Introduction of Bulky Substituents at the Bridgehead Position of a 9-Silatriptycene: Pentacoordinate Hydridoorganylsilicates as Intermediates

Article abstract of DOI:10.1021/om990976o

The reaction of 10-phenyl-10-germa-9-silatriptycene (1b) with phenyllithium in THF/HMPA produced 9,10-diphenyl-10-germa-9-silatriptycene (1a), which had hitherto not been accessible. The reaction proceeds by attack of the phenyl anion on 1b via the intermediate silicate anion 7, which was identified by ²⁹Si NMR spectroscopy. The structure of 1a was confirmed by X-ray crystallography. As a model reaction, the attack of phenyllithium on triphenylsilane (8) and diphenylsilane (14) was investigated by ²⁹Si NMR spectroscopy, revealing the formation of a number of novel hydridosilicates analogous to 7.

Full text of DOI:10.1021/om990976o

Organometallics 2000, 19, 1319-1324
1319
In tr od u ction of Bu lk y Su bstitu en ts a t th e Br id geh ea d
P osition of a 9-Sila tr ip tycen e: P en ta coor d in a te
Hyd r id oor ga n ylsilica tes a s In ter m ed ia tes
Nicolette Rot,† Tom Nijbacker,† Rutger Kroon,† Frans J . J . de Kanter,†
Friedrich Bickelhaupt,*,† Martin Lutz,‡ and Anthony L. Spek‡
Scheikundig Laboratorium, Vrije Universiteit, De Boelelaan 1083,
NL-1081 HV Amsterdam, The Netherlands, and the Bijvoet Center for Biomolecular Research,
Crystal and Structural Chemistry, Utrecht University, Padualaan 8,
NL-3584 CH Utrecht, The Netherlands
Received December 8, 1999
The reaction of 10-phenyl-10-germa-9-silatriptycene (1b) with phenyllithium in THF/HMPA
produced 9,10-diphenyl-10-germa-9-silatriptycene (1a ), which had hitherto not been acces-
sible. The reaction proceeds by attack of the phenyl anion on 1b via the intermediate silicate
anion 7, which was identified by 29Si NMR spectroscopy. The structure of 1a was confirmed
by X-ray crystallography. As a model reaction, the attack of phenyllithium on triphenylsilane
(8) and diphenylsilane (14) was investigated by 29Si NMR spectroscopy, revealing the
formation of a number of novel hydridosilicates analogous to 7.
In tr od u ction
hindrance; otherwise, 6 reacts intermolecularly to form
polymeric products.
We recently reported a novel one-pot synthesis of
9,10-dimetallatriptycenes of group 14 (1) by reaction of
(o-phenylene)magnesium (2) with organometal triha-
lides RMCl3 (3) (Scheme 1).1 Evidence was obtained that
the process occurred in two stages: reaction of 3 molar
equiv of C6H4Mg (2, which is the tetramer (C6H4Mg)4)2
with 1 molar equiv of 3 furnished selectively the
intermediate tri-Grignard species 4; in the second stage,
4 reacted with a second equivalent of 3 to give trip-
tycenes which are symmetrical (1; RM ) R′M′), while
with a different trihalide R′M′Cl3 (3′), unsymmetrical
triptycenes (1, RM * R′M′) were obtained.
Surprisingly, there was strong dependence of the yield
of 1 on the size of both the metal M and the substituent
R; reasonable yields were obtained only if R was small
and/or M was large.1 In all cases, the intermediate 4
had been formed nearly quantitatively, as demonstrated
by deuteriolysis after the first stage, which furnished 5
in high yield. Therefore, it seemed reasonable to assume
that the combination of a large R and a small M leads
to derailment in the second stage: reaction of 4 with
the first chlorine of the second equivalent of 3 obviously
will give rise to the formation of an intermediate such
as 6, which undergoes the desired intramolecular double
ring closure leading to 1 only in the absence of steric
In view of the fact that a number of triptycenes are
known in the parent series with the smaller carbon at
the bridging positions carrying bulky substituents,3 it
was unlikely that analogous derivatives of 1 such as
9,10-diphenyl-10-germa-9-silatriptycene (1a ) were not
capable of existence; rather, one had to devise a different
approach for their synthesis.
Resu lts a n d Discu ssion
Syn th esis a n d Cr ysta l Str u ctu r e of 1a . An ad-
ditional reason for choosing 1a as the synthetic goal was
the availability of 10-phenyl-10-germa-9-silatriptycene
(1b)1b as a precursor; its Si-H functionality was ex-
pected to serve as a starting point for transformation
to the desired Si-Ph group. Indeed, the reaction of 1b
with an excess of phenyllithium in diethyl ether at room
temperature did furnish 1a ; after 1 min, the reaction
was complete, and on workup 1a was isolated in 70%
yield (Scheme 2).
