Synthesis, molecular structure, properties, and electronic structures of [Cp* (dppe)FeC≡C-TTFMe3][PF6]n (n = 0, 1): Electronic coupling between the inorganic and organic electrophores

Article abstract of DOI:10.1021/om100763x

Sequential treatment of trimethylsilylethynyl-TTF (TTFMe₃) 3a dissolved in methanol with potassium fluoride, the chloro complex Cp*(dppe)FeCl (8; Cp* = η⁵-C₅(CH ₃)₅, dppe = 1,2-bis(diphenylphosphino)ethane) in the presence of ammonium hexafluorophosphate, and finally with KOBu^t provided Cp*(dppe)FeC≡C-TTFMe₃ (1), which was isolated in 69% yield as an air- and moisture-sensitive orange powder. The X-ray crystal structure, IR, cyclic voltammetry (CV), UV-vis, Moessbauer, and DFT data obtained for 1 show that only weak interactions take place between the iron center and the TTFMe₃ core and that the latter acts as a rather poor electron donor vis-a-vis the organometallic group. When reacted with 1.0 equiv of [(C₅H₅)₂Fe][PF₆] in tetrahydrofuran (THF) at -60 °C, 1 gives the thermally stable radical cation 1[PF₆], which was isolated in a pure form as a purple powder (77%). The Fe^III derivative was characterized by the same spectroscopic methods as 1 complemented by HRMS and elemental analyses also obtained for neutral 1. All experimental and theoretical data obtained for 1[PF₆] indicate that after one-electron oxidation the relaxation of the molecular structure is characterized by an increase of the Fe-C carbon bond order, the localization of the positive charge on the iron nuclei, and the delocalization of the spin density on the whole molecule. In the radical cation, the interaction between the iron center and the TTFMe₃ fragment is much stronger than in the neutral species, and a good electronic communication characterized by a rather large electronic coupling for a class II MV system takes place (H_ab = 320 cm^-1 determined from the analysis of the NIR spectrum). Furthermore, the dicationic species 1[PF₆] ₂ was in situ generated and characterized by IR, NIR, and EPR spectroscopies.

Full text of DOI:10.1021/om100763x

4628 Organometallics 2010, 29, 4628–4638
DOI: 10.1021/om100763x
Synthesis, Molecular Structure, Properties, and Electronic Structures
of [Cp*(dppe)FeCtC-TTFMe3][PF6]n (n=0, 1): Electronic Coupling
between the Inorganic and Organic Electrophores
,†
Akira Miyazaki,* Yoshiyuki Ogyu, Frederic Justaud, Lahcene Ouahab,*
†
‡
,‡
ꢀ ꢀ
ꢁ
Thomas Cauchy,‡,§ Jean-Franc-ois Halet,*,‡ and Claude Lapinte*,‡
‡
ꢀ
UMR CNRS 6226 Sciences Chimiques de Rennes, Universite de Rennes 1, Campus de Beaulieu,
§
ꢀ
F-35042 Rennes, France, Institut des Sciences et Technologies Moleculaires d’Angers (Moltech Anjou),
UMR 6200 CNRS-Universiteꢀ d’Angers, 2 Boulevard Lavoisier, F-49045 Angers, France, and
†Department of Environmental Applied Chemistry, University of Toyama, Toyama, Japan
Received August 6, 2010
Sequential treatment of trimethylsilylethynyl-TTF (TTFMe3) 3a dissolved in methanol with
potassium fluoride, the chloro complex Cp*(dppe)FeCl (8; Cp*=η5-C5(CH3)5, dppe=1,2-bis-
(diphenylphosphino)ethane) in the presence of ammonium hexafluorophosphate, and finally with
KOBut provided Cp*(dppe)FeCtC-TTFMe3 (1), which was isolated in 69% yield as an air- and
moisture-sensitive orange powder. The X-ray crystal structure, IR, cyclic voltammetry (CV),
€
UV-vis, Mossbauer, and DFT data obtained for 1 show that only weak interactions take place
between the iron center and the TTFMe3 core and that the latter acts as a rather poor electron donor
ꢁ
vis-a-vis the organometallic group. When reacted with 1.0 equiv of [(C5H5)2Fe][PF6] in tetrahydro-
furan (THF) at -60 °C, 1 gives the thermally stable radical cation 1[PF6], which was isolated in a
pure form as a purple powder (77%). The FeIII derivative was characterized by the same spectros-
copic methods as 1 complemented by HRMS and elemental analyses also obtained for neutral 1.
All experimental and theoretical data obtained for 1[PF6] indicate that after one-electron oxidation
the relaxation of the molecular structure is characterized by an increase of the Fe-C carbon bond
order, the localization of the positive charge on the iron nuclei, and the delocalization of the spin
density on the whole molecule. In the radical cation, the interaction between the iron center and the
TTFMe3 fragment is much stronger than in the neutral species, and a good electronic communication
characterized by a rather large electronic coupling for a class II MV system takes place (Hab =
320 cm-1 determined from the analysis of the NIR spectrum). Furthermore, the dicationic species
1[PF6]2 was in situ generated and characterized by IR, NIR, and EPR spectroscopies.
Introduction
extensively studied.1,4 We have ourselves shown that species
containing redox Cp*(dppe)Fe fragments (Cp* = C5Me5,
dppe = 1,2-bis(diphenylphosphino)ethane) linked to π-
conjugated polyynediyl ligands are ideally suited for studies
of electronic and magnetic coupling between the redox-active
centers. These assemblies proved usually to be stable (and
isolable) in different redox states. However, previous stud-
ies have clearly shown that the stability of the oxidized
species is limited by lengthening of the polyynediyl linkers.5
Introduction of aromatic rings such as benzene or thiophene
in the polyynediyl spacer constitutes an attractive alternative
Organometallics comprising two redox active centers con-
nected by a carbon bridge provide an ideal template upon which
intricate mechanistic details of the factors controlling electron
transfer and electron delocalization can be extracted.1-3 Conse-
quently, over the last years, one thrust has involved extremes in
oxidation states of this type of compound with different metal
end-groups, and the consequences for electronic, magnetic, and
geometric structure as well as electron transfer have been
*To whom correspondence should be addressed. E-mail: lapinte@
univ-rennes1.fr; miyazaki@eng.u-toyama.ac.jp; ouahab@univ-rennes1.fr;
jean-francois.halet@univ-rennes1.fr.
(4) (a) Ren, T. Chem. Rev. 2008, 108, 4185–4207. Ren, T. Organo-
metallics 2005, 24, 4854–4870. (b) Schwab, P. F. H.; Levin, M. D.; Michl, J.
Chem. Rev. 2005, 105, 1197–1279. (c) Qi, H.; Noll, B.; Snider, G. L.; Lu, Y.;
Lent, S. S.; Fehlner, T. P. J. Am. Chem. Soc. 2005, 127, 15218–15227.
(d) Blum, A. S.; Ren, T.; Parish, D. A.; Trammell, S. A.; Moore, M. H.;
Kushmerick, J. G.; Xu, G. L.; Deschamps, J. R.; polack, S. K.; Shashidar, R.
J. Am. Chem. Soc. 2005, 127, 10010–10011. (e) Xu, G. L.; Crutchley, R. J.;
DeRosa, M. C.; Pan, Q.-J.; Zhang, H.-X.; Wang, X.; Ren, T. J. Am. Chem.
Soc. 2005, 127, 13354–13365. (f) Akita, M.; Koike, T. Dalton Trans. 2008,
3523–3530.
(1) Crutchley, R. J. Adv. Inorg. Chem. 1994, 41, 273–325.
(2) (a) Ward, M. D. Chem. Soc. Rev. 1995, 24, 121–134. (b) Ward,
M. D. Chem. Ind. 1996, 568–573. (c) Paul, F.; Lapinte, C., Magnetic
communication in binuclear organometallic complexes mediated by carbon-
rich bridges. In Unusual Structures and Physical Properties in Organo-
metallic Chemistry; Gielen, M.; Willem, R.; Wrackmeyer, B., Eds.; John
Wiley & Sons: London, 2002; pp 220-291. (d) Demandis, K. D.; Hartshorn,
C. M.; Meyer, T. J. Chem. Rev. 2001, 101, 2655–2685. (e) Brunschwig, B. S.;
Creutz, C.; Sutin, N. Chem. Soc. Rev. 2002, 31, 168–184.
(3) Paul, F.; Lapinte, C. Coord. Chem. Rev. 1998, 178-180, 427–505.
