Exceptional molecular architectures via cycloadditions to pyrenequinones

Article abstract of DOI:10.1016/j.tet.2010.07.046

The thermal reaction of phencyclone (2) with a 1:1 mixture of 1,8-pyrenequinone (4) and 1,6-pyrenequinone (5) yields 2:1 adducts only of compounds 2 and 4. The observed polycyclic aromatic hydrocarbon 8 is formed via double Diels-Alder addition of 2 to 4, and the polycyclic ketone 9 arises from a combination of Diels-Alder and hetero-Diels-Alder reactions of 2 and 4. In contrast, Lewis acid-catalyzed reactions of 2, 4, and 5 give 2:1 adducts only of 2 and 5. The chief product, polycyclic diketone 10, is derived from a double hetero-Diels-Alder addition of 2 to 5. X-ray analysis of compound 8 shows it to be an exceptionally large polycyclic aromatic arch, and the X-ray structure of 10 reveals it to be a chiral molecular tweezer.

Full text of DOI:10.1016/j.tet.2010.07.046

Tetrahedron 66 (2010) 7933e7938
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Exceptional molecular architectures via cycloadditions to pyrenequinones
,
,
,b,
Qian Qin a b, Douglas M. Ho b y, Joel T. Mague a, Robert A. Pascal, Jr. a
*
a Department of Chemistry, Tulane University, New Orleans, LA 70118, USA
b Department of Chemistry, Princeton University, Princeton, NJ 08544, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 8 June 2010
Received in revised form 14 July 2010
Accepted 16 July 2010
Available online 6 August 2010
The thermal reaction of phencyclone (2) with a 1:1 mixture of 1,8-pyrenequinone (4) and 1,6-pyr-
enequinone (5) yields 2:1 adducts only of compounds 2 and 4. The observed polycyclic aromatic hy-
drocarbon 8 is formed via double DielseAlder addition of 2 to 4, and the polycyclic ketone 9 arises from
a combination of DielseAlder and hetero-DielseAlder reactions of 2 and 4. In contrast, Lewis acid-cat-
alyzed reactions of 2, 4, and 5 give 2:1 adducts only of 2 and 5. The chief product, polycyclic diketone 10,
is derived from a double hetero-DielseAlder addition of 2 to 5. X-ray analysis of compound 8 shows it to
be an exceptionally large polycyclic aromatic arch, and the X-ray structure of 10 reveals it to be a chiral
molecular tweezer.
Keywords:
Polycyclic aromatic hydrocarbons
DielseAlder reaction
Ó 2010 Elsevier Ltd. All rights reserved.
Hetero-DielseAlder reaction
1. Introduction
DielseAlder reactions of various quinones (particularly benzo-
quinone) and dienes (particularly cyclopentadienones) have been
used to prepare numerous complex polycyclic aromatic com-
pounds,1 including the highly twisted2 polycycle 3 (Scheme 1).3e5
Surprisingly, 1,8- (4) and 1,6-pyrenequinone (5) have not been
used for such syntheses, even though they are likely precursors of
large polycyclic aromatic hydrocarbons (PAHs) or ‘nano-
graphenes’.6 Perhaps this omission reflects the fact that these
quinones, though easily prepared as a 1:1 mixture,7 are much more
difficult to obtain in pure form.
To our way of thinking, purification and separation could be
put aside for one synthetic stepda double DielseAlder reaction
with phencyclone8,9 (2)dand that one step might easily gen-
erate the two highly crowded, polycyclic diketones 6 and 7
(Scheme 1), which could be further elaborated if desired. In this
paper we describe both thermal and Lewis acid-catalyzed
cycloadditions of 2 to quinones 4 and 5. Given the straightfor-
ward synthesis of 3 and similar compounds,5 the results of the
pyrenequinone reactions were at the very least unexpected, and
yielded large molecules with highly unusual shapes and struc-
tures, including a new chiral molecular tweezer and an excep-
tionally large polycyclic aromatic arch.
Ph
Ph
Ph
Ph
O
Ph
Ph
O
O
2
– 2 H2
O
3
O
1
– 2 CO
O
Ph
Ph
Ph
2
?