The structure of 1a was confirmed by an X-ray crystal
structure determination (R ) 0.0685; Figure 1). Unfor-
tunately, due to disorder, a differentiation between the
Ge and Si atoms in the molecule could not be made; only
average bond lengths and angles were obtained (see
legend to Figure 1). These structural parameters gener-
ally agree with those of related 9,10-diheterotriptycenes,
in particular with those of 1b.1b It should, however, be
pointed out that an unusual and rather unfavorable
eclipsing of one phenyl substituent (Figure 1: C(23)-
† Vrije Universiteit.
‡ Utrecht University.
(1) (a) Dam, M. A.; Akkerman, O. S.; de Kanter, F. J . J .; Bickelhaupt,
F.; Veldman, N.; Spek, A. L. Chem. Eur. J . 1996, 2, 1139. (b) Dam, M.
A.; de Kanter, F. J . J .; Bickelhaupt, F.; Smeets, W. J . J .; Spek, A. L.;
Fornies-Camer, J .; Cardin, C. J . Organomet. Chem. 1998, 550, 347.
(c) Dam, M. A.; Hoogervorst, W. J .; de Kanter, F. J . J .; Bickelhaupt,
F.; Spek, A. L. Organometallics 1998, 17, 1762. (d) Dam, M. A. 1,2-
Dimetalated Benzene Derivatives of Group 13, 14 and 15; Thesis, Vrije
Universiteit, Amsterdam, 1997.
(2) (a) Tinga, M. A. G. M.; Akkerman, O. S.; Bickelhaupt, F.; Horn,
E.; Spek, A. L. J . Am. Chem. Soc. 1991, 113, 3604. (b) Tinga, M. A. G.
M.; Schat, G.; Akkerman, O. S.; Bickelhaupt, F.; Horn, E.; Kooijman,
H.; Smeets, W. J . J .; Spek, A. L. J . Am. Chem. Soc. 1993, 115, 2808.
(3) (a) Koukotas, C.; Mehlman, S. P.; Schwartz, L. H. J . Org. Chem.
1966, 31, 1970. (b) Ardebili, M. H. P.; Dougherty, D. A.; Mislow, K.;
Schwartz, L. H.; White J . G. J . Am. Chem. Soc. 1978, 100, 7994. (c)
Oki, M.; Fujino, I.; Kawaguchi, D.; Chuda, K.; Moritaka, Y.; Yamamoto,
Y.; Tsuda, S.; Akinaga, T.; Aki, M.; Kojima, H.; Morita, N.; Sakurai,
M.; Toyota, S.; Tanaka, Y.; Tanuma. T.; Yamamoto, G. Bull. Chem.
Soc. J pn. 1997, 70, 457 and references cited therein.
10.1021/om990976o CCC: $19.00 © 2000 American Chemical Society
Publication on Web 03/03/2000

1320 Organometallics, Vol. 19, No. 7, 2000
Rot et al.
Sch em e 1
Sch em e 2
C(28)) with one of the o-phenylene units (C(11)-C(16))
occurs; we have at present no explanation for this
remarkable phenomenon.
Ate In ter m ed ia tes d u r in g th e Nu cleop h ilic Su b-
stitu tion of Si-H Bon d s. Besides achieving our
primary goal to obtain a 9,10-dimetallatriptycene with
two bulky substituents at the bridgehead positions,
interesting information was obtained about details of
the process of substitution at a silicon bridgehead and
on the occurrence of pentacoordinate silicate anions
carrying exclusively carbon and hydrogen as substitu-
ents at silicon.
The transformation of 1b to 1a involves nucleophilic
substitution of an Si-H function, and it occurs at a
bridgehead position; both types of reactions are well-
established.4 Fascinated by the ease of the transforma-
tion in the present case, we investigated the interaction
of 1b with phenyllithium (1.5 equiv) in a mixture of THF
and HMPA (20:1) at -80 °C by means of 29Si NMR
spectroscopy in search for intermediates. The signal of
1
1b (doublet at δ -40 ppm, J SiH ) 215 Hz) had
completely disappeared, but in addition to the singlet
of the final product 1a (δ -43 ppm), a new doublet at
δ -102 ppm (1J SiH ) 220 Hz) was observed (ratio about
1:1); it was assigned to the pentacoordinate silicate
intermediate 7 (Scheme 2). The signal of 7 was still
present after warming to room temperature, but it
disappeared on standing overnight. Afterward, depend-
ing on the conditions (see Experimental Section), several
signals were observed in the silane region (δ(29Si) -19,
-21, -23, -44 ppm); they might in part be due to
products formed by ring opening of the strained trip-
tycene skeleton, but on workup, well-defined products
could not be isolated.
