Zirconocene-mediated, high-yielding macrocyclizations of silyl-terminated diynes

Article abstract of DOI:10.1002/1521-3765(20020104)8:1<74::AID-CHEM74>3.0.CO;2-P

A series of silyl-terminated diynes of varying lengths and substitution patterns have been prepared. These diynes undergo zirconocene coupling with selective formation of trimeric macrocycles from linear alkynes, while nonlinear diynes give cyclodimeric p

Full text of DOI:10.1002/1521-3765(20020104)8:1<74::AID-CHEM74>3.0.CO;2-P

FULL PAPER
Zirconocene-Mediated, High-Yielding Macrocyclizations
of Silyl-Terminated Diynes
Laurel L. Schafer, Jonathan R. Nitschke, Shane S. H. Mao, Feng-Q. Liu, Gabriele Harder,
[
a]
Markus Haufe, and T. Don Tilley*
Abstract: A series of silyl-terminated diynes of varying lengths and substitution
patterns have been prepared. These diynes undergo zirconocene coupling with
selective formation of trimeric macrocycles from linear alkynes, while nonlinear
diynes give cyclodimeric products. The length of the linear diynes can be increased
for the preparation of macrocycles with large nanoscale cavities. Reaction of the
zirconium-containing macrocycles with acid results in the synthesis of metal-free
cyclophanes. All of these macrocycles were prepared in multigram quantities, in the
absence of high-dilution conditions, to give products in >75% yield that are easily
purified as crystalline solids.
Keywords: alkynes ¥ cyclodimeriza-
tion ¥ cyclophanes ¥ cyclotrimeriza-
tion ¥ macrocycles
in high yields. We have utilized this powerful carbon
Àcarbon
bond-forming reaction in the construction of macromolecules
with novel structures.[ The coupling of diynes bearing silyl
substituents allows for controlled regiochemistry, as it is
known that alkynyl silanes are coupled exclusively to a-silyl-
Introduction
8]
Macrocycles are important synthetic targets with applications
in a wide range of fields including host ± guest chemistry,
[
1]
[
2]
[3]
molecular sensors, supramolecular chemistry, and ion
separation.[ These cyclic molecules are typically prepared
using high dilution and/or metal templates to promote
macrocycle formation and reduce the amount of unwanted
oligomeric by-products.[ However, such strategies often
result in low yields and are not amenable to large-scale
preparations. Furthermore, isolation of the macrocyclic
products often requires laborious separations of complex
product mixtures. To address these challenges, previous
reports from our laboratories have described zirconocene
coupling as an efficient strategy in the assembly of CÀC
4]
[9]
substituted zirconacyclopentadienes. It is also known that
the zirconocene coupling of silyl-substituted alkynes is readily
reversible.[ Thus, use of silyl-substituted bis-alkynes allows
the preparation of oligomeric and macrocyclic products upon
zirconocene coupling [Eq. (1)].[ Furthermore, the revers-
ibility of alkyne coupling provides a low-energy pathway for
the conversion of oligomeric products to the thermodynamic
product, which is the smallest strain-free ring that can form.
9]
5]
6a]
Me2
Si
Me2 Cp2
Si
Zr
"Cp2Zr"
THF
bonded macrocyclic compounds in high yields,[ however the
potential of this synthetic route remains to be fully exploited.
Negishi has described a simple procedure for the generation
of zirconocene, stabilized as a 1-butene adduct, by addition of
two equivalents of n-butyllithium to a solution of [Zr(Cp) Cl ]
6]
Me
Me
Si
Me2
Si
Me2
Me
Me
n
∆
2
2
[
7]
Cp2
Zr
in THF at À788C. The coupling of alkynes with zirconocene
ꢀ1†
Me2Si
SiMe2
generated by this method produces zirconacyclopentadienes
[
a] Prof. T. D. Tilley, L. L. Schafer, J. R. Nitschke, S. S. H. Mao, F.-Q. Liu,
G. Harder, M. Haufe
Center for New Directions in Organic Synthesis
Department of Chemistry, University of California
and The Chemical Sciences Division
Me2Si
Cp2Zr
SiMe2
ZrCp2
Si
Me2
Si
Me2
Lawrence Berkeley National Laboratory
Berkeley, California, 94720 (USA)
For the efficient preparation of macrocycles we have shown
Fax : (‡1)510-642-8940
that both internally[
6a, c]
and terminally
[6b]
substituted silyl
E-mail: tilley@cchem.berkeley.edu
74
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0074 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 1

7
4±83
diynes can be used. In the case of internally substituted silyl
bis-alkynes, cyclophanes were isolated in excellent yields, in
cases where small diynes were employed. However, this
synthetic strategy was not useful for the preparation of
macrocycles with large cavities.[ In contrast, preliminary
investigations revealed that terminally substituted silyl diynes
have a greater tendency to form macrocyclic (vs. polymeric)
structures.[ Given the demonstrated potential of the zirco-
nocene coupling of silyl-terminated bis-alkynes to produce
new macrocycles in a highly efficient manner, we have
embarked upon a systematic study to explore the synthetic
possibilities and limitations of this method. This study has
involved the preparation of a series of silyl-terminated bis-
alkynes of varying lengths and alkyne substitution geometries.
Upon zirconocene coupling, the various diyne spacers allowed
the selective preparation of trimeric and dimeric macrocyclic
products in high yields. Here we discuss a variety of cyclo-
phane targets easily accessed with this efficient synthetic
strategy.
I
SnMe3
5
mol% [Pd2(dba)3]
+
Me3Si
SiMe3
6c]
20 mol% AsPh3
DMF, 70°C
4
5
4
SiMe3
I
6b]
Scheme 2. The preparation of tetraphenylene diyne 4.
pled using a modified Stille protocol[ with 1-(trimethyl-
stannyl)-4-(trimethylsilylethynyl)benzene (5)[ in dry DMF
at 708C for 4 days to give the desired tetraphenylene diyne 4.
The crude material could be purified by recrystallization from
hot toluene to give a white solid in 50% yield.
12]
13]
Zirconocene coupling of linear diynes: Diynes 1 ± 3 were used
in a general zirconocene coupling protocol for the high-yield
preparations
(
using the Negishi protocol. This procedure employs a 1:1
stoichiometry of the requisite diyne and Negishi zirconocene
in THF at À788C, with subsequent warming to room temper-
ature and further heating at 658C to generate the correspond-
ing metal-containing macrocycles. Diyne 4 was coupled
successfully using an alternative [Zr (Cp) ] reagent
(
effectively demetallated by reaction with a mild acid (such as
acetic acid or benzoic acid) to give the metal-free macrocycles
of
zirconocene-containing
macrocycles
Scheme 3).[ The zirconocene reagent was generated in situ
6]
[7a]
Results
Syntheses of linear diynes: Linear diynes of various lengths
were desirable starting materials to probe the size of macro-
cycles that could be prepared. These diynes (1 ± 3) were
synthesized using the general Pd-catalyzed coupling route
outlined in Scheme 1. The monophenylene diyne, 1,4-bis(tri-
methylsilylethynyl)benzene (1)[ and the biphenylene diyne,
II
[14]
2
Scheme 4). In all cases, the resultant macrocycles could be
10]
(
e.g., Scheme 3).
In the case of the linear, rigid diynes 1 ± 4, the macrocyclic
H
SiMe3
I
I
n
1
Me3Si
SiMe3
products 6 ± 9 are due to cyclotrimerization. With H NMR
spectroscopy, the formation of macrocyclic products is
displayed as a characteristic upfield shift of the trimethylsilyl
peak due to a shielding effect of the nearby phenyl rings. For
example, in [D ]benzene the SiMe peak of diyne 2 at d ˆ 0.27
n
[
PdCl2(PPh3)2], CuI
NEt3, THF
1
2
3
- n = 1
- n = 2
- n = 3
6
3
Scheme 1. Synthesis of diynes 1 ± 3 by the general Pd-catalyzed
shifts to d ˆ À0.26 in macrocycle 7. Also, the new Cp peak at
approximately d ˆ 6 integrates to 10H, while the trimethyl-
silyl peak integrates to 18H, and the peaks confirm the
incorporation of zirconocene in a 1:1 ratio with the spacer
unit. The fact that we obtained macrocyclic products due to
trimerization was established using a combination of mass
spectrometry analyses and X-ray crystallography (vide in-
coupling route.
4
,4'-bis(trimethylsilylethynyl)biphenyl (2),[10] have been pre-
viously reported. In a similar fashion, 4,4''-bis(trimethylsilyl-
ethynyl)terphenyl (3, n ˆ 3) was prepared using a modified
Sonogashira ± Hagahara route to couple sparingly soluble
[
11]
[6]
4
,4''-diiodoterphenyl with trimethylsilylacetylene. The final
fra).
product was more soluble than the iodinated starting material
and could be purified and isolated by recrystallization from
hot toluene in 86% yield.
