Self-assembled monolayers of alkoxy-substituted octadehydrodibenzo[12] annulenes on a graphite surface: Attempts at peri-benzopolyacene formation by on-surface polymerization

Article abstract of DOI:10.1002/chem.201000711

Self-assembled monolayers of a series of tetraalkoxy-substituted octadehydrodibenzo[12]annulene (DBA) derivatives 1c-g possessing butadiyne linkages were studied at the 1, 2, 4-trichlorobenzene (TCB) or 1-phenyloctane/graphite interface by scanning tunneling microscopy (STM). The purpose of this research is not only to investigate the structural variation of two-dimensional (2D) monolayers, but also to assess a possibility for peri-benzopolyacene formation by two-dimensionally controlled polymerization on a surface. As a result, the formation of three structures, porous, linear, and lamella structures, were observed by changing the alkyl chain length and the solute concentration. The formation of multilayers of the lamella structure was often observed for all compounds. The selection of molecular networks is basically ascribed to intermolecular and molecule-substrate interactions per unit area and network density. The selective appearance of the linear structure of Id is attributed to favorable epitaxial registry matching between the substrate lattice and the overlayer lattice. Even though the closest inter-atomic distance between the diacetylenic units of the DBAs in the lamella structure (≈0.6nm) is slightly larger compared to the typical distances necessary for topochemical polymerization, the reactivity toward external stimuli (electronic-pulse irradiation from an STM tip and UV irradiation) was investigated. Unfortunately, no evidence for polymerization of the DBAs on the surface was observed. The present results indicate the necessity for further designing a suitable system for the on-surface construction of structurally novel conjugated polymers, which are otherwise difficult to prepare.

Full text of DOI:10.1002/chem.201000711

FULL PAPER
DOI: 10.1002/chem.201000711
Self-Assembled Monolayers of Alkoxy-Substituted
Octadehydrodibenzo[12]annulenes on a Graphite Surface: Attempts at peri-
Benzopolyacene Formation by On-Surface Polymerization
Kazukuni Tahara,[a] Koji Inukai,[a] Noritaka Hara,[a] Charles A. Johnson II,[b]
Michael M. Haley,[b] and Yoshito Tobe*[a]
Abstract: Self-assembled monolayers
of a series of tetraalkoxy-substituted
octadehydrodibenzo[12]annulene
solute concentration. The formation of
multilayers of the lamella structure was
often observed for all compounds. The
selection of molecular networks is basi-
cally ascribed to intermolecular and
molecule–substrate interactions per
unit area and network density. The se-
lective appearance of the linear struc-
ture of 1d is attributed to favorable ep-
itaxial registry matching between the
substrate lattice and the overlayer lat-
tice. Even though the closest inter-
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
structure (ꢀ0.6 nm) is slightly larger
compared to the typical distances nec-
essary for topochemical polymeri-
zation, the reactivity toward external
stimuli (electronic-pulse irradiation
from an STM tip and UV irradiation)
was investigated. Unfortunately, no evi-
dence for polymerization of the DBAs
on the surface was observed. The pres-
ent results indicate the necessity for
further designing a suitable system for
the on-surface construction of structur-
ally novel conjugated polymers, which
are otherwise difficult to prepare.
(DBA) derivatives 1c–g possessing bu-
tadiyne linkages were studied at the
1,2,4-trichlorobenzene (TCB) or 1-phe-
nyloctane/graphite interface by scan-
ning tunneling microscopy (STM). The
purpose of this research is not only to
investigate the structural variation of
two-dimensional (2D) monolayers, but
also to assess a possibility for peri-ben-
zopolyacene formation by two-dimen-
sionally controlled polymerization on a
surface. As a result, the formation of
three structures, porous, linear, and la-
mella structures, were observed by
changing the alkyl chain length and the
Keywords: dehydrobenzoannulene ·
interfaces
scanning probe microscopy
self-assembly
·
monolayers
·
·
Introduction
regime.[3] From the crystal engineering concept,[4] to design
and control of the structure and functionality, 2D molecular
networks require investigations of the correlations between
structural features of constituent molecules and the resulting
topologies of surface-confined molecular architectures.[5] In
this respect, various types of 2D molecular networks have
been investigated both under ultra-high vacuum (UHV)
conditions and at liquid/solid interfaces.[1,5]
Creation of two-dimensional (2D) architectures on solid sur-
faces based on molecular self-assembly has received great
attention because of the prospective applications in the
fields of nanoscience and nanotechnology,[1,2] particularly in
view of possible surface patterning in the low nanometer
In principle, to control 2D molecular networks, the fol-
lowing noncovalent interactions and experimental parame-
ters must be taken into account. First, optimization of direc-
tional intermolecular interactions, which determine the net-
work topology, such as hydrogen bonding,[6] dipolar cou-
pling,[7] metal–ligand coordination,[8] and van der Waals in-
teractions between (sometimes interdigitating) alkyl
chains[9,10] is necessary. Molecule–substrate interactions (typ-
ically van der Waals interactions), which can be modulated
by functional groups on molecules and choice of substrate,
[a] Dr. K. Tahara, K. Inukai, N. Hara, Prof. Dr. Y. Tobe
Division of Frontier Materials Science
Graduate School of Engineering Science
Osaka University, Toyonaka, Osaka 560-8531 (Japan)
Fax : (+81)6-6850-6227
E-mail: tobe@chem.es.osaka-u.ac.jp
[b] Dr. C. A. Johnson II, Prof. Dr. M. M. Haley
Department of Chemistry and Materials Science Institute
University of Oregon, Eugene, OR 97403-1253 (USA)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000711.
Chem. Eur. J. 2010, 16, 8319 – 8328
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8319

are also important for the selection of network structures. In
particular, epitaxial matching between the substrate and
overlayer lattices would stabilize specific overlayer struc-
tures.[11] At liquid/solid interfaces, solvents apparently play
roles in network formation.[12] For instance, solvents are
often involved in the network (co-adsorption)[13] and they
affect the adsorption–desorption process too.[14] Finally, vari-
ous experimental conditions, such as timing of measure-
ments,[15] temperature,[16] and solute concentration,[17] are
also known to influence the network formation. Scanning
tunneling microscopy (STM) is a powerful tool to investi-
gate 2D molecular networks with molecular scale precision
and is typically used for this purpose.
Recently, special interest has been focused on covalently
linked 2D molecular networks on surfaces, aiming at not
only fixation of fragile 2D molecular networks, formed
through noncovalent interactions, to construct robust 2D
polymers,[18] but also the formation of molecularly well-de-
fined conducting polymers.[19–21] Additionally, surface-specif-
ic reactions in a confined space would allow for the produc-
tion of unique polymers otherwise unavailable by means of
solution-phase reactions. The programmed reactions of mo-
lecular building blocks on solid surfaces, to generate the de-
sired covalently linked 2D architectures, require precise mo-
lecular design. Pioneering works in this field were undertak-
en for topochemical 1,4-polymerization of diacetylene deriv-
atives. For example, the first oligomerization of 17,19-hexa-
triacontadiyne on graphite by using UV irradiation has been
investigated by means of UHV conditions.[19] At a liquid/
solid interface, De Schryver, De Feyter et al. reported that a
monolayer of a diacetylene-containing isophthalic acid de-
rivative underwent topochemical 1,4-polymerization upon
UV irradiation to form a p-conjugated oligomer.[20a,e] Subse-
quently, Aono et al. demonstrated nanoscale control of poly-
merization of 1,4-butadiyne derivatives on a surface by irra-
diation of an electronic pulse from an STM tip.[20b,c] In these
cases, strict control of spatial arrangement between the diac-
etylene units is necessary to initiate the topochemical 1,4-
polymerization. Other types of covalent bond forming reac-
tions on surfaces have been developed recently, such as
imine formation between amines and aldehydes,[22] ester for-
mation between boronic acids and alcohols,[23] and thermal
and catalytic aryl–aryl coupling reactions.[21,24] In some of
these reactions, specific 2D spatial arrangement of the pre-
cursor molecules is required to achieve covalently-linked 2D
architectures.
