Revisit of the dessy-white intramolecular acetylene-acetylene [2 + 2] cycloadditions

Article abstract of DOI:10.1021/jo061334v

In this experiment, a series of thermal reactions of 4,4′- disubstituted 2,2′-bis(phenylethynyl)biphenyls with 2,3,4,5- tetraphenylcyclopenta-2,4-dienone were carried out under neat conditions and in diphenyl ether at temperatures between 260 and 270 °C t

Full text of DOI:10.1021/jo061334v

Revisit of the Dessy-White Intramolecular Acetylene-Acetylene
[2 + 2] Cycloadditions
Chung-Chieh Lee,† Man-kit Leung,*,†,‡ Gene-Hsiang Lee,† Yi-Hung Liu,† and
Shie-Ming Peng†
Department of Chemistry and Institute of Polymer Science and Engineering, National Taiwan UniVersity,
1 Section 4, RooseVelt Road, Taipei 106, Taiwan
mkleung@ntu.edu.tw
ReceiVed June 28, 2006
In this experiment, a series of thermal reactions of 4,4′-disubstituted 2,2′-bis(phenylethynyl)biphenyls
with 2,3,4,5-tetraphenylcyclopenta-2,4-dienone were carried out under neat conditions and in diphenyl
ether at temperatures between 260 and 270 °C to give rise to 9,10,11,12,13,14-hexaphenylcycloocta[l]-
phenanthrenes as the adducts in 12-23% yields. We traced these results to the intramolecular [2 + 2]
thermal cyclization of 2,2′-bis(phenylethynyl)biphenyls to form 1,2-diphenylcyclobuta[l]phenanthrenes,
which were further trapped as bridged-ketone Diels-Alder adducts, followed by thermal decarbonylative
ring opening, which gave rise to the products.
SCHEME 1
SCHEME 2
Introduction
Intra- and intermolecular cyclization of conjugated enynes
have been of considerable value from synthetic as well as
mechanistic points of view.1 Among the large variety of the
thermal cyclization reactions, transformation of two acetylene
moieties to cyclobutadiene through a [2 + 2] cycloaddition
reaction, as shown in Scheme 1, is an attractive experiment to
attempt.2 However, not many successful examples have been
reported, proving it to be a challenging problem to tackle.
One related example reported in the literature is the Sond-
heimer cyclization of 3,5-octadiene-1,7-diyne (1) that involved
formation of the diallenic cyclic hydrocarbon 2 and benzocy-
clobutene (3) to give the dimeric product 4 (Scheme 2).3 The
presence of the intermediate 3 was further evidenced by a
trapping experiment with cyclopentadiene to give rise to the
Diels-Alder adduct. The presence of 2 was also concluded on
the basis of a hydrogenation experiment, in which 0.5% of
cyclooctane was identified. The dimerization behavior of various
benzocyclobutadiene derivatives under the Sondheimer cycliza-
tion conditions has recently been investigated.4
† Department of Chemistry.
‡ Institute of Polymer Science and Engineering.
(1) For recent reviews and examples, see: (a) Schreiner, P. R.; Navarro-
Va´zquez, A.; Prall, M. Acc. Chem. Res. 2005, 38, 29. (b) Basak, A.; Mandal,
S.; Bag, S. S. Chem. ReV. 2003, 103, 4077. (c) Wang, K. K. Chem. ReV.
1996, 96, 207. (d) Gonza´lez, J. J.; Francesch, A.; Ca´rdenas, D. J.;
Echavarren, A. M. J. Org. Chem. 1998, 63, 2854. (e) Lewis, K. D.; Matzger,
A. J. J. Am. Chem. Soc. 2005, 127, 9968.
Thermal cyclization of the biphenyl homologue 6, however,
occurred in a different manner to afford polynuclear aromatic
(2) Stiles, M.; Burckhardt U.; Haag, A. J. Org. Chem. 1962, 27, 4715.
(3) Mitchell, G. H.; Sondheimer, F. J. Am. Chem. Soc. 1969, 91, 7520.
(4) Wang, K. K.; Liu, B.; Petersen, J. L. J. Am. Chem. Soc. 1996,
118, 6860.