Isolated macrocycle 6 was treated with acetic acid in THFat
room temperature over 12 h to give the metal-free macro-
As the length of the phenyl-
ene spacer was increased, the
Me3Si
Cp2
Zr
3
Me Si
SiMe3
Me3Si
SiMe3
decreased solubility of the halo-
[
Cp2ZrCl2]
H+
genated phenylene starting ma-
terial in the general approach
outlined in Scheme 1 became
problematic. Therefore, for
preparation of the tetraphenyl-
ene diyne (4), the modified
synthetic route given in
Scheme 2 was employed. Here,
2 equiv nBuLi
THF, -78°C
n
n
n
n
SiMe3
n
Me3Si
Cp2Zr
SiMe3
ZrCp2
SiMe3
Me3Si
n
n
SiMe3
SiMe3
SiMe3
SiMe3
1
0 n = 1
11 n = 2
2 n = 3
1
2
3
n = 1
n = 2
n = 3
6 n = 1
7 n = 2
8 n = 3
1
4,4'-diiodobiphenyl was cou-
Scheme 3. Use of diynes 1 ± 3 for the preparation of 10 ± 12.
Chem. Eur. J. 2002, 8, No. 1
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002 0947-6539/02/0801-0075 $ 17.50+.50/0
75

T. D. Tilley et al.
FULL PAPER
Me3Si
Cp2
Zr
Me3Si
SiMe3
Me3Si
SiMe3
Me3Si
Cp2Zr SiMe3
N
H+
Toluene, 65°C
SiMe3
SiMe3 Me Si
Me3Si
Cp2Zr
3
ZrCp2
SiMe3
SiMe3
SiMe3
SiMe3
SiMe3
13
4
9
II
Scheme 4. Coupling of diyne 4 by an alternative [Zr (Cp)
2
] reagent and preparation of 13.
cycle 10 in 92% yield as a colorless crystalline solid. X-ray
quality crystals of 10 were grown from toluene, and the
resulting molecular structure confirmed the trimeric nature of
the macrocycle (Figure 1). Upon demetallation the bulk
formation of macrocycles, and further characterization of 12
(demetallated 8) by mass spectrometry confirmed its cyclo-
trimeric structure (1274, [MH] ). Treatment of 9 with benzoic
‡
acid resulted in poorly soluble material that is consistent with
1
1
material no longer has a Cp peak in the H NMR spectrum,
the formation of 13 (by H NMR spectroscopy and mass
but a new signal appears, which is indicative of a vinylic
proton at d ˆ 6.42 (2H). The trimeric nature of the bulk
material was further confirmed by electron-impact mass
spectrometry), however we did not further pursue its isolation
and purification.
‡
spectrometry, in which case the molecular ion at 816 [M]
Synthesis of nonlinear diynes: In addition to varying the
length of the phenylene spacer, we were interested in probing
the effect of the geometry of the diyne unit to give macro-
was clearly identified. Thus, by inference the trimeric nature
of macrocycle 10 indicates the formation of the trimeric
zirconocene-containing macrocycle 6.
cycles of different shapes. Thus, an alternative biphenyl
Macrocycle 10 is prepared in 78% overall yield as a
crystalline material from the commercially available 1,4-
bis(trimethylsilylethynyl)-benzene. This product strongly re-
sembles the interesting and commercially available [2,2,2]-
paracyclophane that has been shown to be of interest for
cation complexation, both in the interior cavity and on the
[17]
diyne 14 was prepared from 3,3'-dibromobiphenyl
and
SiMe3
Me3Si
[15]
external faces of the arene rings. Note that [2,2,2]-para-
cyclophane is prepared in one step from commercially
available 1,4-a,a'-dibromobenzene to give a mixture of cyclic
products in <20% yield.[
16]
Macrocycles 7 and 11 have been fully characterized and
[6b]
14
15
discussed in a preliminary communication. Macrocycles 8
and 9 were characterized by 1H NMR spectroscopy and
elemental analysis. The data were consistent with the
Me3Si
SiMe3
0
trimethylsilylacetylene using a Pd catalyst in toluene at 808C
for 14 h. The crude product was purified by silica gel column
chromatography using hexanes as an eluent to give the
desired compound 14 in good yield as a white crystalline solid.
This investigation was extended to include the preparation of
an alternative terphenyl diyne 15 using the same Stille
coupling protocol outlined in the preparation of compound 4.
Pale yellow crystalline 15 was isolated in 74% yield by
removing solvent under reduced pressure and washing the
resultant material with minimal pentane.
Another strategy for varying the geometry of the diyne
3
included the incorporation of an sp hybridized center as
shown in Equation (2). Diyne 16 was prepared using a
lithium/halogen exchange reaction of 1-bromo-4-trimethyl-
silylethynylbenzene[ in a 1:1 mixture of THF/Et O with
18]
2
Figure 1. ORTEP depiction of the solid-state molecular structure of 10.
dropwise addition of 1.6m n-butyllithium at À788C. After
76
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0076 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 1

Macrocyclizations of Silyl-Terminated Diynes
74±83
stirring this mixture for 30 minutes, the resulting anion was
quenched with dichlorodimethylsilane, and the reaction
solution was warmed to room temperature over 6 h. The
crude material was isolated by removal of all volatile
materials and extraction into a 1:1 mixture of dichloro-
methane/hexanes. This solution was reduced in volume, and
the desired product was purified by silica gel column
chromatography to give 16 in excellent yield as a white
crystalline material.
lent Cp peaks at d ˆ 6.19 and 6.14, each integrating to 5H.
These inequivalent signals are proposed to arise from a
conformationally constrained, boatlike structure that has one
Cp in a pseudoequatorial orientation and the other in a
pseudoaxial position. Usually the zirconacyclopentadiene
units are coplanar in the resultant macrocycles, however in
this case, the 3,3'-alkyne substitution pattern of diyne 14
precludes this possibility. The dimeric nature of this macro-
cycle was suggested by FAB mass spectrometry with the
observation of a peak corresponding to the molecular ion at
1
7
136 m/z. The demetallated macrocycle 20 was isolated in
5% yield after treating 17 with benzoic acid in toluene at
Me3Si
Br
room temperature for 16 h. X-ray quality crystals were grown
by slow evaporation from a solution in benzene, and the
crystals thereby confirmed a dimeric structure (Figure 2).
nBuLi
1
)
THF / Et2O
-78°C
SiMe2
ꢀ2†
2
) 0.5 equiv
Me2SiCl2
Me3Si
16
Me3Si
Zirconocene coupling of nonlinear diynes: In the case of
nonlinear diynes (14 ± 16) the observed macrocyclic products
(
17 ± 19) were dimeric. As in the previous cases, the dimeric
nature of the resulting macrocyclic products was confirmed by
both mass spectrometry and X-ray crystallography of either
the Zr-containing cyclophanes or their demetallated ana-
logues.
1
Once again, an upfield shift of the H NMR resonance for
the trimethylsilyl peak to d ˆ À0.05indicated formation of a
macrocycle (17). This macrocycle displays solution-phase
behavior consistent with molecular C (rather than D2)
symmetry. In the H NMR spectrum there are two inequiva-
2
1
Figure 2. ORTEP depiction of the solid-state molecular structure of 20.
Cp2
Zr SiMe3
Macrocycle 18 was prepared
Me3Si
Cp2
Zr
using Negishi zirconocene cou-
pling conditions to give the
anticipated dimeric macrocycle
in good yield as a pale yellow
Me3Si
SiMe3
Cp2
Zr
Me3Si
SiMe3
1
powder. The H NMR spectrum
Me2Si
SiMe2
displayed a Cp signal at d ˆ
6
.21, and the SiMe signal was
3
1
7
observed at d ˆ 0.01. Yellow
X-ray quality crystals of this
compound were grown by slow
evaporation from a solution in
benzene (Figure 3). The deme-
tallated macrocycle was pre-
pared by reaction of 18 with
benzoic acid in toluene to give
macrocycle 21 in excellent
yield. The preparation of a
dimeric metal-free macrocycle
was confirmed by NMR spec-
troscopy (loss of the Cp peak
and appearance of a singlet in
the vinylic region) and elec-
tron-impact mass spectrometry,
1
9
Me3Si
Zr SiMe3
Cp2
18
Me3Si
Me3Si
Zr
SiMe3
Cp2
Me3Si
Me3Si
Zr SiMe3
Cp2
SiMe3
SiMe3
Me3Si
SiMe3
Me2Si
SiMe2
2
0
2
2
Me3Si
SiMe3
2
1
Me3Si
SiMe3
Me3Si
SiMe3
Chem. Eur. J. 2002, 8, No. 1
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0077 $ 17.50+.50/0
77

T. D. Tilley et al.
FULL PAPER
the tetraphenyl spacer permits the preparation of macrocycles
with a cavity of nanoscale dimensions in which the distance
from the center of the molecular cavity to the tetraphenylene
[
20]
spacer group is estimated to be approximately 7 ä.
