Steps toward the synthesis of a geodesic C60H12 end cap for a C3v carbon [6,6]nanotube

Article abstract of DOI:10.1016/j.tet.2008.09.087

Several shape-persistent carbon-rich nanomolecules with diameters exceeding 1.7 nm have been prepared by the aldol trimerization of 20-carbon aromatic ketones bearing chlorine atoms at various sites. These C₆₀H₂₇Cl₃ and C<

Full text of DOI:10.1016/j.tet.2008.09.087

Tetrahedron 64 (2008) 11360–11369
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Steps toward the synthesis of a geodesic C60H12 end cap for a C3v carbon
[6,6]nanotube
*
Thomas J. Hill, Richard K. Hughes, Lawrence T. Scott
Department of Chemistry, Merkert Chemistry Center, Boston College, 2609 Beacon Street, Chestnut Hill, MA 02467-3860, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Several shape-persistent carbon-rich nanomolecules with diameters exceeding 1.7 nm have been pre-
pared by the aldol trimerization of 20-carbon aromatic ketones bearing chlorine atoms at various sites.
These C60H27Cl3 and C60H24Cl6 polycyclic aromatic compounds represent attractive intermediates for the
synthesis of a geodesic C60H12 end cap of a C3v carbon [6,6]nanotube.
Received 16 July 2008
Received in revised form
22 September 2008
Accepted 26 September 2008
Available online 8 October 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Dedicated to Professor Reginald H. Mitchell
on the occasion of his 65th birthday
1. Introduction
10 coaxial tubes nested together.3,12,13 Carbon nanotubes from all of
these classes are already known, and their properties vary as
Carbon nanotubes have been widely touted for their potential to
fulfill dreams in materials science and in the emerging realm of
nanotechnology.1–4 They also hold considerable intrinsic scientific
interest owing to their unusual curved networks of trigonal carbon
atoms. Despite intense scrutiny by scientists and engineers
worldwide for nearly two decades, however, these fascinating
carbon-rich materials are still being made today by poorly un-
derstood empirical methods.5–8 We contend that such targets
should be accessible by rational chemical synthesis and that the
successful realization of the requisite synthetic methods will
revolutionize the science of carbon-rich materials. Our 12-step
laboratory synthesis of fullerene-C60 by chemical methods in
20029–11 represents one milestone in the journey toward that goal.
In recent years, we have begun to focus our research on the
development of chemical methods for the rational, structure-spe-
cific synthesis of single-chirality, all-carbon, single walled
nanotubes (SWNTs). The structural variety possible for carbon
nanotubes is virtually limitless.3,12,13 They come in all different dia-
meters. The orientation of the benzene rings along the shaft can be
chiral (helical) or achiral, and the chiral tubes vary according to the
pitch of the helix. Achiral tubes are classified according to the ap-
pearance of their rims as either ‘zig-zag’ or ‘armchair’. Both ends
can be open, or both can be closed, or a single tube may have one
end of each type. Superimposed on all of that, carbon nanotubes
can be single walled, double walled, or multiwalled, with more than
a function of structure.3,12,13 Unfortunately, empirical methods5–8
do not produce homogeneous samples in which all the carbon
nanotubes have the same predefined diameter and chirality (ring
orientation). The problem is compounded by the fact that, unlike
the fullerenes, carbon nanotubes made in this way cannot be sep-
arated and purified to homogeneity by chromatographic methods,
because they are totally insoluble.14,15 This problem represents
a clear challenge to synthetic organic chemists!
Where should one begin? We have chosen the achiral, armchair,
all-carbon, single walled nanotubes (e.g., [6,6]SWNTs) as our
highest priority targets. Such nanotubes are expected to exhibit
metal-like behavior, regardless of diameter.1,3,12,13 Consequently,
they all hold the potential to find use as ultra-thin, super-strong,
electrically conducting nanowires in myriad nanoelectronic de-
vices. Most zig-zag tubes and chiral tubes, by contrast, are expected
not to exhibit metal-like behavior.1,3,12,13 The choice of initial targets
thus seems obvious to us. This paper describes one approach to the
chemical synthesis of [6,6]SWNTs.
1.1. General strategy
The course we are pursuing builds on our previous experience
with syntheses of open geodesic polyarenes.11,16,17,18 Figure 1 il-
lustrates the general strategy, as applied to a proposed synthesis of
[6,6]SWNTs that are closed at one end and open at the other. As we
see it, two major hurdles must be surmounted to produce SWNTs
by this general strategy: (1) synthesis of a geodesic polyarene that
has the initial armchair rim and six fully unsaturated five-mem-
bered rings, as required for a hemisphere by Euler’s theorem19 and
* Corresponding author. Tel.: þ1 617 552 8024; fax: þ1 617 552 6454.
E-mail address: lawrence.scott@bc.edu (L.T. Scott).
0040-4020/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2008.09.087

T.J. Hill et al. / Tetrahedron 64 (2008) 11360–11369
11361
1
2
2
repetitive
addition of
C2units
Figure 1. General plan for building a hemispherical end cap and extending it to produce an armchair SWNT.
(2) extension of the rim by adding C2 units across the ‘bay regions’
to produce new six-membered rings.
represents an extreme example of the cascade effect; in our
C60H27Cl3 synthetic intermediate, three chlorine atoms were in-
corporated at strategic sites, and FVP effected the closure of 15 new
C–C bonds to produce isolable quantities of C60, with an average
yield of >60% per bond.10,11
We recognized at the outset of this project that the most difficult
part of stitching together the arms of 1 by FVP would be formation
of the first three bonds, those closest to the core, marked a in
Figure 2. Closing these three new six-membered rings would
introduce curvature comparable to that in C60, and the cascade
effect might then take over to form the remaining six bonds.
Unfortunately, the incorporation of halogen atoms as radical pre-
cursors in the crowded fjord regions of molecules related to 1 has
proven problematical, presumably for steric reasons.24,25 In such
For such a synthesis to be practical, the last step must be fully
self-replicating, i.e., it must transform hydrocarbon bay regions into
new, fully unsaturated, six-membered rings under a single set of
conditions, without intervention by the chemist. Such a process
would extend an armchair rim indefinitely, until the C2 feedstock
was consumed or the source shut off (not unlike a polymerization
with ‘living polymers’). In fact, catalytic methods for growing car-
bon nanotubes to lengths in excess of 1 mm, using acetylene as the
C2 feedstock, have already been reported,20,21 so this aspect of the
problem may already be solved, at least in principle. Conceivably,
methods may even be found to extend such a nanotube template by
uncatalyzed processes.