Scheme 1. Chemical structures of DBAs 1a–h, tetraiododibenzoindacene
derivative 2, hypothetical tetraradical intermediate 3, and plausible peri-
benzopolyacene 4 formed by lateral topochemical polymerization of the
DBAs.
et al. reported gel formation of DBA derivative 1h, but no
topochemically controlled polymerization was attempted.[28]
In the stacked geometry of DBAs in crystals, the butadiyne
units in a DBA molecule are expected to polymerize indi-
vidually to form poly(butadiyne) moieties in the resultant
tubular polymers.[29] Contrary to polymerization, Swager
et al. also reported the formation of tetraiododibenzoinda-
cene derivative 2 upon treatment of 1b with iodine through
transannular bond formation. Though the mechanism of this
reaction is not certain, one may speculate tetraradical inter-
mediate 3 formed by spontaneous transannular bond forma-
tion in 1b, similar to the formation of the 1,4-benzene dirad-
ical generated by a Bergman reaction.[30,31] Consequently, in
two-dimensionally aligned DBAs, such as 1, we hypothesize
a topochemical polymerization could occur involving both
butadiyne units in a molecule with transannular bond forma-
tion to yield peri-benzopolyacene 4 with a fluoranthene sub-
structure. Moreover, the dehydrogenation reaction (Scholl
type reaction)[32] of benzopolyacene 4 might convert it into a
more planarized, graphene-like ribbon.[33] Regarding this
connection, we became interested in structural control of
2D molecular networks of DBAs on a surface to investigate
their reactivity toward external stimuli. To this end, the
supramolecular networks of tetraalkoxy-substituted DBAs
1c–g at the 1,2,4-trichlorobenzene (TCB) or 1-phenyloctane/
Among the various molecular building blocks, we chose
octadehydrodibenzo[12]annulene derivatives (DBAs, 1c–g,
Scheme 1)[25] that possess two strained 1,3-butadiyne units,
as potential precursors of p-conjugated polymers. DBA de-
rivative 1a was previously shown by Swager et al. to under-
go oligomerization upon heating in the solid state, forming
structurally undefined oligomers.[26] Komatsu, Nishinaga
et al. succeeded in controlling crystal structures of related
DBA derivatives by utilizing stacking interactions of ben-
zene and heavily fluorinated benzene moieties, but no well-
defined polymers were obtained.[27] Recently, Miyata, Hisaki
8320
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8319 – 8328

An STM Investigation of Self-Assembled Monolayers
FULL PAPER
graphite interface were probed systematically using STM.
Three different structural motifs (porous, dense linear, and
lamella) were observed. We discuss here the factors which
determine 2D network structures of DBAs, highlighting the
alkyl chain length effect and concentration dependency.
Moreover, since we found that in the lamella structure the
intermolecular distance between two butadiyne units of
DBAs is relatively short compared to those in other struc-
tures, their reactivity toward irradiation by an electronic
pulse from an STM tip or UV irradiation was investigated.
Results and Discussion
Syntheses of DBAs: Syntheses of DBAs 1c–g were per-
formed by homocoupling reaction of dialkoxydiethynylben-
zenes under Hay conditions.[13c] Details are described in the
Supporting Information.
STM Observation of 2D molecular networks of DBAs at a
liquid/solid interface: All STM observations were performed
at the TCB or 1-phenyloctane/graphite interface at 20–258C.
All images were recorded within 3 h after dropping a solu-
tion of the DBA. Two different solute concentrations
(ꢀ10ꢁ4 m and ꢀ10ꢁ5 m) for each DBA were probed to evalu-
ate the concentration dependent change of the 2D molecu-
lar networks. Illustrated in Figure 1 are the three network
structures observed and their models optimized by using
molecular mechanics simulations (MM3 parameters).
Table 1 summarizes the experimental unit cell parameters of
the observed structures, as well as the network density
(dstructure), which was estimated by dividing the number of
molecules in a unit cell (Z) by the unit cell area (area).
The porous structure of 1c was imaged by dropping a
TCB or 1-phenyloctane (only at low concentration) solution
on the graphite surface (Figure 1a for a phenyloctane solu-
tion and Figure S1 in the Supporting Information for a TCB
solution). In the STM images, aromatic moieties are often
characterized by bright contrast, owing to high tunneling ef-
ficiency compared to other parts of the molecules.[34] Thus,
the bright features correspond to the p-conjugated part of
1c and the lines consisting of the small dots connecting the
bright features are ascribed to adsorbed alkyl chains. All
alkyl chains of 1c are adsorbed on the surface, two of them
align parallel to one of the three equivalent graphite main
axes (the <1,ꢁ2,1,0> directions). The porous structure con-
tains rectangular pores, the sizes of which are measured to
be 3.0 nm for corner-to-corner and 2.0 nm for edge-to-edge.
The pores are most likely occupied by mobile solvent mole-
cules, which were sometimes imaged as brighter contrast
(see Figure S2 in the Supporting Information). The unit cell
parameters in the two different solvents are the same (a=
2.8ꢂ0.1 nm, b =2.8ꢂ0.2 nm, and g=82ꢂ18 for phenyloc-
tane and a=2.8ꢂ0.1 nm, b=2.8ꢂ0.1 nm, and g=82ꢂ18
for TCB). The network density (dporous) of this structure is
low (0.129 molecules per nm2). A tentative network model
is shown in Figure 1b.
Figure 1. STM images of three types of 2D molecular networks of DBAs
and the corresponding network models obtained by molecular mechanics
simulations. a) The porous structure of 1c at the 1-phenyloctane/graphite
interface (Iscl =0.20 nA, Vbias =ꢁ0.15 V, 7.8ꢁ10ꢁ6 m). b) A tentative net-
work model of the porous structure. Periodic boundary condition (5ꢁ5
supercell) is a=5.563 nm, b=5.546 nm, and g=80.828. c) The linear
structure of 1d at the 1-phenyloctane/graphite interface (Iscl =0.26 nA,
V
bias =ꢁ0.21 V, saturated). d) A tentative network model of the linear
structure. Periodic boundary condition (3ꢁ2 supercell) is a=3.076 nm,
b=7.040 nm, and g=76.348. Two alkyl chains per molecule orienting to
the solution phase are omitted for clarity. e) The lamella structure of 1 f
at the 1-phenyloctane/graphite interface (Iscl =0.26 nA, Vbias =ꢁ0.21 V,
2.7ꢁ10ꢁ5 m). f) A tentative network model of the lamella structure. Peri-
odic boundary condition (3ꢁ1 supercell) is a=3.023 nm, b=6.040 nm,
and g=87.158. The pink colored arrows in the STM images point the di-
rection of the three equivalent graphite axes (the <1,ꢁ2,1,0> direc-
tions).
The dense linear structure (Figure 1c) was observed only
for a phenyloctane solution of 1d at a high concentration. In
a domain, all p cores of 1d adopt the same orientation
forming a molecular row. The rows were connected through
interdigitated alkyl chains. Two alkyl chains per molecule
are adsorbed on the surface aligning parallel to one of the
graphite axes (the <1,ꢁ2,1,0> directions), although the
other two chains were not observed on the surface, they are
most probably moving freely in the solution phase. A tenta-
tive network model of the linear structure is shown in Fig-
ure 1d. The unit cell parameters are a=1.0ꢂ0.1 nm, b=
Chem. Eur. J. 2010, 16, 8319 – 8328
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8321

