Zr-promoted 'pair'-selective and regioselective synthesis of penta-substituted benzene derivatives

Article abstract of DOI:10.1016/j.tet.2003.09.102

Tetra-substituted zirconacyclopentadiene derivatives, obtainable via in situ generation of zirconacyclopropenes and their cyclic carbozirconation with alkynes, can be treated with alkynyllithiums to induce 1,2-migration accompanied by aromatization and pr

Full text of DOI:10.1016/j.tet.2003.09.102

Tetrahedron 60 (2004) 1345–1352
Zr-promoted ‘pair’-selective and regioselective synthesis of
penta-substituted benzene derivatives
*
Yves R. Dumond and Ei-ichi Negishi
Herbert C. Brown Laboratories of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, IN 47907-2084, USA
Received 22 August 2003; accepted 2 September 2003
Abstract—Tetra-substituted zirconacyclopentadiene derivatives, obtainable via in situ generation of zirconacyclopropenes and their cyclic
carbozirconation with alkynes, can be treated with alkynyllithiums to induce 1,2-migration accompanied by aromatization and protonolysis,
leading to the formation of penta-substituted benzene derivatives, in which all five substituted may be different.
q 2003 Elsevier Ltd. All rights reserved.
1. Introduction
and regioselective synthesis of benzene derivatives is the
Co-catalyzed alkyne cyclotrimerization developed by
Vollhardt,5 which has been successfully applied to the
synthesis of natural products including estrone (1) and
protoberberine alkaloids (2). However, satisfactory results
have been obtained mostly with tethered a,v-diynes and a
large excess of symmetrical Me3SiCuCSiMe3 requiring
subsequent differentiation of the two Me3Si groups. The
strategy of tethering alkynes has been further extended by
the use of tethered triynes, as exemplified by a Rh-catalyzed
synthesis of a tricyclic precursor to calomelanolactone (3).6
Synthesis of benzenes via cyclotrimerization of three
molecules of alkynes is a thermodynamically favorable
process reported as early as 1866.1 Almost a century later,
Reppe reported what appears to be the first transition metal-
catalyzed cyclooligomerization of alkynes producing ben-
zenes and cyclooctatetraenes.2 Since then, most, if not all, of
the d-block transition metals have been shown to catalyze
cyclotrimerization of alkynes.3 In cases where unsymme-
trically substituted alkynes and/or two or three different
alkynes are employed, synthetically unattractive mixtures of
products are generally obtained. Thus, random cyclotrimer-
ization of three different alkynes can, in principle, produce
10 different products, each of which can exist as two or more
regioisomers. These reactions can still be very attractive for
the synthesis of compounds of material chemical interest but
not of fine chemicals. Development of ‘pair’-selective4 and
regioselective synthesis of benzenes from three different
unsymmetrically substituted alkenes, as represented by
Scheme 1, has indeed been very difficult.
Although only one alkyne molecule is incorporated into a
7
benzene ring, the Dotz benzannulation via Cr-carbene
¨
addition to alkynes is another noteworthy example of
transition metal-mediated selective synthesis of arenes.
Although no systematic historical presentation is intended
here, various pair-selective and regioselective procedures
involving other late transition metals have also been
developed. In particular, considerable efforts have been
expended in the development of Pd-catalyzed procedures
with the use of tethered triynes,8a haloenynes8b and
halodienes9 by Negishi, haloendiynes also by him8a and
de Meijere,10 and haloarenediynes by Grigg.11 Recent
investigations of Pd-catalyzed benzannulation of enynes
and diynes by Yamamoto and Gevorgyan are also
noteworthy.12
Scheme 1.
Until recently, the use of early transition metals in the
alkyne-based selective arene synthesis had been less well-
developed. However, conversion of zirconacyclopenta-
dienes to benzene derivatives in a pair-selective manner
was reported by Takahashi in 199513 and has since been
further developed.14,15 Even so, those procedures that
permit regiocontrolled syntheses of benzene derivatives
had not been developed at the time of our unexpected
One of the notable earlier investigations of pair-selective
Keywords: Zr-Promoted; ‘Pair’-and regioselective synthesis; Penta-
substituted benzenes; Zirconacyclopentadienes; Alkynylzirconate
migratory insertion.
*
Corresponding author. Tel.: þ1-765-494-5301; fax: þ1-765-494-0239;
e-mail address: negishi@purdue.edu
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2003.09.102

1346
Y. R. Dumond, E. Negishi / Tetrahedron 60 (2004) 1345–1352
Scheme 2.
discovery of arene formation by the reaction of zirconacy-
clopentadienes with alkynyllithiums that has led to the
results described below. We report herein a Zr-promoted
synthesis of benzene derivatives containing up to five
different non-hydrogen groups in the predetermined pos-
itions with or without the use of tethering. It may well
represent the first demonstration of transition metal-
mediated alkyne-based synthesis of such benzene deriva-
tives without two or more of one kind of substituent
(Scheme 2).