(5) Coat, F.; Lapinte, C. Organometallics 1996, 15, 477–480.
r
pubs.acs.org/Organometallics
Published on Web 09/23/2010
2010 American Chemical Society

Article
Organometallics, Vol. 29, No. 20, 2010 4629
to circumvent this instability and eventually to tune their
physical properties.6-9 Various nonlinear rigid new geome-
tries can then be envisioned for the carbon-rich spacers
incorporating such aromatic units, leading to structural
variation, which can deeply modify the electronic properties
of the molecules.7,8,10
Chart 1
On the other hand, since the first report on TTF (tetra-
thiafulvalene = 2,20-bis(1,3-dithiolylidene)) in 1970,11 TTF
and its derivatives remain the most studied redox-active organic
systems.12-14 They are now recognized as the gateway to
organic metals, superconductors, and semiconductors.14,15
Because of the wide range of physical properties of TTF-
containing compounds as potential superconductors, charge-
transfer complexes, ferromagnets, nonlinear optical materials,
sensors, electroactive Langmuir-Blodgett films, and organic
field effect transistors, they were the subject of a huge number
of reports.12 More recently, the high electron-donating ability
of these redox-active moieties led to important studies focusing
on their functionalization toward their use in coordination
chemistry.15,16
materials. To establish stronger interactions between these
functional groups, the introduction of a covalent-bonded
linker is expected to be preferable to intermolecular van der
Waals contacts used in the conventional π-d interaction-
based systems. On the basis of this approach, we report here
the synthesis and characterization of the new hybrid complex
1 (Chart 1).
In this compound, two electroactive units, namely, the
TTFMe3 and FeCp*(dppe) groups, are linked by an ethynyl
bridge to favor π-d interactions between them. In contrast
to the complexes 4 and 5, for which the first oxidation was
found to be TTF or TTFMe3 centered,17,18 it can be antici-
pated that the first oxidation of 1 should be iron centered
since the redox potential of the compounds 2, 3a, and 6
are -0.14, 0.35, and -0.10 V vs SCE, respectively.19,20
The isolation of the associated radical cation 1[PF6] is also
reported. This novel species was investigated by various
means in order to assess the extent of the electronic interac-
tion between the metal center and the TTF core through the
acetylide bridge. Attempts to isolate the associated dication
are also briefly described with some spectroscopic character-
izations.
Multifunctional molecular systems based on the assem-
blies incorporating photo- and/or redox-active groups have
been the subject of considerable recent research as suitable
candidates for switchable optical, conductive, and magnetic
(6) (a) Le Narvor, N.; Lapinte, C. Organometallics 1995, 14, 634–639.
(b) Le Stang, S.; Paul, F.; Lapinte, C. Organometallics 2000, 19, 1035–1043.
(7) de Montigny, F.; Argouarch, G.; Costuas, K.; Halet, J.-F.;
Roisnel, T.; Toupet, L.; Lapinte, C. Organometallics 2005, 24, 4558–
4572.
(8) Lohan, M.; Justaud, F.; Roisnel, T.; Ecorchard, P.; Lang, H.;
Lapinte, C. Organometallics 2010, DOI: 10.1021/om100003z.
(9) Ghazala, S. I.; Paul, F.; Toupet, L.; Roisnel, T.; Hapiot, P.;
Lapinte, C. J. Am. Chem. Soc. 2006, 128, 2463–2476.
(10) Tanaka, Y.; Shaw-Taberlet, J. A.; Justaud, F.; Cador, O.;
Roisnel, T.; Akita, M.; Hamon, J.-R.; Lapinte, C. Organometallics
2009, 28, 4656–4669.
(11) Wudl, F.; Smith, G. M.; Hufnagel, E. J. J. Chem. Soc., Chem.
Commun. 1970, 490–491.
(12) Chem. Rev. 2004, 104, 4887-4890, a thematic issue on “molecular
conductors”, P. Batail, Ed.
(13) (a) Ishiguro, T.; Yamaji, R.; Saito, G. Organic Superconductors;
Springer-Verlag: Berlin, 1988. Ouahab, L.; Yagubskii, E. Organic Conduc-
tors, Superconductors and Magnets: From Synthesis to Molecular Elec-
ꢀ
tronics; Kluwer:Dordrecht, 2003. (b) Foumigue, M.; Ouahab, L. Topics in
Organometallic Chemistry Dedicated to Conducting and Magnetic Orga-
nometallic Materials; Springer: Berlin, 2009. Ouahab, L.; Enoki, T. Eur. J.
Inorg. Chem. 2004, 933. (c) Saito, G.; Y. Yoshida, Y. Bull. Chem. Soc.
Results and Discussion
ꢀ
Jpn. 2007, 80, 1. (d) Lorcy, D.; Bellec, N.; Fourmigue, M.; Avarvari, N.
Coord. Chem. Rev. 2009, 253, 1398.
1. Synthesis and Characterization of 1. As illustrated in
Scheme 1, 1 was synthesized from Cp*(dppe)Fe-Cl21 (8) and
TTFMe3 containing three methyl groups (3a). The
TTFMe3 precursor 3a was obtained from treatment of
lithiated TTFMe3 with perfluorohexyliodide, followed by
the Sonogashira coupling reaction with trimethylsilylacety-
lene. Treatment of 3a with potassium fluoride in methanol
provided the ethynyl-TTFMe3 3b, which was in situ reacted
with the chloro complex 8 in the presence of ammonium
hexafluorophosphate. TLC monitoring showed that the
organic reagent was consumed after 16 h, and following
recipes established for the complexation of various alkynes
(14) Yamada, J.; Sugimoto, T., TTF Chemistry-Fundamentals and
Applications of Tetratiafulvalene; Kodansha and Springer: Tokyo, 2004.
(15) (a) Iwahori, F.; Golhen, S.; Ouahab, L.; Carlier, R.; Sutter, J.-P.
Inorg. Chem. 2001, 40, 6541. (b) Setifi, F.; Ouahab, L.; Golhen, S.; Yoshida,
Y.; Saito, G. Inorg. Chem. 2003, 42, 1791. (c) Pellon, P.; Gachot, G.; Le Bris,
J.; Marchin, S.; Carlier, R.; Lorcy, D. Inorg. Chem. 2003, 42, 2056. (d) Devic,
ꢀ
T.; Batail, P.; Fourmigue, M.; Avarvari, N. Inorg. Chem. 2004, 43, 3136.
ꢀ
(e) Avarvari, N.; Fourmigue, M. Chem. Commun. 2004, 1300. (f) Jia, C.; Liu,
S.-X.; Ambrus, C.; Neels, A.; Labat, G.; Decurtins, S. Inorg. Chem. 2006, 45,
3152. (g) Ichikawa, S.; Kimura, S.; Mori, H.; Yoshida, G.; Tajima, H. Inorg.
Chem. 2006, 7575. (h) Wang, L.; Zhang, B.; Zhang, J. Inorg. Chem. 2006,
45, 6860. (i) Massue, J.; Bellec, N.; Chopin, S.; Levillain, E.; Roisnel, T.;
ꢀ
Clerac, R.; Lorcy, D. Inorg. Chem. 2005, 44, 8740. (j) Benbellat, N.; Le Gal,
Y.; Golhen, S.; Gouasmia, A.; Ouahab, L.; Fabre, J.-M. Eur. J. Org. Chem.
2006, 4237. (k) Zhu, Q.-Y.; Liu, Y.; Lu, W.; Zhang, Y.; Bian, G.-Q.; Niu,
G.-Y.; J. Dai, J. Inorg. Chem. 2007, 46, 1006. (l) Gavrilenko, K. S.; Le Gal,
Y.; Cador, O.; Golhen, S.; Ouahab, L. Chem. Commun. 2007, 280.
(m) Pointillart, F.; Gal, L.; Y.; Golhen, S.; Cador, O.; Ouahab, L. Inorg.
Chem. 2008, 47, 9730. (n) Umezono, Y.; Fujita, W.; Awaga, K. J. Am. Chem.
(17) Iyoda, M.; Takahiro, T.; Otani, N.; Ugawa, K.; Yoshida, M.; H.,
M.; Kuwatani, Y. Chem. Lett. 2001, 1310–1311.
ꢁ
(18) Vacher, A.; Barriere, F.; Roisnel, T.; Lorcy, D. Chem. Commun.
ꢀ
Soc. 2006, 128, 1084. (o) Herve, K.; Liu, S.-X.; Cador, O.; Golhen, S.; Le
2009, 7200–7202.
(19) Roue, S.; Lapinte, C. J. Organomet. Chem. 2005, 690, 594–604.
(20) Le Narvor, N.; Toupet, L.; Lapinte, C. J. Am. Chem. Soc. 1995,
117, 7129–7138.
(21) Roger, C.; Hamon, P.; Toupet, L.; Rabaa, H.; Saillard, J.-Y.;
ꢀ
Gal, Y.; Bousseksou, A.; Stoeckli-Evans, H.; Decurtins, S.; Ouahab, L. Eur.
