. C. J. Coord. Chem. 1993, 29, 109–119.

Mixed-Valent Dinuclear Fe-Ru Complexes
Organometallics, Vol. 27, No. 6, 2008 1065
Scheme 1. Synthesis of 2 and 5 ([Fe] ) (η2-dppe)(η5-C5Me5)Fe and [Ru] ) trans-(η2-dppe)2Ru)
Scheme 2. Synthesis of 2[PF6] and 5[PF6] ([Fe] ) (η2-dppe)(η5-C5Me5)Fe and [Ru] ) trans-(η2-dppe)2Ru)
Table 1. Electrochemical Data for Selected Complexesa
E° (V)b
•-
complex
Ru(III)/Ru(II) Fe(III)/Fe(II) NO2/NO2
[Fe]Ct C(C6H5) (8-H)
-0.15c/-0.14c,e
Cl[Ru]Ct C(C6H5) (9-H) 0.44d/0.46d,e
(C6H5)Ct C[Ru]Ct
(4-C6H4NO2) (10)
[Fe]Ct C(C6H4)Ct
[Ru]Cl (2)
C
0.55d
0.41d
-1.25d
C
-0.24d
-0.23d
[Fe]Ct C(C6H4)Ct C[Ru] 0.50d
Ct C(4-C6H4NO2) (5)
-1.25d
Figure 1. ESR spectrum of 5[PF6] in CH2Cl2/1,2-C2H4Cl2 (1:1) at
80 K.
a [Fe] ) (η2-dppe)(η5-C5Me5)Fe and [Ru] ) trans-(η2-dppe)2Ru.
b All E° values in V vs SCE. Conditions: CH2Cl2 solvent, 0.1 M (NnBu4)-
for the dinuclear complexes 2 and 5.31 Following the addition
of the oxidant, the solutions were immediately frozen and the
samples were kept at 80 K until the ESR measurements were
completed. Similar ESR spectra were obtained from isolated
samples of 2[PF6] and 5[PF6], obtained according to Scheme
2. In each case, rhombic signals exhibiting three g tensors were
observed (Figure 1 and Table 2), in line with the presence of
metal-centered radicals.12,29 The presence of some 31P hyperfine
structure (ca. 13 G) is apparent on the g3 tensor of 5[PF6] (Figure
1).32 This feature is also observed, but less resolved, in the case
of 2[PF6].
(PF6) supporting electrolyte, 20 °C, Pt electrode, sweep rate 0.100 V s-1
.
c The ferrocene/ferrocenium (Fc/Fc+) couple was used as an internal
reference for potential measurements. d The [Fe]Ct C(4-C6H4NO2)
complex was used as an internal reference for potential measurements.27
e Measurement carried out in acetonitrile.
a region where Fe(III)28 or Ru(III)29,30 acetylide stretches
typically appear.
The ESR spectra of these compounds in solvent glasses were
then recorded along with the spectra of the monomeric
compounds 8-H[PF6], 9-H[BF4], and 10[BF4], the latter being
used as reference compounds. All of these radical cations were
chemically generated in situ in CH2Cl2/1,2-C2H4Cl2from their
neutral parents at ambient temperature. Acetylferrocenium
tetrafluoroborate was used to oxidize the monomeric complexes
9-H and 10, whereas ferrocenium hexafluorophosphate was used
In the case of compounds 2 and 2[PF6], the Mössbauer spectra
were also recorded at 80 K. Oxidation is accompanied by a
decrease of the 57Fe isomer shift from 0.26(1) to 0.25(2) mm
s-1 and by a concomitant decrease of the quadrupolar splitting
from 1.97(7) to 0.96(8) mm s-1
.
In order to further investigate the electronic structure of these
MV complexes, a variable-temperature (VT) H NMR study
1
(27) Costuas, K.; Paul, F.; Toupet, L.; Halet, J.-F.; Lapinte, C.
Organometallics 2004, 23, 2053–2068.
(28) Paul, F.; Mevellec, J.-Y.; Lapinte, C. Dalton Trans. 2002, 1783–
1790.
(31) Connelly, N. G.; Geiger, W. E. Chem. ReV. 1996, 96, 877–910.
(32) Connelly, N. G.; Gamasa, M. P.; Gimeno, J.; Lapinte, C.; Lastra,
E.; Maher, J. P.; Le Narvor, N.; Rieger, A. L.; Rieger, P. H. J. Chem. Soc.,
Dalton Trans. 1993, 2575–2578.
(29) Paul, F.; Ellis, B. J.; Bruce, M. I.; Toupet, L.; Roisnel, T.; Costuas,
K.; Halet, J.-F.; Lapinte, C. Organometallics 2006, 25, 649–665.
(30) Adams, C. J.; Pope, S. J. A. Inorg. Chem. 2004, 43, 3492–3499.

1066 Organometallics, Vol. 27, No. 6, 2008
Gauthier et al.
Table 2. ESR Data for Compounds in Frozen CH2Cl2/1,2-C2H4Cl2 Solutions at 80 Ka
b
b
b
complex
g1
g2
g3
∆gb
ref
[[Fe]Ct C(C6H5)][PF6] (8-H[PF6])
[Cl[Ru]Ct C(C6H5)][BF4] (9-H[BF4])
[(C6H5)Ct C[Ru]Ct C(4-C6H4NO2)][BF4] (10[BF4])
[[Fe]Ct C(C6H4)Ct C[Ru]Cl][PF6] (2[PF6])
[[Fe]Ct C(C6H4)Ct C[Ru]Ct C(4-C6H4NO2)][PF6] (5[PF6])
2.46(4)
2.51(9)
2.48(9)
2.33(0)
2.33(7)
2.03(3)
2.03(9)
2.03(6)
2.03(9)
2.03(7)
1.97(5)
1.89(6)
1.89(6)
1.98(9)
1.98(8)c
0.48(9)
0.62(3)
0.59(3)
0.34(1)
0.34(9)
12
this work
this work
this work
this work
a [Fe] ) (η2-dppe)(η5-C5Me5)Fe and [Ru] ) trans-(η2-dppe)2Ru. b The last digit is given in parentheses due to the experimental uncertainty of the
c
measurements. APRu ≈ 13.3 G.