10.1021/jo061334v CCC: $33.50 © 2006 American Chemical Society
Published on Web 10/07/2006
J. Org. Chem. 2006, 71, 8417-8423
8417

Lee et al.
SCHEME 3
SCHEME 4
SCHEME 5
intermediates were successfully carried out by using 2,3,4,5-
tetraphenylcyclopentadienone as a trapping reagent.
product 7 as a major product (Scheme 3).5 Diradical intermedi-
ates 8 were proposed to account for the formation of 7. No direct
evidence for the formation of phenanthrocyclobutadiene deriva-
tive 9 was provided.
Results and Discussion
Although the reaction of 6 with (η5-C5H5)Rh(CO)2 led to an
unequivocal η4-C4-ring complex 10,6 whether 10 was formed
through the intermediate 9 is uncertain.7 On the other hand, the
reaction of 6 with air in the presence of RhCl3‚H2O and Aliquat
336 afforded 11 in 60% yield.8 All of these reactions suggested
that proximity interactions between two adjacent acetylene
moieties were significant.
Similar thermal cyclization of 1,8-bis(phenylethylenyl)naph-
thalene (12) proceeded to give rise to polynuclear aromatic
compound 13. A diradical intermediate 14 was proposed for
this reaction (Scheme 4). A repeated attempt to entrap the
reactive cyclobutadiene intermediate 15 with iron pentacarbonyl
only gave rise to the acecycloneirontricarbonyl (16) instead of
cyclobutadieneirontricarbonyl (17).9
Preparation of 4,4′-Disubstitued 2,2′-Bis(phenylethynyl)-
biphenyls. Iodination of 4,4′-dinitrobiphenyl (18)10 was achieved
with a two-step sequence reaction of iodonium salt formation11
followed by nucleophilic aromatic substitution12 with CuI/NaI
to afford 19 in 70% yield (Scheme 5). Since the Sonogashira
coupling of 19 with phenylacetylene would be influenced by a
side reaction of palladium-catalyzed intramolecular cyclization,13
copper phenylacetylide was therefore used instead as the reagent
in the Castro coupling reaction to give 20 in 57% yield.14 The
nitro groups of 20 were reduced by SnCl2‚2H2O to give diamine
21 in 72% yield.15 Compound 21 was subjected to diazotiza-
tion-iodination conditions using HCl/NaNO2/KI as reagents to
We hereby revisit the [2 + 2] thermal cycloaddition reactions.
Trapping experiments for the reactive phenanthrocyclobutadiene
(10) (a) Gull, H. C.; Turner, E. E. J. Chem. Soc. 1929, 491. (b) Mizuno,
Y.; Simamura, O. J. Chem. Soc. 1958, 3875.
(11) (a) Koser, G. F. In The Chemistry of Halides, Pseudo-halides and
Azides; Patai, S., Rappoport, Z., Executive Eds.; Wiley & Sons: Chichester,
1983; Part 1, p 721. (b) Koser, G. F. In The Chemistry of Halides, Pseudo-
halides and Azides; Patai, S., Rappoport, Z., Executive Eds.; Wiley &
Sons: Chichester, 1983; Part 2, p 1265. (c) Ajemian, R. S.; Boyle, A. J. J.
Org. Chem. 1959, 24, 1818.
(5) (a) Kandil, S. A.; Dessy, R. E. J. Am. Chem. Soc. 1966, 88, 3027.
(b) White, E. H.; Sieber, A. A. F. Tetrahedron Lett. 1967, 28, 2713. (c)
Blum, J.; Baidossi, W.; Badrieh, Y.; Hoffman, R. E.; Schumann, H. J. Org.
Chem. 1995, 60, 4738.
(6) (a) Rausch, M. D.; Gardner, S. A.; Tokas, E. F.; Bernal, I.; Reisner,
G. M.; Clearfield, A. J. Chem. Soc., Chem. Commun. 1978, 187. (b) Reisner,
G. M.; Bernal, I.; Rausch, M. D.; Gardner, S. A.; Clearfield, A. J.
Organomet. Chem. 1980, 184, 237.
(12) Larkem, A.; Larkem, H. Indian J. Chem., Sect. B 2002, 41B, 175.
(13) (a) Dougherty, T. K.; Lau, K. S. Y.; Hedberg, F. L. J. Org. Chem.
