Synthesis of polycyclic aromatic hydrocarbon-based nanovehicles equipped with triptycene wheels

Article abstract of DOI:10.1002/chem.201102893

Two new nanovehicles that have extended aromatic platforms as the cargo zones have been obtained. Two strategies were considered for the formation of the perylene core from two naphthalene precursors. The first was based on a Scholl-type reaction involving an oxidant, and the second used a brominated derivative to perform a homocoupling reaction. The first strategy failed under diverse coupling conditions in the presence of several strong oxidants. Nevertheless, the use of CoF₃ in trifluoroacetic acid triggered a dimerization reaction between two ester groups of one molecule and the naphthalene unit of another, thereby surprisingly yielding a ten-membered carbon macrocycle. The second strategy encountered a lack of reactivity of the substrate under several homocoupling conditions. The dimerization was not easily performed but Ullmann-type conditions ultimately gave the expected product. The low yield and low solubility of the product encouraged us to modify our initial design. The synthesis of a new chassis that incorporated additional tert-butyl groups improved the solubility of the molecules and also prevented overcyclization of the aromatic platform by blocking these positions. Some p-phenylene spacers were also intercalated between the iodine and perylene centers to increase the reactivity of the halide towards coupling reactions. Two new chassis were obtained by Scholl-type oxidative coupling using FeCl ₃ as the oxidant. The introduction of four triptycene wheels allowed the formation of the two corresponding nanovehicles. Deux nouveaux nanovehicules ont ete obtenus avec une plateforme polyaromatique pouvant servir de zone de chargement. Deux strategies de synthese ont ete considerees pour la formation du noyau perylene a partir de deux precurseurs naphtaleniques. La premiere a fait intervenir une reaction de type Scholl mettant en jeu un oxydant, tandis que la seconde a mis en uvre un derive brome afin de realiser un homocouplage. La premiere strategie a echoue malgre de nombreuses conditions de couplage testees. Neanmoins, les conditions utilisant CoF₃ dans l'acide trifluoroacetique ont abouti a une dimerisation entre deux fonctions esters d'une molecule et le noyau naphtalene d'une seconde, conduisant a la formation d'un macrocycle a 10 chaenons. La seconde strategie a egalement souffert d'un manque de reactivite du substrat lors de l'homocouplage. La dimerisation n'a pu s'effectuer facilement, mais les conditions de type Ullmann ont finalement permis d'aboutir au compose souhaite. Le faible rendement combine a la faible solubilite du produit de cette reaction nous ont conduits a corriger le design initial. La synthese d'un nouveau chassis a ete entreprise en incorporant des groupements tert-butyles supplementaires, afin d'ameliorer la solubilite des molecules mais aussi pour eviter de surcycliser le chassis polyaromatique en bloquant ces positions. Des espaceurs p-phenylene ont egalement ete intercales entre l'atome d'iode et le noyau perylene pour ameliorer la reactivite de l'halogenure vis-a-vis des reactions de couplage. Deux nouveaux chassis ont ainsi ete obtenus par couplage oxydant de type Scholl en utilisant FeCl₃ comme oxydant. L'introduction de quatre roues triptycenes a permis d'obtenir les deux nanovehicules correspondants. En route! Two new molecular nanovehicles with a large dedicated cargo zone have been synthesized according to a modular strategy based on sequential double Knoevenagel and Diels-Alder reactions. Copyright

Full text of DOI:10.1002/chem.201102893

FULL PAPER
DOI: 10.1002/chem.201102893
Synthesis of Polycyclic Aromatic Hydrocarbon-Based Nanovehicles
Equipped with Triptycene Wheels
Henri-Pierre Jacquot de Rouville,[a] Romain Garbage,[a] Rita E. Cook,[a]
Adeline R. Pujol,[a] Agnꢀs M. Sirven,[a] and Gwꢁnaꢂl Rapenne*[a, b]
Abstract: Two new nanovehicles that
have extended aromatic platforms as
the cargo zones have been obtained.
Two strategies were considered for the
formation of the perylene core from
two naphthalene precursors. The first
was based on a Scholl-type reaction in-
volving an oxidant, and the second
used a brominated derivative to per-
form a homocoupling reaction. The
first strategy failed under diverse cou-
pling conditions in the presence of sev-
eral strong oxidants. Nevertheless, the
use of CoF3 in trifluoroacetic acid trig-
gered a dimerization reaction between
two ester groups of one molecule and
the naphthalene unit of another, there-
by surprisingly yielding a ten-mem-
bered carbon macrocycle. The second
strategy encountered a lack of reactivi-
ty of the substrate under several homo-
coupling conditions. The dimerization
was not easily performed but Ullmann-
type conditions ultimately gave the ex-
pected product. The low yield and low
solubility of the product encouraged us
to modify our initial design. The syn-
thesis of a new chassis that incorporat-
ed additional tert-butyl groups im-
proved the solubility of the molecules
and also prevented overcyclization of
the aromatic platform by blocking
these positions. Some p-phenylene
spacers were also intercalated between
the iodine and perylene centers to in-
crease the reactivity of the halide to-
wards coupling reactions. Two new
chassis were obtained by Scholl-type
oxidative coupling using FeCl3 as the
oxidant. The introduction of four trip-
tycene wheels allowed the formation of
the two corresponding nanovehicles.
Keywords:
nanotechnology
hydrocarbons
perylene
·
·
·
polycycles · triptycenes
Introduction
are able to perform tasks as varied as nanoscale manipula-
tion, information storage, molecular electronics, and me-
chanics.
In the bottom-up approach of nanoscience, functional devi-
ces are built from atoms and small molecules to meet the
challenge of combining the smallest size with the highest
functionality. Multistep chemical synthesis plays a major
role in the constant quest for molecular mechanical compo-
nents because it allows chemists to prepare tailor-made
compounds of predetermined shape and programmed mo-
tions or functions.[1] In recent years, a variety of compounds
that resemble macroscopic machines have been designed
and synthesized. Such molecules are called “technomimet-
ic”[2] because they are designed to transpose macroscopic
objects on the molecular level. Among such devices, one
may cite molecular gears,[3] wheels,[4] wheelbarrows,[5] and
nanovehicles[6] as systems where rotational movements
occur. These nano-objects provide the basis for the future
design of bottom-up nanoscale systems and materials that
Triptycenyl fragments that are linked to axles have shown
their aptitude to act as wheels;[4a] this result has opened the
way to more-complex molecules that are designed to under-
go simultaneous translational and rotational motions, which
are capable of transporting ad-atoms or other molecules.
After the seminal description and synthesis of a molecular
wheelbarrow,[5] one of the earliest examples of molecular
nanovehicles, was described by Tour and co-workers[6a–e]
a
whole family of technomimetic objects have been synthe-
sized, such as nanotrucks,[7] nanodragsters,[8] and nano-
trains.[9] However, it is difficult to imagine such molecules
transporting other molecules because they contain no identi-
fiable “cargo zone”. With this aim in view, we designed two
polycyclic aromatic hydrocarbon (PAH)[10] chassis, which
were designed to contain a cargo zone for the transport of
molecules and to be rigid and resistant to deposition tech-
niques. Herein, we report the synthesis of these two new
chassis, which were built around perylene fragments. In a
last step, the chassis were equipped with four ethynyltripty-
cenyl wheels to yield two prototype nanovehicles.
Our target molecules consisted of a polyaromatic central
platform, based on a PAH skeleton from the “lander”
family,[11] and four wheels. The PAH skeleton was built
around a perylene center, which is known to have very low
solubility because it can easily aggregate through p-stacking
[a] H.-P. Jacquot de Rouville, R. Garbage, R. E. Cook, A. R. Pujol,
A. M. Sirven, Prof. Dr. G. Rapenne
NanoSciences Group, CEMES-CNRS, 29 rue Jeanne Marvig
BP 94347, 31055 Toulouse Cedex 4 (France)
E-mail: rapenne@cemes.fr
[b] Prof. Dr. G. Rapenne
Universitꢀ de Toulouse, UPS, CNRS
CEMES, 118 route de Narbonne, 31062 Toulouse (France)
Chem. Eur. J. 2012, 18, 3023 – 3031
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3023

interactions.[12] Therefore, tert-butyl groups connected to
PAHs are often used to increase organic solubility and are
also used as “fingerprints” because they are easily observed
by STM techniques, as they induce a good contrast in the
image (Scheme 1). The most-important features of these
molecules are the wheels, which enable the lateral motion of
the entire molecule on the surface upon rotation, by analogy
with a macroscopic vehicle. The four 9-triptycenyl wheels
were covalently attached onto this platform through ethynyl
spacer groups, which gave a very low barrier to rotation.[13]
Scheme 1 shows the two three-cogged wheels, which can
freely rotate around their axle owing to the acetylenic
spacer groups.