Herein lies the big payoff to the synthetic organic chemist. Only
a few micrograms of the appropriate geodesic polyarene ‘seed’ need
to be extended to a length of 1 mm or more in order to produce
gram quantities of homogeneous SWNTs! The SWNTs derived in
this manner will all necessarily have the same diameter and
armchair rim structure, by virtue of the growth mechanism. The
tremendous amplification in dimension (ca. 1 nm/1 mm¼106
increase) will compensate for any low yields endured earlier in the
synthesis.
What follows is a summary of our progress toward a synthesis of
the C3v nanotube end cap 2, the geodesic C60H12 hydrocarbon that
would result from ‘stitching together’ the arms of the C60H30 hy-
drocarbon 1,
a carbon-rich C2nHn molecule of nanometer
2
2
dimensions.
1.2. Retrosynthetic analysis
Figure 2 shows the loci of the nine new bonds that must be
formed in stitching up 1 to give 2.
Flash vacuum pyrolysis (FVP) has proven effective as a general
method for transforming planar polycyclic aromatic hydrocarbons
(PAHs) and halogenated derivatives thereof into their bowl-shaped
counterparts by joining together carbon atoms that lie far apart in
the planar conformations.16–18 It has been our experience, however,
that respectable yields of the desired geodesic polyarenes require
the orchestrated generation of aryl radicals (or carbenes) at the
sites where the new bonds must be formed.22,23 Once curvature has
been introduced, a ‘cascade effect’ seems to operate to close addi-
tional rings by thermal cyclodehydrogenations, without the need
for additional radical generating groups.18 Our FVP synthesis of C60
c
a
b
1
Figure 2. Nine new bonds must be formed in stitching up 1 to give 2.

11362
T.J. Hill et al. / Tetrahedron 64 (2008) 11360–11369
situations, a reliance on the 1,2-shift of hydrogen atoms in aryl
radicals under FVP conditions26 has allowed us to circumvent this
problem by attaching the halogen to an adjacent carbon atom in the
precursor (Fig. 3).27
H
H
H
Even this gambit is not possible with 1, however, because all of
the carbon atoms to be joined by a bonds are flanked on both sides
by quaternary ring junctions. These constraints prompted us to
consider whether aryl radicals might be capable of rearranging
under FVP conditions by 1,3-shifts of hydrogen atoms between the
peri-positions of a PAH (Fig. 4). If so, then a halogen atom
incorporated in each fjord region, on a terminal ring, might pro-
mote the initial a-ring closures, following 1,3(peri)-shifts of interior
hydrogens.
Figure 4. Proposed 1,3(peri)-shift of hydrogen in an aryl radical.
certainly favor those ring closures, but they might also present
synthetic difficulties. Having the chlorine atoms moved over one
position (3c) looked appealing from the synthesis perspective,
and their attachment there might still achieve the desired effect,
by exploiting the 1,2-shift of hydrogens, as in Figure 3. Accord-
ingly, we chose 3c as our initial target compound. Figure 5
summarizes our retrosynthetic analysis, which relies on an acid
catalyzed aldol cyclotrimerization24,25 to construct the triply
fused benzene ring at the center.
A preliminary examination of this process by B3LYP/6-31G
*
calculations revealed that the free energy of activation for
1,3(peri)-shift of a hydrogen atom is not higher than that for the
known26 1,2-shift of hydrogen atoms in aryl radicals (64.6 kcal/
mol and 66.0 kcal/mol, respectively).28 We subsequently obtained
experimental evidence for the feasibility of this unusual rear-
rangement,29 and that emboldened us to incorporate three
1,3(peri)-shifts of hydrogen atoms into our plan. Our synthetic
target was thus refined to include three chlorine atoms as depic-
ted in 3a. Chlorine atoms were chosen to serve as the radical
precursors in preference to more weakly bonded bromine atoms
so that the FVP precursor would be as volatile as possible and
sufficiently robust to withstand the temperatures required to
promote vacuum sublimation.
2. Results and discussion
2.1. Synthesis of hexachlorinated trimer 3c
Bromination of the commercially available 1-(2-naph-
thyl)ethanol (4) was achieved using phosphorus tribromide in
refluxing benzene to give 2-(1-bromoethyl)naphthalene (5) in 95%
yield. Subsequent displacement of the benzylic bromide with tri-
phenylphosphine gave the corresponding phosphonium salt (6) in
93% yield (Fig. 6).
We then carried out the Wittig reaction with phosphonium
salt 6, using 2,4-dichlorobenzaldehyde (7) as the coupling partner
and n-butyl lithium to preform the ylide. A mixture of E and Z
olefins (8) was obtained in 82% yield. There was no need to
separate the E and Z isomers since they equilibrate under UV light
in the next step. Photocyclization of 8 worked best in cyclohexane
as the solvent, giving the substituted [4]helicene 9 in 68% yield by
recrystallization, without any chromatography being necessary
(Fig. 7).
Before initiating a synthesis of 3a, we recognized the oppor-
tunity to incorporate radical precursors also for the c-bond
closures.
A
chlorine atom in each cove region (3b) would
Br
Br
FVP
1050 °C
With an efficient route to the [4]helicene intermediate 9 in
hand, we turned to elaborating the methyl handle into the desired
Br
Y
Y
X
X
–Br•
twice
more
Cl
Cl
Cl
X
Br
Br
H
O
Y
X
Y
Y
Cl
X
Br
Br
Cl
close
– H•
~ H [1,2]
Br
3a X = H, Y = H
3b X = Cl, Y = H
3c X = H, Y = Cl
CH3
H
Y
X
+
Ph3P
Cl
C
CH3
Br
HC
O
Figure 3. FVP synthesis of a C30H12 hemifullerene that capitalizes on 1,2-shifts of
hydrogen atoms in the intermediate aryl radicals.27
Figure 5. Retrosynthetic analysis of the FVP precursor 3.

T.J. Hill et al. / Tetrahedron 64 (2008) 11360–11369
11363
Br
PBr3
benzene
95%
PPh3
OH
Br
toluene
93%
NBS
Cl
Cl
benzoyl
peroxide
CCl4
4
5
Cl
Cl
79%
PPh3Br
9
10
N
6
C
Figure 6. Synthesis of phosphonium salt 6.
KCN,
Bu4N+ HSO4
-
H2SO4, H2O
HOAc
75%
Cl
CH2Cl2, H2O
96%
cyclic ketone 13. This was accomplished in four steps (Fig. 8), be-
ginning with benzylic bromination of 9, which gave 10 very cleanly
in 79% yield. The bromide was then displaced with cyanide ion
under biphasic conditions in 96% yield to produce nitrile 11, which
was hydrolyzed under acidic conditions, using a 1:1:1 mixture of
water, concentrated sulfuric acid, and acetic acid at reflux,30,31
giving carboxylic acid 12 cleanly in 75% yield. The final cyclization
was accomplished by a two step, one pot procedure in which car-
boxylic acid 12 was first transformed into the corresponding acid
chloride with thionyl chloride, and that product was subjected to
Friedel–Crafts acylation conditions to generate cyclic ketone 13 in
63% yield.