Y. Tobe et al.
Table 1. Structural parameters of monolayers of DBAs 1c–g.[a]
Compound
Structure
Concentration
[ꢁ 10ꢁ4 m]
a [nm]
b [nm]
g [8]
Z[b]
m[c]
Azimuthal
angle q [8]
Area
[nm2]
Density
G
G
[molecules
per nm2][d]
porous
1.2
2.8ꢂ0.1
2.8ꢂ0.1
1.0ꢂ0.1
1.0ꢂ0.1
1.0ꢂ0.1
1.0ꢂ0.1
0.94ꢂ0.09
1.0ꢂ0.1
1.0ꢂ0.2
1.0ꢂ0.1
1.1ꢂ0.1
1.0ꢂ0.1
2.8ꢂ0.1
2.8ꢂ0.2
4.3ꢂ0.1
4.8ꢂ0.2
3.4ꢂ0.2
4.9ꢂ0.2
5.4ꢂ0.1
5.4ꢂ0.1
5.9ꢂ0.2
5.9ꢂ0.1
6.4ꢂ0.2
6.4ꢂ0.1
82ꢂ1
82ꢂ1
83ꢂ1
85ꢂ1
75ꢂ1
85ꢂ1
87ꢂ2
87ꢂ2
87ꢂ1
87ꢂ1
88ꢂ1
87ꢂ1
1
1
1
1
1
1
1
1
1
1
1
1
4
4
4
4
2
4
4
4
4
4
4
4
36
36
25
32
41
27
30
30
29
30
31
31
7.76
7.76
4.27
4.78
3.29
4.88
5.07
5.39
5.89
5.89
7.04
6.39
0.129
ACHTUNGTRENNUNG
A
porous
0.078
1c
1d
(phenyloctane)[e]
lamella
ACHTUNGTRENNUNG
0.15–1.0
0.14–4.1
saturated
0.39
(phenyloctane)
lamella
(TCB)
linear
(phenyloctane)
lamella
(phenyloctane)
lamella
(TCB)
lamella
(phenyloctane)
lamella
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
0.28–2.8
0.28–saturated
0.25–2.4
0.27–saturated
0.29–2.9
0.12–saturated
ACHTUNGTRENNUNG
1e
1 f
1g
ACHTUNGTRENNUNG
(TCB)
lamella
(phenyloctane)
lamella
ACHTUNGTRENNUNG
HACTUNGTRENNUNG
A
ACHTUNGTRENNUNG
lamella
(phenyloctane)
ACHTUNGTRENNUNG
[a] For all compounds, the same structures were observed in the different solvents. Structural parameters of networks observed in different solvents are
identical within experimental errors. [b] Number of molecules per unit cell. [c] Number of adsorbed alkyl chains per DBA molecule. [d] The value,
(number of adsorbed carbon and oxygen atoms)/(area per unit cell), is given in parentheses. [e] The lamella structure was also observed at this concen-
tration (surface coverage, porous/lamella=7:13).
3.4ꢂ0.2 nm, and g=75ꢂ18. Owing to the dense packing
mode, the unit cell area is small and the network density is
relatively high (dlinear =0.305 molecule per nm2).
trasts, as shown in Figure 2a. At the lower part in Figure 2a,
the relatively brighter lines (the blue line is overlaid) corre-
spond to the p cores of 1 f in the upper layer, whereas the
fuzzy lines (the red line is overlaid) are those in the lower
layer. Average distances between the lines of same contrast
are 5.8 nm (upper layer, l1) and 5.9 nm (lower layer, l2).
These values are similar to the unit cell vector b of the la-
mella structure of 1 f, indicating that both layers adopt the
lamella structure. In addition, though areas are quite small,
three different contrasts were observed in an image, indicat-
ing the formation of a third layer (the yellow circle in Fig-
ure 2b). The observed angles between the bright lines of dif-
ferent layers are mostly 0 or 608 (Figure 2b), but are rarely
908, suggesting specific interaction between the layers. To es-
timate the dependence of interlayer interaction on the inter-
layer angle, we carried out MM3 simulations for an alkane
bilayer on graphite, as a model system, by varying the inter-
layer angle (see the Supporting Information). Interestingly,
minimum potentials were located at the angle of 608 (and
1208) and the second minima with nearly equal potentials at
08 (1808) and 908. These results suggest that the multilayer
formation is driven by favorable alkyl–alkyl interactions be-
tween layers. A tentative model structure of a bilayer of la-
mella structure of 1 f with an interlayer angle of 608 is
shown in Figure 2c.
The lamella structure was observed for 1c (Figure S3) and
1d (Figure S4 and S5) under the conditions (i.e., solvents
and concentrations) other than those described above. On
the other hand, DBAs 1e–g with longer alkyl chains form
only the lamella structure under all conditions examined
(Figures S6–S10). A representative STM image and a struc-
tural model of the lamella structure are shown in Figures 1e
and f, respectively, for 1 f at the phenyloctane/graphite inter-
face. The DBA core of 1 f adopts the same orientation form-
ing a molecular row. All alkyl chains of 1 f align parallel to
one of the graphite axes (the <1,ꢁ2,1,0> directions) filling
the space between the rows of the DBA core. Unlike 1d,
the alkyl chains are not interdigitated, forming a 1D lamella
structure. A modeling study reveals two methylene units in
two of the alkyl chains most likely adopt an unfavorable
eclipsed conformation to achieve favorable intermolecular
van der Waals interactions. For the unit cell parameters of
the lamella structures of 1c–f, the vector a, which corre-
sponds to intermolecular distance within a row and angle g
are constant (ꢀ1.0 nm and from 838 to 878), whereas the
vector b which corresponds to the lamella distance becomes
larger upon elongation of the alkyl chains (Table 1).
Notably, we often observed the formation of multilayers
of the lamella structures for all compounds.[35] For instance,
in the case of 1 f, each layer was imaged with different con-
Alkyl chain length effect and concentration dependency: It
is well known that 2D molecular network structures of
8322
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8319 – 8328

An STM Investigation of Self-Assembled Monolayers
FULL PAPER
to exclusive formation of the lamella structure at all concen-
trations investigated. Although the reason for the different
solvent effect is difficult to elucidate, the differences in sol-
vation and solvent-substrate interactions most likely play a
major role. As for the formation of the porous structure of
1d, both solvents most likely stabilize this pattern by co-ad-
sorption in the pores (Figure S2).
The dependency of the network structures on alkyl chain
and solute concentration can be basically interpreted in
term of intermolecular and molecule-substrate interactions
per unit area.[17] The density (dstructure) of the three network
structures of a DBA with a given alkyl chain length follows
the order of porous<lamellaꢃlinear (Table 1). To avoid the
formation of large void space on the surface, the DBAs with
long alkyl chains are prone to form the dense linear and la-
mella structures because of large intermolecular and mole-
cule-substrate interactions per unit area. The concentration
dependency observed for 1c and 1d, which have relatively
short alkyl chains, in phenyloctane is in accord with this
idea. At high concentration the molecules tend to pack
more densely, whereas the structure with low molecular den-
sity is favored upon dilution, because the difference of inter-
molecular and molecule–substrate interactions per unit area
between the nonporous and porous structures is not so
large. In other words, the molecules tend to cover the sur-
face as widely as possible upon dilution.[17a,f] Moreover, the
unique formation of the porous structure of 1c can be at-
tributed to stabilization, owing to co-adsorption of solvent
(vide supra) and/or epitaxial effect (vide infra).[37]
Figure 2. a) An STM image of the double layered lamella structure of 1 f
at the TCB/graphite interface (Iscl =0.22 nA, Vbias =ꢁ0.30 V, 2.5ꢁ10ꢁ4 m).
The blue and red lines are overlaid on the upper and lower layers, re-
spectively. Values l1 and l2 correspond to the distances between the mo-
lecular rows with same contrast. b) An STM image of the multilayered
lamella structure of 1g at the TCB/graphite interface (Iscl =0.16 nA,
V
bias =ꢁ1.69 V, 5.6ꢁ10ꢁ5 m). The yellow circle indicates a triple layer.
c) Tentative model of the bilayered lamella structure (interlayer angle is
608) of 1 f on graphite. The white-colored molecules form the underlying
lamella structure.
For 1d, the unique appearance of the linear structure can
be ascribed to epitaxial registry matching between the over-
layer and the graphite lattice. Indeed, in the STM images of
the linear structure, an apparent Moirꢂ pattern of the alkyl
chains along the unit cell vector a was observed (see Fig-
ure S11 in the Supporting Information), indicating the exis-
tence of long-range ordering (beyond the unit cell).[38,39] To
analyze possible lattice registry, an EpiSearch subroutine of
the EpiCalc program was performed.[40] The program calcu-
lates a V/V0 value, which is a dimensionless potential for
each azimuthal angle, showing the degree of commensurism
between the overlayer and the substrate layer (V/V0 =1 for
incommensurism, V/V0 =0.5 for point-on-line coincidence,
and V/V0 =0 for commensurism on nonhexagonal substrate).