To further explore the scope of these migratory insertion
reactions, a zirconacyclopentadiene (4a, generated in situ
n
from Bu2ZrCp2 and 2 equiv. of 3-hexyne,18 was treated
with 2 equiv. of 1-octynyllithium in THF at 23 8Cs for 2 h.
After quenching the reaction mixture with 3 N HCl, 1-(n-
hexyl)-2,3,4,5-tetralthylbenzene (5a) was produced in 65%
NMR yield rather than the expected acyclic, conjugated
triene. As a penta-substituted arylzirconium derivative (6a)
was suspected as the organometallic precursor to the
product (5a), the reaction mixture was quenched with 3 N
DCl. However, the extent of D incorporation at the C-6
position was only 5%. When only 1 equiv. of nHexCuCLi
was used, the yield of 5a was 30%. Examination of the
reaction mixture before quenching by NMR spectroscopy
revealed the following. Firstly, 5a had already been formed
in about 30% yield before quenching, indicating that one H
atom was internally supplied and transferred to the product.
Secondly, about 50% of the starting 4a remained unreacted.
2. Results and discussion
We have previously discovered a novel migratory insertion
reactions of zirconates containing alkynyl and/or aryl
groups shown in Scheme 3.16 During the course of our
investigation of related reactions of alkenylzirconates, a
similar migratory insertion reaction shown in Scheme 4 was
also discovered.17 In this reaction, an alkyl group rather than
an alkenyl group preferentially migrated from Zr to an
alkynyl C atom.
n
Addition of the 2 equiv. of HexCuCLi to this mixture
consumed essentially all of 4a, and the yield of 5a increased
to about 60%, thereby establishing the 2:1 stoichiometry
n
between HexCuCLi and 4a. Thirdly, examination of the
2:1 reaction mixture by 1H NMR spectroscopy revealed the
presence of nearly 1 equiv. of LiCp (d 6.30 ppm).19 This has
provided a plausible explanation for the requirement of
n
2 equiv. of HexCuCLi.
When PhCuCPh was used in place of EtCuCEt, the
aromatization process must have been significantly slowed
down. Upon quenching the reaction mixture with 3 N DCl,
Scheme 3.
1
examination of the complex product mixture by H NMR
spectroscopy revealed the formation of the expected
benzene derivative (5a) and its monodeuterated derivative
with an aromatic ring-bound D as well as 1,5,6-trideuterio-
1,2,3,4-tetraphenyl-1,3,5-dodecatriene (7). The above
results have supported our interpretation that the reaction
must have proceeded via a bicyclic zirconate (8) similar to
that identified in the previously reported reaction of
Scheme 4.

Y. R. Dumond, E. Negishi / Tetrahedron 60 (2004) 1345–1352
1347
Scheme 5.
zirconacyclopentenes.17 All of the results presented above
may be summarized as shown in Scheme 5.
process to generate a seven-numbered zirconcycle may then
be followed by a migratory insertion and aromatization via
allylic rearrangement in Mechanism III. At this moment, it
is not possible to choose one over the others, but it would be
interesting and important to pursue the question of whether
or not direct reductive elimination of diorganylzirconocenes
without the involvement of p-bonds is feasible.
At this point, the mechanism for the conversion of the
bicyclic zirconate (8) into 6 still remains to be clarified,
although the following three alternatives may be considered
as being plausible (Scheme 6). Mechanism I involving
direct formation of 6 from 8 via reductive elimination is
simple and attractive, and a similar reductive elimination
process for the formation of cyclobutenes has been
proposed.20 However, the direct C–C bond formation
from diorganylzirconocenes via reductive elimination is
essentially unknown, and a more indirect path of lower
energy barriers involving active participation of p-bonds
and other strained bonds are likely to be operative. With this
rationalization in mind, two indirect paths, i.e., Mechanisms
II and III, are also proposed for the conversion of 8 into 6. In
Mechanism II, a series of two 1,2-migratory insertion
processes are thought to take place to effect an overall
reductive elimination, while a six-electron electrocyclic
One important mechanistric issue that remained to be
clarified was the source of H for converting 6 into 5. As
stated earlier, quenching the reaction mixture derived from
4a and nHexCuCLi gave the expected monodeuterated
derivative of 5a only in 5% yield. So, the putative
intermediate 6a either was not actually formed or was
indeed formed but converted into 5a by abstraction of H
from one or more compounds present in the reaction
mixture before quenching with HCl or DCl. Irrespective of
what the correct mechanism might be, the source of the
majority of H must be one or more of the compounds in the
reaction mixture. Accordingly, some deuterated compounds
were used in place of their non-deuterated counterparts. No
D incorporation was observed when THF-d8 was used in
place of THF. On the other hand, incorporation of D to the
extent of 14% was observed above and beyond 5% from
DCl quenching, when cyclopentadiene-d5 of 96% D was
used for the preparation of Cl2ZrCp2-d10. Although cleaner
D incorporation was never achieved, it was clear that there
were more than one source of H and that the organo-
zirconium precursor to 5a, be it 6a or not, must have been
competively quenched by internal H sources.
