It takes alkynes to make a world - New methods for dehydrobenzoannulene synthesis

Article abstract of DOI:10.1055/s-1998-5736

Recent advances in the assembly of dehydrobenzoannulenes (DBAs) are the subject of this account. Using in situ generated acetylenic intermediates, novel α,ω-polyynes can be easily constructed using Pd-catalyzed alkynylation methodology. Subsequent cyclization via intra- and intermolecular Cu-mediated reactions can provide a wide variety of DBAs, including heretofore inaccessible topologies, systems with an unprecedented number of acetylenic linkages, structures with derivatized aromatic rings, and unusual organometallic species.

Full text of DOI:10.1055/s-1998-5736

ACCOUNTS
557
It Takes Alkynes to Make a World – New Methods for Dehydrobenzoannulene Synthesis
Michael M. Haley*
Department of Chemistry, University of Oregon, Eugene, Oregon 97403-1253, USA
Fax: 541-346-0487; E-mail: haley@oregon.uoregon.edu
Received 2 October 1997
Abstract: Recent advances in the assembly of dehydrobenzoannulenes
(DBAs) are the subject of this account. Using in situ generated
acetylenic intermediates, novel α,ω-polyynes can be easily constructed
using Pd-catalyzed alkynylation methodology. Subsequent cyclization
via intra- and intermolecular Cu-mediated reactions can provide a wide
variety of DBAs, including heretofore inaccessible topologies, systems
with an unprecedented number of acetylenic linkages, structures with
derivatized aromatic rings, and unusual organometallic species.
1.
2.
3.
4.
5.
6.
7.
8.
9.
An Introduction to DBAs
Setting the Stage
Monoacetylenic DBAs
Diacetylenic DBAs
Triacetylenic DBAs
Tetraacetylenic DBAs
Derivatized DBAs
Organometallic DBAs
Conclusion and Outlook
1.
An Introduction to DBAs
It was early April 1990, and I was under the gun. Every graduate student
in the chemistry program at Rice University had to write and defend an
original research proposal during the spring semester of their third year.
The problem was that I was 3000 miles away working in Professor
Roland Boese's laboratory in Essen, Germany. I was there as part of a
collaborative program between the Boese group and my Ph.D. mentor's
group (W. E. Billups), obtaining X-ray structural data on a variety of
1
cyclopropenes and cycloproparenes. Although the work was
progressing well, I was having a hard time with the proposal. Like most
graduate students, I had never written a research proposal and had no
idea where to begin. Needing inspiration, Roland allowed me to look
through his reprints. It was there I found the spark that I needed. In
1986, Vollhardt and co-workers, in collaboration with Boese, reported
2.
Setting the Stage
The 1960s and 1970s saw a tremendous explosion in the area of
dehydrobenzoannulene chemistry. The groups of Nakagawa, Staab,
2
6
the synthesis and crystal structure of hexaethynylbenzene. I was
immediately taken by the beauty of the high symmetry and by the ease
of preparation of the molecule. Later that same day I came across a
paper by Michael Wright where he had successfully prepared
and Sondheimer, the principal protagonists of this era, prepared an
impressive array of DBA structures. Molecules 3–5 are representative
7
examples from each group, respectively. The main factor driving this
3
phenylacetylene polymers via Stille-like coupling. What if it were
research was the dominating question of ring currents in the DBA
macrocycles. How diatropic, paratropic, or atropic were each of these
systems? While there were and still are an infinite number of DBA
topologies that could be synthesized, after awhile it became apparent
that there needed to be additional impetus for this research; thus, interest
waned and the area languished in the 1980s.
possible to combine the results in these two articles? I began to doodle
and soon filled up a page with a hybrid all-carbon network comprised of
acetylenes and benzenes. Upon doing a search of the literature, I was
surprised to find that a group of theoreticians had proposed the same
4
idea. Ray Baughman and his co-workers had published a detailed
article describing structure-property predictions for a new planar form
2
of carbon containing equal numbers of sp and sp atoms, which they
nicknamed graphyne (1). In addition, they had included calculations on
some model dehydrobenzoannulene (DBA) subunits of the larger
polymeric network (e.g., 2). This was just exactly what I was looking
for! Although I knew next to nothing about acetylene chemistry, I wrote
the proposal describing the synthesis of graphyne and four DBA
5
subunits using Stille chemistry and defended it shortly after I returned
The renaissance for DBA chemistry began in the early 1990s. Two key
events made this resurgence possible: (1) the isolation and
to Houston in mid-May. I passed the exam and soon tried to put the
"unpleasant" experience behind me. Little did I know that this did not
represent the end, but just the beginning of my odyssey with
dehydrobenzoannulenes..........