Perfect commensurism, “point-on-point” coincidence, with
graphite lattice was found for the linear structure by assum-
ing a 6ꢁ8 supercell (V/V0 =0.01, q=40.88, b1 =6.510 nm,
b2 =25.920 nm, b=74.408). The Moirꢂ pattern observed ex-
perimentally (Figure S11), however, does not fit the above
results, it indicates the existence of periodicity by four unit
cells each along the unit-cell vector a. To assess the epitaxial
effect for the porous structure of 1c the same analysis was
undertaken. As a result, it also revealed perfect commensur-
ism (4ꢁ6 supercell, V/V0 =0.00, q=36.59, b1 =10.720 nm,
b2 =17.220 nm, b=83.408). Conversely, the lamella struc-
tures of all DBAs 1c–g did not exhibit perfect commensur-
ism hits even assuming supercells (up to 8ꢁ8). These results
indicate that the unique formation of the linear structure of
alkyl-substituted molecules are often affected by the alkyl-
chain length because of the shift of balance between inter-
molecular and molecule–surface van der Waals interac-
tions.[36] In addition, in some cases, the network structures
are concentration-dependent too and thus different struc-
tures emerge upon the modulation of solute concentra-
tion.[17] For example, the molecular networks of alkoxy-sub-
stituted triangular dehydrobenzo[12]annulene derivatives
with short alkyl chains (C10–C12) tend to form porous honey-
comb patterns, whereas those with long chains (C14–C20) ex-
hibit dense nonporous linear structures.[17a] However, the
porous honeycomb structures become dominant upon dilu-
tion of the solution even in the case of those DBAs with
long alkyl chains.
A similar trend was observed in the present system. The
structural change is relatively simple if TCB is used as a sol-
vent. DBA 1c, with the shortest alkyl chain, forms the
porous structure, whereas DBAs 1d–g form lamella struc-
tures at all concentrations examined. On the other hand, the
concentration dependency of network structure was more
variable in phenyloctane; the lamella structure of 1c was ob-
served at higher concentration, whereas the porous structure
appears upon dilution. In the case of 1d, the dense linear
structure was observed at high concentration (Figures 1c
and d), whereas the lamella structure formed at low concen-
tration. Further elongation of the alkyl chains, however, led
Chem. Eur. J. 2010, 16, 8319 – 8328
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8323

Y. Tobe et al.
1d and the porous structure of 1c is attributed, at least par-
tially, to the epitaxial stabilization.
Attempted 2D topochemical polymerization: Topochemical
1,4-polymerization of diacetylenic compounds induced by
heat and photoirradiation to produce poly(butadiyne) deriv-
atives in the crystalline state has been well studied.[41]
Design and control of the arrangement of butadiyne units in
crystals that utilize weak intermolecular interactions have
proven successful to achieve topochemically controlled reac-
tions.[42] On the basis of the 3D crystal structures of numer-
ous diacetylene monomers and those of the resulting poly-
mers, it was found that maximal reactivity toward polymeri-
zation was observed if the distance between two adjacent di-
acetylene moieties is shorter than 5.0 ꢃ and the tilt angle
between the axis of the array and the diacetylene rod is 458.
There are also examples of two-dimensionally confined
polymerization of diacetylenic compounds induced by exter-
nal stimuli. For example, the reaction of diacetylene-con-
taining isopthalic-acid derivative at the liquid/graphite inter-
face upon UV irradiation has been reported by De Schryver,
De Feyter et al.[20a,e] Aono et al. demonstrated polymeri-
zation of carboxy-substituted butadiyne derivatives on a
graphite surface triggered by irradiation of electronic pulse
from an STM tip.[20b] In some of these reports, the structural
parameters of butadiyne units on the graphite surface were
also discussed. The intermolecular distances were found to
be shorter than 5.0 ꢃ and the tilt angles were approximately
508, indicating the geometrical requirement for topochemi-
cal polymerization on the surfaces is similar to those in 3D
crystals.[20]
Figure 3. a) Chemical structures of rubicene 5, a indenorubicene deriva-
tive 6, and a model compound 7 consisting of five fluoranthene units.
b) Structure of 7 optimized by B3LYP/6-31G(d) level of theory under C2
symmetric constraint.
As described in the introduction, attempted topochemical
polymerization of DBA derivatives in the crystalline state
has not been successful.[27,29] On the other hand, in view of
the high reactivity of DBA 1b toward a transannular reac-
tion,[26] we envisioned that, in two-dimensionally aligned
DBAs, such as 1, topochemical polymerization would occur
involving both butadiyne units in a molecule with transannu-
lar bond formation to yield peri-benzopolyacene 4 (see
Scheme 1) with a fluoranthene substructure. A structural
subunit of this polymer, rubicene (5), has been known for
over 100 years (Figure 3a).[43] The largest known fragment
of 4, to our knowledge, is indenorubicene derivative 6
formed by a domino radical cycloaromatization of an arylal-
kyne precursor having three butadiyne units.[44] To assess the
feasibility of an even larger structure like polymer 4, struc-
ture optimization of a model compound 7 consisting of five
fluoranthene units was undertaken by using DFT methods
at the B3LYP/6-31G(d) level of theory (Figure 3b). As
shown in Figure 3b, each peri-fused indene units winds up
and down from the hexacene backbone because of the over-
crowding commonly observed in helicenes. Indeed, the dihe-
dral angles of the bonds forming the fiord region of 7 range
from 14.9 to 22.78 for the inner aromatic rings and 4.9 to
9.08 for the five-membered rings. These torsion angle are
smaller than those observed in [6]helicenes (see Figure S15
in the Supporting Information).[45] The edge-to-edge twist
angle of the hexacene backbone of 7 is 49.78, which is larger
than those of benzannelated pentacenes (41–438),[46] but is
much smaller than those observed for the Pascalꢄs twisted
acenes.[47] These results suggest that peri-benzopolyacene 4
is a realistic target at least in terms of structural and thermo-
dynamic aspects. Oligomeric short fragments may well exist.
Molecular modeling of the lamella structures of all DBAs
by molecular mechanics simulations reveal that the closest
distance between C1 of the butadiyne units of neighboring
DBAs should be 5.9–6.1 ꢃ (Figure 4a). For the dense linear
structure of 1d, the corresponding distance is longer
(ꢀ6.6 ꢃ, Figure 4b). Clearly the distance between the buta-
diyne units in the porous structure of 1c is much longer.
Therefore, we investigated possibility of reaction of DBAs
in the lamella and linear structures toward external stimuli.
We attempted first in-situ irradiation of electronic pulse
from an STM tip (2.0–4.0 V) to the lamella structures of 1c
and 1g at the 1-phenyloctane/graphite interface.[48] However,
we did not observe any change of the monolayer structures
of the DBAs (see the Supporting Information). Second, we
investigated the effect of UV irradiation to the lamella
structures of DBAs 1c, 1 f, and 1g and the linear structure
of 1d under air (i.e., by using a dry film obtained by slow
8324
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8319 – 8328

An STM Investigation of Self-Assembled Monolayers
FULL PAPER
The reactivity of DBAs in the lamella or linear structures
by external stimuli (electronic pulse from an STM tip and
UV irradiation) was investigated aiming at 2D topochemical
polymerization to afford a novel polymer, peri-benzopoly-
AHCTUNGTREGaNNUN cene 4. The structural feasibility of 4 was assessed based on
the DFT calculations for the model compound 7 having five
fluoranthene units. Even though each fluoranthene unit of 7
is highly deformed from planarity, such deformation is not
surprising in view of the presence of helicene-like substruc-
tures. These results suggest that polymer 4, at least for its
small fragments, is not unrealistic in terms of its structural
aspect. However, no evidence for the proposed polymeri-
zation was obtained. This is probably owed to unfavorable
geometries, that is, long diacetylene–diacetylene distances,
and large structural change imposed on planer DBAs to
transform into nonplanar oligomers. The present results in-
dicate the necessity for further changes to design a suitable
system for the on-surface construction of structurally novel
conjugated polymers, which are otherwise difficult to pre-
pare.
Figure 4. The network models for the a) lamella and the b) linear struc-
ꢁ
tures showing the closest intermolecular C C distance between the diace-
tylene units.
evaporation of the solvent, see Figure S12 in the Supporting
Information), as well as the lamella structure of 1g at the 1-
phenyloctane/graphite interface. Again, no changes of the
monolayer structures were observed by subsequent STM in-
vestigations (see the Supporting Information). Thus, we
were unable to detect any evidence for 2D topochemical oli-
gomerization/ploymerization of the DBAs. Plausible reasons
for the inertness of the DBAs are unfavorable geometries
for the reaction, that is, long diacetylene–diacetylene dis-
tance (the distance between C1 and C4 of the butadiyne
units of neighboring DBAs in the lamella structure is
ꢀ6 ꢃ), and the large structural change imposed on planar
DBAs to transform into nonplanar oligomers.