a
Table 3. UV–vis Data for Selected Complexes in CH2Cl2
complex
abs (nm) (10-3ꢀ (M-1 cm-1))
ref
[Fe]Ct C(C6H5) (8-H)
[[Fe]Ct C(C6H5)][PF6] (8-H[PF6])
277 (sh, 14.5), 350 (13.6)
261 (sh, 32.6), 280 (sh, 27.4), 301 (sh, 18.8), 342 (sh, 5.9), 379 (sh,
3.6), 575 (sh, 2.3), 662 (3.1)
27
12
Cl[Ru]Ct C(C6H5) (9-H)
(C6H5)Ct C[Ru]Ct C(4-C6H4NO2) (10)
[Fe]Ct C(C6H4)Ct C[Ru]Cl (2)
[[Fe]Ct C(C6H4)Ct C[Ru]Cl][PF6] (2[PF6])
[Fe]Ct C(C6H4)Ct C[Ru]Ct C(4-C6H4NO2) (5)
[[Fe]Ct C(C6H4)Ct C[Ru]Ct C(4-C6H4NO2)][PF6] (5[PF6])
319 (23.0)
10
240 (63.1), 266 (sh, 32.7), 316 (22.9), 486 (21.2)
330 (sh, 24.0), 378 (32.3)
362 (23.5), 448 (sh, 12.1), 498 (15.1), 1124 (16.1)
328 (26.7), 378 (32.3), 490 (22.0)
this work
this work
this work
this work
this work
354 (28.5), 464 (sh, 33.0), 493 (36.2), 1080 (18.3)
a [Fe] ) (η2-dppe)(η5-C5Me5)Fe and [Ru] ) trans-(η2-dppe)2Ru.
Chart 2. Mononuclear Model Complexes Used throughout This Work
Scheme 3. Synthesis of 11 ([Ru] ) trans-(η2-dppe)2Ru)
to those of their neutral parents 2 and 5. The latter show a couple
of strong absorptions in the visible range, centered at ca. 330
and 380 nm. In addition, 5 also exhibits an absorption at 490
nm (Table 3). These absorptions are responsible for the red-
orange color of the neutral parents. After monooxidation, a new
and strong absorption appears in the near-IR range near 1100
nm for both 2[PF6] and 5[PF6], and the absorptions previously
observed near 330 and 380 nm are now red-shifted by ca. 120
nm and weakened in intensity. As a result, the oxidized
complexes are red-brown. It is noteworthy that the low-energy
absorption of the nitro complex 5 at 490 nm is apparently not
significantly affected by oxidation.
1
was undertaken with 2[PF6]. All H signals detected could be
assigned by comparison with previous results obtained with the
mononuclear model compounds 8-X[PF6].33 Except for H3 and
H4, which overlap with the dppe protons on ruthenium, the
temperature-dependent shifts of the signals of the protons
H1-H11 correspond to those of previous observations (Figure
2); the dppe protons on ruthenium are only slightly affected by
temperature. 31P NMR spectra were also recorded for 2[PF6]
in the +400 to -3500 ppm range. A broad signal (ν1/2 ) 700
( 100 Hz) is present in the diamagnetic region, above 100 ppm,
but no additional signal could be detected at high field, where
the dppe phosphorus atoms bound to Fe(III) are expected to
appear.33 The former signal is thus tentatively assigned to the
four dppe phosphorus atoms bound to ruthenium. Very similar
features were observed in the 1H and 31P NMR spectra of 5[PF6].
Notably, the protons belonging to the 4-nitrophenylacetylide
ligand on ruthenium were not detected outside the diamagnetic
region (0–10 ppm), consistent with a small spin density residing
on the ruthenium center in this complex.
Given the possibility that we would observe an intervalence
charge-transfer (IVCT) transition with these dinuclear MV
complexes, particular attention was paid to the low-energy
absorptions detected for 2[PF6] and 5[PF6]. These absorptions
are very intense and exhibit a composite line shape, with the
occurrence of several underlying processes (Figure 3a). An
overall negative solvatochromism (∆νj ≈ 900 cm-1) is evident
for this absorption in 2[PF6], the most intense contribution being
clearly blue-shifted in more polar solvents (Figure 4). Regardless
of the solvent used, the near-IR absorption of 2[PF6] can be
deconvoluted into three Gaussian-shaped contributions, two of
which (II and III) at higher energies are always separated by
ca. 2200 cm-1 and appear much more solvent-sensitive than
the third one (I) near 5900 cm-1 (Table 4). Similar to the case
for 2[PF6], the intense near-IR absorption of 5[PF6] can be
deconvoluted into three Gaussian contributions resembling those
observed for 2[PF6] (Figure 3b and Table 4).
Electronic Spectroscopy of 2, 2[PF6], 5, and 5[PF6]. The
electronic spectra of the MV complexes 2[PF6] and 5[PF6] were
then investigated in the UV–vis-near-IR range and compared
Synthesis and Characterization of Cl(η2-dppe)2Ru
[Ct C(3-C6H4F)] (11). The new mononuclear complex Cl(η2-
dppe)2Ru[Ct C(3-C6H4F)] (11) was needed in order to deter-
mine the electronic substituent parameter (ESP) of the “-(Ct C)-
(33) Paul, F.; da Costa, G.; Bondon, A.; Gauthier, N.; Sinbandhit, S.;
Toupet, L.; Costuas, K.; Halet, J.-F.; Lapinte, C. Organometallics 2007,
26, 874–896.

Mixed-Valent Dinuclear Fe-Ru Complexes
Organometallics, Vol. 27, No. 6, 2008 1067
Scheme 4. The Two Redox Isomers of the MV Complexes 2+
(X ) Cl) and 5+ (X ) Ct C(4-C6H4NO2)) in the GS
Chart 3. Mononuclear Model Complexes Used to Derive the
ESPs of the Organoruthenium Substituent
diffractometry confirms the structure of complex 11 (Figure 5).
The complex crystallizes in the monoclinic P21/n space group
with one molecule of complex in the asymmetric unit and two
dichloromethane solvates (see the Experimental Section for
details). The bond lengths and angles of this complex are
unexceptional, in comparison with available X-ray data on
related mononuclear Ru(II) acetylide complexes,26,34 apart from
the rather short length of the Ct C triple bond (ca. 1.10 Å).
The latter feature might, however, be an artifact resulting
from the poor quality of the diffraction data set (Rint ) 0.0869).
Figure 2. Temperature dependence of the 1H NMR shifts of 2[PF6]
in CD2Cl2 with proposed assignment for selected protons.
Discussion
Electronic Structure of the Ground State (GS) of the
MV Complexes 2[PF6] and 5[PF6]. While the rhombic ESR
spectra obtained for 2[PF6] and 5[PF6] at 80 K evidence the
metal-centered nature of the unpaired electron in these com-
pounds, unambiguous assignment of the location of the unpaired
electron is not possible from these data alone, since very similar
rhombic spectra were obtained for the mononuclear iron(III)
(8-H[PF6]) and ruthenium(III) (9-H[BF4], 10[BF4]) complexes
used as references (Table 2). However, the slight change in
anisotropy (∆g) between 2[PF6] and 5[PF6] reveals that the
Ru(η2-dppe)2Cl” organoruthenium fragment by 19F NMR (see
below). This complex was thus synthesized from the triflate
precursor complex 7[OTf] (Scheme 3), following an original
synthesis inspired from that used to obtain 2. The desired
complex was isolated in fair yields and fully characterized by
the usual means (see the Experimental Section).