1983, 48, 5273. (b) Ko¨nig, B.; Bubenitschek, P.; Jones, P. G. Liebigs. Ann.
1995, 195. (c) Chou, M.-Y.; Mandal, A. B.; Leung, M.-k. J. Org. Chem.
2002, 67, 1501.
(7) Mu¨ller, E. Synthesis 1974, 11, 761.
(8) Baidossi, W.; Schumann, H.; Blum, J. Tetrahedron 1996, 52, 8349.
(9) (a) Bossenbroek, B.; Shechter, H. J. Am. Chem. Soc. 1967, 89, 7111.
(b) Bossenbroek, B.; Sanders, D. C.; Curry, H. M.; Shechter, H. J. Am.
Chem. Soc. 1969, 91, 371.
(14) Stephens, R. D.; Castro, C. E. J. Org. Chem. 1963, 28, 3313.
(15) Ge, J. J.; Li, C. Y.; Xue, G.; Mann, I. K.; Zhang, D.; Wang, S.-Y.;
Harris, F. W.; Cheng S. Z. D.; Hong, S.-C.; Zhung, X.; Shen, Y. R. J. Am.
Chem. Soc. 2001, 123, 5768.
8418 J. Org. Chem., Vol. 71, No. 22, 2006

Dessy-White Intramolecular Acetylene-Acetylene [2 + 2] Cycloadditions
SCHEME 6
tatetraene products were isolated in the reactions of 20 and 21
under similar conditions.
The structural identification was further supported by X-ray
crystallographic analysis. Single crystals of 28 were successfully
created by slow evaporation of the solvent from a corresponding
chloroform solution (Figure 1). The tub-shape cyclooctatetraene
skeleton (C4-C5, C19-C24) was clearly shown. In addition, the
product containing a phenanthrene ring (C1-C14), two A-rings
(C37-C42 and C43-C48), two B-rings (C25-C30 and C55-C60),
and two C-rings (C31-C36 and C49-C54) was the same as
proposed before. The geometry of the tub-shape eight-membered
ring was normally shaped with four sets of carbon-carbon π
bonds localized between C4-C5, C19-C20, C21-C22, and C23-
C24. The bond lengths were 1.364(5), 1.354(5), 1.353(6), and
1.342 (5), respectively.
The relative orientation of the phenyl groups and the
phenanthrene group clearly explained the NMR pattern of 26
described in the previous section. The important finding to
mention is the fact that the forth group of the aryl signals of 26
were abnormally upfield shifted to δ ) 6.32-6.81 ppm.
According to the crystal structure of 28, we concluded that the
phenyl protons on ring A were placed right over the phenan-
threne units, so therefore the phenyl protons fell into a shielded
region that arise from the diamagnetic anisotropic field of the
phenanthrene moiety. This led to the unusual upfield shifts.
afford diiodo compound 22.16 Although the percent yield of this
step is relatively low, the product was easily purified by liquid
chromatography to create yellowish crystals. The iodide group
was then removed by iodo-lithium exchange, followed by
protonation of the corresponding aryllithium intermediate to
give 6.17
Carbomethoxy-substituted 23 was prepared by a similar
coupling reaction of the corresponding 2,2′-diiodobiphenyl 24
with copper phenylacetylide in 75% yield (Scheme 6).18
Thermal Cyclization and Trapping Reaction of 4,4′-
Disubstituted 2,2′-Bis(phenylethynyl)biphenyls with 2,3,4,5-
Tetraphenylcyclopenta-2,4-dienone. Although benzocyclob-
utadiene had been trapped by many reactive dienes such as
cyclopentadiene, furan, and 1,3-diphenylisobenzofuran,19 an
attempted trapping experiment of 9 with 1,3-diphenylisoben-
zofuran was proved to be unsuccessful.19e Since 2,3,4,5-
tetraphenylcyclopenta-2,4-dienone (25) had been proved as a
good trapping reagent for benzocyclobutadiene, we therefore
adopted 25 as the trapping reagent for 9.20 In a typical trapping
experiment (Scheme 7), a mixture of 2,2′-bis(phenylethynyl)-
biphenyls (6) and an excess amounts of 25 under neat conditions
under nitrogen was heated at 260-270 °C for 4 h. Two major
products were isolated by liquid chromatography on silica gel.