It then occurred to us that, if 6 or an alternate precursor to 5
could be either stabilized or competitively protonated by an
added protonating agent to give 5 without interfering with the
desired process, 5 would be produced in higher yields.
Accordingly, various Lewis bases as election pair donors and
proton acids as protonating agents were added to the reaction
mixture. However, addition of Et3N, 4-(N,N-dimethylamino)-
pyridine (DMAP), dppp, and dppe had no detectable effect,
while that of PMe3 totally inhibited the reaction.
Scheme 6.

1348
Y. R. Dumond, E. Negishi / Tetrahedron 60 (2004) 1345–1352
synthesis of fine chemicals, even though they may be very
useful in the synthesis of oligomers and polymers of
material chemical interest. For the former objective, it was
essential to be able to synthesize zirconacyclopentadienes
containing four different non-hydrogen groups, one or more
of which may be replaced with H and/or some other groups,
such as SiMe3, that can be differentiated later. Of various
methods thus far developed for the synthesis of zircon-
acyclopentadienes,21 two of them often referred to as the
Erker–Buchwald protocol22 and the Negishi–Takahashi
protocol23 have been most widely used. A more recently
developed method of Rosenthal24 and its modification by
Tilley25 as well as earlier contributions by Farona26 and
Nugent27 are also noteworthy. For the purpose of develop-
ing pair-selective synthesis of benzene derivatives, modifi-
cations of the Negishi–Takahashi protocol through the use
of Et2ZrCp2 have proved to be useful.28 At the time of this
investigation, however, there was only one reported
procedure by Buchwald29 for the pair-selective and
regioselective synthesis of zirconacyclopentadienes with
four different groups in the predetermined positions, as
exemplified by the process shown in Scheme 7.
Accordingly, we generated 9 in about 65% yield as a 94:6
regiosomeric mixture of 9 with 10. Under non-optimized
conditions, the 94:6 mixture of 9 and 10 was treated with
2 equiv. of nHexCuCLi. One regioisomer of .93%
regiosomeric purity thus obtained was identified as 11
(Scheme 7). Thus, the conjugated diene moiety containing
nBu, H, Me, and Me3Si groups retained its regiochemistry
throughout the reaction. Noteworthy is the nearly 100%
regioselectivity observed for the incorporation of 1-octyne
in the conversion of 9 to 11. The results indicate that the
migratory insertion of 1-octyne took place almost exclu-
Scheme 7.
Fortunately, addition of proton acids was much more
fruitful. The best proton source proved to be the same
terminal alkyne as that used for the preparation of the
alkynyllithium. Thus, addition of 1 equiv. of 1-octyne to the
mixture of the reaction of 4a with 2 equiv. of nHexCuCLi
at 23 8C for 3 h led to the formation of 5a in 90% yield.
When nHexCuCD was used, D incorporation in the
expected position took place to the extent of 86%. These
results have not only provided a much superior procedure
for the conversion of 4 into 5 but also strongly supported the
intermediacy of 6 as the immediate precursor to 5. Other
proton acids were less effective. Methanol was evidently too
acidic, as no desired product was formed. Presumably,
nHexCuCLi was prematurely quenched by MeOH. Indene
and t-butanol did produce the desired product 5a, but with
no improvement.
n
sively on the side of Bu or away from Me3Si (Scheme 7).
These results were obtained before the development of an
improved procedure involving addition of a 1-alkyne, and
its application to this case has not yet been performed.
As has been amply demonstrated (vide supra), the use of
tethered diynes substantially reduced the level of difficulty
in controlling regioselectivity. Some preliminary results
along this line are summarized in Scheme 8. In cases where
unsymmetrically substituted diynes were used, the reaction
proceeded in approximately 85% regioselectivity. When Zr
was flanked with nBu and Ph, the nBu side migrated
preferentially. In addition to steric factors, electronic
As stated earlier, any methods for the synthesis of benzene
derivatives that must necessarily incorporate two or more of
one kind of substituent are of very limited utility for the
Scheme 8.

Y. R. Dumond, E. Negishi / Tetrahedron 60 (2004) 1345–1352
1349
factors, such as the well-documented benzylic, allylic, and
propergylic interaction with Zr, appears to be significant.
Although only one symmetric tether, i.e., tetramethylene,
was used in this study, a variety of unsymmetric tethers
including those containing heteroatoms can, in principle, be
used in the reaction described herein, and efforts are being
made along this line.
mined using mesitylene as an internal standard. IR spectra
were recorded on a Perkin–Elmer Spectrum 2000 FTIR
spectrometer. THF was distilled from sodium benzophenone
ketyl.
4.2. Preparation of symmetrically tetra-substituted
zirconacyclopentadienes
All of the reactions discussed above employed Cp2ZrCl2,
but we also used (Ind)2ZrCl2, where Ind is indenyl, and
Cp2HfCl2 for the synthesis of 5a from 3-hexyne. Without
the benefit of added 1-octyne, 5a was obtained in 60 and
35% yields, respectively. Prior to the experiments described
thus far in this paper, we briefly investigated the reaction of
dialkynylzirconocoenes with alkenyllithiums. The reaction
of (PhCuC)2ZrCp2 with (E)-PhCuCLi gave, upon treat-
ment with I2, a rather complex mixture containing (E)-
PhCHvCHCuCPh and (E,E)-PhCHvCHCHvCHPh
among others. In no case was a clean reaction observed.