Ph
Ph
Ph
O
O
O
O
O
4
6
2
Ph
?
O
O
5
Ph
7
* Corresponding author. Tel.: þ1 504 862 3547; fax: þ1 504 865 5596; e-mail
Scheme 1.
address: rpascal@tulane.edu (R.A. Pascal Jr.).
y
Present address: P.O. Box 5284, Trenton, NJ 08638, USA.
0040-4020/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.07.046

7934
Q. Qin et al. / Tetrahedron 66 (2010) 7933e7938
2. Results and discussion
one singlet). This was subsequently confirmed by X-ray analysis, as
described below. Compound 8 forms by addition of two molecules
of 2 to one molecule of 1,8-pyrenequinone (4), followed by the
formal loss of two molecules of CO and two molecules of H2O.
Whether the loss of water is a true dehydration or a de-
hydrogenation followed by a deoxygenation, all of which are pos-
sible at high temperature, is unknown.
The chief products of the cycloadditions of phencyclone and the
two pyrenequinones are succinctly illustrated in Scheme 2, but
the reaction conditions, structural characterization, and possible
mechanisms of formation of these complex molecules merit some
discussion.
The more polar product (Rf z0.3; 29% yield based on 4) is quite
unusual. This bright yellow-orange compound has complex 1H and
13C NMR spectra and no apparent symmetry, a MALDI molecular
ion at m/z 950, and a high resolution ESI mass spectrum that sug-
gests a formula of C73H42O2. Although these data should have been
sufficient to form a reasonable structural proposal, in truth, the
identity of the compound baffled us until the solution of the X-ray
structure of the Lewis acid-catalyzed adduct 10 (see below). At that
point, it was obvious that this material is the hybrid DielseAlder/
hetero-DielseAlder double adduct 9 or its isomer 12. Compound 9
is derived from the 1,8-quinone 4, and 12 from the 1,6-quinone 5.
Unfortunately, no single piece of spectroscopic data serves to dis-
tinguish the two possibilities. However, GIAO calculations of the 13C
NMR spectra of compounds 9 and 12, using their HF/6-31G(d)//HF-
6-31G(d) wavefunctions, make a compelling case that the isolated
material is the adduct 9 (see the Supplementary data).
Ph
O
Ph
O
Ph
Ph
Ph
+
Ph
Ph
Ph
8
9
(M = 906)
(M = 950)
O
Ph
O
heat
Ph2O (or PhNO2)
Ph
2
+ 4/5
FeCl3 (or BF3•Et2O)
benzene
Ph
Ph
Ph
O
Ph
O
12
Ph
O
Ph
O
+
It seems, therefore, that both prominent products of the ther-
mal cycloaddition reaction are derived from 1,8-pyrenequinone.
The reaction was performed in hot nitrobenzene with essentially
the same result, and when a 1:4 mixture of the 1,8- and 1,6-pyr-
enequinones was employed, either in diphenyl ether or nitro-
benzene, the only result was to lower the yield of compounds 8
and 9. Numerous minor components were examined by MALDI-
TOF mass spectrometry, but none showed the characteristic ions at
m/z 906, 950, or 994 that correspond to double adducts to
pyrenequinone.
O
O
Ph
O
Ph
O
Ph
Ph
10
(M = 994)
11
(M = 994)
Scheme 2.
There is at present no obvious reason for the selectivity ob-
served in the thermal reactions. The transition states for the first
DielseAlder addition of phencyclone to pyrenequinones 4 and 5
were located at the B3PW91/6-31G(d) level, and they differ in en-
ergy by only 0.1 kcal/mol (see the Supplementary data). Similarly,
the transition states for the first hetero-DielseAlder additions to
the two quinones were located, and while they differ by 2.3 kcal/
mol, this value is still quite small. In any event, in both cases it is
addition to the 1,6-quinone 5 that is favored, but as just stated, only
addition products to the 1,8-quinone 4 were observed. A remaining
possibility is that the products of addition to 5 are selectively
destroyed. This seems unlikely for the ‘trans’ isomer of hydrocarbon
8, but the hetero-DielseAlder adducts might be more sensitive.
