An organosilicon cluster with an octasilacuneane core: A missing silicon cage motif

Article abstract of DOI:10.1002/anie.201208506

Cagey silicon: A silicon cluster, consisting of sixteen silicon atoms and composed of an octasilacuneane core (red, see picture) fused with cyclotetrasilanes (orange), was synthesized by reductive tetramerization of tetrachlorocyclotetrasilane. Analytical and theoretical studies reveal the unique structural and electronic features of this organosilicon cluster. Copyright

Full text of DOI:10.1002/anie.201208506

Angewandte
Chemie
DOI: 10.1002/anie.201208506
Organosilicon Chemistry
An Organosilicon Cluster with an Octasilacuneane Core: A Missing
Silicon Cage Motif**
Shintaro Ishida,* Kyohei Otsuka, Yuki Toma, and Soichiro Kyushin*
Organosilicon clusters (caged oligosilanes) are a fascinating
class of molecules because 1) they can be regarded as partial
structures of bulk crystalline or amorphous silicon and 2) the
electronic properties of silicon clusters depend on silicon
numbers and structural modification.[1] Therefore, synthetic
studies on organosilicon clusters have been extensively
developed. Persilapolyhedranes,[2–4] decasilaadamantane,[5]
pentasila[1.1.1]propellane,[6] persila[n]staffane (catenated sil-
icon cages),[7] and related siliconoids[8,9] have all been shown
to possess unique electronic characteristics owing to devel-
around 1–2 ppm, owing to an intractable polymeric mixture.
Compound 1 was sparingly soluble in common organic
solvents.
À
oped s conjugation among the caged Si Si bonds. Octasila-
cuneane,
a
silicon analog of cuneane[10] (pentacy-
clo[3.3.0.02,4.03,7.06,8]octane), is a missing saturated silicon
cluster motif, but potentially important because octasilacu-
neane is an isomeric silicon cage of octasilacubane,[4] one of
the most well-known silicon polyhedra. Moreover, through
theoretical studies at the MP2/6-31G(d) + ZPE level, Ver-
steeg and Koch have demonstrated that parent octasilacu-
neane (Si8H8) is only 3.3 kcalmolÀ1 less stable than the
isomeric octasilacubane, which is in sharp contrast to the
energetic relation between cubane and cuneane: cubane
(C8H8) is much more unstable (by 43.7 kcalmolÀ1) than
cuneane in their calculations.[11] Cubanes readily isomerize
to the corresponding cuneanes by metal-ion catalyzed skeletal
rearrangements.[10a]
The construction of the silicon cage could be ascribed to
the formation of an energetically reasonable structure by
successive bond formation during the reaction. To elucidate
this idea, we estimated the relative energies of two fused
persilacubanes and three fused persilacuneanes (3-D2d, 3-D4d,
4-C2, 4-C2v, and 4-Cs) by theoretical calculations at the
B3LYP/6-31G(d) level, as shown in Figure 1.[14] Whereas
Herein, we report the synthesis, structure, and properties
of cyclotetrasilane-fused persilacuneane (1), which consists of
sixteen silicon atoms. Compound 1 was found in the course of
our study on ladder oligosilanes.[12] As shown in Equation (1),
À
the reductive tetramerization (formation of eight Si Si bonds
during the reaction) of tetrachlorocyclotetrasilane 2[13] with
sodium gave silicon cluster 1 as air-sensitive, thermally stable
orange crystals in 15% yield. No structural isomers of 1 were
1
found in the reaction mixture. The H NMR spectrum of the
reaction mixture showed a set of eight singlet signals from the
tert-butyl protons of 1 together with a very broad signal
Figure 1. Relative energies (kcalmolÀ1) of compounds 3 and 4.
[*] Dr. S. Ishida,[+] Dr. K. Otsuka, Y. Toma, Prof. Dr. S. Kyushin
Department of Chemistry and Chemical Biology, Graduate School of
Engineering, Gunma University, Kiryu, Gunma 376-8515 (Japan)
E-mail: kyushin@gunma-u.ac.jp
[+] Present address: Department of Chemistry, Graduate School of
Science, Tohoku University, Aoba-ku, Sendai 980-8578 (Japan)
E-mail: sishida@m.tohoku.ac.jp
parent persilacuneane is less stable than persilacubane by
3.3 kcalmolÀ1 at the MP2/6-31G(d) + ZPE level[11] (8.3 kcal
molÀ1 at the B3LYP/6-31G(d) level),[14] 4-C2, a model of 1, is
the most stable among the five isomers. The introduction of
cyclotetrasilane units into the silicon cluster can change the
relative energies of different silicon cages.