More recently, Hirao30 reported the reaction of alkenyl-
zirconocene chlorides with 2 equiv. of alkynyllithiums to
give (E,Z)-conjugated dienes in good yields, thereby further
extending the scope of migratory insertion reactions of
alkynylzirconates.
Zirconacyclopentadienes containing four Et groups (4a) and
four Ph groups (4b) were prepared by the reaction of
nBu2ZrCp218 with 2 equiv. of 3-hexyne and diphenylethyne,
respectively, according to the previously reported pro-
cedure,18 as described below for the synthesis of 5a.
4.2.1. Preparation of 1,1-bis(h5-cyclopentadienyl)-
2,3,4,5-tetraethylzirconacyclopentadiene (4a). To a sol-
ution of Cl2ZrCp2 (584 mg. 2.0 mmol) in THF (6 mL) were
added dropwise at 278 8C via syringe nBuLi (2.5 M in
hexane, 1.6 mL, 4.0 mmol) and 3-hexyne (0.33 g, 0.45 mL,
4 mmol). After stirring for 10 min at 278 8C, the reaction
mixture was warmed to 23 8C and stirred further for 2 h.
After addition of mesitylene (240 mg, 0.28 mL, 2 mmol),
clean quantitative formation of 1,1-bis(h5-cyclopenta-
dienyl)-2,3,4,5-tetraethylzirconacyclopentadiene (4a) was
1
observed by GLC and H NMR (C6D6, THF) Cp signal at
5.94 ppm.
3. Conclusions
4.2.2. Preparation of 1,1-bis(h5-cyclopentadienyl)-
2,3,4,5-tetraphenylzirconacyclopentadiene (4b). This
compound was prepared as above in 94% NMR yield31
by using 0.71 g (4 mmol) of diphenylethyne in place of
The migratory insertion reaction of alkynylzirconates,
generated in situ by treating zirconacyclopentadienes with
2 equiv. of alkynyllithiums, unexpectedly produced the
corresponding benzene derivatives rather than the expected
conjugated trienes. Evidently, the expected migratory inser-
tion reaction took place to give bicyclic zirconacycles (8),
which then underwent reductive C–C bond formation by an
as yet unclear mechanism to produce arylzirconates (6)
Their protonolysis provided benzene derivatives (5). Pro-
tonolysis can be best achieved in situ by the addition of
alkynes corresponding to the alkynyllithium used, since the
precursors to 5, i.e., 6, are unstable, and they tend to undergo
further complex reactions before full consumption of the
starting compounds. This novel synthesis of benzene
derivatives is, in principle, applicable to pair-selective and
regioselective synthesis of benzene derivatives containing
up to five different non-hydrogen groups. As such, it appears
to represent the first demonstration of transition metal-
mediated pair-selective and regioselective synthesis of
benzene derivatives from three different and untethered
alkynes.
1
3-hexyne: H NMR(C6D6, THF) Cp signal at 6.22 ppm.
4.3. Reaction of zirconacyclopentadienes (4) with
alkynyllithiums
This reaction was carried out as detailed below for the
conversion of 4a into 5a.
4.3.1. Conversion of 4a into 1-(n-hexyl)-2,3,4,5-tetra-
ethylbenzene (5a). Representative procedure. 1-Octynyl-
lithium, prepared in THF at 278 8C from 1-octyne (0.44 g,
n
0.59 mL, 4 mmol) and BuLi (2.5 M in hexane, 1.6 mL,
4 mmol), was added dropwise at 278 8C via cannula to the
reaction mixture containing 4a described in Section 4.2.1,
and the resultant mixture was stirred at 278 8C for 1 h,
warmed to 23 8C, but the zirconacyclopentadiene 2a was
observed unchanged by 1H NMR (C6D6/THF¼6/1) (Cp
signal at 5.94 ppm). The reaction mixture was stirred for an
additional 1 h to generate the title compound (5a) observed
in the reaction mixture by 1H NMR (C6D6/THF¼6/1)
(aromatic proton signal at d 6.91 ppm, 50% NMR yield).
Formation of CpLi (d 6.30 ppm) up to 100% was also
observed. The reaction mixture was quenched with 3 N HCl,
extracted with pentane, washed with NaHCO3 dried over
4. Experimental
4.1. General
1
MgSO4, and concentrated. GLC and H NMR examination
All manipulations were conducted under an atmosphere of
dry argon, unless otherwise noted. Flash chromatographic
separations were carried out on 230–400 mesh silica gel 60.