Accordingly, a sample of compound 10, generated in the Lewis acid-
catalyzed reaction, was subjected to the thermal cycloaddition
conditions in diphenyl ether. Although decomposition was ob-
served, the extent was insufficient to account for the absence of
such products from the thermal cycloaddition reactions.
2.1. Thermal cycloaddition products
When compounds 2, 4, and 5 were heated overnight in diphenyl
ether at 280 ꢀC, the starting materials were entirely consumed, and
two prominent, relatively nonpolar components (as well as several
lesser bands) were visible upon TLC analysis of the reaction mix-
ture. The less polar of these (Rf z0.7 in 2:1 hexaneseCH2Cl2; 51%
yield based on 4) is a deep red compound that gave a molecular ion
at m/z 906 in its MALDI mass spectrum (matrix: TCNQ10), but did
not yield significant ionization in an ESI mass spectrum. This be-
havior is consistent with a hydrocarbon of the formula C72H42. This
formula corresponds to a deoxygenated derivative of one of the
double adducts 6 or 7. The relatively sparse 13C NMR spectrum (31
observed resonances) suggested twofold symmetry, and the pres-
ence of three singlets in the 1H NMR spectrum (
d 7.11, 7.53, and 8.05
from the pyrene core) indicated that the correct structure was the
‘cis’ isomer 8 (the corresponding ‘trans’ isomer would show only

Q. Qin et al. / Tetrahedron 66 (2010) 7933e7938
7935
2.2. Lewis acid-catalyzed cycloaddition products
crystallized on a general position in the space group P1 (no. 2), but
the molecule possesses approximate Cs symmetry. The molecular
structure is illustrated in Figures 1 and 2.
The major products of the Lewis acid-catalyzed reaction of
compounds 2, 4, and 5 are completely different from those of the
thermal reaction. After heating the starting materials overnight
with FeCl3 in benzene11 in a sealed tube at 120 ꢀC, preparative TLC
afforded a bright yellow, relatively nonpolar product (Rf z0.4 in
benzene; 21% yield based on 5). When BF3 was employed as the
Lewis acid12 for this reaction at room temperature, the same
product was observed, but in substantially lower yield. The MALDI
mass spectrum of this material shows a molecular ion at m/z 994,
and a high resolution ESI mass spectrum suggests a formula of
C74H42O4, both indicating some sort of double adduct of 2 to the
pyrenequinones that had not, however, undergone decarbon-
ylation. The 13C NMR spectrum is composed of two very similar
subspectra in a 3:1 ratio, each of which contains one carbonyl
group. The stronger subspectrum possesses 32 resonances, clearly
indicating the presence of symmetry. The two components were
not separable by chromatography, but fortunately, the major
component crystallized as orange prisms from CH2Cl2eEtOH, and
an X-ray structure determination (see below) showed it to be
compound 10 (Scheme 2), a cis double hetero-DielseAlder adduct
of 2 to 1,6-pyrenequinone (5). There is little doubt that the other
isomer is 11, the trans adduct to 5, because the corresponding
double adducts to 1,8-quinone 4 are extremely strained. For
example, B3PW91/6-31G(d) calculations indicate that compound
13 (which is difficult enough even to draw) is 58 kcal/mol higher in
energy than the observed adduct 10!
Figure 1. Molecular structure of compound 8. Thermal ellipsoids are drawn at the 50%
probability level.
Figure 2. Stereoview of compound 8.
Although many crowded polycyclic aromatic hydrocarbons
(PAHs) adopt nonplanar shapes, typically twists and saddles,2
compound 8 possesses a unique geometry: it is a very large PAH
arch. The breadth and depth of this arch is most clearly seen in
Figure 2. The arch is 16.3 Å wide at the base, as measured from C7 to
C36 (see Fig. 1), and its depth is 3.6 Å, as measured from the cen-
troid of C7 and C36 to the centroid of the pyrene ‘keystone’ ring
C20eC25 directly above. These dimensions are unprecedented for
an unconstrained, or ‘free-standing’, molecular arch formed from
a single polycyclic aromatic ring system. Two literature structures
illustrate this point. If one is willing to constrain a PAH by in-
corporation into a fullerene or cyclophane, then it may be strongly
bent. Perhaps the most exceptional structure of this type is Bod-
well’s cyclophane 16,13 which is 9.0 Å wide at the base and 5.4 Å in
depth. However, if no constraining strap is employed, then the
previous record holder for a free-standing arch is the saddle-
shaped polycycle 17,14,15 for which the maximum width and depth
are 14.1 Å and 2.0 Å, respectively, well short of the dimensions of
compound 8.
