Organometallics 2010, 29, 4787–4789
DOI: 10.1021/om1004385
4787
The First Linear, Homoleptic Triple-Decker Sandwich Complex of an
f-Element: A Molecular Model for Organolanthanide Nanowires†
Volker Lorenz, Steffen Blaurock, Cristian G. Hrib, and Frank T. Edelmann*
Chemisches Institut der Otto-von-Guericke-Univerit€
at Magdeburg, Universit€
atsplatz 2,
D-39106 Magdeburg, Germany
Received May 7, 2010
Summary: The first structurally authenticated linear, homoleptic triple-decker sandwich complex of an f-element, (η8COT00 )Nd(μ-η8:η8-COT00 )Nd(η8-COT00 ) (2; COT00 = [C8H6(SiMe3)2-1,4]2-) has become available through a rather unexpected
synthetic pathway. Treatment of [Li(THF)4][Nd(COT00 )2] (1)
with anhydrous cobalt(II) chloride (molar ratio ca. 2:1) in a
toluene suspension afforded 2 in 72% isolated yield. The most
notable structural feature of 2 is the near-linear arrangement
of the COT00 rings with (ring centroid)-Nd-(ring centroid)
angles of 176.1 and 175.6°, respectively, and a Nd-(ring centroid)-Nd angle of 177.9°, making the title compound the first
true molecular model for lanthanide-based nanowires.
Molecular spintronics is an exciting new and emergent
subarea of spintronics that benefits from achievements in
molecular electronics and molecular magnetism.1 During the
past few years, molecular spintronics using single molecules
has attracted enormous attention both experimentally and
theoretically, since it holds promise for the next generation of
electronic devices with enhanced functionality and improved
performance, especially in high-density information storage
and quantum computing.2 In this context, various onedimensional organometallic sandwich molecular wires
(SMW’s) have been extensively studied due to their unique
electronic and magnetic properties. Among the promising
examples are lanthanide-based multidecker sandwich complexes of the type Lnn(COT)nþ1 (COT = η8-cyclooctatetraenyl),
†
Part of the Dietmar Seyferth Festschrift.
*To whom correspondence should be addressed. E-mail: frank.
[email protected].
(1) (a) Emberly, E. G.; Kirczenow, G. Chem. Phys. 2002, 281, 311–
324. (b) Rocha, A. R.; Garcia-Suarez, V. M.; Bailey, S. W.; Lambert, C. J.;
Ferrer, J.; Sanvito, S. Nat. Mater. 2005, 4, 335–339. (c) Sanvito, S.; Rocha,
A. R. J. Comput. Theor. Nanosci. 2006, 3, 624–642.
(2) (a) Bogani, L.; Wernsdorfer, W. Nat. Mater. 2008, 7, 179–186.
(b) Coronado, E.; Epstein, A. J. J. Mater. Chem. 2009, 19, 1670–1671.
(c) Mas-Torrent, M.; Crivillers, N.; Mugnaini, V.; Ratera, I.; Rovira, C.;
Veciana, J. J. Mater. Chem. 2009, 19, 1691–1695. (d) Ferrer, J.; GarciaSuarez, V. M. J. Mater. Chem. 2009, 19, 1696–1717.
(3) (a) Kurikawa, T.; Negishi, Y.; Hayakawa, F.; Nagao, S.; Miyajima, K.; Nakajima, A.; Kaya, K. J. Am. Chem. Soc. 1998, 120, 11766–
11772. (b) Kurikawa, T.; Negishi, Y.; Hayakawa, F.; Naguo, S.; Miyajima, K.;
Nakajima, A.; Kaya, K. Eur. Phys. J. D 1999, 9, 283–287. (c) Miyajima, K.;
Kurikawa, T.; Hashimoto, M.; Nakajima, A.; Kaya, K. Chem. Phys. Lett.