Yellow crystals of 11 · 2CH2Cl2 could be grown by slow
evaporation of the solvent from a dichloromethane solution of
the complex. The solid-state structure obtained by X-ray
Table 4. Near-IR Data for 2[PF6] and 5[PF6] in Dichloromethane and Acetonitrilea
νjmax (cm-1 b
(ꢀ (M-1 cm-1))c
)
[Fe+]Ct C(C6H4)Ct C[Ru]X
X ) Cl (2+)
band
I
(∆νj1/2
)
(cm-1
)
∆G° (cm-1
)
(∆νj1/2
)
(cm-1 c,d
theor
)
Hab (cm-1 b
)
exptl
5900 (1750)
1500
4760
4840e
4760
4840e
4760
4840e
1620
5950 (1100)e
8900 (16100)
9500 (12600)e
11100 (1100)
11750 (1700)e
1500e
2670
1600
3090
3280
3830
4000
II
1060f/2070g
2830e
2670
1000e,f/2330e,g
III
2830e
X ) Ct C(C6H4)NO2 (5+)
I
II
III
5900 (1200)
9230 (18100)
11800 (1150)
1500
2660
2750
5650
5650
5650
760
2880
3770
1150f/1790g
a [Fe] ) (η2-dppe)(η5-C5Me5)Fe and [Ru] ) trans-(η2-dppe)2Ru. b Values (50 cm-1 c Values (5%. d Calculated following eq 1. Note that these
.
values are only appropriate for IVCT bands; they are included here to illustrate the incompatibility between the experimental half-widths measured for
some bands and those calculated from the classical Hush model. e Values found in CH3CN. f Calculated following eq 2 with dab ) 12 Å. g Calculated
for a class III MV complex ([νmax - ∆G°]/2).

1068 Organometallics, Vol. 27, No. 6, 2008
Gauthier et al.
unpaired electron is only slightly affected by the change in the
apical ligand on ruthenium, suggesting at most a weak delo-
calization to this metal center.35
Stronger indications about the localization of the electronic
hole/unpaired electron in these MV complexes can be gathered
from CV data. The first oxidation potentials of 2 and 5 at -0.24
and -0.23 V, respectively, vs the saturated calomel electrode
(SCE) are more cathodic than that observed for the mononuclear
iron model complex 8-H (-0.15 V vs SCE)24 but are even more
cathodic than those of the corresponding mononuclear ruthenium
models 9-H10,26,34 (0.44 vs SCE) and 1026 (0.55 vs SCE). Thus,
the oxidation likely corresponds to an iron-centered process in
2 and 5.
Firmer evidence for the dominant localization of the electronic
vacancy on iron in 2[PF6] or 5[PF6] comes from the NMR and
1
Mössbauer investigations performed on 2[PF6]. First, the H
NMR spectrum of 2[PF6] is diagnostic of an iron-centered
radical. Thus, the proton signals are significantly shifted
compared to their corresponding values in the diamagnetic
parent 2 when located in the coordination sphere of iron,
whereas they stay very close to their position in the spectrum
of the diamagnetic complex 9-H when located in the coordina-
tion sphere of ruthenium, in line with a negligible spin
delocalization to this metal center.33 In addition, the quadrupolar
separation (∆Q) value of 0.97 ( 0.01 of the 57Fe Mössbauer
doublet for 2[PF6] is also diagnostic of a Fe(III) center in the
solid state.36 Thus, the electronic structure of 2[PF6] or 5[PF6]
in the GS might be more accurately described by the valence
bond (VB) isomer A instead of B (Scheme 4).
Figure 3. Deconvolution of the broad near-IR absorption (a) at
1124 nm for 2[PF6] and (b) at 1080 nm for 5[PF6] in CH2Cl2.
Figure 4. Solvatochromic behavior of the broad near-IR absorption
for 2[PF6]. The absorption recorded in THF has been omitted for
clarity.
On the basis of Hammett correlations previously established,27
2 or 5 can be considered as mononuclear compounds such as
8-X, in which a strongly electron-releasing organometallic
substituent is appended to the arylacetylide ligand in place of
X. From the linear free energy relationships (LFERs) involving
the first oxidation potentials, Hammett σp electronic substituent
parameters (ESPs) of -0.54 ( 0.06 and of -0.48 ( 0.06 would
be obtained for the “-(Ct C)Ru(η2-dppe)2Cl” and “-(Ct C)-
Ru(η2-dppe)2[Ct C(4-C6H4NO2)]” substituents, respectively.
The first of these values is, however, significantly larger than
the σp value of -0.34 ( 0.04, which can be derived indepen-
dently for the “-(Ct C)Ru(η2-dppe)2Cl” organoruthenium frag-
ment by 19F NMR from the p- (9-F) and m-fluoro derivatives
(11; Chart 3), according to the method originally proposed by
Taft and co-workers.37 This latter value actually suggests that
the “-(Ct C)Ru(η2-dppe)2Cl” substituent should be slightly
more electron-releasing than a methoxy group in 2; this
statement is also supported by the anisotropy of the ESR
spectrum of 2[PF6], on the basis of correlations between the
anisotropy and the σ+ ESPs previously established for 8-X[PF6]
(34) (a) Hurst, S. K.; Cifuentes, M. P.; Morrall, J. P. L.; Lucas, N. T.;
Whittall, I. R.; Humphrey, M. G.; Asselberghs, I.; Persoons, A.; Samoc,
M.; Luther-Davies, B.; Willis, A. C. Organometallics 2001, 20, 4664–4675.
(b) Morrall, J. P.; Cifuentes, M. P.; Humphrey, M. G.; Kellens, R.; Robijns,
E.; Asselberghs, I.; Clays, K.; Persoons, A.; Samoc, M.; Willis, A. C. Inorg.
Chim. Acta 2006, 359, 998–1005.
Figure 5. ORTEP plot of 11. Thermal ellipsoids are at the 30%
probability level. Hydrogen atoms have been omitted for clarity.
radicals and considering that the electronic vacancy in 2[PF6]
is mostly located on iron (Scheme 4).12
(35) In line with that statement, the poorly resolved 31P hyperfine
structure apparent on the g3 tensor in the ESR spectrum of 5[PF6] (Figure
1) is more reminiscent of a triplet than of a quintet, suggesting that the
unpaired electron mostly sees two phosphorus atoms rather than four, in
accordance with its dominant localization on the iron center.
(36) Argouarch, G.; Thominot, P.; Paul, F.; Toupet, L.; Lapinte, C. C. R.
Chim. 2003, 6, 209–222.
Thus, at least for 2, and solely on the basis of E° values, it
seems that the electron-releasing capability of the organoruthe-
nium unit is slightly overestimated in the LFERs established
for the model complexes 8-X. Conversely, this also means that
the E° value corresponding to the Fe(II)/Fe(III) oxidation is
too large to result solely from inductive/mesomeric substitu-
ent effects. We will come back to this statement later on.