The first product was identified as the polynuclear aromatic
adduct 7, which has been reported in other studies.5b The second
1
major product isolated in 12% yield was identified with an H
In addition to that, strong NOE enhancements between the
phenanthrene doublet at 8.10 with the phenyl doublets at 6.32
and 6.95 were observed in the NOESY spectrum of 26, which
indicated that protons on the A-rings, B-rings, and phenanthrene
were spatially close to each other. This was consistent with the
prediction based on the tub-shape structure.
NMR spectrum with a relatively symmetrical pattern. On the
basis of the 1H,1H-COSY experiment, the aromatic resonance
signals was assigned into four groups: (1) δ 8.74 (d, 2H), 8.10
(d, 2H), 7.60 (t, 2H), 7.44 (t, 2H); (2) 7.20 (t, 2H), 7.14 (t, 4H),
7.07 (d, 4H); (3) 6.95 (m, 6H), 6.87 (t, 4H); (4) 6.81 (t, 2H),
6.68 (t, 4H), 6.32 (d, 4H). These 1H NMR patterns were in good
agreement with structure 26, in which one group of phenan-
threne protons and three groups of phenyl protons were
expected. Similar cyclooctatetraene products 27 and 28 were
found in the other trapping experiments. The yield percentage
of diiodo derivative 27 was relatively high in the trapping
reaction. On the contrary, for unknown reasons no cyclooc-
Mechanistic Aspects. Benzocyclobutadiene is a well-known
reactive antiaromatic dienophile that would form Diels-Alder
cycloadducts with the reactive cyclopentadienones, even under
very mild conditions.20,21 As shown in Scheme 8, bridged
ketones were formed accordingly as the adduct under the
reaction conditions.20 The thermal stabilities of the bridged
ketone cycloadducts were different, depended on the substituents
on the cyclopentadienone. For example, adducts 31 and 32 were
stable and could be identified at room temperature. Thermal
decarbonylation in diglyme at 160 °C gave 34 and 35 as its
products. Adduct 33 from acecyclone, on the other hand, was
decarbonylated directly under the reaction conditions to form
cyclooctatetraene 36. When cyclobutadiene 9 was formed from
the intramolecular acetylene-acetylene cyclization of 6 at high
temperature, 9 was trapped in a similar fashion by 25 to give
bridged ketone cycloadduct 37 (Scheme 9). However, due to
the steric repulsions between the phenyl rings, it was expected
(16) Vogel, A. I.; Furniss, B. S.; Hannaford, A. J.; Rogers, V.; Smith,
P. W. G.; Tatchell, A. R. Vogel’s Textbook of Practical Organic Chemistry,
5th ed.; Longman: England, 1989; p 928.
(17) Wakefield, B. J. Organolithium Methods; Academic Press: London,
1988.
(18) Compounds 23 and 24 were synthesized according to Marvel’s
procedure. (a) Kwong, C.-Y.; Chan, T.-L.; Chow, H.-F.; Lin, S.-C.; Leung,
M.-k. J. Chin. Chem. Soc. 1997, 44, 211. (b) Hanson, M. P.; Marvel, C. S.
J. Polym. Sci., Polym. Lett. Ed. 1978, 16, 653. (c) Lin, S.-C.; Marvel, C. S.
J. Polym. Sci., Part A: Polym. Chem. 1979, 17, 2337.
(19) (a) Nenitzescu, C. D.; Avram, M.; Dinn, D. Chem. Ber. 1957, 90,
2541. (b) Cava, M. P.; Mitchell, M. J. J. Am. Chem. Soc. 1959, 81, 5409.
(c) Cava, M. P.; Pohlke, R. J. Org. Chem. 1962, 27, 1564. (d) Miyamoto,
T.; Odaira, Y. Tetrahedron Lett. 1973, 1, 43. (e) Cava, M. P.; Mangold, D.
Tetrahedron Lett. 1964, 26, 1751.
(21) (a) Cava, M. P.; Mitchell, M. J. Cyclobutadiene and Related
Compounds; Academic Press: New York, 1967. (b) Toda, F.; Garratt, P.
Chem. ReV. 1992, 92, 1685.