F igu r e 1. Displacement ellipsoid plot of 1a drawn at the
50% probability level. Germanium and silicon atoms were
constrained to occupy the same positions in a 1:1 ratio with
the same displacement parameters. Selected bond lengths
(Å) and angles (deg): M(1)-R(1) ) 1.878(5), M(2)-R(2) )
1.886(5), M(1)-Cipso ) 1.918(5)-1.936(5), M(2)-Cipso
1.927(5)-1.934(5); Cipso-M(1)-Cipso ) 101.8(2)-103.4(2),
Cipso-M(2)-Cipso ) 101.7(2)-103.7(2), M(1)-Cipso-Cendo
)
)
Two aspects of 7 drew our attention. First, while
many examples of pentacoordinate silicates are known,
they generally carry one or more electronegative ligands
114.1(4)-117.8(4), M(2)-Cipso-Cendo ) 113.4(4)-117.1(4),
M(1)-Cipso-Cexo ) 122.5(4)-126.3(4), M(2)-Cipso-Cexo
123.3(4)-128.3(4).
)

Pentacoordinate Hydridoorganylsilicates
Organometallics, Vol. 19, No. 7, 2000 1321
Ta ble 1. Ra tio of P r od u cts F or m ed in th e
Rea ction of 8 a n d P h Li (1.5 equ iv) in THF (/HMP A)
(Deter m in ed by 29Si{1H} NMR Sp ectr oscop y)a
coupling were assigned to 9-12, as shown in Scheme
3. As in the case of 1b, the starting material 8 was
completely converted. Besides tetraphenylsilane (10),
the straightforward phenyl substitution product corre-
sponding to 1a , three ate adducts were identified by
their strongly shielded 29Si chemical shifts: 9 (doublet
ratio (%)
entry
T (°C)
solventb
8
9
10 11 12 [13]c
1d
2d
-80
A
A
A
A
A
A
A
A
B
B
A
A
5
6
6
5
4
20 28 36
13 32 39
11 35 35
1
at δ(29Si) -94 ppm, J SiH ) 140 Hz), the phenyl adduct
-80e
-80f
-93
of 8, 11 (δ(29Si) -103 ppm, no 1J SiH coupling), the known
phenyl adduct of 10,5a and, at first surprisingly, 12
3d
4d
8
6
36 38
34 32
5d
-65
1
(triplet at δ(29Si) -70 ppm, J SiH ) 130 Hz), the formal
6d
-50
10 18 24
4
9
4
lithium hydride adduct of 8.
7d
-6
19
34
38
52
23
22
6
8d
-50
At low temperatures, the composition of the reaction
mixture changed little with time (Table 1, entries 1-4).
While the sum of 10 and 11 was constant, 11 increased
at the expense of 10; this may be due to a slow reaction
of 10 with the excess of phenyllithium still present at
that time. When the temperature was raised to -6 °C,
the silicates 9 and 11 disappeared (entries 5-8); only
10 and 12 remained. At T g -50 °C, the unidentified
silane 13 (δ(29Si) -15 ppm) was formed in low yield
(Table 1, entries 6-8).
In a second experiment (Table 1, entries 9-12), the
reaction was initially performed in THF in the absence
of HMPA. Under these conditions, the conversion was
much slower, as indicated by a large amount of unre-
acted 8. Only the substitution product 10 and minor
amounts of 12 were observed, but no 9 and 11 (entries
9 and 10). This illustrates that the formation of silicate
anions depends on effective solvation of the lithium
counterion.4-7 Indeed, after addition of HMPA to this
reaction mixture at -80 °C, the reaction went to
completion (entry 11); under these conditions some 9
and 11 survived after warming to room temperature for
about 14 h (entry 12).
9g
-80
56
40
10g
11g
12g
-100
-80
8
3
3
30 35 29
25 27 39
3
6
room temp
a
Relative yields based on the integrals of the decoupled singlets
b
(corrected for the number of scans). Legend: A, in THF/HMPA
(7:1); B, in THF. c Unidentified product at δ -15 ppm. Experi-
d
ment 1. e After 24 min. f After 42 min. Experiment 2.
g
at silicon.4 Only very recently, the first silicate com-
plexes with five carbon-bonded ligands have been
observed by NMR spectroscopy;5 apart from gas-phase
species,6a,b only one silicate has so far been identified
in solution with silicon bonded to carbons and hydrogens
only.6c The second interesting aspect of 7 is the coupling
1
constant J SiH ) 220 Hz, which is rather large for a
hydridosilicate (vide infra).
Therefore, we studied triphenylsilane (8) as an acyclic
model for 1b. Depending on the temperature (Table 1,
entries 1-8), a solution of 8 and phenyllithium (1.5
equiv) in THF/HMPA (7:1) showed four 29Si signals,
which on the basis of their chemical shift and SiH
(4) (a) Corriu, R. J . P.; Gue´rin, C. J . Organomet. Chem. 1980, 198,
231. (b) Armitage, D. A. In Comprehensive Organometallic Chemistry;
Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.: Pergamon: Oxford,
U.K., 1982; Chapter 9.1, Vol. 2, p 1. (c) Holmes, R. R.; Day, R. O.;
Harland, J . J .; Sau, A. C.; Holmes, J . M. Organometallics 1984, 3, 341.
