Synthesis of a pentasilapropellane. Exploring the nature of a stretched silicon-silicon bond in a nonclassical molecule

Article abstract of DOI:10.1021/ja104810u

We report on the successful synthesis of Si₅Mes₆ (Mes = 2,4,6-trimethylphenyl), which consists of an archetypal [1.1.1] cluster core featuring two ligand-free, inverted tetrahedral bridgehead silicon atoms. The separation between the bridgehead Si atoms is much longer, and the bond strength much weaker, than usually observed for a regular Si-Si single bond. A detailed analysis of the electronic characteristics of Si ₅Mes₆ reveals a low-lying excited triplet state, indicative of some biradical(oid) character. Reactivity studies provide evidence for both closed-shell and radical-type reactivity, confirming the unusual nature of the stretched silicon-silicon bond in this nonclassical molecule.

Full text of DOI:10.1021/ja104810u

Published on Web 07/13/2010
Synthesis of a Pentasilapropellane. Exploring the Nature of a Stretched
Silicon-Silicon Bond in a Nonclassical Molecule
Dominik Nied, Ralf Ko¨ppe, Wim Klopper,* Hansgeorg Schno¨ckel, and Frank Breher*
Faculty of Chemistry and Biosciences, Karlsruhe Institute of Technology, Kaiserstrasse 12,
D-76131 Karlsruhe, Germany
Received June 12, 2010; E-mail: klopper@kit.edu; breher@kit.edu
Scheme 1
Abstract: We report on the successful synthesis of Si5Mes6 (Mes
) 2,4,6-trimethylphenyl), which consists of an archetypal [1.1.1]
cluster core featuring two ligand-free, “inverted tetrahedral” bridge-
head silicon atoms. The separation between the bridgehead Si
atoms is much longer, and the bond strength much weaker, than
usually observed for a regular Si-Si single bond. A detailed analysis
of the electronic characteristics of Si5Mes6 reveals a low-lying excited
triplet state, indicative of some biradical(oid) character. Reactivity
studies provide evidence for both closed-shell and radical-type
reactivity, confirming the unusual nature of the stretched silicon-silicon
bond in this “nonclassical” molecule.
below). The elemental analysis and electron impact (EI) mass spectra
were consistent with the composition Si5Mes6. 1 is EPR-silent at room
temperature and 100 K. The 1H and 13C NMR spectra of 1 (in C6D6)
are very similar to those found for Ge5Mes6,17 further supporting the
formation of the pentasila analogue. Most revealingly, two resonances
were observed in the 29Si NMR spectrum at δ ) 25.5 and -273.2
ppm for the bridging (Sibr) and bridgehead (Sib) atoms in 1, respectively
(δcalcd ) 34 and -270 ppm; see Supporting Information).
Despite the sensitivity of 1, we were able to grow single crystals
suitable for X-ray structure analysis (Figure 1). As expected for
[1.1.1]propellane core structures, 1 consists of two ligand-free bridge-
head silicon atoms (Si1 and Si2), which are bonded to three bridging
silicon atoms (Si3, Si4, and Si5) at distances between 233.2(1) and
236.0(1) pm. Of particular interest is the separation of 263.6(1) pm
between the two bridgehead silicon atoms, which is ca. 30 pm (13%)
longer than usually observed for a regular Si-Si single bond. Based
on this very long separation and previous quantum chemical calcula-
tions,17 it appears that the Sib · · ·Sib interaction in 1 is fairly weaksin
this case considerably stretched, though not fully broken. Note that a
comparably elongated Si-Si bond of 269.7 pm was found by Wiberg
et al. for the sterically encumbered disilane tBu3Si-SitBu3.18 The latter
was shown to be a convenient source (heating to ca. 50 °C) for very
reactive supersilyl radicals (tBu3Si•). The bond strength in
tBu3Si-SitBu3 was estimated to be considerably weaker than in normal
disilanes.
