Synthesis and Photochromism of Chloro- and tert-Butyl-Functionalized Terarylene Derivatives for Surface Deposition

Article abstract of DOI:10.1002/ejoc.201601657

In the field of extreme miniaturization of electronic devices, it is necessary to study candidate molecules at the single-molecular level by scanning tunneling microscopy (STM). This necessitates molecules with specific functionalizations to control molec

Full text of DOI:10.1002/ejoc.201601657

A Journal of
Accepted Article
Title: Synthesis and photochromism of chloro and tert-butyl terarylenes
derivatives for surface deposition
Authors: Jan Patrick Dela Cruz Calupitan, Olivier Galangau, Olivier
Guillermet, Roland Coratger, Takuya Nakashima, Gwenael
Rapenne, and Tsuyoshi Kawai
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201601657
Link to VoR: http://dx.doi.org/10.1002/ejoc.201601657
Supported by

10.1002/ejoc.201601657
European Journal of Organic Chemistry
FULL PAPER
Synthesis and photochromism of chloro and tert-butyl
terarylenes derivatives for surface deposition
Jan Patrick Dela Cruz Calupitan,[a,b,c] Olivier Galangau,[a,b] Olivier Guillermet, [c] Roland Coratger, [c]
Takuya Nakashima,[b] Gwénaël Rapenne,*[a,c] and Tsuyoshi Kawai*[a,b]
Abstract: In the field of extreme miniaturization of electronic devices, photon-quantitative
reactivity,[6]
efficient
oxidative
it is necessary to study candidate molecules at the single molecular
level by scanning tunneling microscopy (STM). This necessitates
molecules to harbor specific functionalization to control molecule-
surface interactions to avoid molecular diffusion on the surface and
molecule-molecule interactions to prevent aggregation of molecules.
In this context, we present the synthesis and photochromism of
terarylene derivatives with tert-butyl and chloride groups. We show
that these groups preserve the photochemical and switching
properties of molecules. DFT calculations on the conformation of the
molecules predict the presence of several conformations during
vacuum deposition. Initial STM experiments show that tert-butyl
groups give images with a better contrast to identify the molecules
on surface without any aggregation.
cycloreversion,[7] fluorescence switching,[8] circularly-polarized
luminescence switching,[9] and selective photo-release of organic
fragments[10]. These new optimizations now give access to new
functions which may be incorporated to advanced molecular
networks and circuits.
For the aforementioned applications, it is necessary to study the
electronic properties, molecule-surface interactions, and
switching events of these molecules at the single molecular level.
Scanning tunneling microscopy (STM) offers unique possibilities
to study single molecules on metallic, semi-conducting, and
insulating surfaces. Since the flickering of diarylethenes at the
single molecular level was observed,[11] several groups started to
investigate the electronic and switching properties of
diarylethenes at the single molecular level[12] and of 2D self-
assembled structures at the solid-liquid interface[13]
.
For
terarylenes, two the best of our knowledge, only two studies on
a self-assembled monolayer of a derivative attempted to
approach single molecular level properties.[14] It remains
therefore to investigate terarylenes under STM at the single
molecular level to exploit their unique properties. In this paper,
we present the synthesis and photochemical characterization of
terarylene derivatives specifically designed for STM studies
(Figure 1).
Introduction
Considerable interest in organic chemistry is being focused on
the search for new molecular building blocks which could be
exploited
for
molecular-scale
electronic
devices.[1]
Photochemically reactive bistable molecules have been
extensively studied as new promising platforms for the next
generation molecular devices and memories.[2] Diarylethenes
and their derivatives have been sought for molecular memories
and switches because of their high fatigue-resistance, thermal
stability, solid-state reactivity, and high photosensitivity.[3]
Chemical modification of their central ethene units has been
S
S
UV
Vis
N
N
N
N
R
R
R
R
S
S
S
S
subject to molecular designs for flexible tuning of properties and
functionalities.[2c,
closed (c)
open (o)
3c, 4]
The authors have also studied some of
them so-called terarylenes, which have central aromatic units
such as thiophene, thiazole and benzothiophene.[5] Some of
them displayed unique and advanced functionality, such as
1: R =
3: R =
5: R =
Cl
Ph
Ph
[a]
[b]
[c]
J.P.D.C. Calupitan, Dr. O. Galangau, Prof. G. Rapenne, Prof. T. Kawai
NAIST-CEMES International Collaborative Laboratory
CEMES-CNRS (UPR 8011)
29 rue Jeanne Marvig, BP94347, 31055 Toulouse, France
E-mail: gwenael.rapenne@cemes.fr
URL: http://www.cemes.fr/rapenne?lang=fr
J.P.D.C. Calupitan, Dr. O. Galangau, Dr. T. Nakashima, Prof. T. Kawai
Graduate School of Materials Science
Nara Institute of Science and Technology
8916-5 Takayama, Ikoma, Nara, Japan
E-mail: ntaku@ms.naist.jp, tkawai@ms.naist.jp
URL: http://mswebs.naist.jp/LABs/kawai/english/index.html
J.P. Calupitan, Dr O. Guillermet, Prof. R. Coratger and Prof. G.