1999, 306, 256–262. (d) Nakajima, A.; Kaya, K. J. Phys. Chem. A 2000,
104, 176–191. (e) Hosoya, N.; Takegami, R.; Suzumura, J.; Yada, K.;
Koyasu, K.; Miyajima, K.; Mitsui, M.; Knickelbein, M.; Yabushita, A.;
Nakajima, A. J. Phys. Chem. A 2005, 109, 9–12. (f) Takegami, R.; Hosoya,
N.; Suzumura, J.; Yada, K.; Nakajima, A.; Yabushita, A. Chem. Phys. Lett.
2005, 403, 169–174. (g) Miyajima, K.; Knickelbein, M.; Nakajima, A.
Polyhedron 2005, 24, 2341–2345. (h) Miyajima, K.; Knickelbein, M.;
Nakajima, A. J. Phys. Chem. A 2008, 112, 366–375. (i) Xu, K.; Huang,
J.; Lei, S.; Su, H.; Boey, F. Y. C.; Li, Q.; Yang, J. J. Chem. Phys. 2009, 131,
104704. (j) Zhang, X; Ng, M.-F.; Wang, Y.; Wang, J.; Yang, S.-W. ACS Nano
2009, 3, 2525–2522.
r 2010 American Chemical Society
which can be synthesized via a combination of laser vaporization and molecular beam methods.3 This way nanowires
up to 8 nm in length (Ln = Eu, n = 18) have been obtained.
A major obstacle making these oligomers difficult to manipulate and to investigate is their low solubility. It has recently
been pointed out by Cloke et al. that lanthanide triple-decker
sandwich complexes would provide excellent molecular
models for the Ln(COT) sandwich nanowires.4 Previously
reported triple-deckers based on a COT bridging ligand include a series of lanthanide(II) complexes of the type (μ-COT)[Ln(CpR)]2 (R = Me5, Ln = Sm, Yb, Eu; R = iPr5, Ln =
Yb, Eu), as well as a number of solvated and COT-ring-substituted analogues.4,5 However, unlike the case for the nanowires, all these triple-decker sandwich complexes comprise
significantly bent molecular structures.4,5 Homoleptic
lanthanide triple-deckers of the type Ln2(COT)3 (Ln = Ce,
Nd, Eu, Lu), for which a linear configuration is more likely,
have been known since 1976,6 and Ce2(COT)3 has recently
been reinvestigated by EPR and EXAFS methods, with the
latter showing consistency with a triple-decker sandwich
structure.7 A series of ring-substituted derivatives of the type
Ln2(COT00 )3 (Ln = Ce, Nd, Sm; COT00 = [C8H6(SiMe3)21,4]2-) were reported by us in 1998.8 Thus far, unfortunately,
positive proof of the linear triple-decker structure is lacking,
as no X-ray diffraction data are known to exist for any of
these species. We report here a novel preparation and the
first structural authentication of a homoleptic, linear tripledecker sandwich complex of an f-element.
Reactions of anhydrous lanthanide trichlorides with
Li2(COT00 ) have been shown to strongly depend on the reaction conditions and starting material ratios. Products isolated from such reactions include not only the anionic sandwich complexes [Ln(COT00 )2]- 9 and chloro-bridged dimers
[(COT00 )Ln(μ-Cl)THF)]210 but also unusual cluster-centered
(4) Summerscales, O. T.; Jones, S. C.; Cloke, F. G. N.; Hitchcock,
P. B. Organometallics 2009, 28, 5896–5908 and references cited therein.
(5) (a) Evans, W. J.; Clark, R. D.; Ansari, M. D.; Ziller, J. W. J. Am.
Chem. Soc. 1998, 120, 9555–9563. (b) Evans, W. J.; Johnston, M. A.; Greci,
M. A.; Ziller, J. W. Organometallics 1999, 18, 1460–1464. (c) Walter, M. D.;
Wolmersh€auser, G.; Sitzmann, H. J. Am. Chem. Soc. 2005, 127, 17494–
17503.
(6) (a) Ely, S. R.; Hopkins, T. E.; DeKock, C. W. J. Am. Chem. Soc.