Gas chromatography was performed on a HP 6890 Gas
Chromatograph using HP-5 capillary column (30 m£
0.32 mm, 0.5 mm film) with mesitylene as an internal
using mesitylene as an internal standard indicated about
65% yields of the title compound over several runs. After
filtration on a silica gel pad and evaporation in vacuo, the
1
title compound was isolated in 51% yield (279 mg): H
NMR (CDCl3, Me4Si) d 0.90 (t, J¼5.7 Hz, 3H), 1.15–1.4
(m, 20H), 2.55–2.7 (m, 10H), 6.87 (s, 1H); 13C NMR
(CDCl3) d 14.13, 15.59 (2C), 15.68, 15.92, 21.71, 21.79,
standard. H and 13C NMR spectra were recorded on a
Varian Gemini-300 spectrometer. NMR yields were deter-
1

1350
Y. R. Dumond, E. Negishi / Tetrahedron 60 (2004) 1345–1352
22.03, 22.69, 25.60, 29.83, 31.77, 31.82, 33.10, 127.23,
137.32, 137.40, 138.36, 139.27, 139.77; HRMS (FAB
KIPEG) calcd for C20H34 (Mþ1) 275.2739, found
275.2737.
Addition of 12 equiv. of Et3N or DMAP, 6 Mequiv. of 2,20-
bipyridyl or dppp, or 1.6 Mequiv. of dppe did not detectably
change either the rate of formation or the yield of 5a, the
latter of which ranged 50–70%. Addition of (MeOCH2)2 or
TMEDA (6 equiv.) merely slowed down the reaction, and
addition of either PMe3 (12 equiv.) or 1,10-phenanthroline
(6 Mequiv.) inhibited the desired benzene formation. In all
cases, 1H NMR examination of the reaction mixture in
C6D6–THF (6:1) showed the displacement of up to 1 equiv.
of CpLi (d 6.30 pm).
4.3.2. 1-(n-Hexyl)-2,3,4,5-tetraphenylbenzene (5b). To
the reaction mixture containing 4b prepared in Section
4.2.2 was added 1-octynyllithium, prepared in THF at
278 8C from 1-octyne (0.44 g, 0.59 mL, 4 mmol) and
nBuLi (2.5 M in hexane, 1.6 mL, 4 mmol), at 278 8C via
cannula, and the reaction mixture was stirred at 278 8C for
1 h and warmed to 23 8C. The reaction mixture was further
stirred for 3 h, quenched with 3 N HCl, extracted with
pentane, washed with NaHCO3 dried over MgSO4, and
concentrated. GLC and GC/MS examination showed the
formation of the title compound (5b) and 1,2,3,4-tetra-
phenyl-1,3,5-dodecatriene. In case of quenching with 3 N
DCl, GLC and GC/MS examination showed the formation
of the benzene derivative (5b) and 7 containing three
deuterium atoms, supporting the proposed structure of 8.
4.6. Reactions of zirconacyclopentadienes with
alkynyllithiums in the presence of the corresponding
1-alkynes. An improved synthesis of benzene derivatives
These reactions were run in the same manner as described in
Section 4.3.1 except that 1-alkyne (1 equiv. relative to Zr)
was added at the beginning of the reaction of zirconacyclo-
pentadienes with 1-alkynyllithium (2 equiv.).
4.6.1. Improved synthesis of 1-(n-hexyl)-2,3,4,5-tetra-
ethylbenzene (5a). Improved representative procedure.
To the reaction mixture containing 4a (Section 4.2.1), were
added nHexCuCLi, prepared from 1-octyne (0.44 g,
4.3.3. Preparation of 1,1-bis(h5-cyclopentadienyl-2-tri-
methylsilyl-3-methyl-5-(n-butyl)zirconacyclopenta-
diene(9) and its conversion to 1-trimethylsilyl-2-(n-
hexyl)-4-(n-butyl)-6-methylbenzene(11). To HZrCp2Cl
(1.40 g, 5.43 mmol) in 16 mL of toluene was added at
23 8C 1-hexyne (0.45 g, 0.62 mL, 5.43 mmol). After 10 h at
23 8C, MeLi (1.4 M in Et2O, 3.88 mL, 5.43 mmol) was
added at 0 8C. One hour later, 1-trimethylsilyl-1-propyne
(0.61 g, 0.80 mL, 5.43 mmol) was added. The reaction
mixture was then heated at 80 8C for 2 days to afford
regioisomerically 94% pure 9 in 42% yield, only other
regiosomer produced being the 3-(n-butyl) isomer (6%).
n
0.59 mL, 4.0 mmol) and BuLi (2.5 M in hexane, 1.6 mL,
4.0 mmol), and 1-octyne (0.22 g, 2.0 mmol) at 278 8C.
After stirring first at 278 8C for 1 h and then at 223 8C for
several hours, 5a was produced in 90% yield by NMR. Its
spectral data are presented in Section 4.3.1.
4.6.2. 1-(n-Octyl)-6-deuterio-2,3,4,5-tetraethylbenzene.