Silicon-based analogues of hydrocarbons intrigue chemists for a
number of reasons. They are often fundamentally different from their
carbon counterparts and have remained a challenge for both experi-
mentalists and theoreticians for a long time.1 After the first isolation
of a kinetically stabilized disilene R2SidSiR2 by West et al. in 1981,2
the synthesis and characterization of several silicon-silicon multiple
bonds, spiro- and bicyclic structures, and clusters followed.3 The
isolation of the first spirocyclic pentasiladiene4 fused the chemistry of
unsaturated SidSi moieties with that of annulated ring systems such
as tetrasilabicyclo[1.1.0]butanes.5 However, the parent structure of these
annulated ring systems, that is, the pentasila[1.1.1]propellane Si5R6,
remained elusive until now. This heavier homologue of the widely
investigated carbon propellanes C5R66 was targeted by theory more
than 2 decades ago7 and designated a synthetic challenge.8 Unsur-
prisingly, only three closely related silicon clusters consisting of
“naked” Si atoms have been reported so far, by Wiberg9 and
Scheschkewitz.10
The intrinsic [1.1.1] scaffold of propellanes belongs to the “nonclas-
sical” structures showing “inverted tetrahedral” bridgehead atoms. The
bonding between the bridgehead atoms is far from trivial. Even the
well-established all-carbon [1.1.1]propellane has attracted renewed
interest from both experimentalists11 and theoreticians.12 The heavy
[1.1.1]propellanes [E5R6] (E ) Si, Ge, Sn) have frequently been
described as biradicaloids13 due to the considerably stretched bond
between their bridgeheads, Eb. Although numerous quantum chemical
calculations7 have been performed on these species, synthetically
accessible and structurally characterized species are very rare.14-17
In the present contribution, we report on the synthesis and characteriza-
tion of pentasila[1.1.1]propellane Si5Mes6 (1), the last missing member
within the Group 14 element series, finally closing the gap between
theoretical predictions and discussions.
With these observations on the related though topologically different
Si2R6 molecule in mind, we tried to explore the weak Sib · · ·Sib bond
For the synthesis of 1 (Scheme 1), we reacted Si2Cl6 with 3 equiv
of Mes2SiCl2 and 12 equiv of a freshly prepared lithium naphthalenide
solution (Li+[Naphth]•-, 0.8 M in THF).17 1H NMR spectroscopic
monitoring of the crude product of the reaction showed that the target
compound 1 is formed in about 10% yield. After a nonoptimized
workup procedure, 1 was isolated by column chromatography and
recrystallization from toluene/acetonitrile in analytically pure form as
bright yellow crystals. 1 is extremely sensitive to air and moisture (see
Figure 1. Molecular structure of Si5Mes6 (1).
9
10264 J. AM. CHEM. SOC. 2010, 132, 10264–10265
10.1021/ja104810u  2010 American Chemical Society

C O M M U N I C A T I O N S
to 1 as compared to the heavier homologues. In contrast, however,
we found a higher reactivity of the silicon propellane.
The first experimental evidence came from observations regarding
the moisture sensitivity of 1. Note that Sn5Dep6 (Dep ) 2,6-Et2C6H3)14
and Ge5Mes617 are stable toward degassed water, without any sign of
decomposition and/or reaction. Hydrocarbon solutions of 1, however,
rapidly decolorize when exposed to traces of water. In order to address
this as well as the general reactivity of 1 a little further, we performed
some preliminary NMR tube scale reactions using selected reagents.
It appears that 1 shows both closed-shell and radical-type reactivity.
Several reagents, such as H2O, PhSH, PhOH, and Me3SnH (not
Me3SiH), can readily be added across the bridge, furnishing the
corresponding bicyclo[1.1.1]pentasilane derivatives (see Supporting
Information). Evidence for biradicaloid reactivity of 1 came from
studies using typical reagents for radical-type reactivity, such as
Me3SnH or 9,10-dihydroanthracene. The latter gave the dihydrogen
adduct H21 in low yield after prolonged reaction times.
Figure 2. (a) Experimental UV-vis spectrum of 1 in THF and most relevant
TD-DFT calculated UV-vis transitions of A2 and E symmetry (b) DFT
calculated frontier orbitals of Si5Dmp6 (1q): HOMO-1 (a1, -5.86 eV); HOMO
(e, -5.64 eV); and LUMO (a2, -1.77 eV).
in 1 a little further, both experimentally by Raman spectroscopy and
theoretically by time-dependent (TD)-DFT calculations on Si5Dmp6
(1q, Dmp ) 2,6-Me2C6H3, see Supporting Information). However, the
aim to extract the contribution of the restoring Sib · · ·Sib bond forces
from the totally symmetric Raman bands cannot be achieved for 1
with the desired accuracy, because too many internal coordinates are
involved in each of the totally symmetric normal modes observed at
ν ) 479, 370, and 340 cm-1 (νcalcd ) 455, 349, and 312 cm-1). In
order to obtain at least an estimation of the Sib · · ·Sib bond strength in
1, we calculated the reaction energy for the process 1q + H2 f H21q.
