Fluoride ion mediated reactions of disubstituted acetylenes Me3SiCCMMe3 (M=C, Si, Ge, Sn) with terminal acetylenes

Article abstract of DOI:10.1016/S0022-328X(01)01109-3

The CsF-18-crown-6 mediated reactions of disubstituted acetylenes Me₃SiCCMMe₃ (M=C, Si, Ge, Sn) with phenylacetylene and 2-methyl-5-pyridylacetylene in benzene have been studied. The first step of the reaction is the deprotonation of

Full text of DOI:10.1016/S0022-328X(01)01109-3

Journal of Organometallic Chemistry 634 (2001) 69–73
www.elsevier.com/locate/jorganchem
Fluoride ion mediated reactions of disubstituted acetylenes
Me3SiCꢀCMMe3 (M=C, Si, Ge, Sn) with terminal acetylenes
(
Edmunds Lukevics *, Kira Rubina, Edgars Abele, Mendel Fleisher, Pavel Arsenyan,
Solveiga Gr¯ınberga, Juris Popelis
Lat6ian Institute of Organic Synthesis, 21 Aizkraukles Street, Riga LV-1006, Lat6ia
Received 18 April 2001; accepted 4 July 2001
Dedicated to Professor M.G. Voronkov on the occasion of his 80th birthday
Abstract
The CsF–18-crown-6 mediated reactions of disubstituted acetylenes Me3SiCꢀCMMe3 (M=C, Si, Ge, Sn) with phenylacetylene
and 2-methyl-5-pyridylacetylene in benzene have been studied. The first step of the reaction is the deprotonation of aryl acetylene
by F−-ion. The carbanion formed interacts with the disubstituted acetylene. The silylated, germylated and stannylated acetylenes
were formed in yields 26–91%. Quantum chemical calculations of trimethylsilyl and trimethylgermyl group transfer have been
performed. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: 1-tert-Butyl-2-trimethylsilyl-, bis(trimethylsilyl)-, 1-trimethylgermyl-2-trimethylsilyl-, and 1-trimethylstannyl-2-trimethylsilylacetylenes;
Phenylacetylene and 2-methyl-5-pyridylacetylene; Fluoride ion mediated; Quantum chemical calculations
1. Introduction
Recently we have studied the CsF–18-crown-6 medi-
ated trimethylsilyl group transfer from trimethylsily-
lacetylene to an aryl or hetaryl acetylene [19]. In the
present paper we are reporting the results of the investi-
gations of the cesium fluoride mediated metallation of
phenylacetylene and 2-methyl-5-pyridylacetylene with
Me3SiCꢀCMMe3 (M=C, Si, Ge, Sn) to compare the
influence of Group 14 elements on this substitution
reaction. Semiempirical AM1 calculations were used to
Aryl and hetaryl trimethylsilylacetylenes are usually
prepared by the reactions of aryl(hetaryl)acetylene with
BuLi–Me3SiCl or EtMgBr–Me3SiCl [1], ArCu with
trimethylsilyliodoacetylene [2], as well as by the Pd
catalyzed alkynylation of aryl triflates [3] or halides [4].
Phenyltrimethylsilylacetylene is also obtained from
phenylacetylene by the reaction with ethyl trimethylsily-
lacetate in the presence of Bu4NF [5] or trimethylsily-
lacetylene in the presence of KF-Al2O3 [6]. According
to the literature data germyl and stannyl substituted
acetylenes were obtained using the three general meth-
ods: interaction of germyl(stannyl)halogenides with
ethynylmagnesium bromides [1,7–11] or lithium deriva-
tives of the corresponding acetylenes [12–15], reaction
of triorganosubstituted germyl(stannyl) amines with
acetylenes RꢁCꢀCH [16]. The reactions of Group 14
element containing compounds catalyzed with nucle-
ophiles have also been studied extensively [17,18].
1
describe the mechanism of the reaction. MS and H-
NMR spectroscopy confirmed the structure of the final
products.
2. Results and discussion
Phenylacetylene (1) and 2-methyl-5-pyridylacetylene
(2) have been chosen as model substances for investi-
gating the cesium fluoride mediated metallation of aro-
matic and heteroaromatic acetylenes (Scheme 1).
The synthesis of Group 14 element substituted ary-
lacetylenes 3a–c and 4a–c was carried out in the pres-
ence of catalytic amount of CsF as the fluoride ion
source and 18-crown-6 in benzene (Table 1). The two
* Corresponding author. Tel.: +371-755-1822; fax: +371-755-
0338.