The molecular structure of 1 was determined by X-ray
crystallography, as shown in Figure 2.[15] Compound 1 has
crystallographic C2 symmetry with the axis through the
[**] This work was supported in part by Grants-in-Aid for Scientific
Research from the Ministry of Education, Culture, Sports, Science
and Technology (Japan) and the Japan Society for the Promotion of
Science. We also thank Prof. Dr. Takeshi Yamanobe, Gunma
University (Japan) for measurement of the CP-MAS NMR spectrum.
À
À
midpoints of the Si1 Si1* and Si5 Si5* bonds, and a well-
developed network structure consisting of sixteen silicon
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201208506.
Angew. Chem. Int. Ed. 2013, 52, 2507 –2510
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2507

.
Angewandte
Communications
Figure 4. 29Si cross polarization magic angle spinning (CP-MAS) NMR
spectrum of 1 at room temperature with signals calculated by the
GIAO method at the B3LYP/6-311+G(2d,p) level. Atom numbers
correspond to those in Figure 2.
consistent with the molecular structure in the solid state.
These signals were essentially reproduced by GIAO calcu-
lation at the B3LYP/6-311 + G(2d,p) level using the opti-
mized structure, and can be assigned as follows: Three
characteristic signals shifted upfield at À71.8, À62.4, and
À56.6 ppm are due to the 29Si nuclei of the cyclotrisilane rings
(Si1, Si4, and Si8).[20,21] The signals at 25.0, 29.1, 35.4, and
39.0 ppm can be assigned to any of 29Si nuclei of Si2, Si3, Si5,
and Si6.[22,23]
Figure 2. Molecular structure of 1. Thermal ellipsoids are drawn at the
30% probability level. Selected bond lengths [pm]: Si1–Si2 238.2(1),
Si1–Si4 238.4(1), Si1–Si1* 238.0(2), Si1–Si8* 233.2(1), Si2–Si3
252.2(1), Si3–Si4 241.3(1), Si4–Si5 240.4(1), Si4–Si8* 235.4(1), Si5–Si6
241.8(1), Si5–Si8 236.1(1), Si5–Si5* 241.0(2), Si6–Si7 251.0(1), Si7–Si8
242.0(1).
À
atoms. The Si Si bond lengths of peripheral cyclotetrasilane
moieties (238.2(1)–252.2(1) pm) are slightly elongated
because of steric hindrance from the tert-butyl groups.
Although the recently reported compound, pentasila-
[1.1.1]propellane, has inverted tetrahedral silicon atoms with
Notably, the 29Si signal of Si7 at 66.4 ppm is shifted rather
downfield. A possible explanation may be given by the large
paramagnetic term (sp) of Si7. This term has a predominant
effect on the chemical shift of 29Si NMR. According to
a theoretical equation,[24] sp is roughly inversely proportional
to the averaged transition energy. The downfield shift of the
29Si signal of Si7 is probably due to its contribution to the
HOMO and the LUMO of 1.
Theoretical calculations were carried out to obtain insight
into the electronic characterization of 1. Frontier orbitals are
located on the silicon skeleton, indicating well-developed
s conjugation (see the Supporting Information). The major
contribution of the HOMO is an out-of-phase interaction
À
a significantly long bridgehead Si Si bond (263.6(1) pm; its
bonding character has been controversial)[6] the bridgehead
À
Si Si bond lengths of 1 (Si1–Si4 238.4(1) pm, Si5–Si8
236.1(1) pm) are within standard values.[16,17] However, the
bridgehead silicon atoms (Si1, Si4, Si5, and Si8) have highly
deformed trigonal monopyramidal structures, as illustrated in
Figure 3. Each silicon atom has a planar structure, in which
among the sSi1 Si4, sSi1 Si8*, and sSi7* Si8* orbitals and among the
À
À
À
sSi1* Si4*, sSi1* Si8, and sSi7 Si8 orbitals. The major contribution of
À
À
À
the LUMO consists of an out-of-phase interaction between
the pseudo-p* orbitals of two peripheral cyclotetrasilane rings
(Si5-Si6-Si7-Si8 and Si5*-Si6*-Si7*-Si8* rings).