J. Inorg. Chem. 2006, 3498. (p) Cosquer, G.; Pointillart, F.; Gal, L.; Y.;
Golhen, S.; Cador, O.; Ouahab, L. Dalton Trans. 2009, 3495.
(16) Fourie, E.; Swarts, J.; Lorcy, D.; Bellec, N. Inorg. Chem. 2010,
49, 952–959.
^
Hamon, J.-R.; Lapinte, C. Organometallics 1991, 10, 1045–1054.

4630 Organometallics, Vol. 29, No. 20, 2010
Miyazaki et al.
Figure 1. ORTEP drawing of 1A. Thermal ellipsoids are at the 30% probability level. Hydrogen atoms are omitted for clarity.
Scheme 1a
a Key reagents: (i) KF, NH4PF6, CH3OH, 25 °C, 16 h, then KOBut; (ii) [(C5H5)2Fe][PF6], THF, -60 °C, 2 h.
at the iron center, 1 equiv of KOBut was added to favor
the deprotonation of the vinylidene intermediate.7,22 After
workup 1 was isolated in 69% yield as an air- and moisture-
sensitive orange powder. A FAB mass spectrum of 1
showed the expected parent ion, and the microanalysis was
satisfactory.
of bond lengths and bond angles is given in Table 1. As it is
invariably observed for all the members of the Cp*(dppe)-
Fe-CtC family of complexes, the metal adopts a pseudo-
octahedral geometry with bond lengths and angles in the
previously established ranges.3,22-26
The structural parameters within the TTFMe3 units are in
the range of what is usually observed for neutral molecules
with, however, a small difference between the CdC double
bonds.27 Indeed, the double bond of the ring connected to
Well-resolved 1H, 13C, and 31P NMR spectra were
obtained and confirmed the structure. The most striking
1
features of the H NMR spectrum include the Cp* methyl
˚
singlet at 1.46 ppm, the methyl resonances at 1.74, 1.37, and
1.36 ppm, two dppe methylene multiplets at 2.63 and 1.84 ppm,
and the aromatic peaks integrating to 20 protons. The 31P
NMR spectrum displayed one singlet at 100.77 ppm, as usual
for these acetylide iron derivatives.23 Finally, the 13C NMR
spectrum showed all expected resonances including the triplet
(δ 151.32, 2JCP = 38.4 Hz) and singlet (δ 116.6) signatures for
the acetylide linkage and the three resonances characteristic of
the methyl substituents of the TTFMe3 moiety (δ 14.62, 12.99,
and 12.97).
2. Molecular Structure of 1. Single crystals suitable for
X-ray crystallography were grown by slow diffusion of
n-pentane into a toluene solution containing 1 at 20 °C.
Complex 1 crystallizes in the monoclinic space group P21/c.
The X-ray data conditions, crystal data, data collection, and
structure refinement parameters are summarized in the
Experimental Section (Table 8).
the iron-alkynyl group is longer (1.340(3) A) than the outer
˚
one, namely, the C44-C46 bond distance (1.332(4) A). The
TTFMe3 unit is not planar; it is bent around the S1S2
segment with a dihedral angle of ca. 28.0° between the planes
defined by (S1S2S3S4) and (S1S2C39C40). The dihedral
angle between the planes defined by P1-Fe-P2 and
(S1S2S3S4) is equal to ca. 69.7°.
3. Cyclic Voltammetry of 1 and Synthesis of the Oxidized
Species 1[PF6]. The initial scan in the CV of 1 from -1
to þ1.0 V shows three chemically reversible oxidation waves
separated by 0.49 and 0.46 V, respectively (Figure 2). The
analysis of the peak currents shows that the anodic and
cathodic currents are strictly identical for the three waves,
(24) Weyland, T.; Costuas, K.; Toupet, L.; Halet, J.-F.; Lapinte, C.
Organometallics 2000, 19, 4228–4239.
(25) Coat, F.; Guillevic, M.-A.; Toupet, L.; Paul, F.; Lapinte, C.
The asymmetric unit is composed of two distinct but very
similar units A and B. Molecules A and B contain the Fe1 and
Fe2 atoms, respectively. We describe here only the unit A.
The molecular structure of A is shown in Figure 1, and a set
Organometallics 1997, 16, 5988–5998.
(26) Hamon, P.; Toupet, L.; Hamon, J.-R.; Lapinte, C. Organo-
metallics 1996, 15, 10–12.
(27) (a) Mas-Torrent, M.; Durkut, M.; Hadley, P.; Ribas, X.; Rovira,
C., J. Am. Chem. Soc. 2004, 126. (b) Mas-Torrent, M.; Hadley, P.;
ꢀ
Bromley, S. T.; Ribas, X.; Tarres, J.; Mas, M.; Molins, E.; Veciana, J.;
(22) Weyland, T.; Lapinte, C.; Frapper, G.; Calhorda, M. J.; Halet,
J.-F.; Toupet, L. Organometallics 1997, 16, 2024–2031.
(23) Denis, R.; Toupet, L.; Paul, F.; Lapinte, C. Organometallics
2000, 19, 4240–4251.
Riovira, C. J. Am. Chem. Soc. 2004, 126, 8546. (c) Guionneau, P.; Kepert,
C. J.; Bravic, G.; Chasseau, D.; Truter, M. R.; Kurmoo, M.; Day, P. Synth.
Met. 1997, 1973–1974. (d) Flandrois, S.; Chasseau, D. Acta Crystallogr.
1977, B33, 2744–2750.

Article
Organometallics, Vol. 29, No. 20, 2010 4631
˚
Table 1. Experimental Selected Bond Distances (A) and Angles (deg) for 1A and Calculated Bond Distances for
1nþ, 3bnþ, and 6nþ a
1A
1
1þ
12þ(S)
12þ(T)
12þ(BS)
3b
3bþ
6
6þ
Fe-P1
2.1751(6)
2.1731(6)
1.898(2)
1.220(3)
1.418(3)
1.346(3)
1.340(3)
1.332(4)
1.744
2.213
2.223
1.873
1.247
1.402
1.366
1.357
2.252
2.253
1.827
1.257
1.382
1.388
1.373
2.279
2.289
1.793
1.269
1.365
1.408
1.393
2.302
2.324
1.854
1.250
1.394
1.382
1.392
2.291
2.295
1.819
1.261
1.376
1.397
1.393
2.207
2.215
1.890
1.236
2.283
2.299
1.865
1.237
Fe-P2
Fe-C37
C37-C38
1.218
1.408
1.363
1.356
1.216
1.405
1.374
1.394
C38-C39
C39-C40
C42-C43
C44-C46
Fe-Cp*b
1.767
1.783
1.783
1.769
1.802
1.752
1.807
1.755
1.803
1.753
1.763
1.800
C-S (av)
1.783
1.753
P1-Fe-P2
Fe-C37-C38
C37-C38-C39
C38-C39-C40
S1-C42-C43-S3
S2-C42-C43-S4
85.42(2)
176.4(2)
175.4(2)
127.9(2)
2.73
4.87
a The atom numbering of the computed compounds is the same as that in 1A. b Centroid of the η5-C5Me5 ligand.
Table 2. Electrochemical Data for Selected Complexesa
very similar to that of the reference complex 6, suggesting
again that the TTFMe3 moiety acts as a very weak electron-
releasing substituent for the Cp*(dppe)Fe moiety.
compd
E01 (ΔEp)
E02 (ΔEp)
ΔE03 (ΔEp)
ref
1
2
3a
4
5
-0.11 (0.07)
-0.14 (0.09)
0.41 (0.06)
0.35
0.07
-0.10 (0.07)
0.38 (0.07)
0.84 (0.08)
this work
19
this work
17
18
20
The potential of the second redox event is 0.03 V more
favored than the first oxidation of 3a, indicating that the
oxidized [Cp*(dppe)FeIII-CtC-]þ• moiety does not act as
an electron-withdrawing group with respect to the TTFMe3
part, but rather as a weakly electron-donating fragment.
The hybrid systems TTF-ferrocene 4 and Me3TTF-
alkynyl-ruthenium 5 also showed three reversible redox
waves in the CVs. It is worth noting that the first oxidation
potential of 4 is slightly less positive than that of 3a,
indicating that ferrocene can stabilize the radical cation
centered on the TTF moiety (provided that TTF is less
electron rich than 3a).17 In the case of compound 5, the
first oxidation potential was found at a significantly less
anodic potential than that of the first oxidation potential
of 3a. Nevertheless, the first oxidation of 5 was assigned to
the TTFMe3 core, strongly affected by the electron-releasing
ruthenium moiety.18 Interestingly, while the second oxidation
can be assigned to the first oxidation of the TTFMe3 group in
1, it was attributed to the ferrocene in the case of 4 and to the
second oxidation of the TTFMe3 core in 5 (by the authors of
ref 18). As a result, the three redox families [1]nþ, [4]nþ, and
[5]nþ somewhat differ.