Accordingly, the addition of H2 to the bridgehead Si atoms is
exothermic by ∆H°(0 K) ) -90 kJ mol-1. By using D(H2) ) 436 kJ
mol-1 and assuming a Si-H bond strength of ∼350 kJ mol-1 for silyl-
substituted silanes,19 the Sib · · ·Sib bond strength can be estimated to
amount to ∼174 kJ mol-1. Based on this approximation and as
expected, the Sib · · ·Sib interaction is considerably weaker than in
normal disilanes (ca. 306-332 kJ mol-1).19
All these promising results clearly indicate distinctive peculiarities
of the stretched bond between the bridgehead atoms in heavy
propellanes in general, and in particular in 1. Once designated as
synthetic challenge and targeted by theory over 2 decades ago,
pentasila[1.1.1]propellane is indeed revealed to be an intriguing species
from the perspectives of both bonding and reactivity.
Acknowledgment. We thank the Fonds der Chemischen Industrie,
the German Science Foundation (CFN, Project No. C3.3), and the
Ministry of Science, Research and the Arts of Baden-Wu¨rttemberg
(Az: 7713.14-300).
Supporting Information Available: Details of experimental proce-
dures, preliminary reactivity studies, analytical data, X-ray structure
determination, and quantum chemical calculations (PDF, CIF). This
material is available free of charge via the Internet at http://pubs.acs.org.
Alongside these remarkable bonding features, we determined some
interesting electronic properties for 1. The electronic transitions
observed in the experimental UV-vis spectrum of 1 in THF fit
perfectly to the TD-DFT calculated singlet excitations for the model
compound 1q (Figure 2).17 The calculated vertical singlet excitation
wavelength of 325.7 nm (1A2, HOMO-1 (a1) f LUMO (a2)) directly
corresponds to the experimentally observed absorption at λ ) 325 nm,
whereas λmax ) 396 nm belongs to electronic transitions between the
cluster-bonding HOMO (e) and the LUMO (a2) (Figure 2). The
excitation to the first excited triplet state (3A2) was calculated to
correspond to a wavelength of 546.8 nm.17 Usually, these transitions
are not visible in the UV-vis spectrum. The experimental spectrum
of 1, however, shows a broad absorption of very low intensity at λ )
546 nm (Figure 2). Its wavelength corresponds to an excitation energy
of 219 kJ mol-1, that is, about 25 and 10 kJ mol-1 below the respective
excitation energies of its Ge and Sn analogues.17 This is remarkable,
because the first singlet A2 excitation energy of 1q is larger than those
of its analogues. Whereas the increase in the HOMO-1 f LUMO
gaps along the series Sn f Ge f Si is consistent with the computed
singlet excitation energies, the close proximity (reflected by the rather
localized excited ground state difference densities of the 1A2 and 3A2
states, see Supporting Information) of the orbitals involved in 1q leads
to a larger exchange interaction 〈a1a2|a2a1〉 and extra stabilization of
the 3A2 state compared to the Ge and Sn analogues.
The relatively high LUMO energy of 1q (-1.77 eV; cf. -2.10 eV
for Ge5Dmp6 and -2.48 eV for Sn5Dmp5) was experimentally
confirmed by electrochemical studies using cyclic voltammetry under
strictly anaerobic and dry conditions. By analogy to the heavy Ge17
and Sn propellanes,14 1 is quasi-reversibly reduced to the radical anion
[Si5Mes6]•- and dianion [Si5R6]2- at potentials of E°1/2 ) -2.88 and
-3.12 V, respectively. Compared to the heavier homologues, however,
both half-wave potentials are more cathodically shifted, which is in
accord with an energetically destabilized LUMO. On the basis of these
findings, one would a priori expect a less facile addition of nucleophiles
References
(1) (a) Power, P. P. Nature 2010, 463, 171. (b) Chivers, T.; Konu, J. Comments
Inorg. Chem. 2009, 30, 131.
(2) West, R.; Fink, M. J.; Michl, J. Science 1981, 214, 1343.