(d) Holmes, R. R.; Day, R. O.; Harland, J . J .; Holmes, J . M. Organo-
metallics 1984, 3, 347. (e) Holmes, R. R.; Day, R. O.; Chandrasekhar,
V.; Holmes, J . M. Inorg. Chem. 1985, 24, 2009. (f) Tandura, S. N.;
Voronkov, M. G.; Alekseev, N. V. In Topics in Current Chemistry;
Boschke, F. L., Ed.; Springer: Berlin, 1986; Vol. 131, p 99. (g)
Bassindale, A. R.; Taylor, P. G. In The Chemistry of Organic Silicon
Compounds; Patai, S., Rappoport, Z., Eds.; Wiley: Chichester, U.K.,
1989; Part 1, Chapter 13, p 839. (h) Corriu, R. J . P.; Young, J . C. In
The Chemistry of Organic Silicon Compounds; Patai, S., Rappoport,
Z., Eds.; Wiley: Chichester, U.K., 1989; Chapter 20, p 1241. (i) Holmes,
R. R. Chem. Rev. 1990, 90, 17. (j) Corriu, R. J . P.; Gue´rin, C.; Henner,
B. J . L.; Wang, Q. Organometallics 1991, 10, 2297. (k) Corriu, R. J . P.;
Gue´rin, C.; Henner, B. J . L.; Wang, Q. Organometallics 1991, 10, 3574.
(l) Chuit, C.; Corriu, R. J . P.; Reye, C.; Young, J . C. Chem. Rev. 1993,
93, 1371. (m) Holmes, R. R. Chem. Rev. 1996, 96, 927. (n) Deiters, J .
A.; Holmes, R. R. Organometallics 1996, 15, 3944. (o) Kost, D.;
Kalikhman, I. In The Chemistry of Organic Silicon Compounds;
Rappoport, Z., Apeloig, Y., Eds.; Wiley: Chichester, U.K., 1998; Vol.
2, Part 2, Chapter 23, p 1339. (p) Bassindale, A. R.; Taylor, P. G.;
Glynn, S. J .; Taylor, P. G In The Chemistry of Organic Silicon
Compounds; Rappoport, Z., Apeloig, Y., Eds.; Wiley: Chichester, U.K.,
1998; Vol. 2, Part 2, Chapter 9, p 495. (q) Takeuchi, Y.; Takayama, T.
In The Chemistry of Organic Silicon Compounds; Rappoport, Z.,
Apeloig, Y., Eds.; Wiley: Chichester, U.K., 1998; Vol. 2, Part 2, Chapter
6, p 267. (r) Hermanns, J .; Schmidt, B. J . Chem. Soc., Perkin Trans. 1
1999, 81. (s) Maggiarosa, N.; Tyrra, W.; Naumann, D.; Kirij, N. V.;
Yagupolski, Y. L. Angew. Chem. 1999, 111, 2392; Angew. Chem., Int.
Ed. 1999, 38, 2252.
The surprising thermal stability of the dihydrido-
triphenylsilicate 12 might be ascribed to its two small
apical hydride ligands, which reduce steric hindrance.
Its identity and unusual stability was confirmed by
independent synthesis from diphenylsilane (14) and
phenyllithium (Scheme 4), yielding 12 as the exclusive
product when phenyllithium was used in (slight) excess,
and only traces of 9, 10, or 11 were sometimes observed
by 29Si NMR spectroscopy. With an excess of 14,
interesting exchange reactions occurred. These phenom-
ena, and the mechanism of ligand exchange4k,l,o,p,8 which
leads to the products formed from 8 shown in Scheme
3, are currently under investigation.
Ster eoch em istr y of Hyd r id osilica tes. All new
pentacoordinate silicate anions are characterized by
strongly shielded 29Si signals which are typical for such
species;4q,5-8 this supports their general structural
assignment. However, the matter of their stereochem-
ical arrangement is more difficult for several reasons.
There is general agreement that, in the absence of
special factors, the trigonal bipyramid (TBP) is the
preferred geometry of pentacoordinated phosphoranes
and silicates; minor distortions of the TBP toward the
square pyramid (SP) along the Berry exchange coordi-
nate9 are frequently observed.4 In the absence of ex-
(5) (a) de Keijzer, A. H. J . F.; de Kanter, F. J . J .; Schakel, M.;
Schmitz, R. F.; Klumpp, G. W. Angew. Chem. 1996, 108, 1183; Angew.
Chem., Int. Ed. Engl. 1996, 35, 1127. (b) de Keijzer, A. H. J . F.; de
Kanter, F. J . J .; Schakel, M.; Osinga, V. P.; Klumpp, G. W. J .
Organomet. Chem. 1997, 548, 29.
(6) (a) Hajdasz, D. J .; Squires, R. R. J . Am. Chem. Soc. 1986, 108,
3139. (b) Hajdasz, D. J .; Ho, Y.; Squires, R. R. J . Am. Chem. Soc. 1994,
116, 10751. (c) Hong, J .-H.; Boudjouk, P. Organometallics 1995, 14,
574.
