Evidence for LiBr-assisted generation of a silylene from a 1,2-diaryl-1,2-dibromodisilene

Article abstract of DOI:10.1002/asia.201100833

Instant silylene! LiBr assists the dissociation of 1,2-diaryl-1,2- dibromodisilenes. Reactions with various trapping agents gave products identical to those from corresponding reactions with aryl bromosilylenes. The reaction provides a method for the in situ generation of silylenes under mild conditions, and the reaction rate is increased by the addition of LiBr.

Full text of DOI:10.1002/asia.201100833

COMMUNICATION
DOI: 10.1002/asia.201100833
Evidence for LiBr-Assisted Generation of a Silylene from a 1,2-Diaryl-1,2-
dibromodisilene
Joon Soo Han,[b] Takahiro Sasamori,*[a] Yoshiyuki Mizuhata,[a] and Norihiro Tokitoh*[a]
Silylenoids R2SiMX (structures denoted as A–C) are the
silicon analogues of carbenoids, where X denotes a leaving
group (e.g. halogen or alkoxy) and M is a metal atom.[1] Sily-
lenoids have comparable reactivity to silylenes (D) and
react with both electrophiles and nucleophiles. Tamao and
Kawachi showed that Ph2tBuOSiLi underwent a self-con-
densation to give a product of structure type E.[1b] Nowa-
days, the synthesis of disilenes (F) by reduction of the corre-
sponding halosilane is well established, and silylenoids were
postulated as intermediates in this reaction.[2] The first
of prototypical trapping reagents for reactive intermediates
(Scheme 1) to give the products 2–10, which are identical to
the trapping products of the corresponding aryl bromosily-
stable silylenoid (tBu2MeSi)2SiFLi·ACTHNUTRGNE(UNG THF)3 was reported re-
cently, and its molecular structure was revealed as C.[2d] The
synthesis of halosilylenoids and their silylene-like reactivity
have been reported extensively.[3]
Scheme 1. Silylene-like reactivity of 1 in benzene at room temp. All reac-
tions yield the respective products quantitatively.
Wiberg et al. reported the synthesis of the first stable bis-
ACHTUNGTRENNUNG(silyl)dihalodisilene and showed that the dihalodisilene has
À
typical reactivity of a disilene, giving products with Si Si
bonds derived from Si=Si bonds.[4] Recently, we successfully
lene.[6] All reactions proceeded very cleanly when the re-
spective trapping agent was added to a solution of 1 in ben-
zene and the reaction mixture was left to stand at room tem-
perature until the yellow color of 1 disappeared. Generally,
to trap reactive silylenes, a large excess of trapping agent
(about 40-fold) was required. In the case of 1, approximately
a sixfold excess was enough to complete the reaction in
about 5 h (except for cyclohexene, which took 24 h). The
most beneficial advantage of these reactions is the clean and
quantitative conversion in all reactions shown in Scheme 1.
After the completion of the reactions, excessive reagent can
be removed by evaporation. However, in the case of benzo-
phenone or diphenylacetylene, which are difficult to
remove, a stoichiometric amount (2 equiv) was used, and
the reactions were completed within 6.5 h and three days,
respectively.
synthesized
a
stable 1,2-dibromodisilene Bbt(Br)Si=
[bis(trimethylsilyl)methyl]-4-[tris-
Si(Br)Bbt (1, Bbt=2,6-bisAHCTNUGTRENNNUG
(trimethylsilyl)methyl]phenyl).[5b] Disilene 1 is highly sensi-
tive towards oxygen, both in solution and in the solid state,
but stable at room temperature under the exclusion of air
and moisture.
During the course of our studies using 1, we found unex-
pected reactivity patterns. Disilene 1 reacted with a variety
[a] Prof. Dr. T. Sasamori, Prof. Dr. Y. Mizuhata, Prof. Dr. N. Tokitoh
Institute for Chemical Research
Kyoto University
Gokasho, Uji, Kyoto 611-0011 (Japan)
Fax : (+81)774-38-3209
E-mail: sasamori@boc.kuicr.kyoto-u.ac.jp
tokitoh@boc.kuicr.kyoto-u.ac.jp
At first, we believed the silylene-like reactivity of 1 to be
a result of the thermal dissociation of 1 into two equivalents
of the corresponding silylenes.[7] We have previously demon-
strated that extremely hindered disilenes, for example, (E)-
[b] Dr. J. S. Han
Nano-Materials Research Center
Korea Institute of Science and Technology
39-1 Hawolgok-dong, Seongbuk-gu, Seoul 136-791 (Korea)
and (Z)-Mes
phenyl, Tbt=2,4,6-tris
G
ACHTUNGTRENN(UNG Tbt)Mes (12, Mes=2,4,6-trimethyl-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201100833.
AHCTUNGTRENNUNG
298
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2012, 7, 298 – 300