1976, 98, 1624–1625. (b) Greco, A.; Cesca, S.; Bertolini, G. J. Organomet.
Chem. 1976, 113, 321–330.
(7) Walter, M. D.; Booth, C. H.; Lukens, W. W.; Andersen, R. A.
Organometallics 2009, 28, 698–707.
(8) (a) Poremba, P.; Edelmann, F. T. J. Organomet. Chem. 1998, 553,
393–395. (b) Edelmann, F. T. New J. Chem. 1995, 19, 535–550. (c) Edelmann,
F. T.; Freckmann, D. M. M.; Schumann, H. Chem. Rev. 2002, 102, 1851–
1896.
(9) Poremba, P.; Reissmann, U.; Noltemeyer, M.; Schmidt, H.-G.;
Br€
user, W.; Edelmann, F. T. J. Organomet. Chem. 1997, 544, 1.
Published on Web 08/09/2010
pubs.acs.org/Organometallics
4788
Organometallics, Vol. 29, No. 21, 2010
Lorenz et al.
Scheme 1
multidecker sandwich complexes11 as well as the bimetallics
Ln2(COT00 )3.8 Although the latter could be isolated from
LnCl3/Li2(COT00 ) reactions in a molar ratio of 2:3, all
attempts to grow single crystals failed within the course of
our initial investigation.8 In a continuation of this work we
have now found a rather unexpected and novel access to the
triple-decker molecules. Treatment of [Li(THF)4][Nd(COT00 )2]
(1)8 with anhydrous cobalt(II) chloride (molar ratio ca. 2:1)
in a toluene suspension produced a dark green solution
accompanied by formation of a black precipitate of metallic
Co (admixed with LiCl).12 Crystallization of the toluenesoluble material from cyclopentane afforded dark green,
prismlike crystals, which were identified as the pure tripledecker sandwich complex (η8-COT00 )Nd(μ-η8:η8-COT00 )Nd(η8-COT00 ) (2). Its formation (72% isolated yield) in this
unexpected redox process is illustrated in Scheme 1. In the
course of the reaction, one of the dianionic COT00 ligands in
1 is oxidized to 1,4-bis(trimethylsilyl)cyclooctatetraene (admixed
with its 1,6 isomer), which was isolated from the concentrated mother liquid and identified unambiguously by comparison with the NMR data (1H, 13C) of an authentic specimen.10b,13 Apparently two important factors contribute to
the success of this novel synthetic route. One is certainly the
enhanced solubility of the product imparted by the use of the
(10) (a) Burton, N. C.; Cloke, F. G. N.; Hitchcock, P. B.; de Lemos,
H. C.; Sameh, A. A. J. Chem. Soc., Chem. Commun. 1989, 1462–1464.
(b) Burton, N. C.; Cloke, F. G. N.; Joseph, S. C. P.; Karamallakis, H.; Sameh,
A. A. J. Organomet. Chem. 1993, 462, 39–43.
(11) Lorenz, V.; Edelmann, A.; Blaurock, S.; Freise, F.; Edelmann,
F. T. Organometallics 2007, 26, 4708.
(12) Synthesis of 2: solid [Li(THF)4][Nd(COT00 )2] (1; 2.13 mmol, 2.0
g) was added to a stirred suspension of CoCl2 (1.16 mmol, 0.15 g) in
toluene (150 mL) at room temperature. The reaction mixture was stirred
for 24 h and refluxed for an additional 5 h, causing the formation of a
black precipitate. After filtration the solvent was evaporated under
vacuum. The solid residue was extracted with 150 mL of cyclopentane.