(7) (a) Damrauer, R.; Danahey, S. E. Organometallics 1986, 5, 1490.
(b) Damrauer, R.; O’Connell, B.; Danahey, S. E.; Simon, R. Organo-
metallics 1989, 8, 1167. (c) Schomburg, D.; Krebs, R. Inorg. Chem.
1984, 23, 1378.
(8) Becker, B.; Corriu, R. J . P.; Gue´rin, C.; Henner, B. J . L. J .
Organomet. Chem. 1989, 369, 147.

1322 Organometallics, Vol. 19, No. 7, 2000
Rot et al.
Sch em e 3
Sch em e 4
Ch a r t 1
perimental data, one can only speculate about the
structures of the silicates 9, 11, and 12, but very likely,
they are close to a TBP, as there are no distorting forces
imposed by unusual ligands or small rings. This as-
sumption is supported by the NMR data. Hydrogen is
usually more apicophilic than the phenyl group and is
therefore expected to occupy an apical position in a
hydridosilicate of the kind discussed here. Because the
apical bonds are longer and weaker, one expects the
apical 29Si-H coupling of a silicate to be smaller than
that of a related tetracoordinate silane, and indeed, it
decreases from the silanes 8 (1J SiH ) 198 Hz) and 14
(1J SiH ) 197 Hz) to the silicates 9 (1J SiH ) 140 Hz) and
12 (1J SiH ) 130 Hz).
It is against this background that the Si-H coupling
for 7 (1J SiH ) 220 Hz) is highly unusual, as it is even
slightly larger than that of the parent silane 1b
(1J SiH ) 215 Hz). This means that the s character in
the Si-H bond increases from 1b to 7, which, in a TBP
structure, would indicate that hydrogen is in an equato-
rial position in spite of the presence of four aryl ligands
which normally are less apicophilic! As far as we are
aware, there is only one report on two neutral com-
pounds, i.e., o-Me2NCH2C6H4SiHMePh and o-Me2-
NCH2C6H4SiH2Ph, with pentacoordinate silicon arising
from intramolecular coordination of a tertiary amine to
a tetracoordinate silicon center, both of which have the
hydrogens in equatorial positions and the phenyl group
in an axial position.10
Of course, in the case of 7, the situation is more
complicated because the three phenylene groups of the
triptycene skeleton are restricted in adopting ideal
positions both in the TBP and in the SP. For reasons of
angle strain, the triptycene ipso carbons in a TBP can
only be arranged in an approximately apical-diequato-
rial fashion, while in an SP, an apical-dibasal arrange-
ment would be preferred. Presumably, the actual con-
figuration is somewhere between these extremes, but
as pointed out above, the increase of s character of the
Si-H bond from 1b to 7 is more in line with a quasi-
equatorial position of the hydrogen in a (slightly dis-
torted) TBP. This configuration is the one initially
formed when the phenyl anion approaches the silicon
of 1b from an apical direction.4p,10 Apparently 7 is stuck
in this unusual configuration instead of relaxing to the
normally more stable stereoisomer 7′ with an apical
hydrogen (Chart 1). Possibly, steric effects7b,11 are
responsible; a TBP model of 7 showed that the free
phenyl group in the (usually preferred) equatorial
position as in 7′ would experience considerable steric
hindrance from the three ortho hydrogens of the trip-
tycene unit. For the formation of 1a , however, the
hydrogen has to be transferred from the equatorial to
the apical position by a pseudorotation from 7 to 7′
before it can act as a leaving group.4p
The proposed structural assignment of 7 is supported
1
by J SiH values of other hydridosilicates. Couplings for
equatorial Si-H bonds have been assigned in [HSi(OR)4]-
(J eq ) 209-296 Hz),4j [H2Si(OR)3]2- (J eq ) 222 Hz,
J ap ) 194 Hz),4k and 15 (Chart 1; J eq ) 192.5 Hz, J ap
)
(9) (a) Holmes, R. R.; Deiters, J . A. J . Am. Chem. Soc. 1977, 99,
3318. (b) Holmes, R. R. In Progress in Inorganic Chemistry; Lippard,
S. J ., Ed.; Wiley: New York, 1984; Vol. 32, p 119.
179.6 Hz).6c In these compounds, the expected trend of
(10) Brelie`re, C.; Carre´, F. Corriu, R. J . P.; Poirier, M.; Royo, G.
Organometallics 1986, 5, 388.
(11) Harland, J . J .; Payne, J . S.; Day, R. O.; Holmes, R. R. Inorg.
Chem. 1987, 26, 760.