On the basis of the full reversibility of the CV waves, the
mono- and dioxidized complexes 1[PF6] and 1[PF6]2 were
thought as being accessible synthetic targets. According to a well-
established procedure, complex 1 was reacted with 1.0 equiv of
[(C5H5)2Fe][PF6] in tetrahydrofuran (THF) at -60 °C. Upon
gradual warming to room temperature, the thermally stable
radical cation 1[PF6] was formed. Partial precipitation and
crystallization provide pure samples that were isolated in good
yield (77%) as a dark purple powder. The Fe(III) derivative was
characterized by usual spectroscopic methods complemented by
HRMS and elemental analysis. As expected, this dark complex
presents a CV identical to that of 1. Similar efforts to prepare the
dication were unsuccessful. However, treatment of 1with 2 equiv
of [(C5H5)2Fe][PF6] in tetrahydrofuran at -60 °C provided a
greenish-brown liquor, from which addition of cold pentane
allowed the isolation of 1[PF6]2 in almost pure form. However,
the large kinetic reactivity of this salt does not allow further
purification. For this reason, the charactization of 1[PF6]2 was
achieved only using FTIR and EPR spectroscopies.
0.92 (0.09)
0.71
0.52
0.85
1.07
6
a In CH2Cl2, 0.1 M [Bu4nN]PF6, scan rate 0.1 V/s, Pt electrodes, V vs
SCE (Fc/Fcþ = 0.46 V vs SCE used as internal reference for potential
measurements).
Figure 2. CV of 1 (V vs SCE, 10-3 M solution in dichloro-
methane at 298 K, 0.1 M [Bu4 N]PF6, scan rate = 0.100 V s-1).
n
so that the electron-deficient species 1þ, 12þ, and 13þ are
apparently stable at the platinum electrode.
Comparison of the first oxidation potential of 1 with those
found for 2 (Table 2) and other related compounds of
the [Cp*(dppe)Fe-CtC-] series23 strongly suggests that
the first redox process should be metal centered ({[FeII]-
CtC-TTFMe3}/{[FeIII þ•
other oxidations are expected to mainly concern the
TTFMe3 moiety ({[FeIII þ•-CtC-TTFMe3}/{[FeIII þ•
-Ct
] -CtC-TTFMe3}), while the two
]
]
C-TTFMe3þ•} and {[FeIII þ•-CtC-TTFMe3þ•}/{[FeIII þ•
] ] -
CtC-TTFMe32þ}, respectively). In particular, compari-
son of the first redox potentials (E0 ) given in Table 2
1
indicates that the TTFMe3 moiety does not behave as a
strong electron donor when connected to the alkynyl-iron
site. Apparently, it is only slightly less electron-donating
than the thiophene unit. Moreover, the potential of 1 is

4632 Organometallics, Vol. 29, No. 20, 2010
Table 3. IR νCtC Bond Stretching for 1[PF6]n and Related Complexesa
Miyazaki et al.
compd
n = 0
n = 1
n = 2
ref
1
2022b (2035)
1917b (1965)
2021 (w), 1885 (s)
1890) (S); (1970) (T); (1870) (BS)
this work
2
2044
2140
2140 (2124)
NA
2029
1977, 1965
19
this work
this work
17
18
20
3a
3b
4
5
6
(2139)
NA
NA
1910 (1936)
(1893)
a Nujol, cm-1 unless otherwise specified. Computed values are given in parentheses b KBr.
Scheme 2. Possible Electronic Structures for 1 and 1þ
˚
(1.879(5) A for R = 1,4-C6H4CN). These results suggest
that 1b is only a minor contributor (at best) to the overall
structure of 1.
Upon one-electron oxidation, the IR absorption correspond-
ing to the νCtC stretch is shifted to lower wave numbers. This
shift follows the same trend as previous observations.23
However, its amplitude is exceptionally large in the present
case (Δν = 105 cm-1). Indeed, up to now the largest lowering
oftheνCtC frequency induced by a single electron transfer was
observed for the complex having a strongly electron-releasing
substituent (R = 1,4-C6H4NMe2),23 and the effect was not as
large as in the present case (Δν = 94 cm-1). The weight of the
VB structure 1þb (Scheme 2) is probably important in the
descriptionofthe electronicstructure of1þ. Asa consequence,
the positive charge is apparently mainly iron centered, while
the spin density should be shared between the iron center and
the organic moiety. Our results suggest that when coordinated
to the Cp*(dppe)Fe moiety the TTFMe3 group presents an
original behavior. Indeed, up to now we have always observed
that in the Cp*(dppe)FeIII series the spin density is strongly
metal centered regardless of the substituent on the alkynyl
ligand. In contrast, in the closely related Cp*(dppe)RuIII series,
the spin density is largely distributed on the whole molecule.32
The IR spectrum of the deoxidized species 1[PF6]2 displays
two bands corresponding to the νCtC stretch at 2020 and
1885 cm-1 with a constant ratio of relative intensities of 1:4.
Following the DFT data discussed below these two bands
might be assigned to the triplet (S=1) and singlet or
“broken symmetry” (S=0) spin isomers.33
4. IR Spectroscopy. A large number of mononuclear FeII
and FeIII acetylides of Cp*(dppe)Fe-CtC-R were subject to
FTIR analyses. The stretching frequency of the triple bond
(νCtC) was found to be dependent on the electronic nature of
the appended substituent, and its evolution could be quantita-
tively rationalized using valence bond (VB) formalism.23,28-30
It was shown that respectively electron-withdrawing substituents
in FeII acetylides or electron-releasing substituents in the FeIII
congeners favor the cumulenic character of the alkynyl bridge.23
The neutral complex was characterized by a strong ab-
sorption at 2022 cm-1, which corresponds to the CtC
stretching mode. In comparison with other members of this
Cp*(dppe)Fe-CtC-R family of FeII complexes, the fre-
quency is relatively low. This is confirmed by DFT calcula-
tions (see Table 3). For purpose of comparison, the lowest
frequencies ever observed for this kind of complex were
found for R = 9,10-C14H8-CN (νCtC=1986 cm-1), and
a frequency as low as 2025 cm-1 was found for R = 1,
4-C6H4CN.23,31 These data suggest that CtC bond order of
the alkynyl linker is probably weakened by the presence of
the TTFMe3 core. However, the VB structure 1b (Scheme 2)
should not play a significant role in the description of the
electronic structure of this complex because the CV data
indicate that the TTFMe3 group is weakly electron releasing.
Moreover, it is also noteworthy that the Fe-C bond distance
57
57
€
€
5. Fe Mossbauer Spectroscopy. Fe Mossbauer spec-
troscopy is a very sensitive probe for identifying the iron
oxidation state and the nature of the Fe-C bond.34 More-
over, in a homogeneous series of iron compounds like Cp*-
(dppe)Fe-CtC-, it has been shown that the isomeric shift is
related to the electron density at the metal center.35 Addi-
€
tionally, Mossbauer spectroscopy constitutes a sensitive and
useful means to probe the selectivity of the redox reactions.
57
€
The Fe Mossbauer spectra of microcrystalline samples
of the complexes 1and 1[PF6] were run at 80 K and least-squares
fitted with Lorentzian line shapes.36 The isomer shift (IS) and the
quadrupole splitting (QS) parameters are given in Table 4. The
spectrum of the neutral complex 1displays a unique doublet, and
˚
in 1 is significantly longer (1.898(2) A) than in the iron
acetylides bearing an electron-withdrawing aryl group
(28) 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.
(32) Paul, F.; Ellis, B. E.; Bruce, M. I.; Toupet, L.; Roisnel, T.;
Costuas, K.; Halet, J.-F.; Lapinte, C. Organometallics 2006, 25, 649–665.
(33) Guillaume, V.; Mahias, V.; Mari, V.; Lapinte, C. Organometallics
2000, 19, 1422–1426. Lapinte, C. J. Organomet. Chem. 2008, 693, 793–801.
(34) Guillaume, V.; Thominot, P.; Coat, F.; Mari, A.; Lapinte, C.
J. Organomet. Chem. 1998, 565, 75–80.
(29) Paul, F.; Mevellec, J.-Y.; Lapinte, C. Dalton Trans. 2002, 1783–
1790.
ꢀ
(30) Paul, F.; Toupet, L.; Thepot, J.-Y.; Costuas, K.; Halet, J.-F.;
Lapinte, C. Organometallics 2005, 24, 5464–5478.
(31) de Montigny, F.; Argouarch, G.; Roisnel, T.; Toupet, L.;
Lapinte, C.; Lam, S. C.-F.; Tao, C.-H.; Yam, V. W.-W. Organometallics
2008, 27, 1912–1923.
(35) Argouarch, G.; Thominot, P.; Paul, F.; Toupet, L.; Lapinte, C.