Results and Discussion
First strategy: Our first intermediate target is shown in
Scheme 2. This target would be able to form alkyne bonds
in three steps (see below) from esters groups that were di-
rectly attached onto the PAH platform. The high symmetry
of our target molecule allowed us to follow a convergent
retrosynthetic analysis (Scheme 2).
Two synthetic routes were considered for the formation of
the chassis: The first strategy involved a Scholl oxidative
coupling reaction,[14] which would allow the formation of the
perylene core in one step from compound 1. On the contra-
ry, the second route would form the perylene core in two
steps from compound 2. The first step of this latter route
Scheme 1. Target nanovehicles. Linker (L)=phenyl or no linker, R1 =
tBu, R2 =H or R1 =H, R2 =tBu.
ꢀ
was a Csp2 Csp2 transition-metal-catalyzed coupling reaction,
followed by a cyclodehydrogenation. In both strategies, the
synthetic intermediates were polyaromatic, and therefore re-
ꢀ
quired solubilizing functional groups to prevent p p stack-
Abstract in French: Deux nouveaux nanovꢀhicules ont ꢀtꢀ
obtenus avec une plateforme polyaromatique pouvant servir
de zone de chargement. Deux stratꢀgies de synthꢁse ont ꢀtꢀ
considꢀrꢀes pour la formation du noyau pꢀrylꢁne ꢂ partir de
deux prꢀcurseurs naphtalꢀniques. La premiꢁre a fait interve-
nir une rꢀaction de type Scholl mettant en jeu un oxydant,
tandis que la seconde a mis en œuvre un dꢀrivꢀ bromꢀ afin
de rꢀaliser un homocouplage. La premiꢁre stratꢀgie a ꢀchouꢀ
malgrꢀ de nombreuses conditions de couplage testꢀes. Nꢀan-
moins, les conditions utilisant CoF3 dans l’acide trifluoroacꢀ-
tique ont abouti ꢂ une dimꢀrisation entre deux fonctions
esters d’une molꢀcule et le noyau naphtalꢁne d’une seconde,
conduisant ꢂ la formation d’un macrocycle ꢂ 10 chaꢃnons. La
seconde stratꢀgie a ꢀgalement souffert d’un manque de rꢀacti-
vitꢀ du substrat lors de l’homocouplage. La dimꢀrisation n’a
pu sꢄeffectuer facilement, mais les conditions de type Ullmann
ont finalement permis d’aboutir au composꢀ souhaitꢀ. Le
faible rendement combinꢀ ꢂ la faible solubilitꢀ du produit de
cette rꢀaction nous ont conduits ꢂ corriger le design initial.
La synthꢁse d’un nouveau chꢅssis a ꢀtꢀ entreprise en incorpo-
rant des groupements tert-butyles supplꢀmentaires, afin
d’amꢀliorer la solubilitꢀ des molꢀcules mais aussi pour ꢀviter
de surcycliser le chꢅssis polyaromatique en bloquant ces posi-
tions. Des espaceurs p-phꢀnylꢁne ont ꢀgalement ꢀtꢀ intercalꢀs
entre l’atome d’iode et le noyau pꢀrylꢁne pour amꢀliorer la
rꢀactivitꢀ de l’halogꢀnure vis-ꢂ-vis des rꢀactions de couplage.
Deux nouveaux chꢅssis ont ainsi ꢀtꢀ obtenus par couplage
oxydant de type Scholl en utilisant FeCl3 comme oxydant.
L’introduction de quatre roues triptycꢁnes a permis d’obtenir
les deux nanovꢀhicules correspondants.
ing and aggregation. Ester functional groups are well-suited
for this purpose because they not only drastically increase
the solubility of the compounds, but can also be efficiently
converted into alkyne groups in three steps, that is: conver-
sion of the ester into a carboxylic acid, decarboxylative bro-
mination,[15] and finally a Sonogashira coupling. The key
step in the first route was the formation of the perylene
core, which was performed by an oxidative coupling reac-
tion. Known for a century, Scholl-type reactions allow the
synthesis of the perylene core by a coupling reaction be-
tween two naphthyl fragments and an oxidant, which also
acts as a Lewis acid.[14]
The first route (Scheme 3) began with a double Knoeve-
nagel reaction between acenaphthenequinone and diethyl
acetonedicarboxylate under basic conditions (Scheme 3).[16]
Triethylamine was used as a base to avoid any hydrolysis of
the ester groups. Subsequently, diethylestercyclopentadie-
none (3) underwent aromatization in the presence of 1,2-
di(4-tert-butylphenyl)ethyne by
a Diels–Alder reaction
under high temperatures.[5b] The formation of carbon mon-
oxide shifted the equilibrium of the reaction in favor of the
aromatization and compound 1 was obtained in two steps.
After obtaining the half-chassis (1), the formation of the
perylene core by an oxidative coupling was performed by an
oxidative coupling reaction. The mechanism of Scholl-type
reactions is still not well-understood, but it is generally ac-
cepted that the formation of a radical cation on the aromatic
substrate, which is triggered by the oxidant, is involved. This
cation can react as an electrophile in a Friedel–Crafts reac-
tion on a second aromatic core. After the formation of the
3024
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 3023 – 3031

Synthesis of Polycyclic Aromatic Nanovehicles
FULL PAPER
system would be drastically ex-
tended. The color of this type
of aromatic cores usually varies
from red to blue.
To form the dimer by oxida-
tive coupling, a series of strong
oxidants were tested, among
which were CoF3,[18] Pb-
ACHUTNGRENNUG CAHTUNGTRENNUGN
(OAc)4,[18] Sc(OTf)3,[19] DDQ
(2,3-dichloro-5,6-dicyano-1,4-
benzoquinone),[20] AlCl3,[21] and
MoCl5.[21] However, the desired
product was not obtained in
Scheme 2. Retrosynthetic analysis of the perylene-based polycyclic aromatic hydrocarbon.
ꢀ
C C bond, a second equivalent of oxidant promotes rearo-
any of these cases, despite full consumption of the starting
material and formation of several inseparable products. This
lack of reactivity towards oxidative coupling could be attrib-
uted to the electron-deficient nature of the molecule.
Indeed, the involved radical cation would either be not sta-
bilized enough or even not formed.
Interestingly, using CoF3 in TFA allowed the formation of
an orange compound in quantitative yield. After purification
and characterization, coupled MS-NMR analysis indicated
that dimerization of compound 1 had occurred by the reac-
tion of an ester group of one molecule with the naphthalene
unit of another, thereby yielding the undesired ten-mem-
bered macrocycle 4. This reaction may have occurred
through the hydrolysis of an ester group, thus forming the
corresponding acylium ion, which could then react with a
naphthalene unit in a Friedel–Crafts reaction.
matization, thereby providing a biaryl derivative. Iron
N
chloride in dichloromethane is commonly used to perform
the synthesis of the perylene core.[17] In addition, the pres-
ence of nitromethane can speed up the reaction in some
cases.[17] This effect is presumably because of the stabiliza-
tion of the intermediate radical cation, thus enhancing the
rate of the reaction. The oxidative coupling between two
molecules of compound 1 was then attempted, both with
and without nitromethane. When the reaction was per-
formed without nitromethane, MS (methane chemical ioni-
zation) showed a highest mass at m/z=1881.9, which could
correspond to more than three fragments of compound 1
(which has a mass of 610.4). In the presence of nitrome-
thane, the ion with the highest mass (m/z=1189.5) could be
assigned as two coupled subunits that have lost a fragment,
because the mass of the desired molecule is 1216.6. Howev-
er, in both cases, the obtained products were yellow, which
suggested that only one bond was formed between the two
biaryl groups. Indeed, a significant red shift would be ex-
pected if both bonds were formed, because the aromatic
After this unexpected result, we explored new coupling
conditions. By analogy with the synthesis of perylene dii-
mides, a basic coupling reaction was performed on com-
pound 3 in the presence of 1,5-diazabicyclo
ACHTUNGTREN[NUNG 4.3.0]nonene
and potassium tert-butoxide.[11] The mixture of these two
Scheme 3. a) Diethyl acetonedicarboxylate (1 equiv), NEt3 (1.5 equiv), MeOH, RT, 30 min, Ar, 47–79%; b) 1,2-di(4-tert-butylphenyl)ethyne (1 equiv),
Ph2O, 2508C, 15 h, 37%; c) FeCl3 (25 equiv), CH2Cl2/CH3NO2 (9:1), RT, 4 h, Ar, 0%; d) CoF3 (1 equiv), TFA, Ar, 708C, 15 h, 99%; e) Cu powder, pyri-
dine, reflux, 3 h, 11%.