Pentacoordinate Hydridoorganylsilicates
Organometallics, Vol. 19, No. 7, 2000 1323
Syn th esis of 9,10-Dip h en yl-9,10-d ih yd r o-10-ger m a -9-
sila -9,10[1′,2′]ben zen oa n th r a cen e (1a ). To a solution of 1b
(0.0558 g, 0.137 mmol) in Et2O (25 mL) was added phenyl-
lithium in cyclohexane/Et2O (0.5 mL, 1.8 M, 0.9 mmol). After
1 min, a saturated aqueous NH4Cl solution was added to the
slightly yellow reaction mixture. After extraction with CHCl3,
the combined organic layers were filtered and dried by passing
through a column of MgSO4. The solvent and biphenyl (present
in the PhLi solution) were evaporated in vacuo at 110 °C. The
resulting yellow solid was washed with a small amount of
Et2O, yielding 1a as a white powder (0.0463 g, 0.0958 mmol,
70% pure according to 1H NMR spectroscopy). Crystals for an
X-ray crystal structure determination were obtained from C6D6
by slow evaporation.
1a : mp 337 °C dec.1H NMR (400.1 MHz, CDCl3): δ 8.35
(m, J ) 7.5, 1.4, 1.3, 0.6 Hz, 2H; H(2′,6′)), 8.19 (m, J ) 7.4,
1.4, 1.4, 0.6 Hz, 2H; H(2′′,6′′)), 7.81 (m, J ) 7.0, 1.7, 0.8 Hz,
3H; H(1,8,13)), 7.79 (m, J ) 7.9, 0.8, 0.7 Hz, 3H; H(4,5,16)),
7.72 (m, J ) 7.5, 7.4, 1.3, 0.6 Hz, 2H; H(3′,5′)), 7.72 (m, J )
7.4, 1.3 Hz, 1H; H(4′)), 7.71 (m, J ) 7.6, 7.4, 1.3, 0.6 Hz, 2H;
H(3′′,5′′)), 7.68 (m, J ) 7.6, 1.4 Hz, 1H; H(4′′)), 7.23 (m, J )
7.6, 7.0, 0.8 Hz, 3H; H(2,7,14)), 7.23 (m, J ) 7.9, 7.6, 1.7 Hz,
3H; H(3,6,15)). 1H NMR (400.1 MHz, C6D6): δ 8.24 (m, J )
7.4, 1.8 Hz, 1.4, 0.7 Hz, 2H; H(2′,6′)), 8.04 (m, J ) 7.4, 1.4,
1.3, 0.6 Hz, 2H; H(2′′,6′′)), 7.88 (m, J ) 7.1, 1.2, 0.8, 3H; H(1,8,-
13)), 7.81 (m, J ) 7.6, 1.0, 0.8 Hz, 3H; H(4,5,16)), 7.40 (m, J )
7.5, 1.4 Hz, 1H; H(4′)), 7.38 (m, J ) 7.5, 7.4, 1.6, 0.7 Hz, 2H;
H(3′,5′)), 7.37 (m, J ) 7.6, 7.4, 1.4, 0.6 Hz, 2H; H(3“,5”)), 7.36
(m, J ) 7.6, 1.3 Hz, 1H; H(4′′)), 7.06 (m, J ) 7.5, 7.1, 1.0 Hz,
3H; H(2,7,14)), 7.05 (m, J ) 7.6, 7.5, 1.2 Hz, 3H; H(3,6,15)).
13C{1H} NMR (100.6 MHz, CDCl3): δ 148.9 (C(4a,10a,11)),
142.6 (C(8a,9a,12)), 137.2 (C(2′,6′)), 136.6 (C(2“,6”)), 133.9
(C(1,8,13)), 132.2 (C(4,5,16)), 131.0 (C(3′)), 130.5 (C(4′′)), 129.9
(C(1′′)), 129.4 (C(3′′,5′′)), 129.0 (C(3′,5′)), 128.2 (C(1′)), 127.6
(C(2,7,14)), 127.4 (C(3,6,15)). 29Si NMR (79.5 MHz, CDCl3):
δ - 43 (1J SiC ) 68 Hz). 13C{1H} NMR (100.6 MHz, C6D6): δ
149.1 (C(4a,10a,11)), 142.9 (C(8a,9a,12)), 137.3 (C(2′,6′)), 136.7
(C(2“,6”)), 134.1 (C(1,8,13)), 132.4 (C(4,5,16)), 130.8 (C(3′)),
130.2 (C(4′′)), 130.1 (C(1′′)), 129.3 (C(3′′,5′′)), 128.9 (C(3′,5′)),
128.5 (C(1′)), 127.6 (C(2,7,14)), 127.4 (C(3,6,15)). HRMS (EI):
calcd for [C30H2228Si74Ge]+, 484.0703; found, 484.0705.
F igu r e 2. Numbering system of 1a .
J eq > J ap is realized. The same order has been reported
1
for J PF couplings in pentacoordinate fluorophospho-
ranes.12 However, the opposite order J ap > J ex has been
1
assigned for J SiF in fluorosilicates7,13 and, by analogy,
1
for J SiC in a pentaorganyl silicate.5b The reason for this
discrepancy is not clear at the moment.