Cl
11
O
O
HO
1. SOCl2
2. AlCl3/CH2Cl2
63%
Cl
Cl
Cl
To access the 60-carbon target 3c, we planned to rely on the
aldol trimerization of ketone 13. Our laboratory has enjoyed good
success in the past using titanium tetrachloride as the Lewis acid
catalyst in hot o-dichlorobenzene to promote aldol trimerizations
of other aromatic ketones,9,10,22,24,25 so we treated ketone 13 with
this catalyst system at 115 ꢀC. MALDI-TOF MS analysis of the
product mixture (see Section 4) revealed that the desired cyclic
trimer 3c was indeed formed, but we also saw evidence for
Cl
12
13
Figure 8. Elaboration of [4]helicene 9 into ketone 13.
avail. Likewise, all attempts to separate and purify cyclic trimer 3c
by recrystallization proved fruitless.
We believe that the poor behavior of ketone 13 in the aldol
trimerization reaction is a direct consequence of the very poor
significant amounts of the
a,b-unsaturated dimer 14 and cyclic
tetramer 15 (Fig. 9).
This aldol trimerization reaction was repeated many times,
varying the concentrations, the rate of addition, the amount of
catalyst, the reaction time, the Lewis acid, the solvent, and the
temperature (up to 180 ꢀC); however, no significant improvement
in the ratio of 3c to other products could ever be achieved. Sepa-
ration of the very insoluble 3c from other products was attempted
using Soxhlet extraction with several different solvents but to no
solubility of the
pathway to cyclic trimer 3c. Ketone 13 itself is only sparingly sol-
uble in o-dichlorobenzene, and planar -unsaturated aldol dimers
of aromatic ketones generally exhibit even lower solubility than the
corresponding monomeric ketones. Previous studies from our
laboratories24,25 have shown that aldol trimerization reactions can
be easily derailed by the poor solubility of any intermediates along
the pathway to a cyclic trimer. Consequently, we elected to modify
our plan to address this solubility problem.
a,b-unsaturated dimer 14, which lies on the
a,b
As a general rule, the presence of multiple chlorine atoms
around the rim of a polycyclic aromatic hydrocarbon diminishes its
solubility, relative to the hydrocarbon. We therefore decided to
reduce the number of chlorine atoms in our target and aim for the
synthesis of trimer 3a. Although this would result in a 60-carbon
pyrolysis precursor having only three radical generators, we anti-
cipated that they would still promote the three initial C–C bond
closures, each following a 1,3-shift of hydrogen. Formation of the
first three new rings should then impose enough curvature on the
molecule to set up the cascade closure of the remaining six bonds
by thermal cyclodehydrogenations. Remembering that our syn-
thesis of C60 had only three chlorine atoms attached to the pyrolysis
precursor,10 we viewed 3a as still constituting a reasonable target.
n-BuLi
PPh3Br
Cl
then
H
O
Cl
Cl
82%
6
8
hν
cyclohexane
I2
Cl
68%
7
O
2.2. Synthesis of trichlorinated trimer 3a
Cl
The synthesis of monochloroketone 21 is completely analo-
gous to our synthesis of ketone 13. The same phosphonium salt
(6) was used for the Wittig reaction, and n-butyl lithium was
again used as the base. The ylide was combined this time with
2-chlorobenzaldehyde to give the new stilbene derivative 16 as
Cl
9
Figure 7. Synthesis of the substituted [4]helicene 9.

11364
T.J. Hill et al. / Tetrahedron 64 (2008) 11360–11369
Cl
Cl
As can clearly be seen from the mass spectrum, we were again
producing mixtures of -unsaturated dimer 22 (m/z 586/588), the
O
a,b
desired trimer 3a (m/z 854/856), and cyclic tetramer 23 (m/z 1136/
1138), under conditions that consumed 100% of the starting ma-
terial. This time, however, we also observed a dione of the acyclic
tetramer (24, m/z 1168/1170), which could be separated as a red
band by preparative TLC.32
This aldol trimerization reaction was repeated many times,
varying the conditions as in the attempted synthesis of 3c, but no
improvement could ever be achieved, and the challenge of isolating
pure 3a from these product mixtures proved to be insurmountable.
We again ascribe the poor results of the aldol trimerization reaction
TiCl4
Cl
+
o-DCB
115 °C
ca 70%
O
Cl
Cl
13
Cl
14
to poor solubility of the intermediate a,b-unsaturated dimer (22, in
this case). Reducing the number of chlorine atoms from six to three
did not solve the problem.
Cl
Cl
Faced with this setback, we devised a new plan to address the
frustrating solubility problem. In our experience, large polycyclic
aromatic compounds that are nonplanar exhibit significantly
greater solubility than their planar counterparts, presumably be-
cause they are less able to achieve strong intermolecular attractions
Cl
+
Cl
through
p–p stacking. Accordingly, we began to reconsider the
chlorinated derivative of 3 that bears a chlorine atom in the cove
region of each arm (3b). We reasoned that such a large substituent
protruding into such a tight space ought to enforce a severe helical
twist in each arm, thereby enhancing the solubility of the trimer
significantly. Molecular modeling28 confirms that 3b deviates from
planarity substantially more than either 3a or 3c. Moreover, as
noted earlier, the cove region chlorine atoms in trimer 3b are po-
sitioned to generate aryl radicals precisely where they are needed
to promote the c-bond closures. As an added benefit, the extra twist
might even improve the volatility of the trimer (3b), thereby fa-
cilitating sublimation in the pyrolysis. As it turned out, synthesizing
the required ketone proved less troublesome than we had origi-
nally feared.
Cl
Cl
3c
Cl
Cl
Cl
Cl
Cl
2.3. Synthesis of hexachlorinated trimer 3b
Cl
Using the same phosphonium salt 6 as before, the Wittig re-
action with 2,5-dichlorobenzaldehyde (25)33,34 gave a mixture of
E and Z isomers of 26 in 85% yield. Photocyclization of this stil-
bene derivative proceeded in yields comparable to the prior cases,
giving the dichloro[4]helicene 27 in 62% yield, despite the steric
problems, and bromination of the methyl group was accom-
plished with N-bromosuccinimide to give benzylic bromide 28 in
96% yield.
Cl
Cl
15
Figure 9. Mixture of products obtained from the attempted aldol trimerization of
ketone 13 using TiCl4 in o-dichlorobenzene at 115 ꢀC.