This reaction was carried out as described in Section 4.6.1
except that 1-decyne (0.55 g, 4.0 mmol) was used to prepare
nOctCuCLi and that 1-deuterio-1-decyne (95% D, 0.28 g,
2.0 mmol) was added as a D source The title compound was
obtained in 90% yield with incorporation of D to the extent
of 86%: 1H NMR(CDCl3, Me4Si) d 0.90 (t, J¼5.7 Hz, 3H),
1.15–1.4 (m, 24H), 2.55–2.7 (m, 10H), 6.87 (s, 14% of 1H);
13C NMR (CDCl3) d 13.91, 15.47 (2C), 15.54, 15.79, 21.58,
21.66, 21.90, 22.58, 25.47, 28.89, 29.39, 31.53, 31.73,
32.95, 33.75, no peak at 127 because of D on the benzene
ring, 137.07, 137.17, 138.04, 138.98, 139,50; LRMS calcd
(Mþ) 303, found 303 (86%).
To the mixture containing 9 obtained above was added at
278 8C 1-octynyllithium, prepared in THF from 1-octyne
n
(1.32 g, 1.77 mL, 12.0 mmol) and BuLi (2.5 M in hexane,
4.8 mL, 12.0 mmol). After stirring at 23 8C overnight, the
standard workup as in Section 4.3.1 afforded 11 in 53%
yield based on 9 as a .93% regioisomerically pure
1
substance: H NMR (CDCl3, Me4Si) d 0.49 (s, 9H), 1.0–
1.1 (m, 6H), 1.4–2.35 (m, 15H), 2.61 (t, J¼7.7 Hz, 2H),
2.78 (t, J¼8.0 Hz, 2H), 6.92 (s, 2H); 13C NMR (CDCl3) d
3.82 (3C), 14.10 (2C), 22.62, 22.68, 24.84, 29.56, 31.92,
33.41, 33.82, 35.31, 36.98, 127.40, 128.52, 132.69, 143.30,
144.08, 149.60; HRMS calcd for C20H36Si (Mþ) 304.2586,
found 304.2588.
4.6.3. Preparation of 3,9-dodecadiyne (12) and its
conversion to 5,8-diethyl-6-(n-hexyl)-1,2,3,4-tetra-
hydronaphthalene (13). 3,9-Dodecadiyne (12) was pre-
pared in 70% yield by successively treating 1,7-octadiyne
n
4.4. Investigation of the proton source
(5.3 g, 6.64 mL, 50 mmol) in THF (100 mL) with BuLi
(2.5 M in hexane, 42 mL, 105 mmol, 278 8C, 1 h) and
iodoethane (17.9 g, 9.2 mL, 115 mmol) in DMPU (150 mL):
1H NMR (CDCl3, Me4Si) d 1.11 (t, J¼7.5 Hz, 6H), 1.57
(broad t, J¼6.5 Hz, 4H), 2.1–2.2 (m, 8H); 13C NMR
(CDCl3) d 12.35 (2C), 14.28 (2C), 18.26 (2C), 28.19 (2C),
79.02 (2C), 81.75 (2C).
Following the same procedure as in Section 4.3.1 for the
preparation of 5a, the reaction mixture was quenched with
3 N DCl and ,5% D incorporation was observed by LRMS.
Following the same procedure as above for the preparation
of 5a but using THF-d8, 0% D incorporation was obtained.
By using Cp2ZrCl2-d10 (.96% D incorporation) for the
synthesis of 5a, 14% D incorporation was recorded by 13C
NMR and LRMS.
3,9-Dodecadiyne (12) (0.324 g, 2.0 mmol) prepared above
was treated with nBu2ZrCp2 generated in THF (6 mL) from
Cp2ZrCl2 (0.584 g, 2.0 mmol) and nBuLi (2.5 M in hexane,
1.6 mL, 4.0 mmol) as described in Section 4.2.1. After
confirming the formation of a zirconabicycle in nearly
4.5. Effects of added bases
1
n
The reaction of 4a with 2 equiv. of nHexCuCLi was run in
the presence of various tertiary amines and phosphines.
quantitative yield by H NMR spectroscopy, HexCuCLi
generated in situ from 1-octyne (0.44 g, 4.0 mmol) and an

Y. R. Dumond, E. Negishi / Tetrahedron 60 (2004) 1345–1352
1351
additional 2.0 mmol (0.22 g) of 1-octyne were added at
Acknowledgements
278 8C, and the reaction mixture was stirred overnight at
23 8C to produce 13 in 90% NMR yield: 1H NMR (CDCl3,
Me4Si) 0.89 (t, J¼6.8 Hz, 3H), 1.11 (t, J¼6.7 Hz, 3H), 1.19
(t, J¼7.6 Hz, 3H), 1.2–1.55 (m, 12H), 1.8–2.8 (m, 10H),
6.85 (s, 1H); 13C NMR (CDCl3) d 13.99, 14.23, 14.36,
21.21, 22.91, 23.28, 25.47, 26.46, 26.94, 29.72, 31.57,
31.80, 32.00, 33.03, 126.41, 132.54, 134.87, 137.41, 137.61,
139.29; HRMS calcd for C20H32 (Mþ) 272.2504, found
272.2501.