8
characterization of the fullerenes; and (2) the extensive development of
9
palladium-catalyzed cross-coupling reactions using alkynes. Fullerene
chemistry has single-handedly made research on all-carbon and carbon-
rich molecular and polymeric systems, of which DBAs are a prime

558
Michael M. Haley
SYNLETT
example, in vogue once more. Many of these molecules, including
DBAs, can potentially serve as precursors for variety of
It was during this DBA resurgence that I decided to jump into the fray. I
believed my postdoctoral stay in Peter Vollhardt's labs had prepared me
for the task that lay ahead. I specifically went to Peter's group as I knew
I could pick up a wealth of experience with acetylene chemistry,
something I could have used back in Essen. Armed with this new-found
knowledge, I joined the faculty at Oregon in July 1993 and set out to
build the molecules I had been thinking about for the previous three
years.
a
technologically important materials, such as novel allotropes of carbon,
8b,c,10-12
molecular scaffolds, and ladder polymers.
Palladium-catalyzed
alkynylation has revolutionized the way DBAs are constructed (vide
infra). Not only have new topologies now become accessible, but also
the number of steps to known macrocycles has been shortened and
overall yields dramatically increased in some cases. For example,
Staab's original synthesis of DBA 4 required six steps (8% yield)
starting from o-tolane, itself the product of a four-step reaction
7b
sequence.
Conversely, using commercially available 1-bromo-2-
iodobenzene, molecule 4 could be assembled quickly in two steps (23%
13
yield) via palladium chemistry. A drawback to this route though are
that higher cyclooligomers are produced as well, making isolation of
14
pure 4 somewhat tedious.
3.
Monoacetylenic DBAs
Our first synthetic attempts at Oregon were directed towards the
graphyne substructures 2, 8, and 9. Even though I was eager to begin
work on the molecules, I had difficulty convincing graduate students
that these systems were worthwhile targets. Nevertheless, eager
undergraduates were abundant and soon the ball began to roll. The
initial experiments for 2 and 8 utilized an intermolecular approach,
The group that instigated the DBA renaissance was that of Francois
Diederich (ETH). During the early 1990s, their research focused on the
preparation of cyclic polyacetylene macrocycles and of planar
15
13
polymeric networks based on tetraethynylethene (e.g., 6). Another
similar to what had been used previously to prepare other DBAs. We
group influential in this field was that of Jeff Moore (Illinois). The last
six years has seen a large number and variety of phenylacetylene
quickly learned that this was not the route to pursue. Instead of
furnishing the desired macrocycles, we obtained copious amounts of
polymeric materials. Use of high dilution conditions had no effect. This
was not how I wanted things to start.
16
macrocycles and dendrimers produced by Moore's team (e.g., 7).
Although neither research group was working directly on DBAs, many
of their discoveries potentially could be applied to such systems (vide
infra).
Reassessment of the situation suggested use of an intramolecular
approach. We redesigned the synthetic routes and Josh Kehoe, a very
talented undergraduate that I "inherited" when I moved into my labs at
Oregon, set off to make these molecules a reality. Since we were going
back to step 1, it was best to test and fine-tune our methods on a known
system; thus we first prepared the simplest graphyne subunit, annulene 4
17
(Scheme 1). Alkyne protection/deprotection,
Pd-catalyzed
9
18
alkynylation, and conversion of masked iodides comprised most of
the chemistry involved in the construction of 4, as well as 2 and 8. Using
the proven methods mentioned above, we were able to quickly assemble
diyne 10, then triyne 11. Iodination, desilylation, and intramolecular
19
alkynylation with Pd(dba) under high dilution conditions furnished 4
2
20
as the sole product. Although our route had several more steps than
the two-step bromoiodobenzene pathway, the overall yield was higher –
Biographical Sketch
Michael M. Haley was born in 1965 in Lake Charles, LA. After growing up in Tulsa,
Oklahoma, he studied cyclopropene and cycloproparene chemistry with Prof. W. E.
Billups at Rice University where he received both his Bachelor's (1987) and Ph.D.
degrees (1991). In 1991 he received a National Science Foundation Postdoctoral
Fellowship to work with Prof. K. P. C. Vollhardt on [N]phenylene chemistry at the
University of California, Berkeley. In 1993 he joined the faculty at the University of
Oregon where he is currently an Assistant Professor of Chemistry and an associate
member of the Materials Science Institute. In 1995 he was selected as a recipient of a
National Science Foundation CAREER Award. His current research focuses on
molecules with a high C:H ratio, novel aromatic and anti-aromatic systems, molecules of
theoretical interest, and the reaction of strained-ring compounds with transition metal
complexes. Outside the lab he can be found enjoying the many splendors of the state of
Oregon – gardening, wine tasting, snowboarding, or whitewater kayaking.

June 1998
It Takes Alkynes to Make a World − New Methods for Dehydrobenzoannulene Synthesis
559
13
35% for Scheme 1 versus the reported 23% yield. More importantly,
no cyclooligomeric molecules were detected, thus making product
isolation and purification extremely easy.
Iodination, desilylation, and double intramolecular coupling
gratifyingly furnished 2 as a bright yellow solid. To date, attempts to
obtain X-ray diffracting crystals have failed; nevertheless, the spectral
properties were fully in accordance with the proposed structure. The
most disappointing feature of this work was the abysmal overall yield –
0.6% for 13 steps. Three key steps (g, i, and j) each proceeded in under
30% yield. Despite numerous efforts (varying reagent amounts, altering
the order of addition, changing catalysts and solvents), we have been
unable to dramatically improve yields or adequately scale-up the
reactions for 2.