Experimental Section
Syntheses of DBAs: Syntheses of DBAs 1c–g were performed by homo-
coupling reaction of dialkoxydiethynylbenzenes under Hay conditions.[13c]
Details are described in the Supporting Information.
STM investigation: All experiments were performed at 20–258C by using
a Nanoscope IIIa (Digital Instruments Inc.) with an external pulse/func-
tion generator (model HP 8111a or Agilent 33220A) with a negative
sample bias. The STM image in Figure 1b was acquired in the constant
current mode, and the other images were acquired in the variable current
mode. Tips were electrochemically etched in a 2m KOH/6m KCN solu-
tion in water or mechanically cut from Pt/Ir wire (80%/20%, diameter
0.2 mm). Prior to imaging, a compound under investigation was dissolved
in commercially available anhydrous 1,2,4-trichlorobenzene (TCB, Al-
Conclusion
We have carried out STM investigations on self-assembled
monolayers of tetraalkoxy-substituted DBAs 1c–g at the
TCB or phenyloctane/graphite interface. Three types of
structures, porous, linear, and lamella structures, were
formed and the appearance of which was changed by alter-
ing the alkyl chain length, concentration, and the choice of
solvent. DBA 1c, which has the shortest alkyl chain, favors
formation of the porous structure in both solvents, whereas
for phenyloctane, the dense lamella structure appears at a
high solute concentration. DBA 1d adopts the linear pattern
at high solute concentration in phenyloctane, whereas it
transforms to the lamella pattern upon dilution. The lamella
structure of 1d was also observed in TCB. The other DBA
derivatives 1e–g form only the dense lamella structure at all
conditions examined. The selection in the networks is basi-
cally interpreted in terms of intermolecular and molecule–
substrate interactions per unit area and network density.
The unique appearance of the linear structure of 1d is at-
tributed to a favorable epitaxial registry matching between
the substrate lattice and overlayer lattice. Similarly, the for-
mation of porous structure of 1c is owed, partly, to the epi-
taxial effect and solvent co-adsorption. The multilayer for-
mation of the lamella structures was observed and was at-
tributed to favorable interlayer alkyl–alkyl van der Waals in-
teractions.
drich) or 1-phenyloctane (TCI) at two different concentrations (ꢀ10ꢁ4
m
and 10ꢁ5 m), and a drop of this solution was applied on a freshly cleaved
surface of HOPG (grade ZYB, Momentive Performance Material Quartz
Inc., Strongsville, OH). The STM investigations were then performed at
the liquid/solid interface. By changing the tunneling parameters during
the STM imaging, namely, the voltage applied to the substrate and the
average tunneling current, it was possible to switch from the visualization
of the adsorbate layer to that of the underlying HOPG substrate. This en-
abled us to correct for drift effects by the use of SPIP software (Image
Metrology A/S). The unit cell parameters were determined from more
than 47 experimental data of at least two calibrated STM images.
Molecular mechanics simulations: Molecular mechanics simulations were
performed by using the Tinker package and the MM3 force field,[49]
which has been recently reparameterized to take into account weak non-
bonding interactions, such as p–p stacking, CH–p interactions, and hy-
drogen bonds.[50]
Each starting geometry was built from a molecular model, which was ob-
tained from optimization by the PM3 method under D2h symmetric con-
straints. Then, the orientation of the p system and alkyl chains was deter-
mined from the STM image, and the conformation of some alkyl chains
ꢁ
was adjusted by rotating only a minimum number of C C bonds to fit
the experimentally obtained unit cell parameters. The network models
were placed at 0.35 nm above the first layer of a periodic two-layer sheet
of graphite (interlayer distance of graphite is also 0.35 nm) and adjusted
to make the alkyl chains align parallel to the directions of graphite sym-
metry axes (the <1,ꢁ2,1,0> directions). Periodic boundary conditions
were employed to fit the experimental unit cell or sometimes the super-
Chem. Eur. J. 2010, 16, 8319 – 8328
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8325

Y. Tobe et al.
lattice (a few unit cells). The graphite structure was frozen during the
simulation, and a cutoff of 2.0 nm was applied form the van der Waals in-
teractions (Lennard–Jones type).
Surin, P. Samorꢆ, Small 2007, 3, 190–194; l) K. E. Plass, A. L. Grze-
siak, A. J. Matzger, Acc. Chem. Res. 2007, 40, 287–293;
m) J. A. A. W. Elemans, S. Lei, S. De Feyter, Angew. Chem. 2009,
121, 7434–7469; Angew. Chem. Int. Ed. 2009, 48, 7298–7332.
[2] a) C. Joachim, J. K. Gimzewski, A. Aviram, Nature 2000, 408, 541–
548; b) A. Nitzan, M. A. Ratner, Science 2003, 300, 1384–1389.
[3] J. R. Sheats, B. W. Smith, Microlithographic Science and Technology,
1st ed., Marcel Dekker, New York, 1999.
[4] a) G. M. J. Schmidt, Pure Appl. Chem. 1971, 27, 647–678; b) Crystal
Engineering: The Design of Organic Solid (Ed.: G. R. Desiraju),
Elsevier, Amsterdam, 1989; c) G. R. Desiraju, Angew. Chem. 1995,
107, 2541–2558; Angew. Chem. Int. Ed. Engl. 1995, 34, 2311–2327;
d) N. Blagden, R. J. Davey, Cryst. Growth Des. 2003, 3, 873–885.
[5] S. Furukawa, S. De Feyter, Top. Curr. Chem. 2009, 287, 87–133.
[6] There are a number of reports about the formation of 2D molecular
networks through hydrogen bonding as connectivity. See, for exam-
ple: a) S. Griessl, M. Lackinger, M. Edelwirth, M. Hietschold, W. M.
Heckl, Single Mol. 2002, 3, 25–31; b) S. J. H. Griessl, M. Lackinger,
F. Jamitzky, T. Markert, M. Hietschold, W. M. Heckl, Langmuir
2004, 20, 9403–9407; c) S. B. Lei, C. Wang, S. X. Yin, H. N. Wang, F.
Xi, H. W. Liu, B. Xu, L. J. Wan, C. L. Bai, J. Phys. Chem. B 2001,
105, 10838–10841; d) J. Otsuki, E. Nagamine, T. Kondo, K. Iwasaki,
M. Asakawa, K. Miyake, J. Am. Chem. Soc. 2005, 127, 10400–
10405; e) W. Mamdouh, R. E. A. Kelly, M. Dong, L. N. Kantorovich,
F. Besenbacher, J. Am. Chem. Soc. 2008, 130, 695–702.
[7] a) T. Yokoyama, S. Yokoyama, T. Kamikado, Y. Okuno, S. Mashiko,
Nature 2001, 413, 619–621; b) M. de Wild, S. Berner, H. Suzuki, H.
Yanagi, D. Schlettwein, S. Ivan, A. Baratoff, H.-J. Guentherodt,
T. A. Jung, ChemPhysChem 2002, 3, 881–885; c) Y. Wei, W. Tong,
C. Wise, X. Wei, K. Armbrust, M. Zimmt, J. Am. Chem. Soc. 2006,
128, 13362–13363; d) Y. Wei, W. Tong, M. B. Zimmt, J. Am. Chem.
Soc. 2008, 130, 3399–3405.
[8] a) M. A. Lingenfelder, H. Spillmann, A. Dmitriev, S. Stepanow, N.
Lin, J. V. Barth, K. Kern, Chem. Eur. J. 2004, 10, 1913–1919; b) S.
Stepanow, N. Lin, D. Payer, U. Schilckum, F. Klappenberger, G.
Zoppellaro, M. Ruben, H. Brune, J. V. Barth, K. Kern, Angew.
Chem. 2007, 119, 724–727; Angew. Chem. Int. Ed. 2007, 46, 710–
713; c) M. Surin, P. Samorꢆ, A. Jouaiti, N. Kyritsakas, M. W. Hossei-
ni, Angew. Chem. 2007, 119, 249–253; Angew. Chem. Int. Ed. 2007,
46, 245–249.