Ca. 75 mL of cyclopentane was evaporated under vacuum. (η8COT00 )Nd(η8-COT00 )Nd(η8-COT00 ) (2) crystallized upon standing at
room temperature as dark green prisms (0.85 g, 72%), which were
suitable for X-ray crystallography. Anal. Calcd for C47H82Nd2Si6
(Mr = 1104.2): C, 51.13; H, 7.49. Found: C, 50.57; H, 7.32. IR (KBr
disk): ν 2995 w, 2953 s, 2895 m, 1445 w, 1404 m, 1313 w, 1247 vs, 1213 m,
1046 s, 1037 m, 981 m, 937 m, 909 m, 840 vs, 782 m, 749 s, 733 s, 686 m,
651 w, 636 m, 548 m, 501 w cm-1. Mass spectrum (EI): m/z 1032 (100%)
[{C8H6(SiMe3)2}3Nd2], 1017 (14%), 959 (18%) [{C8H6(SiMe3)2}3Nd2 Si(CH3)3], 784 (58%) [{C8H6(SiMe3)2}2Nd2], 711 (18%), 639 (13%)
[{C8H6(SiMe3)2}2Nd]. 1H NMR (600 MHz, d8-toluene, 25 °C): δ 1.01 (br
s, outer ring Si(CH3)3), -1.35 (br s, inner ring Si(CH3)3) (intensity ratio
2:1); -3.86 (br s, COT-H), -7.77 (br s, COT-H), -9.01 (br s, COT-H),
-13.77 (br s, COT-H), -14.57 (br s, COT-H), -18.25 (br s, COT-H)
ppm. 13C NMR (100.6 MHz, d8-toluene, 25 °C): δ 177.3, 170.5, 163.6,
136.3, 103.9 (br, COT00 ), -1.5, -1.7 (Si(CH3)3) ppm. 29Si NMR (79.5
MHz, d8-toluene, 25 °C): δ -44.7 (inner ring SiMe3), -63.3 (outer ring
SiMe3) ppm (intensity ratio 1:2).
(13) (a) Wadepohl, H.; Merkel, R.; Pritzkow, H.; Ruhm, S. Dalton
Trans. 2001, 3617–3626. (b) Wadepohl, H.; Gebert, S.; Merkel, R.; Pritzkow,
H. J. Organomet. Chem. 2002, 641, 142–147.
Figure 1. Molecular structure of 1. Selected bond lengths (Å)
and angles (deg): XCOT0 0 A-Nd1 = 1.894, Nd1-XCOT0 0 B = 2.164,
XCOT0 0 B-Nd2 = 2.156, Nd2-XCOT0 0 C = 1.906, CCOT0 0 A-Nd1 =
2.622(4)-2.688(4), Nd1-CCOT0 0 B = 2.811(4)-2.917(4), CCOT0 0 BNd2 = 2.805(4)-2.885(4), Nd2-CCOT0 0 C = 2.629(4)-2.673(4);
XCOT0 0 A-Nd1-XCOT0 0 B = 176.1, XCOT0 0 B-Nd2-XCOT0 0 C = 175.6,
Nd1-XCOT0 0 B-Nd2 = 177.9. XCOT0 0 = COT00 ring centroid.
bulky, silyl-substituted COT00 ligands.10 The other is the
heterogeneous reaction in a nonpolar solvent (toluene),
which prevents the formation of solvated species. It was
shown in early work by DeKock et al. that Ln2(COT)3 tripledeckers formed by cocondensation techniques afforded the
solvated species [Ln(COT)(THF)2][Ln(COT)2] upon contact
with THF.6a
A mass spectrum of 2 showed the molecular ion of the
triple-decker at m/z 1032 with 100% relative intensity. Due to
the paramagnetic nature of the Nd3þ ions, the 1H NMR resonances are spread over a range of ca. 20 ppm. Dark green,
prismlike, X-ray-quality crystals were grown by slow cooling
of a saturated solution in cyclopentane. An X-ray diffraction
analysis of 214 clearly established the presence of the first
homoleptic, linear triple-decker sandwich complex of an
f-element (Figure 1).