The 400 MHz 1H NMR spectrum of 28, unlike those of the pre-
vious benzylic bromides (10 and 18), shows an AB-quartet for the
two benzylic hydrogens (Jgem¼10.5 Hz). This indicates that [4]hel-
icene 28 is not interconverting between P and M atropisomers on
the NMR timescale. A barrier of 19.8 kcal/mol was calculated for
this interconversion in the parent 1-chlorobenzo[c]phenanthrene
a mixture of E and Z olefins in 87% yield. Photocyclization of the
stereoisomeric mixture was carried out as before in cyclohexane
in the presence of iodine and propylene oxide to give the
substituted [4]helicene 17 in 57% yield. Subsequent benzylic
bromination proceeded cleanly, giving 18 in 90% yield, and dis-
placement of the bromide with cyanide ion provided nitrile 19 in
92% yield. Acid catalyzed hydrolysis of the nitrile group gave an
89% yield of carboxylic acid 20, which was converted to the acid
chloride as before, followed by addition of aluminum chloride to
produce the desired monochloroketone 21 in 57% yield (Fig. 10).
It was gratifying to find that our prediction about improved
solubility was correct.
With the new, more soluble substrate, we began working on
conditions to effect the aldol trimerization. Treatment of ketone 21
with titanium tetrachloride in hot o-dichlorobenzene was carried
out as before and the product was precipitated with methanol.
Figure 11 shows the MALDI-TOF mass spectrum of the material
obtained by filtration.
at the B3LYP/6-31G level of theory (no thermodynamic correc-
*
tions).28 The carboxylic acid (30) derived from 28 likewise shows
diastereotopic hydrogens on the methylene group in its 1H NMR
spectrum.
Displacement of the bromide with cyanide ion gives the ben-
zylic nitrile 29 in 90% yield, and acid catalyzed nitrile hydrolysis
proceeds cleanly to give a 93% yield of the carboxylic acid 30. The
synthesis of ketone 31 is completed by in situ generation of the acid
chloride of 30, followed by intramolecular Friedel–Crafts acylation
in 83% yield (Fig. 12).
As predicted, dichloroketone 31 is indeed more soluble than
either of the other ketones described above (13 and 21). Much
to our disappointment, however, treatment of ketone 31 with
titanium tetrachloride in hot o-dichlorobenzene produced an
inseparable mixture of the
a,b-unsaturated dimer, the cyclic

T.J. Hill et al. / Tetrahedron 64 (2008) 11360–11369
11365
PPh3Br
hν
n-BuLi
NBS
Cl
Cl
cyclohexane
O
H
benzoyl
peroxide
CCl4
I2
O
Cl
90%
57%
87%
6
16
17
O
N
O
Br
C
HO
H2SO4
H2O
HOAc
89%
KCN,
Bu4N+HSO4
-
1. SOCl2
Cl
Cl
Cl
Cl
2. AlCl3/CH2Cl2
57%
CH2Cl2,H2O
92%
18
19
20
21
Figure 10. Synthesis of monochloroketone 21.
trimer, and the cyclic tetramer, very similar to those we had
seen before. The Lewis acid catalyzed aldol trimerizations of
these ketones (13, 21, and 31) simply were not clean enough to
be useful.
Fortunately, at the time when these results were obtained, other
work in our laboratory had just uncovered a new protocol for
effecting the aldol trimerization of cyclic ketones using a Brønsted
acid system. To our great relief, these new conditions worked
beautifully, not only for the aldol trimerization of dichloroketone 31
but also for the monochloroketone 21, as described below.
2.4. A superior aldol trimerization method
The new Brønsted acid conditions developed by Amick and
Scott25 for the aldol trimerization call for a combination of p-tol-
uenesulfonic acid and propanoic acid in hot o-dichlorobenzene. The
role of the propanoic acid remains uncertain, but its presence was
found to raise the yields in other aldol trimerizations tested by as
much as 10–15%. We began by heating a solution of ketone 31,
p-toluenesulfonic acid, and propanoic acid in o-dichlorobenzene
overnight at 110 ꢀC. MALDI-TOF analysis of the crude product
Cl
O
Cl
22
Cl
3a
Cl
Cl
Cl
Cl
O
O
Cl
Cl
Cl
Cl
24
Cl
Cl
23
Figure 11. MALDI-TOF mass spectrum of the product mixture from attempted aldol trimerization of 21 with titanium tetrachloride in hot o-dichlorobenzene.

11366
T.J. Hill et al. / Tetrahedron 64 (2008) 11360–11369
a powerful method for the rapid construction of carbon-rich
nanomolecules that have well-defined, relatively rigid shapes.
We are pursuing the proposed conversion of cyclic trimer 3b to
the C3v nanotube end cap 2 but have no results to report at this
time.
PPh3Br
n-BuLi
Cl
O
H
Cl
Cl
85%
Cl
4. Experimental section
4.1. General
6
26
25
Tetrahydrofuran, dichloromethane, and o-dichlorobenzene were
obtained anhydrous and air-free from a solvent purification system
from Glass Contour Solvent Systems, Inc. All other solvents and
commercial chemicals were of the best available grade and were
used without further purification. Proton NMR chemical shifts are
reported in parts per million downfield from tetramethylsilane with
hν
NBS
Cl
cyclohexane
benzoyl
peroxide
CCl4
I2
O
Cl
27
96%
62%
Br
deuteriochloroform (
d
¼7.26 ppm) as the reference standard, unless
otherwise specified. Carbon NMR chemical shifts are reported in
parts per million downfield from tetramethylsilane with deuterio-
N
C
chloroform (
d
¼77.0 ppm) as the reference standard. Preparative
KCN,
Bu4N+HSO4
H2SO4
-
thin layer chromatographies were performed on 20 cmꢁ20 cm
H2O
Cl
Cl
Analtech Uniplate Taper plates, Silica GF. For column chromatog-
CH2Cl2,H2O
90%
HOAc
93%
raphies, silica gel 32–63 mm was used. High-performance liquid
Cl
28
Cl
29
chromatographies (HPLC) were performed on a Waters Delta 600
with a Supelcosil LC-PAH (21.2ꢁ250 mm) reversed-phase column.
Gas Chromatograph–Mass Spectrometer (GC–MS) analysis was
performed on a Thermo Finnigan Trace DSQ with electron impact
O
ionization with a Thermo TR-5MS (15 mꢁ0.25 mmꢁ0.25
mm film)
O
HO
column. MALDI-TOF MS analyses were performed in the Boston
College Mass Spectrometry Center using a solvent-free sample
preparation with TCNQ as the matrix.35 High-resolution mass
spectrometry (HRMS) was performed in the Boston College Mass
Spectrometry Center, or by the Mass Spectrometry Laboratory at the
University of Illinois. Elemental analysis was performed by Rob-
ertson Microlit Laboratories. Melting points are uncorrected.