Rapenne
S
N
S
N
UV
Vis
N
N
N
N
R
R
R
R
S
S
S
S
closed (c)
open (o)
2
: R =
4
: R =
6: R =
Cl
Figure 1. 1o and 2o are the unfunctionalized model compounds. 3o
and 4o are the target molecules with tert-butyl groups and 5o and 6o
the molecules functionalized with peripheric chlorine centers.
Université de Toulouse
29 rue Marvig, BP94347, 31055 Toulouse Cedex 4, France
1
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201601657
European Journal of Organic Chemistry
FULL PAPER
For STM studies, we have selected two terarylenes 1o and 2o
as model compounds. 1o was chosen because it preserves a
1o[6b] and 2o[5b] were previously synthesized by a double Suzuki
coupling of the side moieties to the central rings 2,3-
reactive
photocyclization
conformation
to
display
photon-quantitative
dibromobenzothiophene
and
4,5-dibromo-2-phenylthiazole
[6b]
while 2o was selected because it displayed
respectively. However, the carbons involved in these double
coupling reactions have markedly different reactivities. For
example, for 2o, the same one-pot conditions were used to
attach the 5-methyl-2-phenylthiazol-4-yl moiety to the reactive
carbon at the 5-position and the non-reactive 4-position of the
thiazole ring. We therefore chose a different route for the
synthesis that respects the different reactivities of the carbons in
these positions, taking inspiration from previous works. [18]
highly efficient oxidative cycloreversion reaction with 2000 %
electron efficiency [7a]. Tert-butyl groups (3o and 4o) and chloro-
substituents (5o and 6o) were incorporated on both sides of
these molecules. Tert-butyl groups are routinely used for
molecules studied under STM because they show bright
contrast in STM images, minimize aggregation on the surface,
and slightly decouple the electronic properties of the molecule
from the surface; their steric bulky groups elevate aromatic
cores from the surface to avoid physisorption or chemisorption
that would otherwise cause a loss of molecular properties.[15]
The chlorine group on the other hand has been selected
because halogen-substituted molecules have been shown to
direct their surface assembly by surface-molecule dipole
interactions on insulating substrates composed of crystalline
We opted for consecutive coupling to these different carbon
positions. For
phenylthiazole
1o
(Scheme 2), 4-bromo-5-methyl-2-
[17b]
was first connected to the 3-position of the
benzothiophene substituent by Suzuki coupling with the
commercially available benzo[b]thien-3-ylboronic acid to yield 4-
(benzothiophen-3-yl)-2-phenyl-5-methylthiazole (12). Only then
was the second 5-methyl-2-phenylthiazol-4-yl derivative
attached to the electron-poor carbon in the 2-position by C-H
direct arylation using Pd(OAc)2, Cs2CO3, pivalic acid, di-tert-
butyl(methyl)phosphonium tetrafluoroborate in mesitylene to
produce 1o. It should be noted that this gave a small increase of
the yield of the synthesis from the reported 57%[6b] to an overall
yield of 62% with the two consecutive coupling reactions (87 %
and 71 % for each step).
NaCl or KCl bilayer previously grown over
a metallic
substrate.[12c] In addition, halogens are especially useful for
forming stable assemblies of molecules on semi-conducting
surfaces by a combination of molecule-molecule and surface-
molecule interactions. [16]
Results and Discussion
Scheme 2. Synthesis of model compounds 1o and 2o by a new synthetic
pathway. a) benzo[b]thien-3-ylboronic acid Pd(PPH3)4, K3PO4, DME, H2O; b) 1.
Syntheses
To prepare the tert-butyl and chloro derivatives, two thiazole
precursors were necessary. Scheme 1 shows the synthetic way
to access the 5-methyl-2-phenylthiazolyl groups for molecules
3o to 6o. 4-bromo-2-(3,5-di-tert-butylphenyl)-5-methylthiazole
(8) was synthesized from the commercially available 2,5-di-tert-
butyl-1-bromobenzene. It is first coupled with 4,5-
dibromothiazole to produce 4-bromo-2-(3,5-di-tert-butylphenyl)-
5-methylthiazole (7) followed by SN2 reaction of its deprotonated
adduct (by LDA) with methyl iodide. Meanwhile, a similar
synthesis route for 4-bromo-5-methyl-2-phenylthiazole from
thiobenzamide[17] was used to synthesize the chloro derivative 4-
l
j
Br
h
S
S
k
S
N
S
B(OH)2
a
N
i
N
N
l
c
S
S
S
12
1o
Br
N
S
Ph
Br
Ph
N
S
S
m
N
S
N
c
b
N
N
N
Br
N
S
S
S
bromo-2-(p-chlorophenyl)-5-methylthiazole
(11)
from
p-
Ph
S
13
2o
chlorothiobenzamide.
n-BuLi, B(OMe)3, THF 2. 4-bromo-2-phenylthiazole, Pd(PPH3)4, K3PO4, DME,
H2O; c) 4-bromo-5-methyl-2-phenylthiazole, Pd(OAc)2, Cs2CO3, pivalic acid,
di-tert-butyl(methyl)phosphonium tetrafluoroborate, mesitylene.