C. R. Chim. 2003, 6, 209–222.
(36) Varret, F.; Mariot, J.-P.; Hamon, J.-R.; Astruc, D. Hyperfine
Interact. 1988, 39, 67–81.

Article
Organometallics, Vol. 29, No. 20, 2010 4633
Table 4. 57Fe Mossbauer Parameters for 1nþ (n = 0, 1) and
Related Compounds 2 and 7
Table 5. EPR Parameters for 1[PF6] and Related Fe(III)
Compounds at 66 K
€
g2 (a2)a
g3 (a3)a
giso
Δgc
ref
b
ISa (QS) Γb, mm s-1
compd
g1
1[PF6] 2.450 2.036 (38) 2.002 (5)
1[PF6]2 2.312 2.063 2.063
2.163 0.448 this work
this work
compd
n = 0
n = 1
1
0.240 (1.979) 0.136
0.259 (1.969) 0.116
0.153 (1.114) 0.220
0.227 (0.974) 0.146
0.141 (1.123) 0.171
D = -88 G, E = 8 G, Δms = ( 2, g = 4.267
2[PF6] 2.366 2.036 (14) 1.986 (16) 2.129 0.380 39
2c
7d
6[PF6] 2.457 2.034
1.977
2.156 0.480 20
a The velocity is referenced to iron metal. b Γ is the half-width at half-
a In Gauss. b giso = (g1 þ g2 þ g3)/3. c Δg = g3 - g1.
height. c From ref 19. d From ref 35.
the components of the g-tensors, characteristic of d5 low-spin
Fe(III) in a pseudooctahedral environment. The g values
extracted from the spectra are collected in Table 5. The
spectrum of 1[PF6] is characterized by a large isotropic
tensor (giso) and a large tensor of anisotropy (Δg).
the parameters are very close to those obtained for 2, chosen as a
representative member of the Cp*(dppe)FeII-CtC- series.3,34,35
In the case of complex 1[PF6], the spectrum displays a
single doublet, which probes the thermal stability of this
radical. However, the QS parameter is exceptionally large for
a FeIII-alkynyl derivative, evidencing that the electronic
structure of the new complex is unusual. Indeed, for the
17-electron FeIII radical cation of this series, it has been
observed that the quadrupole splitting is systematically
Optimal resolution of the spectral features allows the
observation of the hyperfine coupling with two phosphorus
nuclei. In contrast with previous observations of related FeIII
radicals,19,28,40 the coupling constants are quite different for
the g2 and g3 components. Such an original anisotropy possibly
originates from the partial delocalization of the unpaired
electron on the TTFMe3 moiety.
The dicationic complex 1[PF6]2 was obtained by treatment
of complex 1 with 2.2 equiv of [(C5H5)2Fe][PF6] in THF
at -80 °C. The resulting brownish solution was transferred
at this temperature into an EPR tube before being cooled to
66 K. A well-resolved EPR spectrum displaying three com-
ponents for the Δms = (1 transition was obtained, and the
EPR parameters were extracted by simulation (Table 5).
Additionally, fine structures were clearly visible in the g = 2
region, reflecting the zero-field splitting of the ms components
of the triplet state. Furthermore, the Δms = (2 transition
characteristic of the triplet state was clearly observed at g =
4.267. These interesting results clearly demonstrate that com-
plex 1[PF6]2 carries two unpaired electrons and magnetic
exchange interactions between them take place.
7. UV-Vis Absorption Spectroscopy. The UV-vis spectra
of the complexes 1[PF6]n (n = 0, 1) were recorded in the range
260-800 nm. The spectrum of the neutral complex resembles
the spectrum obtained for the previously reported function-
alized organoiron(II) compounds (Table 6).41 Besides the
high-energy transition that can apparently be attributed to
π-π* ligand-centered transitions, two less intense transi-
tions were also observed. These absorptions located in the
visible range are at the origin of the orange color of 1 and can
be assigned to MLCT transitions.23 However, in comparison
with 2 and other related complexes bearing a functional aryl
group, these transitions are red-shifted by ca. 60 nm.19,23,42
This feature suggests that the low-energy transition results
from some admixture of dπ(Fe) f π*(CtC-TTFMe3) and is
in line with the atomic character of the HOMOs and LUMOs
(see Section 9).
below 1.0 mm s-1 23,37 In the [Cp*(dppe)Fe-]þ series, QS
.
values ranging between 1.0 and 1.5 are characteristic of
metallacumulenylidene cationic complexes with the general
structure Cp*(dppe)Fe(dC(dC)nR2)]X, and complex 7 can
be regarded as a representative example of this type of com-
pound. For these derivatives an empiric dependence between
the FedC distances and the QS values was observed.35 Accord-
ing to this relationship the iron-carbon distance in 1[PF6] is
˚
˚
expected to be close to 1.80 (0.02 A (computed value 1.827 A).
The Fe-C distance shortens upon one-electron oxidation by ca.
0.07 A. Such a behavior strongly contrasts with all the previous
˚
observations for the known redox couple of [Cp*(dppe)-
Fe-CtC-R]0/þ, for which it was found that the Fe-C bond
distance shortens much less (ca. 0.04 A at maximum).
30
˚
The IS parameters reflect the electron density at the iron
nucleus and decrease as the positive charge is more and more
localized on the metal center or less stabilized by electrostatic
interaction between cations and anions.38 In the case of the
complex 1[PF6], the IS value is very small and compares well
with the data obtained in the case of the iron carbene 7 (IS =
0.141 mm s-1; QS = 1.123 mm s-1), for which the charge is
localized on the metal center.35
Taken as a whole, the variation of the QS and IS param-
eters upon one-electron oxidation indicates that the Fe-C
carbon bond order increases significantly, and the positive
charge density is concentrated on the iron nuclei. As a
consequence, in terms of VB description, the structures 1þb
might have a significant weight in the depiction of the
electronic structure of 1[PF6]; in contrast, 1þa and even more
1þb0 structures should have a negligible contribution. These
observations are in line with the FTIR data and are fully
supported by the DFT calculations (see Section 9). Note that
such an electronic structure also explains why the second
oxidation of 1 is less positive than the first oxidation of 3a.
6. Glass EPR Spectroscopy. Complex 1[PF6] was dis-
solved in a 1:1 CH2Cl2/C2H4Cl2 mixture, and the resulting
solution was transferred into an EPR tube before being
cooled to 68 K. The X-band EPR spectrum of the mono-
cation shows three well-resolved features corresponding to
The UV-vis spectrum of the mono-oxidized species
1[PF6] displays several absorptions that are also found for
other FeIII-alkynyl complexes.7,37 Indeed, upon one-electron
ꢀ
(39) Roue, S.; Le Stang, S.; Toupet, L.; Lapinte, C. C. R. Chim. 2003,
6, 353–366.
(40) Hamon, P.; Justaud, F.; Cador, O.; Hapiot, P.; Rigaut, S.;
Toupet, L.; Ouahab, L.; Stueger, H.; Hamon, J.-R.; Lapinte, C. J. Am.
Chem. Soc. 2008, 130, 17372–17383.
(41) Costuas, K.; Paul, F.; Toupet, L.; Halet, J.-F.; Lapinte, C.
(37) Weyland, T.; Costuas, K.; Mari, A.; Halet, J.-F.; Lapinte, C.
Organometallics 1998, 17, 5569–5579.
Organometallics 2004, 23, 2053–2068.
ꢁ
(42) Weyland, T. Synthese et Proprietes de Complexes Organo-fer
Bi- et Tri-nucleaires sous Differents Degres d’Oxydation. Ph.D. Thesis,
ꢀ ꢀ
ꢀ
€
€
ꢀ
(38) Gutlich, P.; Link, R.; Steinhauser, H. G. Inorg. Chem. 1978, 17,
2509.
Rennes 1, Rennes, 1997.

4634 Organometallics, Vol. 29, No. 20, 2010
Table 6. Visible Absorption Data for 1[PF6]n (n = 0, 1), in CH2Cl2
Miyazaki et al.
a
absorption λ/nm
compd
(103 ε/dm3 mol-1 cm-1
)
ref
1
288 (27.2), 346 (22.3), 400b (12.3)
232 (54.0), 276 (33.0), 336 (12.3), 514 (4.1), 572 (3.8)
233 (31.0), 273b (13.7), 362 (6.9)
402 (5.0), 434a (4.4), 736 (6.6)
297 (9000), 337 (7700), 386 (2000)
282 (6200), 338 (4900), 404 (2600), 466 (4400), 540 (1500), 682 (2600)
270 (4600), 351 (3000), 467 (6000), 540 (2100), 682 (3600)
this work
this work
19
1[PF6]
2
2[PF6]
3a
19
this work
this work
this work
3a[ClO4]d
3a[ClO4]2
e
a Unless otherwise specified. b Shoulder. c UV range not available. d In situ generated by treatment of 3a with 1 equiv of Fe(ClO4)3. e In situ generated
by treatment of 3a with 2 equiv of Fe(ClO4)3.
oxidation, the intense bands at ca. 346-400 nm in 1 are
replaced by two less intense bands at ca. 514-572 nm. It is
likely that upon oxidation of FeII to FeIII, the band due to the
dπ(Fe) f π*(CtC-TTFMe3) MLCT transition vanishes.
