Palladium-catalyzed cross-coupling reactions of allyl, phenyl, alkenyl, and alkynyl germatranes with aryl iodides

Article abstract of DOI:10.1021/om020578c

The potential of using organogermatranes in palladium-catalyzed cross-coupling with aryl iodides has been investigated. We have found that organogermatranes are less reactive in this analogue of Stille coupling than trialkylorganostannanes; nevertheless activation by fluoride promotes the reaction. The hypervalent germanium species produced from the germatranes provide a more efficient and more easily handled reagent than trialkoxygermanium analogues.

Full text of DOI:10.1021/om020578c

Organometallics 2002, 21, 5911-5918
5911
P a lla d iu m -Ca ta lyzed Cr oss-Cou p lin g Rea ction s of Allyl,
P h en yl, Alk en yl, a n d Alk yn yl Ger m a tr a n es w ith Ar yl
Iod id es
J . W. Faller* and Roman G. Kultyshev
Department of Chemistry, Yale University, New Haven, Connecticut 06520
Received J uly 19, 2002
The potential of using organogermatranes in palladium-catalyzed cross-coupling with aryl
iodides has been investigated. We have found that organogermatranes are less reactive in
this analogue of Stille coupling than trialkylorganostannanes; nevertheless activation by
fluoride promotes the reaction. The hypervalent germanium species produced from the
germatranes provide a more efficient and more easily handled reagent than trialkoxyger-
manium analogues.
In tr od u ction
of triphenylarsine as a ligand in reactions with slower
reacting aryl iodides. Their intramolecular variant of
alkenyl iodide-alkenyl siloxane coupling resulted in an
effective synthesis of medium-sized rings with an in-
ternal 1,3-cis-cis diene unit.6b They also demonstrated
mild cross-coupling conditions for a number of silyl
compounds that employed tetrabutylammonium fluoride
as the activator.6c These studies led to the development
of fluoride-free cross-couplings using organosilanols.6d,e
Organogermanes, however, received much less atten-
tion as possible replacements for organostannanes,
presumably owing to their higher cost, lower reactivity,
and less developed chemistry. Kosugi et al.7 reported
that 1-aza-5-germa-5-organobicyclo[3.3.3]undecanes (car-
bagermatranes, 1) are much more reactive in the cross-
coupling reaction with aryl bromides than tetracoordi-
nated germanes owing to the internal coordination of
nitrogen to germanium. A similar observation for the
internally coordinated carbastannatranes N(CH2CH2-
CH2)3SnR (2) has been previously made by Vedejs et
al.8a This intramolecular nucleophilic assistance allowed
alkyl group transfer, which does not usually occur in
the case of tetracoordinated stannanes, to aryl halides8a
and alkenyl triflates.8b
In the past decade the palladium-catalyzed cross-
coupling reactions between organic electrophiles and
various organometallic reagents became a standard tool
for construction of carbon-carbon bonds. Among them
the reaction of trialkylorganostannanes (the Stille-
Migita-Kosugi reaction) is of great importance owing
to the high reactivity and availability of the tin reagents
as well as their stability toward air and moisture.1
Nevertheless, a serious drawback of the Stille reaction,
that is, the toxicity of the starting materials and
byproducts, has prompted many research groups to
search for alternative reagents. In particular, silicon
reagents have received significant attention. This area
of research was pioneered by Hiyama, who found that
silanes could be activated toward transmetalation in the
presence of a fluoride-ion source via formation of hy-
pervalent silicates.2 Later Shibata3 and DeShong4 re-
ported on the palladium-catalyzed cross-coupling of aryl
siloxanes with aryl halides in the presence of tetrabu-
tylammonium fluoride (TBAF). In some cases, the
presence of a phosphine was necessary to achieve high
yields. N-Heterocyclic carbene ligands instead of phos-
phines have also been used in similar systems.5 Reac-
tions of aryl bromides with alkynylsilanes can also be
catalyzed by a palladium/imidazolium salt system that
could potentially involve carbene intermediates. Den-
mark and co-workers6a demonstrated the validity of
vinylpolysiloxanes as cross-coupling partners in reac-
tions with aryl iodides and noted the beneficial effect
In our hands, however, an attempted synthesis of
N(CH2CH2CH2)3GeCl, a precursor necessary for the
(1) Farina, V.; Krishnamurthy, V.; Scott, W. J . Org. React. 1997,
50, 1.
(2) (a) Hiyama, T.; Hatanaka, Y. Pure Appl. Chem. 1994, 66, 1471-
1478. (b) Hiyama, T. In Metal-Catalyzed Cross-Coupling Reactions;
Diederich, F., Stang, P. J ., Eds.; Wiley-VCH: Weinheim, 1998; Chapter
10.
(3) Shibata, K.; Miyazawa, K.; Goto, Y. Chem. Commun. 1997,
1309-1310.
(4) (a) Mowery, M.; DeShong, P. J . Org. Chem. 1999, 64, 1684-
1688. (b) Mowery, M.; DeShong, P. J . Org. Lett. 1999, 1, 2137-2140.
(c) DeShong, P.; Handy, C. J .; Mowery, M. E. Pure Appl. Chem. 2000,
72, 1655-1658. (d) Manoso, A. S.; DeShong, P. J . Org. Chem. 2001,
66, 7449-7455.
(5) (a) Lee, H. M.; Nolan, S. P. Org. Lett. 2000, 2, 2053-2055. (b)
Yang, C. L.; Nolan, S. P. Organometallics 2002, 21, 1020-1022.
(6) (a) Denmark, S. E.; Wang, Z. J . Organomet. Chem. 2001, 624,
372-375. (b) Denmark, S. E.; Yang, S.-M. J . Am. Chem. Soc. 2002,
124, 2102-2103. (c) Denmark, S. E.; Pan, W. T. Org. Lett. 2001, 3,
61-64, and references therein. (d) Denmark, S. E.; Sweis, R. F. J . Am.
