Synthesis of poly(para-phenylene)(2-isocyano-2-tosylpropane-1,3-diyl), poly(para-phenylene)(2-oxopropane-1,3-diyl) and oligo(cyclopentadienones) via carbonylative coupling of α,α′-dibromoxylene

Article abstract of DOI:10.1039/b706807g

Poly(para-phenylene)(2-oxopropane-1,3-diyl), a potential precursor of linear graphene, is generated in low yield from hydrolysis of a previously unknown poly(para-phenylene)(2-isocyano-2-tosylpropane-1,3-diyl) generated from inexpensive, commercially available starting materials. The Royal Society of Chemistry.

Full text of DOI:10.1039/b706807g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Synthesis of poly(para-phenylene)(2-isocyano-2-tosylpropane-1,3-diyl),
poly(para-phenylene)(2-oxopropane-1,3-diyl) and
oligo(cyclopentadienones) via carbonylative coupling of
a,a9-dibromoxylene{
Robert G. Potter and Thomas S. Hughes*
Received (in Bloomington, IN, USA) 8th May 2007, Accepted 3rd September 2007
First published as an Advance Article on the web 24th September 2007
DOI: 10.1039/b706807g
Carbonylative couplings with Na2Fe(CO)4 have been conducted
in our laboratories to provide symmetrical and asymmetrical
heterosubstituted diphenylacetones in high yield.8 However, when
the aromatic rings are substituted with electron withdrawing
groups, the yields are significantly lower. We have now found that
the van Leusen9 TOSMIC reaction is capable of producing these
electron poor dibenzyl ketones in moderate yields that are at least
as good as those reported in the literature10 (Fig. 1). van Leusen
couplings have also been conducted asymmetrically,9 but this has
not yet been pursued by our group.
Poly(para-phenylene)(2-oxopropane-1,3-diyl), a potential pre-
cursor of linear graphene, is generated in low yield from
hydrolysis of a previously unknown poly(para-phenylene)-
(2-isocyano-2-tosylpropane-1,3-diyl) generated from inexpen-
sive, commercially available starting materials.
Linear1 and cyclic2 fused polyaromatic hydrocarbons have been
synthesized by many groups, and are of increasing importance due
to their processability, relatively low expense, and low band gaps.
Oligomeric and polymeric graphitic materials3 and nanotubes4
have been investigated; however no synthesis of a monodisperse
carbon nanotube or a purely linear, oligomeric graphene has been
reported to date.{ These materials are of great interest for their
potential as components for photovoltaic cells, components of
nanoscale field effect transistors,5 and as shape persistent structures
for supermolecular assemblies.6
(para-Phenylene)(2-oxopropane-1,3-diyl) macrocycles have been
reported previously by Sasaki and Kitagawa11 using a van Leusen
carbonylative coupling reaction at high dilution. We have
reproduced these results, and have successfully purified and
characterized the cyclic trimer, tetramer and pentamer. It was
reasoned that performing this reaction, or the Collman coupling,
at higher concentrations would lead to linear oligomers.
Concentration of the TOSMIC coupling conditions does produce
the previously unreported polymer poly(para-phenylene)(2-iso-
cyano-2-tosylpropane-1,3-diyl) (PPIT) (Fig. 2).
Alternatively, the reaction of a,a9-dibromoxylene with
Na2Fe(CO)4 at high dilution led to a complex mixture of products
which did not contain a significant amount of macrocycle of any
degree of polymerization, presumably due to the slow rate-
determining second alkylation required for the ring closure step.8,12
This reaction was not run under more concentrated conditions to
obtain linear polymer because the kinetics would still be
disfavorable, and because of the success encountered using the
TOSMIC reagent.
More recently7 polymers consisting of tetraarylcyclopentadie-
none units have been suggested as low band gap semiconductors
and possible superconductors due to the low calculated HOMO–
LUMO gap of the monomer (ca. 2.72 eV) and the subsequent
reduction in band gap after polymerization which arises from the
increase in the extent of conjugation.
Due to our interest in polyaromatic macrocycles and polymers
we have decided to pursue the synthesis of oligo(para-pheny-
lene)(2-oxopropane-1,3-diyl) for later conversion to oligocyclopen-
tadienones and ultimately to oligohexaarylphenylenes, both as
macrocycles for the synthesis of carbon nanobelts and as linear
oligomers for the products of graphitic ribbons.
Department of Chemistry, Cook Physical Sciences Building, 82
University Place, University of Vermont, Burlington, VT 05401, USA.
E-mail: Thomas.S.Hughes@uvm.edu; Fax: +1 802-656-8705;
Tel: +1 802-656-0161
{ Electronic supplementary information (ESI) available: Synthetic meth-
ods and spectroscopic characterization of novel compounds. See DOI:
10.1039/b706807g
Fig. 1 van Leusen coupling of benzyl bromides.
Chem. Commun., 2007, 4665–4667 | 4665
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Fig. 4 1H NMR of polymer 4.