1.SOCl2
Cl
Cl
2.AlCl3/CH2Cl2
83%
Cl
30
Cl
31
Figure 12. Synthesis of dichloroketone 31.
4.2. 2-(1-Bromoethyl)naphthalene (5)
This compound has previously been synthesized by benzylic
bromination of 2-ethylnaphthalene.36 To a flame-dried 500 mL
three-necked round-bottom flask equipped with a reflux condenser
were added 1-(2-naphthyl)ethanol (20.90 g, 121.4 mmol) and
250 mL of benzene. The solution was brought to reflux and stirred
vigorously while phosphorus tribromide (7.00 mL, 74.5 mmol) was
added dropwise. After 20 min, the solution was allowed to cool to
room temperature and was then quenched with 1 M NaOH. The
organic layer was washed with water (3ꢁ100 mL) until the yellow
color no longer remained and then with a solution of saturated
aqueous sodium chloride (100 mL). Drying over magnesium sulfate,
filtration, and removal of the solvent under reduced pressure gave
27.16 g (95%) of the title compound as an oil, which solidified upon
standing. The spectroscopic properties matched with those of the
commercially available material.
revealed the desired trimer 3b and substantial quantities of the a,b-
unsaturated ketone dimer; however, for the first time, no peaks
were seen for the cyclic tetramer or for other compounds larger
than the trimer. By varying the temperature, we soon found that
this catalyst system gives the desired cyclic trimer 3b almost ex-
clusively at 180 ꢀC in o-dichlorobenzene (Fig. 13).
Furthermore, the isolation of cyclic trimer 3b also turned out to
be trivial. At the end of the reaction, the product mixture is cooled
back to room temperature, and methanol is added to precipitate the
clean trimer in 79% yield. Despite the nonplanarity induced by
chlorine atoms in the three cove regions, 3b proved to be too in-
soluble for 1H NMR analysis in a variety of solvents, even at elevated
temperatures.
Exposure of the monochloroketone 21 to these same reaction
conditions gave cyclic trimer 3a cleanly in 74% isolated yield. We
did not go back to try these new Brønsted acid trimerization con-
ditions with ketone 13, because access to ample supplies of the
superior nanotube end cap precursor 3b quickly curtailed our in-
terest in trimer 3c.
4.3. 1-(2-Naphthyl)ethyltriphenylphosphonium bromide (6)
2-(1-Bromoethyl)naphthalene (27.16 g, 115.5 mmol) and tri-
phenylphosphine (33.33 g, 127.1 mmol) were dissolved in 70 mL of
toluene, and the mixture was heated to reflux (110 ꢀC) with vig-
orous stirring for 12 h, during which time a white solid precipitated
from solution. The reaction mixture was placed in the freezer
at ꢂ10 ꢀC for 4 h and then filtered to collect the solid. Any large
crystals were pulverized using a mortar and pestle. The solid was
washed again, collected by filtration, and washed with cold hex-
anes to give a total of 49.37 g (93%) of the phosphonium salt as
3. Concluding remarks
Figure 14 shows a space filling model of the parent C60H30
hydrocarbon 1 to which all of the trimers reported here are struc-
turally related. By conventional criteria, this is a large molecule. The
diameter across the widest part measures somewhat more than
1.7 nm. Clearly, the aldol trimerization of aromatic ketones offers
a white powder, mp 207 ꢀC. 1H NMR (400 MHz, CDCl3):
d 7.86 (dd,

T.J. Hill et al. / Tetrahedron 64 (2008) 11360–11369
11367
O
p-TsOH·H2O
propanoicacid
THEORETICAL
Cl
o-DCB,180 °C
79%
Cl
31
Cl
Cl
Cl
Cl
Cl
Cl
EXPERIMENTAL
3b
MALDI-TOF MS (TCNQ matrix)
Figure 13. Clean cyclic trimer 3b is the only product seen by MALDI-TOF analysis of the material isolated from aldol trimerization of 31 using the new Brønsted acid conditions.
JPH¼11.0 Hz, JHH¼7.4 Hz, 6H), 7.80–7.71 (m, 4H), 7.63 (td, J¼8.0,
3.4 Hz, 7H), 7.51 (d, J¼8.1 Hz, 1H), 7.47–7.33 (m, 4H), 6.83 (qd,
J¼14.3, 7.3 Hz, 1H), 1.89 (dd, J¼19.0, 7.1 Hz, 3H). 13C NMR (100 MHz,
34.28 mmol) and 100 mL of anhydrous THF. The mixture was
cooled to ꢂ78 ꢀC and stirred while n-butyl lithium (15.1 mL of
a 2.5 M solution in hexanes, 38 mmol) was added dropwise. The
solution turned orange/red while stirring for 45 min, at which point
it was allowed to come to room temperature and stirred for an
additional 45 min. At this point, the solution was deep red. The 2,5-
dichlorobenzaldehyde (6.00 g, 34.3 mmol) was added as a solution
in THF (5 mL). The reaction mixture turned light yellow in color
after being stirred for 14 h. THF was removed and the resulting
material was deposited onto silica gel by evaporation of the solvent
under reduced pressure. The compound was then passed through
a short silica gel plug using hexanes as the eluent. Concentration of
the hexanes gave 9.09 g (85%) of the title compound as a white
solid. The E and Z isomers were obtained in a ratio of 5.3 to 1, re-
spectively, as judged by 1H NMR spectroscopic analysis, but they
were not separated. The major isomer had its methyl hydrogens
chemical shift at 2.30 ppm (d, J¼1.3 Hz), and those of the minor
isomer appeared at 2.36 ppm (d, J¼1.5 Hz). The corresponding 13C
chemical shifts were 17.69 ppm and 26.53 ppm, respectively. HRMS
DART (m/z): [MþH]þcalcd for C19H15Cl2 313.0551; found 313.0563.
CDCl3):
d
134.6, 134.4 (d, J¼9.2 Hz), 132.7, 132.5, 130.6 (d, J¼5.6 Hz),
130.0, 129.9, 129.8, 128.2, 127.7, 127.3, 126.5, 126.3, 117.5 (d,
J¼82.5 Hz), 34.7 (d, J¼42.3 Hz), 17.1. Elemental analysis calcd for
C30H26BrP: C, 72.44; H, 5.27. Found: C, 72.42; H, 5.21.