Chem. Soc. 2001, 123, 6439-6440. (e) Denmark, S. E.; Pan, W. T. J .
Organomet. Chem. 2002, 653, 98-104.
(7) Kosugi, M.; Tanji, T.; Tanaka, Y.; Yoshida, A.; Fugami, K.;
Kameyama, M.; Migita, T. J . Organomet. Chem. 1996, 508, 255-257.
(8) (a) Vedejs, E.; Haight, A. R.; Moss, W. O. J . Am. Chem. Soc. 1992,
114, 6556-6558. (b) J ensen, M. S.; Yang, C.; Hsiao, Y.; Rivera, N.;
Wells, K. M.; Chung, J . Y. L.; Yasuda, N.; Hughes, D. L.; Reider, P. J .
Org. Lett. 2000, 2, 1081-1084.
10.1021/om020578c CCC: $22.00 © 2002 American Chemical Society
Publication on Web 11/27/2002

5912 Organometallics, Vol. 21, No. 26, 2002
Faller and Kultyshev
synthesis of organocarbagermatranes, met with only
very limited success. Consequently, we decided to look
for a similar class of molecules with internal coordina-
tion, which would allow more facile synthesis. Ger-
matranes (3) appeared to be good candidates owing to
the ease of their preparation in good yields from
organotrihalogermanes and triethanolamine.9 In this
report we present our findings with respect to the
behavior of this class of compounds in palladium-
catalyzed cross-coupling reactions with aryl iodides.
matranes are unreactive. On the other hand, an addi-
tion-elimination or Heck-type mechanism (Scheme 1)
can explain the formation of the internally arylated
product, R,p-methylstyrene. In the Heck reaction, aryl
palladation of a double bond is followed by â-hydrogen
elimination. In the case of allyl germatrane elimination
of GeO3N is probably more favorable since we do not
observe any arylated germatranes among the products.
Addition of the aryl group to the double bond can occur
either terminally or internally (pathways “a” and “b”,
respectively). In pathway “a” GeO3N is attached to the
â-carbon and can be eliminated, yielding the terminally
arylated product that is also expected from a Stille
reaction pathway. In pathway “b” GeO3N is attached to
the γ-carbon; therefore â-hydrogen elimination takes
place, resulting in an internally arylated germatrane
π-bound to LPd(H)(I). Insertion of this olefin into the
Pd-H bond can lead to the σ-alkyl complex from which
a GeO3N group can be eliminated to yield R,p-dimeth-
ylstyrene.
Although we found no examples of a similar Heck
reaction featuring allyl germanes in the literature,
similar allylsilane systems have been studied. Thus
Karabelas et al.13a observed the formation of silylated,
as well as desilylated, products in the reaction of
iodobenzene with allyltrimethylsilane. Olofsson et al.13b
showed that internally arylated allyltrimethylsilanes
can be obtained preferentially from allyltrimethylsilanes
and aryltriflates if the chelating ligand DPPF is used
with Pd(OAc)2. J effery demonstrated13c recently that the
direction of the palladium-catalyzed arylation of allyl-
trimethylsilane toward the formation of either silylated
or desilylated product can be controlled by choice of the
base used in the reaction. Allylsilanes have been also
used as terminating moieties in the asymmetric in-
tramolecular Heck reaction leading to the stereoselec-
tive construction of sp3 carbon centers bearing a vinyl
substituent.13d There are a few examples of Heck
reaction of vinylgermanes and vinylsilanes. Thus,
Kosugi et al.7 reported that vinyltributylgermane af-
forded p-methylstyrene in 60% yield upon reaction
with p-bromotoluene. Ikenaga et al.14a published the
first example of aryldegermylation of styryltrimeth-
ylgermanes by arenediazonium tetrafluoroborates lead-
ing to both R- and â-arylated styrenes. Reaction of a
hypervalent alkenyl silicon reagent with aryl iodides has
also been reported to produce both R- and â-styrene
derivatives.14b
Resu lts a n d Discu ssion .
P r elim in a r y Stu d ies w ith Allyl, P h en yl, a n d
P h en yleth yn yl Ger m a tr a n es. Our attempts to repeat
the results of Kosugi et al.7 for the cross-coupling
between 1 (R ) allyl, phenyl) and aryl bromides using
analogous germatranes (R ) allyl, 3a ; R ) phenyl, 3b)
were unsuccessful. Only a trace amount of allylbenzene
was detected after the reaction of 3a with PhBr in THF
in the presence of Pd2(dba)3CHCl3 (1 mol %)-PPh3 (4
mol %) for 45 h at 115 °C in a sealed glass ampule.
Moreover, phenyl germatrane was found to be com-
pletely inert to phenyl group transfer upon Kosugi’s
conditions or heating at 120 °C for 24 h in the presence
of Pd(PPh3)4 (3 mol %). In further experiments 4-iodot-
oluene (4a ) was used as a substrate with a catalyst
obtained in situ from Pd(dba)2 and a ligand. Under
”ligandless” conditions, formation of small amounts of
4-allyltoluene was observed. Surprisingly, in a similar
experiment in the presence of PPh3 (20 mol %) no
coupling product could be detected. Although replacing
PPh3 with bulkier phosphines, such as P(o-tol)3 and
2-(dicyclohexylphosphino)biphenyl, resulted in low con-
versions to 4-allyltoluene, addition of AsPh3 as a ligand
(1:1) was more effective, resulting in 27% conversion to
4-allyltoluene. In all of these experiments a second,
minor coupling product was also detected. It was identi-
1
fied as R,p-dimethylstyrene by comparison with the H
NMR data reported in the literature.10 Similarly, forma-
tion of R-methylstyrene was observed in a stoichiometric
reaction between allyl germatrane and (PPh3)2Pd(Ph)-
(I)11 in toluene at 100 °C.