Buoyed by this success, we turned our attention to the larger graphyne
models. The routes used to construct 2 and 8 were quite similar to the
one for 4 in that we took advantage of many of the same reaction
sequences; unfortunately, these compounds proved to be synthetically
more difficult to prepare. At this stage, we decided to introduce tert-
butyl groups strategically placed on the aromatic rings to aid with any
solubility problems we might encounter. As mimics of carbon networks,
the larger models would be expected to exhibit poor solubility
characteristics.
The next system, molecule 8, proved to be equally challenging. Since
Josh had already prepared compound 10 earlier (Scheme 1), we figured
we would go ahead and try for the parent system. Using the same
general sequence of reactions, the carbon backbone (16a→17a) was
easily assembled (Scheme 3). Closure of 17a gave after vacuum
sublimation a yellow solid which proved to be sparingly soluble in
organic solvents. This problem prevented the acquisition of NMR data;
nevertheless, mass spectrometry, IR, and UV-Vis data confirmed the
20
formation of 8a. Even though it was eventually expected for large
models, we were surprised by the severity of the solubility problem at
this stage. We attributed the low yield of the cyclization step (<15%)
mainly to this problem. Repetition of the synthetic sequence with p-t-
butylaniline (Scheme 3) resulted in a worse outcome. Although the
same general set of reactions was used, the conversion of (16b→17b)
proved to be exceedingly difficult as, quite inexplicably, yields varied
wildly. To date we have been unable to isolate cyclized material 8b via
this route.
Over the course of 18 months, we tried several variations of the above
theme hoping to prepare 8b. In the process, three valiant undergraduates
22
expended their research lives, but all our efforts came to naught. Given
our synthetic problems, work on the more complex model 9 never even
got off the ground. Christmas 1995 was quickly approaching. For my
first 2.5 years at Oregon, I had very little to show for the large amount of
work – a "new" synthesis of known 4, an abysmal route to 2, horribly
insoluble 8a, and many dead-end intermediates for 8b. This would
I assigned the synthesis of 2 to a new undergraduate, Jamie Kiley, and
the elaborate route he accomplished is depicted in Scheme 2. Starting
with p-t-butylaniline, Jamie was able to construct first 13, then 14.
21

560
Michael M. Haley
SYNLETT
certainly not get me tenure. Something had to change, and change soon,
or I would be forced to "pull the plug" on my pet project.
4.
Diacetylenic DBAs
Two important events happened shortly after New Year's Eve 1996: (1)
two graduate students began to work on the acetylene molecules –
Stephen Brand changed research areas and Josh Pak joined my group.
Although the undergraduates had made good progress, now I had the
manpower that could be devoted full-time to this project. (2) The
decision was made to move from monoacetylenes to diacetylenes.
Preliminary DSC experiments on the thermal chemistry of 2 and 4
showed these molecules to be exceedingly stable, polymerizing only
23
above 350 °C and over a wide 50+ °C range. These results did not
bode well for preparing ordered, carbon-rich materials. Strained
diacetylenes (e.g., 18), on the other hand, had been recently shown by
Swager et al. to polymerize below 200 °C and over a relatively narrow
24
range. Although they had been unable to conclusively identify the
thermal products, their results looked promising. Shortly before the end
of 1995, Vollhardt and Youngs published their results on strained
25
diacetylene 19. This came as some surprise as I was unaware that
26
Peter's group was involved with DBA chemistry at that time. Their
beautiful results for 19 seemed to suggest that the molecule underwent
topochemical polymerization to give a tube-like structure with a
polydiacetylene backbone. After consulting with both groups to ensure
we did not tread on their turf, we set off on making diacetylenic
macrocycles.
Macromolecular network 20 and model compounds 21-24 were the next
set of molecules we focused on. Graphdiyne 20 differs from graphyne 1
by inclusion of an extra alkyne unit between the aromatic rings. Our
interest in these larger systems is fourfold: (1) graphdiyne should exhibit
most, if not all the interesting materials properties predicted for
In order to ensure formation of a single product, an intramolecular
dimerization of α,ω-polyynes was envisaged to prepare models 21-24.
Synthesis of each polyyne would necessitate use of a suitably
4
graphyne. (2) The diacetylenic composition of 20-24 is particularly
attractive as the topochemical polymerization of diacetylenes (vide
supra) creates single crystals of organic polymers with conjugated
functionalized phenylbutadiyne. In the case of 21, the synthon we
28
needed was 25 (Scheme 4). Selective desilylation with K CO should
2
3
27
backbones. The resultant polydiacetylenes are known to exhibit NLO
afford the free phenylbutadiyne, which would then be subjected to
standard Pd-coupling conditions. When this was indeed tried,
concentration of the reaction mixture gave a dark brown gum instead of
a free-flowing oil. Disaster had struck! Had we done our homework, we
would have known that the parent molecule, 1-phenyl-1,3-butadiyne, is
a highly reactive compound which polymerizes rapidly when neat or in
concentrated solution; even a dilute solution at –20 °C polymerizes
properties and are photogenerated carriers with mobilities much higher
27a
than other organic polymers. (3) The extra alkyne units in 20 result in
a pore size of about 2.5 Å, a cavity which can easily incorporate atoms
as large as cesium. This should lead to interesting redox applications, as
the large holes in the planar sheets can accommodate through-sheet
transport of metal ions and can provide a unique method of dopant
storage by intrasheet intercalation. (4) The designed synthetic pathways
to these molecules should be significantly shorter. This hopefully will
translate into easier preparation of greater quantities of material.