(20) Barton, J. W.; Lee, D. V.; Shepherd, M. K. J. Chem. Soc., Perkin
Trans. 1 1985, 7, 1407.
J. Org. Chem, Vol. 71, No. 22, 2006 8419

Lee et al.
SCHEME 7
SCHEME 8
FIGURE 1. ORTEP diagram of 28.
SCHEME 9
unit. The formation of 9 directly through [2 + 2] cycloaddition
or through a tetrahedrane intermediate 38, on the other hand,
was feasible at high temperature due to the proximity of the
acetylene units (Scheme 10).5a,b Of course, the possibility of
having the diradical intermediate Z,Z-8 trapped by 25 to lead
to the same product cannot be completely eliminated.
All these unsolved questions lead us to further investigate
the involving mechanism. To compare the results, we adopted
22 as a probe to carry out a series of mechanistic studies under
different conditions at a lower temperature of 150 °C in PhOPh.
The results are summarized in Tables 1 and 2. Although the
percent yields of 27 were lower in comparison, the amounts of
unidentified products were reduced under these conditions. To
examine if any diradical intermediates were involved in the
Diels-Alder cycloaddition reaction, we employed 2,4,6-tri-tert-
butylphenol (TBP) as a radical scavenger to test the assumption.
The results are denoted in the last entry of Table 1. If the
diradical intermediates were involved in the bimolecular cy-
cloaddition process, the diradicals would be trapped by the
scavengers, and therefore, the cycloaddition with 25 would be
suppressed. However, our experimental results in Table 1
revealed that the presence of scavenger did not alter the isolated
yields of 27, leading to a contradictory implication against the
diradical mechanisms. The almost identical isolated yields of
29 in all three cases are worthy of note. The presence of 25 did
not inhibit the formation of 29, indicating that 27 and 29 were
formed in noncompetitive fashion. These unusual results urged
us to follow the reaction kinetics in detail.
SCHEME 10
that thermal decarbonylation would subsequently occur at high
temperature and give rise to 26. In other words, the present
results of the trapping reactions imply that the formation of 9
from 6 is feasible under thermal conditions.
The mechanism for the formation of 9 from 6 was not
immediately obvious; nevertheless, the bisallenlic process in the
Sondheimer cyclization was unlikely because it required an
intermediate that can break down the aromaticity in the biphenyl
In the kinetic studies, we compared the decay rate of 22 in
the presence and in the absence of 25. The two parallel
experiments were run simultaneously. To calibrate the amount
8420 J. Org. Chem., Vol. 71, No. 22, 2006

Dessy-White Intramolecular Acetylene-Acetylene [2 + 2] Cycloadditions
TABLE 1. Effects of Radical Scavenger TBP on the Cycloaddition
Reaction of 22 (0.088 M) with 25 in PhOPh at 150 °C
isolated yield (%)
entry
25 (equiv)
TBP (equiv)
3
27
29
1
2
3
50
55
51
8
11
18
17
TABLE 2. Chemical Kinetics Data for the Intramolecular
Cyclization of 22 (0.070 M) and the Cycloaddition Reaction of 22
(0.070 M) with 25 (5.5 equiv) in PhOPh at 145 °C
without 25
with 25
time (min)
22a
29a
22a
29a
27a
FIGURE 2. Relative plot of the calculated heats of formation of
compounds 6, 7, 9, 38, and 39.
0
40
110
210
100
74
31
0
9
51
60
100
69
31
9
0
10
47
63
0
2
7
10
10
a Against the internal standard 24 (0.052 M).
of all the components in the reaction mixture, we adopted 24
as an internal standard. The amounts of 22, 27, and 29 were
determined on the basis of their integrations at δ ) 8.00 (d, J
) 1.5 Hz, 2H), 8.41 (d, J ) 1.7 Hz, 2H), and 9.00, 9.01(2s,
2H) ppm, respectively, against the standard at δ ) 8.63 (s, 2H)
ppm. These signals corresponded to the aromatic protons
adjacent to the iodo groups. The experiments were run at 145
°C under argon in order to avoid any interference from oxygen.8
The reaction mixtures were sampled using a syringe and were
dissolved in CDCl3 for 1H NMR measurements. The results are
summarized in Table 2. Some critical observations based on
the experimental results can be concluded as follows: (1) the
rates of decay of 22 were very close in both cases; (2) the rates
of formation of 29 were almost identical in two experiments;
and (3) the presence of 25 did not suppress formation of 29.