As the length of the rigid-rod spacer increases, the solubility
of both the diyne precursor and resultant macrocycle de-
creases significantly. Thus, as in the case of diyne 4, the
Negishi zirconocene reagent[ was not useful, since removal
of the desirable organic products from the LiCl by-product
became problematic. This difficulty was overcome by using an
alternative zirconocene reagent (Scheme 4).[ In this case,
the inorganic salt by-products and the solubility problems
associated with the requisite low temperatures (À788C) for
Negishi zirconocene were avoided. Even with this alternative
zirconocene coupling reagent, the larger macrocycle was
isolated in a somewhat smaller yield, undoubtedly due to
isolation and purification challenges associated with reduced
solubility of the resultant macrocyclic product 9. We are
currently exploring the synthesis of very large macrocycles
from even longer spacers.[ In these cases, the synthetic
challenge is the preparation of soluble oligophenylene diyne
precursors for use in subsequent zirconocene-mediated mac-
rocyclizations.
7]
14]
Figure 3. ORTEP depiction of the solid-state molecular structure of 18.
in which case the major peak was that corresponding to the
molecular ion at 848 m/z.
21]
Finally, macrocycle 19, which was prepared from the diyne
3
with an sp hybridized center, was shown to be the product of
cyclodimerization. In this case, the zirconocene-containing
macrocycle 19, which was prepared in the standard fashion,
was not isolated and characterized, but rather was treated in
situ with HCl (6m) to give macrocycle 22 as a white solid in
The ORTEP diagram presented in Figure 1 shows the key
features associated with these cyclotrimerized products. The
phenyl substituents are not conjugated with the diene units of
the demetallated macrocycles, as they are perpendicular to
the diene moiety as a result of steric factors. The diene units of
the demetallated species are not coplanar; instead they
display a torsion angle about the CÀC single bond that
94% yield. This macrocycle was shown to be dimeric by
electron-impact mass spectrometry, which allowed observa-
tion of a molecular ion at 814 m/z.
averages 258. This torsion is necessary to accommodate both
the vinyl protons and bulky trimethylsilyl groups. These
features are observed for all trimeric macrocycles of this
type.[
Discussion
6]
Previous results from these laboratories have demonstrated
the usefulness of terminally substituted diynes as building
In contrast to the use of precursor diynes with linear alkynyl
substituents, it was observed that nonlinear diynes (14 ± 16)
enabled the isolation of products due to cyclodimerization
(17 ± 22). The macrocycles that formed from diynes 14 and 15,
were isolated in somewhat reduced yields (75± 80%) with the
other by-products being due to complex mixtures of oligo-
mers. However, because of the crystalline nature of macro-
cycles 17 and 18, we were able to easily obtain the purified
products of cyclodimerization. In the solid-state molecular
structure of 20, the macrocycle adopts a twisted conformation,
with the biphenyl units having a dihedral angle of approx-
imately 478, while the CÀC single bond of the diene unit
blocks for the preparation of macrocyclic products in excep-
tionally high yields.[
6b]
Here, we discuss the utility of
terminally substituted diynes for the preparation of macro-
cycles of various sizes. We have also investigated the effect of
precursor diyne geometry on the shape of the resulting
macrocyclic products.
The use of linear, rigid-rod spacers allows the preparation
of trimeric macrocycles of variable size. In contrast to
previously published work on internally substituted silyl
alkynes,[ the application of this strategy toward the synthesis
of increasingly larger trimeric macrocycles by increasing the
length of the precursor diyne works efficiently, and the
synthetic protocol has proven to be very general. Conse-
quently, zirconocene coupling of diynes 1 ± 4 resulted only in
cyclotrimerization products (6 ± 13). We have investigated a
range of spacer lengths from 6 to 19 ä,[ and were able to
consistently prepare the resulting macrocycles in excellent
yield (ꢁ80%) with no modification of the experimental
procedure. In contrast, for the internally substituted cases, as
the spacer became larger, an equilibrium between oligomeric
and macrocylic structures was observed, in which case dilute
solutions favored the preparation of macrocycles.[
6c]
displays a torsion angle of approximately 258. The distance
across the cavity of this small macrocycle is approximately
4.3 ä (as measured from the CÀC single bond connecting the
phenyl units of one spacer to the CÀC single bond connecting
19]
the phenyl units of the opposite side). In the solid state, 20
possesses crystallographically imposed C molecular symme-
2
try, like that observed in the solution phase of its zirconium-
1
containing precursor (17). However, the H NMR data for 20
are consistent with solution-phase C2v symmetry (as seen in
the vinyl region of the spectrum, in which only one signal at
d ˆ 6.55 is observed), which suggests enhanced conforma-
tional mobility in this compound relative to its progenitor, 17.
The zirconium-containing dimeric macrocycle 17 also dis-
6c]
Here, the range of spacer lengths has allowed the prepa-
ration of macrocycles with variable cavity sizes. For example,
78
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0078 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 1

Macrocyclizations of Silyl-Terminated Diynes
74±83
played increased air and mois-
ture sensitivity in comparison
with the cyclotrimeric macro-
cycles. These observations and
Cp2
Zr
SiMe3
N
Me Si
3
SiMe3
SiMe3
[
Cp2ZrCl2]
N
N
the somewhat reduced synthet-
2 equiv nBuLi
N
ic yield for 17 suggest the pres-
ence of more ring strain com-
N
THF, -78°C
N
N
N
SiMe3
pared with the cyclotrimeric
Me3Si
Cp Zr
N
N
ZrCp2
macrocycles previously dis-
cussed.
SiMe3
2
Me3Si
24
SiMe3
SiMe3
The effect of nonlinear
Scheme 5. Cyclotrimerization with diyne 24.
diynes on the shape of the
resultant macrocycle was also noted in the coupling of bis-
alkyne 15; as predicted, crystalline dimeric macrocycles were
obtained upon zirconocene coupling (Figure 3). The distance
across the cavity of this macrocycle is 5.6 ä with the two
protons that are pointing toward the center of the cavity being
slightly out of the plane of the zirconacyclopentadiene groups.
ence may be attributed to the preference of the bipyridyl
diyne for adopting an anti-geometry, and the alkyne units are
placed in the requisite linear arrangement for cyclotrimeriza-
tion to take place. This example represents a case, in which the
energy gained from the entropically favored formation of the
smallest macrocyclic ring is not sufficient to overcome the
preferred anti-conformation of the diyne precursor.
3
In the case of diyne 8, in which an sp hybridized Si center
was incorporated into the spacer to give a modified angle
between the nonlinear alkynyl substituents, the yields were
much improved (94%) for the isolation of macrocycle 22. In
this case demetallation was achieved with HCl (6m). These
reaction conditions are not general for demetallation, and in
particular, dimeric macrocycles 20 and 21 are known to
decompose under these harsh reaction conditions. Thus, the
observed higher yields and chemical robustness of this system
are believed to be due to the reduced ring strain of this
cyclodimeric product.
These new zirconium-containing macrocycles are also
attractive intermediates in the synthesis of a range of
functionalized macrocycles through substitution chemistry at
the zirconocene site of the zirconacyclopentadiene units.[
However, such substitution chemistry has proven difficult in
the presence of the a-silyl substituents of the zirconacyclo-
pentadienes. Current investigations are focusing on develop-
ment of general reagents to effect these desirable trans-
formations.
25]
The effect of ring strain on macrocycle formation was
further noted in our attempt to prepare monophenylene meta-
substituted macrocycle 23 [Eq. (3)]. Here, the H NMR
Conclusion
1
spectra obtained for isolated products from use of both
zirconocene reagents were consistent with complex mixtures
of oligomeric materials. In the case of the ortho-substituted
substrate it has been reported that an alkyne p-complex of
zirconocene is the isolable product.[ However, with a
methylene spacer unit the cyclodimeric macrocycle was
obtained in low yield.[
We have prepared a range of new cyclophanes from various
terminally substituted silyl bis-alkynes. The geometry of the
initial diyne has been shown to determine the shape of the
resultant macrocyclic product in a predictable fashion. Thus, it
is possible to selectively prepare cyclotrimeric products from
linear, rigid spacers, while cyclodimeric materials are accessed
with nonlinear spacers. The size of the spacer can be increased
to access large cyclophanes with nanoscale cavities. The
application of this strategy in the preparation of very large
macrocycles has been shown to proceed with no appreciable
loss in yield for the larger substrates. The product yields for
this synthetic methodology are remarkably high in all the
cases presented, with a slight reduction in yield being
observed when the resultant macrocyclic products are some-
what strained. The Negishi protocol for generating zircono-
22]
23]
Cp2
Zr
SiMe3
Me3Si
SiMe3
"
Cp2Zr"
ꢀ
3†
23
Me3Si
SiMe3
Zr
SiMe3
Cp2
[7]
cene in situ is preferred for the preparation of these
unfunctionalized macrocycles as a result of the ease of
generation of the zirconocene. However, in cases, in which
substrate and/or product solubility is low, the alternative
zirconocene reagent (Rosenthal×s complex) has proven to be
the reagent of choice.[
All of these macrocycles can be prepared on a multigram
scale, in the absence of high dilution conditions, and are
isolated and purified as crystalline solids. Consequently, this
synthetic method provides a simple and versatile method for
the preparation of cyclophanes of different sizes and shapes
by simple, strategic modifications of the diyne precursor.