Surprisingly, the simplest model, compound 21, had eluded synthesis,
though some alkylated derivatives were prepared by Swager's group by
the same cyclooligomerization reaction of 1,2-diethynylbenzenes used
29
within a few hours. This extreme reactivity has limited the synthetic
utility of phenylbutadiynes to date. Indeed, all of our attempts to use
unprotected phenylbutadiynes in Pd-catalyzed alkynylation reactions
provided intractable polymeric gums. We had just started down this
track and we were already derailed. Still, my 4+ years training as a
cyclopropene chemist had taught me that there are many ways to
circumvent instability problems.
24
to produce 18. As with the two-step route to 4, complex product
mixtures were produced which made separation of the dimers, trimers,
and tetramers very difficult and resulted in low isolated yields of a given
macrocycle. Additionally, the variation in product structures was
severely limited by the ease of construction, or lack thereof, of the
starting o-diethynylbenzene.
The answer to this impasse was in situ generation of the
30
phenylbutadiynes under standard Pd-coupling conditions.
The
question was what base should we use to remove the trimethylsilyl
group? The first attempt using K CO gave us back starting materials.
2
3

June 1998
It Takes Alkynes to Make a World − New Methods for Dehydrobenzoannulene Synthesis
561
Addition of NaOMe did furnish the desired product, but in less than 5%
yield; again, we isolated mainly starting materials. The third time was
indeed a charm – addition of a few milliliters of a concentrated KOH
solution provided the bis-coupled product in 71% yield. Desilylation
and use of high dilution conditions in the oxidative dimerization
reaction gave 21 as the sole product in moderate yield. As with 8,
compound 21 is poorly soluble in common organic solvents;
nevertheless, all spectral data (NMR, IR, UV, MS) support this structure.
The minimal solubility of the product is no doubt responsible for the
low isolated yield.
α,ω-polyynes in very good yields. Subsequent deprotection and
cyclization provided DBAs that were inaccessible by traditional routes
32
33
(28, 30, 31) or previously available only in low yield (29, 32 ).
Fortunately, product solubility was not an issue as these macrocycles
either possessed lower symmetry (e.g., 28 vs. 21) or were non-planar
(29-32).
5.
Triacetylenic DBAs
1997 saw two new graduate students come on board. The first task for
Dave Kimball and Brad Wan was to extend our process to substituted
34
35
phenylhexatriynes, as illustrated in Scheme 6. The necessary
polyyne 33 was prepared with some difficulty from 34, an intermediate
in the syntheses of 28-30 that is readily available in multigram
28
Assembly of bismacrocycle 22 is depicted in Scheme 5. From the
36
quantities.
Modified
Cadiot-Chodkiewicz
coupling
using
outset, we recognized the need for including solubilizing substituents.
The required triynes 26 were easily prepared by standard Pd-catalyzed
reactions. Fourfold in situ desilylation/alkynylation gave the fluorescent
dodecaynes 27 in ca. 35% yield. Although modest, this yield implies an
average conversion of about 90% for each of the eight transformations
necessary. Desilylation of 27 and intramolecular oxidative dimerization
gave 22. Once again, poor solubility was a factor. Even with four t-butyl
groups, 22a proved to be virtually insoluble in common solvents, hence
the low yield. Fortunately, the n-decyl moieties worked quite well, with
22b now isolated in 83%(!) yield after purification.
37
monodesilylated 34 and known TMS-C≡CC≡C-Br
gave 33 in
moderate yield, which in turn was converted into 35 by our standard
procedures. As with DBA 28, 35 proved to be readily soluble.
Additionally, since our syntheses produce only single macrocycles,
extensive separation and purification procedures which result in material
loss were avoided. We have successfully prepared 36-38 using similar
methodology. Together, these macrocycles represent the very first
examples
of
hexatriyne-
and/or
hexaenediyne-connected
dehydroannulenes, benzannelated or otherwise.
The in situ deprotection/alkynylation protocol has proven to be
6.
Tetraacetylenic DBAs
exceedingly useful in our laboratory. In addition to macrocycles 21 and
28,31
22,
we have prepared a wide variety of other DBA topologies. As a
The more time I spend in the chemistry arena, the more I am convinced
serendipity plays a role. Sometimes it can be a major part of the
discovery. The following is a perfect case in point. Although Scheme 4
makes the assembly of 25 look trivial, anyone who has worked with
butadiynyltrimethylsilane can tell you that synthesis of this molecule is
project for summer 1996, three capable undergraduate students (Mike
Bell, Jamieson English, and Charles Johnson) synthesized DBAs 28-
30
32. Extension of the simple, one-pot procedure to the iodoarenes in
Table 1 allowed us to prepare a series of bis(triisopropylsilyl) protected

562
Michael M. Haley
SYNLETT
our knowledge, the 32-membered macrocycle, possessing an
unprecedented number of four acetylenes between the aromatic rings,
makes 40 the largest tetrabenzo-DBA yet characterized.