E-mail address: kira@osi.lv (E. Lukevics).
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(01)01109-3

70
E. Luke6ics et al. / Journal of Organometallic Chemistry 634 (2001) 69–73
The interaction of arylacetylenes with the 1-trime-
thylsilyl-2-trimethylgermylacetylene occurred smoothly
and gave the mixture of silylated and germylated ary-
lacetylenes 3a,b and 4a,b, respectively. The 30–40%
excess of trimethylgermyl derivatives (3b, 4b) was ob-
served in both cases.
Transmetallation of stannyl substituted acetylenes
has been described in Ref. [21]. We have found that
1-trimethylsilyl-2-trimethylstannylacetylene in the CsF–
18-crown-6–benzene system at room temperature forms
symmetrical bis(trimethylsilyl)- and bis(trimethylstan-
nyl)acetylenes. The latter is inactive under the described
phase transfer catalysis conditions (GCMS data).
Therefore, the aryl trimethylsilylacetylenes 3a and 4a
were formed in 30–60% excess relative to the corre-
sponding trimethylstannyl derivatives 3c and 4c (Table
1).
Scheme 1.
fold excess of cesium fluoride to 18-crown-6 was used
because of possible side reactions, e.g. Me3SiF forma-
tion or HF evolution. The Molar ratio of ArCꢀCH–
Me3SiCꢀCMMe3–CsF–18-crown-6,
appeared to be optimal [19].
1:1:0.2:0.1
It has been found that the interaction of 1-tert-butyl-
2-trimethylsilylacetylene with arylacetylenes 1 and 2
affords silylated acetylenes 3a (34%) and 4a (27%). The
conversion of the phenylacetylene 1 did not exceed
34%, and was not influenced by the increase in the base
(CsF) amount up to 1.5 equivalents and a rise in
temperature from 25 to 50 °C.
On the contrary, bis(trimethylsilyl)acetylene reacted
with arylacetylenes giving the desired products
ArCꢀCSiMe3 3a and 4a at room temperature with
excellent yields (90–91%). We tried to decrease the
ratio of bis(trimethylsilyl)acetylene to 1 or 2 from 1:1 to
0.5:1.0 taking into account the presence of two
trimethylsilyl groups in Me3SiCꢀCSiMe3, but the pro-
cess stopped at the 50% yield of silylated product and
trimethylsilylacetylene. The silylation of terminal
acetylenes with Me3SiCꢀCH needs activation and oc-
curs only at 50 °C [19].
The 1H-NMR and mass spectroscopic data of
acetylenes 3a–c, 4a–c are shown in Tables 2 and 3.
Semiempirical methods like AM1 provide a quite
effective compromise between the accuracy of the re-
sults and the expense of computer time required. Calcu-
lations performed with AM1 reflect the experimental
results as effectively as ab initio calculations using a
small basis set [22]. Reactions of bis(trimethylsi-
lyl)acetylene and 1-trimethylsilyl-2-trimethylgermy-
lacetylene with phenylacetylene have been chosen as
models for the calculations due to their smooth
running.
The Cs+–crown ether complex formation leads to
‘organic masking’ of the metal and F−-ion solubilisa-
tion. Such systems have been described as involving the
‘reaction of naked anions’ [23].
At the beginning we supposed that the first step of
the fluoride ion mediated reactions of silicon containing
Table 1
Reactions of disubstituted acetylenes Me3SiCꢀCMMe3 (M=C, Si, Ge, Sn) with terminal acetylenes (molar ratio ArCꢀCH–Me3SiCꢀCMMe3–CsF–
18-crown-6, 1:1:0.2:0.1)
Starting acetylene
Ar
M
Reaction time (h)
Reaction product a
Yield (%)
1
1
1
Ph
Ph
Ph
C
Si
Ge
5
6
5
3a b
3a
3a
34
91
31
45
38
29
27
90
36
46
41
26
3b
3a
1
Ph
Sn
5
3c c
4a b
4a
2
2
2
2-Me-5-pyridyl
2-Me-5-pyridyl
2-Me-5-pyridyl
C
Si
Ge
7
6
5
4a
4b
2
2-Me-5-pyridyl
Sn
5
4a
4c c
a Reaction products were described: 3a, 3b [13]; 3c [20]; 4a [19].
b The conversion of the phenylacetylene 1 did not exceed 34%.
c It was detected that 16–20% of the 1-trimethylsilyl-2-trimethylstannylacetylene was converted to the symmetrical bis(trimethylstan-
nyl)acetylene.