4.4. (E) and (Z)-1-(2,5-Dichlorophenyl)-2-(2-naphthyl)-
propene (26)
Into a flame-dried round-bottom flask under nitrogen were
added 1-(2-naphthyl)ethyltriphenylphosphonium bromide (17.05 g,
4.5. 1,4-Dichloro-6-methylbenzo[c]phenanthrene (27)
A solution of (E)- and (Z)-1-(2,5-dichlorophenyl)-2-(2-naph-
thyl)propene (3.980 g, 12.02 mmol), iodine (3.200 g, 12.62 mmol),
and propylene oxide (80.0 mL, 1.14 mol) in 4 L of cyclohexane was
photolyzed in a quartz vessel for 35 h in a Rayonet photochemical
apparatus equipped with 15 mercury lamps of 254 nm and 35 W.
The solvent was removed by rotary evaporation to give an orange
solid. Recrystallization of the crude mixture from ethanol yielded
Figure 14. Space filling model of the parent C60H30 hydrocarbon 1.

11368
T.J. Hill et al. / Tetrahedron 64 (2008) 11360–11369
2.15 g of 1,4-dichloro-6-methylbenzo[c]phenanthrene as a yellow
crystalline solid, mp 131 ꢀC. A second crop of crystals was obtained
from the mother liquor to give an additional 0.30 g of product
80 mL each of glacial acetic acid, concd sulfuric acid, and water. The
mixture was heated to reflux (approximately 110 ꢀC) with stirring
for 16 h. The reaction mixture was then cooled to room tempera-
ture and diluted with water. The precipitate was collected by vac-
uum filtration and washed with water. Any large solids were
broken up with a mortar and pestle, washed again with water, and
filtered. The gray powder was dried thoroughly under vacuum to
yield 4.05 g (93%) of the title compound, mp 209–211 ꢀC. 1H NMR
(combined yield 62%). 1H NMR (500 MHz, CDCl3):
d 8.13 (d,
J¼1.1 Hz, 1H), 8.11 (dd, J¼8.2, 1.2 Hz, 1H), 8.05 (d, J¼8.8 Hz, 1H), 8.02
(d, J¼8.8 Hz, 1H), 7.98 (dd, J¼7.9, 1.4 Hz, 1H), 7.62 (d, J¼8.1 Hz, 1H),
7.59 (ddd, J¼8.4, 7.0, 1.5 Hz, 1H), 7.56 (ddd, J¼8.2, 6.9, 1.6 Hz, 1H),
7.53 (d, J¼8.1 Hz, 1H), 2.82 (d, J¼1.1 Hz, 3H). 13C NMR (125 MHz,
CDCl3):
d
134.7, 132.2, 132.0, 131.8, 131.4 130.9, 130.0, 129.0, 128.9,
(400 MHz, CDCl3):
d
8.13 (s, 1H), 8.01 (d, J¼8.0 Hz, 1H), 7.95
127.8, 127.7, 127.1, 126.6, 126.3, 126.1, 124.8, 122.8, 121.3 20.1. HRMS
(d, J¼8.8 Hz, 1H), 7.89 (d, J¼7.8 Hz, 1H), 7.86 (d, J¼8.8 Hz, 1H), 7.56
DART (m/z): [MþH]þcalcd for C19H13Cl2 311.0394; found 311.0397.
(d, J¼8.1 Hz, 1H), 7.54–7.46 (m, 3H), 4.19 (d, J¼16.2 Hz, 1H), 4.09
(d, J¼16.2 Hz, 1H). 13C NMR (100 MHz, CDCl3):
d 176.3, 131.7, 131.5,
4.6. 6-Bromomethyl-1,4-dichlorobenzo[c]phenanthrene (28)
131.3, 131.1, 130.9, 130.5, 130.2, 129.4 128.9, 128.6, 128.4, 127.1, 126.8,
126.7, 126.4, 124.93, 124.91, 120.7, 39.1. IR (KBr) n(O–H): 3046 cmꢂ1
,
Into a 250 mL round-bottom flask equipped with a reflux con-
denser were added 1,4-dichloro-6-methylbenzo[c]phenanthrene
(1.53 g, 4.92 mmol), N-bromosuccinimide (0.962 g, 5.41 mmol), and
a pinch (approximately 10 mg) of dibenzoylperoxide. Carbon tetra-
chloride (75 mL) was added, and the mixture was heated to reflux
(approximately 80 ꢀC) and monitored by NMR spectroscopy; all
starting material was consumed typically within 24 h, but some
larger scale reactions required 3 days.37 After being allowed to cool
to room temperature, the mixture was filtered through filter paper
to remove solid succinimide, which was then rinsed with hexane.
The filtrate was deposited onto silica gel by evaporation of the sol-
vent under reduced pressure and passed through a short plug of
silica using dichloromethane as the eluent. The solvent was con-
centrated to give 1.84 g (96%) of the title compound as a tan solid, mp
n(C]O): 1708 cmꢂ1. HRMS DART (m/z): [MþH]þcalcd for C20H13Cl2O2
355.0293; found 355.0299.
4.9. 1,4-Dichloro-6H-benzo[a]acephenanthrylen-7-one (31)
(1,4-Dichlorobenzo[c]phenanthren-6-yl)acetic acid (1.01 g,
2.84 mmol) was added to a flame-dried 1 L three-necked round-
bottom flask equipped with a reflux condenser under nitrogen. To
this was added thionyl chloride (40 mL, 548 mmol) and the
resulting solution was heated to reflux (approximately 80 ꢀC).
After 1 h, the excess thionyl chloride was removed by vacuum
distillation to leave a yellow-brown oil in the flask. The system
was refilled with nitrogen, and the oil was dissolved in 650 mL of
anhydrous dichloromethane, which was then cooled to 0 ꢀC with
an ice bath. The yellow solution turned dark red upon addition of
aluminum chloride (1.14 g, 8.53 mmol). The solution was stirred at
0 ꢀC for 1 h, at which point it was heated to reflux (approximately
40 ꢀC) for 15 min. After being cooled to room temperature, the
reaction was quenched by addition of 1 N aqueous hydrochloric
acid (50 mL). The yellow solution was washed first with water
(2ꢁ250 mL), then with saturated sodium chloride solution
(250 mL), and finally dried over magnesium sulfate and filtered.
The solvent was removed by rotary evaporation, and the crude
material was passed through a short plug of silica with dichloro-
methane as the eluent. Removal of the solvent gave 0.96 g (83%) of
1,4-dichlorobenzo[a]acephenanthrylen-7-one as a yellow solid,
150 ꢀC. 1H NMR (400 MHz, CDCl3):
d
8.37 (s, 1H), 8.19 (d, J¼8.9 Hz,
1H), 8.13 (d, J¼8.9 Hz,1H), 8.10 (dd, J¼8.1, 0.8 Hz,1H), 8.03 (dd, J¼7.7,
1.6 Hz, 1H), 7.68 (d, J¼8.2 Hz, 1H), 7.64 (td, J¼7.3, 1.5 Hz, 1H), 7.62 (d,
J¼8.1 Hz, 1H), 7.59 (td, J¼6.9, 1.5 Hz, 1H), 5.10 (d, J¼10.5 Hz, 1H) 5.03
(d, J¼10.5 Hz, 1H). 13C NMR (100 MHz, CDCl3):
d 133.5, 131.9, 131.4,
131.2, 131.1, 130.8, 130.0, 129.4, 129.2, 129.0, 128.9, 127.21, 127.19,
126.9, 126.6, 124.9, 124.2, 120.7, 31.4. HRMS DART (m/z):
[MþH]þcalcd for C19H12BrCl2 388.9499; found 388.9480.