The authors thank the National Science Foundation (CHE-
0309613) and Purdue University for support of this
research. Y. R. D. was a Purdue Research Foundation
Graduate Fellow (1997–1998). Generous assistance in the
procurement of Zr compounds by Boulder Scientific Co. is
gratefully acknowledged. We also thank Professor
T. Takahashi for sharing unpublished results and Professor
V. Gegorgyan for providing us with useful information.
Technical assistance provided by Ms Q. Hu is gratefully
acknowledged.
4.6.4. Preparation of 1-phenyl-1,7-dodecadiyne (14) and
its conversion to 5-phenyl-6-(n-hexyl)-8-(n-butyl)-
1,2,3,4-tetrahydronaphthalene (15). 1,7-Octadiyne (10.6 g,
n
13.3 mL, 100 mmol) was mono-butylated by using BuLi
(100 mmol), nBuI (22.1 g, 13.7 mL, 120 mmol), and DMPU
(90 mL). After the usual workup, distillation afforded
1,7-dodecadiyne in about 35% yield. Successive treatment
References and notes
1. Berthelot, M. C. R. Acad. Sci. 1866, 905.
2. Reppe, W.; Schlichting, O.; Klager, K.; Toepel, T. Liebigs
Ann. Chem. 1948, 560, 1.
n
of 1,7-dodecadiyne (0.45 g, 2.8 mmol) in THF with BuLi
(3.36 mmol 278 8C), dry ZnBr2 (4.2 mmol, 278 8C for 1 h
and then 23 8C for 0.5 h), iodobenzene (0.86 g, 0.47 mL,
4.2 mmol), and Pd(PPh3)4 (65 mg, 0.056 mmol, 14 h,
23 8C).32 The reaction mixture was successively treated
with 3 N HCl, pentane, aqueous NaHCO3, and MgSO4 to
3. See, for example Collman, J. P.; Hegedus, L. S.; Norton, J. R.;
Finke, R. G. Principles and Applications of Organotransition
Metal Chemistry; 2nd ed. University Science Books: Mill
Valley, CA, 1987; p 989.
1
give 14 in 91% yield: H NMR (CDCl3Me4Si) d 0.97 (t,
4. In place of ‘pair’-selective, copuloselective (copulo in Latin¼
to pair) has been proposed. See Anastasia, A.; Negishi, E. In
Handbook of Organopalladium Chemistry for Organic
Synthesis; Negishi, E., Ed.; Wiley-Interscience: New York,
2002; p 311.
J¼7 Hz, 3H), 1.45–1.55 (m, 4H), 1.7–1.8 (m, 4H), 2.2–2.3
(m, 4H), 2.49 (t, J¼6.8 Hz, 2H), 7.3–7.35 (m, 3H), 7.45–
7.5 (m, 2H); 13C NMR (CDCl3) d 13.59, 18.30, 18.39,
18.92, 21.90, 27.79, 28.22, 31.18, 79.54, 80.52, 80.74,
89.89, 123.94, 127.43, 128.11 (2C), 131.47 (2C).
5. Vollhardt, K. P. C. Angew Chem. Int. Ed. 1984, 23, 539, and
references therein.
1-Phenyl-1,7-dodecadiyne (14) (0.476 g, 2.0 mmol) was
converted to 15 in 71% NMR yield by its reaction with
nHexCuCLi (4.0 mmol) and 1-octyne (2.0 mmol) as
described in Section 4.6.3. The product (15) yielded the
following spectral data: 1H NMR (CDCl3, Me4Si) d 0.88 (t,
J¼6.7 Hz, 6H), 125–1.6 (m, 16H), 1.8–2.75 (m, 8H), 6.93
(s, 1H), 7.1–7.55 (m, 5H); 13C NMR (CDCl3) d 13.92 (2C),
22.28, 23.12, 26.37, 26.84, 28.77, 28.87, 29.07, 29.33,
31.71, 31.84, 32.41, 32.61, 126.19, 126.66, 127.99 (2C),
129.55 (2C), 132.06, 134.92, 137.53, 139.00, 139.70,
141.06; HRMS calcd for C26H36 (Mþ) 348.2817, found
348.2816.
6. Neeson, S. J.; Stevenson, P. J. Tetrahedron 1989, 45, 6239.
7. For a review, see (a) Wulff, W. D. Comprehensive Organo-
metallic Chemistry II, Abel, E. W., Stone, F. G. A., Wilkinson,
¨
G., Eds.; Pergamon: New York, 1995; Vol. 12, p 469. (b) Dotz,
K. H.; Tomuschat, P. Chem. Soc. Rev. 1999, 28, 187.
8. (a) Negishi, E.; Harring, L. S.; Owczarczyk, Z.; Mohamud,
M. M.; Ay, M. Tetrahedron Lett. 1992, 33, 3253. (b) Negishi,
E.; Ay, M.; Sugihara, T. Tetrahedron 1993, 49, 5471.
9. Zhang, Y.; Negishi, E. J. Am. Chem. Soc. 1989, 111, 3454.
10. Meyer, F. E.; de Meijere, A. Synlett 1999, 777.
11. Grigg, R.; Rasul, R.; Savic, V. Tetrahedron Lett. 1997, 38,
1825.