Lattice registry analysis: An epitaxial interface can be described by seven
parameters, an overlayer lattice (b1, b2, and b) and a substrate lattice (a1,
a2, and a), and azimuthal angles (q), between the lattice vectors a1 and
b1. The substrate and overlayer lattice vectors for a given azimuthal ori-
entation q are related through the transformation matrix [C] in Equa-
tion (1) in which the matrix coefficients (p, q, r, and s) are defined by
Equations (2)–(5).[11]
"
#
"
#
"
# "
#
b1
b2
a1
a2
p q
r s
a1
a2
¼ ½Cꢄ
¼
ð1Þ
p ¼ b1sinðaꢁqÞ=a1sinðaÞ
q ¼ b1sinðqÞ=a2sinðaÞ
ð2Þ
ð3Þ
ð4Þ
ð5Þ
r ¼ b2sinðaꢁqꢁbÞ=a1sinðaÞ
s ¼ b2sinðq þ bÞ=a2sinðaÞ
The analyses of registry between the molecular layer and substrate layer
were performed by using EpiCalc program developed by Ward and co-
workers.[40] EpiCalc determines the lattice registry by rotating an overlay-
er lattice (b1, b2, and b) on a substrate lattice (a1, a2, and a) through a
series of azimuthal angles (q). In the present study, following values were
employed as the unit cell parameters of the substrate (graphite; a1 =a2 =
0.246 nm and a =608). The program calculates V/V0 value which is di-
mensionless potential for each azimuthal angle, showing the degree of
commensurism between overlayer and substrate layer (V/V0 =1 for in-
commensurism, V/V0 =0.5 for point-on-line coincidence, and V/V0 =0 for
commensurism on nonhexagonal substrate). All analyses were performed
within the experimental errors of unit cell parameters by increment of
0.01 nm for unit cell vectors and 0.18 for unit cell angle. The variation of
azimuthal angle (Dq) was 0.018. Overlayer size was set to 25ꢁ25. A su-
percell was assumed up to 8ꢁ8. Threshold value of V/V0 was set 0.1 to
find “point-to-point (POP) coincident” as well as to exclude the numer-
ous “point-on-line (POL) coincident” hits.
DFT calculation: All theoretical calculations were performed with the
Gaussian 03 package.[51] B3LYP calculations with the 6-31G(d) bases sets
were used for the geometry optimization of model compound 7. Com-
pound 7 adapts C2 symmetric structure at the energy minimum, and the
structures without symmetric constraint (C1 symmetry) has slightly larger
energy (1.00 kcalmolꢁ1). The nature of the stationary points was assessed
by means of vibration frequency analysis.
[9] a) X. Qiu, C. Wang, Q. Zeng, B. Xu, S. Yin, H. Wang, S. Xu, C. Bai,
J. Am. Chem. Soc. 2000, 122, 5550–5556; b) D. Blꢂger, D. Kreher, F.
Mathevet, A.-J. Attias, G. Schull, A. Huard, L. Douillard, C. Fiorini-
Debuischert, F. Charra, Angew. Chem. 2007, 119, 7548–7551;
Angew. Chem. Int. Ed. 2007, 46, 7404–7407; c) D. Blꢂger, D.
Kreher, F. Mathevet, A.-J. Attias, I. Arfaoui, G. Metgꢂ, L. Douil-
lard, C. Fiorini-Debuisschert, F. Charra, Angew. Chem. 2008, 120,
8540–8543; Angew. Chem. Int. Ed. 2008, 47, 8412–8415.
[10] a) S. Furukawa, H. Uji-i, K. Tahara, T. Ichikawa, M. Sonoda, F. C.
De Schryver, Y. Tobe, S. De Feyter, J. Am. Chem. Soc. 2006, 128,
3502–3503; b) K. Tahara, S. Furukawa, H. Uji-i, T. Uchino, T. Ichi-
kawa, J. Zhang, W. Mamdouh, M. Sonoda, F. C. De Schryver, S. De
Feyter, Y. Tobe, J. Am. Chem. Soc. 2006, 128, 16613–16625.
[11] a) A. C. Hillier, M. D. Ward, Phys. Rev. B 1996, 54, 14037–14051;
b) D. E. Hooks, T. Fritz, M. D. Ward, Adv. Mater. 2001, 13, 227–241.
[12] Y. Yang, C. Wang, Curr. Opin. Colloid Interface Sci. 2009, 14, 135–
147.
Acknowledgements
This work was supported by a Grant-in-Aid for Scientific Research from
the Ministry of Education, Culture, Sports, Science, and Technology,
Japan, and the US National Science Foundation. C.A.J. acknowledges the
NSF-IGERT program for a fellowship and for a travel grant to Japan.
[13] a) W. Mamdouh, H. Uji-i, J. S. Ladislaw, A. E. Dulcey, V. Percec,
F. C. De Schryver, S. De Feyter, J. Am. Chem. Soc. 2006, 128, 317–
325; b) K. G. Nath, O. Ivasenko, J. A. Miwa, H. Dang, J. D. Wuest,
A. Nanci, D. F. Perepichka, F. Rosei, J. Am. Chem. Soc. 2006, 128,
4212–4213; c) K. Tahara, C. A. Johnson II, T, Fujita, M. Sonoda,
F. C. De Schryver, S. De Feyter, M. M. Haley, Y. Tobe, Langmuir
2007, 23, 10190–10197.
[14] a) B. Venkataraman, J. J. Breen, G. W. Flynn, J. Phys. Chem. 1995,
99, 6608–6619; b) C.-J. Li, Q.-D. Zeng, C. Wang, L.-J. Wan, S.-L.
Xu, C.-R. Wang, C.-L. Bai, J. Phys. Chem. B 2003, 107, 747–750;
c) M. Lackinger, S. Griessl, W. M. Heckl, M. Hietschold, G. W.
Flynn, Langmuir 2005, 21, 4984–4988; d) L. Kampschulte, M. Lack-
[1] a) J. Frommer, Angew. Chem. 1992, 104, 1325–1357; Angew. Chem.
Int. Ed. Engl. 1992, 31, 1298–1328; b) D. M. Walba, F. Stevens,
N. A. Clark, D. C. Parks, Acc. Chem. Res. 1996, 29, 591–597; c) J. V.
Barth, Annu. Rev. Phys. Chem. 2007, 58, 375–407; d) J. A. Theobald,
N. S. Oxtoby, M. A. Phillips, N. R. Champness, P. H. Beton, Nature
2003, 424, 1029–1031; e) L. C. Giancarlo, G. W. Flynn, Acc. Chem.
Res. 2000, 33, 491–501; f) P. Samorꢅ, J. P. Rabe, J. Phys. Condens.
Matter 2002, 14, 9955–9973; g) S. De Feyter, F. C. De Schryver, J.
Phys. Chem. B 2005, 109, 4290–4302; h) J. V. Barth, G. Costantini,
K. Kern, Nature 2005, 437, 671–679; i) L.-J. Wan, Acc. Chem. Res.
2006, 39, 334–342; j) B. A. Hermann, L. J. Scherer, C. E. House-
croft, E. C. Constable, Adv. Funct. Mater. 2006, 16, 221–235; k) M.
8326
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8319 – 8328

An STM Investigation of Self-Assembled Monolayers
FULL PAPER
inger, A.-K. Maier, R. S. K. Kishore, S. Griessl, M. Schmittel, W. M.
Heckl, J. Phys. Chem. B 2006, 110, 10829–10836; e) X. Shao, X.
Luo, X. Hu, K. Wu, J. Phys. Chem. B 2006, 110, 1288–1293; f) R.
Gutzler, S. Lappe, K. Mahata, M. Schmittel, W. M. Heckl, M. Lack-
inger, Chem. Commun. 2009, 680–682.
[28] I. Hisaki, H. Shigemitsu, Y. Sakamoto, Y. Hasegawa, Y. Okajima, K.
Nakano, N. Tohnai, M. Miyata, Angew. Chem. 2009, 121, 5573–
5577; Angew. Chem. Int. Ed. 2009, 48, 5465–5469.