O
O
Ph
O
Ph
O
Ph
Ph
13
It is easy to understand why no double adducts to 4 are ob-
served, but where are the single adducts? For this apparent selec-
tivity there is a plausible, if unproven explanation. The initial
adduct of 2 and 4 is compound 14, but because a second hetero-
DielseAlder addition to the remaining enone is sterically blocked,
this compound eventually tautomerizes to the phenol 15, which
may undergo oxidation and/or polymerization, especially in the
presence of FeCl3.
R
O
N
O
R
N
O
O
I
I
I
OH
O
H
O
O
I
Ph
O
Ph
O
O
N
R
O
16
Ph
O
N
R
O
Ph
17
2.4. Molecular and crystal structure of compound 10
14
15
Compound 10 gave orange plates from solution in
a
2.3. Molecular structure of compound 8
CH2Cl2eEtOH. X-ray analysis showed that the compound crystal-
lized on a general position in the space group P1 (no. 2), but the
molecule possesses approximate C2 symmetry. The molecular
structure is illustrated in Figure 3.
Compound 8 gave red-orange needles from a solution in
CHCl3eDCEeether. X-ray analysis showed that the compound

7936
Q. Qin et al. / Tetrahedron 66 (2010) 7933e7938
insolubility of the product limited its characterization to a MALDI-
TOF mass spectrum (m/z 892) that clearly shows the loss of 14
hydrogen atoms. The seven presumed sites of dehydrogenation are
marked with bold bonds.
Cu(OTf)2
AlCl3
CS2
8
Figure 3. Molecular structure of compound 10. Thermal ellipsoids are drawn at the
50% probability level, and hydrogen atoms have been omitted for clarity.
18
The core of the molecule consists of a flat pyrene base and,
roughly perpendicular to it, two flanking cyclopentaphenanthrenes.
This relatively rigid core is extended by two of the four attached
phenyl groups, and overall the molecule clearly falls into the class of
‘molecular tweezers’.16 Indeed, compound 10 is a chiral molecular
tweezer17 prepared in only two steps from commercial starting
materials. More importantly, the sides of its chiral cleft contain
carbonyl groups that permit further functionalization; for example,
hydride reduction of these carbonyls will flank the cleft with two
inwardly directed hydroxyl groups that might aid in guest
complexation.
Compound 10 has not yet been resolved, but its potential to
form clathrates rather than to self-associate (a problem with many
tweezers) is evident from its crystal structure. Figure 4 illustrates
one of the boxlike channels in the crystal that are filled with
dichloromethane molecules. The channel’s cross-section is 7 Å by
7 Å, and it lies roughly parallel to a body diagonal of the unit
cell. The pyrene base of 10 is wide enough that the cyclo-
pentaphenanthrene wing of one molecule cannot nest between
two wings of another while maintaining van der Waals contact to
both. Such dimensions in a rigid molecule promote the formation of
cavities or channels in its crystals, and, in the case of a pure en-
antiomer, chiral cavities or channels that may selectively enclose
a chiral guest.