4.7. (1,4-Dichlorobenzo[c]phenanthren-6-yl)acetonitrile (29)
6-Bromomethyl-1,4-dichlorobenzo[c]phenanthrene
(5.280 g,
13.52 mmol), potassium cyanide (3.520 g, 54.09 mmol), and tetra-
butylammonium hydrogensulfate (4.590 g, 13.52 mmol) were
placed in a round-bottom flask. To this were added 250 mL of
dichloromethane and 125 mL of water. The reaction mixture was
stirred vigorously for 10 h at room temperature to ensure that the
two layers were mixing. The layers were then allowed to separate,
and the organic layer was washed with water (2ꢁ150 mL) and finally
with saturated aqueous sodium chloride (150 mL). After the organic
layer was dried over magnesium sulfate, it was filtered, deposited
onto silica gel by evaporation of the solvent under reduced pressure
and passed through a short plug of silica using dichloromethane as
the eluent. The product was concentrated to dryness and yielded
4.12 g (90%) of the title compound as a pale yellow solid, mp 158 ꢀC.
mp 190 ꢀC. 1H NMR (400 MHz, CDCl3):
8.35 (d, J¼8.4 Hz, 1H), 8.24 (d, J¼8.4 Hz, 1H), 7.81–7.68 (m, 4H),
3.99 (s, 2H). 13C NMR (100 MHz, CDCl3):
201.9, 140.2, 134.6, 133.6,
133.5, 131.5, 131.31, 131.28, 131.2, 130.9, 130.4, 128.6, 127.7, 127.2,
d 8.60 (s, 1H), 8.39 (s, 1H),
d
127.1, 126.8, 126.4, 123.3, 118.5, 41.8. IR (KBr) n(C]O): 1719 cmꢂ1
.
HRMS DART (m/z): [MþH]þcalcd for C20H11Cl2O 337.0187; found
337.0188.
4.10. Trimerization of 1,4-dichloro-6H-benzo[a]-
acephenanthrylen-7-one (3b)
Into a flame-dried 50 mL sealable tube were added 1,4-dichloro-
6H-benzo[a]acephenanthrylen-7-one (0.500 g, 1.48 mmol) and
p-toluenesulfonic acid monohydrate (0.987 g, 5.19 mmol). The tube
was flushed with nitrogen, and then 10 mL of o-dichlorobenzene
and 0.04 mL of propanoic acid were added via syringe. The tube
was sealed and immersed in an oil bath at 180 ꢀC for 16 h. After the
reaction mixture had cooled to room temperature, it was diluted
with sufficient methanol to precipitate the crude product (200 mL).
From vacuum filtration, 0.374 g (79%) of a brown solid was
obtained. The product was too insoluble for NMR characterization.
MALDI-TOF MS analysis matched with the calculated isotope pat-
tern and showed no other major peaks: (m/z) [M]þ 954/956/958/
960/962. HRMS DART negative ion mode (m/z): [Mþ] calcd for
C60H24Cl6 954.0009; found 953.9963. HRMS DART positive ion
mode (m/z): [Mþ1]þ calcd for C60H25Cl6 955.0082; found 955.0069.
1H NMR (400 MHz, CDCl3):
d
8.42 (s, 1H), 8.14 (d, J¼8.8 Hz, 1H), 8.12
(d, J¼8.1 Hz, 1H), 8.03 (dd, J¼7.8, 1.4 Hz, 1H), 7.91 (d, J¼8.8 Hz, 1H),
7.71 (d, J¼8.2 Hz, 1H), 7.66 (td, J¼7.2, 1.3 Hz, 1H), 7.65 (d, J¼8.3 Hz,
1H), 7.61 (ddd, J¼8.4, 7.0,1.6 Hz,1H), 4.27 (s, 2H).13C NMR (100 MHz,
CDCl3):
d 131.8, 131.2 (2C?), 130.9, 130.7, 129.9, 129.6 129.1, 128.6,
128.5, 127.2, 127.1, 126.9, 126.7, 126.3, 125.3, 123.3, 119.2, 117.0, 22.3
(one overlapping peak). IR (KBr) n(C]N): 2248 cmꢂ1. HRMS APPI
(m/z): [M]þcalcd for C20H10Cl2N 334.0190; found 334.0195.
4.8. (1,4-Dichlorobenzo[c]phenanthren-6-yl)acetic acid (30)
A 500 mL round-bottom flask was charged with (1,4-dichloro-
benzo[c]phenanthren-6-yl)acetonitrile (4.120 g, 12.24 mmol) and

T.J. Hill et al. / Tetrahedron 64 (2008) 11360–11369
11369
12. Carbon Nanotubes; Endo, M., Iijima, S., Dresselhaus, M. S., Eds.; Elsevier: New
York, NY, 1997.
13. Carbon Nanotubes Synthesis, Structure, Properties, and Applications; Dresselhaus,
M. S., Dresselhaus, G., Avouris, P., Eds.; Topics in Current Physics; Springer:
Berlin/Heidelberg, 2001; Vol. 80.
14. Park, T.-J.; Banerjee, S.; Hemraj-Benny, T.; Wong, S. S. J. Mater. Chem. 2006, 16,
141–154.
15. Zheng, M.; Jagota, A.; Strano, M. S.; Santos, A. P.; Barone, P.; Chou, S. G.; Diner, B. A.;
Dresselhaus, M. S.; McLean, R. S.; Onoa, G. B.; Samsonidze, G. G.; Semke, E. D.;
Usrey, M.; Walls, D. J. Science 2003, 302, 1545–1548 and references cited therein.
16. Scott, L. T. Pure Appl. Chem. 1996, 68, 291–300.
17. Scott, L. T.; Bronstein, H. E.; Preda, D. V.; Ansems, R. B. M.; Bratcher, M. S.;
Hagen, S. Pure Appl. Chem. 1999, 71, 209–219.
18. Tsefrikas, V. M.; Scott, L. T. Chem. Rev. 2006, 106, 4868–4884.
19. Fowler, P. W.; Manolopoulos, D. E. Atlas of Fullerenes; Oxford University
Press: Oxford, UK, 1995; A discussion of Euler’s theorem can be found in
chapter 2.