The UV/Vis spectra of 1 and our heptacyclic ladder
oligosilane, iPr20Si16 (5),[25] both of which consist of sixteen
Figure 3. Geometries around the deformed silicon atoms of 1.
À À
the sum of three Si Si Si bond angles is almost 3608. The
À
remaining Si Si bond is almost perpendicular to the plane in
À
the cases of Si5 and Si8, whereas the remaining Si Si bond is
silicon atoms, are compared in Figure 5. Compound 5 has
well-developed s conjugation along the long axis of the
molecule and shows the lowest energy absorption as a dis-
tinctive band at 464 nm.[25] In contrast, the absorption of
1 does not show clear bands above 250 nm and tails to ca.
570 nm, except for some shoulders at 345 and 380 nm. The
absorption tailing is characteristic of silicon polyhedral
tilted to form a three-membered ring in the cases of Si1 and
Si4. Similar trigonal monopyramidal silicon atoms have been
found in the silicon clusters,[8,9d] tris(triisopropylsilyl)silane,[18]
and unsolvated tris(di-tert-butylmethylsilyl)silyllithium.[19]
Figure 4 shows the solid state 29Si CP-MAS NMR
spectrum of 1. Eight signals were observed, which is
2508 www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 2507 –2510

Angewandte
Chemie
Vol. 2 (Eds.: Z. Rappoport, Y. Apeloig), Wiley, Chichester, 1998,
pp. 2177 – 2216.
[2] For tetrasilatetrahedranes, see: a) N. Wiberg, C. M. M. Finger,
K. Polborn, Angew. Chem. 1993, 105, 1140 – 1142; Angew. Chem.
Int. Ed. Engl. 1993, 32, 1054 – 1056; b) M. Ichinohe, M.
Toyoshima, R. Kinjo, A. Sekiguchi, J. Am. Chem. Soc. 2003,
125, 13328 – 13329.
[3] For hexasilaprismane, see: A. Sekiguchi, T. Yatabe, C. Kabuto,
H. Sakurai, J. Am. Chem. Soc. 1993, 115, 5853 – 5854.
[4] For octasilacubanes, see: a) H. Matsumoto, K. Higuchi, Y.
Hoshino, H. Koike, Y. Naoi, Y. Nagai, J. Chem. Soc. Chem.
Commun. 1988, 1083 – 1084; b) K. Furukawa, M. Fujino, N.
Matsumoto, Appl. Phys. Lett. 1992, 60, 2744 – 2745; c) A.
Sekiguchi, T. Yatabe, H. Kamatani, C. Kabuto, H. Sakurai, J.
Am. Chem. Soc. 1992, 114, 6260 – 6262; d) H. Matsumoto, K.
Higuchi, S. Kyushin, M. Goto, Angew. Chem. 1992, 104, 1410 –
1412; Angew. Chem. Int. Ed. Engl. 1992, 31, 1354 – 1356; e) K.
Furukawa, M. Fujino, N. Matsumoto, J. Organomet. Chem. 1996,
515, 37 – 41; f) M. Unno, T. Matsumoto, K. Mochizuki, K.
Higuchi, M. Goto, H. Matsumoto, J. Organomet. Chem. 2003,
685, 156 – 161.
Figure 5. UV/Vis spectra of 1 in methylcyclohexane and 5 in hexane at
room temperature. 1 (c), 5 (a).
[5] J. Fischer, J. Baumgartner, C. Marschner, Science 2005, 310, 825.
[6] D. Nied, R. Kçppe, W. Klopper, H. Schnçckel, F. Breher, J. Am.
Chem. Soc. 2010, 132, 10264 – 10265, and references therein.
[7] T. Iwamoto, D. Tsushima, E. Kwon, S. Ishida, H. Isobe, Angew.
Chem. 2012, 124, 2390 – 2394; Angew. Chem. Int. Ed. 2012, 51,
2340 – 2344.
compounds: UV/Vis absorption of hexasilaprismane (DipSi)6
(Dip = 2,6-diisopropylphenyl) tails to ca. 500 nm,[3] and those
of octasilacubanes (DepSi)8 (Dep = 2,6-diethylphenyl) and
[(tBuMe2Si)Si]8 tail to ca. 470 nm.[1c,4d] The absorption edge of
1 is remarkably red-shifted compared with those of these
hexasilaprismane and octasilacubanes.
The characteristic tailing of 1 is probably due to close
energy levels of molecular orbitals with a small band gap. The
HOMO and the LUMO have energy levels close to other
molecular orbitals, and therefore, absorption bands overlap to
give a closely overlapped absorption curve without peaks.