Despite the limited success with allyl germatrane,
phenyl and phenylethynyl germatranes were inert in
attempted coupling reactions with 4a in the presence
of Pd(dba)2/AsPh3. The observed reactivity of ger-
matranes bearing different apical substituents clearly
contradicted the order established for the Stille reac-
tion, that is, ethynyl > alkenyl > aryl > allyl.12 In
addition, formation of the internally arylated product
in the allyl germatrane case cannot be explained by a
Stille-like mechanism. If we assume that allyl ger-
matrane reacts with 4-iodotoluene in a Heck-type reac-
tion by virtue of the double bond present, then it
becomes clear why phenyl and phenylethynyl ger-
In agreement with a Heck-type mechanism, addition
of 1 and 2 equiv of triethylamine as a base improved
the conversion of 4a to 4-allyltoluene to 44 and 51%,
respectively. Similarly, upon reaction of 3-butenylger-
matrane with 4-iodotoluene under the conditions that
gave maximum conversion in the case of allyl ger-
matrane, the major product was 1-(4-tolyl)-2-butene
(33%), with one of the minor products being identified
as 2-(4-tolyl)-2-butene (ratio 2.4:1).
(9) Gar, T. K.; Khromova, N. Yu.; Sonina, N. V.; Nikitin, V. S.;
Polyakova, M. V.; Mironov, V. F. Zh. Obshch. Khim. 1979, 49, 1516-
1522.
(10) Barton, D. H. R.; Parekh, S. I. Synth. Commun. 1989, 19, 3353-
3361.
(11) Fitton, P.; J ohnson, M. P.; McKeon, J . E. J . Chem. Soc., Chem.
Commun. 1968, 6-7.
(12) (a) Spessard, G. O.; Miessler, G. L. Organometallic Chemistry;
Prentice Hall: Upper Saddle River, NJ , 1996; p 422. (b) Stille, J . K.
Angew. Chem., Int. Ed. Engl. 1986, 25, 508-523.
(13) (a) Karabelas, K.; Westerlund, C.; Hallberg, A. J . Org. Chem.
1985, 50, 3896-3900. (b) Olofsson, K.; Larhed, M.; Hallberg, A. J . Org.
Chem. 1998, 63, 5076-5079. (c) J effery, T. Tetrahedron Lett. 2000,
41, 8445-8449. (d) Tietze, L. F.; Raschke, T. Synlett. 1995, 6, 597-
598.
(14) (a) Ikenaga, K.; Matsumoto, S.; Kikukawa, K.; Matsuda, T.
Chem. Lett. 1990, 185-188. (b) Hojo, M.; Murakami, C.; Aihara, H.;
Komori, E.; Kohra, S.; Tominaga, Y.; Hosomi, A. Bull. Soc. Chem. Fr.
1995, 132, 499-508.

Cross-Coupling Reactions of Germatranes
Organometallics, Vol. 21, No. 26, 2002 5913
Sch em e 1
Ta ble 1. Cr oss-Cou p lin g of P h en yl Ger m a tr a n e w ith 4a
entry
3b, equiv
L
TBAF, equiv
time, h
yield of 5, %
conversion of 4a to 6, %
1
2
3
4
5
6
7
8
1
0
1
2
2
2
2
2
AsPh3
AsPh3
1
1
1
2
2
2
2
4
16
20
20
16
12
12
12
15
28a
32a
42a
28a
34b
14a
13a
24a
60b
18a
AsPh3
43b/38c
31a
2-P(t-Bu)2-biphenyl
2-PCy2-biphenyl
(2-furyl)3P
31a
36a
AsPh3
22b
Determined by integration of p-CH3 signals of 5, 6, and 4a in the 1H NMR spectra of crude products. Determined by H NMR using
a
b
1
1,3,5-trimethoxybenzene as an internal standard. c Isolated yield.
Rea ction s of P h en yl Ger m a tr a n e (3b) in th e
P r esen ce of TBAF . To avoid possible complications
from a Heck reaction pathway in the case of ger-
matranes containing isolated double bonds, we exam-
ined the activation of 3b toward transmetalation using
TBAF (Table 1). The reaction of 4a with phenylger-
matrane (1:1) in THF in the presence of Pd(dba)2, AsPh3,
and 1 equiv of TBAF (entry 1) resulted in 28% conver-

5914 Organometallics, Vol. 21, No. 26, 2002
Faller and Kultyshev
Sch em e 2
Sch em e 3
Ta ble 2. Cr oss-Cou p lin g of Alk en yl Ger m a tr a n es w ith 4a
entry
alkenyl germatrane
R1
R2
L
yield of 12, %
yield of 6, %
1
2
3
E-11a
E-11a
Ph
Ph
H
H
H
Ph
2-P(t-Bu)2-biphenyl
AsPh3
AsPh3
53a
21a
16a
15b
64a
Z-11a
73b
(E/Z ) 1:3.5)c
Z-11b
E/Z ) 1:2.6d
71a
4
5
H
H
p-Tol
AsPh3
AsPh3
16a
12b
(E/Z ) 1:7.5)c
Z-11c
E/Z ) 1:5.4d
73b
p-ClC6H4
(E/Z ) 1:5.5)c
E/Z ) 1:3.4d
a
b
Determined by 1H NMR using 1,3,5-trimethoxybenzene as an internal standard. Determined by 1H NMR using 1,2-diphenoxyethane
as a standard. c Determined by integration of olefinic hydrogens. Determined by integration of methyl hydrogens.
d
sion of the starting iodide to 5 and 32% conversion to 6.