Chem. Eur. J. 2012, 18, 3023 – 3031
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3025

G. Rapenne et al.
bases enhanced the basicity of the mixture and utilized the
electron-deficient nature of the molecule to trigger the cou-
pling reaction. Unfortunately, only the tetracarboxylic acid
derivative of the desired product was obtained in 5% yield
To minimize the electron-deficient nature of cyclopenta-
dienone 2, a Diels–Alder reaction was performed with bis(-
para-tert-butylphenyl)acetylene[5b] under microwave irradia-
tion. Again, the substrate remained unreacted. Indeed, the
bromine atom decreased the energy of the HOMO, which
was already stabilized by the ester groups, thereby prevent-
ing the reaction from occurring.
1
and a purity of close to 80% (based on H NMR spectrosco-
py). The loss of the solubilizing ester groups, presumably
owing to their hydrolysis, made this molecule highly insolu-
ble and prevented any further purification or synthesis. The
lack of reactivity of this substrate forced us to consider a
palladium coupling reaction.
Finally, Ullmann-type conditions,[28] which are known to
favor the homocoupling of electron-deficient substrates,
were used. Cyclopentadienone 2 was heated to reflux in pyr-
idine in the presence of copper powder. Dimer 5 was ob-
tained in 11% yield but the subsequent oxidative coupling
reaction with ironACTHNUTRGNEUNG(III) chloride to co-planarize the two sub-
units and generate the perylene core was unsuccessful. This
result encouraged us to consider a new target molecule.
The second synthetic route was based on the homocou-
pling of a brominated derivative as the key step (Scheme 3,
route 2). In contrast with the first route, the formation of
the perylene core would be performed in two steps instead
of one. Thus, a first bond would be formed between the two
aromatic compounds and then the formation of the second
bond would be carried out by a cyclodehydrogenation reac-
tion. Bromination of acenaphthenequinone by using a Frie-
del–Crafts reaction was successfully performed in bromine
under refluxing conditions[22] in 62% yield. Bromoacenaph-
thenequinone was then reacted with the acetonedicarboxy-
late under basic conditions to give cyclopentadienone 2 in
79% yield.
Alternative strategy: In our new design, phenyl groups re-
placed ester groups in the chassis. This modification was ex-
pected to avoid the strongly electron-deficient nature of the
first chassis, and hence increase its reactivity during the syn-
thesis. These phenyl groups must be functionalized to intro-
duce alkynes in subsequent steps; therefore, iodine atoms
were introduced onto the aromatic rings in the para posi-
tions with respect to the chassis.
In our first approach, we started with the synthesis of
di(4-iodophenyl)cyclopentadienone (6; Scheme 4). Com-
pound 6 was obtained from a double Knoevenagel reaction
between acenaphthenequinone and 1,3-di(4-iodophenyl)pro-
pan-2-one under basic conditions.[5b] In this case, potassium
hydroxide was chosen as the base, because hydrolysis was
no longer possible. Di(4-iodophenyl)cyclopentadienone (6)
was obtained in 44% yield and submitted to a Diels–Alder
reaction with 1,2-di(4-tert-butylphenyl)ethyne.[5b] After aro-
matization by evolution of carbon monoxide, compound 7
was obtained in 44% yield. With a convergent synthesis in
mind, it seemed preferable to introduce alkynes before the
oxidative coupling; however, the Scholl coupling did not
work in the presence of TMS-protected acetylene groups.
Therefore, we had to first carry out the oxidative coupling
then perform a quadruple Sonogashira reaction on the poly-
aromatic platform to introduce the alkyne groups.
After obtaining compound 2, several transition-metal-cat-
alyzed coupling reactions were attempted. First, Suzuki-type
conditions with a palladium-based catalyst were considered.
The formation of the dimer in one step by reaction of
0.5 equivalent of bis-pinacolborane was unsuccessful, despite
the use of catalysts such as [PdCl2ACTHNUTRGNEUNG
(dppf)],[23] which con-
tained a chelating dppf (1,1’-bis(diphenylphosphino)ferro-
cene) ligand that can facilitate the reductive elimination
step during the catalytic cycle, or even the universal Buch-
wald catalyst.[24] Therefore, to isolate the boronic ester, we
considered the reaction with bis-pinacolborane in the pres-
ence of a palladium catalyst.[25] In all cases, the starting ma-
terial was recovered in quantitative yield, which again dem-
onstrated the lack of reactivity of this substrate.
We also tried Stille homocoupling reactions with bis-
ACHTUNGTRENNUNG(tributyl)stannane. In this case, yellowish compounds were
obtained but none were the target dimer. A coupling reac-
tion based on a bimetallic Ni/Zn system in the presence of
triphenylphosphine was also attempted, which again did not
give a favorable outcome.[26] Even the coupling product that
contains phosphine substituents, which is usually observed
as a side-product, was not obtained here. Finally, coupling
conditions based on a bimetallic Co/Mn system that was re-
cently described in the literature did not give the desired
molecule.[27]
To explain this lack of reactivity, the first hypothesis we
put forward was that the intermediate formed after the oxi-
dative addition step was too stable and, therefore, unlikely
to react further to give the desired product. The second hy-
pothesis was that the PdII species could not be reduced to
Pd0 to complete the cycle during the reductive elimination
step. There is no evidence to confirm either of these hypoth-
eses.
Scholl coupling conditions were applied to the diiodo pre-
cursor (7) with ironACTHNUTRGNE(NUG III) chloride acting as the oxidant.
After addition of the oxidant to the mixture, the solution
immediately changed color from yellow to red. Unfortunate-
ly, despite several attempts at purification, MS analysis
showed a mixture of the product with other chlorinated de-
rivatives that could not be separated by column chromatog-
raphy. When the coupling was performed in the presence of
10% nitromethane, the solution showed a different color
change from the previous conditions, from yellow to green,
with a strong red shift that is characteristic of such target
compounds. However, the desired compound was not ob-
tained. After purification and characterization, we estab-
lished that we had obtained an overcyclized compound (8),
ꢀ
which contained six additional C C bonds (Scheme 4,
shown in bold).[18] The green compound, which underwent
3026
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 3023 – 3031

Synthesis of Polycyclic Aromatic Nanovehicles
FULL PAPER
Scheme 4. a) 1,2-Di(4-tert-butylphenyl)ethyne (1 equiv), Ph2O, 2608C, MW (300 W), 45 min, 44%; b) FeCl3 (25 equiv), CH2Cl2/CH3NO2 (9:1), RT,
15 min, Ar, 19%; c) 9-ethynyltriptycene (8 equiv), [PdACTHUNTRGNE(UNG PPh3)4] (10%), CuI (20%), piperidine/THF (1:1), 808C, 6 h, Ar, 34%. In molecules 8 and 9, the
ꢀ
additional unexpected C C bonds obtained in the oxidative coupling are highlighted in bold.
six intramolecular cyclodehydrogenation reactions, was iso-
lated in 19% yield.
Alder reaction under microwave irradiation with 1,2-bis(3,5-
di-tert-butylphenyl)ethyne (11). After aromatization by evo-
lution of carbon monoxide, the half-chassis (12), which pos-
sessed not two but four tert-butyl groups, was obtained in
50% yield.[5b] With diiodide derivative 12 in hand, we per-
formed an oxidative coupling reaction to expand the aro-
matic platform. In the absence of nitromethane, the product
was formed together with chlorinated derivatives that were
inseparable by column chromatography. However, the pres-
ence of 10% of nitromethane in the reaction medium al-
lowed the formation of the coupling product 13 in 62%
yield.