(3) See, for instance: (a) Sekiguchi, A.; Kinjo, R.; Ichinohe, M. Science 2004,
305, 1755. (b) Wiberg, N.; Vasisht, S. K.; Fischer, G.; Mayer, P. Z. Anorg.
Allg. Chem. 2004, 630, 1823. (c) Weidenbruch, M.; Willms, S.; Saak, W.;
Henkel, G. Angew. Chem., Int. Ed. 1997, 36, 2503. (d) Ishida, S.; Iwamoto,
T.; Kabuto, C.; Kira, M. Nature 2003, 421, 725. (e) Wang, Y.; Xie, Y.;
Wei, P.; King, R. B.; Schaefer, H. F., III; von, P.; Robinson, G. H. Science
2008, 321, 1069.
(4) Iwamoto, T.; Tamura, M.; Kabuto, C.; Kira, M. Science 2000, 290, 504.
(5) Rohmer, M.-M.; Be´nard, M. Chem. Soc. ReV. 2001, 30, 340.
(6) Wiberg, K. B. Chem. ReV. 1989, 89, 975.
(7) (a) Schleyer, P. von R.; Janoschek, R. Angew. Chem., Int. Ed. 1987, 26,
1267. (b) Schoeller, W. W.; Dabisch, T.; Busch, T. Inorg. Chem. 1987,
26, 4383. (c) Nagase, S.; Kudo, T. Organometallics 1987, 6, 2456. (d)
Nagase, S.; Kudo, T. Organometallics 1988, 7, 2534. (e) Nagase, S.
Polyhedron 1991, 10, 1299. (f) Gordon, M. S.; Nguyen, K. A.; Carroll,
M. T. Polyhedron 1991, 10, 1247.
(8) Tsumuraya, T.; Batcheller, S. A.; Masamune, S. Angew. Chem., Int. Ed. 1991,
30, 902.
(9) Fischer, G.; Huch, V.; Mayer, P.; Vasisht, S. K.; Veith, M.; Wiberg, N. Angew.
Chem., Int. Ed. 2005, 44, 7884.
(10) (a) Scheschkewitz, D. Angew. Chem., Int. Ed. 2005, 44, 2954. (b)
Abersfelder, K.; White, A. J. P.; Rzepa, H. S.; Scheschkewitz, D. Science
2010, 327, 564.
(11) Messerschmidt, M.; Scheins, S.; Grubert, L.; Pa¨tzel, M.; Szeimies, G.;
Paulmann, C.; Luger, P. Angew. Chem., Int. Ed. 2005, 44, 3925.
(12) (a) Wu, W.; Gu, J.; Song, J.; Shaik, S.; Hiberty, P. C. Angew. Chem., Int. Ed.
2009, 48, 1407. (b) Shaik, S.; Danovich, D.; Wu, W.; Hiberty, P. C. Nature
Chem. 2009, 1, 443.
(13) Breher, F. Coord. Chem. ReV. 2007, 251, 1007.
(14) (a) Sita, L. R.; Bickerstaff, R. D. J. Am. Chem. Soc. 1989, 111, 6454. (b)
Sita, L. R.; Kinoshita, I. J. Am. Chem. Soc. 1990, 112, 8839. (c) Sita, L. R.;
Kinoshita, I. J. Am. Chem. Soc. 1991, 113, 5070. (d) Sita, L. R.; Kinoshita,
I. J. Am. Chem. Soc. 1992, 114, 7024.
(15) Drost, C.; Hildebrand, M.; Lo¨nnecke, P. Main Group Metal Chem. 2002,
25, 93.
(16) Richards, A. F.; Brynda, M.; Power, P. P. Organometallics 2004, 23, 4009.
(17) Nied, D.; Klopper, W.; Breher, F. Angew. Chem., Int. Ed. 2009, 48, 1411.
(18) (a) Wiberg, N.; Schuster, H.; Simon, A.; Peters, K. Angew. Chem., Int. Ed.
1986, 25, 79. (b) Wiberg, N. Coord. Chem. ReV. 1997, 163, 217.
(19) Becerra, R. Walsh, R. In The Chemistry of Organic Silicon Compounds,
Vol. 2; Rappoport, Z., Apeloig, Y., Eds.; Wiley: Chichester, 1998.
JA104810U
9
J. AM. CHEM. SOC. VOL. 132, NO. 30, 2010 10265