This was encouraging since no coupling could be de-
tected in the absence of TBAF. The homocoupling prod-
uct was also obtained in the absence of phenyl ger-
matrane (entry 2), in agreement with data of Mowery
and DeShong4a obtained under “ligandless” conditions.
Our attempt to use Pd(dba)2 without added ligand re-
sulted in lower conversions of the iodide to both hetero-
and homocoupled products compared to the experiment
with triphenylarsine (entry 3). Increasing the ger-
matrane-to-iodide ratio to 2:1 significantly improved the
conversion to 4-phenyltoluene (entry 4). Both 2-P(t-Bu)2-
biphenyl and 2-PCy2-biphenyl were less effective than
AsPh3, resulting in 31% conversion to the cross-coupling
product (entries 5, 6). Tris(2-furyl)phosphine (entry 7)
was also less effective than triphenylarsine. Increasing
the ratio of TBAF to 4a to 4:1 resulted in promoting
the homocoupling of 4a rather than the cross-coupling
reaction.
of Guillemin and co-workers.15 Attempted cross-coupling
between vinyl germatrane and 4a in the presence of
TBAF and Pd(dba)2/AsPh3 (1:2) led to a mixture of
products, one of them being the desired 4-methylstyrene
(Scheme 2). Two more products have been identified,
one as (E)-â-(4-methyl)styrylgermatrane and another as
(E)-4,4′-dimethylstilbene. Although product 8 may be
interpreted as a Stille product, both 8 and 9 could result
from a Heck reaction. In the last step of this reaction
either a â-germatranyl moiety or a â-hydrogen is
eliminated from two possible regioisomers of the σ-alky-
lpalladium intermediate (Scheme 3). The stilbene prod-
uct 10 could form as a result of (1) a Heck reaction
between 8 and 4a ; (2) a cross-coupling reaction between
9 and 4a ; or (3) a combination of both pathways.
As shown in Table 2, â-aryl-substituted ethenyl
germatranes prepared by rhodium-catalyzed hydroger-
mylation of arylacetylenes by HGe(CH2CH2O)3N16 were
found to form stilbenes upon reaction with 4a in THF.
Rea ction s of Alk en yl Ger m a tr a n es in th e P r es-
en ce of TBAF . Vinyl germatrane (7) was prepared from
vinylgermanium trichloride obtained by the procedure
(15) Guillemin, J .-C.; Lassalle, L.; J anati, T. Planet. Space Sci. 1995,
43, 75-81.
(16) Faller, J . W.; Kultyshev, R. G. Submitted to Organometallics.

Cross-Coupling Reactions of Germatranes
Organometallics, Vol. 21, No. 26, 2002 5915
Ta ble 3. Cr oss-Cou p lin g of Alk yn yl Ger m a tr a n es
w ith Ar yl Iod id es
of diarylacetylenes were obtained with 4a as a substrate
(entries 1 and 4), while 2-iodotoluene (4b) (entries 2 and
5) and 4-iodoanisole (4c) (entries 3 and 6) gave lower
yields of cross-coupling products. When 4-chloro-1-
iodotoluene (4d ) was used as a substrate, the expected
cross-coupling product 14bd was isolated only in mod-
erate yield (entry 7). This was due to the presence of a
second product, which was identified as 1,4-bis(p-
tolylethynyl)benzene22 by its 1H NMR spectrum and
mass spectroscopy. This result suggests that aryl chlo-
rides and bromides could also be activated under the
reaction conditions.
acetylene
entry germatrane R1 iodide
R2
product yield, %
The fluoride could be involved in several steps in
catalysis, the most important of which would seem to
be the production of a more active hypervalent germa-
nium complex. Evidence shedding light onto the fate of
the germatranyl group was found upon careful exami-
nation of products formed after reaction of 13b with 4c
and 4d . In both cases addition of dichloromethane to
the reaction mixture after removal of THF caused
precipitation of white crystalline material, which was
1
2
3
4
5
6
7
13a
13a
13a
13b
13b
13b
13b
H
H
H
CH3
CH3
CH3
CH3
4a
4b
4c
4a
4b
4c
4d
p-CH3
o-CH3
p-CH3O
p-CH3
o-CH3
p-CH3O
p-Cl
14a a
14a b
14a c
14ba
14bb
14bc
14bd
91a
85a
68b
89a
72a
72b
56a,c
Isolated yield. Determined by 1H NMR using 1,3,5-tri-
a
b
methoxybenzene as an internal standard. c Lower yield partially
due to the formation of 1,4-bis(p-tolylethynyl)benzene as a second
cross-coupling product.
1
identified as fluorogermatrane by comparison of its H
and 19F NMR (δ ) -75.5 vs CHF2Cl) spectra with those
of the authentic compound prepared by the method of
Lukevics et al.18 Iodogermatrane formation was not
detected by 1H NMR. Instead, the presence of free iodide
was confirmed by treatment of aqueous ethanol solution
of the residue left after removal of diarylacetylene and
fluorogermatrane with aqueous AgNO3. This does not
necessarily imply the absence of a reductive elimination
from palladium forming a Ge-I bond, as one would
expect a preference of germanium for fluoride. That is,
iodide might be lost after any germanium-palladium
interaction.
Thus, (E)-â-styrylgermatrane (11a ) resulted in 53%
conversion of the iodide to 4-methylstilbene (12a ) with
2-P(t-Bu)2-biphenyl as a ligand and 64% with triphen-
ylarsine (Table 2, entries 1, 2). Interestingly, introduc-
tion of an aryl substituent into â-position completely
shuts down the pathway that would lead to a â,â-biaryl
ethenyl germatrane, as only the cross-coupled and
homocoupled products are observed in these reactions.