Note also that the potential separation ∆E° between the two
(37) (a) Taft, R. W.; Price, E.; Fox, I. R.; Lewis, I. C.; Andersen, K. K.;
Davies, G. T. J. Am. Chem. Soc. 1963, 85, 3146–3156. (b) Taft, R. W.;
Price, E.; Fox, I. R.; Lewis, I. C.; Andersen, K. K.; Davies, G. T. J. Am.
Chem. Soc. 1963, 85, 709–724.

Mixed-Valent Dinuclear Fe-Ru Complexes
Organometallics, Vol. 27, No. 6, 2008 1069
reversible redox potentials corresponding to the metal-
centered oxidations in this compound (0.65 V) is quite close
to the difference between the redox potentials of the
corresponding oxidations in the monomer model complexes
8-H and 9-H (0.59 V), consistent with the redox centers in
2[PF6] not being so strongly coupled that it renders the Hush
model inapplicable.
Electronic Delocalization in MV Complexes 2[PF6] and
5[PF6]. On the basis of the data available for 8-H, 9-H, and 10
(Table 3), the low-energy absorptions observed in the range
330–500 nm most likely correspond to MLCT transitions for 2
and 5.10,26,27,34 More specifically, by comparison with the low-
energy transition observed for 10 at 486 nm (Table 3), the low-
energy transition of 5 observed at 493 nm is proposed to
rendered more electron-poor. This observation, along with the
fact that both transitions have the same half-widths, suggests
that contributions II and III are associated with similar types of
electronic transitions in 2[PF6] and 5[PF6] and confirms that
they involve the ruthenium center to some extent, in line with
their purported IVCT nature.
Note that low-energy transitions in the 600–900 nm range
have also been observed for the Fe(III) model complexes
8-X[PF6], featuring electron-releasing substituents.12 Such ab-
sorptions were observed at 718 nm for the methoxy-substituted
complex (X ) OMe) and at 894 nm for the dimethylamino
substituent (X ) NMe2) and have been previously attributed to
LMCT transitions. A closer inspection reveals, however, that
the most intense contribution (i.e., band II) in both 2[PF6] and
5[PF6] is significantly more intense and broader than the low-
energy transitions previously observed for 8-X[PF6] (νj1/2 < 1800
cm-1 for X ) OMe and νj1/2 < 2400 cm-1 for X ) NMe2). In
addition, the solvatochromic behavior of contributions II and
III for 2[PF6] is markedly more pronounced than that of the
low-energy bands of 8-OMe[PF6] (∆νj ≈ 200 cm-1) or
8-NMe2[PF6] (∆νj ≈ 370 cm-1), which can be rationalized by
the enhanced charge transfer taking place in 2[PF6] and 5[PF6].
All these observations strongly suggest that the nature of the
near-IR transitions observed for the heterodinuclear MV com-
plexes 2[PF6] and 5[PF6] might be somewhat different from
those corresponding to the low-energy transitions previously
observed with 8-X[PF6].
correspond to a dRu f (π*)ArNO excitation. Consistent with the
2
localization of the unpaired electron being predominantly on
iron (Scheme 4), this electronic transition does not seem to be
strongly affected by oxidation. According to previous work with
the mononuclear Fe(III) complexes 8-X[PF6], the LMCT
transitions πAr f dFe are expected to arise in the same energy
range for 2[PF6] and 5[PF6].12
The appearance of an intense absorption in the near-IR range
for both MV complexes is more puzzling. On the basis of the
data available for the corresponding symmetric MV complexes
1[PF6] and 3[PF6]4,13 and of the ∆G° gap deduced from the
first oxidation potentials of the model complexes 8-H and 9-H
by cyclic voltammetry (4760 cm-1),38 an IVCT is expected in
this spectral region on the basis of the Hush model.39 Indeed,
considering the λFe ≈ 4000 cm-1 and λRu ≈ 6500 cm-1 values
previously reported for the reorganization energies of 1[PF6]4
and 3[PF6],13 respectively, the IVCT band should be found
at around 10 000 cm-1 for 2[PF6]. This value is reasonably
close to the intense absorption detected for this compound
at ca. 8890 cm-1 (1124 nm). In addition, the negative
solvatochromism exhibited by this absorption (Figure 4)
would also be consistent with an IVCT transition of a weakly
coupled MV complex. However, since this absorption exhibits
a composite shape (Figure 3a), the various underlying
contributions must now be identified in order to further
support our hypothesis.
(∆ν1/2)theor ) [2310(νmax - ∆G°)]1/2
Hab ) (2.06 × 10-2 ⁄ dab)(ꢀmax max∆ν1⁄2)1⁄2
(1)
(2)
ν
According to the Hush model applied to class-II unsym-
metrical MV complexes (eq 1), the theoretical bandwidth of an
IVCT transition is related to ∆G°, the energy gap between the
two potential wells corresponding to the two diabatic redox
states A and B (Scheme 4) on the reaction coordinate axis, and
to νjmax, the energy of the peak maximum.2,5 For 2[PF6], this
quantity (∆G°) was estimated from the difference in the
oxidation potentials of the monomeric model complexes 8-H
and 9-H in dichloromethane and amounts to ca. 4760 cm-1 (see
above). As shown in Table 4, the theoretical half-width
The contribution at lowest energy (I) probably corresponds
to a d-d ligand field (LF) transition. Such transitions are usually
observed with related Fe(III) complexes.4,12 The weak solva-
tochromism of this contribution established with 2[PF6], as well
as its spectral characteristics in 2[PF6] and 5[PF6], is consistent
with this attribution.
(∆νj
that found experimentally for curve II [(νj1/2 exptl
corresponds well with the (νj1/2 exptl value measured experimen-
)
1/2 theor of the IVCT computed according to eq 1 is close to
)
II] and also
)
tally for the whole near-IR absorption of 2[PF6] (Table 4). This
brings further credence to its assignment as an IVCT transition.
Another experimental observation that supports such an assign-
ment is the fact that band II becomes broader and is also shifted
to higher energy in more polar solvents, a trend in line with
predictions of eqs 1 and 3.2,39 In contrast, the third contribution
(III) is too narrow to correspond to another IVCT transition in
the GS but could likely correspond to an IVCT transition
involving an excited state of 2[PF6], as observed previously with
1[PF6].2,40 Such transitions often result when low-lying LF states
are present.