All the results presented are consistent with the fact that
the shape and size of the resultant macrocycle can be
predicted based on the geometry of the precursor diyne,
given that these reaction conditions favor formation of the
smallest strain-free ring that can form. However, it should be
noted that functionality can be an important factor in
determining the structure of the macrocyclic products.[ In
contrast to the cyclodimeric product formed upon zircono-
cene coupling of diyne 6, its bipyridine analogue (24) yields
the product of cyclotrimerization (Scheme 5).[ This differ-
14]
24]
24]
Chem. Eur. J. 2002, 8, No. 1
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0079 $ 17.50+.50/0
79

T. D. Tilley et al.
FULL PAPER
The THF was then removed in vacuo, and distilled water (90 mL) was
added to dissolve the by-product LiCl. The yellow suspension was stirred
vigorously for 1 h. The solid product was collected using a Hirsch funnel.
The water wash was repeated, and the (highly insoluble) product was then
Experimental Section
All manipulations were performed under an inert atmosphere of nitrogen
using either standard Schlenk techniques or a Vacuum Atmospheres glove
box. Dry, oxygen-free solvents were employed throughout. All solvents
washed similarly with Et
to ensure that the isolated product was finely divided, as chunks of material
can occlude H O. The yellow microcrystalline product was dried under
vacuum to give of the desired material (3.05g, 85% yield).
2
O (90 mL) and toluene (90 mL). Care was taken
6
were distilled from sodium/benzophenone ketyl. [D ]Benzene was purified
2
by vacuum distillation from Na/K alloy. Elemental analyses were per-
formed by the UCB Microanalytical Laboratory or Desert Analytics. NMR
spectra were recorded on a BrukerAMX (300 MHz), a Varian Bruker-
AMX (400 MHz), or a BrukerDRX (500 MHz) spectrometer at 258C and
1H NMR (300 MHz, C
): d ˆ 6.54 (s, 2H; C ), 6.19 (s, 5H; C
À0.06 (s, 9H; Si(CH ); elemental analysis calcd (%) for C78
(1474.2): C 63.48, H 6.56; found: C 63.42, H 6.18.
Trimeric macrocycle (7): A round-bottom flask (250 mL) was charged with
Zr(Cp) Cl ] (2.3 g, 8.0 mmol), 4,4'-bis(trimethylsilylethynyl)biphenyl (2)
D
H
),
Zr
6
6
4
H
4
5
5
3
)
3
H
96Si
6
3
1
3
referenced to residual solvent resonances. C spectra were proton
decoupled. Mass spectrometry was carried out on a VGProSpec or a
VGZAB using electron-impact or FAB sources. Zirconocene dichloride
[
2
2
(
4
Strem), n-butyllithium solution (Aldrich), 4,4'-dibromobiphenyl (Aldrich),
,4'-diiodobiphenyl (Aldrich), and trimethylsilylacetylene (GFS) were used
(
2.8 g, 8.0 mmol), and THF (150 mL). The flask was cooled to À788C, and
a solution of nBuLi in hexanes (1.6m, 10 mL, 16 mmol) was added
dropwise. The reaction mixture was warmed to room temperature then
stirred for 16 h, and then heated to reflux for a further 24 h. All volatile
materials were removed in vacuo, and the residue was washed with hexanes
[
10]
as received. The compounds 1,4-bis(trimethylsilylethynyl)benzene 1,
[
10]
4
,4'-bis(trimethylsilylethynyl)biphenyl 2,
1-bromo-4-(trimethylsilylethyn-
4-(trimethylsilylethynyl)-1-(tri-
[
13]
[12]
yl)benzene,
4,4''-diiodoterphenyl,
methylstannyl)benzene 5,[ and 3,3'-dibromobiphenyl[17] were prepared
13]
(
3 Â 40 mL). The hexane-insoluble residue was then extracted with hot
by literature procedures.
toluene (5 Â 40 mL), and the toluene extracts were combined and cooled to
4
,4''-Bis(trimethylsilylethynyl)terphenyl (3): A PTFE-stoppered Schlenk
À408C to afford 7 as yellow crystals (4.1 g, 90%).
[
12]
flask (1 L) was charged with 1,4''-diiodoterphenyl (30.0 g, 62.2 mmol),
Pd(PPh ] (0.180 g, 0.156 mmol), and CuI (0.0593 g, 0.311 mmol). iPr NH
100 mL) and dry THF (400 mL) were added to these reagents, followed by
1
H NMR (300 MHz, C
6
D
6
): d ˆ 7.16 (d, J(H,H) ˆ 8.1 Hz, 4H; ArH), 6.71
[
(
3
)
4
2
(
d, J(H,H) ˆ 8.1 Hz, 4H; ArH), 6.13 (s, 10H; C
5
H
5
), À0.26 (s, 18H;
(Si(CH ), 149.4
), 3.00
3
(1702.2): C 67.66, H
13
SiCH
(
(
6
3
); C NMR (100.6 MHz, C
6
D
6
): d ˆ 206.1 (ZrC
4
3 3 2
) )
H
5 5
trimethylsilylacetylene (17.0 mL, 131 mmol). The reaction was stirred
overnight at 508C, and the solvents were removed in vacuo. The product
mixture was suspended in toluene (2 L), filtered on a Buchner funnel, and
passed through a plug of silica (100 mL) that was washed with additional
toluene (200 mL). The resulting light yellow solution was reduced in
volume to 100 mL, and the cream-colored solid was collected by filtration.
The crude product was recrystallized from hot toluene to give white crystals
ZrC
SiCH
.39; found: C 66.93, H 6.36.
4
(Si(CH
3
)
3
)
2
), 142.7, 134.2, 128.6, 128.2 (ArC), 111.4 (C
3
); elemental analysis calcd (%) for C96
H
108Si
6
Zr
Trimeric macrocycle (8): A PTFE-stoppered Schlenk flask (100 mL) was
charged with [Zr(Cp) Cl ] (0.692 g, 2.37 mmol) that was dissolved in dry
2
2
THF (20 mL). A separate Schlenk flask (100 mL) was loaded with 4,4''-
bis(trimethylsilylethynyl)terphenyl (3) (1.00 g, 2.37 mmol), to which was
(
22.5g, 86% yield).
added THF (50 mL). The [Zr(Cp)
2
Cl
2
] flask was cooled to À788C, and a
1H NMR (500 MHz, CDCl
): d ˆ 7.68 (s, 2H; ArH), 7.59 (d, J(H,H) ˆ
3
solution of nBuLi in hexane (1.6m, 2.96 mL, 4.73 mmol) was added
dropwise. The reaction was stirred at À788C for 20 min, and then the 4,4''-
bis(trimethylsilylethynyl)terphenyl solution was added dropwise by can-
nula. The reaction mixture was warmed to room temperature, and heated
to 608C for 12 h. The THF was then removed in vacuo, and distilled water
(90 mL) was added to dissolve the by-product LiCl. The yellow suspension
was stirred vigorously for 1 h. Then the solid product was collected using a
Hirsch funnel. The water wash was then repeated, and the (highly
8
.5Hz, 2H; ArH), 7. 56 (d, J(H,H) ˆ 8.5Hz, 2H; ArH), 0.29 (s, 9H;
13
Si(CH
1
for C28
3
)
3
); C NMR (125MHz, CDCl
3
): d ˆ 140.7, 139.8, 132.7, 127.6, 126.9,
); elemental analysis calcd (%)
(422.2): C 79.56, H 7.15; found: C 79.76, H 7.06.
22.4 (ArC), 105.1, 95.3 (CꢂC), 0.2 (SiCH
3
H
30Si
2
4
,4'''-Bis(trimethylsilylethynyl)tetraphenyl (4): In a Schlenk flask (100 mL)
equipped with a stir bar, [Pd (dba) ] (0.340 g, 0.371 mmol) and AsPh
0.454 g, 1.48 mmol) were loaded in the glove box. In a separate Schlenk
2
3
2
(
[
13]
flask (100 mL) Me
3
SnC
6
H
4
CꢂCSiMe
3
5
(2.50 g, 7.42 mmol) and 4,4'-
insoluble) product was washed similarly with Et O (50 mL), and toluene
2
diiodobiphenyl (1.00 g, 2.46 mmol) were loaded. This Schlenk flask was
evacuated and backfilled before suspending the organics in dry DMF
(50 mL). Care was taken to ensure that the isolated product was finely
divided, as chunks of material can occlude H O. The product was dried
2
(
30 mL). This mixture was transferred by cannula to the reaction flask
under vacuum to give yellow microcrystalline material (1.39 g, 91%).
containing the catalyst mixture. The flask with the organic reagents was
washed with dry DMF (20 mL), and this solution was transferred to the
reaction flask as well. This flask was then sealed and heated to 708C with
stirring for 4 days, and a dark gray suspension of white precipitate resulted.