Con clu sion s
While 9,10-dimetallatriptycenes with two bulky aryl
substituents at the metals have not been accessible by
previous approaches, the first representative of this
group, 9,10-diphenyl-10-germa-9-silatriptycene (1a ), has
been prepared by reaction of its 9-H substituted analo-
goue 1b with phenyllithium. The role of hydridosilicates
as intermediates in the substitution of 1b and of model
silanes PhnH4-nSi with phenyllithium has been dem-
onstrated by 29Si NMR spectroscopy.
Exp er im en ta l Section
Gen er a l Com m en ts. All reactions were carried out under
an argon atmosphere. Glassware and syringes were dried in
an oven at 120 °C for at least 24 h or flame-dried (glassware)
before use. The solvents Et2O and THF were predried on
NaOH and KOH, respectively, distilled from LiAlH4, and kept
on sodium wire under nitrogen. THF-d8 and HMPA were dried
on molecular sieves and kept under argon. 1H NMR spectra
were measured at 25 °C on a Bruker AC 200 spectrometer and
2H NMR, 13C NMR, and 29Si NMR spectra on a Bruker MSL
400 spectrometer. Melting points (uncorrected) were measured
in a sealed capillary. The chemical shifts and J HH coupling
constants were determined by simulation using the gNMR
program14 and from CH correlation spectra. The HRMS
measurement of 1a was performed on a Finnigan MAT 90
mass spectrometer. GC/MS measurements were performed on
a HP 5890 II GC/5971 MS combination (70 eV, Chrompack
BP 1 (QSGE) 50 m × 0.25 mm column). Preparative layer
chromatography (PLC) was performed on a Merck plate (silica
gel 60 F254 20 × 20 cm, layer thickness 2 mm), using an
n-hexane/Et2O mixture (9:1) as eluent unless stated otherwise.
The concentration of phenyllithium was determined by HCl/
NaOH titration of the “total” base formed after hydrolysis
minus the amount of “rest” base determined after reaction of
an aliquot of phenyllithium with 1,2-dibromoethane (dried on
molecular sieves). Triptycene 1b was prepared according to
reported procedures.1b,c The numbering system of 1a is shown
in Figure 2.
NMR Sp ectr oscop y of Solu tion s of 1b a n d P h en yl-
lith iu m in THF (/HMP A). Under argon, THF (2.7 mL), THF-
d8 (0.4 mL), and HMPA (0.15 mL) were added to 1b (0.0492
g, 0.121 mmol) in an NMR tube of 10 mm diameter. Subse-
quently, the tube was capped with a septum and cooled (-78
°C). After addition of a solution of phenyllithium in cyclohex-
ane/Et2O (0.10 mL, 1.8 M, 0.18 mmol), the dark brown reaction
mixture was homogenized on a vortex. 29Si NMR (-80 °C): -19
(?), -43 (1a ), -102 (7, 1J SiH ) 220 Hz) in varying ratios due to
solubility problems. 29Si NMR spectroscopy at room temper-
ature showed that 1a and 7 were still present. Upon addition
of MeOD (0.25 mL) the brown reaction mixture turned yellow.
After the addition of water, the mixture was extracted with
CHCl3 and the combined organic layers were evaporated to
dryness in vacuo; a yellow, HMPA-containing oil remained.
On addition of water, a white solid precipitated, and the
supernatant was removed carefully. The precipitate was
washed several times with water and twice with a small
amount of EtOH and dried in vacuo. 1H and 13C NMR
spectroscopy showed the white product to be pure 1a (42.8 mg,
0.089 mmol, 71% yield).
In a second experiment, the reaction was initially performed
in THF/THF-d8 with about 1 equiv of phenyllithium in
cyclohexane/Et2O (80 µL, 1.8 M, 0.14 mmol). 29Si NMR (-50
°C, -80 °C, -100 °C): -40 (1b). Subsequently, HMPA (0.5 mL)
and about 1 equiv of phenyllithium in cyclohexane/Et2O (40
mL, 1.8 M, 0.07 mmol) were added. 29Si NMR (-40 °C): -43
(1a ), -102 (7), ratio about 1:1. After standing for 14 h at room
temperature, three signals at δ -21, -23, -44 (1a ), ratio
(12) (a) Muetterties, E. L.; Mahler, W.; Schmutzler, R. Inorg. Chem.
1963, 2, 613. (b) Muetterties, E. L.; Mahler, W.; Packer, K. J .;
Schmutzler, R. Inorg. Chem. 1964, 3, 1298.
(13) Klanberg, F.; Muetterties, E. L. Inorg. Chem. 1968, 7, 155.
(14) Budzelaar, P. H. gNMR.; Ivorysoft, Amsterdam, 1992.

1324 Organometallics, Vol. 19, No. 7, 2000
Rot et al.
1:1:1, were observed, but no ate complex. Workup of the
reaction mixture as described above, gave a brown sticky
powder: 1H NMR (200.1 MHz, CDCl3): δ 8.0-6.5 (bm).
Attempted separation of the products by PLC was unsuccess-
ful.