Geometry optimization (GAMESS)[29] showed that this
overcyclized chassis (8) had a curved shape at both extremi-
ties (Figure 1). This specificity was of interest because it was
an appropriate shape to act as a cargo zone. As in fullerenes,
this geometry can be explained by the presence of alternat-
ing five and six-membered rings in the polycyclic aromatic
hydrocarbon platform. The final step, which involved con-
necting the four triptycene wheels through a quadruple So-
nogashira coupling process,[30] successfully afforded nanove-
hicle 9 in 34% yield.
To test the reactivity of the four iodine centers towards
Sonogashira coupling,[30] we performed such reactions with
(3-cyanopropyl)dimethylsilylacetylene. The use of a (3-cya-
nopropyl)dimethylsilyl protecting group, which is a polar an-
alogue of trimethylsilyl,[31] allowed the easy separation of
the target molecule from the expected side-products (triyne,
diyne, and monoyne). Chassis 14, which had four protected
alkyne groups, was obtained in 24% yield, which corre-
sponded to a satisfying 70% yield for each Sonogashira cou-
pling reaction. Deprotection of the four alkyne groups was
performed under basic conditions and gave compound 15 in
97% yield (Scheme 5). Nanovehicle 16 was also formed by
quadruple Sonogashira coupling in 66% yield. The reason
for this higher yield compared to the synthesis of nanovehi-
cle 9 was the higher solubility of the precursor, which al-
lowed the reaction to be performed in five times more-con-
centrated solutions.
Figure 1. Optimized geometry of chassis 8, showing the curvature of the
polyaromatic fragment.
A second approach with additional tert-butyl groups to
avoid overcyclization of the PAH platform: Our second ap-
proach strived to facilitate the purification of the chassis
after the oxidative coupling reaction through the introduc-
tion of additional solubilizing groups. Importantly, tert-butyl
substituents were introduced at the meta positions of precur-
Conclusion
We have designed and synthesized polycyclic aromatic hy-
drocarbons nanovehicles based on analogy with macroscopic
vehicles. The syntheses of two new chassis has been ach-
ieved, both of which included a dedicated cargo zone. Sub-
sequent quadruple Sonogashira coupling with ethynyltripty-
cene wheels allowed us to obtain the corresponding nanove-
hicles in moderate to good yields. Molecules in the overcycl-
ꢀ
sor 6 to prevent subsequent additional C C bond-formation
during the dimerization of the aromatic platform. These ad-
ditional tert-butyl groups also reduced the number of p-
stacking interactions between the molecules, thereby pre-
venting aggregation in solution, which is another source of
poor solubility. Cyclopentadienone 6 underwent a Diels–
Chem. Eur. J. 2012, 18, 3023 – 3031
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3027

G. Rapenne et al.
Scheme 5. a) KOH (1 equiv), RT, 2 h, Ar, 44%; b) Compound 11 (1 equiv), Ph2O, 2508C, MW (300 W), 45 min, 50%; c) FeCl3 (25 equiv), CH2Cl2/
CH3NO2 (9:1), Ar, 62%; d) (3-cyanopropyl)dimethylsilylacetylene (20 equiv), [PdACHTNUGTRENNUNG
e) K2CO3, THF/MeOH (1:1), RT, 6 h, 98%; f) 9-ethynyltriptycene (10 equiv), [PdACHTNUGTRENNUNG
NMR Spectra were recorded on Bruker Avance 300 or Avance 500 spec-
trometers and full assignments were made using COSY, ROESY, HMBC,
and HMQC methods. Chemical shifts are defined with respect to TMS=
ized series are very interesting owing to their high curva-
tures, but are much more difficult to handle owing to their
low solubilities and lower yields.
1
0 ppm for H and 13C NMR spectra and were measured relative to residu-
Work is now underway to image and manipulate the mo-
lecular nanovehicles with an STM tip and, in particular, we
are investigating the possibility of loading the cargo zone
with, for example, co-deposited C60 molecules that have
good affinities for such aromatic nanostructures.
Because polyaromatic hydrocarbons have the potential to
transport atoms over small distances of a few nanometers,[33]
we hope that a molecule equipped with wheels will be capa-
ble of transporting atoms or even molecules over larger dis-
tances across surfaces, with potential applications in molecu-
lar electronics.
al solvent peaks. The following abbreviations were used to describe the
signals: s for singlet, d for doublet, t for triplet, q for quadruplet, m for
multiplet. Compound numbering is given in Schemes 3–5. FAB and DCI
(desorption chemical ionization) mass spectrometry were performed
using a Nermag R10–10. Elemental analyses were measured on a Perkin-
Elmer CHN 2400. UV/Vis spectra were recorded at RT on a Varian Cary
5000 spectrometer (sh for shoulder).
Computational details: The calculations were performed using the
GAMESS program,[29a] with the semi-empirical RM1 Hamiltonian
model.[29b] Geometry optimization was achieved in the gas phase without
symmetry constraints. Calculation of vibrational frequencies confirmed
that the optimized geometry corresponded to a minimum.
Diethyl 8,9-bis(4-tert-butylphenyl)fluoranthene-7,10-dicarboxylate (1): In
a Schlenk flask, compound 3 (0.45 g, 1.3 mmol, 1 equiv) and 1,2-di(4-tert-
butylphenyl)ethyne (0.38 g, 1.3 mmol, 1 equiv) were stirred in diphenyl
ether (12 mL) overnight at 2508C. The solvent was removed by using a
Kugelrohr apparatus. Purification of the crude material was performed
by column chromatography on silica gel (pentane/CH2Cl2 6:4). The prod-
uct was obtained as a yellow solid in 37% yield (290 mg). 1H NMR
(300 MHz, CDCl3): d=7.97 (d, 2H, J=7.2 Hz; Hc), 7.90 (d, 2H, J=
8.1 Hz; He), 7.63 (dd, 2H, J=8.1, 7.2 Hz; Hd), 7.16 (AA’BB’, 4H, J=
8.4 Hz; Hb), 7.03 (AA’BB’, 4H, J=8.4 Hz; Ha), 4.14 (q, 4H, J=7.2 Hz;
CH2CH3), 1.24 (s, 18H; tBu), 0.86 ppm (t, 6H, J=7.2 Hz; CH2CH3);
13C NMR (75 MHz, CDCl3): d=168.9, 149.7, 138.6, 135.4, 134.9, 133.9,
133.2, 130.4, 130.0, 129.9, 128.0, 127.8, 124.2, 122.9, 61.4, 34.4, 31.3,
13.5 ppm; MS (DCI, NH3): m/z: 628.4 [M+NH4]+ (calcd 628.3); elemen-
tal analysis calcd (%) for C42H42O4: C 82.59, H 6.93; found: C 81.92,
H 6.77.
Experimental Section
Synthesis: All commercially available chemicals were reagent grade and
used without further purification. Cobalt
ACHTUNGTRNE(NUNG III) fluoride, copper iodide, ni-
tromethane, [Pd(PPh3)4], and 1-bromo-3,5-di-tert-butylbenzene were pur-
ACHTUNGTRENNUNG
chased from Aldrich. Trimethylsilylacetylene, triisopropylsilylacetylene,
trifluoroacetic acid (TFA), diethyl acetonedicarboxylate, acenaphthene-
quinone, and FeCl3 were purchased from Acros. Ethynylbenzene and pyr-
idine were purchased from Lancaster and piperidine was purchased from
Fluka. Triethylamine was dried over KOH, THF over sodium with benzo-
phenone, MeOH over magnesium, and CH2Cl2 over CaH2. All reactions
were carried out using standard Schlenk techniques under an argon at-
mosphere. Flash column chromatography was carried out on silica gel
230–400 mesh from SDS.
Diethyl 3-bromo-8-oxo-8H-cyclopenta-(a)-acenaphthylene-7,9-dicarboxy-
late (2): In
a Schlenk flask, 4-bromo-acenaphthenequinone (1.67 g,
(3-Cyanopropyl)dimethylsilylacetylene,[31]
4-bromoacenaphthenequi-
6.4 mmol, 1 equiv), diethyl acetonedicarboxylate (1.16 mL, 6.4 mmol,
1 equiv), and triethylamine (9.2 mmol) were stirred in distilled MeOH
(17 mL) for 30 min at RT. The solvent was removed under reduced pres-
sure and the crude material was purified by column chromatography on
none,[32] 1,2-di(4-tert-butylphenyl)ethyne,[5b] 1,3-di(4-iodophenyl)propan-
2-one,[5b] and 9-ethynyltriptycene[5b] were prepared according to literature
procedures.