[29] In the case of octadehydrotribenzo[14]annulene, however, topo-
chemical polymerization did take place, which is due to the suitable
packing for polymerization in crystals. K. P. Baldwin, A. J. Matzger,
D. A. Scheiman, C. A. Tessier, K. P. C. Vollhardt, W. J. Youngs, Syn-
lett 1995, 1215–1218.
[30] For Bergman reactions, see: a) R. R. Jones, R. G. Bergman, J. Am.
Chem. Soc. 1972, 94, 660–661; b) R. G. Bergman, Acc. Chem. Res.
1973, 6, 25–31; c) T. P. Lockhart, P. B. Comita, R. G. Bergman, J.
Am. Chem. Soc. 1981, 103, 4082–4090.
[15] a) A. Stabel, R. Heinz, F. C. De Schryver, J. P. Rabe, J. Phys. Chem.
1995, 99, 505–507; b) L. Piot, A. Marchenko, J. Wu, K. Mꢇllen, D.
Fichou, J. Am. Chem. Soc. 2005, 127, 16245–16250; c) G. M. Florio,
J. E. Klare, M. O. Pasamba, T. L. Werblowsky, M. Hyers, B. J. Berne,
M. S. Hybertsen, C. Nuckolls, G. W. Flynn, Langmuir 2006, 22,
10003–10008.
[16] a) D. Rohde, C.-J. Yan, H.-J. Yan, L.-J. Wan, Angew. Chem. 2006,
118, 4100–4104; Angew. Chem. Int. Ed. 2006, 45, 3996–4000; b) C.-
J. Li, Q.-D. Zeng, Y.-H. Liu, L.-J. Wan, C. Wang, C.-R. Wang, C.-L.
Bai, ChemPhysChem 2003, 4, 857–859; c) R. T. Baker, J. D. Mou-
gous, A. Brackley, D. L. Patrick, Langmuir 1999, 15, 4884–4891.
[17] a) S. Lei, K. Tahara, F. C. De Schryver, M. Van der Auweraer, Y.
Tobe, S. De Feyter, Angew. Chem. 2008, 120, 3006–3010; Angew.
Chem. Int. Ed. 2008, 47, 2964–2968; b) L. Kampschulte, T. L. Wer-
blowsky, R. S. K. Kishore, M. Schmittel, W. M. Heckl, M. Lackinger,
J. Am. Chem. Soc. 2008, 130, 8502–8507; c) C.-A. Palma, M. Bonini,
A. Llanes-Pallas, T. Breiner, M. Prato, D. Bonifazi, P. Samorꢅ,
Chem. Commun. 2008, 5289–5291; d) C.-A. Palma, J. Bjork, M.
Bonini, M. S. Dyer, A. Llanes-Pallas, D. Bonifazi, M. Persson, P.
Samorꢅ, J. Am. Chem. Soc. 2009, 131, 13062–13071; e) C.-A. Palma,
M. Bonini, T. Breiner, P. Samorꢆ, Adv. Mater. 2009, 21, 1383–1386;
f) K. Tahara, S. Okuhata, J. Adisoejoso, S. Lei, T. Fujita, S. De Feyt-
er, Y. Tobe, J. Am. Chem. Soc. 2009, 131, 17583–17590.
[18] J. Sakamoto, J. van Heijst, O. Lukin, A. D. Schlꢇter, Angew. Chem.
2009, 121, 1048–1089; Angew. Chem. Int. Ed. 2009, 48, 1030–1069.
[19] a) H. Ozaki, T. Funaki, Y. Mazaki, S. Masuda, Y. Harada, J. Am.
Chem. Soc. 1995, 117, 5596–5597; b) O. Endo, H. Ootsubo, N. Toda,
M. Suhara, H. Ozaki, Y. Mazaki, J. Am. Chem. Soc. 2004, 126,
9894–9895.
[20] a) P. C. M. Grim, S. De Feyter, A. Gesquiꢈre, P. Vanoppen, M.
Rꢇcker, S. Valiyaveettil, G. Moessner, K. Mꢇllen, F. C. De Schryver,
Angew. Chem. 1997, 109, 2713–2715; Angew. Chem. Int. Ed. Engl.
1997, 36, 2601–2603; b) Y. Okawa, M. Aono, Nature 2001, 409, 683–
684; c) Y. Okawa, M. Aono, J. Chem. Phys. 2001, 115, 2317–2322;
d) Y.-H. Qiao, Q.-D. Zeng, Z.-Y. Tan, S.-D. Xu, D. Wang, C. Wang,
L.-J. Wan, C.-L. Bai, J. Vac. Sci. Technol. B 2002, 20, 2466–2469;
e) A. Miura, S. De Feyter, M. M. S. Abdel-Mottaleb, A. Gesquiꢈre,
P. C. M. Grim, G. Moessner, M. Sieffert, M. Klapper, K. Mꢇllen,
F. C. De Schryver, Langmuir 2003, 19, 6474–6482; f) S. Nishio, D. I.-
i. H. Matsuda, M. Yoshidome, H. Uji-i, H. Fukumura, Jpn. J. Appl.
Phys. 2005, 44, 5417–5420.
[31] The tetraradical intermediate 3 apparently seems much less favora-
ble than 1,4-benzene diradical. However, a very rough estimate of
bond dissociation energies indicates that the conversion of DBA 1b
into tetraradical 3 would be endoergic by about 5–6 kcalmolꢁ1
.
Namely, the bond dissociation energy (56.6 kcalmolꢁ1) of one of p
bonds of a triple bond is estimated from the difference between
those of ethyne (230.7 kcalmolꢁ1) and ethene (174.1 kcalmolꢁ1).
(Four time of this energy) minus (twice of the dissociation energy
(116 kcalmolꢁ1) of a vinyl–vinyl bond) yields 5.6 kcalmolꢁ1. The
thermochemical data are taken from the following review: S. J.
Blanksby, G. B. Ellison, Acc. Chem. Res. 2003, 36, 255–263.
[32] a) R. Scholl, J. Mansfeld, Ber. Dtsch. Chem. Ges. 1910, 43, 1734–
1746; b) M. Mꢇller, C. Kꢇbel, K. Mꢇllen, Chem. Eur. J. 1998, 4,
2099–2109; c) P. Rempala, J. Kroulꢅk, B. T. King, J. Am. Chem. Soc.
2004, 126, 15002–15003.
[33] a) J. Wu, L. Gherghel, M. D. Watson, J. Li, Z. Wang, C. D. Simpson,
U. Kolb, K. Mꢇllen, Macromolecules 2003, 36, 7082–7089; b) X.
Yang, X. Dou, A. Rouhanipour, L. Zhi, H. J. Rꢊder, K. Mꢇllen, J.
Am. Chem. Soc. 2008, 130, 4216–4217.
[34] R. Lazzaroni, A. Calderone, G. Lambin, J. P. Rabe, J. L. Brꢂdas,
Synth. Met. 1991, 41, 525–528.
[35] There are a number of reports about the formation of mutilayer.
See, for example: a) F. Jꢊckel, M. Ai, J. Wu, K. Mꢇllen, J. P. Rabe, J.
Am. Chem. Soc. 2005, 127, 14580–14581; b) S. Lei, J. Puigmartꢅ-
Luis, A. Minoia, M. Van der Auweraer, C. Rovira, R. Lazzaroni,
D. B. Amabilino, S. De Feyter, Chem. Commun. 2008, 703–705;
c) D. Takajo, E. Fujiwara, S. Irie, T. Nemoto, S. Isoda, H. Ozaki, N.
Toda, S. Tomii, T. Magara, Y. Mazaki, G. Yamamoto, J. Cryst.
Growth 2002, 237–239, 2071–2075.
[36] a) P. Wu, Q. Zeng, S. Xu, C. Wang, S. Yin, C. L. Bai, ChemPhys-
Chem 2001, 2, 750–754; b) F. Charra, J. Cousty, Phys. Rev. Lett.
1998, 80, 1682–1685; c) L. Askadskaya, C. Boeffel, J. P. Rabe, Ber.
Bunsen-Ges. 1993, 97, 517–521; d) S. Ito, M. Wehmeier, J. D. Brand,
C. Kꢇbel, R. Epsch, J. P. Rabe, K. Mꢇllen, Chem. Eur. J. 2000, 6,
4327–4342; e) E. Mena-Osteritz, Adv. Mater. 2002, 14, 609–616.