E. Luke6ics et al. / Journal of Organometallic Chemistry 634 (2001) 69–73
71
Table 2
1H-NMR spectroscopic data of acetylenes 3a–c, 4a–c
ArCꢀCMMe3
M
l (ppm), J (Hz)
MMe3 (s)
2-Me (s)
Ring protons (m)
3a
3b
3c
Si
Ge
Sn
0.25
0.43
0.36
–
–
–
7.28 and 7.47
7.28 and 7.46
7.28 and 7.47
(J117,119Sn–H=60, 53 and 59.27)
4a
4b
4c
Si
Ge
Sn
0.25
0.48
0.36
2.54
2.58
2.57
7.05 (H-3), 7.59 (H-4), 8.61 (H-6)
7.10 (H-3), 7.66 (H-4), 8.62(H-6)
7.09 (H-3), 7.63 (H-4), 8.57 (H-6)
(J117,119Sn–H=60, 53 and 59.27)
compounds could be the F− attack at the silicon with
the formation of pentacoordinated species. The calcula-
tions showed that two mechanisms of fluoride ion
interaction with bis(trimethylsilyl)acetylene are possi-
ble. The pentacoordinated silicon formation could take
place only if the starting position F···Si-CꢀC was
strongly linear. All others fluoride ion steric positions
led to the elimination of the proton from the methyl
group and HF formation. Due to the low probability of
the sterical position necessary for the pentacoordinated
silicon formation, we had to find another possible
reaction route.
Fig. 3c. The heat of the reaction is −159.6 kcal mol−1
.
All the steps of the reaction studied occur without
activation energy.
The quantum-chemical calculations and computer
visualisation of the process allow proposing the follow-
ing reaction mechanism (Scheme 2).
The interaction of phenylacetylene
1 with 1-
trimethylsilyl-2-trimethylgermylacetylene was calculated
Table 3
Mass spectra data for acetylenes 3a–c, 4a–c
Deprotonation of phenylacetylene 1 with fluoride ion
turned out to be more advantageous under PTC condi-
tions. The reaction heat is −73.3 kcal mol−1 (Fig. 1a
and b). Thus, it can be concluded that the first step of
the process is deprotonation of the starting ary-
lacetylene. In favour of this mechanism is the fact that
alkynes with less acidic terminal proton (propargyl
ethers, for example) did not react under the described
conditions.
The interaction of the arylacetylene carbanion with
bis(trimethylsilyl)acetylene is the next reaction step
(Fig. 2). The calculation starting state is shown in Fig.
2a.
ArCꢀCMMe3 m/z (I, %)
3a
3b
174(M+, 17), 160(16), 159(M+−Me, 100),
143(4), 131(4), 129(9), 105(8), 53(6), 43(10)
220(M+, 11), 207(21), 206(12), 205(M+−Me,
100), 204(33), 203(76), 201(56), 175(31), 174(10),
173(23), 171(17), 115(13), 89(11)
3c
266(M+, 8), 255(18), 253(14), 252(11),
251(M+−Me, 100), 250(36), 249(76), 248(31),
247(44), 225(11), 221(62), 220(23), 219(46),
218(19), 217(29), 145(17), 143(12), 120(23),
118(17), 116(12), 115(12)
4a
4b
189(M+, 18), 175(16), 174(M+−Me, 100),
144(5), 77(5), 53(5), 43(7)
235(M+, 12), 233(9), 222(21), 221(11),
220(M+−Me, 100), 219(32), 218(76), 216(58),
190(21), 188(17), 186(13), 163(5), 123(6), 87(6)
281(M+, 8), 270(18), 268(14), 267(11),
266(M+−Me, 100), 265(37), 264(78), 263(30),
262(45), 236(51), 235(19), 234(40), 233(16),
232(24), 169(11), 135(11), 133(10), 120(32),
119(12), 118(30), 117(15), 116(18)
During the anion approach the tetrahedral configura-
tion of silicon was changed and turned to the bipyrami-
dal in the equilibrium state (Fig. 2b). The heat of the
4c
intermediate formation is −27.5 kcal mol−1
.