Meanwhile, the carbon in 4-position of the thiazole ring is known
to be the least reactive so this was prioritized for the first
attachment. The boronic acid derivative of 5-bromo-4-methyl-2-
phenylthiazole were first generated in-situ before adding 4-
bromo-2-phenylthiazole and the other reagents for Suzuki
coupling to produce 2-phenyl-5-methyl-4-(2-phenylthiazol-4-yl)-
thiazole (13). The electron-poor carbon in the 5-position of the
thiazole ring is then free for C-H direct arylation to yield 2o. In
that case, with a yield of 96% and 26% for the two consecutive
steps, the overall yield for the two coupling reactions has been
strongly improved from 6% for the previously reported
synthesis[5b] to 25%.
Scheme 1. Synthesis of the tert-butyl and chloro derivatives of side 5-methyl-
2-phenylthiazolyl derivatives. a) n-BuLi, B(OMe)3, THF; b) 2,4-dibromothiazole,
Pd(OAc)2, Xantphos, K3PO4, THF, 60 oC; c) LDA, THF, CH3I -78 oC to RT; e)
2-bromo-1,1-diethoxyethane, EtOH, 70 oC; (e) NBS, THF, 60 oC; f) LDA, CH3I,
THF.
2
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201601657
European Journal of Organic Chemistry
FULL PAPER
The same consecutive coupling reactions were utilized to
synthesize respective tert-butyl derivatives 3o and 4o from 8
(Scheme 3) and chloro-derivatives 5o and 6o from 11 (Scheme
4). Detailed synthetic procedures may be found in the
Experimental Section while NMR spectra of precursors and
target materials may be found in the Supporting Information (SI
2-16).
Table 1. Photochemical properties of compounds
λmax,o(ε)[a]
308(3.51)
327(3.21)f
310(3.43)
318(3.21)
313(3.40)
322(3.36)
λmax,c(ε)[b]

o-c
c-o
Ec-o
[c]
[d]
[e]
1
2
3
4
5
6
600(1.04)
0.95
0.5
0.02
0.03f
0.03
0.04
0.02
0.03
0.591
0.706
0.585
0.693
0.596
0.702
591(1.32)f
595(1.06)
594(1.31)
607(1.12)
600(1.27)
0.94
0.5
S
S
S
B(OH)2
a
N
N
N
c
S
S
S
0.95
0.5
14
3
Br
N
S
Ph
Ph
8
S
N
S
N
All experimental measurements were done in dichloromethane. [a]
max of open
b
c
N
N
N
[b]
forms in nm,  in 104 M-1cm-1

max of closed forms in the visible region in nm,
Br
S
[c]
[d]
in 104 M-1cm-1
photocyclization quantum yield of open forms
cycloreversion quantum yield of closed forms; [e]Energy differences of DFT-
S
S
N

Ph
15
4
S
[f]
optimized structures of closed and open forms forms in eV in vacuum; data
from ref 7b.
Scheme 3. Synthesis of the tert-butyl-functionalized terarylenes 3o and 4o. a)
benzo[b]thien-3-ylboronic acid Pd(PPH3)4, K3PO4, DME, H2O; b) 1. n-BuLi,
B(OMe)3, THF 2. 4-bromo-2-phenylthiazole, Pd(PPH3)4, K3PO4, DME, H2O; c)
8, Pd(OAc)2, Cs2CO3, pivalic acid, di-tert-butyl(methyl)phosphonium
tetrafluoroborate, mesitylene.
Figure 2.Changes in spectra of terarylenes upon illumination of UV-light (254
nm). Spectra in blue shows photostationary state.
S
S
S
B(OH)2
a
N
N
N
c
S
S
S
Cl
Cl
Cl
16
5
Br
N
S
Cl
Ph
Ph
11
S
N
S
N
b
c
N
N
N
Br
S
S
S
N
Cl
Cl
Cl
Ph
17
6
S
Scheme 4. Synthesis of the compounds chloro-functionalized terarylenes 5o
and 6o. a) benzo[b]thien-3-ylboronic acid Pd(PPH3)4, K3PO4, DME, H2O; b) 1.
n-BuLi, B(OMe)3, THF 2. 4-bromo-2-phenylthiazole, Pd(PPH3)4, K3PO4, DME,
H2O; c) 11, Pd(OAc)2, Cs2CO3, pivalic acid, di-tert-butyl(methyl)phosphonium
tetrafluoroborate, mesitylene.
Photochemistry
Upon irradiation of solutions of all the compounds in
dichloromethane with UV light (256 nm), bands increased in the
550-650 nm region of the visible spectrum accompanied by a
decrease and red-shifting of peaks in the UV region. (Figure 2)
These are due to the simultaneous loss of aromaticity of the
thiazole rings and the formation of a planar -conjugated
structure in the center of the molecule. Similar peaks centered
around 600 nm were observed for the closed forms, with a fine
structure appearing for those of 4 and 6. TD-DFT calculations
Photochemical properties of respective derivatives of 1o and 2o
do not differ much from each other. (Table 1) Maximum
absorbance peaks recorded in dichloromethane for 1o, 3o, and
5o are very similar with absorption bands at 308 nm, 310 nm,
and 313 nm respectively. Meanwhile, those of 2o, 4o, and 6o
are at 327 nm, 318 nm, and 322 nm respectively. They also
displayed a second peak at lower wavelengths. No systematic
solvatochromism was observed as these peaks were very
similar to those reported for 1o[6b] and 2o[5b] in hexane.