Additional insight about the nature of the two remaining
contributions is available by comparing the deconvolutions
obtained for 2[PF6] and 5[PF6] (Table 4). Thus, replacement of
the apical chloride atom by a less π-donating ligand such as
the 4-nitrophenylacetylide should have a sizable influence on
any transition involving the Ru(II) center, such as an IVCT
transition, but should affect to a much lesser extent any transition
involving the bridging ligand, such as an LC or LMCT
transition. In accord with our hypothesis, the band I at lowest
energy is shifted very slightly on proceeding from 2[PF6] to
5[PF6], as might be expected for a LF transition centered on
the Fe(III) terminus. As is the case with 2[PF6], contributions
II and III are separated by ca. 2500 cm-1 in 5[PF6] but are
now shifted toward higher energies by ca. 350 cm-1, as would
be expected for IVCT transitions when the Ru(II) center is
Assuming that band II corresponds to an IVCT transition,
the electronic coupling (Hab) of the MV complexes 2[PF6] and
5[PF6] amounts to 1060 ( 50 and 1150 ( 50 cm-1 (Table 1),
respectively, according to eq 2, when the Fe-Ru (dab) distances
(40) Note that this weak contribution (III) is required by the slight
asymmetry of the band shape of the near-IR absorption. It could also be an
artifact resulting from an unresolved vibronic progression. See for
instance: Bailey, S. E.; Zink, J. I.; Nelsen, S. F J. Am. Chem. Soc. 2003,
125, 5939–5947.
(38) ∆G° ) e[E°9-H - E°8-H] (in this expression, e represents the charge
of the electron).
(39) Creutz, C. Prog. Inorg. Chem. 1983, 30, 1–73.

1070 Organometallics, Vol. 27, No. 6, 2008
Gauthier et al.
are taken as 12 Å.41 Despite the sizable uncertainties associated
with the spectroscopic measurements and the sensitivity of the
compounds, these values provide support for the proposition
that quite similar electronic couplings are operative in 2[PF6]
and 5[PF6].
The solvent has only a minor influence on the magnitude of
the electronic coupling (Hab) in 2[PF6]. Thus, proceeding from
dichloromethane to acetonitrile does not significantly affect the
stability of the MV complex 2[PF6], as evidenced by the similar
∆E° values found (Table 2), nor does it modify the electronic
coupling (see Table 4).
When the entropy associated with the solvent reorganization
is neglected, an estimate of λ can then be derived for these
heterodinuclear MV compounds using eq 3 and ∆G° (Table
4). For both compounds the reorganization energy is larger than
twice the electronic coupling (2Hab < λ); thus, these compounds
can be considered as class II MV complexes in the classification
of Robin and Day,42 as is also suggested from electrochemical
data and from the negative solvatochromic behavior exhibited
by the IVCT transition of 2[PF6].2 Considering that 2Hab/λ >
Figure 6. Solvent dependence of the near-IR band maximum (νmax
)
for 2[PF6]: (a) toluene (not included in the correlation coefficient);
(b) thf; (c) dichloromethane; (d) acetone; (e) acetonitrile.
model compounds,43 the rest of the difference being attributable
to synergistic, statistical, and magnetic contributions.44
λout ) (∆e)2[(1 ⁄ 2aFe) + (1 ⁄ 2aRu) - (1 ⁄ dab)](1 ⁄ Dop - 1 ⁄ Ds)
(5)
[1 - (∆νj1/2)theor/2λ], they can be further categorized as class
Finally, we have also plotted νmax against the solvent function
(Figure 6). If the point corresponding to toluene is not
considered,45 a linear relationship is found for the most polar
solvents, in accordance with the Hush model (eqs 3 and 4).
According to eq 5, the intercept at 7310 cm-1 corresponds to
λin + ∆G°, which permits us to obtain an approximate value
for the internal contribution to the reorganization energy λin )
2550 cm-1. This estimate seems somewhat high for the internal
reorganization energy according to theoretical and experimental
estimates available in the literature for purely “inorganic” MV
complexes.39 However, it is well-known that such extrapolations
often overestimate the true value of λin. Alternatively, the high
value presently found might also reflect the large “organic”
character of the “ruthenium-centered” excited-state of 2[PF6].
In comparison, λout ) 1590 cm-1 constitutes a smaller com-
ponent of the reorganization energy involved during the electron
transfer in 2[PF6].
IIA MV compounds according to the subcategorization devel-
oped for symmetric MV complexes by Brunschwig et al.19
However, because the sizable electronic coupling of these MV
complexes is certainly somewhat underestimated by the Hush
treatment, they might be close to borderline class IIA/class IIB
MV complexes. In other words, they are strongly coupled MV
complexes with a “localized” valency on the NMR time scale
in the GS.21
While the strong intensity of the bands II and III can appear
surprising for IVCTs in localized MV complexes such as 2[PF6]
and 5[PF6], it is quite likely that the strong electronic coupling
and the less stringent symmetry restrictions operative for these
nonsymmetric compounds are responsible for the strong intensity
and asymmetric shape of the bands. In these unsymmetrical MV
complexes, charge transfer occurs in the same direction for the
LMCT and the IVCT transitions, and symmetry rules do not
forbid the mixing of these transitions; the present data are thus
consistent with the low-energy “LMCT” transitions previously
observed in mononuclear Fe(III) complexes 8-X[PF6] acquiring
an increasing “substituent” character as X becomes more and
more able to accommodate an electronic vacancy, eventually
transforming into IVCT transitions for the heterodinuclear MV
compounds 2[PF6] and 5[PF6].
Conclusion
We have shown in this contribution that the intense absorp-
tions present in the near-IR region (near 1100 nm) of electronic
spectra of unsymmetrical MV complexes such as 2[PF6] and
5[PF6] likely correspond to IVCT transitions. In contrast to
related symmetrical dinuclear MV compounds featuring only
electron-rich organoruthenium end groups such as 3[PF6], the
spectral features of these bands in the unsymmetrical complexes
investigated herein obey predictions based on the Hush model.
This behavior might be ascribed to the presence of the electron-
rich iron end group “(η2-dppe)(η5-C5Me5)Fe”, which tends to
localize the electronic vacancy on the iron in the GS. On the
basis of the assumption that the near-IR transition corresponds
to an IVCT transition, an effective electronic coupling of ca.
1100 cm-1 can be derived between this end group and the
second Ru-based redox-active end group “-(Ct C)Ru(η2-
dppe)2(X)” (X ) Cl, Ct C(4-C6H4NO2)) in both 2[PF6] and
νmax ≈ λ + ∆G°
(3)
∆Gdeloc ≈ (Hab)2 ⁄ (λ + ∆G°) ≈ (Hab)2 ⁄ νmax
(4)
Using eq 45,39 and the available data to derive an estimate of
the contribution of the electronic coupling to the separation
between the redox waves (∆E deloc ) ∆G deloc/e), we find values
of ca. 0.015 V for 2[PF6] and 0.018 V for 5[PF6]. Despite the
experimental uncertainties associated with the redox potentials
under our conditions ((0.01 V), it is clear that the electronic
coupling contributes a significant part of the small increase in
the separation between the first and second oxidation potentials
in the dinuclear complexes, relative to those of the mononuclear
(43) ∆E°1 – [E°6-H - E°8-H] ) 0.06 V and ∆E°5 – [E°11 - E°8-H] )
0.03 V.
(41) In eq 3, ꢀmax, νjmax and ∆νj1/2 are the extinction coefficient, the energy
of the maximum, and the half-width of the IVCT band, respectively. The
Fe-Ru (dab) distance was estimated from DFT modeling and from X-ray
data on related complexes.