Similar loss of the dπ(Fe) f π*(CtC-Ar) MLCT intensities
has been reported in the oxidation of the related compounds
Cp*(dppe)Fe-CtC-Ar to [Cp*(dppe)Fe-CtC-Ar][PF6].24,30,37
Absorption bands lower in energy are also observed in the
spectrum of 3a[ClO4], suggesting that these transitions can be
ascribed to a LMCT transition derived from the whole structure
[Cp*(dppe)FeIII-CtC-TTFMe3] and not only from the or-
ganometallic moiety, as it was found in some instances.43,44
This finding is corroborated by the distribution of the spin
density over the iron-ethynyl and TTFMe3 moieties (see
Section 9).
Figure 3. Near-IR spectrum for 1[PF6] in dichloromethane and
proposed deconvolution.
8. NIR Absorption Spectroscopy. The spectrum of the
radical cation 1[PF6] dissolved in CH2Cl2 exhibits absorp-
tions of moderate intensities in the NIR range. A spectrum
run in CH3CN, a much more polar solvent than CH2Cl2,
evidences an overall solvatochromism toward the higher
energies. Given the possibility that photoinduced electron
transfer can take place from the TTFMe3 moiety to the FeIII
center of 1[PF6], NIR bands might be assigned to ligand-
to-metal charge transfer (LMCT) transitions. An analogy
can be made between these transitions and intervalence
transitions (IVCT) observed in the case of mixed-valence
compounds, and the Marcus-Hush theory may apply.
Assuming that the absorption bands can be approximated
in terms of Gaussian band shapes, the experimental spectra
were deconvoluted in the range 3300-14000 cm-1. Regard-
less of the solvent used, the results show that three Gaussian
curves with similar intensities can describe very accurately
the experimental spectrum (Figure 3). Indeed, the fits are
good enough to allow an almost exact overlay of the sum of
the spectral components with the experimental traces. Table
7 summarizes spectral parameters extracted from the band
shape analysis.
Assignment of the NIR absorptions is not obvious. The
IVCT character of low-energy and high-energy bands was
excluded by the observation of the electronic spectrum of
the unstable dioxidized species 1[PF6]2. Despite its high
kinetic instability due to unavoidable and unknown side
reactions in the experimental conditions, this compound
can be generated in situ. Thus, the complex 1[PF6]2 was pre-
pared in a Schlenk tube by addition of ferrocenium hexa-
fluorophosphate to a CH2Cl2 solution of 1[PF6] cooled
at -60 °C. NIR spectroscopic monitoring of this solution
shows that the band around 7000 cm-1 has disappeared,
while a weak absorption is still present around 5500 cm-1
and a more intense band can be observed at ca. 13 000 cm-1
.
This result clearly indicates that the band found at 6960 cm-1
in the spectrum of 1[PF6] can safely be ascribed to a photo-
induced electron transfer. Consequently, the contribution at
lowest energy probably corresponds to a d-d ligand field
transition. Such transitions are usually observed for mono-
nuclear and polynuclear alkynyl iron(III) complexes.3,37
It has been shown that they are characteristic of the
Cp*(dppe)FeIII fragment and are attributed to a ligand field
(LF) transition from the (SOMO-2) to the SOMO.30 The
weak solvatochromism of the band, as well as its spectral
parameters, is consistent with this attribution.9,45 On the
other hand, the band at 11 200 cm-1, which is blue-shifted
upon one-electron oxidation of 1[PF6], can probably be ascribed
to a LMCT transition.
The presence of an IVCT band in the NIR spectrum shows
that complex 1[PF6], which possesses both inorganic and
organic redox centers, behaves as a mixed-valence complex.
This band can furthermore allow the calculation of the
electronic coupling parameter Hab, which can be computed
using eq 1, where dab is the through-space distance between
˚
the redox centers (A) and the other terms are as given in
Table 7. In the present case determination of dab is not easy.
We have approximated this parameter taking the through-
space distance between the iron atom and the centroid of the
central carbon-carbon double bond of the TTFMe3 core
(dab = 7.46 A). This formula gives a Hab value of 320 cm-1
˚
(43) Wong, K. M.-C.; Lam, S. C.-F.; Ko, C.-C.; Zhu, N.; Yam, V.
ꢀ
Halet, J.-F. Inorg. Chem. 2003, 42, 7086–7097.
(44) Paul, F.; Meyer, W. E.; Toupet, L.; Jiao, H.; Gladysz, J. A.;
W.-W.; Roue, S.; Lapinte, C.; Fathallah, S.; Costuas, K.; Kahlal, S.;
(45) Gauthier, N.; Olivier, C.; Rigaut, S.; Touchard, D.; Roisnel, T.;
Humphrey, M. G.; Paul, F. Organometallics 2008, 27, 1063–1072.
(46) Robin, M. B.; Day, P. Adv. Inorg. Chem. Radiochem. 1967, 10,
247–422.
Lapinte, C. J. Am. Chem. Soc. 2000, 122, 9405–9414.

Article
Organometallics, Vol. 29, No. 20, 2010 4635
Table 7. NIR Absorption Data for 1[PF6] in CH2Cl2 and CH3CN
solvent
CH2Cl2
ν
max (cm-1
)
ε
max (M-1cm-1
)
Δν1/2exp (cm-1
)
Δν1/2cal (cm-1 a
)
Hab (cm-1 b
)
ref
4990
6780
11200
5350
7580
770
1120
750
380
880
850
1750
2100
950
1800
2300
this work
2441
2794
320
337
CH3CN
this work
11900
730
a Calculated from eq 2. b Calculated from eq 1.
ethylene group.47 Knowing that the atomic separations are
particularly sensitive to the charge of the TTFMe3, this
seems to indicate that the Me3TTF moiety acts as a rather
weak electron donor to the iron redox group, confirming
experimental results discussed above.
Scheme 3. Redox Isomers of 1[PF6]
This good agreement between experimental and computed
distances observed for the neutral compound 1 gives con-
fidence in the computed bond distances in the cationic forms
of 1, for which no X-ray data are available yet. Indeed, upon
oxidation, the largest change in the atomic separation occurs
on the Fe-ethynyl group, with some lengthening of the Fe-P
distance, some shortening of the Fe-C37 distance, and a
slight lengthening of the C37-C38 separation, as usually
observed for organometallic compounds containing this
kind of terminus.24,32,37,41-43,48 The TTFMe3 part of the
molecule is less affected—the C-S(av) and central C42-C43
for 1[PF6]. This value is high for a class II compound.46
Hab ¼ 0:0206ðνmaxεmaxΔν1=2
Þ
1=2=dab
ð1Þ
ð2Þ
exp
1=2
ðΔν1=2Þtheo ¼ ½2310ðνmax - ΔG0Þꢀ
Physically, the NIR transition corresponds to a photo-
induced electron transfer from the TTFMe3 core to iron,
giving rise to the excited state II with dominant radical
character on TTFMe3 as depicted in Scheme 3, while the
˚
˚
˚
distances change from 1.783 A to 1.769 A and from 1.357 A
˚
to 1.373 A upon oxidation, respectively—indicating that the
H
ab parameter represents the electronic coupling of the redox
first oxidation occurs mainly at the iron end group as pro-
posed above. Indeed, these atomic separations are very sensi-
tive to oxidation. The corresponding C-C distance increases
isomers (I and II) of 1[PF6]. The former best represents the
ground state with the charge localized on the iron center
based upon the many spectroscopic probes given above and
the DFT calculations discussed below. The free energy
difference (ΔG0) can be approximated by the difference in
oxidation potentials between the mononuclear alkynyl iron
complex 6 and the alkyne-substituted TTFMe3 3a inTable 2
(0.52 V or 4200 cm-1). Equation 2 can then be used to
calculate the theoretical bandwidth of the IVCT band.1,3
The value obtained is ca. 25% larger than that observed
(2441 vs 1750 cm-1). In view of the approximations made,
uncertainty in the numbers, and the relatively large electro-
nic coupling, which favors deviation from the Hush equa-
tions, this provides a valuable a posteriori validation of the use
of the two-level model to extract the electronic coupling
parameter.