3028
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 3023 – 3031

Synthesis of Polycyclic Aromatic Nanovehicles
FULL PAPER
silica gel (CH2Cl2/EtOAc, 98:2). The product was obtained as a purple
solid in 79% yield (2.15 g). 1H NMR (300 MHz, CDCl3): d=8.67 (dd,
1H, J=7.2, 0.5 Hz; He), 8.47 (d, 1H, J=7.7 Hz; Hd’), 8.22 (dd, 1H, J=
8.5, 0.5 Hz; Hc), 7.83 (dd, 1H, J=8.5, 7.2 Hz; Hd), 7.63 (d, 1H, J=
7.7 Hz; Hc’), 4.46 (q, 2H, J=7.1 Hz; CH2), 4.45 (q, 2H, J=7.1 Hz; CH2),
1.46 (t, 3H, J=7.1 Hz; CH3), 1.45 ppm (t, 3H, J=7.1 Hz; CH3);
13C NMR (75 MHz, CDCl3): d=191.9, 164.8, 164.5, 162.1, 162.0, 146.2,
132.5, 130.9, 130.1, 129.8, 129.2, 128.7, 128.6, 128.1, 126.7, 112.6, 112.3,
61.2, 14.3 ppm; MS (DCI, NH3): m/z: 427.2 [M]+ (calcd 427.0); elemental
analysis calcd (%) for C21H15BrO5: C 59.04, H 3.54; found: C 58.34,
H 3.42.
121.4, 93.5 ppm; MS (DCI, NH3): m/z: 1129.4 [M]+ (calcd 1129.5); ele-
mental analysis calcd (%) for C27H14I2O: C 53.32, H 2.32; found: C 53.12,
H 2.21.
8,9-Bis(4-tert-butylphenyl)-7,10-bis(4-iodophenyl)fluoranthene (7): In
a
Schlenk flask, compound 6 (281 mg, 0.46 mmol, 1 equiv) and 1,2-di(4-tert-
butylphenyl)ethyne (134 mg, 0.46 mmol, 1 equiv) were heated in diphenyl
ether (2 mL) at 2608C under microwave irradiation (300 W) for 45 min.
After evaporation of the solvent by using a Kugelrohr apparatus, the
crude material was purified by column chromatography on silica gel
(pentane/CH2Cl2 6:4). The product was obtained as a yellow solid in
44% yield (176 mg). 1H NMR (300 MHz, CD2Cl2): d=7.75 (dd, 2H, J=
8.2, 1.9 Hz; He), 7.65 (AA’BB’, 4H, J=8.3 Hz; Hc), 7.35 (t, 2H, J=8.2,
7.2 Hz; Hf), 7.07 (AA’BB’, 4H, J=8.3 Hz; Hd), 7.05 (dd, 2H, J=8.2,
1.9 Hz; Hg), 6.88 (AA’BB’, 4H; Hb), 6.71 (AA’BB’, 4H; Ha), 1.13 ppm (s,
18H; tBu); 13C NMR (75 MHz, CDCl3): d=148.1, 141.0, 139.4, 137.2,
136.2, 136.1, 136.1, 136.0, 133.1, 132.1, 130.7, 129.6, 127.6, 126.6, 123.4,
123.0, 92.5, 34.1, 31.1 ppm; MS (DCI, NH3): m/z: 888.1 [M+NH4]+ (calcd
888.1); elemental analysis calcd (%) for C48H40I2: C 66.22, H 4.63; found:
C 64.97, H 4.27.
Diethyl 8-oxo-8H-cyclopenta-(a)-acenaphthylene-7,9-dicarboxylate (3):
In a Schlenk flask, acenaphthenequinone (0.5 g, 2.75 mmol, 1 equiv), di-
ethyl acetonedicarboxylate (0.5 mL, 2.75 mmol, 1 equiv), and triethyla-
mine (0.6 mL, 4.25 mmol, 1.5 equiv) were stirred in distilled MeOH
(7 mL) for 30 min at RT. The solvent was removed under reduced pres-
sure and the crude material was purified by column chromatography on
silica gel (CH2Cl2/EtOAc, 0!5%). The product was obtained as a purple
solid in 47% yield (450 mg). 1H NMR (300 MHz, CDCl3): d=8.63 (d,
2H, J=7.2 Hz; Ha), 8.05 (d, 2H, J=8.4 Hz; Hc), 7.75 (dd, 2H, J=8.4,
7.2 Hz; Hd), 4.45 (q, 4H, J=7.1 Hz; CH2), 1.45 ppm (t, 6H, J=7.1 Hz;
CH3); 13C NMR (75 MHz, CDCl3): d=186.9, 172.4, 164.8, 132.5, 131.7,
131.1, 130.1, 128.8, 126.7, 125.4, 61.2, 14.3 ppm; MS (DCI, NH3): m/z:
348.1 [M]+ (calcd 348.1); elemental analysis calcd (%) for C21H16O5:
C 72.41, H 4.63; found: C 72.11, H 4.46.
Overcyclized chassis 8: To a solution of compound 7 (50 mg, 0.057 mmol,
1 equiv) in distilled CH2Cl2 (6 mL) was added dropwise a solution of
ironACTHNUGRTENUNG(III) chloride (232 mg, 1.43 mmol, 25 equiv) in nitromethane
(0.25 mL) at RT. A stream of argon was bubbled through the reaction
mixture during the reaction. After stirring for 15 min at RT, MeOH was
added and the precipitate was filtered. After drying under reduced pres-
sure, the crude product was purified by column chromatography on silica
gel (cyclohexane/CH2Cl2, 0!10%). The product was obtained as a green
Macrocycle 4: In
a
Schlenk flask, compound
1 (0.10 g, 0.16 mmol,
1 equiv) and cobaltACHTUNGTRENNUNG(III) fluoride (0.02 g, 0.16 mmol, 1 equiv) were stirred
1
solid in 19% yield (9.3 mg). H NMR (300 MHz, CDCl3): d=8.30 (d, 4H,
in trifluoroacetic acid (1 mL) overnight at 708C. CH2Cl2 (5 mL) was
added and the mixture was poured into water (50 mL) and extracted
with CHCl3 (50 mL). The organic layers were dried over MgSO4. After
the solvent had been evaporated, the crude material was purified by
column chromatography on silica gel (CH2Cl2/cyclohexane 1:1). The
product was obtained as an orange solid in 99% yield (89 mg). 1H NMR
(300 MHz, CDCl3): d=9.35 (d, 4H, J=7.0 Hz), 7.97 (d, 4H, J=8.0 Hz),
7.91 (d, 4H, J=8.1 Hz), 7.83 (m, 8H), 7.60 (dd, 4H, J=8.0, 7.0 Hz), 7.53
(d, 8H, J=8.3 Hz), 7.39 (d, 8H, J=8.3 Hz), 7.20 (dd, 4H, J=8.0, 2.0 Hz),
6.29 (4H, d, J=8.0 Hz), 4.17 (q, 4H, J=7.1 Hz; CH2), 0.92 ppm (t, 6H,
J=7.1 Hz; CH3); 13C NMR (75 MHz, CDCl3): d=177.4, 171.8, 149.4,
138.7, 135.6, 134.9, 133.8, 133.4, 130.6, 130.0, 129.9, 128.0, 127.8, 124.2,
122.9, 61.6, 34.6, 31.4, 13.6 ppm; MS (DCI, NH3): m/z: 1129.4 [M]+ (calcd
1129.5); elemental analysis calcd (%) for C80H72O6: C 85.07, H 6.43;
found: C 84.92, H 6.30.
J=2.0 Hz), 7.90 (m, 4H), 7.80–7.55 (m, 16H), 7.21 (d, 4H, J=7.2 Hz),
1.28 ppm (s, 36H; tBu); 13C NMR (75 MHz, CDCl3) the sample was not
soluble enough to obtain a satisfactory signal-to-noise ratio; MS (DCI,
NH3): m/z: 1725.2 [M+H]+ (calcd 1725.1); HRMS (FAB): m/z: calcd for
C96H64I4: 1724.1187 [M]+; found: 1724.1166.