7.
Derivatized DBAs
One of the limitations of the traditional copper-mediated
cyclooligomerization reaction is generation of differentially substituted
DBAs. The substitution pattern in the starting o-diethynylbenzene must
be maintained on each and every benzene moiety in the oligomeric
mixture of DBAs that is produced. Thus, it is impossible to prepare less
symmetric systems like 41 via the older route.
large quantities is most certainly non-trivial. Thus triyne 25 is guarded
jealously in the lab and is not to be wasted. Nevertheless, failure to
degas a solution properly in a Pd-catalyzed alkynylation results in a side
reaction where an appreciable amount of the terminal acetylene
dimerizes. Although formation of such a side product is part of the
II→
0
initiation process of the catalytic cycle (Pd Pd ), usually the amount
of alkyne dimer is quite small (< 5 mg). Unfortunately, during one of his
reactions making the α,ω-polyyne precursor to 31, Mike Bell forgot to
degas the mixture before adding 25. The first band off the
13
Until recently, we had focused solely on preparation of the parent
hydrocarbons. With our synthetic approach, however, it should be
possible to construct a wide variety of derivatized structures. Due to the
stepwise pattern of molecule assembly, double introduction of one or
two functional groups, thus forming macrocycles based on 41, should be
a straightforward process. One pattern in which we are particularly
interested is "push-pull" derivatives like 41f and 41g. It should be
possible to enhance the physical properties of the molecule for potential
use of the annulenic monomers or polymers as conjugated materials for
chromatography column was 50 mg of a light brown gum. C NMR
data showed only 6 resonances in the alkyne region, and the aromatic
1
region of H NMR spectrum was far too simple to be the desired α,ω-
polyyne. Conclusion – it had to be dimer 39a.
I had mixed emotions about 39a. I was disappointed with Mike as he
had worked in my group long enough to know better than not to degas a
reaction solution. On the other hand, I knew that 39a could be useful in
its own right. I started wondering out loud in the lab. Surely there must
be a way to intentionally produce 39a?!? Someone piped up that it
would be neat if we could do the dimerization in situ as well. That was
it! The answer was so easy – addition of K CO to Eglinton
40
electronics and photonics. Specifically, we hope that our systems
could find use as highly active, second and third-order non-linear optical
41
2
3
materials. To this end, we have successfully prepared several donor
dimerization conditions (Scheme 7). Mike set up the reaction late that
afternoon and redeemed himself by noon the next day – a 61% yield of
pure 39a. We have tried this modification on a variety of monoynes and
42
and/or acceptor macrocycles, as illustrated in Table 2. The assembly
of 41a-g is easily accomplished as the starting arenes 42 and 43 are
readily prepared via literature methods. UV-Vis spectral results for 41f
and 41g show enhanced delocalization in these macrocycles – the
absorption bands are significantly broadened and are red-shifted by up
38
diynes, obtaining product yields as high as 98%. With an ample
supply of 39a in hand, desilylation with Bu NF and cyclization with
4
39
42
CuCl/Cu(OAc)
under pseudo-high dilution conditions provided the
2
to 75 nm.
orange cyclodimer 40a as the sole product; neither higher
cyclooligomers nor the highly strained macrocycle arising from
intramolecular ring closure were detected. Despite being non-planar,
compound 40a proved to be somewhat soluble in common solvents.
Repetition of the synthetic sequence using 26a furnished macrocycle
40b. As expected, inclusion of the t-butyl substituents noticeably
improved product solubility and thus the isolated yield. To the best of
8.
Organometallic DBAs
While "push-pull" derivatization is one method for modifying the
electrooptical properties of our annulenes, another promising route is
incorporation of a metal fragment.
interaction in transition metal σ-acetylide complexes
43
Specifically, the electronic
44
leads to

June 1998
It Takes Alkynes to Make a World − New Methods for Dehydrobenzoannulene Synthesis
563
absorption in 21. These data are contrary to those found with a highly
conjugated, acyclic system, where insertion of the
bis(triethylphosphine)platinum fragment resulted in a hypsochromic
44b
shift. Our work strongly suggests that electronic delocalization in the
platinum complex is indeed functional throughout the entire
macrocycle.
An alternate use of organometallic fragments that we have been
pursuing is use of such species to stabilize highly strained DBAs with
the subsequent liberation of the free hydrocarbons. Despite significant
efforts on the part of several research groups, systems containing
15
contiguous, bent triynes and higher polyynes have proven elusive;
only strained monoyne and diyne derivatives (such as 18) are
24,25
known.