9. Electronic Structures. In order to better understand
some of the experimental results, a theoretical investigation
was conducted at the DFT level (see computational details)
on the neutral and oxidized forms of 1, as well as on 3b and
Cp*(dppe)Fe(CtCH) (6) for comparison. Optimized dis-
tances and angles computed for the neutral compound 1
compare rather well with available experimental data found
in Table 1. The largest deviations of the experimental atomic
˚
˚
to 1.394 A, whereas the C-S(av) bonds decrease to 1.753 A
for the parent TTFMe3 molecule 3bþ (see Table 1). Interest-
ingly, the Fe-C37 distance shortens more in 1 than in the
˚
˚
parent molecule 6 upon oxidation (from 1.873 A to 1.827 A,
˚
˚
and from 1.890 A to 1.865 A for 1 and 6, respectively). This
shows a more carbenic character for 1þ and gives some weight
to the Lewis formula 1þb (Scheme 3) to describe its electronic
structure.
A few other computational data such as the energies of the
frequencies of the ethynyl CtC vibrator confirm the con-
clusions that first oxidation occurs mainly at the Fe-ethynyl
moiety. They are given in Table 3 for the three compounds
1nþ, 3bnþ, and 6nþ (n = 0-1). Less energetic vibrational
frequencies are computed for the ethynyl groups upon
oxidation in 1 and 6, but not in 3b, for which oxidation
should occur on the TTFMe3 part.
The molecular orbital diagrams of 1, 3b, and 6 are com-
pared in Figure 4. They all present a large HOMO-LUMO
gap, characteristic of electronically saturated molecules.
Nevertheless, that for 1 is smaller, with the HOMO some-
what antibonding between the Fe-ethynyl and the TTFMe3
moieties higher in energy (vide infra). Accordingly, the
computed first ionization potential (IP) is smaller for 1 than
for the two parent molecules 3b and 6 (4.78, 6.01, and 5.31 eV,
respectively). The fact that the first IP of 1 is substantially
lower than that of 6 indicates that the TTFMe3 group acts as
˚
separations (<0.03 A) are observed for the Fe-P distances,
which are computed somewhat longer than the experimental
ones, a tendency generally noted for this kind of compound.
The Fe-C distance in 1 is slightly shorter than the corre-
˚
sponding one in complex 6 (1.873 vs 1.890 A), whereas the
C-C ethynyl bond is computed to be slightly longer (1.247 vs
1.237 A). C-C and C-S distances in 1 are nearly identical to
those computed for 3b (see Table 1) and somewhat differ
from those computed in Me3TTF-CHdCH-Py, in which the
TTFMe3 and acceptor pyridine moieties are linked by an
(47) Prabusankar, G.; Molard, Y.; Cordier, S.; Golhen, S.; Le Gal,
Y.; Perrin, C.; Ouahab, L.; Kahlal, S.; Halet, J.-F. Eur. J. Inorg. Chem.
2009, 2153–2161.
(48) Bruce, M. I.; Costuas, K.; Davin, T.; Ellis, B. E.; Halet, J.-F.;
Lapinte, C.; Low, P. J.; Smith, K. M.; Skelton, B. W.; Toupet, L.; White,
A. H. Organometallics 2005, 24, 3864–3881.
˚

4636 Organometallics, Vol. 29, No. 20, 2010
Miyazaki et al.
Figure 6. Spin density distribution and Mulliken population
analysis for 1þ (top) and 12þ(BS) (bottom). Contour values
are (0.005 e-/bohr3.
The HOMO of these systems is important in the description
of their geometrical and physical properties upon oxidation.
They are depicted in Figure 5. In the case of compound 1, it is
substantially localized on the Fe center and the ethynyl group
and to a lesser extent on the TTFMe3 part. Analogously, the
HOMO of its parent molecule 6 is heavily located at the
Fe-C2 backbone. Not surprisingly, being mainly Fe-C37
π-antibonding and C37-C38 π-bonding in character, oxida-
tion of 1 and 6 leads to some shortening of Fe-C37 and a
slight lengthening of C37-C38 (see Table 1). In line with the
absence of ethynyl character in the HOMO of 3b, the C37-38
bond length does not change upon oxidation (see Table 1).
The atomic spin densities of the monocationic compound
1þ are given at the top of Figure 6. Their analysis gives some
insight on the localization of the unpaired electron. Interest-
ingly, the spin density is almost equally distributed over the
Fe-ethynyl and the TTFMe3 moieties. A possible explana-
tion for this observation is some reorganization of electron
density after relaxation of the molecular geometry following
oxidation, which mainly occurs on the Fe-ethynyl group
(vide supra).
Figure 4. Molecular orbital diagrams of 1 (center), 3b (right),
and 6 (left).
What about the possible second oxidation experimentally
observed at 0.38 V? The geometries of the dicationic form of
complex 1 within three different spin states, i.e., the closed-
shell singlet (low-spin (LS) diamagnetic state), the triplet
state (high-spin (HS) magnetic state), and the broken sym-
metry singlet (BS) featuring the antiferromagnetic state, were
investigated. Pertinent atomic separations associated with
these different magnetic states are given in Table 1. The three
different geometries hardly differ except the Fe-C(R) dis-
tance, which is computed slightly longer in the singlet state
with respect to the triplet and broken-symmetry states (see
Table 1). Regardless of the magnetic state of 12þ, the
TTFMe3 part is the most affected upon the second oxidation,
with some substantial lengthening of the central C-C dis-
tance and some shortening of the outer C-C bonds
(compare 1þ and 12þ in Table 1). It is important to note
that the most stable electronic configuration is that of the
Figure 5. Plots (contour values: (0.05 [e/bohr3]1/2) and energy
of the HOMO and HOMO-1 of 1 (bottom), 3b (middle), and 6
(top).
a weak electron donor to the iron-ethynyl group during the
first oxidation, as proposed on the basis of the electrochemical
data (see above).

Article
Organometallics, Vol. 29, No. 20, 2010 4637
antiferromagnetic 12þ(BS) structure (with a large exchange
magnetic coupling J (J = E(BS) - E(T)) of -805 cm-1). Its
corresponding spin density plot is shown at the bottom of
Figure 6. Interestingly, unpaired spin density resides both at
the iron-ethynyl group and the TTFMe3 fragment.
(HRMS) were recorded on a high-resolution ZabSpec TOF VG
analytical spectrometer operating in the ESIþ mode, at the
ꢀ
Centre Regional de Mesures Physiques de l’Ouest (CRMPO),
Rennes. Polyethyleneglycol (PEG) was used as internal reference,
and dichloromethane was used as solvent. All mass measure-
ments refer to peaks for the most common isotopes (1H, 12C, 31P,
and 56Fe). EPR spectra were recorded on a Bruker EMX-8/2.7
Conclusion
(X-band) spectrometer. The 57Fe Mossbauer spectra were recorded
€
with a 2.5 ꢁ 10-2 C (9.25 ꢁ 108 Bq) 57Co source using a
In summary, this contribution shows that while weak elec-
tronic interactions take place between the electron-rich iron
center Cp*(dppe)Fe and the TTFMe3 core in the neutral
complex 1, in the oxidized species 1[PF6] a very good electronic
coupling exists between the two electrophores. Consequently,
the TTFMe3 moiety can be regarded as a synthetic metal,12
and the assembly behaves as a heterobimetallic MV complex.
Comparison with MV heterobimetallic complexes possessing
the same Cp*(dppe)Fe unit associated with less electron rich
ends containing iron,25 rhenium,48 and ruthenium44 centers,
which all behave as class II Robin-Day systems, shows that the
electronic communication in the MV complex 1[PF6] is very
good, and the Hab parameter is as large as the largest value
found for these related systems with comparable distances
between the redox centers. Additionally, trends previously
reported for bimetallic related compounds have been further
supported. The reversible redox behavior, polarized electronic
structures with delocalization of the odd electron on the whole
molecule, capabilities for photoinduced electron transfer, and
other properties demonstrated above illustrate the potential for
organometallics associated with synthetic metals (like TTF
derivatives12) in nanoscale devices, including components for
molecular electronics and NLO active assemblies. Work is
currently underway to further explore, define, and elaborate
the electronic and magnetic properties of these hybrid systems
through construction of tailored bridge microenvironments.
symmetric triangular sweep mode. Computer fitting of the
€
Mossbauer data to Lorentzian line shapes was carried out with
a previously reported computer program.36 The isomer shift
values are reported relative to iron foil at 298 K. Elemental
analyses were conducted on a Thermo-Finnigan Flash EA 1112
CHNS/O analyzer by the Microanalytical Service of the CRMPO
at the University of Rennes 1, France.