The precipitate was collected by vacuum filtration on a Buchner funnel and
redissolved into boiling toluene (600 mL). The solution was then filtered
hot, and the collected precipitate was washed with hot toluene (2 Â
1H NMR (300 MHz, CD
Cl
): d ˆ 7.55 (s, 2H; ArH), 7.22 (d, J(H,H) ˆ
),
3 3
) ); No C NMR data were obtained because of poor
2
2
8
.5Hz, 2H; ArH), 6.76 (d, J(H,H) ˆ 8.5Hz, 2H; ArH), 6.39 (s, 10H; C
5 5
H
1
3
À0.29 (s, 9H; Si(CH
solubility; elemental analysis calcd (%) for C114
H 6.26; found: C 70.72, H 6.04.
6 3
H120Si Zr (1930.2): C 70.86,
Trimeric macrocycle (9): A Schlenk flask (50 mL), equipped with stir bar,
was charged with crystalline [Zr(Cp)
[14]
(pyr)(Me
3
SiC
ꢂCSiMe )] (94 mg,
3
1
00 mL). The resulting yellow solution was concentrated to 600 mL, at
2
which point a hazy solution developed. The heating was stopped, and the
product was permitted to crystallize overnight. The resulting white waxy,
flaky material was collected by filtration (0.600 g, 50% yield). This
compound required no further purification.
0.20 mmol) and diyne 4 (100 mg, 0.200 mmol). The mixture was suspended
in toluene (15mL). This solution was heated to 65 8C for 36 h to give a pale
yellow solution. The solvent was removed under vacuum to give a dark
yellow powder that was further purified by recrystallization from benzene,
and this afforded pale yellow crystals (119 mg, 82% yield).
1H NMR (300 MHz, CDCl
H; ArH), 2.72 (s, 18H; SiCH
poor solubility; MS (70 eV): m/z (%): 498 [M] , 483 [M À CH
analysis calcd (%) for C34 (498.2): C 81.87, H 6.87; found: C 81.74, H
.74.
Trimeric macrocycle (6): A PTFE-stoppered Schlenk flask (100 mL) was
charged with [Zr(Cp) Cl ] (2.16 g, 7.39 mmol) and dissolved in dry THF
20 mL). separate Schlenk flask (100 mL) was loaded with 1,4-
bis(trimethylsilylethynyl)benzene (1) (2.00 g, 7.39 mmol), to which THF
20 mL) was added. The [Zr(Cp) Cl
): d ˆ 7.74 ± 7.67 (m, 8H; ArH), 7.62 ± 7.54 (m,
3
13
1H NMR (300 MHz, C
8
3
); No C data could be collected because of
6
D
6
): d ˆ 7.36 (m, 12H; ArH), 6.93 (d, J(H,H) ˆ
‡
‡
13
] ; elemental
8 Hz, 4H; ArH), 6.23 (s, 10H; C
5
H
5
), À0.06 (s, 18H; SiCH
3
); No C NMR
3
H
34Si
2
data were obtained because of poor solubility; elemental analysis calcd (%)
Zr (2160.8): C 73.31, H 6.11; found: C 73.70, H 6.47.
6
for C132
H
132Si
6
3
Trimeric cyclophane (10): A Schlenk flask (100 mL) was charged with
monophenylene trimer (6) (0.500 g, 1.02 mmol), and it was then suspended
in dry THF (50 mL). Glacial acetic acid (1.16 mL, 20.3 mmol) was added to
this yellow suspension, and the mixture was stirred for 12 h at room
temperature. Over this time, the color gradually became white, and all
solids dissolved. The THF and excess acetic acid were removed in vacuo,
and the residue was dissolved in toluene (100 mL). The slightly hazy,
colorless solution was passed through a plug of silica (20 mL) and washed
through with toluene (40 mL). The toluene was removed in vacuo, and the
2
2
(
A
(
2
2
] flask was cooled to À788C, and a
solution of nBuLi in hexane (2.46m, 6.0 mL, 14.78 mmol) was added
dropwise. The reaction was stirred at À788C for 20 min, and then the 1,4-
bis(trimethylsilylethynyl)benzene solution was added dropwise. The reac-
tion mixture was warmed to room temperature and heated to 608C for 12 h.
80
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0080 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 1

Macrocyclizations of Silyl-Terminated Diynes
74±83
product was washed with hexanes (20 mL). The yield of tiny colorless
crystals was 0.255 g, 92%.
this resulted in a greenish/yellow suspension. The reaction mixture was
cooled and diluted with toluene (50 mL). The organic fraction was washed
1
with NaCO
3
(1m, 50 mL), followed by water (3 Â 50 mL), and finally brine
H NMR (500 MHz, C
6
D
6
): d ˆ 6.81 (s, 2H; ArH), 6.42 (s, 1H; HCˆC), 0.09
1
3
(50 mL). The solvents were removed in vacuo, and the resultant light brown
oil was redissolved in hexanes (100 mL) and filtered through a silica plug
(30 mL). After the removal of hexanes under vacuum a pale yellow
crystalline solid was isolated in 74% yield (0.42 g). This material was then
washed with minimal pentane to further purify.
(
(
s, 9H; Si(CH
3
)
3
); C NMR (125MHz, C
6
D
6
): d ˆ 164.4, 140.7, 130.6, 129.4,
‡
ArC and CˆC), 1.1 (Si(CH
)
); MS (70 eV): m/z (%): 816 [M] , 801 [M À
3
‡
3
‡
CH
(
3
] , 743 [M À Si(CH
) ] ; elemental analysis calcd (%) for C48H72Si
3 3 6
816.5): C 70.51, H 8.88; found: C 70.47, H 9.00.
Trimeric cyclophane (11): A round-bottom flask (100 mL) was charged
with macrocycle 7 (0.95g, 1.9 mmol) and THF (20 mL). An aqueous HCl
solution (6m, 20 mL) was added to the flask, and the resulting mixture was
stirred for 2 h at room temperature. The reaction mixture was poured into
methanol (200 mL), and the white powder 11 was collected (0.52 g, 98%).
Crystals were obtained by cooling of a saturated solution (ca. 0.5g/ 50 mL)
in toluene/acetonitrile (ca. 3:1) to À108C.
1H NMR (300 MHz, CDCl
ArH), 0.26 (s, 18H, SiCH
): d ˆ 7.76 (s, 1H; ArH), 7.60 ± 7.52 (m, 11H;
13
3
3
); C NMR (125MHz, CDCl
3
): d ˆ 141.0, 140.9,
132.4, 129.4, 126.9, 126.3, 125.8, 122.2 (ArC), 104.9, 95.0 (CꢂC), 0.0
‡
‡
3
] ; elemental
(SiCH
analysis calcd (%) for C28
3
); MS (70 eV): m/z (%): 422 [M] , 407 [M À CH
2
H30Si (394.1): C 79.56, H 7.15; found: C 79.54, H
7.11.
1
H NMR (300 MHz, C
)H ), 0.03 (s, 18H; SiCH
(SiMe ), 140.8 (C (SiMe
11.4 (C ), 0.43 (SiCH ); MS (70 eV): m/z (%): 1045[ M] ; elemental
analysis calcd (%) for C66 (1044.5): C 74.79, H 8.10; found: C 73.52, H
.43.
6
D
6
): d ˆ 7.05(s, 8H; ArH), 6. 54 (s, 2H; C
4
(Si-
): d ˆ 162.6
3 2 2
) H ), 138.0, 130.5, 130.0, 125.9 (ArC),
Diyne (16): A round-bottom flask (250 mL) was charged with 1-bromo-4-
trimethylsilylethynyl)benzene (6.00 g, 23.7 mmol) and a mixture of THF/
diethyl ether (1:1, 120 mL). The flask was cooled to À788C, and then nBuLi
1.6m/hexane, 14.8 mL, 23.7 mmol) was added dropwise. The mixture was
SiCl (1.10 mL, 12.0 mmol) was then
1
3
Me
(
1
3
2
3 6 6
); C NMR (100.6 MHz, C D
(
C
4
3
)
2
H
2
4
‡
5
H
5
3
(
H
84Si
6
stirred for 30 min at À788C. Me
2
2
8
added dropwise, and the temperature of the flask was raised to room
temperature over 6 h. The volatile materials were removed in vacuo, and
the residue was extracted with dichloromethane/hexanes (1:1, 120 mL).
The solution was concentrated to 20 mL and then passed through a silica
gel (15 Â 4.2 cm) column using a mixture of dichloromethane/hexane (1:7)
as an eluting solvent. The product was obtained as a crystalline solid in
89% yield (4.30 g) after removal of solvents.