predict these max values quite well. (SI22-28)
All these
compounds displayed clear isosbestic points at about 350 nm,
supporting the well-defined two-component reaction scheme in
Figure 1. No marked spontaneous bleaching was observed in
3
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201601657
European Journal of Organic Chemistry
FULL PAPER
the colored solution at room temperature similar with mother
diarylethene derivatives. Molar absorption coefficients of both
open and closed forms of respective derivatives of 1 and 2 are
similar within 4% relative standard deviation. The bulky electron-
donating tert-butyl groups and electronegative chlorine atoms do
STM sample preparations. Photocyclization quantum yield of 1o
and its derivatives slightly decreased to around 95 %. It is
suspected that solvent polarity caused a small disruption of the
intramolecular tethering forces as previously mentioned. The 1H-
NMR signal responsible for H-bonding (noted h in Scheme 2) is
not seem to affect much the electronic structure of the molecules. well resolved in non-polar solvents such as toluene but it mixes
Photoreactivity of diarylethenes and their derivatives is
predominantly affected by their conformation in the excited state,
which, due to the fast dynamics of the reaction, is determined by
the ground state conformation. In solution at room temperature,
diarylethenes oscillate between two conformers: the non-
with other aromatic proton signals in dichloromethane. (SI29)
This means that CH-N interactions are not strong enough to fully
prevent rotation to the non-reactive conformation in polar
1
solvents. Meanwhile, H-NMR signals of two methyl groups are
indistinguishable in non-polar solvents but are split in polar
solvents. (SI29) X-ray crystal structure and DFT calculations of
1o show that: (1) the aromatic rings are skewed from each
others’ planes so that there is no communication among them by
-conjugation and; (2) the closeness of the methyl groups to the
aromatic thiazole planes associates with CH- interactions. Due
to these results, the methyl protons are chemically equivalent in
the NMR spectra of anti-parallel conformation. The observed
resolution of methyl peaks in dichloromethane therefore
suggests a disruption of the structure that dissymetrizes methyl
protons causing the resolution of the two peaks. These
observations are similar to when intramolecular forces are
disrupted by heating or when measuring the spectra in H-
bonding-disruptive solvents such as methanol.[6b]
reactive parallel conformation and the reactive anti-parallel
[3a]
conformation.
Various strategies that shift the equilibrium
between these two conformations to the anti-parallel one were
therefore employed to control the quantum yield of the
photocyclization reaction. These mostly included introducing
bulky groups or bridging the two side rings to prevent the
3c, 19]
rotation of the molecule to the parallel conformation.[3b,
These strategies were perfected in the design and synthesis of
1o, which displayed 98 % photocyclization quantum yield
because it is locked in the anti-parallel conformation by multiple
non-covalent interactions.[6b] Consequently, 2o equilibriates to
the parallel conformation so that only 40 % photocyclization
quantum yield was observed in a non-polar solvent. [5b]
In the present study, photochemical properties were measured
in dichloromethane because it is the preferred solvent used for
On the other hand, photocyclization quantum yields of 2o and its
derivatives slightly increased to 50 %. A similar comparison of
Figure 3. Rotational conformers of 3o and 4o. Stronger N-H interactions in 3o are highlighted in blue while weaker S-N interactions are highlighted in green.
4
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201601657
European Journal of Organic Chemistry
FULL PAPER
NMR spectra in cyclohexane and in dichloromethane showed no
difference in the aromatic peaks. On the other hand, methyl
peaks are closer to each other (SI30). Consequent to the
analysis above, slightly stronger CH3-aryl interactions may
contribute to a slight shift of the equilibrium between the reactive
anti-parallel conformation and the non-reactive parallel
conformation to the reactive one. Compounds 3o and 5o again
showed “photon-quantitative” nature with photochemical ring-
cyclization quantum yield close to unity. It must also be noted
that cycloreversion quantum yields are unsurprisingly
maintained at a low level smaller than 4 % for all compounds.
of the 2-phenylthiazolyl moieties were rotated. Torsional energy
scans yielded 3 minima: the first one corresponds to the anti-
parallel conformation which has both the S-N and N-H
interactions (Conformation 0, 3o-C0); the second one preserved
only the N-H (3o-C1) and; the last one only the S-N interaction
(3o-C2) (Figure 3, see SI33 for energy scan). Then, the
structures of these minima were further geometry-optimized by
the same method as above. The lack of S-N interactions in 3o-
C1 caused a destabilization of 7.8 kJ/mol (81 meV) while the lack
of the stronger N-H interactions in 3o-C2 caused
a
destabilization of up to 9.3 kJ/mol (96 meV). Using the
calculated energy differences of the conformations and following
the Boltzmann equation, we predict that 93.7 % of the molecule
stays in the reactive C0 conformation at room-temperature in
vacuum. This goes down to 77.8 % at elevated temperatures
(250°C) necessary for sublimating the molecule during vacuum-
deposition of samples for STM measurements. (Table 2) It is
therefore expected, due to surface deposition that exposes the
molecule to higher temperatures, that STM images of 3o may
reveal more than one conformation of the molecule.