It is natural to assume that the negatively charged
intermediate interacts with the counterion. The lithium
ion was used as a counterion, because the cesium ion
parameters are absent in the MOPAC6 program. The
biggest negative charge of the intermediate is localised
at the b-carbon atom of the CꢀC bond. That is why we
suppose that Li+ would attack the SiꢁC bond (Fig. 3).
The starting state of this process is shown in Fig. 3a.
The change in the linearity caused by the interaction
with the lithium ion is shown in Fig. 3b. The CꢁSi bond
cleavage and PhCꢀCSiMe3 (3a) formation are shown in
Fig. 1. (a) F− attack at the CꢁH bond; (b) proton elimination.

72
E. Luke6ics et al. / Journal of Organometallic Chemistry 634 (2001) 69–73
Fig. 2. (a) Starting state; (b) equilibrium state.
Fig. 3. (a) Starting state; (b) Li+ attack at the SiꢁC bond; (c) SiꢁC bond cleavage.
Scheme 2.
Fig. 4. (a) Starting state; (b) PhCꢀCꢁGe containing intermediate.
in a similar way. The first step of the reaction was the
deprotonation of the starting arylacetylene. There are
two competing reaction centres for aryl carbanion at-
tack in the 1-trimethylsilyl-2-trimethylgermylacetylene
molecule. The PhCꢀCꢁSi bond formation had the heat
of −25.5 kcal mol−1, but the formation of PhCꢀCꢁGe
bond was characterised by the reaction heat of −43.6
kcal mol−1 (Fig. 4). The PhCꢀCꢁGe containing inter-
mediate destruction leads to the 1-trimethylgermyl-2-
lacetylene at room temperature [19]. However, it was
shown that the reaction of 1-trimethylsilyl-2-
trimethylgermylacetylene with phenylacetylene gave not
only 1-trimethylgermyl-2-phenylacetylene 3b but also
1-trimethylsilyl-2-phenylacetylene 3a. Thus, it can be
concluded that both the competing reactions were ob-
served. As the heat of PhCꢀCꢁSi bond formation
(−25.5 kcal mol−1) is lower than that of PhCꢀCꢁGe
(−43.6 kcal mol−1) the germylacetylene 3b yield (45%)
is higher than the yield of the corresponding silyl
analogue 3a (31%).
phenylacetylene
3b
and
trimethylsilylacetylene
formation. The latter does not react with pheny-

E. Luke6ics et al. / Journal of Organometallic Chemistry 634 (2001) 69–73
73
3. Materials and methods
Supporting Information is a6ailable from the authors
on request. The multi-structure XYZ files describing the
geometry changes of the molecular systems during the
optimisation process are available from the author
(M.F.) at misha@osi.lv.
3.1. Instrumental
1H-NMR spectra were recorded on a Varian 200
Mercury spectrometer (200 MHz) using CDCl3 as the
solvent. Mass spectra were registered on a GCMS HP
6890 (70 eV) apparatus. GC analysis was performed on
a Chrom-5 instrument equipped with a flame-ionisation
detector using a glass column packed with 5% OV-101/
Chromosorb W-HP (80–100 mesh, 1.2 m×3 mm).
Bis(trimethylsilyl)acetylene (Aldrich) was used without
purification. 1-tert-Butyl-2-trimethylsilylacetylene, 1-
trimethylsilyl-2-trimethylgermylacetylene and 1-tri-
methylsilyl-2-trimethylstannylacetylene were prepared
and characterised as outlined in Refs. [21], [24–26]. CsF
was calcinated at ca. 200 °C during 1 h. Benzene was
distilled over CaH2 and kept over molecular sieves of 4
References
[1] C. Eaborn, D.R.M. Walton, J. Organomet. Chem. 2 (1964) 95.
[2] R. Oliver, D.R.M. Walton, Tetrahedron Lett. (1972) 5209.
[3] T. Okita, M. Isobe, Synlett (1994) 589.
[4] A. Carpita, R. Rossi, C.A. Veracini, Tetrahedron 41 (1985) 1919.
[5] E. Nakamura, T. Murofushi, M. Shimizu, U. Kuwajima, J. Am.
Chem. Soc. 98 (1976) 2346.
[6] T. Baba, A. Kato, H. Yuasa, F. Toriyama, H. Handa, Y. Ono,
Catal. Today 44 (1998) 271.
[7] S.D. Ibekwe, M.J. Newlands, J. Chem. Soc. (1965) 4608.