The nature of a substituent on the phenyl ring does not
appear to significantly affect the yield of the cross-
coupling product. Thus, the yields were almost identical
for electron-releasing methyl, electron-neutral hydrogen,
and electron-withdrawing chloro substituents in para-
position (entries 3-5). In general, when germatranes
containing mostly Z-isomer were used as the cross-
coupling partners, the E/ Z ratios of stilbenes obtained
were higher than the corresponding ratios in the start-
ing germatranes (entries 3-5).
Rea ction s of Alk yn yl Ger m a tr a n es in th e P r es-
en ce of TBAF . Alkynyl germatranes are readily ac-
cessible from the corresponding alkynyl germanium
trichlorides, which in turn can be prepared via the
method of Lutsenko and co-workers.17 Preliminary
experiments with 4a , phenylethynyl germatrane, and
10 mol % Pd(dba)2 demonstrated that in the presence
of TBAF transfer of the phenylethynyl group occurs
readily, and no homocoupled product is observed with
either AsPh3 or 2-P(t-Bu)2-biphenyl (20 mol %). How-
ever, in the case of arsine formation of diphenylacety-
lene was detected by GLC (ratio 1:16, yield of p-tolyl-
phenylacetylene was 92% from 1H NMR). No such
scrambling of groups was observed when the 2-P(t-Bu)2-
biphenyl ligand was used. For this reason the latter was
chosen for the further experiments shown in Table 3.
The yield of the cross-coupling product was not affected
when the amounts of both Pd(dba)2 and ligand were
reduced to 5 and 10 mol %, respectively.
Rea ction of E,Z-â-Meth ylcr otylger m a tr a n e (E,Z-
15) in t h e P r esen ce of TBAF . This compound was
prepared from E,Z-â-methylcrotylgermanium trichloride
and triethanolamine.19 This substrate with a trisubsti-
tuted double bond was chosen to minimize a possible
Heck reaction pathway. Several ligands were used
(Table 4); among them 2-P(t-Bu)2-biphenyl gave the
highest conversion to the cross-coupling products. In-
crease of catalyst loading from 5 to 10 mol % led only
to increased conversion of the iodide to the product of
homocoupling. Although the starting germatrane con-
tained only a small amount of the 3-(2-methyl)butenyl
isomer (11:1), the corresponding ratio of cross-coupling
products 16a :16b was significantly lower (6:1). This
result indicates that allylic rearrangement occurred,
similar to the observations of others in the Stille
reaction.20
Tr ia lk oxyger m a n es vs Ger m a tr a n es in TBAF -
Assist ed Cr oss-Cou p lin g R ea ct ion s. From our re-
(18) Lukevics, E.; Belyakov, S.; Arsenyan, P.; Popelis, J . J . Orga-
nomet. Chem. 1997, 549, 163-165.
(19) Faller, J . W.; Kultyshev, R. G. To be submitted to Organome-
tallics.
(20) (a) Godschalx, J .; Stille, J . K. Tetrahedron Lett. 1980, 21, 2599-
2602. (b) Trost, B. M.; Keinan, E. Tetrahedron Lett. 1980, 21, 2595-
2598. (c) Echavarren, A. M.; Stille, J . K. J . Am. Chem. Soc. 1987, 109,
5478-5486. (d) Labadie, S. S. J . Org. Chem. 1989, 54, 2496-2498. (e)
Obora, Y.; Tsuji, Y.; Kobayashi, M.; Kawamura, T. J . Org. Chem. 1995,
60, 4647-4649.
(21) Zaitseva, G. S.; Karlov, S. S.; Churakov, A. V.; Howard, J . A.
K.; Avtomonov, E. V.; Lorberth, J . Z. Anorg. Allg. Chem. 1997, 623,
1144-1150.
(22) Nguyen, P.; Yuan, Z.; Agocs, L.; Lesley, G.; Marder, T. B. Inorg.
Chim. Acta 1994, 220, 289-296.
The results obtained with different germatranes and
aryl iodides are summarized in Table 3. Very good yields
(17) Efimova, I. V.; Kalganov, B. E.; Kazankova, M. A.; Lutsenko,
I. F. Zh. Obshch. Khim. 1984, 54, 459-460.

5916 Organometallics, Vol. 21, No. 26, 2002
Faller and Kultyshev
Ta ble 4. Cr oss-Cou p lin g of E,Z-â-Meth ylcr otyl
tography was performed on a Hewlett-Packard 5890 GC.
Melting points were determined in open capillaries and are
uncorrected.
Ger m a tr a n e w ith 4a
Vin ylger m a n iu m Tr ich lor id e. A 50 mL solv-seal flask
equipped with a stirbar was charged with 0.0313 g of AIBN
and 2.21 g of 97% vinyltributylstannane (6.76 mmol), cooled
to -78 °C, and evacuated. The flask was transferred to a
glovebox, and 0.77 mL of GeCl4 (6.75 mmol) was added. The
reaction mixture was heated at 90 °C for 3 h, during which
time the mixture became light brown. Pure CH2dCHGeCl3
(1.24 g, 89%) was obtained by vacuum distillation. The 1H
NMR agreed with that reported by Guillemin and co-workers.15
Vin yl Ger m a tr a n e (7). A 100 mL Schlenk flask equipped
with a stirbar was charged under nitrogen with 1.52 g of CH2d
CHGeCl3 (7.38 mmol) and 20 mL of anhydrous toluene. EtOH
(1.30 mL, 22.2 mmol) and triethylamine (3.09 mL, 22.2 mmol)
were added. After stirring for 1 h the white precipitate was
filtered in the glovebox and washed with an additional amount
of anhydrous toluene (90 mL). The filtrate combined with
washings was added to a 250 mL round-bottom flask contain-
ing 1.115 g of 98% triethanolamine (7.324 mmol) in 5 mL of
toluene. The resulting mixture was stirred under nitrogen at
110 °C for 5.5 h. After cooling to room temperature the
resulting solution was filtered and the toluene removed under
reduced pressure. Compound 7 was obtained as a fluffy white
yield of conversion
Pd(dba)2, E,Z-16a + of 4a to
entry
L, mol %
mol %
16b, %a
6, %a
1
2
3
4
2-P(t-Bu)2-biphenyl, 10
2-P(t-Bu)2-biphenyl, 20
P(t-Bu)3, 10
5
10
5
41
40
26
28
31
44
15
49
AsPh3, 10
5
a
Determined by integration of p-CH3 signals of E,Z-16a + 16b,
6, and 4a in the 1H NMR spectra of crude products.