From these results, organosilicon cluster 1 may be regarded as
an intermediate model between small organosilicon mole-
cules and silicon semiconductors that have the band struc-
tures of orbitals.[26]
In summary, we have synthesized a cyclotetrasilane-fused
octasilacuneane 1, which consists of sixteen silicon atoms, by
reductive tetramerization of tetrachlorocyclotetrasilane.
X-ray crystallography, theoretical studies, and analysis by
NMR and UV/Vis spectroscopy revealed the unique struc-
tural and electronic features of 1, which are due to ring strain
and developed s conjugation. Investigations of the reactivity
of 1 and the expansion of s conjugation by introducing
functional groups are now ongoing and will appear in the near
future.
[8] For a Si8(SitBu3)6 cluster, see: G. Fischer, V. Huch, P. Mayer,
S. K. Vasisht, M. Veith, N. Wiberg, Angew. Chem. 2005, 117,
8096 – 8099; Angew. Chem. Int. Ed. 2005, 44, 7884 – 7887.
[9] Scheschkewitz and co-workers have proposed the term “silico-
noids” for silicon clusters with unsubstituted silicon vertexes;
see: a) D. Scheschkewitz, Angew. Chem. 2005, 117, 3014 – 3016;
Angew. Chem. Int. Ed. 2005, 44, 2954 – 2956; b) K. Abersfelder,
A. J. P. White, H. S. Rzepa, D. Scheschkewitz, Science 2010, 327,
564 – 566; c) K. Abersfelder, A. J. P. White, R. J. F. Berger, H. S.
Rzepa, D. Scheschkewitz, Angew. Chem. 2011, 123, 8082 – 8086;
Angew. Chem. Int. Ed. 2011, 50, 7936 – 7939; d) K. Abersfelder,
A. Russell, H. S. Rzepa, A. J. P. White, P. R. Haycock, D.
Scheschkewitz, J. Am. Chem. Soc. 2012, 134, 16008 – 16016.
[10] For cuneanes, see: a) L. Cassar, P. E. Eaton, J. Halpern, J. Am.
Chem. Soc. 1970, 92, 6366 – 6368; b) R. Gleiter, S. Brand,
Tetrahedron Lett. 1994, 35, 4969 – 4972; c) H. Irngartinger, S.
Strack, R. Gleiter, S. Brand, Acta Crystallogr. Sect. C 1997, 53,
1145 – 1148; d) R. Gleiter, S. Brand, Chem. Eur. J. 1998, 4, 2532 –
2538; e) M. V. Roux, J. Z. Dꢁvalos, P. Jimꢂnez, R. Notario, O.
CastaÇo, J. S. Chickos, W. Hanshaw, H. Zhao, N. Rath, J. F.
Liebman, B. S. Farivar, A. Bashir-Hashemi, J. Org. Chem. 2005,
70, 5461 – 5470.
[11] U. Versteeg, W. Koch, J. Comput. Chem. 1994, 15, 1151 – 1162.
[12] For reviews on ladder oligosilanes, see: a) H. Matsumoto, S.
Kyushin, M. Unno, R. Tanaka, J. Organomet. Chem. 2000, 611,
52 – 63; b) S. Kyushin, H. Matsumoto, Adv. Organomet. Chem.
2003, 49, 133 – 166.
Received: October 22, 2012
Published online: January 30, 2013
[13] Tetrachlorocyclotetrasilane 2 was prepared by the head-to-head
reductive dimerization of 1,2-dichlorodisilane S3 and subsequent
dephenylchlorination. For details of the synthesis and X-ray
crystal analyses of 1 and 2, see the Supporting Information.
[14] We also calculated the energies of the parent cubane, cuneane,
octasilacubane, and octasilacuneane at the B3LYP/6-31G(d)
level. Our results were comparable to the previous report by
Versteeg and Koch.[11] See the Supporting Information for
details.
[15] CCDC 906219 (S4), 906220 (2), and 906221 (1) contain the
supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Keywords: cluster compounds · cuneanes · organosilicon ·
propellanes · structure elucidation
.
[1] For reviews on cyclic oligosilanes and silicon cage compounds,
see: a) E. Hengge, R. Janoschek, Chem. Rev. 1995, 95, 1495 –
1526; b) A. Sekiguchi, H. Sakurai, Adv. Organomet. Chem. 1995,
37, 1 – 38; c) A. Sekiguchi, S. Nagase in The Chemistry of
Organic Silicon Compounds, Vol. 2 (Eds.: Z. Rappoport, Y.