(42) Robin, M. B.; Day, P. AdV. Inorg. Chem. Radiochem. 1967, 10,
247–422.
(44) Evans, C. E. B.; Naklicki, M. L.; Rezvani, A. L.; White, C. A.;
Kondratiev, V. V.; Crutchley, R. J. J. Am. Chem. Soc. 1998, 120, 13096–
13103.
(45) We have presently no explanation for this discrepancy, but specific
solute-solute or solute-solvent interactions might be at the origin of the
“higher-than-expected” νjmax value found.

Mixed-Valent Dinuclear Fe-Ru Complexes
Organometallics, Vol. 27, No. 6, 2008 1071
in dichloromethane. Stirring was maintained for 1.5 h at 20 °C,
and the solution was concentrated in vacuo. Addition of diethyl
ether resulted in the precipitation of 2[PF6], which could be isolated
(0.138 g, 0.077 mmol, 60%) after subsequent washings with toluene
(5 mL), diethyl ether (10 mL), and n-pentane (10 mL). FT-IR (ν,
KBr/CH2Cl2, cm-1): 2034/2038 (m, RuCt C), 1943/1947 (s,
FeCt C). 31P NMR (81 MHz, CDCl3, δ in ppm): 103.3 (broad s,
P, Ru(dppe)2). 1H NMR (500 MHz, CD2Cl2, δ in ppm, numbering
according to Figure 2): 29.1 (s, 2H, H2), 9.6 (broad s, 2H, H9),
8.1–6.8 (m, 52H, HAr/dppe), 4.4 (s, 4H, H6 or H7), 4.2 (s, 4H, H6 or
H7), 2.6 (s, 8H, (CH2)dppe/Ru), -4.0 (s, 2H, H10), -9.0 (s, 15H, H11),
-66.8 (s, 2H, H1). Mössbauer (mm s-1 vs 57Fe, 80 K): IS 0.25(2),
QS 0.96(8).
5[PF6]. This value would unambiguously categorize these open-
shell species as sizably coupled class II MV complexes in the
classification of Robin and Day.
Experimental Section
General Data. All manipulations were carried out under an inert
atmosphere. Solvents and reagents were used as follows: Et2O and
n-pentane, distilled from Na/benzophenone; CH2Cl2, distilled from
CaH2 and purged with Ar, opened/stored under Ar. The [(η5-
C5H5)2Fe][PF6] ferrocenium salt was prepared by previously
published procedures.31 High-field NMR spectra experiments were
performed on a multinuclear Bruker 500, 300, or 200 MHz
instrument (AVANCE 500, AM300WB, and 200DPX). Chemical
shifts are given in parts per million relative to tetramethylsilane
(TMS) for 1H and 13C NMR spectra, external H3PO4 for 31P NMR
spectra, and external fluorotrichloromethane for 19F NMR spectra.
Transmittance-FTIR spectra were recorded using a Bruker IFS28
spectrometer (400–4000 cm-1). Near-IR and UV–visible spectra
were recorded using a Cary 5000 spectrometer. ESR spectra were
recorded on a Bruker EMX-8/2.7 (X-band) spectrometer. The
Mössbauer spectra were recorded with a 2.5 × 10-2 C (9.25 ×
108 Bq) 57Co source using a symmetric triangular sweep mode.46
MS analyses were performed at the “Centre Regional de Mesures
Physiques de l’Ouest” (CRMPO, University of Rennes) on a high-
resolution MS/MS ZABSpec TOF Micromass spectrometer. El-
emental analyses were performed at the “Centre Regional de
Mesures Physiques de l’Ouest” (CRMPO, University of Rennes).
The solid-state structure (X-ray) was resolved at the “Centre de
Diffractométrie X” (UMR CNRS 6226, University of Rennes).
Unless specified, all reagents were of commercial grade. The
complexes 6,22 7,23 9-H,25 9-F34b and 1025 were obtained according
to reported syntheses.
Synthesis of (η2-dppe)(η5-C5Me5)Fe[Ct C-1,4-(C6H4)Ct C]-
Ru(η2-dppe)2[Ct C(4-C6H4NO2)] (5). In a Schlenk tube, (η2-
dppe)(η5-C5Me5)Fe[Ct C-1,4-(C6H4)Ct C]Ru(η2-dppe)2Cl (2; 0.250
g, 0.152 mmol), 4-nitrophenylacetylene (0.089 g, 0.608 mmol), and
KPF6 (0.084 g, 0.456 mmol) were dissolved in 30 mL of
dichloromethane and the mixture was stirred for 2 days at room
temperature. After removal of the solvent, the red residue was
dissolved in THF in the presence of excess KOtBu (0.034 g, 0.304
mmol) and stirred for 2 h. After evacuation of the solvent, the
product was chromatographed through an alumina column using
toluene as eluant. The fraction containing the product was then
concentrated to dryness and the residual solid washed with
n-pentane (3 × 5 mL) to afford the desired red complex 5 (0.130
g, 0.074 mmol, 49%). Anal. Calcd for C106H95FeNO2P6Ru: C, 72.43;
H, 5.45. Found: C, 73.19; H, 5.85. MS (ESI): m/z 1757.4182 (M•+),
calcd for C106H95FeNO2P6Ru 1757.4263 (M•+). FT-IR (ν, KBr,
cm-1): 2070 (sh), 2044 (FeCt C and RuCt C’s). 31P NMR (81
MHz, C6D6, δ in ppm): 101.7 (s, 2P, (dppe)Fe), 54.6 (s, 4P,
1
(dppe)2Ru). H NMR (200 MHz, C6D6, δ in ppm): 8.12 (m, 6H,
HAr), 7.89 (m, 8H, J ) 6 Hz, HAr), 7.40–6.74 (m, 52H, HAr), 6.62
(d, 2H, J ) 8.4 Hz, HAr), 2.84 (m, 2H, CH2), 2.52 (m, 8H, CH2),
1.99 (m, 2H, CH2), 1.59 (s, 15H, C5(CH3)5). 13C{1H} NMR (125
MHz, C6D6, δ in ppm): 154.6 (q, 2JCP ) 14.4 Hz, RuCt C), 144.1
(s, CAr), 140.6–137.0 (m, CAr/dppe + FeCt C), 135.8–135.0 (m,
CHAr/dppe + CAr), 131.0–128.1 (m, CHAr/dppe + CHAr + RuCt C),
126.2, 124.3, 121.8, 120.6, 119.4 (s, CAr and MCt C’s), 88.5 (s,
C5(CH3)5), 32.4 and 31.9 (m, (CH2)dppe/Ru&Fe), 11.2 (s, C5(CH3)5).