Trimeric cyclophane (12): A Schlenk flask (100 mL) was charged with
trimeric macrocycle 8 (0.236 g, 0.366 mmol) and dry benzene (50 mL).
Glacial acetic acid (0.2 mL, 3.6 mmol) was added to this yellow suspension,
and the mixture was vigorously stirred for 24 h at room temperature. Over
this time the yellow suspension gradually became white. The benzene and
excess acetic acid were removed in vacuo, and the residue was dissolved in
toluene (100 mL). The slightly hazy colorless solution was passed through a
plug of silica (10 mL) and washed through with toluene (100 mL). The
toluene was removed in vacuo, and the product was washed with methanol
1H NMR (400 MHz, C
J(H,H) ˆ 8.4 Hz, 4H; ArH), 0.27 (s, 6H; Si(CH
D
6
): d ˆ 7.47 (d, J(H,H) ˆ 8.4 Hz, 4H; ArH), 7.21 (d,
), 0.19 (s, 18H;
): d ˆ 136.0, 135.4, 131.0
ArC), 125.1, 105.1, 96.7 (CꢂC), À0.1 (CꢂCSi(CH
), À2.5(SiCH );
elemental analysis calcd (%) for C24 (404.3): C 71.22, H 7.97; found: C
1.08, H 7.74.
6
3 2
)
1
3
CꢂCSi(CH
3 3 6 6
) ); C NMR (100.6 MHz, C D
(
20 mL) to give colorless microcrystals (0.14 g, 90%).
1H NMR (500 MHz, C
): d ˆ 7.29 (d, J(H,H) ˆ 8.5Hz, 2H; ArH), 7.28 (s,
H; ArH), 7.22 (d, J(H,H) ˆ 8.5Hz, 2H; ArH), 6. 59 (s, 1H; HC ˆC), 0.09 (s,
(
3
)
3
3
6
D
6
H
32Si
3
2
9
1
7
1
3
H; Si(CH
3
)
3
); C NMR (125MHz, C
6
D
6
): d ˆ 163.5, 141.3, 139.7, 130.6,
Dimeric macrocycle (17): A Schlenk flask (100 mL), equipped with stir bar,
was charged with [Zr(Cp) Cl ] (0.845g, 2.89 mmol) dissolved in dry THF
29.7, 128.9, 127.7, 126.4 (ArC and CˆC), 0.8 (Si(CH
)
); MS (70 eV): m/z
3
3
‡
‡
‡
(
%): 1274 [MH] , 1260 [M À CH
3
] , 1202 [M À Si(CH
3
)
3
] ; elemental
2
2
(
20 mL). In a separate Schlenk flask diyne 14 (1.00 g, 2.89 mmol) was
analysis calcd (%) for C84
7
H
96Si
6
(1272.5): C 79.18, H 7.59; found: C 78.96, H
dissolved in dry THF (10 mL). The zirconocene dichloride solution was
.39.
cooled to À788C before the addition of a solution of nBuLi in hexane
3
,3'-Bis(trimethylsilylethynyl)biphenyl (14): A PTFE-stoppered Schlenk
(
1.6m, 3.61 mL, 5.9 mmol). This reaction mixture was stirred at low
flask (500 mL) was charged with copper(i) iodide (26.0 mg, 0.135mmol)
and [Pd(PPh ] (104 mg, 0.0898 mmol) in the glove box. In a separate
temperature for 15minutes before addition of the diyne solution by
cannula. The reaction mixture was slowly warmed to room temperature
with stirring and then heated to 608C for a further 12 h. All volatile
materials were removed in vacuo, and the residue was extracted with
toluene (30 mL), and the supernatant was removed from the insoluble salts
by cannula filter. The solution in toluene was then evaporated to dryness,
and the crude material was isolated as a dark yellow solid material. This
material was purified by washing with minimum amounts of hexanes (2 Â
3 4
)
[
17]
Schlenk flask 3,3'-dibromobiphenyl (1.4 g, 4.49 mmol) was dissolved in
toluene (20 mL) and transferred by cannula to the reaction flask containing
the catalyst mixture. Degassed diisopropylamine (15mL) was added by
cannula to this suspension. The reaction mixture was stirred for 15minutes,
after which it was transparent. Finally, trimethylsilylacetylene (3.20 mL,
2
.20 g, 22.4 mmol) was added by syringe. The reaction mixture was stirred
for 14 h at 808C. The resulting mixture was cooled and filtered by gravity
before removal of the volatile material under vacuum. The resulting solid
was suspended in hexanes (60 mL) and filtered through a silica plug. The
silica plug was washed with hexanes (3 Â 30 mL), and the combined
filtrates were dried under vacuum and purified by silica gel column
chromatography using hexanes as a eluent. The resulting white crystalline
material was isolated in 70% yield (1.1 g).
3
mL) to give a bright yellow solid material (1.2 g, 1.1 mmol) which could
be recrystallized from a concentrated solution in toluene (73% yield).
1H NMR (300 MHz, C
): d ˆ 6.89 (t, J(H,H) ˆ 9 Hz, 2H; ArH), 6.82 (s,
2H; ArH), 6.57 (d, J(H,H) ˆ 6 Hz, 2H; ArH), 6.19 (s, 5H; CpH), 6.17 (d,
);
D
6
6
J(H,H) ˆ 3 Hz, 2H; ArH), 6.14 (s, 5H; CpH), À0.05(s, 18H; Si(CH
3 3
)
1
3
C data were not obtained because of low solubility; MS (70 eV): m/z (%):
‡
1H NMR (400 MHz, CDCl
(
1136 [M] ; HRMS calcd: 1133.2820; found: 1133.2823; elemental analysis
3
): d ˆ 7.70 (s, 2H; ArH), 7.54 (m, 2H; ArH), 7.41
13
64 72 4 2
calcd (%) for C H Si Zr (1134.8): C 67.66, H 6.39; found: C 67.66, H 6.63.
m, 2H; ArH), 7.30 (m, 2H; ArH), 0.28 (s, 18H; Si(CH
3
)
3
); C NMR
(
1
100.6 MHz, CDCl
3
): d ˆ 140.4, 131.0, 130.7, 128.8, 127.3, 123.7 (ArC),
Dimeric macrocycle (18): A Schlenk flask (20 mL), equipped with stir bar,
was charged with [Zr(Cp) Cl ] (51.9 mg, 0.177 mmol) in the dry box. This
‡
04.9, 94.5(C ꢂC), À0.1 (Si(CH
)
3
); MS (70 eV) m/z (%): 346 [M] , 331
3
2
2
‡
[
M À CH
.56; found: C 75.96, H 7.22.
Diyne (15): A Schlenk flask (100 mL), equipped with stir bar, was charged
with [Pd (dba) ] (0.124 g, 0.135mmol) and AsPh (0.165g, 0. 54 0 mmol) in
3
] ; elemental analysis calcd (%) for C22H26Si (346.2): C 76.23, H
2
material was dissolved in THF (5mL) and then cooled to À788C before
the addition of a solution of nBuLi (2.5m, 0.14 mL, 0.35mmol) in hexanes.
This solution was stirred for 15minutes at À788C, before the cannula
addition of a solution of diyne 15 (75mg, 0.177 mmol) in THF (5mL) from
a separate Schlenk flask. The resulting reaction mixture was stirred for 4 h
while slowly warming to room temperature. Then the solution was heated
to 658C for a further 12 h. The deep orange solution was evaporated to
dryness, and the desired product was redissolved in toluene and filtered
away from the insoluble material by cannula filtration. This solution was
evaporated to dryness to isolate the crude product, which was further
purified by washing with dry hexanes to give a pale yellow powder (97 mg,
0.0753 mmol, 85% yield). X-ray quality crystals were grown by slow
evaporation of a solution in benzene.
7
2
3
3
the glove box. In a separate Schlenk flask (100 mL), 4-(trimethylsilyl-
ethynyl)-1-(trimethylstannyl)benzene (1.00 g, 2.97 mmol) was loaded. The
compound 1,3-dibromobenzene (0.163 mL, 0.318 g, 1.35mmol) was added
to this Schlenk flask by syringe. This mixture of organic compounds was
2
evacuated and backfilled with N three times before being dissolved in dry
DMF (20 mL). This solution of organic reagents was added to the Schlenk
flask containing the Pd catalyst by cannula. The Schlenk flask containing
the organic reagents was rinsed with dry DMF (15mL) and added to the
catalyst suspension. The dark suspension was stirred at 658C for 3 days, and
Chem. Eur. J. 2002, 8, No. 1
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0081 $ 17.50+.50/0
81

T. D. Tilley et al.
FULL PAPER
1H NMR (400 MHz, C
obscured by solvent signal, ArH), 7.04 (t, J(H,H) ˆ 8 Hz, 1H; ArH), 6.79
d, J(H,H) ˆ 8 Hz, 4H; ArH), 6.21 (s, 10H; CpH), 0.01 (s, 18H; Si(CH );
C NMR data were not obtained because of poor solubility; repeated
attempts to obtain satisfactory analytical data of crystalline material were
not successful. The X-ray crystal structure (Figure 3) and further reactivity
to compound 21 confirmed its identity.