Although we have prepared triyne- and tetrayne-linked
45
systems (vide supra), calculations show these to be relatively strain-
free. The most "highly strained" is compound 36. Even then, the
largest bending in the acetylenes is calculated to be only 5.1°.
electronic delocalization across the metal center. This characteristic
has indeed been suggested to influence the electrooptical properties of
metal acetylenic system, as recent studies of organometallic rigid-rod
polymers with bridging polyyne units found enhanced nonlinear optical
48
In hindsight, our first attempt was overly ambitious – macrocycle 45
46
49
behavior. With the above objective in mind, Josh Pak set about
was our target. With sp bond angles around 162°, 45 would be highly
exploring the inclusion of trans-bis(triethylphosphine)platinum
fragments in our annulenic systems.
strained and therefore was expected to be quite reactive. The octacobalt
complex 46, on the other hand, should be readily isolable. Indeed, we
prepared 46 as stable, deep maroon crystals (Scheme 9). All
spectroscopic data supported formation of the strain-free dimeric
structure. Unfortunately, all attempts to liberate 45 from the cobalt units
led only to insoluble materials. Diederich et al. have observed similar
Shown in Scheme 8 is our successful preparation of platinum-containing
47
macrocycle 44. Whereas the parent hydrocarbon 21 proved to be
poorly soluble, inclusion of the trans-bis(triethyl-phosphine)platinum
fragment in the carbon backbone dramatically improved product
solubility. The X-ray crystal structure of 44 (Figure 1) clearly shows
retention of the trans geometry and the additional strain placed on the
molecule by metal incorporation – the bond angles on the side
containing the platinum fragment are distorted up to 10° from linearity.
15
problems trying to prepare the cyclocarbons. Whether the failure to
prepare these two classes of macrocycles is due to the extreme reactivity
of the distorted polyyne moiety or to the lack of a viable synthetic route
is not certain. Thus, isolation and characterization of smaller bent
hexatriyne– and octatetrayne–containing systems should help answer
these questions.
9.
Conclusion and Outlook
The chemistry presented herein provides an entry into the synthesis of a
diverse array of dehydrobenzoannulene topologies. Although the Pd-
catalyzed alkynylation/Cu-mediated dimerization strategy is not new,
our ability to construct a plethora of previously inaccessible α,ω-
polyynes breathes new life into this old route.
A rough estimate for the extent of electron delocalization in platinacycle
44 can be obtained by comparison of its absorption spectrum with that
Where do we go from here? There are several important avenues which
we are currently pursuing: (1) bent polyynes. Is it possible to prepare
highly strained polyynes like 45 or 47? We are hoping the answer is
of 21. The λ
of the highest end adsorption in 44 is shifted
max
bathochromically by about 25 nm when compared with the analogous

564
Michael M. Haley
SYNLETT
"yes". (2) Aromaticity. Systems like 21 are Hückel [4n+2] π-electron
aromatics. To date, though, it has been difficult to quantify just how
weak the diatropic ring current is. Annelating the DBA to a
Buckminsterfullerenes, Billups, W. E; Ciufolini, M. A. (Eds.);
VCH Publishers: New York, NY, 1993.
(9) Heck, R. F. Palladium Reagents in Organic Syntheses, Academic
50
dimethyldihydropyrene core (e.g., 48) should provide the answer. (3)
Press: London, 1985.
Organometallics. The use of DBAs as ligands in transition metal
(10) Bachmann, P. K.; Messier, R. Chem. Eng. News 1989, 67(20), 24.
(b) Simpson, M. New Sci. 1988, 117(1603), 50.
14,47,49
chemistry is very limited.
We feel that there is a wealth of
unexplored chemistry utilizing these systems. Macrocycle 49 is but one
such complex we are actively seeking to prepare. (4) Materials
properties. As mentioned several times in this account, DBAs have the
potential for forming novel materials with useful properties. We have
only scratched the surface of this facet. Our synthetic methods allow us
to prepare a wide variety of DBAs in sufficient quantity to permit further
study. This should allow us to systematically study the reaction
chemistry and possible materials properties of our macrocycles. With
perseverance and a little luck, we just might "find a diamond in the
rough". On the other hand, given the subject of this account, it might be
more appropriate to "find the fullerene in the soot"!
(11) (a) Neenan, T. X.; Callstrom, M. R.; Scarmoutzos, L. M.; Stewart,
K. R.; Whitesides, G. M.; Howes, V. R. Macromolecules 1988, 21,
3528. (b) Callstrom, M. R.; Neenan, T. X.; Whitesides, G. M. ibid.
1988, 21, 3530.
(12) (a) Balaban, A. T.; Rentia, C. C.; Ciupitu, E. Rev. Roum. Chim.
1968, 13, 231. (b) Hoffmann, R.; Hughbanks, T.; Kertész, M.;
Bird, P. H. J. Am. Chem. Soc. 1983, 105, 4831. (c) Johnston, R. L.;
Hoffmann, R. ibid. 1989, 111, 810. (d) Baughman, R. H.; Galvao,
D. S. Nature (London) 1993, 365, 735. (e) Best, S. A.; Bianconi, P.
A.; Merz, K. M., Jr. J. Am. Chem. Soc. 1995, 117, 9251.
(13) Huynh, C.; Linstrumelle, G. Tetrahedron 1988, 44, 6337.