CV (CH2Cl2, 0.1 M n-Bu4N+PF6-, 20 °C, 0.1 V s-1; E° in V vs
SCE (∆Ep in V, ip,a/ip,c)): -1.25 (0.14, 1.0), -0.23 (0.07, 1.0), 0.50
(0.07, 1.0).
Synthesisof[(η2-dppe)(η5-C5Me5)Fe[Ct C-1,4-(C6H4)Ct C]-
Ru(η2-dppe)2{Ct C(4-C6H4NO2)}][PF6] (5[PF6]). [(η5-C5H5)2Fe]-
[PF6] (0.024 g; 0.074 mmol) was added to a solution of 5 (0.130
g, 0.074 mmol) in dichloromethane. Stirring was maintained for
2 h at 20 °C, and the solution was concentrated in vacuo. Addition
of diethyl ether resulted in the precipitation of 5[PF6], which could
be isolated (0.120 g, 0.063 mmol, 85%) after successive washings
with toluene (5 mL), diethyl ether (5 mL), and n-pentane (5 mL).
FT-IR (ν, KBr/CH2Cl2, cm-1): 2037/2045 (s, RuCt C), 1945/1950
(s, FeCt C). 31P NMR (81 MHz, CD2Cl2, δ in ppm): 135.2 (broad
s, P, Ru(dppe)2). 1H NMR (500 MHz, CD2Cl2, δ in ppm, numbering
according to Figure 2): 31.5 (s, 2H, H2), 9.1 (broad s, 2H, H9),
8.3–5.7 (m, 56H, HAr/dppe), 4.2 (s, 4H, H6 or H7), 3.6 (s, 4H, H6 or
H7), 2.7 (s, 8H, (H2)dppe/Ru), -3.8 (s, 2H, H10), -9.4 (s, 15H, H11),
-65.6 (s, 2H, H1).
Synthesis of (η2-dppe)(η5-C5Me5)Fe[Ct C-1,4-(C6H4)Ct C]-
Ru(η2-dppe)2Cl (2). In a Schlenk tube, [RuCl(dppe)2][OTf] (7;
0.242 g, 0.224 mmol) and KPF6 (0.058 g, 0.315 mmol) were
dissolved in 20 mL of THF. The complex (η2-dppe)(η5-
C5Me5)Fe[Ct C(4-C6H4Ct CH)] (6; 0.160 g, 0.224 mmol) was then
slowly added before addition of 50 mL of methanol. The solution
was stirred for 2 days at room temperature, and the solvent was
removed in vacuo. The remaining residue was extracted with
dichloromethane and the extract concentrated in vacuo. Several
washings with diethyl ether yielded a slightly air-sensitive brown
solid. This solid was stirred for 4 h in THF in the presence of excess
of KOtBu (0.041 g, 0.362 mmol). After removal of the solvent,
extraction with toluene, concentration of the extract to dryness, and
subsequent washing with n-pentane, the desired red complex 2 was
isolated (0.250 g, 0.152 mmol; 84%). Anal. Calcd for
C98H91ClFeP6Ru: C, 71.47; H, 5.57. Found: C, 71.43; H, 5.56. MS
(ESI): m/z 1646.3652 (M•+), calcd for C98H91ClFeP6Ru 1646.3628
(M•+). FT-IR (ν, KBr, cm-1): 2056 (FeCt C and RuCt C). 31P
NMR (81 MHz, C6D6, δ in ppm): 101.7 (s, 2P, (dppe)Fe), 50.9 (s,
4P, (dppe)2Ru). 1H NMR (300 MHz, CD2Cl2, δ in ppm): 7.97 (m,
4H, HAr/dppe), 7.44–6.99 (m, 56H, HAr/dppe), 6.75 (d, 2H, 3JHH ) 8.2
3
Hz, HAr), 6.50 (d, 2H, JHH ) 8.2 Hz, HAr), 2.74 (m, 10H, CH2),
2.10 (m, 2H, CH2), 1.49 (s, 15H, C5(CH3)5). 13C{1H} NMR (75
MHz, CD2Cl2, δ in ppm): 140.3–125.3 (CAr/dppe + FeCt C +
RuCt C), 126.3 (s, CAr), 125.0 (s, CAr), 120.0 (s, FeCt C), 114.0
(s, RuCt C), 87.7 (s, C5(CH3)5), 30.9 (m, (CH2)dppe/Ru&Fe), 10.2 (s,
Synthesis of Cl(η2-dppe)2RuCt C(3-C6H4F)] (11). In a Schlenk
tube, [RuCl(dppe)2][OTf] (7; 0.160 g, 0.148 mmol) and HCt C(3-
C6H4F) (0.20 mL, 1.48 mmol) were dissolved in 10 mL of
dichloromethane. The brown-orange solution was stirred for 2 days
at room temperature, and the solvent was removed in vacuo. The
remaining brownish solid was washed twice with diethyl ether (2
× 3 mL), extracted with dichloromethane (2 × 5 mL), and filtered
through paper. Subsequently, ca. 3 mL of NEt3 was added and the
solution was stirred for a further 2 h at 25 °C. After removal of the
C5(CH3)5). CV (CH2Cl2, 0.1 M n-Bu4N+PF6-, 20 °C, 0.1 V s-1
;
E° in V vs SCE (∆Ep in V, ip,a/ip,c)): -0.24 (0.08, 1.0), 0.41 (0.08,
1.0). Mössbauer (mm s-1 vs 57Fe, 80 K): IS 0.26(1), QS 1.97(7).
Synthesisof[(η2-dppe)(η5-C5Me5)Fe[Ct C-1,4-(C6H4)Ct C]-
Ru(η2-dppe)2Cl][PF6] (2[PF6]). [(η5-C5H5)2Fe][PF6] (0.040 g,
0.121 mmol) was added to a solution of 2 (0.210 g, 0.128 mmol)
(46) Greenwood, N. N. Mössbauer Spectroscopy; Chapman and Hall:
London, 1971.

1072 Organometallics, Vol. 27, No. 6, 2008
Gauthier et al.