Cp*(dppe)Fe-CtC-TTFMe3 (1). A Schlenk tube was charged
with 0.678 g (1.08 mmol, 1.05 equiv) of (η2-dppe)(η5-C5-
(CH3)5)Fe-Cl, 0.357 g (1.032 mmol, 1.0 equiv) of Me3TTF-C2-
TMS, 0.185 g (1.13 mmol, 1.1 equiv) of NH4PF6, and 0.066 g
(1.13 mmol, 1.1 equiv) of KF along with 40 mL of MeOH. The
mixture was stirred for 16 h at 25 °C before adding KOBut
(0.100 g). The solvent was then cryogenically trapped, and the
dark orange residue was washed with 10.0 mL of pentane,
partially precipitated from CH2Cl2/pentane, washed three times
with 10 mL of pentane, and dried under vacuum, yielding
complex 1 as an orange powder (0.643 g, 69%). Crystals were
grown by slow diffusion of pentane from a toluene solution of 1
(layer/layer). Anal. Calcd for C47H48P2S4Fe þ 0.5C7H8: C,
67.02; H, 5.79. Found: C, 67.18; H, 5.96. ESI MS (m/z): calcd
for [C47H48P2S4Fe] 858.14636, found (CH2Cl2) 858.1484 ([M]þ,
100%). FT-IR (ν, KBr, cm-1): 2022 (s, CtC). 31P NMR (δ,
C6D6, 81 MHz): 101.8 (s, dppe). 1H NMR (δ, C6D6, 200 MHz):
7.89 (t, JHH = 8.0 Hz, 4H, Hpara), 7.23 (dd, JHH = 7.8 and
8.0 Hz, 8H, Hmeta), 7.08 (dd, 3JHH = 7.8 and 8.0 Hz, 4H, Hortho),
7.02 (t, 3JHH = 7.8 Hz, 4H, Hpara), 2.63 and 1.84 (m, 2H, 2 ꢁ
CHdppe), 1.74 (s, 3H, CH3), 1.46 (s, 15 H, C5(CH3)5), 1.36 and
1.37 (2 x CH3). 13C35 NMR (δ, C6D6): 151.3 (CR, 2JPH = 38.4
Hz), 138.9 and 137.4 (2 ꢁ Cipso/Ar-dppe) 134.1, 134.0, 129.2, 128.8,
127.4, and 127.1 (6 ꢁ CHAr-dppe), 122.5, 122.3, and 122.2 (3 ꢁ
C-CH3/TTF), 116.6 (Cβ), 109.6, 109.5 (2 ꢁ CdCTTF) and 106.0 (Cδ),
87.8 (CCp*), 30.60 (CH2/dppe), 14.62, 12.99, 12.97 (3 ꢁ CH3/TTF),
10.06 (CH3/Cp*).
[Cp*(dppe)Fe-CtC-TTFMe3][PF6] (1[PF6]). A Schlenk tube
was charged with complex 1 (0.210 g, 0.244 mmol, 1.0 equiv),
which was dissolved in THF (10 mL). The solution was then
cooled to -60 °C prior to adding ferricinium hexafluorophos-
phate (0.072 g, 0.22 mmol, 0.9 equiv) at once. The temperature
of the solution was allowed to reach room temperature in 2 h
before adding pentane (20 mL) under vigorous stirring. The
resulting precipitate was subsequently washed with pentane (2 ꢁ
20 mL) to yield 1[PF6] (0.190 g, 77%) as an air-stable, dark violet
powder. FT-IR (ν, KBr, cm-1): 1917 (s, CtC), 835 (PF6). The
CV was identical to that of 1. Anal. Calcd for C47H48F6P3S4Fe:
C, 56.23; H, 4.82. Found: C, 56.37; H, 5.10. ESI-MS (m/z):
calcd for [C47H48P2S4Fe] 858.14636, found (CH2Cl2) 858.1448
[M - PF6].
3
3
Experimental Section
General Procedures. Manipulations of air-sensitive com-
pounds were performed under an argon atmosphere using
standard Schlenk techniques or in an argon-filled Jacomex 532
drybox. Tetrahydrofuran (THF), diethyl ether, toluene, and
pentane were dried and deoxygenated by distillation from
sodium/benzophenone ketyl. Acetone was distilled from P2O5.
Dichloromethane and dichloroethane were distilled under ar-
gon from P2O5 and then from Na2CO3. Methanol was distilled
over dried magnesium turnings. The following compounds were
prepared following published procedures: ferrocenium hexa-
fluorophosphate, [Fe(η5-C5H5)2][PF6],49 and Cp*(dppe)FeCl.21
Potassium tert-butoxide (ACROS) was used without further
purification. Infrared spectra were obtained in KBr with a Bruker
IFS28 FTIR infrared spectrophotometer (4000-400 cm-1).
UV-visible spectra were recorded on a Varian CARY 5000
spectrometer. 1H, 13C, and 31P NMR spectra were recorded on a
Bruker DPX200, an Avance 300, and anAvance500NMR multi-
nuclear spectrometer at ambient temperature, unless otherwise
noted. Chemical shifts are reported in parts per million (δ)
relative to tetramethylsilane (TMS), using the residual solvent
resonances as internal references. Coupling constants (J) are
reported in hertz (Hz), and integrations are reported as numbers
of protons. The following abbreviations are used to describe
peak patterns: br = broad, s = singlet, d = doublet, dd =
double doublet, t = triplet, q = quartet, m = multiplet. 1H and
13C NMR peak assignments are supported by the use of COSY,
HMQC, and HMBC experiments. High-resolution mass spectra
X-ray Crystal Structure Determination. Single crystals were
mounted on a APEXII Bruker-AXS diffractometer equipped
with a CCD camera and a graphite-monochromated Mo KR
˚
radiation source (λ = 0.71073 A), from the Centre de Diffrac-
ꢀ
ꢀ
tometrie (CDFIX), Universite de Rennes 1, France. Data were
collected at 150 K. The structure was solved with a direct
method using the SIR-97 program and refined with a full matrix
least-squares method on F2 using the SHELXL-97 program.50
Crystal data, data collection, and structure refinement param-
eters for 1: empirical formula, C47H48FeP2S4; formula mass, g
(50) Sheldrick, G. M. SHELXL97, Program for Refinement of
€ €
Crystal Structures; University of Gottingen: Gottingen, Germany, 1997.
(49) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877–910.

4638 Organometallics, Vol. 29, No. 20, 2010
Miyazaki et al.
mol-1, 858.88; collection T, K, 150(2); crystal system, mono-
total reflns, 67343; no. unique reflns, 19393 [R(int) = 0.0289];
data/restraints/parameters, 19393/0/989; final R, R1 = 0.0355,
wR2 = 0.0853; R indices (all data), R1 = 0.0489, wR2 = 0.0938;
˚
clinic; space group, P21/c #14; a (A), 15.7427(3); b (A),
˚
3
˚
˚
26.8761(5); c (A), 20.0397(4); β (deg), 92.510(1); V (A ),
8470.7(3); Z, 8; Dcalcd (g cm-3), 1.347; crystal size (mm), 0.08
ꢁ 0.06 ꢁ 0.06; F(000), 3600; abs coeff (mm-1), 0.662; θ range
(deg), 1.50 to 27.50; range h,k,l, -20 e h e 20, -30 e k e 34; no.
2
goodness of fit/F , 1.114; largest diff peak and hole (e A-3), 0.508
˚
and -0.317.
Density Functional Theory Calculations. DFT calculations
were carried out on compounds 1nþ, 3bnþ, and 6nþ (n = 0-2)
with the Gaussian 09 program.51 The geometric structures were
fully optimized without any symmetry constraint using the B86
functional52 within triple-ζ valence basis sets augmented with
polarization functions (TZVP) due to Ahlrichs and co-workers.
Plots of molecular orbitals and spin densities were constructed
using the MOLEKEL program 4.1.53
(51) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.;
Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.;
Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J., J. A.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.;
Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo,
C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin,
A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma,
K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Acknowledgment. This work was financially supported
by Japan Society for Promotion of Science and the Centre
ꢀ
National de la Recherche Scientifique (GDRI joint project
no. 91), which are gratefully acknowledged. T.C. thanks
the CNRS for a postdoctoral grant. We are grateful to
Dr. K. Costuas (Rennes) for helpful discussions.
€
Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.;
Cioslowski, J.; Fox, D. J. Gaussian 09, Revision A.02; Gaussian Inc.:
Pittsburgh, PA, 2009.
(52) Becke, A. D. Phys. Rev. 1988, A38, 3098–3200. Perdew, J. P. Phys.
Rev. 1986, B33, 8822.
Supporting Information Available: Complete set of crystal-
lographic data in CIF format for the X-ray crystal structure of 1.
This material is available free of charge via the Internet at http://
pubs.acs.org.
€
€
(53) Flukiger, P.; Luthi, H. P.; Portmann, S.; Weber, J. Molekel 4.3,
Swiss Center for Scientific Computing: Manno, 2000-2002; http://www.
cscs.ch/molkel.