Apeloig), Wiley, Chichester, 1998, pp. 119 – 152; d) E. Hengge,
H. Stꢀger in The Chemistry of Organic Silicon Compounds,
Angew. Chem. Int. Ed. 2013, 52, 2507 –2510
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 2509

.
Angewandte
Communications
[16] The silicon skeleton of 1 contains hitherto unknown persila-
[21] a) H. Matsumoto, A. Sakamoto, Y. Nagai, J. Chem. Soc. Chem.
Commun. 1986, 1768 – 1769; b) M. Kira, T. Iwamoto, T. Maru-
yama, T. Kuzuguchi, D. Yin, C. Kabuto, H. Sakurai, J. Chem. Soc.
Dalton Trans. 2002, 1539 – 1544.
[3.2.1]- and [3.2.2]propellanes and spiro[3.3]- and [4.3]oligosi-
lanes as substructures. Compound
1 also has a spiro-
[3.2]hexasilane moiety. Some spiro[3.2]hexasilane derivatives
have been reported.[8,17d]
[22] The 29Si signals of Si2, Si3, and Si6 were observed within the
range of reported tert-butyl-substituted cyclotetrasilanes:
[17] For spirooligosilanes, see: a) P. Boudjouk, R. Sooriyakumaran, J.
Chem. Soc. Chem. Commun. 1984, 777 – 778; b) T. Iwamoto, M.
Tamura, C. Kabuto, M. Kira, Science 2000, 290, 504 – 506; c) J.
Hlina, C. Mechtler, H. Wagner, J. Baumgartner, C. Marschner,
Organometallics 2009, 28, 4065 – 4071; d) T. Iwamoto, Y. Furiya,
H. Kobayashi, H. Isobe, M. Kira, Organometallics 2010, 29,
1869 – 1872.
[18] a) S. Kyushin, H. Sakurai, H. Matsumoto, Chem. Lett. 1998, 107 –
108; b) P. P. Gaspar, A. M. Beatty, T. Chen, T. Haile, D. Lei,
W. R. Winchester, J. Braddock-Wilking, N. P. Rath, W. T.
Klooster, T. F. Koetzle, S. A. Mason, A. Albinati, Organometal-
lics 1999, 18, 3921 – 3932.
tBu7Si4H = 31.7
and
41.8 ppm,[23a]
trans-tBu6Si4H2 =
32.0 ppm,[23a] tBu2(tBu2MeSi)3Si4Br3 = 37.3 ppm,[23b] and trans-
tBu6Si4Br2 = 24.7 ppm.[23c]
[23] a) S. Kyushin, H. Sakurai, H. Matsumoto, J. Organomet. Chem.
1995, 499, 235 – 240; b) A. Sekiguchi, T. Matsuno, M. Ichinohe, J.
Am. Chem. Soc. 2001, 123, 12436 – 12437; c) S. Kyushin, H.
Kawai, H. Matsumoto, Organometallics 2004, 23, 311 – 313.
[24] a) N. F. Ramsey, Phys. Rev. 1950, 78, 699 – 703; b) A. Saika, C. P.
Slichter, J. Chem. Phys. 1954, 22, 26 – 28.
[25] S. Kyushin, Y. Ueta, R. Tanaka, H. Matsumoto, Chem. Lett. 2006,
35, 182 – 183.
[19] M. Nakamoto, T. Fukawa, V. Ya. Lee, A. Sekiguchi, J. Am.
Chem. Soc. 2002, 124, 15160 – 15161.
[26] For reviews on group 13 and group 14 clusters, and their
intermediate role between small molecules and the bulk state,
see: a) H. Schnçckel, Dalton Trans. 2005, 3131 – 3136; b) A.
Schnepf, Chem. Soc. Rev. 2007, 36, 745 – 758; c) A. Schnepf, New
J. Chem. 2010, 34, 2079 – 2092. See also Reference [9d].
[20] The 29Si signals of silyl-substituted cyclotrisilanes appear in
upfield regions: (Et3Si)6Si3 = À174.4 ppm,[21a] (tBuMe2Si)6Si3 =
À156.5 ppm,[21b] and (iPr2MeSi)6Si3 = À148.6 ppm.[21b]
2510
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 2507 –2510