DFT Calculations
We performed calculations by using the Gaussian09 package [20]
to gain insight about the photochemistry observed and predict
the behavior of molecules under vacuum. We used the
functional B3LYP with basis set 6-31G(d,p). Geometry
optimizations of the reactive conformations of the open forms
and of the closed forms were first performed with zero-point
correction. IR spectra calculations showed no negative
vibrations, ensuring that calculated energies corresponded to
absolute minima.
Table 2. Boltzmann distribution of
Similar results as published for 1o[6b] were obtained for its
derivatives. For 5o, distance between sulfur of the
benzothiophene moiety and the nitrogen of the side-thiazolyl
group was 3.14 Å and between Hh (Scheme 2) and the other
side-thiazolyl N at 2.64 Å. Interestingly, the S-N and N-H
distances for 3o were slightly different at 3.18 Å and 2.60 Å
respectively. (Figure 3, 3o-C0) This could be due to a slight
twisting of the C-C bonds between the thiazolyl and phenyl
moieties to avoid steric hindrance. All measurements were
smaller than the sum of the van der Waals radii of respective
atoms involved (3.35 Å for S-N and 2.75 Å for N-H). All these
measurements indicate that the strong intramolecular
interactions involved in keeping 1o in its reactive conformation
are maintained in its derivatives 3o and 5o. This is in reasonable
agreement with similarly high photocyclization quantum yields
observed for these three compounds. Geometry-optimizations
on the closed forms 1c, 3c, and 5c were also achieved so that
the relative energy differences between the open and closed
forms of 1, 3, and 5 were computed. All these were found to lie
at around 590 meV. (Table 1, SI 31)
Similarly, the geometry of reactive conformations of 4o and 6o
are relatively the same to that of 2o. The closed forms 2c, 4c,
and 6c are also similar. (SI 32) The differences in energies
between the optimized closed-forms and the open forms were
found at 700 meV. (Table 1, SI32) All these calculations showed
that the peripheral groups, including the bulky tert-butyl groups,
have a little effect on the geometry, energy, and thus the
photochemistry and switching properties of the molecules.
To probe further how the molecules would behave under ultra-
high vacuum conditions of the STM chamber, DFT calculations
were employed on the molecules with tert-butyl groups in the
periphery, 3o and 4o. We investigated the equilibrium between
the parallel and anti-parallel conformations in vacuum by rotating
the dihedral angles involved and calculating the total energies.
For 3o, the dihedral angle between the benzothiophene and one
conformers in vacuum
Distribution / %
298.15 K
523.15 K
3o
C0
94
4
78
13
9
C1
C2
2
4o
C0
C1
C2
6
89
5
16
71
14
For 4o, the geometry-optimized energy of the reactive
conformation (4o-C0) was found to lie between those of two non-
reactive conformations (4o-C1, C2). (Figure 3, see SI34 for
scans) Thus, the ratio of the molecules found in the reactive vs
the non-reactive conformation are switched relative to those of
3o. Majority of the molecules are in the non-reactive
conformation at room temperature in vacuum. This goes down to
70.7 % once the temperature is increased. This predicts that
majority of the molecule would be in the non-reactive
conformation in STM images.
STM Experiments
To probe the impact of tert-butyl groups on STM images, we
performed initial STM experiments under low-temperature ultra-
high vacuum condition. We have chosen 1o and its tert-butyl
derivative 3o for the first experiments because DFT predictions
and photocyclization quantum yields indicate that a majority of
molecules would remain in the reactive conformation. Figure 4
shows the DFT-optimized structure of 1o (Figure 4a) and 3o
(Figure 4b) with their respective measurements.
5
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201601657
European Journal of Organic Chemistry
FULL PAPER
Samples were first deposited onto a metallic filament by
immersion in a dichloromethane solution of the compound. After
drying, the filament was heated to about 523 K for the
sublimation of molecules. Successful deposition of 1o was
observed on a Ag(111) substrate which was cooled down to 4.5
K. As shown in Figure 4c, the molecule appeared as a unique
and large almost circular bright spots and no apparent structure
were found. This symmetric shape did not reveal the fine
structure of the molecule. The bright contrast could be partly due
to a strong physisorption of the molecule on the surface as well
as a continuous electronic density throughout the molecule due
to -conjugation. In the optimized structure of 1o (Figure 4a), the
distance between the side phenyl groups is 1.5 nm, which is
considerably larger than observed size, 1.2nm (Figure 4c). This
claim about the electronic structure of the molecule and this
anomaly about the size could have been explained by how the
STM tip approaches the molecule on the surface, i.e., its surface
configuration upon absorption. However, the lack of geometric
information on the STM image renders this impossible.
spots, consistent with previous STM observations of molecules
with tert-butyl groups (Figure 4d).[15] Meanwhile the central bright
spot could correspond to the central aromatic structure of the
molecule. Aside from -conjugation due to the aromatic rings,
the sulfur atoms which are known to interact well with silver
could contribute to the bright contrast. To the best of our
knowledge, these are the first single molecule images of
photochromic terarylenes. Further manipulations and switching
of the compound are currently underway to confirm these
observations.