FIGURE 3. Estimation of the energy of Z,Z-8 by mapping the Ca-
Ca′ bond formation from 6 and the Ca-Ca′ bond cleavage of 9.
Semiempirical Calculations. To explore the plausible reac-
tion intermediates, semiempirical PM5 calculations22 were
carried out in order to estimate the heats of formation of the
related structures. According to other studies, compounds 6, 7,
9, 38, 39, and the diradical intermediates 8 are targets for
evaluation. The closed-shell structures of 6, 7, 9, 38, and 39
could be optimized by Hartree-Fock methods, and the estimated
heats of formation (∆Hf°) of 186, 114, 175, 245, and 184 kcal/
mol were respectively calculated (Figure 2). Perhaps due to the
large ring-strains of the tetrahedran skeleton, the calculated ∆Hf°
of 38 is 59 kcal/mol higher than that of diyne 6. Due to the
high heat of formation of 38, involvement of 38 as an
intermediate in the reactions was unlikely. The estimated ∆Hf°
of 9 and 39, on the other hand, were slightly lower than that of
6 by 11 and 2 kcal/mol, respectively. We attribute the relatively
high ∆Hf° of 6 to the weak carbon-carbon π bonds of the
acetylene groups.23 Although 7 was thermodynamically more
stable than 6 by 72 kcal/mol, it seemed to be formed from 6 in
stepwise fashion. The most likely intermediate was 39, which
could be formed either through a concerted or stepwise diradical
mechanism from 6.
Although there was a possibility of involvement of the
diradical 8 in the reaction, direct calculation for the heat of
formation of the diradicals are complicated. Therefore, we
attempted to estimate the energy of the diradicals by mapping
the reaction profile. In this approach, formation of Z,Z-8 from
9 was mapped by locking the Ca-Ca′ bond length (L) at a
specified value and optimized the remaining structural param-
eters by PM5 methods. The calculated ∆Hf°’s that corresponded
to each specific L were plotted in Figure 3. As expected, the
∆Hf° increased when L increased gradually. When L was kept
shorter than 2.0 Å, the optimized geometry fell into a distorted
cyclobutadiene-like structure 40, with the dihedral angle θ
around 2 ( 1° for the nearly coplanar biphenyl moiety. On the
other hand, when L was set around 2.1 Å, two structures with
similar heats of formation could be identified. One was in the
form of 40 while the other was in the form of 41. When L was
locked at a distance larger than 2.2 Å, the distorted diynelike
skeleton 41 became energetically charged. In these cases, the
biphenyl unit was twisted, with the dihedral angle θ ≈ 53-
54°. A reaction profile could therefore be constructed. When
the phenylacetylene carbons Ca and Ca′ of 41 became close to
a certain short distance around 2.1 Å, a crossover from 41 to
40 was expected. The form of diradical Z,Z-8 occurred as a
(22) The calculations were carried out by using the package software of
CAChe developed by Oxford Molecular Ltd. and Fujitsu Ltd.
(23) (a) Zavitas, A. A. J. Phys. Chem. A 2003, 107, 897. (b) Afeefy, H.
Y.; Liebman, J. F.; Stein, S. E. Neutral Thermochemical Data. In NIST
Chemistry WebBook; Linstrom, P. J., Mallard, W. G., Eds.; NIST Standard
Reference Database 69; National Institute of Standards and Technology:
Gaithersburg, MD 20899, July 2001 (http://webbook.nist.gov). (c) Luo, Y.
Handbook of Bond Dissociation Energies in Organic Compounds; CRC:
Boca Raton, 2003.
J. Org. Chem, Vol. 71, No. 22, 2006 8421

Lee et al.
reaction. Since phenanthrocyclobutadiene 43 was the key
intermediate involved in the Diels-Alder reaction step, forma-
tion of the adduct 27 would not be interrupted by TBP, the
radical scavenger.
In summary, our results provided a mechanistic point of view
for the Dessy-White intramolecular acetylene-acetylene [2 +
2] cycloaddition reaction. The reaction may involve formation
of a phenanthrocyclobutadiene intermediate from two proximate
acetylene groups at high temperature.