20. Pan, Z. W.; Xie, S. S.; Chang, B. H.; Wang, C. Y.; Lu, L.; Liu, W.; Zhou, W. Y.; Li,
W. Z.; Qian, L. X. Nature 1998, 394, 631–632.
21. Lee, Y. T.; Kim, N. S.; Park, J.; Han, J. B.; Choi, Y. S.; Ryu, H.; Lee, H. J. Chem. Phys.
Lett. 2003, 372, 853–859.
4.11. Trimerization of 4-chloro-6H-benzo[a]-
acephenanthrylen-7-one using p-TsOH (3a)
Into a flame-dried 10 mL sealable tube were added 4-chloro-
6H-benzo[a]acephenanthrylen-7-one (0.050 g, 0.17 mmol) and
p-toluenesulfonic acid monohydrate (0.110 g, 0.578 mmol). The
tube was flushed with nitrogen, and then 1 mL of o-di-
chlorobenzene and 0.04 mL of propanoic acid were added via sy-
ringe. The tube was sealed and immersed in an oil bath at 180 ꢀC for
16 h. After the reaction mixture had cooled to room temperature, it
was diluted with sufficient methanol to precipitate the crude
product (approximately 50 mL). From vacuum filtration, 0.035 mg
(74%) of a dark brown solid was obtained. The product was too
insoluble for NMR characterization. MALDI-TOF MS showed no
other major peaks: (m/z) [M]þ 852/854.
22. Ansems, R. B. M.; Scott, L. T. J. Am. Chem. Soc. 2000, 122, 2719–2724.
23. Scott, L. T.; Bratcher, M. S.; Hagen, S. J. Am. Chem. Soc. 1996, 118, 8743–8744.
24. Boorum, M. M.; Scott, L. T. In Modern Arene Chemistry; Astruc, D., Ed.; Wiley-
VCH: Weinheim, Germany, 2002; pp 20–31.
25. Amick, A. W.; Scott, L. T. J. Org. Chem. 2007, 72, 3412–3418.
26. Brooks, M. A.; Scott, L. T. J. Am. Chem. Soc. 1999, 121, 5444–5449.
27. Hagen, S.; Bratcher, M. S.; Erickson, M. S.; Zimmermann, G.; Scott, L. T. Angew.
Chem., Int. Ed. 1997, 36, 406–408.
Acknowledgements
We thank the National Science Foundation for financial support
of this work through Grant No. 0414066 and for funds to purchase
mass spectrometers through Grant No. 0619576.
28. Geometry optimizations of ground state molecules and transition states were
carried out using the Spartan computational package (Wavefunction, Inc., Ir-
vine, CA 92612, USA), with complete vibrational analysis; thermodynamic
corrections to the enthalpy and entropy were adjusted to typical FVP condi-
tions (1100 ꢀC/0.1 mmHg).
29. Hill, T. J. Ph.D. dissertation; Boston College: Chestnut Hill, MA, 2007.
30. Hydrolysis conditions adapted from those reported by Fernandez, F.; Gomez,
G.; Lopez, C.; Santos, A. Synthesis 1988, 802–803.
Supplementary data
The supplementary data contains experimental procedures for
all the reactions depicted in Figures 7–11, as well as 1H NMR and 13
C
NMR spectra for compounds 6, 8–13, 16–21, and 26–31 and mass
spectra for compounds 3a, 3c, 14, and 15. Supplementary data as-
sociated with this article can be found in the online version, at
doi:10.1016/j.tet.2008.09.087.
31. Hydrolysis of nitrile 11 was first attempted under basic conditions using KOH in
ethylene glycol; however, the reaction did not proceed cleanly, giving de-
composition products that were difficult to separate.
32. Chromatography of the mixture gave a slow moving red band that was ten-
tatively identified as the tetramerdione 24 on the basis of its mobility and its 1H
References and notes
NMR spectrum; the most characteristic signals appeared far downfield at
(singlet), 9.21 (doublet), and 9.09 (triplet).
d 9.80
33. We have found that 2,5-dichlorobenzaldehyde (25) can be easily prepared in
67% yield by ortho-lithiation of p-dichlorobenzene with 1 equiv of n-butyl
lithium,34 followed by 1 equiv of N,N-dimethyl formamide and aqueous
hydrolysis.
34. Anctil, E. J. G.; Snieckus, V. In Metal-Catalyzed Cross-Coupling Reactions, 2nd ed.;
De Meijere, A., Diederich, F., Eds.; Wiley-VCH: Weinheim, 2004; Vol. 2,
pp 761–813.
35. Pryzbilla, L.; Brand, J.-D.; Yoshimura, K.; Reader, H. J.; Mu¨llen, K. Anal. Chem.
2000, 72, 4591–4597.
36. Bull, S. D.; Davies, S. G.; Epstein, S. W.; Garner, A. C.; Mujtaba, N.; Roberts, P. M.;
Savory, E. D.; Smith, A. D.; Tamayo, J. A.; Watkin, D. J. Tetrahedron 2006, 62,
7911–7925.
1. Collins, P. G.; Avouris, P. Sci. Am. 2000, 283, 62–69.
2. Carbon Nanotubes: Science and Applications; Meyyappan, M., Ed.; CRC LLC: Boca
Raton, FL, 2005.
3. Carbon Nanotubes: Properties and Applications; O’Connell, M. J., Ed.; CRC LLC:
Boca Raton, FL, 2006.
4. Endo, M.; Hayashi, T.; Kim, Y.-A. Pure Appl. Chem. 2006, 78, 1703–1713.
5. Arepalli, S. J. Nanosci. Nanotechnol. 2004, 4, 317–325.
6. Resasco, D. E.; Herrera, J. E.; Balzano, L. J. Nanosci. Nanotechnol. 2004, 4, 398–407.
7. Nikolaev, P. Encyclopedia Nanosci. Nanotechnol. 2004, 3, 917–927.
8. Mishra, L. N.; Shibata, K.; Ito, H.; Yugami, N.; Nishida, Y. IEEE Trans. Plasma Sci.
2004, 32, 1727–1733.
9. Boorum, M. M.; Vasil’ev, Y. V.; Drewello, T.; Scott, L. T. Science 2001, 294, 828–831.
10. Scott, L. T.; Boorum, M. M.; McMahon, B. J.; Hagen, S.; Mack, J.; Blank, J.;
Wegner, H.; de Meijere, A. Science 2002, 295, 1500–1503.
11. Scott, L. T. Angew. Chem., Int. Ed. 2004, 43, 4994–5007.
37. Irradiation of the reaction mixture to promote the radical bromination process
gave mixtures of doubly brominated material and unreacted starting material
along with the desired product.