Conclusions
We thus synthesized terarylene derivatives especially designed
for STM studies on metallic surfaces (tert-butyl-functionalities)
and insulating or semi-conducting surfaces (chloro-groups). 1o
was chosen as a model because of its highly sensitive
photocyclization while 2o was chosen because of its efficient
oxidative cycloreversion. Synthesis of terarylene derivatives with
tert-butyl and chloride groups has been achieved in very good
yields and fully characterized. Functional groups attached to
them do not have a significant effect on the photochemical and
switching properties of the molecules. DFT calculations in
vacuum predict that the rotational equilibrium between the
parallel and anti-parallel conformations of these molecules may
contribute to a multiplicity of forms observed under STM. Thanks
to the presence of the tert-butyl groups, STM imaging of 3o did
result to a nice improvement of the imaging with a fine structure
observed on Ag(111). In addition, the presence of 4 bright
circular signals attributed to tert-butyl groups helped interpret
STM images. We therefore present a simple molecular design
principle to aid in STM imaging of molecules whilst preserving
their photo-functions.
These new terarylenes make now possible their study as single
molecules on metallic surfaces for the tert-butyl derivatives and
on NaCl or KCl / metal for the chloride derivatives. Then, this
family of molecules could find applications in future molecular-
electronics and circuits.
Figure 4. DFT-optimized structure of 1o (a) and 3o (b) with measured sizes. c)
Deposition of 1o on Ag(111) produced bright spots with no apparent fine
structure. d) Deposition of 3o on Ag(111) revealed a much more resolved
structure which show prominent circular lobes charactertistic of tert-butyl
groups.
Experimental Section
General syntheses, purification, and characterization
The syntheses are shown in Schemes 1-4. 1-bromo-3,5-di-tert-
butylbenzene, THF (2.5 M), 2,4-dibromothiazole, thiobenzamide, 4-
chlorothiobenzamide, palladium (II) acetate, trimethyl borate, 4,5-Bis-9,9-
dimethylxanthene (Xantphos), lithium diisopropylamide, iodomethane,
tetrakis(triphenylphosphine)palladium, benzothiophene-3-boronic acid,
di-tert-butylmethylphosphonium tetrafluoroborate, pivalic acid, and dry
solvents were purchased from Sigma Aldrich and were used without any
further treatment. HPLC grade solvent purchased from Sigma Aldrich
were used for purification and spectroscopic analyses. 4-bromo-5-
methyl-2-phenylthiazole was prepared according to literature.[6b] 1H and
13C NMR spectra were recorded on a Bruker Avance NMR (300 MHz or
500 MHz) at 25°C. Full assignments were made by comparing with
literature and by using COSY, ROESY, HMBC and HMQC methods
when necessary. Chemical shifts are defined with respect to TMS = 0
Meanwhile, sublimation of 3o under the same conditions
showed at least two shapes of spots. (SI 35) Figure 4d shows
an image of the molecules with a bright spot and four lobes
skewed around it. The distance between respective maxima of
these spots are at 1.5 nm and 1.4 nm. The width of the central
bright spot is 9 Å. These measurements correspond well with
DFT calculations on 3o that place the distance between
respective tert-butyl groups at 1.5 nm and between the benzo-
group of the benzothiophene moiety and the methyl groups at
8.3 Å (Figure 4b). Following DFT prediction, we propose to place
the tert-butyl groups at the rounded corners of the rectangular
6
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201601657
European Journal of Organic Chemistry
FULL PAPER
ppm for 1H and 13C NMR spectra and were measured relative to residual
solvent peaks. The following abbreviations have been used to describe
the signals: s for singlet; d for doublet; t for triplet; q for quadruplet; m: for
multiplet. The numbering scheme is given in Schemes 1-4. Mass spectra
were measured from Q TRAP 2000 (Applied Biosystems, High resolution
mode) and API-365 (Perkin Elmer Sciex) spectrometers for ESI
measurements, while a DSQ Thermo Fisher Scientific spectrometer was
used for DCI measurements. Melting points were measured on a Kofler
Heizbank melting point apparatus or with a capillary Melting Point Meter
M5000 apparatus. Purifications by automatic flash chromatography were
performed with HPLC grade solvents on a Spot ultimate automatic
chromatography system (ARMEN Instrument). Two types of prepacked
column were used either Versaflash column from Aldrich or Universal
column from Yamazen Corporation. High purity grade silica from Fluka
(pore size of 60 Å, particle size of 40-63 µm) was used for manual flash
chromatography. TLC analyses were achieved on silica gel bought from
Fluka (silica gel matrix containing a fluorescent indicator at 254 nm).
A solution of p-chlorothiobenzamide (5 g, 1 eq, 29 mmol) and 2-bromo-
1,1-diethoxyethane (8.59 g, 1.5 eq, 43.6 mmol) in 20 mL of dry ethanol
was refluxed overnight. Then, the mixture was allowed to cool to room
temperature, diluted with water, extracted with ether, and washed with
brine. The combined organic phases were dried over MgSO4, filtered,
and dried in vacuo. Purification by column chromatography (or
hexane/ethyl acetate 40:1) afforded 3.14 g of 9 as yellow oily crystals
(65% yield). Rf=0.42 (CH2Cl2/hexane 1/1); 1H NMR (300 MHz, CDCl3): δ
7.92 – 7.87 (m, AA’XX’, 4H, 2H, Hd), 7.86 (d, J = 3.3 Hz, 1H, He), 7.41 (m,
AA’XX’, 2H, He), 7.33 ppm (d, J = 3.3 Hz, 1H, Hg); 13C NMR (75 MHz,
CDCl3): δ 167.17, 143.94, 136.04, 132.20, 129.30, 127.88, 119.26 ppm;
HRMS (EI): m/z calcd for C9H7ClNS+ [M+H]+: 194.9909; found: 194.9915.