Table 5. Crystal Data, Data Collection, and Refinement Parameters
Table 6. Selected Bond Lengths (Å) and Angles (deg) for
for 11 · 2CH2Cl2
11 · 2CH2Cl2
formula
fw
temp (K)
cryst syst
space group
a (Å)
C60H52ClFP4Ru · 2CH2Cl2
Selected Bond Lengths
1222.27
Ru-P1A
Ru-P2A
Ru-P1B
Ru–P2B
Ru1-Cl
Ru1-C1
C1-C2
C2-C3
C3-C4
C4-C5
C5-C6
C6-C7
C7-C8
C8-C3
C7-F7
2.3947(17)
2.3689(16)
2.3852(16)
2.3562(16)
2.5370(18)
2.043(8)
100(2)
monoclinic
P21/n
14.0228(16)
17.5293(17)
90.0
100.962(7)
90.0
b( Å)
R (deg)
ꢀ (deg)
γ (deg)
V (Å3)
1.096(9)
1.524(11)
1.383(11)
1.379(13)
1.373(15)
1.342(14)
1.383(13)
1.430(12)
1.434(12)
5665.8(10)
4
1.433
Z
Dcalcd (g cm-3
)
cryst size (mm)
F(000)
0.55 × 0.25 × 0.05
2504
abs coeff (mm-1
θ range (deg)
hkl range
)
0.669
2.92–27.40
-17 to +18, -22 to +22, -30 to +29
Selected Bond and Dihedral Angles
P1A-Ru1-P2A
P1B-Ru1-P2B
P1A-Ru1-C1
Cl-Ru1-C1
P2A- Ru1-C1
Ru1-C1-C2
C1-C2-C3
82.54(6)
82.38(6)
92.48(17)
175.52(17)
90.10(16)
176.7(6)
172.7(7)
123.1(8)
121.3(10)
123.7(10)
118.9(9)
148.2
no. of reflns: total/unique
R(int)
no. of restraints/params
final R
43 235/12 775
0.0869
0/658
0.0846
0.1883
0.1362
0.2149
1.035
Rw
R indices (all data)
Rw (all data)
C2-C3-C4
goodness of fit/F2 (Sw)
C4-C5-C6
∆Fmax (e Å-3
)
)
2.186
-2.263
C6-C7-C8
∆Fmin (e Å-3
C6-C7-F7
P1A-Ru1/C3-C4a
P2B-Ru1/C3-C4a
48.6
solvent and extraction with CH2Cl2, the pale orange solution was
chromatographed on Al2O3 (neutral grade), using CH2Cl2 as eluant.
The desired complex 11 was eluted as a yellow band, affording a
pale yellow solid (0.095 g, 0.09 mmol; 61%). Anal. Calcd for
C60H52ClFP4Ru: C, 68.47; H, 4.98; F, 1.81. Found: C, 68.11; H,
4.99; F, 1.94. MS (ESI): m/z 1052.1799 (M•+), calcd for
C60H52ClFP4Ru 1052.1749 (M•+). FT-IR (ν, KBr/CH2Cl2, cm-1):
2057/2065 (RuCt C). 31P NMR (81 MHz, CDCl3, δ in ppm): 50.7
(s, dppe). 19F NMR (188 MHz, CDCl3, δ in ppm): -116.1 (s, FAr).
1H NMR (200 MHz, CDCl3, δ in ppm): 7.50 (m, 8H, HAr/dppe),
7.35–6.95 (m, 33H, HAr), 6.70 (m, 1H, HAr), 6.45 (m, 1H, HAr),
6.25 (m, 1H, HAr), 2.71 (m, 8H, (CH2)dppe). 13C{1H} NMR (125
MHz, CDCl3, δ in ppm): 163.3 (d, 1JCF ) 243 Hz, CF), 137.0 and
136.3 (2 × m, CAr/dppe), 135.1 and 134.9 (2 × s, CAr/dppe), 132.6 (d,
3JCF ) 9.7 Hz, CAr), 129.6 and 129.5 (2 × s, Cpara/dppe), 129.1 (d,
3JCF ) 9.3 Hz, CHAr), 127.9 and 127.6 (2 × s, CAr/dppe), 120.3 (q,
2JCP ) 15.4 Hz, RuCt C), 117.1 (d, 2JCF ) 20.7 Hz, CHAr), 113.4
(s, RuCt C), 110.4 (d, 2JCF ) 21.4 Hz, CHAr), 31.3 (m, (CH2)dppe).
UV–vis (λmax, nm (10-3 ꢀ, M-1cm-1); CH2Cl2): 328 (14.0). CV
(CH2Cl2, 0.1 M n-Bu4N+PF6-, 20 °C, 0.1 V s-1; E° in V vs SCE
(∆Ep in V, ip,c/ip,a)): 1.37 (0.148, 0.45), 0.51 (0.072, 1.0).
a Dihedral angle.
and 658 parameters converged at Rw(F2) ) 0.1883 (R(F) ) 0.0846)
for 8075 observed reflections with I > 2σ(I). Selected bond lengths
and angles are given in Table 6.
Determination of the ESPs of the “(Ct C)Ru(η2-dppe)2Cl”
End Group. Pure samples of 9-F and 11 (ca. 0.05 mmol) were
each dissolved in 1 mL of freshly distilled and deoxygenated CCl4.
These solutions were transferred to NMR tubes under argon, and
5 µL (0.076 mmol) of fluorobenzene was syringed in each tube as
internal reference. After homogenization of the solution, the
19F{1H} NMR spectra of both samples were recorded at 25 °C.
The values of σI and σR, as well as the σp values, were then derived
using the appropriate equations.50
Acknowledgment. M.G.H. thanks the Australian Research
Council for financial support and an ARC Australian
Professorial Fellowship. F.P. and N.G. thank the CNRS for
financial support and Region Bretagne for a Ph.D. grant. A.
Mari (LCC, Toulouse), A. Bondon (PRISM, UMR CNRS
6026), and S. Sinbandhit (CRMPO, Rennes) are acknowl-
edged for their assistance in recording the Mössbauer and
the NMR spectra.
Crystallography. Data collection of crystals of 11 · 2CH2Cl2 was
performed on an APEXII Bruker-AXS diffractometer (see Table 5
for details). The structure was solved by direct methods using the
SIR97 program47 and then refined with full-matrix least-squares
methods based on F2 (SHELX-97)48 with the aid of the WINGX
program.49 All non-hydrogen atoms were refined with anisotropic
thermal parameters. H atoms were included in their calculated
positions. A final refinement on F2 with 12 775 unique intensities
Supporting Information Available: A CIF file giving crystal
data for 11 · 2CH2Cl2. This material is available free of charge via
the Internet at http://pubs.acs.org. Final atomic positional coordi-
nates, with estimated standard deviations, bond lengths and angles,
and anisotropic thermal parameters, have been deposited at the
Cambridge Crystallographic Data Centre and allocated the deposi-
tion number CCDC 668838.
(47) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G.; Giaco-
vazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R.
J. Appl. Crystallogr. 1999, 32, 115–119.
(48) Sheldrick, G. M. SHELX97-2: Program for the refinement of crystal
structures; University of Göttingen, Göttingen, Germany, 1997.
(49) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837–838.
OM700584M
(50) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165–195.