3. Conclusion
The formation of PAH 8 from the thermal reaction of phency-
clone and 1,8-pyrenequinone is at least similar to the initially
expected result, but the highly curved, arch-like structure of 8,
instead of the more common twisting distortion, was a distinctly
unexpected outcome. In contrast, the formation of hetero-Dielse
Alder products 9 and 10 is highly unusual and in some respects
unprecedented reactivity. Although cyclopentadienones occasion-
ally act as dienophiles (most commonly when undergoing
dimerization), we have found no examples of such molecules
acting as dienophiles in a hetero-DielseAlder reaction. On the
other hand,
a,b-unsaturated carbonyl compounds as dienes are
legion,19 but it is uncommon for them to react with electron-de-
ficient dienophiles such as 2. However, we have found no example
of a DielseAlder reaction of such a heterodiene where the ene was
part of an aromatic ring and addition of the dienophile caused
a significant loss of aromatic stabilization. Indeed, in the present
cases the initial additions of phencyclone to the pyrenequinones
are calculated to be highly endothermic (see the Supplementary
data), but the reaction must be driven by eventual rear-
omatization to yield the observed products. Whatever the reason
for the unusual reactions described here, any lasting impact will be
due to the easy accessibility of the structurally unique products.
4. Experimental
4.1. General
Phencyclone (2) was prepared as described previously.8,9 A 1:1
mixture of 1,8-pyrenequinone (4) and 1,6-pyrenequinone (5) was
prepared by the method of Cho and Harvey.7 The 1,6-quinone is
slightlylesspolarthanthe1,8-quinone, andsmallsamplesenrichedin
the former (up to 4:1 5:4) were prepared by collecting the lead frac-
tions from chromatography on silica gel (solvent, 50:1 tolueneeethyl
acetate). All other solvents and reagents were commercial, reagent
grade materials, and they were used without further purification. 1H
NMR spectra were recorded at 400 MHz or 500 MHz on Varian Unity
INOVA and Bruker AVANCE spectrometers, respectively; samples
were dissolved in CDCl3. 13C NMR spectra were recorded on the latter
instrument at 126 MHz. MALDI-TOF mass spectra were recorded on
a Bruker Autoflex III spectrometer; TCNQ was employed as the ma-
trix.10 High-resolution ESI-TOF mass spectra were recorded on an
Agilent 6220 spectrometer.
Figure 4. Stereoview of a dichloromethane-containing channel in the crystal structure
of 10.
2.5. Dehydrogenation of compound 8
As noted in the introduction, polycyclic aromatic compounds,
such as 8, are likely precursors of ‘nanographenes’. Accordingly,
compound 8 was subjected to dehydrogenation with copper(II)
triflate and AlCl3 in CS2.18 This gave the fully dehydrogenated
compound 18 in roughly 30% yield. Unfortunately, the extreme

Q. Qin et al. / Tetrahedron 66 (2010) 7933e7938
7937
4.2. Data for compounds
NaHCO3, the resulting mixture was extracted twice with CHCl3, and
the combined organic extracts were dried over Na2SO4 and con-
centrated. This material was purified via preparative TLC (solvent,
benzene). A bright yellow band (Rf z0.4) proved to be the double
hetero-DielseAlder adduct of the 1,6-pyrenequinone (5) and
4.2.1. Thermal DielseAlder reactions of phencyclone (2) and pyr-
enequinones
4 and 5. Phencyclone (2, 233 mg, 0.610 mmol),
a mixture of 1,8-pyrenequinone and 1,6-pyrenequinone (1:1 4:5,
70.0 mg, 0.301 mmol), and diphenyl ether (0.5 mL) were heated
overnight at 280 ꢀC in a screw-capped tube. After cooling, MeOH
was added, and the resulting dark precipitate was collected by fil-
tration. Preparative TLC (solvent, 2:1 hexaneseCH2Cl2) gave an ar-
ray of colored bands. The least polar, luminescent, red-orange band