Synthesis of 4-bromo-2-(4-chlorophenyl)-thiazole (10)
A solution of 9 (2.00 g, 1eq, 10.2 mmol) and N-bromosuccinimide (2.00g,
1.1eq, 11.2 mmol) in 50 mL DMF was stirred and refluxed overnight.
Then, the mixture was allowed to cool to room temperature, diluted with
water, extracted with ether and washed with brine. The combined organic
phases were dried over MgSO4, filtered and dried in vacuo. Purification
by silica-gel column chromatography (hexane/chloroform 8/2) afforded
2.4 g of 10 as a white powder (86 % yield). M.p. 103 oC; Rf=0.67
(CH2Cl2/hexane 1:1); 1H NMR (300 MHz, CDCl3): δ 7.85 – 7.75 (m,
AA’XX’, 2H, Hd), 7.72 (s, 1H, He), 7.47 – 7.35 ppm (m, AA’XX’, 2H, Hf);
13C NMR (75 MHz, CDCl3): δ 168.26, 145.12, 136.55, 131.68, 129.44,
127.51, 109.03 ppm; HRMS (EI): m/z calcd for C9H5BrClNS+ [M]+:
272.9015; found: 272.9016.
Synthesis of 4-bromo-2-(3,5-di-tert-butylphenyl)thiazole (7)
A solution of 1-bromo-3,5-di-tert-butylbenzene (3 g, 1.68 eq, 11.14 mmol)
in 30 mL dry THF was charged with a hexane solution of n-butyllithium
(12.26 mmol, 1.1 eq) at -78 oC under Ar atmosphere and stirred for 30
min. Then, trimethyl borate was added (1.9 mL, 16.71 mmol, 2.4 eq) and
the solution was slowly heated to room temperature overnight. Then, 5
mL of aq. 6M HCl was added and the solution was stirred for another
three hours. 200 mL of water was added and the precipitate was
collected, washed with water, and dried in vacuo to generate a white
precipitate. The product was used without further purifications. This was
then charged with 2,4-dibromothiazole (1.648 g, 1 eq, 6.79 mmol), and
potassium phosphate (4.32 g, 3 eq, 20.4 mmol). Meanwhile, a mixture of
palladium (II) acetate (0.038 g, 0.025 eq, 0.17 mmol, 2.5mol%) and 4,5-
Bis-9,9-dimethylxanthene (Xantphos) (0.098 g, 0.025eq, 0.17 mmol) in
25 mL of dried and degassed THF was stirred and bubbled with Ar gas.
A darker color was observed after 15 mins and this was added to the
solids above. This resulting mixture was refluxed at 60 oC for 18 hrs. The
solids were filtered on celite and the crude was extracted with CH2Cl2.
The solvent was removed and the product was purified by silica gel
column chromatography (CH2Cl2) and 7 was collected as a yellow oil to
yield 1.6 g of product (67% yield). Rf=0.67 (CH2Cl2/hexane 1:1); 1H NMR
(300 MHz, CDCl3): δ 7.75 (d, J = 1.8 Hz, 2H, Ha), 7.52 (t, J = 1.8 Hz, 1H,
Hb), 7.19 (s, 1H, Hc), 1.37 ppm (s, 18H, tert-Bu); 13C NMR (75 MHz,
CDCl3): δ 170.38, 151.85, 132.16, 125.90, 125.20, 120.89, 116.21, 35.13,
31.52 ppm; HRMS (ESI): m/z calcd for C17H23BrNS [M+H]+: 352.0735;
found: 352.0721.
Synthesis of 4-bromo-2-(4-chlorophenyl)-5-methylthiazole (11)
Synthesis of 11 was accomplished by a halogen-dance reaction from
7.[17b] 8.18 mmol of a THF solution of lithium diisopropylamide (1.5 eq)
was added to a solution of 1.5 g of 7 (1 eq, 5.45 mmol) in 20 mL of dry
THF at -78 oC and stirred for 15 mins under Ar. Then, iodomethane (2.32
g, 3 eq, 16 mmol) was added and the mixture was stirred at the same
temperature for an additional 2 hours. The solution was allowed to warm
to room temperature and stirred overnight before quenching with 20 mL
of 2M aqueous ammonium chloride. The product was extracted with
ether, washed with brine and dried in vacuo. Separation by column
chromatography (CH3Cl/hexane 3:7) afforded 1.45 g of 11 as a white
o
1
powder (92 % yield). M.p. 98 C; Rf=0.71 (CH2Cl2/hexane 1/1); H NMR
(300 MHz, CDCl3): δ 7.84 – 7.73 (m, AA’XX’, 2H, Hd), 7.43 – 7.32 (m,
AA’XX’, 2H, Hf), 2.42 ppm (s, 3H, CH3).; 13C NMR (75 MHz, CDCl3): δ
164.18, 136.32, 131.43, 129.29, 127.27, 125.65 , 13.17 ppm; HRMS (EI):
m/z calcd for C10H7BrClNS+ [M]+: 286.9171; found: 286.9171.