Overcyclized nanovehicle 9: In a Schlenk tube, a solution of compound 8
(30 mg, 0.017 mmol, 1 equiv) and 9-ethynyltriptycene (25 mg,
0.089 mmol, 8 equiv) in THF/piperidine (1:1, 30 mL) was degassed under
argon during 20 min. Copper iodide (15 mg, 78.7 mmol) and [PdACHTUNGTRENNUNG(PPh3)4]
(15 mg, 13 mmol) were added and the mixture was heated for 6 h at 808C.
After the solvent had been concentrated, the crude product was purified
by column chromatography on silica gel (CH2Cl2/toluene, 0!100%).
The product was obtained as a dark-green solid in 34% yield (67 mg).
1H NMR (300 MHz, C6D6): d=8.13 (d, 4H, J=2.0 Hz), 7.96 (d, 12H, J=
8.0 Hz; H5), 7.65 (dd, 4H, J=7.8, 1.1 Hz; Hc), 7.52 (m, 4H), 7.3–7.0 (m,
16H), 7.0–6.7 (m, 36H, H2,3,4), 6.13 (s, 4H; H1), 1.07 ppm (s, 36H; tBu);
13C NMR (75 MHz, CDCl3) the sample was not soluble enough to obtain
a satisfactory signal-to-noise ratio; UV/Vis (CH2Cl2): lmax (e)=359 sh
(40800), 397 (36300), 421 (37300), 460 sh (40800), 572 sh (57600), 616
(40800), 669 nm (57600 molꢀ1 dm3 cmꢀ1); MS (DCI, NH3): m/z: 2324.7
[M]+ (calcd 2324.9); HRMS (FAB): m/z: calcd for C184H116: 2324.9077
[M]+; found: 2324.8957.
Tetraethyl-7’,8-dioxo-2’,3-bi-(8H)-cyclopenta-(a)-acenaphthylene)-
6’,7,8’,9-tetracarboxylate (5): To
a solution of compound 2 (65 mg,
0.15 mmol, 1 equiv) in pyridine (2 mL) was added copper powder
(copper bronze) (40 mg, 6.32 mmol, 4 equiv). The medium was heated to
reflux for 3 h. After evaporation of the solvent, the crude material was
purified by column chromatography on silica gel (CH2Cl2/AcOEt, 0!
5%). The product was obtained as a yellow solid in 11% yield (6.2 mg).
1H NMR (300 MHz, CDCl3): d=8.84 (d, 2H, J=7.3 Hz; Hd’), 8.74 (dd,
2H, J=5.8, 2.0 Hz; He), 7.88 (d, 2H, J=7.3 Hz; Hc’), 7.71–7.68 (m, 4H;
Hcꢀd), 4.49 (q, 4H, J=7.1 Hz; CH2), 4.48 (q, 4H, J=7.1 Hz; CH2), 1.49
(t, 6H, J=7.1 Hz; CH3), 1.48 ppm (t, 6H, J=7.1 Hz; CH3); 13C NMR
(75 MHz, CDCl3): d=192.3, 191.9, 165.6, 162.2, 159.0, 158.9, 146.2, 131.3,
130.5, 129.1, 128.9, 127.8, 111.1, 61.0, 14.3 ppm; MS (DCI, NH3): m/z:
695.1 [M+H]+ (calcd 695.6).
[(3,5-Di-tert-butylphenyl)ethynyl]triisopropylsilane (10): In
a Schlenk
flask, a solution of 1-bromo-3,5-di-tert-butylbenzene (1.346 g, 5 mmol,
1 equiv) and triisopropylsilylacetylene (1.3 mL, 5.5 mmol, 1.1 equiv) in pi-
peridine (40 mL) was flushed with argon for 20 min. After addition of
[PdACHTNUGRTNEUNG(PPh3)4] (0.27 g, 1.1 mmol, 22%) and CuI (0.088 g, 0.5 mmol, 9.2%),
the mixture was stirred at 808C for 6.5 h. The reaction mixture was then
washed with a saturated solution of NH4Cl (100 mL) and extracted with
CH2Cl2 (3ꢂ150 mL). The organic layers were washed with 1m HCl
(200 mL) and dried over MgSO4. After evaporation of the solvent, the
crude material was purifed by column chromatography on silica gel (pe-
troleum ether). The product was obtained as a clear oil in quantitative
7,9-Bis(4-iodophenyl)-8H-cyclopenta-(a)-acenaphthylen-8-one (6): To a
solution of acenaphthenequinone (382 mg, 2.1 mmol, 1 equiv) and 1,3-
di(4-iodophenyl)propan-2-one (1.07 g, 2.1 mmol, 1 equiv) in distilled
MeOH (12 mL), was added potassium hydroxide (118 mg, 2.1 mmol,
1 equiv). The solution turned green immediately and was stirred at RT
for a further 2 h. After evaporation of the solvent, the crude material
was purified by trituration with MeOH. The product was obtained as a
blue solid in 44% yield (1.04 g). 1H NMR (300 MHz, CD2Cl2): d=8.07
(dd, 2H, J=7.2, 0.6 Hz; Hg), 7.97 (dd, 2H, J=8.2, 0.6 Hz; He), 7.92
(AA’BB’, 4H, J=8.6 Hz; Hc), 7.63 (dd, 2H, J=8.2, 7.2 Hz; Hf), 7.61 ppm
(AA’BB’, 4H, J=8.6 Hz; Hd); 13C NMR (75 MHz, CD2Cl2): d=202.6,
155.4, 150.7, 146.7, 143.9, 131.0, 128.1, 126.4, 125.3, 123.3, 122.4, 122.3,
1
yield (2.1 g). H NMR (300 MHz, CDCl3): d=7.40 (t, 1H, J=1.8 Hz; Ha),
7.32 (d, 2H, J=1.8 Hz; Hb), 1.33 (s, 18H; tBu), 1.20 (d, 18H; MeiPr),
1.15 ppm (hept, 3H; CHiPr); 13C NMR (75 MHz, CDCl3): d=150.6, 126.1,
122.8, 122.5, 108.3, 88.7, 34.7, 31.3, 18.7, 11.4 ppm; MS (DCI, NH3): m/z:
388.3 [M+NH4]+ (calcd 388.7).
1,2-Bis(3,5-di-tert-butylphenyl)ethyne (11): To a solution of compound 10
(5 mmol) in DMSO (50 mL),
a solution of potassium fluoride
Chem. Eur. J. 2012, 18, 3023 – 3031
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3029

G. Rapenne et al.
(11.5 mmol, 2.3 equiv) in water (5 mL) was added and the mixture was
heated to reflux at 808C for 3 h. The medium was then poured into water
(100 mL), extracted with toluene (3ꢂ100 mL), washed with water
(100 mL), and dried over MgSO4. After evaporation of the solvent, the
crude material was purified by column chromatography on silica gel
(pentane) to afford (3,5)-di-tert-butylphenyl)ethyne as a clear oil in 80%
yield. In a Schlenk flask, a solution of this oil (835 mg, 3.89 mmol,
ꢀ1.75 ppm; MS (DCI, NH3): m/z: 2055.3 [M]+ (calcd 2055.2); HRMS
(FAB): m/z: calcd for C144H156N4Si4: 2053.1407 [M]+; found: 2053.1354.
Tetraethynyl chassis (15): Potassium carbonate (30 mg, 0.21 mmol,
20 equiv) was added to a solution of compound 14 (23 mg, 0.012 mmol,
1 equiv) in THF/MeOH (1:1, 6 mL). The solution was stirred for 6 h at
RT. After evaporation, the crude product was purified by column chro-
matography on silica gel (toluene). The product was obtained as a purple
solid in 97% yield (18 mg). 1H NMR (300 MHz, C6D6): d=7.62 (d, 4H,
J=7.8 Hz; He), 7.37 (AA’BB’, 8H, J=8.2 Hz; Hd), 7.20 (AA’BB’, 8H,
J=8.3 Hz; Hc), 7.06 (d, 4H, J=1.1 Hz; Ha), 6.96 (d, 4H, J=7.7 Hz; Hf),
6.86 (d, 8H, J=1.1 Hz; Hb), 2.74 (s, 4H; Halkyne), 1.13 ppm (s, 72H; tBu);
13C NMR (75 MHz, CDCl3): d=151.7, 149.0, 141.6, 141.1, 138.8, 137.2,
136.8, 136.5, 135.7, 135.6, 132.0, 130.9, 130.6, 126.5, 125.5, 125.0, 124.3,
83.4, 77.5, 34.3, 31.2 ppm; MS (DCI, NH3): m/z: 1554.1 [M]+ (calcd
1553.8).