sults it appears that germatranes are unable to par-
ticipate in cross-coupling reactions without activa-
tion with TBAF. Since aryl and vinyl siloxanes have
been reported in cross-coupling reactions in the pres-
ence of fluoride-ion source, it was of interest to look at
the similar reactions of trialkoxygermanes and compare
the results with those obtained for germatranes. Pre-
liminary study showed that under the “ligandless”
conditions reported by DeShong and Mowery,4a PhGe-
(OEt)3 (17a ) did not produce any 4-phenyltoluene
when treated with 4-iodotoluene, 4a , in DMF. Never-
theless, addition of tri(2-furyl)phosphine resulted in
formation of the cross-coupled product in 29% yield,
which is slightly lower than with phenyl germatrane,
3b. Table 5 summarizes the results. With triphenylars-
ine as a ligand, phenyl germatrane also produced a
higher yield of 5 compared to the reaction with phenyl
triethoxygermane (entries 1 and 2, Table 5). Phenyl-
ethynyl triethoxygermane (17b) was inferior to the
corresponding germatrane 13a when reacted with 2-
iodotoluene, 4b , under similar conditions (entries 3
and 4).
1
solid (1.67 g, 92%) melting at 164-165 °C. H NMR (CDCl3,
400 MHz): δ 6.14-6.05 (m, 1H, CHd), 5.99-5.92 (m, 2H,
CH2d), 3.81 (t, 6H, J ) 5.6, CH2O), 2.85 (t, 6H, J ) 5.6, CH2N).
13C{1H} NMR (CDCl3, 100.6 MHz): δ 135.7, 132.4, 56.8, 51.9.
Anal. Calcd for C8H15O3NGe: C, 39.09; H, 6.15, N, 5.70.
Found: C, 38.96; H, 6.08, N, 5.75. The structure was verified
by X-ray crystallography (see Supporting Information). Crystal
data for 7: monoclinic, space group P21/ n (No. 14) a ) 6.7543-
(5) Å, b ) 9.4464(7) Å, c ) 15.430(2) Å, â ) 95.152(5)°, V )
980.53(12) Å3, Z ) 4, T ) 183.2 K; full matrix least squares
refinement of F (118 parameters)
P r ep a r a tion of Ar yleth yn yl Ger m a tr a n es (13a ,b). Phe-
nylethynyl- and p-tolylethynylgermanium trichlorides were
obtained via the procedure of Lutsenko and co-workers.17
Without distillation these trichlorides were converted into the
corresponding germatranes using the two-step procedure
involving alkoxylation with ethanol followed by transalkoxy-
1
lation with triethanolamine.9 The H and 13C NMR spectra of
13a obtained in 95% yield (based on the trichloride) were
identical to those previously reported.21 A representative
procedure for the synthesis of p-tolylethynyl germatrane (13b)
is described. A 50 mL solv-seal flask equipped with a stirring
bar was charged under nitrogen with 0.78 mL of 97% 4-ethy-
nyltoluene (5.97 mmol), 10 mL of toluene, 0.70 mL of germa-
nium tetrachloride (6.14 mmol), and triethylamine (6.17
mmol). The flask was sealed, cooled, evacuated, and then
placed in an oil bath preheated to 100 °C. After 21 h the white
precipitate was filtered off and washed with 20 mL of toluene.
The filtrate and washings were combined in a 100 mL Schlenk
flask equipped with a stirring bar and a rubber septum
followed by addition of 1.05 mL of absolute EtOH (17.9 mmol)
and 2.50 mL of Et3N (17.9 mmol) and stirring at room
temperature for 50 min. The white precipitate was filtered and
washed with 50 mL of toluene, and the filtrate and washings
were added to a 250 mL round-bottomed flask containing 0.890
g of 98% triethanolamine (5.85 mmol) and a stirbar. A
condenser was attached, and the reaction mixture was heated
at 100 °C for 6 h. After cooling to room temperature the
precipitate was filtered and washed with pentane and 60 mL
of dichloromethane. After dichloromethane removal in vacuo
13b was obtained as a white solid (1.786 g, 87% yield based
on GeCl4) melting at 263 °C. 1H NMR (CDCl3, 500 MHz): δ
7.40 (d, 2H, J ) 8.1, C6H4), 7.03 (d, 2H, J ) 8.1, C6H4), 3.87 (t,
6H, J ) 5.6, CH2O), 2.89 (t, 6H, J ) 5.6, CH2N), 2.29 (s, 3H,
CH3). 13C{1H} NMR (CDCl3, 125.8 MHz): δ 138.5, 132.5, 128.8,
120.1, 99.8, 89.9, 56.9, 51.6, 21.7. Anal. Calcd for C15H19O3-
Con clu sion
We have found that organogermatranes are less
reactive in this analogue of Stille coupling than trialkyl-
organostannanes; nevertheless activation by fluoride
promotes the reaction. In comparison of germatranes
and triethyloxygermanes it appears that germatranes
are more efficient in the cross-coupling reaction with
aryl iodides and furthermore they also have the advan-
tage of being easier to handle and purify.