All three COT00 rings are η8-coordinated to neodymium,
with the central ring acting as a μ-η8:η8 bridging ligand. The
(14) Crystal data for complex 2: C47H82Nd2Si6 3 c-C5H10; Mr =
1104.15, monoclinic, space group P21/n, a = 13.2914(3) Å, b =
17.2442(3) Å, c = 24.3534(4) Å, β = 102.6801(15)°, Z = 4, T =
153(2) K, μ = 2.041 mm-1, green prism. Of 45 491 reflections measured,
11 010 were independent (Rint = 0.0819). Final R1 = 0.0407, wR2 =
0.1017 (all data). Supplementary crystallographic data can be found in
the Cambridge Crystallographic Data file CCDC-740282.
Communication
Organometallics, Vol. 29, No. 21, 2010
4789
Nd-C distances to the outer rings are in the ranges of
2.622(4)-2.688(4) Å for Nd1 and 2.629(4)-2.673(4) Å for
Nd2. This can be compared to average Nd-C distances of
2.660(24), 2.673(16), and 2.787(19) Å reported for the
Nd-C(η8-COT) bond lengths in [Nd(COT)(THF)2][Nd(COT)2].6a Significantly longer Nd-C distances (2.811(4)2.917(4) Å for Nd1 and 2.805(4)-2.885(4) for Nd2) are
found for the bridging COT00 ligand. The corresponding
Nd-(ring centroid) distances are 1.894 Å (Nd1) and 1.906
Å (Nd2) to the outer rings and 2.164 Å (Nd1) and 2.156 Å
(Nd2) to the bridging ring. These values can be favorably
compared to corresponding Nd-C(ring centroid) distances
of, for example, 1.957 Å in (COT)NdI(THF)315 or 1.981 and
2.027 Å in (COT)Nd(μ-η8:η2-COT)Li(THF)3.15 The most
notable structural feature of 2, however, is the near-linear
arrangement of the COT00 rings with (ring centroid)-Nd(ring centroid) angles of 176.1° (Nd1) and 175.6° (Nd2),
respectively, and a Nd-(ring centroid)-Nd angle of 177.9°.
The observed minor deviations from perfect linearity can be
traced back to repulsive interactions between methyl groups
of the ring substituents. Thus far, a comparable near-linear
arrangement of rings has only been achieved in the Yb(II)
quadruple-decker complex Cp*Yb(μ-η8:η8-COT000 )Yb(μ-η8:η8-COT000 )YbCp*, which contains the very bulky 1,3,6-tris-
(trimethylsilyl)cyclooctatetraenyl (=COT000 ) bridging ligand.16
In contrast, all other triple-decker compounds containing
lanthanide elements comprise bent configurations.4 It should
be noted at this stage that the reaction was initially carried out
with the intention to make a heterobimetallic Nd/Co quadruple-decker sandwich complex.
In summarizing the main results of the present study, the
first linear, homoleptic triple-decker sandwich complex of an
f-element, (η8-COT00 )Nd(μ-η8:η8-COT00 )Nd(η8-COT00 ) (2),
has been made available through a novel high-yield synthetic
route involving oxidation of the anionic sandwich complex
[Li(THF)4][Nd(COT00 )2] (1) with CoCl2. Future work in this
direction will show if this method can be employed to pepare
other Ln2(COT00 )3 derivatives and perhaps even the parent
triple-decker sandwiches Ln2(COT)3. In any case the Nd
complex 1 represents the first true molecular model for
organolanthanide-based nanowires.
(15) Meermann, C.; Ohno, K.; T€
ornroos, K. W.; Mashima, K.;
Anwander, R. Eur. J. Inorg. Chem. 2009, 76–85.
(16) Edelmann, A.; Blaurock, S.; Lorenz, V.; Hilfert, L.; Edelmann,
F. T. Angew. Chem., Int. Ed. 2007, 46, 6732–6734.
Supporting Information Available: CIF file giving X-ray
structural data for 2. This material is available free of charge
via the Internet at http://pubs.acs.org.
Acknowledgment. This work was generously supported by the Deutsche Forschungsgemeinschaft (SPP
1166 “Lanthanoid-spezifische Funktionalit€
aten in
Molek€
ul und Material”). Financial support by the Ottovon-Guericke-Universit€at Magdeburg is also gratefully
acknowledged.