Cr yst a l St r u ct u r e Det er m in a t ion of 1a : C30H22GeSi,
Mr ) 483.16, colorless needle, 0.45 × 0.25 × 0.20 mm3, triclinic,
P1h, a ) 8.869(2) Å, b ) 11.176(6) Å, c ) 12.945(6) Å, R ) 99.42-
(3)°, â ) 103.81(2)°, γ ) 110.21(2)°, V ) 1125.8(8) Å3, Z ) 2,
D ) 1.425 g cm-3, 289 parameters, 5442 measured reflections,
5108 unique reflections (Rint ) 0.0584). R (I > 2σ(I)): R1 )
0.0685, wR2 ) 0.1587. R (all data): R1 ) 0.1132, wR2 )
0.1798. S ) 1.049. Intensities were measured on an Enraf-
Nonius CAD4T diffractometer with rotating anode (Mo KR,
λ ) 0.710 73 Å) at a temperature of 150(2) K. The structure
was solved with the program SIR9715 and refined with the
program SHELXL9716 against F2 values of all reflections up
to a resolution of ((sin θ)/λ)max ) 0.65 Å-1. An absorption
correction was not considered necessary. Germanium and
silicon atoms were constrained to occupy the same positions
in a 1:1 ratio with the same displacement parameters. Non-
hydrogen atoms were refined freely with anisotropic displace-
ment parameters. Hydrogen atoms were refined as rigid
groups. The drawing, structure calculations, and checking for
higher symmetry was performed with the program PLATON.17
NMR Sp ectr oscop y of Solu tion s of 8 a n d P h en yl-
lith iu m in THF (/HMP A). Under argon, THF (3 mL), THF-
d8 (0.4 mL) and HMPA (0.5 mL) were added to 8 (0.1049 g,
0.403 mmol) in an NMR tube of 10 mm diameter. Subse-
quently, the tube was capped with a septum and cooled (-78
°C). After addition of a solution of phenyllithium in cyclohex-
ane/Et2O (0.33 mL, 1.8 M, 0.6 mmol), the dark brown reaction
mixture was homogenized on a vortex. 29Si NMR spectroscopy
at low temperature showed several signals, see Table 1 (entries
1-9). After standing for 14 h at room temperature, four signals
at δ -12 (19%), -14.6 (60% 10), -14.8 (11%), and -18 (10%)
were observed. After addition of water and extraction with
diethyl ether, GC/MS analysis showed the presence of Ph3-
SiOH (6%; presumably formed by inadvertent oxidation on
workup) and 10 (94%). Ph3SiOH: MS (EI, m/z (relative
intensity)) 276 (M•+, 57), 199 ([M - C6H5]+, 100), 181 (12), 152
(7), 122 ([M - 2C6H5]+, 15), 77 (16). 10: MS (EI, m/z (relative
intensity)) 336 (M•+, 48), 259 ([M - C6H5]+, 100), 182 ([M -
2C6H5]+, 45), 105 ([M - 3C6H5]+, 12), 77 (4).
In a second experiment, the reaction was initially observed
in THF/THF-d8 (Table 2, entries 9 and 10), and at a later stage
HMPA (0.15 mL) was added (Table 2, entries 11 and 12, see
text); in this experiment, δ(29Si) -14 was observed for 10.
NMR Sp ectr oscop y of Solu tion s of 14 a n d P h en yl-
lith iu m in THF (/HMP A). Under nitrogen, THF-d8 (0.523 mL)
and HMPA (0.077 mL) were added to 14 (0.062 mmol) in an
NMR tube of 5 mm diameter, homogenized, and cooled in
liquid nitrogen. After addition of a solution of phenyllithium
in cyclohexane/Et2O (0.052 mL, 1.8 M, 0.093 mmol), the tube
was capped, warmed to -100 °C, and then homogenized on a
vortex to yield an almost black reaction mixture. 29Si NMR
spectroscopy at -80 °C showed the practically exclusive
formation of 12 (δ(29Si) -69.9) with occasionally traces of 9
(δ(29Si) -94) 10 (δ(29Si) -14), and 11 (δ(29Si) -103).
Ack n ow led gm en t. This research was supported
(N.R., M.L., A.L.S.) by The Council for Chemical Sci-
ences (GCW), with financial aid from The Netherlands
Organization for Scientific Research (NWO).
Su p p or tin g In for m a tion Ava ila ble: Details of the crys-
tal structure analysis. This material is available free of charge
via the Internet at http://pubs.acs.org.
OM990976O
(15) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.;
Moliterni, A. G. G.; Burla, M. C.; Polidori, G.; Camalli, M.; Spagna, R.
SIR97: Program for Crystal Structure Solution; University of Bari,
Bari, Italy, 1997.
(16) Sheldrick, G. M. SHELXL-97: Program for Crystal Structure
Refinemen; University of Go¨ttingen, Go¨ttingen, Germany, 1997.
(17) Spek, A. L. PLATON: A Multipurpose Crystallographic Tool;
Utrecht University, Utrecht, The Netherlands, 1998.