D
): d ˆ 7.33 (brs, 1H; ArH), 7.20 ± 7.13 (m,
16363 measured reflections, 5790 were unique, and 2509 were observed
with [I > 3.00s(I)], data corrected for Lorentz and polarization effects,
analyzed for agreement and absorption using XPREP with an empirical
6
6
(
3 3
)
1
3
absorption correction using SADABS (Tmax ˆ 0.96,
T
min ˆ 0.74). The
structure was solved by direct methods (SIR92) and developed by least-
2
squares refinement against jF j . No. of parameters, 220; H atoms were
À
À3
calculated but not refined; R ˆ 0.076 and R
w
ˆ 0.088; D1max ˆ 0.80 e ä
.
Dimeric cyclophane (20): In a Schlenk flask (25mL), equipped with stir
bar, macrocycle 17 (200 mg, 0.176 mmol) and benzoic acid (1.00 g,
Crystal data for 20 (C44
H
56Si
4
): crystal dimensions 0.10 Â 0.25 Â 0.35mm,
≈
triclinic, space group P1, a ˆ 6.8221(6), b ˆ 11.775(1), c ˆ 14.192(1) ä, b ˆ
3
À3
8
.20 mmol) were dissolved in dry toluene (7 mL). The reaction mixture
was stirred at room temperature overnight, then diluted with toluene
30 mL) before filtering to remove precipitates. The filtrate was washed
with a Na CO
solution (2m, 2 Â 30 mL) and brine (30 mL). The organic
fraction was dried over MgSO , then evaporated to dryness to give the
101.088(1)8; Vˆ 10669(1) ä , 1calcd ˆ 1.085gcm , 2qmax 46.58, MoKa radi-
ation (l ˆ 0.71069 ä), wscans, Tˆ À145.08C, of 5118 measured reflections,
3026 were unique, and 2565 were observed with [I > 3.00s(I)], data
corrected for Lorentz and polarization effects, analyzed for agreement and
absorption using XPREP with an empirical absorption correction using
SADABS (Tmax ˆ 0.96, Tmin ˆ 0.87). The structure was solved by direct
(
2
3
4
crude product as a pale yellow powder (90 mg, 76% yield). This material
was further purified by recrystallization from a solution in benzene with
slow evaporation to give colorless blocklike crystals.
2
methods (SIR92) and developed by least-squares refinement against jF j .
No. of parameters, 217; H atoms were included but not refined; R ˆ 0.035
À
À3
1H NMR (300 MHz, CDCl
and R
w
ˆ 0.050; D1max ˆ 0.21 e ä
.
3
): d ˆ 7.23 (s, 2H; ArH), 7.13 (m, signal obscured
by solvent peak, ArH), 7.05(d, J(H,H) ˆ 8 Hz, 2H; ArH), 6.91 (t, J(H,H) ˆ
Crystallographic data (excluding structure factors) for the structures
reported in this paper have been deposited with the Cambridge Crystallo-
graphic Data Centre as supplementary publication nos. CCDC-162175,
162176, and 162177. Copies of the data can be obtained free of charge on
application to CCDC, 12 Union Road, Cambridge, CB21EZ, UK (fax:
1
3
8
Hz, 2H; ArH), 6.55 (s, 2H; CˆCH), 0.05(s, 18H; SiCH
3
); C NMR data
were not obtained because of poor solubility; MS (70 eV): m/z (%): 696
[
‡
M] ; elemental analysis calcd (%) for C44
H
56Si
4
(696.3): C 75.79, H 8.10;
found: C 75.55, H 8.26.
(
‡44)1223-336-003; e-mail: deposit@ccdc.cam.ac.uk).
Dimeric cyclophane (21): In a Schlenk flask (25mL), equipped with stir
bar, macrocycle 18 (75mg, 0.0 58 mmol) and benzoic acid (71 mg,
0
.58 mmol) were dissolved in dry toluene (5 mL). The reaction mixture
was stirred at room temperature overnight, and then diluted with toluene
20 mL) before filtering to remove precipitates. The supernatant was
washed with Na CO
solution (2m, 2 Â 30 mL) and brine (30 mL). The
organic fraction was dried over MgSO , and then evaporated to dryness to
give the crude product (40 mg, 80% yield). The product was further
purified by recrystallization from a solution in hexanes (5mL) to give
colorless needlelike crystals (30 mg, 61% yield).
Acknowledgements
(
2
3
We thank the National Science Foundation and the Director, Office of
Basic Energy Sciences, Chemical Sciences Division, and U.S. Department
of Energy under Contract No. DE-AC03-76SF00098, for partial support of
this work. L.L.S. thanks NSERC (Canada) for a Postdoctoral Fellowship.
The Center for New Directions in Organic Synthesis is supported by
Bristol-Myers Squibb as Sponsoring Member. We also thank Dr. F. J.
Hollander, Dr. Allen Oliver, and Dr. Dana Caulder of the U.C. Berkeley
X-ray diffraction facility (CHEXRAY) for determinations of the X-ray
structures.
4
1H NMR (300 MHz, CDCl
): d ˆ 7.42 (m, 3H; ArH), 7.21 (d, J(H,H) ˆ
3
9
Hz, 4H; ArH), 7.19 (s, 1H; ArH), 6.93 (d, J(H,H) ˆ 9 Hz, 4H; ArH), 6.31
‡
(
s, 2H; CHSiMe
3
), À0.06 (s, 18H; SiCH
3
); MS (70 eV): m/z (%): 848 [M] ;
elemental analysis calcd (%) for C56H64Si (848.4): C 79.18, H 7.59; found: C
4
7
9.94, H 7.64.
Dimeric cyclophane (22): A Schlenk tube (250 mL) was charged with
Zr(Cp) Cl ] (0.68 g, 2.32 mmol), diyne 16 (0.937 g, 2.32 mmol), and THF
[
2
2
[
1] a) V. R¸diger, H.-J. Schneider, Chem. Eur. J. 2000, 6, 3771 ± 3776; b) J.
B¸gler, N. A. J. M. Sommerdijk, A. J. W. G. Visser, A. van Hoek,
R. J. M. Nolte, J. R. J. Engbersen, D. N. Reinhoudt, J. Am. Chem. Soc.
(
2
80 mL). The flask was cooled to À788C, and then nBuLi (1.6m/hexane,
.90 mL, 4.64 mmol) was added dropwise. The flask was allowed to warm to
room temperature over 7 h and was heated for 1 h at 658C. An aqueous
HCl solution (6m, 20 mL) was added to the flask, and then the resulting
mixture was stirred for 2 h at room temperature. The reaction mixture was
poured into methanol (150 mL), and the white precipitate was collected
and purified by silica gel chromatography (eluted with hexane/ether, 100:1)
to afford 0.89 g (94%) of 22 as a white solid.
1
1
999, 121, 28 ± 33; c) M. F. Hawthorne, Z. Zheng, Acc. Chem. Res.
997, 30, 267 ± 276; d) C. Mueller, J. A. Whiteford, P. J. Stang, J. Am.
Chem. Soc. 1998, 120, 9827 ± 9837; e) B. Dietrich, P. Viout, J.-M. Lehn,
Macrocyclic Chemistry: Aspects of Organic and Inorganic Supra-
molecular Chemistry, VCH, New York, 1993, and references therein.
2] a) P. D. Beer, P. A. Gale, Angew. Chem. 2001, 113, 502 ± 532; Angew.
Chem. Int. Ed. 2001, 40, 486 ± 516; b) K. S. Bang, M. B. Nielsen, R.
Zubarev, J. Becher, Chem. Commun. 2000, 215± 216; c) G. Ulrich, P.
Turek, R. Ziessel, A. De Cian, J. Fischer, Chem. Commun. 1996,
[
[
1H NMR (300 MHz, CDCl
J(H,H) ˆ 8 Hz, 8H; ArH), 6.07 (s, 4H; CˆCH), 0.46 (s, 12H; Si(CH
): d ˆ 7.12 (d, J(H,H) ˆ 8 Hz, 8H; ArH), 6.77 (d,
),
): d ˆ 163.6
CˆCHSi), 141.9, 136.9, 132.9 (ArC), 129.2 (CˆC), 128.6 (ArC), 0.37
3
3 2
)
1
3
À0.15(s, 36H; Si(CH
3 3 3
) ); C NMR (75.5 MHz, CDCl
(
(
2
461 ± 2462, and references therein.
‡
Si(CH
3
)
3
), À2.62 (Si(CH
3
)
2
); MS (70 eV): m/z (%): 814 [M] ; elemental
3] a) A. R. Pease, J. O. Jeppesen, J. F. Stoddart, Y. Luo, C. P. Collier, J. R.
Heath, Acc. Chem. Res. 2001, 34, 433 ± 444; b) K. Campbell, R.