Experimental Section
Preparation and Identification of 6, 19-23, 28, and 30. The
corresponding synthetic procedures and characterization data are
reported in the Supporting Information.
FIGURE 4. Tentatively proposed mechanisms for the Dessy-White
intramolecular cycloadditions of 22.
Thermal Cycloaddition Reactions of 4,4′-Disubstituted 2,2′-
bis(phenylethynyl)biphenyls with 2,3,4,5-Tetraphenylcyclopenta-
2,4-dienone. 9-Phenyldibenz[a,c]anthracene (7) and 9,10,11,-
12,13,14-Hexaphenylcycloocta[l]phenanthrene (26). A General
Procedure under Neat Conditions. Reaction of 2,2′-bis(phenyl-
ethynyl)biphenyl (6) (0.12 g, 0.35 mmol) with 25 (1.42 g, 3.71
mmol) was carried out under nitrogen at 260-270 °C for 4 h. The
mixture was purified by column chromatography on silica gel using
toluene/hexane (1:10) as the eluent to produce 7 (rf ) 0.59) and
26 (Rf ) 0.18). Compound 7 was isolated as a slightly greenish
solid (0.04 g, 33%). The spectra were identical with literature data:
5c mp 228-230 °C (lit.5cmp 229 °C); 1H NMR (400 MHz, CDCl3)
δ 9.12 (s, 1H), 8.72 (d, J ) 8.6 Hz, 1H), 8.48 (d, J ) 8.6 Hz, 1H),
8.43 (d, J ) 6.9 Hz, 1H), 8.10 (d, J ) 8.2 Hz, 1H), 7.73 (d, J )
8.7 Hz, 1H), 7.64-7.52 (m, 7H), 7.44-7.37 (m, 4H), 6.98 (dt, Jd
) 1.4 Hz, Jt ) 7.8 Hz, 1H); 13C NMR (100 MHz, CDCl3) δ 142.3,
137.3, 132.3, 131.9, 131.7, 131.3, 130.7, 130.5, 129.3, 128.3, 127.7,
127.6, 127.4, 127.3, 126.9, 126.8, 126.1, 125.9, 125.4, 124.0,
123.23, 123.21, 122.0; HRMS (EI) m/z calcd for C28H18 354.1409,
1
found 354.1399. Compound 26 is a white solid (0.03 g, 12%): H
FIGURE 5. Linear correlation plot of ln([22]o/[22]t) versus reac-
NMR (500 MHz, CDCl3) δ 8.74 (d, J ) 8.0 Hz, 2H), 8.10 (d, J )
8.0 Hz, 2H), 7.60 (t, J ) 7.0 Hz, 2H), 7.44 (t, J ) 7.5 Hz, 2H),
7.19 (t, J ) 7.5 Hz, 2H), 7.14 (t, J ) 8.0 Hz, 4H), 7.06 (d, J ) 7.5
Hz, 4H), 6.96-6.92 (m, 6H), 6.87 (t, J ) 7.5 Hz, 4H), 6.81 (t, J
) 7.5 Hz, 2H), 6.68 (t, J ) 8.0 Hz, 4H), 6.32 (d, J ) 7.5 Hz, 4H);
13C NMR (125 MHz, CDCl3) δ 143.7, 143.0, 140.1, 139.7, 139.0,
138.8, 138.2, 131.2, 130.5, 130.4, 130.2, 130.0, 128.1, 128.0, 127.6,
127.2, 127.1, 126.7, 126.4, 126.2, 126.1, 122.7; HRMS (EI) m/z
calcd for C56H38 710.2974, found 710.2950.
tion time.
transition state or a short-life reactive intermediate in the reaction
profile, with an energy about 20 kcal/mol higher than that of 6.
Aspects from the Kinetics Study. The above experimental
results serve as very useful information for constructing a view
of reaction mechanisms. First of all, as mention before, since
27 and 29 were formed in noncompetitive fashion, the assump-
tion of formation of 27 and 29 through competition of a common
intermediate should be eliminated. To explain for the above
observations, we tentatively propose a mechanism as shown in
Figure 4.