dergo ready thermal dissociation into the corresponding sily-
lenes.[7b] Variable-temperature UV/Vis and NMR spectros-
copy (À20 to 708C) suggested the thermal dissociation in
the germanium analogue of 1, Bbt(Br)Ge=Ge(Br)Bbt
(13).[5a] In sharp contrast, we were not able to detect any
changes in the variable-temperature UV/Vis and 1H NMR
spectra of 1 (À60 to 408C),[5b] suggesting that 1 does not un-
dergo thermal dissociation into the corresponding silylenes.
How can the silylene-like reactivity of 1 be otherwise ex-
plained? To explore the possibility of a photochemical disso-
ciation process,[8] we conducted comparable control experi-
ments under the exclusion of light and corroborated that the
silylene-like reactivity is not the result of a photochemically
induced dissociation process.
Despite the negative results from a thermal dissociation
mechanism, we could not rule out the presence of undetect-
able amounts of dissociated BbtBrSi: in equilibrium with 1.
However, there is one crucial experimental result, which we
are unable to bring into accordance with a thermal dissocia-
tion mechanism. The thermally induced decomposition of 1
detected spectroscopically, and acceptable values for C, H,
and Br analysis were obtained.[5b] It is therefore unlikely
that organic impurities are responsible for the silylene-like
reactivity of 1. Secondly, we turned our attention to the pos-
sibility of inorganic impurities, especially LiBr. Although
LiBr was removed from the hexane solutions of 1 by filtra-
tion through celite, residual contamination with a trace
amount of LiBr is virtually unavoidable.
The solubility of LiBr in nonpolar solvents such as ben-
zene is too low for equivalent amounts of LiBr and 1 to be
dissolved. However, LiBr is sufficiently soluble in polar sol-
vents such as dimethoxyethane (DME). We therefore tested
the reaction between 1 and excess cyclohexene with LiBr in
DME. Surprisingly, the characteristic yellow color of 1 dis-
appeared within seven minutes, and the corresponding
trapped product 9 was obtained in a quantitative yield.
Without the addition of LiBr, the same reaction required
24 h for quantitative conversion (Scheme 3). Also, increased
reaction rate by the addition of LiBr was observed in the re-
action of 1 with excess triethylsilane in THF.[9]
results in the quantitative formation of cyclized product
[5b]
À
(11), which still contains a Si Si bond. If the thermal dis-
sociation of 1 would result in the formation of low-coordi-
nate silylenes, why does the thermal decomposition product
À
11 still contain a Si Si bond? In previous studies, we estab-
lished that bulky disilene 12 can thermally dissociate into si-
À
lylenes, which undergo intramolecular cyclization by C H
[7a]
À
insertion to give 14, which has no Si Si bond (Scheme 2).
Scheme 3. Effect of LiBr: Reaction rate is increased, thermolysis product
À
15 is obtained, which does not contain a Si Si bond.
Furthermore, we investigated the thermolysis in the pres-
ence of LiBr. A control experiment in DME delivered the
reaction times and product formation (11) identical to those
for the reaction in benzene. However, by the addition of
LiBr (4 equiv), the thermolysis reached completion after
40 min and resulted in the formation of 15 (Scheme 3).
We conclude that the silylene-like reactivity of 1 is proba-
bly the result of a trace amount of LiBr, and the most likely
intermediates are the corresponding silylenoid and silylene
(Scheme 4). Comparable reactions with purified 1 (obtained
by passing hexane solutions of 1 through a short-path neu-
tral silica-gel column in a glove-box) with phenylacetylene
required more than twice the reaction time but resulted in
the same product 7. This result is understandable, since a
complete removal of LiBr from 1 is extremely difficult. Sev-
Scheme 2. Thermal decomposition products of disilenes 1 and 12.
The interesting result obtained from the thermolysis of 1
prompted us to search for another mechanistic pathway in
addition to thermal dissociation. Firstly, we ensured purity
of 1. Compound 1 was prepared by the reduction of
BbtSiBr3 with exactly two equivalents lithium naphthalenide
at À788C in THF, followed by a general workup procedure:
After removal of THF, naphthalene was separated by subli-
mation at 458C/0.01 mmHg. Trace residues of THF were re-
moved by repeated co-evaporation with hexane, and LiBr
was removed by filtration from the hexane solutions of 1.
Analytically pure, powdered samples were obtained from
concentration of the hexane filtrates. No impurities could be
Scheme 4. Possible equilibrium of 1 in the presence of LiBr.
Chem. Asian J. 2012, 7, 298 – 300
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
299