1 equiv)
4.8 mmol, 1.1 equiv), in piperidine (40 mL) was flushed with argon for
20 min. After addition of [Pd(PPh3)4] (0.24 g, 0.4 mmol, 4.5%) and CuI
and
1-bromo-3,5-di-tert-butyl-1-bromobenzyne
(0.84 mL,
AHCTUNGTRENNUNG
(0.08 g, 0.2 mmol, 4.5%), the mixture was stirred overnight at 808C. The
mixture was then poured into a saturated solution of NH4Cl and extract-
ed with CH2Cl2 (100 and 50 mL). The organic layers were washed with
saturated NH4Cl (100 mL) and water (100 and 50 mL) and dried over
MgSO4. After evaporation of the solvent, the crude material was purified
by column chromatography on silica gel (cyclohexane). The product was
obtained as a white solid in 56% yield (1.13 g). 1H NMR (300 MHz,
CDCl3): d=7.44 (d, 4H, J=1.8 Hz; Hb), 7.32 (t, 2H, J=1.8 Hz; Ha),
1.38 ppm (s, 36H; tBu); 13C NMR (75 MHz, CDCl3): d=150.7, 125.0,
122.5, 122.4, 89.2, 34.8, 31.3 ppm; MS (DCI, NH3): m/z: 420.3 [M+NH4]+
(calcd 420.6); elemental analysis calcd (%) for C30H42O: C 89.49,
H 10.51; found: C 89.06, H 10.01.
Nanovehicle (16): In a Schlenk tube, a solution of compound 15 (22 mg,
14 mmol, 1 equiv) and 9-ethynyltriptycene (25 mg, 0.089 mmol, 8 equiv) in
THF/piperidine (1:1, 4 mL) was degassed under argon for 20 min.
Copper iodide (12 mg, 63 mmol) and [PdACHTNUTRGNEUNG(PPh3)4] (12 mg, 10.4 mmol) were
added and the mixture was heated for 6 h at 808C. After the solvent had
been concentrated, the crude product was purified by column chromatog-
raphy on silica gel (CH2Cl2/toluene, 0!100%). The product was ob-
tained as a purple solid in 66% yield (24 mg). 1H NMR (300 MHz,
C6D6): d=8.07 (m, 12H; H5), 7.72 (d, 4H; J=7.8 Hz; He), 7.56 (d, 8H,
J=8.3 Hz; Hc), 7.40 (t, 8H, J=8.3 Hz; Hd), 7.25–7.21 (m, 12H; Hb,f), 7.13
(d, 4H, J=1.8 Hz; Ha), 7.01–6.93 (m, 36H; H2,3,4), 5.84 (s, 4H; H1),
1.20 ppm (s, 72H; tBu); 13C NMR (75 MHz, C6D6): d=150.1, 145.8,
145.7, 142.1, 139.9, 138.4, 137.9, 137.7, 133.1, 132.0, 131.8, 127.8, 126.6,
126.2, 124.4, 123.6, 122.5, 93.9, 85.4, 54.9, 54.5, 35.3, 32.2, 32.1, 30.8, 30.7,
23.6, 15.0, 14.9 ppm; UV/Vis (CH2Cl2): lmax (e)=326 (12900), 466 sh
(3460), 493 (5980), 527 (11340), 569 nm (14800 molꢀ1 dm3 cmꢀ1); MS
(DCI, NH3): m/z: 2563.3 [M]ꢀ (calcd 2563.4); HRMS (FAB): m/z: calcd
for C200H164: 2561.2520 [M]+; found: 2561.2497.
8,9-Bis(3,5-di-tert-butylphenyl)-7,10-bis(4-iodophenyl)fluoranthene (12):
In a Schlenk flask, compounds 6 (923 mg, 1.52 mmol, 1 equiv) and 11
(612 mg, 1.52 mmol, 1 equiv) were heated in diphenyl ether (5 mL) at
2508C under microwave irradiation (300 W) for 45 min. After evapora-
tion of the solvent by Kugelrohr distillation, the crude material was puri-
fied by column chromatography on silica gel (CH2Cl2/pentane, 0!10%).
The product was obtained as a yellow solid in 50% yield (760 mg).
1H NMR (300 MHz, CDCl3): d=7.77 (d, 2H, J=7.9 Hz; He), 7.66 (d,
4H, J=8.1 Hz; Hc), 7.42–7.33 (m, 2H; Hf), 7.15–7.03 (m, 4H, Hd), 6.92–
6.90 (m, 4H; Ha,g), 6.65 (d, 4H, J=1.8 Hz; Hb), 1.03 ppm (s, 36H; tBu);
13C NMR (75 MHz, CDCl3): d=148.8, 141.6, 139.7, 138.1, 137.1, 136.2,
136.1, 133.2, 132.3, 129.6, 127.6, 126.6, 126.3, 123.1, 118.6, 92.4, 34.3,
31.2 ppm; MS (DCI, NH3): m/z: 1000.2 [M]+ (calcd 1000.8); elemental
analysis calcd (%) for C56H56I2: C 68.43, H 5.74; found: C 68.04, H 5.37.
Tetraiodo chassis (13): To
0.057 mmol, 1 equiv) in distilled CH2Cl2 (6 mL), was added dropwise a
solution of iron(III) chloride (232 mg, 1.43 mmol, 25 equiv) in nitrome-
a solution of compound 12 (50 mg,
Acknowledgements
ACHTUNGTRENNUNG
thane (0.25 mL) at RT. An argon stream was bubbled through the reac-
tion mixture during the reaction. After stirring the solution for 15 min at
RT, MeOH was added and the precipitate was filtered. After drying
under reduced pressure, the crude product was purified by column chro-
matography on silica gel (cyclohexane/CH2Cl2, 0!10%). The product
This work was supported by the CNRS, the University Paul Sabatier
(Toulouse), the European Community, and the ANR P3N (AUTOMOL
project n8 ANR 09-NANO-040). H.P.J.deR. thanks the French Ministry
of National Education for a PhD Fellowship. R.G. thanks the French
Ministry of National Education and the Ecole Normale Supꢀrieure of
Cachan for a PhD Fellowship. R.C. thanks the NSF for financial support
(REU US-France project). Dr. C. Joachim and Dr. A. Gourdon are
thanked for fruitful discussions and Dr. Isabelle M. Dixon is acknowl-
edged for her corrections and comments on this manuscript. Christine
Viala is also thanked for her help on the synthesis of the starting materi-
als as well as for NMR measurements.
1
was obtained as a purple solid in 62% yield (35 mg). H NMR (300 MHz,
CDCl3): d=7.92 (m, 4H, He), 7.65 (m, 8H; Ha,f), 7.05 (m, 8H; Hc), 6.89
(m, 8H; Hd), 6.59 (m, 8H; Hb), 0.98 ppm (s, 72H; tBu); 13C NMR
(75 MHz, CDCl3) the sample was not soluble enough to obtain a satisfac-
tory signal-to-noise ratio; MS (DCI, NH3): m/z: 1961.5 [M]+ (calcd
1961.6); HRMS (FAB): m/z: calcd for C112H108I4: 1960.4630 [M]+; found:
1960.4607.
Tetra(1-[2-(3-cyanopropyl)dimethylsilyl]ethynyl) chassis (14): In
a
Schlenk tube, a solution of compound 13 (27 mg, 0.013 mmol, 1 equiv)
and (3-cyanopropyl)dimethylsilylacetylene (42 mg, 0.27 mmol, 20 equiv)
in THF/piperidine (1:1, 4 mL) was degassed under argon for 20 min.
[1] a) V. Balzani, A. Credi, M. Venturi, Molecular Devices and Ma-
chines—A Journey into the Nano World, Wiley-VCH, Weinheim, 2nd
ed. 2008; b) G. Vives, H.-P. Jacquot de Rouville, A. Carella, J.-P.
Launay, G. Rapenne, Chem. Soc. Rev. 2009, 38, 1551.