Two parallel electrocyclic reactions of 22 occurred to lead
to highly reactive 42 and 43 that would further react quickly to
yield the final products. The electrocyclic reactions were
irreversible and rate-determining. According to this mechanism,
a first-order decay of 22, expressed as -d[22]/dt ) (kb + kc)-
[22], was expected. Transformation of the rate equation into
the integrated form gave ln([22]o/[22]t) ) (kb + kc)t, in which
[22]o and [22]t were the concentrations of 22 at time equal to 0
and t, respectively. This prediction was consistent with the linear
correlation in the plot of ln([22]o/[22]t) versus reaction time
shown in Figure 5. Since the decay of 22 was a result of the
intramolecular processes, the decay rate should be constant,
regardless of the presence or absence of the Diels-Alder
trapping reagent 25.
2,7-Diiodo-9-phenyldibenz[a,c]anthracene (29) and 2,7-Di-
iodo-9,10,11,12,13,14-hexaphenylcycloocta[l]phenanthrene (27).
Reaction of 22 (0.12 g, 0.20 mmol) with 25 (0.48, 1.25 mmol)
was carried out in diphenyl ether (1.5 mL) under nitrogen at 270
°C for 4 h. The mixture was cooled to room temperature and washed
with hexane. The insoluble solid was filtered, collected, and washed
with acetone to obtain insoluble 29 as a greenish solid (0.01 g,
1
8%): mp 304-306 °C; H NMR (400 MHz, CDCl3) δ 9.01 (s,
1H), 9.00 (s, 1H), 8.13-8.04 (m, 3H), 7.88-7.84 (m, 2H), 7.74
(d, J ) 8.7 Hz, 1H), 7.67-7.56 (m, 5H), 7.47-7.45 (m, 1H), 7.38
(d, J ) 7.4 Hz, 2H). The amounts of the isolated products are not
enough to provide a high-quality 13C NMR spectrum. However,
the pattern and the high-resolution mass spectrum support the
assignment. HRMS (EI): m/z calcd for C28H16I2 605.9341, found
605.9341. The filtrate was collected, concentrated under reduced
pressure, and purified by column chromatography on silica gel using
CH2Cl2/hexane (1:3) as the eluent to produce 27 (Rf ) 0.23) as a
white solid (0.04 g, 23%): 1H NMR (400 MHz, CDCl3) δ 8.41 (d,
J ) 1.8 Hz, 2H), 8.36 (d, J ) 8.8 Hz, 2H), 7.84 (dd, J ) 1.7, 8.7
Hz, 2H), 7.20 (t, J ) 7.1 Hz, 2H), 7.14 (t, J ) 7.4 Hz, 4H), 7.02
(d, J ) 6.9 Hz, 4H), 6.99-6.96 (m, 2H), 6.94-6.85 (m, 10H),
6.75 (t, J ) 7.7 Hz, 4H), 6.34 (d, J ) 7.2 Hz, 4H); 13C NMR (125
MHz, CDCl3) δ 144.6, 142.7, 139.6, 139.2, 138.5, 138.1, 137.8,
137.0, 135.1, 132.9, 130.3, 130.3, 129.9, 128.9, 128.2, 127.9,
The percent yield of 29, on the other hand, was governed by
the constant rate ratio of kc/(kb + kc). Therefore, the formation
of 29 would not be affected by the Diels-Alder trapping
8422 J. Org. Chem., Vol. 71, No. 22, 2006

Dessy-White Intramolecular Acetylene-Acetylene [2 + 2] Cycloadditions
127.44, 127.35, 126.8, 126.5, 124.2, 92.9; HRMS (EI) m/z calcd
for C56H36I2 962.0906, found 962.0884.
26-28, and 30; H spectrum of 29; 1H-1H COSY, NOESY, HMBC,
and HSQC spectra of 26; CIF data of 28; and PM5 coordinates of
6, 7, 9, 38, and 39. This material is available free of charge via the
Internet at http://pubs.acs.org.
Acknowledgment. We thank the National Science Council
and Ministry of Education of Taiwan for the financial support.
Supporting Information Available: Synthetic procedures for
1
6, 19-23, 28, and 30; H and 13C NMR spectra of 6, 7, 19-23,
JO061334V
J. Org. Chem, Vol. 71, No. 22, 2006 8423