Fig. 2 Tosylmethyl isocyanide mediated polymerization.
macrocycles 2 with HCl in methylene chloride is straightforward,
but the hydrolysis of polymer 4 is less well-behaved. Higher
concentrations of acid lead to product mixtures of low solubility
and poor conversion. However, hydrolysis with dilute HCl gives
much higher conversion to the ketone functionality. Alternatively,
the isonitrile can be converted to the isocyanate with lead
tetraacetate; this polyisocyanate is easily hydrolyzed to 4 upon
stirring with alumina in methylene chloride. This method gives
lower conversion to the ketone, but suggests further chemical
transformations of 3, which are currently being pursued in our
laboratories.
Employing the TOSMIC reagent in a biphasic reaction mixture
using TBAI as the phase transfer catalyst and aqueous sodium
hydroxide as the base, polymer 3 was obtained. GPC showed a
reproducible Xn of 24–25, corresponding to an Mn of 9600. The
polydispersivity was measured to be 1.37, and was reproducible
from batch to batch. 1H NMR (Fig. 3) end-group analysis
comparing the integrations of the diastereotopic benzylic protons
to that of the remaining benzyl bromide groups gave an Xn of 30.5.
The reaction can be conducted easily on a multigram scale, but
decomposition of the polymer occurs slowly at room temperature,
requiring refrigeration if storage for longer than a few weeks is
necessary.
To determine the feasibility of synthesizing a tetraphenylcyclo-
pentadienone polymer that would require multiple Knovenagel
condensations on the 1,3-diphenyl-2-propanone moieties in 4, the
multiple Knovenagel condensation was attempted on shorter
oligomers which were prepared by performing the polymerization
with an excess of monofunctional reagent. By employing a 2 : 1
stoichiometry of benzyl bromide to a,a9-dibromoxylene in the
TOSMIC carbonylative coupling reaction, diketone 5a and
triketone 5b oligomers can be isolated in low yields, as shown in
Fig. 5. The unoptimized yields are low, but it should be noted that
the hydrolysis of these short oligomers is quite facile without any
problems of solubility, a result similar to that observed for small
cyclic oligomers but not for the polymer.
Due to the reactivity of the pendant functional groups we
believe this polymer could prove useful for further functionaliza-
tion to novel polymers. Methodology developed by other groups13
allows the potential for the conversion of PPIT to poly(para-
phenylene)(propane-1,3-diyl), polyisocyanide and polyisocyanate
polymers.
Hydrolysis of PPIT 3 affords the corresponding poly(para-
phenylene)(2-oxopropane-1,3-diyl), 4 (Fig. 4). The hydrolysis of
Initial Knovenagel condensations performed on diketone 5a
(and tri-ketone macrocycle) using potassium hydroxide in ethanol
led to complete decomposition of the dark purple reaction mixture
upon exposure to the atmosphere. Mass spectrometry of the
complex product mixture contains peaks consistent with
Fig. 3 1H NMR of polymer 3.
Fig. 5 Short diphenylacetone oligomers.
4666 | Chem. Commun., 2007, 4665–4667
This journal is ß The Royal Society of Chemistry 2007

View Article Online
oligomers, as well as polymers. We also believe the PPIT polymer
to be a useful precursor for the production of a variety of
functionalized low molecular weight polymers which may not be
accessible through current methology.
GPC data were provided by Prof. Robert B. Grubbs and Dr
Anand Sundararaman. Mass spectrometry was provided by the
Washington University Mass Spectrometry Resource with support
from the NIH National Center for Research Resources (Grant
No. P41RR0954), and by Applied Biosystems/MDS Sciex.
Notes and references
{ Although non-discrete examples are numerous, see refs. 1 and 2.
1 (a) E. Clar, Chem. Dtsch. Ber. Ges., 1937, 72B, 2137; (b) M. Bendikov,
F. Wudl and D. F. Perepichka, Chem. Rev., 2004, 104, 4891; (c) J. Wu,
L. Gherghel, M. D. Watson, J. Li, Z. Wang, C. D. Simpson, U. Kolb
and K. Mu¨llen, Macromolecules, 2003, 36, 7082.
2 (a) L. V. Radushkevich and V. M. Lukyanovich, Zh. Fiz. Khim., 1952,
26, 88; (b) S. Iijima, Nature, 1991, 354, 56.
Fig. 6 Synthesis of tetraphenylcyclopentadienone dimer.
photo-oxidation products.14 After exclusion of light and oxygen
prior to low temperature aqueous workup, condensation of 5a
with benzil led to a low yield of the corresponding tetraphenylcy-
clone 6 (Fig. 6). However, the majority of the product mixture was
composed of a variety of regioisomers and diastereomers resulting
from hydration of the four cyclopentadienone double bonds of 6.
Once purified, 6 and its hydrates are as stable (for several
months after removal of solvent) as substituted monomeric
tetraarylcyclopentadienones.
3 M. M. Haley and W. B. Wan, Adv. Strained Interesting Org. Mol., 2000,
8, 1.
4 (a) R. Saito, M. Fujita, G. Dresselhaus and M. S. Dresselhaus, Appl.