(Rf z0.7) proved to be the ‘cis’ double DielseAlder adduct of 2 and
4, isolated as deep red crystals (compound 8, 70 mg, 0.077 mmol,
51% based on the amount of the 1,8-quinone 4 present in the
starting material). No similar adduct of the 1,6-quinone 5 was ob-
served. A more polar, bright yellow band (Rf z0.3) was isolated as
a red solid; this proved to be the mixed DielseAlder/hetero-
DielseAlder adduct of 2 and 4 (compound 9, 41 mg, 0.043 mmol,
29% based on the amount of the 1,8-quinone 4 present in the
phencyclone (2), isolated as a red-orange solid (23 mg, 23 mmol, 21%
based on the amount of 5 in the starting material). This material
appeared to be a 3:1 mixture of two isomers, most clearly observed
in the two very similar subspectra that comprise its 13C NMR spec-
trum. These are the cis (10, confirmed by X-ray) and (presumably)
trans (11) double adducts of 2 to 5. Crystals of compound 10 suitable
for X-rayanalysis were obtained from CH2Cl2eEtOH, and a relatively
clean 1H NMR spectrum of 10 was obtained by selection of crystals
from the X-ray sample. For 10: 1H NMR (CDCl3)
d
6.88 (t, J¼7 Hz, 2H),
7.14 (m, 6H), 7.23 (m,16H), 7.47 (m, 2H), 7.69 (m,10H), 7.88 (t, J¼8 Hz,
2H), 8.10 (d, J¼8 Hz, 2H), 8.27 (d, J¼8 Hz, 2H); 13C NMR (CDCl3):
d
52.7, 93.1, 113.5, 115.4, 121.6, 124.1, 124.96, 124.98, 126.26, 126.28,
127.2,127.4,127.5,127.7,128.0,128.4,128.6,128.8,129.0,129.3,131.2,
131.5, 131.86, 131.88, 132.5, 132.7, 135.6, 136.8, 140.8, 148.5, 163.7,
201.3 (32 of 33 expected resonances); MS (MALDI) m/z 994 (Mþ);
HRMS (ESI) m/z 995.3143 (MþH), calcd for C74H43O4 995.3161. For
starting material). For 8: 1H NMR (CDCl3)
d
6.93 (t, J¼7.5 Hz, 2H),
7.01 (t, J¼7.5 Hz, 2H), 7.11 (s, 2H), 7.22 (d, J¼8 Hz, 2H), 7.36 (m, 4H),
7.41 (d, J¼8 Hz, 2H), 7.47 (m, 10H), 7.53 (s, 2H), 7.55 (m, 6H), 7.60 (m,
4H), 8.05 (s, 2H), 8.29 (d, J¼7.5 Hz, 2H), 8.31 (d, J¼7.5 Hz, 2H); 13C
11: Partial 13C NMR (CDCl3):
d 52.2, 92.8,113.3,115.5,121.4,124.2, .,
NMR (CDCl3)
d
122.9, 123.3, 123.6, 125.2, 125.5, 125.6, 126.0, 126.6,
140.6, 148.4, 163.5, 201.0. When the same reaction was performed
with BF3$Et2O instead of FeCl3 as the Lewis acid catalyst, but at
a much lower temperature (ꢁ78 ꢀC to room temperature, over-
night), the same hetero-DielseAlder adducts were isolated, but in
much lower yield (4.6 mg, 4%).
126.9,127.8,127.9, 128.1,128.4,128.7,129.0, 129.1,130.1, 130.2, 130.3,
130.4, 130.6, 131.1, 131.2, 131.9, 132.1, 132.8, 133.4, 134.7, 135.8, 141.4,
144.3 (31 of 32 expected resonances); MS (MALDI) m/z 906 (Mþ).
Crystals suitable for X-ray analysis were obtained by the evapora-
tion of a solution in CHCl3eDCEeether. For 9: 1H NMR (CDCl3)
4.3. General computational methods
d
6.77 (t, J¼8 Hz,1H), 6.85 (m, 2H), 7.02 (m, 3H), 7.08 (m, 5H), 7.15 (s,
2H), 7.16 (m, 3H), 7.21 (m, 3H), 7.33 (m, 12H), 7.43 (d, J¼8 Hz, 1H),
7.50, (t, J¼8 Hz, 1H), 7.54 (s, 1H), 7.63 (m, 2H), 7.93 (d, J¼8 Hz, 1H),
7.97 (d, J¼8 Hz, 1H), 8.00 (d, J¼8 Hz, 1H), 8.07 (d, J¼8 Hz, 1H), 8.14
All ab initio and hybrid density functional calculations were
performed with GAUSSIAN 03;20 the built-in default thresholds for
wave function and gradient convergence were employed. Transition
states for cycloadditions of phencyclone and pyrenequinones were
located at the B3PW91/6-31G(d) level by using the QST3 function in
GAUSSIAN03, and therelevantpotential minimaandtransition state
structures were characterized by frequency calculations. HF/6-31G
(d)//HF/6-31G(d) wavefunctions were employed in GIAO calcula-
tions to give the absolute 13C chemical shifts for 9 and 12, and these
were converted to ordinary chemical shifts by comparison to the 13C
chemical shift computed for tetramethylsilane at the same level.