Copper iodide (12 mg, 63 mmol) and [PdACHTNUTRGNEUNG(PPh3)4] (12 mg, 10.4 mmol) were
added and the mixture was heated overnight at RT. After concentration
of the solvent, the crude product was purified by column chromatography
on silica gel (petroleum ether/CH2Cl2, 50!70%). The product was ob-
tained pure as a purple solid in 24% yield (6.4 mg). 1H NMR (300 MHz,
CDCl3): d=7.88 (d, 4H, J=7.7 Hz; He), 7.43 (AA’BB’, 8H, J=8.3 Hz;
Hd), 7.28 (AA’BB’, 8H, J=8.3 Hz; Hc), 6.85 (t, 4H, J=1.7 Hz; Ha), 6.78
(d, 4H, J=7.7 Hz; Hf), 6.62 (d, 8H, J=1.7 Hz; Hb), 2.46 (t, 8H, J=
6.9 Hz; CH2CN), 1.91–1.81 (m, 16H; CH2CH2), 0.98 (s, 72H; tBu), 0.9–
0.85 ppm (m, 24H; SiCH3); 13C NMR (75 MHz, CDCl3): d=148.8, 141.7,
140.9, 138.1, 136.8, 136.3, 136.1, 131.7, 130.4, 130.3, 126.2, 125.0, 124.1,
121.8, 121.0, 119.7, 118.7, 106.6, 91.9, 76.5, 34.3, 31.2, 20.6, 20.4, 15.7,
[2] a) G. Rapenne, Org. Biomol. Chem. 2005, 3, 1165; b) A. Carella, C.
Coudret, G. Guirado, G. Rapenne, G. Vives, J.-P. Launay, Dalton
Trans. 2007, 177; c) E. R. Kay, D. A. Leigh, F. Zerbetto, Angew.
Chem. 2007, 119, 72; Angew. Chem. Int. Ed. 2007, 46, 72.
[3] a) J. M. Chance, J. H. Geiger, K. Mislow, J. Am. Chem. Soc. 1989,
111, 2326; b) J. Clayden, J. H. Pink, Angew. Chem. 1998, 110, 2040;
Angew. Chem. Int. Ed. 1998, 37, 1937; c) A. Carella, R. Poteau, G.
Rapenne, J.-P. Launay, Chem. Eur. J. 2008, 14, 8147; d) A. Carella,
G. Vives, T. Cox, J. Jaud, G. Rapenne, J. P. Launay, Eur. J. Inorg.
Chem. 2006, 980; e) F. Chiaravalloti, L. Gross, K. H. Rieder, S. Stoj-
kovic, A. Gourdon, C. Joachim, F. Moresco, Nat. Mater. 2007, 6, 30.
3030
www.chemeurj.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 3023 – 3031

Synthesis of Polycyclic Aromatic Nanovehicles
FULL PAPER
[4] a) L. Grill, K. H. Rieder, F. Moresco, G. Rapenne, S. Stojkovic, X.
Bouju, C. Joachim, Nat. Nanotechnol. 2007, 2, 95; b) G. Vives, J. M.
Tour, Tetrahedron Lett. 2009, 50, 1427 and references therein.
[5] a) G. Jimenez-Bueno, G. Rapenne, Tetrahedron Lett. 2003, 44, 6261;
b) G. Rapenne, G. Jimenez-Bueno, Tetrahedron 2007, 63, 7018; c) L.
Grill, K. H. Rieder, F. Moresco, G. Jimenez-Bueno, C. Wang, G.
Rapenne, C. Joachim, Surf. Sci. 2005, 584, L153.
[6] a) Y. Shirai, A. J. Osgood, Y. Zhao, K. F. Kelly, J. M. Tour, Nano
Lett. 2005, 5, 2330; b) Y. Shirai, A. J. Osgood, Y. Zhao, Y. Yao, L.
Saudan, H. Yang, C. Yu-Hung, L. B. Alemany, T. Sasaki, J.-F. Morin,
J. M. Guerrero, K. F. Kelly, J. M. Tour, J. Am. Chem. Soc. 2006, 128,
4854; c) J.-F. Morin, Y. Shirai, J. M. Tour, Org. Lett. 2006, 8, 1713;
d) T. Sasaki, A. J. Osgood, L. B. Alemany, K. F. Kelly, J. M. Tour,
Org. Lett. 2008, 10, 229; e) G. Vives, J. M. Tour, Acc. Chem. Res.
2009, 42, 473; f) T. Kudernac, N. Ruangsupapichat, M. Parschau, B.
Macia, N. Katsonis, S. R. Harutyunyan, K.-H. Ernst, B. L. Feringa,
Nature 2011, 479, 208.
[18] A. Mckillop, A. G. Turrell, D. W. Young, E. C. Taylor, J. Am. Chem.
Soc. 1980, 102, 6504.
[19] Y. Li, Z. Wang, Org. Lett. 2009, 11, 1385.
[20] L. Y. Zhai, R. Shukla, R. Rathore, Org. Lett. 2009, 11, 3474.
[21] S. Nagarajan, C. Barthes, A. Gourdon, Tetrahedron 2009, 65, 3767.
[22] R. H. Mitchell, M. Chaudhary, R. V. Williams, R. Fyles, J. Gibson,
M. J. Ashwoodsmith, A. J. Fry, Can. J. Chem. 1992, 70, 1015.
[23] J. Hashim, T. N. Glasnov, J. M. Kremsner, C. O. Kappe, J. Org.
Chem. 2006, 71, 1707.
[24] K. L. Billingsley, S. L. Buchwald, J. Org. Chem. 2008, 73, 5589.
[25] P. E. Broutin, I. Cerna, M. Campaniello, F. Leroux, F. Colobert,
Org. Lett. 2004, 6, 4419.
[26] L. Dudd, M. Hart, D. Ring, E. Sondaz, J. Bonvoisin, Y. Coppel,
Inorg. Chem. Commun. 2003, 6, 1400.
[27] M. Amatore, C. Gosmini, Synlett 2009, 1073.
[28] H. W. Gibson, F. C. Bailey, J. Org. Chem. 1976, 41, 557.
[29] a) M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S.
Gordon, J. H. Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen, S. J.
Su, T. L. Windus, M. Dupuis, J. A. Montgomery, J. Comput. Chem.
1993, 14, 1347; b) G. B. Rocha, R. O. Freire, A. M. Simas, J. J. P.
Stewart, J. Comput. Chem. 2006, 27, 1101.
[7] T. Sasaki, J.-F. Morin, M. Lu, J. M. Tour, Tetrahedron Lett. 2007, 48,
5817.
[8] G. Vives, J. H. Kang, K. F. Kelly, J. M. Tour, Org. Lett. 2009, 11,
5602.
[9] C. J. Villagꢃmez, T. Sasaki, J. M. Tour, L. Grill, J. Am. Chem. Soc.
2010, 132, 16848.
[30] K. Sonogashira, T. Yatake, Y. Tohda, S. Takahashi, N. Hagihara, J.
Chem. Soc. Chem. Commun. 1977, 291.
[10] R. Rieger, K. Mꢄllen, J. Phys. Org. Chem. 2010, 23, 1644.
[11] R. Barattin, A. Gourdon, Eur. J. Org. Chem. 2009, 1022.
[12] F. Wꢄrthner, Chem. Commun. 2004, 1564.
[31] S. Hçger, K. Bonrad, J. Org. Chem. 2000, 65, 2243.
[32] H. G. Rule, S. B. Thompson, J. Chem. Soc. 1937, 1761.
[33] F. Rosei, M. Schunack, P. Jiang, A. Gourdon, E. Lægsgaard, I. Sten-
sgaard, C. Joachim, F. Besenbacher, Science 2002, 296, 328.
[13] S. Toyota, Chem. Rev. 2010, 110, 5398.
[14] R. Scholl, J. Mansfeld, Ber. Dtsch. Chem. Ges. 1910, 43, 1734.
[15] N. J. Bruce, J. Org. Chem. 1972, 37, 664.
[16] S. H. Tucker, J. Chem. Soc. 1958, 1462.
Received: September 15, 2011
Revised: November 26, 2011
[17] M. Wehmeier, M. Wagner, K. Mꢄllen, Chem. Eur. J. 2001, 7, 2197.
Published online: February 1, 2012
Chem. Eur. J. 2012, 18, 3023 – 3031
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3031