Phys. Lett., 1992, 60, 18; R. Saito, M. Fujita, G. Dresselhaus and
M. S. Dresselhaus, Appl. Phys. Lett., 1992, 60, 2204; (b) P. M. Ajayan,
Chem. Rev., 1999, 99, 1787; (c) B. I. Yakobson and R. E. Smalley, Am.
Sci., 1997, 85, 324.
5 H. Ba¨ssler, in Electronic Excitation. In Electronic Materials: The
Oligomer Approach, ed. K. Mu¨llen and G. Wegner, Wiley-VCH,
Weinheim, Germany, 1998.
To avoid the production of the hydrate products, potassium
bis(trimethylsilyl)amide in a solution of THF was employed. These
reaction conditions led to an 84% yield of 6, and no apparent
hydrate or photooxidation side products. The apparent increase in
reactivity of the intermediates leading to 6, or of the product itself,
was originally thought to be attributable to the lowering of the
LUMO arising from the increased conjugation. However, a
B3LYP/6-31G(d) computation reveals a HOMO–LUMO gap
similar to those of other electron-rich tetraarylcyclopentadienones
synthesized in our laboratory with no such sensitivity to oxygen
and a LUMO comparable to that in parent tetraphenylcyclopen-
tadienone which is not so susceptible to nucleophilic attack that it
cannot be prepared using hydroxide base. It should be noted that
the yield obtained in this multiple Knovenagel condensation is
much higher than those reported for the synthesis of other
bis(tetraphenylcyclopentadienone) isomers using hydroxide bases
where strict removal of oxygen and light was not performed.15
Initial attempts to convert oligomer 5b to a trimeric cyclopenta-
dienone were not successful, which we attribute to the small scale
of those reactions and the possibility that the linear trimer 5b may
be long enough to undergo back-biting aldol condensations with
itself that are not possible for shorter 5a.
6 M. Muller, F. Morgenroth, U. Scherf, T. Soczka-Guth, G. Klarner and
K. Mu¨llen, Philos. Trans. R. Soc. London, Ser. A, 1997, 355, 715;
J. P. Schmidtke, R. H. Friend, M. Kastler and K. Mu¨llen, J. Chem.
Phys., 2006, 124, 174704; J. M. Warman, J. Piris, W. Pisula, M. Kastler,
D. Wasserfallen and K. Mu¨llen, J. Am. Chem. Soc., 2005, 127, 14257;
A. Fechtenkotter, K. Saalwachter, M. A. Harbison, K. Mu¨llen and
H. W. Speiss, Angew. Chem., Int. Ed., 1999, 38, 3039.
7 F. Wudl, presented in part at the 25th Annual Esther and Bingham
J. Humphrey Memorial Symposium in Chemistry, Burlington, VT,
September 30th, 2006.
8 R. G. Potter and T. S. Hughes, Org. Lett., 2007, 9, 1187.
9 (a) O. Possel and A. M. van Leusen, Tetrahedron Lett., 1977, 48, 4229;
(b) D. van Leusen and A. M. van Leusen, Org. React., 2001, 57, 417.
10 S.-i. Inaba and D. R. Rieke, J. Org. Chem., 1985, 50, 1373; O. Possel
and A. M. van Leusen, Tetrahedron Lett., 1977, 48, 4229; E. Campaigne
and B. E. Edwards, J. Org. Chem., 1962, 27, 3760; R. P. L’Esperance,
D. Van Engen, R. Rajeev Dayal and R. A. Pascal, J. Org. Chem., 1991,
56, 688; K.-V. Tran and D. Bickar, J. Org. Chem., 2006, 71, 6640.
11 H. Sasaki and T. Kitagawa, Chem. Pharm. Bull., 1983, 31, 2868;
K. Kurosawa, M. Suenaga, T. Inazu and T. Yoshino, Tetrahedron Lett.,
1982, 23, 5335.
12 J. P. Collman, R. G. Finke, J. N. Cawse and J. I. Brauman, J. Am.
Chem. Soc., 1977, 99, 2515.
13 (a) D. van Leusen, E. van Echten and A. M. van Leusen, Recl. Trav.
Chim. Pays-Bas, 1992, 111, 469; (b) U. Heß and H. Brosig, Tetrahedron
Lett., 1991, 32, 5539; (c) D. Johnson and A. Poulos, Tetrahedron Lett.,
1992, 33, 2045; (d) J. S. Yadav, R. Kache and V. K. Reddy, Synth.
Commun., 1998, 28, 4249.
14 N. M. Bikales and E. I. Becker, J. Am. Chem. Soc., 1956, 21, 1405.
15 (a) L. Tong, D. M. Ho, N. J. Vogelaar, C. E. Schutt and R. A. Pascal,
Jr., J. Am. Chem. Soc., 1997, 119, 7291; (b) M. A. Ogliaruso and
E. I. Becker, Org. Chem., 1965, 30, 3354.
Due to the low cost of the starting materials and the ease of the
experimental conditions we believe even a low yielding conversion
of the cyclic or linear oligomers to polycyclopentadienones to be a
potential pathway to linear and cyclic tetraarylcyclopentadienone
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 4665–4667 | 4667