(m, 2H); 13C NMR (CDCl3):
d 52.8, 92.9, 113.1, 117.1, 122.1, 122.4,
123.4, 123.7, 123.9, 124.1, 124.6, 125.2, 125.3, 125.7, 126.1, 126.4,
126.7, 127.0, 127.1, 127.2, 127.3, 127.5, 127.72, 127.74, 127.8, 127.9,
128.0, 128.1, 128.4, 128.5, 128.70, 128.72, 128.8, 129.00, 129.02, 129.1,
129.3, 130.0, 130.3, 130.4, 130.6, 130.9, 131.37, 131.41, 131.7, 131.9,
132.0, 132.3, 132.4, 132.6, 133.2, 133.5, 133.7, 134.1, 134.3, 135.47,
135.55, 136.9, 140.4, 141.3, 144.4, 149.0, 164.0, 201.3 (64 of 65
expected resonances); MS (MALDI) m/z 950; HRMS (ESI) m/z
951.3238, calcd for C73H43O2 951.3263. When the reaction was
performed in nitrobenzene at 180 ꢀC using the 1:1 mixture of 4 and
5, the isolated yield of compound 8 was 25% (once again based on
the amount of 1,8-quinone 4 present in the starting material).
Acknowledgements
This work was supported by National Science Foundation Grant
CHE-0936862 and Petroleum Research Fund Grant 45801-AC4,
which are gratefully acknowledged.
4.2.2. Dehydrogenation of compound 8. Cu(OTf)2 (1.3 mg, 3.7 mmol)
was placed in a 100 mL round-bottom flask, and the flask was
evacuated and heated with a flame to dry the salt. AlCl3 (467 mg,
3.5 mmol) and CS2 (50 mL, previously dried over molecular sieves)
were added. The pink mixture was stirred vigorously for 15 min
and then warmed to 30 ꢀC. A solution of compound 8 (10 mg,
0.011 mmol) in CS2 (5 mL) was added via syringe, and the mixture
instantly turned dark blue. After stirring at room temperature for
36 h, MeOH (50 mL) was added to quench the reaction. The dark
precipitate was collected by filtration, and it was washed succes-
sively with 10% NH4OH, 1 N HCl, water, EtOH, CS2, and CH2Cl2. The
remaining black, insoluble solid was dried overnight, and proved to
Supplementary data
(1) A PDF file containing 1H and 13C NMR spectra of compounds
8e10, the MALDI-TOF spectrumof compound 18, and full Ref. 20, and
(2) an ASCII text file containing the atomic coordinates and energies
of the calculated structures and transition states discussed in the
text. Crystallographic data (excluding structure factors) for the
structures in this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication num-
bers CCDC 779601 and 779602. Copies of the data can be obtained,
free of charge, on application to CCDC, 12 Union Road, Cambridge,
CB2 1EZ, UK [fax: þ44(0) 1223 336033 or e-mail: depos-
it@ccdc.cam.ac.uk]. Supplementary data associated with this article
can be found in online version at doi:10.1016/j.tet.2010.07.046.
be compound 18 (3 mg, 3
m
mol, e30%) as judged by mass spec-
trometry. MS (MALDI) m/z 892 (Mþ).
4.2.3. Lewis acid-catalyzed hetero-DielseAlder reaction of phencyclone
(2) and pyrenequinones
5 and 6. Phencyclone (2, 165 mg,
0.43 mmol), mixture of 1,8-pyrenequinone and 1,6-pyr-
a
References and notes
enequinone (1:1 4:5, 50 mg, 0.22 mmol), FeCl3 (3 mg), and dry
benzene (0.3 mL) were heated for 18 h at 120 ꢀC in a screw-capped
tube. After cooling, the reaction was quenched with saturated
1. Review of classical syntheses employing cyclopentadienones: Ogliaruso, M. A.;
Romanelli, M. G.; Becker, E. I. Chem. Rev. 1965, 65, 261e367.