T. Sasamori, N. Tokitoh et al.
COMMUNICATION
eral attempts to detect intermediate species spectroscopical-
ly were unsuccessful.[10]
[1] a) P. Boudjouk, P. U. Samaraweera, Angew. Chem. 1988, 100, 1406–
1407; Angew. Chem. Int. Ed. Engl. 1988, 27, 1355–1356; b) K.
Tamao, A. Kawachi, Angew. Chem. 1995, 107, 886–888; Angew.
Chem. Int. Ed. Engl. 1995, 34, 818–820.
[2] a) N. Wiberg, W. Niedermayer, J. Organomet. Chem. 2001, 628, 57–
64; b) R. Fischer, J. Baumgartner, G. Kickelbick, C. Marschner, J.
Am. Chem. Soc. 2003, 125, 3414–3415; c) M. Zirngast, M. Flock, J.
Baumgartner, C. Marschner, J. Am. Chem. Soc. 2008, 130, 17460–
17470; d) G. Molev, D. Bravo-Zhivotovskii, M. Karni, B. Tumanskii,
M. Botoshansky, Y. Apeloig, J. Am. Chem. Soc. 2006, 128, 2784–
2785; e) J. Harloff, E. Popowski, H. Reinke, J. Organomet. Chem.
2007, 692, 1421–1441.
[3] a) M. E. Lee, H. M. Cho, M. S. Ryu, C. H. Kim, W. Ando, J. Am.
Chem. Soc. 2001, 123, 7732–7733; b) M. E. Lee, H. M. Cho, Y. M.
Lee, J. K. Choi, C. H. Park, S. E. Jeong, U. Lee, Chem. Eur. J. 2004,
10, 377–381; c) Y. M. Lim, H. M. Cho, M. E. Lee, K. K. Baeck, Or-
ganometallics 2006, 25, 4960–4964; d) Y. M. Lim, C. H. Park, S. J.
Yoon, H. M. Cho, M. E. Lee, K. K. Baeck, Organometallics 2010, 29,
1355–1361; e) H. M. Cho, Y. M. Lee, M. E. Lee, Dalton Trans. 2010,
39, 9232–9234.
[4] a) N. Wiberg, H. Auer, S. Wagner, K. Polborn, G. Kramer, J. Orga-
nomet. Chem. 2001, 619, 110–131; b) N. Wiberg, W. Niedermayer,
G. Fischer, H. Nçth, M. Suter, Eur. J. Inorg. Chem. 2002, 1066–
1070.
[5] a) T. Sasamori, Y. Sugiyama, N. Takeda, N. Tokitoh, Organometallics
2005, 24, 3309–3314; b) T. Sasamori, K. Hironaka, Y. Sugiyama, N.
Takagi, S. Nagase, Y. Hosoi, Y. Furukawa, N. Tokitoh, J. Am. Chem.
Soc. 2008, 130, 13856–13857.
[6] For a recent review on metallylenes, see: Y. Mizuhata, T. Sasamori,
N. Tokitoh, Chem. Rev. 2009, 109, 3479–3511.
[7] a) N. Tokitoh, H. Suzuki, R. Okazaki, K. Ogawa, J. Am. Chem. Soc.
1993, 115, 10428–10429; b) N. Takeda, N. Tokitoh, Synlett 2007,
2483–2491 and references therein; c) S. Tsutsui, E. Kwon, H.
Tanaka, S. Matsumoto, K. Sakamoto, Organometallics 2005, 24,
4629–4638.
Our results presented herein provide evidence for the
LiBr-assisted generation of a silylene from a 1,2-diaryl-1,2-
dibromodisilene. Compared to disilenes, silylenes are highly
reactive, and their synthesis and handling is extremely diffi-
cult due to their limited stability. This work provides the
proof of concept for an alternative and complementary
method for the in situ generation of silylenes from disilenes.
It is noteworthy that the reaction products from this study
À
still contain a Si Br bond, which can be exploited in subse-
quent reactions. Mechanistic studies, scope, and optimization
studies of this new reaction are currently in progress.
Experimental Section
General method for the synthesis of 2--10. An NMR tube was charged
with 1 (20 mg, 14 mmol), trapping agent (2–6 equiv), and C6D6 (0.5 mL)
in an MBraun glove-box. The mixture was left to stand at room tempera-
ture until the characteristic yellow color of 1 disappeared. The resulting
solution contained the reaction product in quantitative yield, as evident
from 1H NMR spectroscopy. Individual reaction times depended on the
amount and the nature of the trapping agent and ranged from three
hours to three days. An excess amount of trapping agent shortened the
reaction time; however, no side product was observed, even when a stoi-
chiometric amount of trapping agent was used. If an excess amount of
trapping agent was used, removal of solvent and unreacted trapping
agent gave an analytically pure product.[10]
Acknowledgements
[8] S. Masamune, Y. Eriyama, T. Kawase, Angew. Chem. 1987, 99, 601–
602; Angew. Chem. Int. Ed. Engl. 1987, 26, 584–585.
This work was supported by a Grant-in Aid for Scientific Research (B)
(No. 22350017), Young Scientist (A) (No. 23685010), and the Global
COE Program B09 (“Integrated Materials Science”, Kyoto University)
from the Ministry of Education, Culture, Sports, Science and Technology
(MEXT), Japan. J.S.H. thanks KIST for financial support (Future Key
Technology Program).
[9] Reactions in THF gave similar results, but small amounts of by-
À
products were observed due to the instability of the Si Br bond in
THF.
[10] For the detailed results see the Supporting Information.
Received: October 6, 2011
Published online: December 1, 2011
Keywords: disilenes · reactive intermediates · silanes ·
silylenes · silylenoids
300
www.chemasianj.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2012, 7, 298 – 300