[8] J.C. Masson, M. le Quan, P. Cadiot, Bull. Soc. Chim. Fr. (1968)
1085.
[9] P. Mazerolles, Bull. Soc. Chim. Fr. (1960) 856.
[10] P. Mazerolles, Compt. Rend. 251 (1960) 2041.
[11] W. Steingross, W. Zeill, J. Organomet. Chem. 6 (1966) 464.
[12] K.H. Birr, D. Kra¨ft, Z. Anorg. Allg. Chem. 311 (1961) 235.
[13] C. Eaborn, D.R.M. Walton, J. Organomet. Chem. 4 (1965) 217.
[14] B.C. Pant, H.F. Reiff, J. Organomet. Chem. 15 (1968) 65.
[15] D. Seyferth, M.H. Cohen, Inorg. Chem. 1 (1962) 913.
[16] J. Satge´, M. Massol, Compt. Rend. 259 (1964) 4733.
[17] C. Chuit, R.J.P. Corriu, C. Reye, J.C. Young, Chem. Rev. 93
(1993) 1371.
[18] G.G. Furin, O.A. Vyazankina, B.A. Gostevsky, N.S. Vyazankin,
Tetrahedron 44 (1988) 2675.
[19] E. Abele, K. Rubina, R. Abele, J. Popelis, I. Mazeika, E.
Lukevics, J. Organomet. Chem. 586 (1999) 184.
[20] H. Hortmann, P. Schrader, W. Reiss, Naturwissenschaften 50
(1963) 373.
[21] D. Seyferth, D.L. White, J. Organomet. Chem. 32 (1971) 317.
[22] H.-D. Ho¨ltje, G. Folkers, Molecular Modeling. Basic Principles
and Applications, vol. 5, VCH, New York, USA, 1997, p. 21.
[23] E.V. Dehmlow, S.S. Dehmlow, Phase Transfer Catalysis, VCH
Verlagsgesellschaft mbH, D-6940 Weinheim, Federal Republic of
Germany, 1993, p. 22.
[24] E. Liepin¸sˇ, I. Birgele, E. Lukevics, E.T. Bogoradovskii, V.S.
Zavgorodnii, J. Organomet. Chem. 393 (1990) 11.
[25] J.J. Eisch, M.W. Foxon, J. Org. Chem. 36 (1971) 3520.
[26] W. Findeiss, W. Davidsohn, M.C. Henry, J. Organomet. Chem.
9 (1967) 435.
[27] M.J.S. Dewar, E.G. Zoebisch, E.F. Healy, J.J.P. Stewart, J. Am.
Chem. Soc. 107 (1985) 3902.
[28] J.J.P. Stewart, MOPAC, version 6.0, Quantum Chemical Program
Exchange (QCPE), Program Number 455, Bloomington, IN,
1984.
,
A.
3.2. General procedure for the synthesis of acetylenes
3a–c, 4a–c
Freshly calcinated CsF (0.03 g, 0.2 mmol) was added
to a mixture of 1 or 2 (1 mmol) and 18-crown-6 (0.026
g, 0.1 mmol) in 1.5 ml of dry benzene under argon
atmosphere. After 5 min of stirring the corresponding
acetylene (1 mmol) was added. The reaction was carried
out for 5–7 h at room temperature (r.t.) to achieve the
substrate disappearance (GC control, 120–250 °C).
The reaction mixture was filtered over a thin layer of
silica gel or Al2O3 and evaporated under reduced
pressure.
3.3. Theoretical calculations
All calculations were carried out using the semiem-
pirical AM1 [27] method as implemented in MOPAC
6
[28]. The equilibrium geometries were obtained with
complete optimisation at the PRECISE level. The fre-
quencies analysis has shown that all optimum struc-
tures present the minimum points on the potential
energy surface. A keyword PARASOK was applied in
the calculations of the lithium cation containing struc-
tures [29]. To obtain the data on the change in geome-
try during the optimisation process, calculations were
performed using a keyword FLEPO. Post-processing
animation was carried out with XMOL [30] and JMOL
[31] programs.
[29] J.J.P. Stewart, MOPAC, version 6.0, Manual, Program Number
455, Bloomington, IN, 1984.
[30] XMOL, version 1.3.1, Minnesota Supercomputer Center Inc.,
Minneapolis MN, 1993.
[31] JMOL: http://www.chem.columbia.edu/ꢁgezelter.