Exp er im en ta l Section
Gen er a l Da ta . Most synthetic manipulations were carried
out using standard Schlenk techniques under an inert atmo-
sphere. Anhydrous solvents were used unless noted otherwise.
1H NMR spectra were recorded on either a Bruker 500 MHz
or a Bruker 400 MHz spectrometer, and chemical shifts are
in ppm relative to residual solvent resonances (1H). 13C{1H}
NMR spectra were recorded on either a Bruker 500 MHz or a
Bruker 400 MHz spectrometer operating at 125.8 and 100.6
MHz, respectively, and referenced to the deuterated solvent
signals. 19F NMR spectra were recorded on a Bruker spec-
trometer operating at 376.3 MHz, and chemical shifts are
measured in ppm relative to external CHF2Cl. Gas chroma-

Cross-Coupling Reactions of Germatranes
Organometallics, Vol. 21, No. 26, 2002 5917
Ta ble 5. Com p a r ison of Ger m oxa n es vs Ger m a tr a n es a s Cr oss-Cou p lin g P a r tn er s
germanium
compound
Pd(dba)2,
mol %
yield of cross-
coupling product, %
entry
iodide
ligand, mol %
AsPh3, 20
AsPh3, 20
2-P(t-Bu)2-biphenyl, 10
2-P(t-Bu)2-biphenyl, 10
1
2
3
4
3b
4a
4a
4b
4b
10
10
5
43a
25a
85b
30b
17a
13a
17b
5
Determined by 1H NMR using 1,3,5-trimethoxybenzene as an internal standard. Isolated yield.
a
b
NGe: C, 53.95; H, 5.74, N, 4.19. Found: C, 53.45; H, 5.77, N,
4.19 (six separate analyses gave acceptable H and N, but
consistently gave slightly low values for carbon).
4.89. Found: C, 79.44; H, 4.90. Later fractions contained both
14bd and 1,4-bis(p-tolylethynyl)benzene.22
The following coupling products have been identified by
1
comparison with the previously reported H NMR spectra: 4-
Gen er a l P r oced u r e for Cr oss-Cou p lin g Rea ction s. A
25 mL round-bottom three-neck flask equipped with a stirring
bar, rubber septum, and stopper was charged with an aryl
iodide, Pd(dba)2, ligand, and organogermatrane (1.2-2 equiv).
A condenser was attached, and the flask evacuated and refilled
with nitrogen. Anhydrous THF (10-12 mL) was added fol-
lowed by a 1 M solution of TBAF in THF (ratio germatrane/
TBAF ) 1:1). The flask was placed in an oil bath and heated
to approximately 75 °C when reflux began. After 12-18 h the
solution was filtered through a pad of Celite. The pad was
washed with three portions of dichloromethane (15 mL), and
the filtrate and washings were combined and volatile materials
removed with a rotary evaporator. The cross-coupling product
was extracted with petroleum ether (bp 35-60 °C) and purified
by column chromatography using petroleum ether as an
eluent. To determine yields by 1H NMR, 25-55 mg of 1,3,5-
trimethoxybenzene or 1,2-diphenoxyethane was added to a
residue followed by CDCl3 (2-3 mL). Integrals of methoxy and
methylene hydrogens, respectively, of these compounds were
used in yield calculations. The representative procedure for
the synthesis of 4-chloro-4′-methyldiphenylacetylene (14bd )
and isolation of fluorogermatrane is as follows. A 25 mL round-
bottom three-neck flask equipped with a stirring bar, rubber
septum, and stopper was charged with 0.1875 g of 99% 4d
(0.778 mmol), 0.0224 g of Pd(dba)2 (0.039 mmol, 5 mol %),
0.0233 g of 2-P(t-Bu)2-biphenyl (0.077 mmol, 10 mol %), and
0.3357 g of 13b (1.005 mmol, 1.3 equiv). After attaching to a
condenser, the flask was evacuated and refilled with nitrogen,
and THF (10 mL) was added followed by 1.0 mL of a 1 M
solution of TBAF in THF (1.0 mmol). The flask was placed in
an oil bath and heated to approximately 75 °C. Twelve hours
after the start of reflux the solution was filtered through a
pad of Celite. The pad was washed with three portions of
dichloromethane (15 mL), and the filtrate and washings were
combined and volatile materials were removed with a rotary
evaporator. Upon addition of CH2Cl2 to the residue, fluoroger-
methylbiphenyl (5),4a phenyl-p-tolylacetylene (14a a ),23 phenyl-
o-tolylacetylene (14ab),23 4-methoxydiphenylacetylene (14ac),23
bis(4-methylphenyl)acetylene (14ba),24 4-methoxyphenyl-p-tolyl-
acetylene (14bc),25 (E)-4-methylstilbene (E-12a ),26a,c,d (Z)-4-
methylstilbene (Z-12a),26d (E)-4,4′-dimethylstilbene (E-12b),26a-c
(Z)-4,4′-dimethylstilbene (Z-12b),26b and (E)-4-chloro-4′-meth-
ylstilbene (E-12c).26c In addition, 5, 6, and 14a a have been
identified by comparison of their GC retention times to those
of the samples of authentic compounds purchased from Aldrich
or synthesized via previously published procedures (14a a ).27
(Z)-4-Ch lor o-4′-m eth ylstilben e (E-12c): colorless oil. 1H
NMR (CDCl3, 400 MHz): δ 7.19-7.03 (m, 8H, p-ClC6H4 and
p-MeC6H4), 6.58 (d, 1H, J ) 12.2, dCH), 6.48 (d, 1H, J ) 12.2,
dCH), 2.32 (s, 3H, CH3). 13C{1H} NMR (CDCl3, 100.6 MHz):
δ 137.3, 136.1, 134.1, 132.8, 131.1, 130.4, 129.2, 128.9, 128.6,
128.4, 21.5.