(14) For an alternate cyclooligomerization synthesis of 4 (three steps
from phenylacetylene, 24% overall yield) and numerous
references of its use as a ligand in organometallic chemistry, see:
Solooki, D.; Ferrara, J. D.; Malaba, D.; Bradshaw, J. D.; Tessier,
C. A.; Youngs, W. J. Inorg. Synth. 1997, 31, 122.
(15) (a) Diederich, F. Nature (London) 1994, 369, 199. (b) Diederich,
F. in Modern Acetylene Chemistry, Stang, P. J.; Diederich, F.
(Eds.); VCH: Weinheim, 1995, p 443.
(16) (a) Moore, J. S. Acct. Chem. Res. 1997, 30, 402. (b) Young, J. K.;
Moore, J. S. in Modern Acetylene Chemistry, Stang, P. J.;
Diederich, F. (Eds.); VCH: Weinheim, 1995, p 415.
Acknowledgments
The author is deeply indebted to the graduate and undergraduate co-
workers mentioned in this account. Without their hard work and tireless
dedication, none of the results described herein would have been
possible. We gratefully acknowledge the National Science Foundation,
The Petroleum Research Fund (administered by the American Chemical
Society), and the University of Oregon for financial support.
(17) (a) Brandsma, L. Preparative Acetylenic Chemistry, 2nd ed.,
Elsevier: Amsterdam, 1988. (b) Colvin, E. W. Silicon Reagents in
Organic Synthesis, Academic Press: London, 1988.
(18) Moore, J. S.; Weinstein, E. J.; Wu, Z. Tetrahedron Lett. 1991, 32,
2465.
(19) Zhang, J.; Pesak, D. J.; Ludwick, J. L.; Moore, J. S. J. Am. Chem.
Soc. 1994, 116, 4227.
References and Notes
(1) (a) Boese, R. In Advances in Strain in Organic Chemistry, Vol. 2;
Halton, B., Ed.; JAI Press: London, 1992, p 191. (b) Billups, W.
E.; Arney, B. E., Jr. In Advances in Strain in Organic Chemistry,
Vol. 6; Halton, B., Ed.; JAI Press: London, 1997, p 35.
(20) Kehoe, J. M.; Haley, M. M., unpublished results.
(21) Kiley, J. H.; Haley, M. M., unpublished results.
(22) Josh Kehoe, Ryan Petersen, and Charles Johnson were the
unfortunate souls to have gotten stuck with all the variations
towards 8b. I gratefully acknowledge their many efforts.
(2) Diercks, R.; Armstrong, J. C.; Boese, R.; Vollhardt, K. P. C.
Angew. Chem., Int. Ed. Engl. 1986, 98, 270.
(23) Robert Schneidmiller of Professor David Johnson's laboratory at
Oregon has kindly provided us with all of the needed DSC results
over the last 2+ years.
(3) Wright, M. E. Macromolecules 1989, 22, 3256.
(4) Baughman, R. H.; Eckhardt, H.; Kertesz, M. J. Chem. Phys. 1987,
87, 6687.
(24) Zhou, Q.; Carroll, P. J.; Swager, T. M. J. Org. Chem. 1994, 59,
1294.
(5) Echavarren, A. M.; Stille, J. K. J. Am. Chem. Soc. 1987, 109,
5478, and references therein.
(25) Baldwin, K. P; Matzger, A. J.; Scheiman, D. A.; Tessier, C. A.;
Vollhardt, K. P. C.; Youngs, W. J. Synlett 1995, 1215.
(6) For a comprehensive review of annulene chemistry, see: Balaban,
A. T.; Banciu, M.; Ciorba, V. Annulenes, Benzo-, Hetero-, Homo-
Derivatives and their Valence Isomers, Vol. 1-3, CRC Press: Boca
Raton, FL. 1987.
(26) Adam Matzger, the talented graduate student who prepared 19,
joined Peter's group just as I left for Oregon; thus, I was unaware
of the projects Peter had him exploring.
(27) (a) Polydiacetylenes, Bloor, D.; Chance, R. R. (Eds.); Martinus
Nijhoff: Boston, 1985, and references therein. (b) Wegner, G. Pure
Appl. Chem. 1977, 49, 443.
(7) (a) Yasuhara, A.; Satake, T.; Iyoda, M.; Nakagawa, M.
Tetrahedron Lett. 1975, 895. (b) Staab, H. A.; Graf, F. Chem. Ber.
1970, 103, 1107. (c) Darby, N.; Cresp, T. M.; Sondheimer, F. J.
Org. Chem. 1977, 42, 1960.
(28) Haley, M. M.; Brand, S. C.; Pak, J. J. Angew. Chem., Int. Ed. Engl.
1997, 36, 836.
(8) (a) Krätschmer, W.; Lamb, L. D.; Fostiropoulos, K.; Huffman, D.