McDonald, N. R. Branda, R. R. Tykwinski, Org. Lett. 2001, 1045±
analysis calcd (%) for C48
.27.
Crystal data for 10 (C48
H68Si (813.1): C 70.86, H 8.42; found: C 70.48, H
6
8
H
≈
72Si
6
): crystal dimensions 0.20 Â 0.10 Â 0.22 mm,
1
1
048; c) S. Hˆger, A.-D. Meckenstock, Chem. Eur. J. 1999, 5, 1686 ±
691; d) M. Fujita, Chem. Soc. Rev. 1998, 27, 417 ± 425; e) M.
Fernandez-Saiz, H.-J. Schneider, J. Sartorium, W. D. Wilson, J. Am.
Chem. Soc. 1996, 118, 4739 ± 4745, and references therein.
triclinic, space group P1, a ˆ 10.1899(4), b ˆ 15.9734(7), c ˆ 17.6277(7) ä,
3
À3
b ˆ 73.955(1)8; Vˆ 2616.6(2) ä , 1calcd ˆ 1.038 gcm , 2qmax 51.58, MoKa
radiation (l ˆ 0.71069 ä), wscans, Tˆ À109.08C, of 11852 measured
reflections, 8333 were unique, and 3188 were observed with [I > 3.00s(I)],
data corrected for Lorentz and polarization effects, analyzed for agreement
and absorption using XPREP with an empirical absorption correction using
SADABS (Tmax ˆ 0.98, Tmin ˆ 0.68). The structure was solved by direct
[
4] a) R. G. Smith, J. D. Lamb, J. Chromotogr., A 1994, 671, 89 ±94; b) B.
Ahlers, K. Cammann, S. Warzeska, R. Kr‰mer, Angew. Chem. 1996,
1
08, 2270 ± 2271; Angew. Chem. Int. Ed. Engl. 1996, 35, 2141 ± 2143;
c) C. Bucher, R. S. Zimmerman, J. Am. Chem. Soc. 2001, 123, 2099 ±
100; d) A. Andrievsky, F. Ahuis, J. L. Sessler, F. Vˆgtle, D. Gudat, M.
2
methods (SIR92) and developed by least-squares refinement against jF j .
2
No. of parameters, 487; H atoms were included but not refined; R ˆ 0.101
Moini, J. Am. Chem. Soc. 1998, 120, 9712 ± 9713, and references
À
À3
and R
w
ˆ 0.128; D1max ˆ 0.71 e ä
.
therein.
Crystal data for 18 (Zr
0
c ˆ 22.5575(3) ä, b ˆ 101.107(2)8; Vˆ 9019.3(3) ä , 1calcd ˆ 1.225gcm
4
Si
4
C
106
H
95): crystal dimensions 0.21 Â 0.13 Â
[5] a) F. Vˆgtle, Cyclophane Chemistry: Synthesis, Structures and Reac-
tions, Wiley, Chichester, 1990; b) D. Parker, Macrocycle Synthesis: A
Practical Approach, Oxford University Press, Oxford, 1996; c) F.
Diederich, P. J. Stang, Templated Organic Synthesis, Wiley-VCH,
.12 mm, monoclinic, space group C2/c, a ˆ 14.8362(1), b ˆ 27.4645(7),
3
À3
,
2
q
max 44.38, MoKa radiation (l ˆ 0.71069 ä), wscans, Tˆ À106 Æ 18C, of
82
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0082 $ 17.50+.50/0
Chem. Eur. J. 2002, 8, No. 1

Macrocyclizations of Silyl-Terminated Diynes
74±83
Weinheim, 2000; d) P. Siemsen, R. C. Livingston, F. Diederich, Angew.
Chem. 2000, 112, 2740 ± 2767; Angew. Chem. Int. Ed. 2000, 39, 2632 ±
A. Ohff, M. Michalik, H. Gˆrls, V. V. Burlakov, V. B. Shur, Angew.
Chem. 1993, 105, 1228 ± 1229; Angew. Chem. Int. Ed. Engl. 1993, 32,
1193 ± 1195.
2
657.
[
6] a) S. S. H. Mao, T. D. Tilley, J. Am. Chem. Soc. 1995, 117, 5365 ± 5366;
b) S. S. H. Mao, T. D. Tilley, J. Am. Chem. Soc. 1995, 117, 7031 ± 7032;
c) S. S. H. Mao, F.-Q. Liu, T. D. Tilley, J. Am. Chem. Soc. 1998, 120,
[15] a) T. Probst, O. Steigelmann, J. Riede, H. Schmidbaur, Angew. Chem.
1990, 102, 1471 ± 1472; Angew. Chem. Int. Ed. Engl. 1990, 29, 1397 ±
1399; b) T. Probst, O. Steigelmann, J. Riede, H. Schmidbaur, Chem.
Ber. 1991, 124, 1089 ± 1093; c) J. Schulz, F. Vˆgtle, Top. Curr. Chem.
1994, 172, 41 ± 86.
[16] a) I. Tabushi, Tetrahedron 1971, 27, 4845± 4853; b) F. Voegtle, W.
Kissener, Chem. Ber. 1984, 117, 2538 ± 2541.
[17] H. R. Snyder, C. Weaver, C. D. Marshall, J. Am. Chem. Soc. 1949, 71,
289 ± 291.
[18] D. L. Pearson, J. M. Tour, J. Org. Chem. 1997, 62, 1376 ± 1387.
[19] Lengths of spacers estimated from Molecular Mechanics Models
generated using MacSpartan Plus.
[20] Cavity size estimated from Molecular Mechanics Models generated
using MacSpartan Plus. This is the average length of the vector that
may be drawn from the center of the molecular cavity to the center of
the rigid spacer.
[21] J. R. Nitschke, T. D. Tilley, J. Am. Chem. Soc. 2001, 123, 10183 ± 10190.
[22] B. P. Warner, W. M. Davis, S. L. Buchwald, J. Am. Chem. Soc. 1994,
116, 5471 ± 5472.
[23] B. Du, M. F. Farona, D. B. McConnville, W. J. Youngs, Tetrahedron
1995, 51, 4359 ± 4370.
1
193 ± 1206.
7] a) E. I. Negishi, R. E. Cederbaum, T. Takahashi, Tetrahedron Lett.
986, 27, 2829 ± 2832; b) S. L. Buchwald, B. T. Watson, J. C. Huffman,
[
[
[
1
J. Am. Chem. Soc. 1987, 109, 2544 ± 2546.
8] a) S. S. H. Mao, T. D. Tilley, Macromolecules 1996, 29, 6362 ± 6364;
b) F.-Q. Liu, G. H. Harder, T. D. Tilley, J. Am. Chem. Soc. 1998, 120,
3
271 ± 3272.
9] a) G. Erker, R. Zwettler, J. Organomet. Chem. 1991, 409, 179 ± 188;
b) G. Erker, R. Zwettler, C. Kruger, I. Hyla-Kryspin, R. Gleiter,
Organometallics 1990, 9, 524 ± 530; c) G. Erker, R. Zwettler, C.
Kruger, I. Hyla-Kryspin, R. Gleiter, J. Organomet. Chem. 1988, 346,
C15± C18.
[
[
[
[
[
10] S. Takahashi, Y. Kuroyama, K. Sonogashira, N. Hagihara, Synthesis
980, 627 ± 630.
11] J. Zauhar, A. D. Bandrauk, K. D. Truong, A. Michel, Synthesis 1995,
03 ± 706.
12] V. Farina, B. Krishnan, D. R. Marshall, G. P. Roth, J. Org. Chem. 1993,
8, 5434 ± 5444.
13] C. Eaborn, A. R. Thompson, D. R. M. Walton, J. Chem. Soc. C 1967,
364 ± 1366.
14] a) J. R. Nitschke, S. Z¸rcher, T. D. Tilley, J. Am. Chem. Soc. 2000, 122,
1
7
5
[24] J. Nitschke, T. D. Tilley, J. Org. Chem. 1998, 63, 3673 ± 3676.
[25] a) P. J. Fagan, W. A. Nugent, J. C. Calabrese, J. Am. Chem. Soc. 1994,
116, 1880 ± 1889; b) T. Takahashi, M. Kotora, R. Hara, Z. Xi, Bull.
Chem. Soc. Jpn. 1999, 72, 2591 ± 2602.
1
1
2
0345± 10352; b) J. R. Nitschke, T. D. Tilley, Angew. Chem. 2001, 113,
200 ± 2203; Angew. Chem. Int. Ed. 2001, 40, 2142 ± 2145; c) U.
Rosenthal, A. Ohff, W. Baumann, A. Tillack, H. Gˆrls, V. V. Burlakov,
V. B. Shur, Z. Anorg. Allg. Chem. 1995, 621, 77 ± 83; d) U. Rosenthal,
Received: June 15, 2001 [F3333]
Chem. Eur. J. 2002, 8, No. 1
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
0947-6539/02/0801-0083 $ 17.50+.50/0
83