12. (a) Saito, S.; Yamamoto, Y. In Handbook of Organopalladium
Chemistry for Organic Synthesis; Negishi, E., Ed.; Wiley-
Interscience: New York, 2002; pp 1635–1646 Chapter
IV.10.2. (b) Gevorgyan, V.; Yamamato, Y. J. Organomet.
Chem. 1999, 576, 232.
4.7. Use of Ind2ZrCl2 and Cp2HfCl2 in the synthesis of 5a
from 3-hexyne and 1-octyne
Following the non-optimized procedure described in
Section 4.3.1, the use of Ind2ZrCl2 and Cp2HfCl2 led to
the formation of 5a in 60 and 35% yields, respectively.
13. Takahashi, T.; Kotora, M.; Xi, Z. J. Chem. Soc., Chem.
Commun. 1995, 361.
14. Takahashi, T.; Hara, R.; Nishihara, Y.; Kotora, M. J. Am.
Chem. Soc. 1996, 118, 5154.
4.8. Reaction of bis(phenylethynyl)zirconocene with
(E)-2-phenylethenyllithium
15. Takahashi, T.; Xi, Z.; Yamazaki, A.; Liu, Y.; Nakajima, K.;
Kotora, M. J. Am. Chem. Soc. 1998, 120, 1672.
Bis(phenylethynyl)-zirconocene was generated by treat-
ing Cp2ZrCl2 (2.0 mmol) with PhCuCLi prepared from
4 mmol each of PhCuCH and nBuLi in THF (75%
NMR yield, Cp signal at d 6.15 ppm). Its treatment with
(E)-2-phenylethenyllithium (4 mmol) at 23 8C for 1 d
followed by addition of I2 (2.5 equiv.) let to a complex
mixture containing 1,4-diphenyl-1-buten-3-yne and 1,4-
diphenyl-1,3-butadiene. This reaction was not further
investigated.
16. Takagi, K.; Rousset, C. J.; Negishi, E. J. Am. Chem. Soc. 1991,
113, 1440.
17. Dumond, Y.; Negishi, E. J. Am. Chem. Soc. 1999, 121, 11223.
18. (a) Negishi, E.; Cederbaum, F. E.; Takahashi, T. Tetrahedron
Lett. 1986, 27, 2829. (b) Negishi, E.; Holmes, S. J.; Tour, J. M.;
Miller, J. A.; Cederbaum, F. E.; Swanson, D. R.; Takahashi, T.
J. Am. Chem. Soc. 1989, 111, 3336.
19. (a) Kondakov, D.; Negishi, E. Chem. Commun. 1996, 963.

1352
Y. R. Dumond, E. Negishi / Tetrahedron 60 (2004) 1345–1352
(b) See also Razuvaev, G.; Vyshinskaya, L.; Vasil’eva, G.;
Malysheva, A. Dokl. Akad. Nauk. SSSR 1978, 243, 1212.
26. Thanedar, S.; Farona, M. F. J. Organomet. Chem. 1982, 235,
65.
20. Liu, Y.; Sun, W.; Nakajima, K.; Takahashi, T. Chem.
Commun. 1998, 1133.
27. Nugent, W. A.; Thorn, D. L.; Harlow, R. L. J. Am. Chem. Soc.
1987, 109, 2788.
21. For a review, see Negishi, E.; Takahashi, T. Houben-Weyl
Science of Synthesis, Imamoto, T., Ed.; Thieme: Stuttgart,
2003; Vol. 2, pp 739–774 Sect. 2.11.5.
28. (a) Takahashi, T.; Kageyama, M.; Denisov, V.; Hara, R.;
Negishi, E. Tetrahedron Lett. 1993, 34, 687. (b) Xi, Z.; Hara,
R.; Takahashi, T. J. Org. Chem. 1995, 60, 4444.
29. (a) Buchwald, S. L.; Watson, B. T.; Huffman, J. C. J. Am.
Chem. Soc. 1987, 109, 2544. (b) Buchwald, S. L.; Nielsen,
R. B. J. Am. Chem. Soc. 1989, 111, 2870.
30. Ishikawa, T.; Ogawa, A.; Hirao, T. J. Organomet. Chem. 1999,
575, 76.
22. For a review, see Buchwald, S. L.; Nielsen, R. B. Chem. Rev.
1988, 88, 1047.
23. For a review, see Takahashi, T.; Li, Y. In Titanium and
Zirconium in Organic Synthesis; Marek, I., Ed.; Wiley-VCH:
Weinheim, 2002; pp 50–85 Chapter 2.
24. For a review, see Rosenthal, V.; Burlakov, V. V. In Titanium
and Zirconium in Organic Synthesis; Marek, I., Ed.; Wiley-
VCH: Weinheim, 2002; pp 355–389 Chapter 10.
25. Nitschke, J. R.; Zu¨rcher, S.; Tilley, D. T. J. Am. Chem. Soc.
2000, 122, 10345.
31. Ref. 21, p 753.
32. King, A. O.; Negishi, E.; Villani, Jr. F. J.; Silveira, Jr. A.
J. Org. Chem. 1978, 43, 358.