[37] The size of each domain in the monolayer of DBAs also depends on
a solute concentration. At higher concentration, DBAs tend to form
many small domains, whereas the size of domain increases upon di-
lution. For example, the average domain size of the lamella struc-
ture formed by dropping a phenyloctane solution of 1e at a concen-
tration of 2.8ꢁ10ꢁ4 m is about 1900 nm2 (calculated from six large
area images (200ꢁ200 nm)), whereas upon dilution the size of
domain becomes over typical scanning size (150 ꢁ150 nm,
22500 nm2).
[38] The Moirꢂ pattern of monolayers of alkane or alkylated molecules
were often observed. See, for example: a) K. E. Plass, K. Kim, A. J.
Matzger, J. Am. Chem. Soc. 2004, 126, 9042–9053; b) S.-B. Lei, C.
Wang, X.-L. Fan, L.-J. Wan, C.-L. Bai, Langmuir 2003, 19, 9759–
9763; c) K. E. Plass, A. J. Matzger, Chem. Commun. 2006, 3486–
3488; d) A. Stabel, R. Heinz, J. P. Rabe, G. Wegner, F. C. De Schryv-
er, D. Corens, W. Dehaen, C. Sꢇling, J. Phys. Chem. 1995, 99, 8690–
8697; e) S. De Feyter, A. Gesquiꢈre, P. C. M. Grim, F. C. De Schryv-
er, S. Valiyaveettil, C. Meiners, M. Sieffert, K. Mꢇllen, Langmuir
1999, 15, 2817–2822; f) D. M. Cyr, B. Venkataraman, G. W. Flynn,
Chem. Mater. 1996, 8, 1600–1615.
[21] L. Lafferentz, F. Ample, H. Yu, S. Hecht, C. Joachim, L. Grill, Sci-
ence 2009, 323, 1193–1197.
[22] a) S. Weigelt, C. Busse, C. Bombis, M. M. Knudsen, K. V. Gothelf,
E. Lægsgaard, F. Besenbacher, T. R. Linderoth, Angew. Chem. 2008,
120, 4478–4482; Angew. Chem. Int. Ed. 2008, 47, 4406–4410; b) M.
Treier, N. V. Richardson, R. Fasel, J. Am. Chem. Soc. 2008, 130,
14054–14055.
[23] N. A. A. Zwaneveld, R. Pawlak, M. Abel, D. Catalin, D. Gigmes, D.
Bertin, L. Porte, J. Am. Chem. Soc. 2008, 130, 6678–6679.
[24] a) L. Grill, M. Dyer, L. Lafferentz, M. Persson, M. V. Peters, S.
Hecht, Nat. Nanotechnol. 2007, 2, 687–691; b) M. Bieri, M. Treier, J.
Cai, K. Aꢉt-Mansour, P. Ruffieux, O. Grçning, P. Grçning, M. Kas-
tler, R. Rieger, X. Feng, K. Mꢇllen, R. Fasel, Chem. Commun. 2009,
6919–6921.
[25] a) O. M. Behr, G. Eglinton, R. A. Raphael, Chem. Ind. 1959, 699–
700. For a recent review on dehydrobenzoannulenes, see: b) E. L.
Spitler, C. A. Johnson II, M. M. Haley, Chem. Rev. 2006, 106, 5344–
5386.
[26] Q. Zhou, P. J. Carroll, T. M. Swager, J. Org. Chem. 1994, 59, 1294–
1301.
[27] T. Nishinaga, N. Nodera, Y. Miyata, K. Komatsu, J. Org. Chem.
2002, 67, 6091–6096.
[39] B. Ilan, G. M. Florio, M. S. Hybertsen, B. J. Berne, G. W. Flynn,
Nano Lett. 2008, 8, 3160–3165.
Chem. Eur. J. 2010, 16, 8319 – 8328
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8327

Y. Tobe et al.
[40] EpiCalc was downloaded on the Wardꢄs group web site (http://
www.nyu.edu/fas/dept/chemistry/wardgroup/Software.html).
[41] a) G. A. Wagner, Z. Naturforsch. B 1969, 24, 824–832; b) R. H.
Baughman, J. Polym. Sci. Polym. Phys. Ed. 1974, 12, 1511–1535. For
a review, see: c) V. Enkelmann, Adv. Polym. Sci. 1984, 63, 91–136.
[42] a) J. J. Kane, R.-F. Liao, J. W. Lauher, F. W. Fowler, J. Am. Chem.
Soc. 1995, 117, 12003–12004; b) Z. Li, F. W. Fowler, J. W. Lauher, J.
Am. Chem. Soc. 2009, 131, 634–643; c) A. Sun, J. W. Lauher, N. S.
Goroff, Science 2006, 312, 1030–1034. For reviews, see: d) F. W.
Fowler, J. W. Lauher, J. Phys. Org. Chem. 2000, 13, 850–857; e) J. W.
Lauher, F. W. Fowler, N. S. Goroff, Acc. Chem. Res. 2008, 41, 1215–
1229.
[43] a) R. Fittig, A. Schltz, Justus Liebigs Ann. Chem. 1873, 166, 361–
382; b) W. Schlenk, M. Karplus, Chem. Ber. 1928, 61, 1675–1680;
c) R. Scholl, H. K. Meyer, Chem. Ber. 1932, 65, 926–927; d) V. Sach-
weh, H. Langhals, Chem. Ber. 1990, 123, 1981–1987; e) M. Smet, R.
Shukla, L. Fꢇlçp, W. Dehaen, Eur. J. Org. Chem. 1998, 2769–2773.
[44] K. Miyawaki, T. Kawano, I. Ueda, Tetrahedron Lett. 2000, 41, 1447–
1451.
[45] The dihedral angles of the central aromatic bonds of [6]helicene de-
rivatives range from 26 to 308 and those next to the central bonds
range from 12 to 188; a) D. A. Lightner, D. T. Hefelfinger, T. W.
Powers, G. W. Frank, K. N. Trueblood, J. Am. Chem. Soc. 1972, 94,
3492–3497; b) C. de Rango, G. Tsoucaris, J. P. Delerq, G. Germain,
J. P. Putzeys, Cryst. Struct. Comm. 1973, 2, 189–192; c) G. W. Frank,
D. T. Hefelfinger, D. A. Lightner, Acta Crystallogr. Sect. B 1973, 29,
223–230; d) F. Aloui, R. E. Abed, A. Marinetti, B. B. Hassine, Tetra-
hedron Lett. 2007, 48, 2017–2020.
[46] a) M. G. Rossmann, J. Chem. Soc. 1959, 2607–2613; b) A. Izuoka,
K. Wakui, T. Fukuda, N. Sato, T. Sugawara, Acta Crystallogr. Sect. C
1992, 48, 900–902; c) J. M. Robertson, J. Trotter, J. Chem. Soc. 1959,
2614–2623.
review on twisted acenes, see: b) R. A. Pascal, Jr., Chem. Rev. 2006,
106, 4809–4819.
[48] To confirm reliability of our experimental conditions, we performed
control experiments by using a monolayer of 10,12-nonacosadiynoic
acid at 1-phenyloctane/graphite interface, which is reported to poly-
merize on a surface upon application of electronic pulse.[20f] By ap-
plying electronic pulse from an STM tip, the polymerization was
indeed observed, indicating validity of our experimental conditions.
See also the Supporting Information.
[49] a) N. L. Allinger, Y. H. Yuh, J. H. Lii, J. Am. Chem. Soc. 1989, 111,
8551–8566; b) J. H. Lii, N. L. Allinger, J. Am. Chem. Soc. 1989, 111,
8566–8575; c) J. H. Lii, N. L. Allinger, J. Am. Chem. Soc. 1989, 111,
8576–8582.
[50] B. Ma, J. H. Lii, N. L. Allinger, J. Comput. Chem. 2000, 21, 813–825.
[51] Gaussian 03, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Na-
nayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wallingford
CT, 2005.
[47] The largest known end-to-end twist angle of a pentacene unit is
1448; a) J. Lu, D. M. Ho, N. J. Vogelaar, C. M. Kraml, R. A. Pas-
cal, Jr., J. Am. Chem. Soc. 2004, 126, 11168–11169. For an excellent
Received: March 21, 2010
Published online: June 30, 2010
8328
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8319 – 8328