7938
Q. Qin et al. / Tetrahedron 66 (2010) 7933e7938
2. Review of twisted acenes: Pascal, R. A., Jr. Chem. Rev. 2006, 106, 4809e4819.
3. Arbuzov, B. A.;Abramov, V. S.; Devyatov, Y. B. J. Gen. Chem. USSR 1939, 9,1559e1563.
4. Pascal, R. A., Jr.; Van Engen, D. Tetrahedron Lett. 1987, 28, 293e294.
5. Similar syntheses of crowded quinones: (a) Qiao, X.; Padula, M. A.; Ho, D. M.;
Vogelaar, N. J.; Schutt, C. E.; Pascal, R. A., Jr. J. Am. Chem. Soc. 1996, 118, 741e745;
(b) Watson, J. A., Jr.; Pascal, R. A., Jr.; Ho, D. M.; Kilway, K. V. Tetrahedron Lett.
2000, 41, 5005e5008.
6. Recent reviews of nanographenes: (a) Grimsdale, A. C.; Müllen, K. Angew. Chem.,
Int. Ed. 2005, 44, 5592e5629; (b) Wu, J.; Pisula, W.; Müllen, K. Chem. Rev. 2007,
107, 718e747; (c) Enoki, T.; Takai, K.; Osipov, V.; Baldakova, M.; Vul’, A. Chem.d
Asian J. 2009, 4, 796e804.
12. Gelman, D. M.; Forsyth, C. M.; Perlmutter, P. Org. Lett. 2009, 11, 4958e4960.
13. Merner, B. L.; Dawe, L. N.; Bodwell, G. J. Angew. Chem., Int. Ed. 2009, 48,
5487e5491.
14. Shi, Y.; Qian, H.; Li, Y.; Yue, W.; Wang, Z. Org. Lett. 2008, 10, 2337e2340.
15. This statement is based on several extensive searches of the Cambridge
Structural Database for large, nonplanar, polycyclic aromatic compounds.
However, given the idiosyncratic nature of such searches, it is hard to know
whether all possibilities have been covered.
16. Reviews of molecular tweezers: (a) Zimmerman, S. C. Top. Curr. Chem.
€
1993, 165, 71e102; (b) Klarner, F.-G.; Kahlert, B. Acc. Chem. Res. 2003, 36,
919e932.
7. Cho, H.; Harvey, R. G. J. Chem. Soc., Perkin Trans. 1 1976, 836e839.
8. Dilthey, W.; Henkels, S.; Schaefer, A. Ber. Dtsch. Chem. Ges. 1938, 71, 974e979.
9. Pascal, R. A., Jr.; Van Engen, D.; Kahr, B.; McMillan, W. D. J. Org. Chem. 1988, 53,
1687e1689.
17. Review of chiral molecular tweezers: Harmata, M. Acc. Chem. Res. 2004, 37,
862e873.
18. Simpson, C. D.; Brand, J. D.; Berresheim, A. J.; Przybilla, L.; Rader, H. J.; Müllen, K.
Chem.dEur. J. 2002, 8, 1424e1429.
19. Review of heterodiene syntheses: Desimoni, G.; Tacconi, G. Chem. Rev. 1975, 75,
€
€
10. Przybilla, L.; Brand, J.-D.; Yoshimura, K.; Rader, H. J.; Müllen, K. Anal. Chem.
2000, 72, 4591e4597.
651e692.
11. Nebois, P.; Bouaziz, Z.; Fillion, H.; Moeini, L.; Aurell Piquer, M. J.; Luche, J.-L.;
Riera, A.; Moyano, A.; Pericas, M. A. Ultrasonics Sonochem. 1996, 3, 7e13.
20. Frisch, M. J., et al. Gaussian 03, Revision C.02; Gaussian: Wallingford CT,
2004.