(p-Tolyl)(o-tolyl)acetylen e (14bb): white solid (72% yield)
1
melting at 43 °C. H NMR (CDCl3, 500 MHz): δ 7.50 (d, 1H,
J ) 7.5), 7.44 (d, 2H, J ) 8.0), 7.24-7.16 (m, 5H), 2.52 (s, 3H,
o-CH3), 2.38 (s, 3H, p-CH3). 13C{1H} NMR (CDCl3, 125.8
MHz): δ 140.3, 138.5, 132.0, 131.6, 129.6, 129.3, 128.3, 125.8,
123.5, 120.7, 93.7, 87.9, 21.7, 21.0. Anal. Calcd for C16H14: C,
93.16; H, 6.84. Found: C, 92.91; H, 6.83.
(E,Z)-4-(4′-Meth ylp h en yl)-3-m eth yl-2-bu ten e (E,Z-16a )
a n d 3-(4′-m eth ylp h en yl)-2-m eth yl-1-bu ten e (16b): color-
1
less oil. H NMR of E,Z-16a (CDCl3, 400 MHz): δ 7.12-7.06
(m, 4H, p-MeC6H4), 5.42-5.35 and 5.34-5.27 (2m, 1H, CHd),
3.34 and 3.25 (2s, 2H, CH2Ar), 2.33 (s, 3H, p-CH3C6H4), 1.73
and 1.62 (2d, 3H, CH3CHdC(Me)CH2), 1.61 and 1.55 (2s, 3H,
dC(CH3)CH2). 1H NMR of 16b (CDCl3, 400 MHz): δ 7.12-
7.06 (m, 4H, p-CH3C6H4), 4.89 (br s, 1H, dCH), 4.85 (br s, 1H,
dCH), 3.36 (q, 1H, J ) 7.1, CH3CHAr), 2.33 (s, 3H, p-CH3C6H4),
1.61 (s, 3H, CH3CdCH2), 1.37 (d, 3H, J ) 7.1, CH3CHAr).
Str u ctu r e Deter m in a tion a n d Refin em en t of 7. Crys-
tals suitable for crystallography were obtained by slow diffu-
sion of ethylcyclohexane into a CH2Cl2 solution of the com-
pound. Data were collected on a Nonius KappaCCD (Mo KR
radiation) and corrected for absorption. The structures were
solved by direct methods and refined on F for all reflections.
1
matrane precipitated as a white solid, mp >300 °C. H NMR
(d6-DMSO, 400 MHz): δ 3.70 (td, 6H, J ) 5.7; 1.5, FGeOCH2),
2.95 (t, 6H, J ) 5.7, CH2N). 19F{1H} NMR (d6-DMSO, 376.3
MHz): δ -75.5. The supernatant liquid was carefully decanted,
and the precipitate was washed with an additional portion of
CH2Cl2. Dichloromethane solutions were combined, the solvent
was removed in vacuo, and the residue was extracted with
petroleum ether. The extract was chromatographed on a silica
gel column using petroleum ether as an eluent. From earlier
fractions compound 14bd was obtained as a white solid in 56%
(23) Gujadhur, R. K.; Bates, C. G.; Venkataraman, D. Org. Lett.
2001, 3, 4315-4317.
(24) Kabalka, G. W.; Wang, L.; Pagni, R. M. Tetrahedron 2001, 57,
8017-8028.
(25) Al-Hassan, M. I. J . Organomet. Chem. 1990, 395, 227-229.
(26) (a) Katritzky, A. R.; Tymoshenko, D. O.; Belyakov, S. A. J . Org.
Chem. 1999, 64, 3332-3334. (b) Shi, M.; Xu, B. J . Org. Chem. 2002,
67, 294-297. (c) Warner, P.; Sutherland, R. J . Org. Chem. 1992, 57,
6294-6300. (d) Ward, W. J ., J r.; McEwen, W. E. J . Org. Chem. 1990,
55, 493-500.
1
yield (0.098 g). H NMR (CDCl3, 400 MHz): δ 7.46-7.40 (m,
4H, p-ClC6H4 and p-MeC6H4), 7.33-7.29 (m, 2H, p-ClC6H4),
7.16 (d, 2H, J ) 7.9), 2.37 (s, 3H, CH3). 13C{1H} NMR (CDCl3,
100.6 MHz): δ 138.9, 134.2, 132.9, 131.7, 129.4, 128.9, 122.2,
120.0, 90.7, 87.8, 21.8. Anal. Calcd for C15H11Cl: C, 79.47; H,
(27) Hundertmark, T.; Littke, A. F.; Buchwald, S. L.; Fu, G. C. Org.
Lett. 2000, 2, 1729-1731.

5918 Organometallics, Vol. 21, No. 26, 2002
Faller and Kultyshev
Non-hydrogen atoms were refined with anisotropic displace-
ment parameters. Hydrogen atoms were included at calculated
positions.
Ack n ow led gm en t. We thank the National Science
Foundation (Grant CHE0092222) for support of this
research.
Cr ysta l Da ta for Vin ylger m a tr a n e (7): monoclinic, space
group P21/n (No. 14) a ) 6.7543(5) Å, b ) 9.4464(7) Å, c )
15.430(2) Å, â ) 95.152(5)°, V ) 980.53(12) Å3, Z ) 4, T )
183.2 K; full matrix least squares refinement of F (118
parameters) was based on 1863 observed reflections (I >
3.00σ(I)) and converged to R ) 0.038, Rw ) 0.054. See
Supporting Information for details.
Su p p or tin g In for m a tion Ava ila ble: An ORTEP drawing
of vinyl germatrane, 7, and a CIF file giving crystallographic
data for 7. This material is available free of charge via the
Internet at http://pubs.acs.org.
OM020578C