R. Nature (London) 1990, 347, 354. (b) Dresselhaus, M. S.;
Dresselhaus, G.; Eklund, P. C. Science of the Fullerenes and
Carbon Nanotubes, Academic Press: San Diego, CA. 1995; (c)
(29) The first edition of reference 17a contained the preparation of the
parent molecule, but stated that the diyne proved to be very
unstable. Other groups have reported similar problems using free

June 1998
It Takes Alkynes to Make a World − New Methods for Dehydrobenzoannulene Synthesis
565
phenylbutadiynes in synthesis; see: Godt, A. J. Org. Chem. 1997,
62, 7471.
1996, 8, 231. (c) Tykwinski, R. R.; Schreiber, M.; Carlon, R. P.;
Diederich, F.; Gramlich, V. Helv. Chim. Acta 1996, 79, 2249.
(42) Pak, J. J.; Haley, M. M., manuscript in preparation.
(30) Haley, M. M.; Bell, M. L.; English, J. J.; Johnson, C. A.; Weakley,
T. J. R. J. Am. Chem. Soc. 1997, 119, 2956.
(43) For a recent review, see: Long, N. J. Angew. Chem., Int. Ed. Engl.
1995, 34, 21.
(31) Work on macrocycles 23 and 24 is currently in progress by
Stephen Brand.
(44) For some recent examples, see: a) Bradshaw, J. D.; Guo, L.,
Tessier, C. A., Youngs, W. J. Organometallics 1996, 15, 2582. b)
Faust, R.; Diederich, F.; Gramlich, V.; Seiler, P. Chem. Eur. J.
1995, 1, 111. c) Tykwinski, R. R.; Stang, P. J. Organometallics
1994, 13, 3203. d) Beck, W.; Niemer, B.; Wieser, M. Angew.
Chem., Int. Ed. Engl. 1993, 32, 923.
(32) Boese, R.; Matzger, A. J.; Vollhardt, K. P. C. J. Am. Chem. Soc.
1997, 119, 2052.
(33) Guo, L.; Bradshaw, J. D.; Tessier, C. A.; Youngs, W. J. J. Chem.
Soc., Chem. Commun. 1994, 243.
(34) Although the synthesis of phenylhexatriynes has been previously
reported, no yields were included in the paper. The only
identification of the products was by UV spectroscopy. See:
Eastmond, R.; Walton, D. M. R. Tetrahedron 1972, 28, 4591.
(45) Lichtenberger, D. L.; Renshaw, S. K.; Bullock, R. M. J. Am.
Chem. Soc. 1993, 115, 3276. b) Lichtenberger, D. L.; Renshaw, S.
K.; Wong, A.; Tagge, C. D. Organometallics 1993, 12, 3522. c)
Lewis, J.; Khan, M. S.; Kakkar, A. K.; Johnson, B. F. G.; Marder,
T. B.; Fyfe, H. B.; Wittmann, F.; Friend, R. H.; Dray, A. E. J.
Organomet. Chem. 1992, 425, 165.
(35) Kimball, D. B.; Wan, W. B.; Haley, M. M., manuscript in
preparation.
(36) Rubin, Y.; Parker, T. C.; Khan, S. I.; Holliman, C. L.; McElvany,
(46) Nalwa, H. S. Adv. Mater. 1993, 5, 341. b) Kanis, D. R.; Ratner, M.
A.; Marks, T. J. Chem. Rev. 1994, 94, 195. c) Marder, S. R.; Perry,
J. W. Adv. Mater. 1993, 5, 804. d) Meyers, L. K.; Langhoff, C.;
Thompson, M. E. J. Am. Chem. Soc. 1992, 114, 7560.
S. W. J. Am. Chem. Soc. 1996, 118, 5308.
(37) de Graaf, W.; Smits, A.; Boersma, J.; van Koten, G.; Hoekstra, W.
P. M. Tetrahedron 1988, 44, 6699.
(38) Haley, M. M.; Bell, M. L.; Brand, S. C.; Kimball, D. B.; Pak, J. J.;
(47) Pak, J. J.; Weakley, T. J. R.; Haley, M. M. Organometallics 1997,
16, 4505.
Wan, W. B. Tetrahedron Lett. 1997, 38, 7483.
(39) Vögtle, F.; Berscheid, R. Synthesis 1992, 58.
(48) Semi-empirical calculations (AM1, PM3) were performed on a
SGI workstation using Spartan molecular modelling software
(Version 4.1.1).
(40) Conjugated Polymers and Related Materials. The Interconnection
of Chemical and Electronic Structure. Salaneck, W. R.;
Lundström, I.; Ränby, B. (Eds.); Oxford University Press: Oxford,
1993.
(49) Haley, M. M.; Langsdorf, B. L. J. Chem. Soc., Chem. Commun.
1997, 1121.
(41) Tykwinski, R. R.; Schreiber, M.; Gramlich, V.; Seiler, P.;
Diederich, F. Adv. Mater. 1996, 8, 226. (b) Bosshard, C.; Spreiter,
R.; Günter, P.; Tykwinski, R. R.; Schreiber, M.; Diederich, F. Ibid.
(50) Mitchell, R. H.; Iyer, V. S.; Khalifa, N.; Mahadevan, R.;
Venugopalan, S.; Weerawarna, S. A.; Zhou, P. J. Am. Chem. Soc.
1995, 117, 1514, and references therein.