Synthesis of 4-bromo-2-(3,5-di-tert-butylphenyl)-5-methylthiazole (8)
Synthesis of 4-(benzothiophen-3-yl)-2-(phenyl)-5-methylthiazole (12)
5.08 mmol of a THF solution of lithium diisopropylamide (1.5 eq) was
added to a solution of 1.2 g of 7 (1 eq, 3.4 mmol) in THF at -78 °C and
stirred for 15 mins. Then, iodomethane (0.77 g, 1.6 eq, 5.42 mmol) was
added and the mixture was stirred at the same temperature for an
additional 2 hours. The solution was then slowly warmed to room
temperature before quenching with 20 mL 2M aqueous ammonium
chloride. The product was extracted with ether, washed with brine and
dried in vacuo. Purification by column chromatography (CH2Cl2) afforded
A Schlenck flask was charged with 4-bromo-5-methyl-2-phenylthiazole[6b]
(0.476 g, 1 eq, 1.87 mmol), benzothiophene-3-boronic acid (0.596 g, 1.5
eq, 2.81 mmol), potassium phosphate (596 g, 1.5 eq, 2.81 mmol), and
tetrakis(triphenylphosphine)palladium (0.064 g, 0.03 eq, 0.056 mmol)
under Ar atmosphere. DME (37.5 mL) and water (18.7 mL) were added
and the solution was refluxed at 60 oC for 24 hours. The solution was
then cooled to room temperature before extracting with ethyl acetate,
washed with water, dried over MgSO4 and dried in vacuo. Separation by
column chromatography (CH2Cl2/hexane, 1/1) afforded 0.5 g of the
product (87 % yield). M.p. 153 oC; Rf=0.46 (CH2Cl2/hexane 1:1); 1H
NMR (300 MHz, CDCl3): δ 8.16 – 8.09 (m, 1H, Hh), 8.05 – 7.98 (m, 2H,
Hi), 7.95 – 7.90 (m, 1H, Hj), 7.54 (s, 1H, Hk), 7.50 – 7.36 (m, 5H, Hl), 2.56
(s, 3H, CH3). 13C NMR (75 MHz, CDCl3): δ 164.15, 147.51, 140.15,
138.72, 134.60, 133.89, 130.66, 130.54, 129.86, 129.00, 127.78, 126.42,
o
0.905 g of 8 (73% yield). M.p. 136 C; Rf=0.65 (CH2Cl2/hexane 1:1); 1H
NMR (300 MHz, CDCl3) δ 7.69 (d, J = 1.8 Hz, 2H, Ha), 7.49 (t, J = 1.8 Hz,
1H, Hb), 2.43 (s, 3H, CH3), 1.36 ppm (s, 18H, tert-Bu); 13C NMR (125
MHz, CDCl3): δ 167.12, 151.84, 132.10, 128.42, 125.05, 124.93, 120.73,
35.15, 31.44, 13.18 ppm; HRMS (EI): m/z calcd for C18H25BrNS+ [M+H]+:
366.0891; found: 366.0876.
Synthesis of 2-(p-chlorophenyl)thiazole (9)
7
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201601657
European Journal of Organic Chemistry
FULL PAPER
125.77, 124.67, 124.50, 124.46, 122.61, 12.81 ppm; HRMS (EI): m/z
calcd for C18H13NS2 [M]+: 307.0489; found: 307.0487.
Synthesis
methylthiazole (16)
of
4-(benzothiophen-3-yl)-2-(4-chlorophenyl)-5-
Synthesis of 2-phenyl-5-methyl-4-(2-phenylthiazol-4-yl)-thiazole (13)
16 was prepared by the same procedure as 12 except that 4-bromo-5-
methyl-2-phenylthiazole was replaced by 11. From 11 (0.5 g, 1 eq, 1.73
mmol), benzothiophene-3-phenyl boronic acid (0.462 g, 1.5 eq, 2.60
mmol), potassium phosphate (0.552 g, 1.5 eq, 2.60 mmol),
tetrakis(triphenylphosphine)palladium (0.06 g, 0.03 eq, 0.05 mmol
3mol%) in 23 mL DME and 12 mL water, 0.490 mg (83 % yield) of 13
A solution of 4-bromo-5-methyl-2-phenylthiazole[6b] (0.5 g, 1.5 eq, 1.97
mmol) in dry THF (5.6 ml) under dry Ar conditions was cooled down to -
78 oC and then added with 1.8 mL 1.2 M n-butyllithium (1.65 eq, 2.16
mmol). The mixture was stirred for another 15 mins before adding
trimethyl borate (1.0 g, 7.5 eq, 9.84 mmol). The mixture was stirred for
another two hours at the same temperature before allowing to slowly
warm-up to room temperature overnight. Then, the solvents DME (3.3
mL) and water (3.3 ml) were slowly added. Next, 4-bromo-2-
phenylthiazole[6b] (0.31 g, 1 eq, 1.31 mmol), potassium phosphate (0.84
g, 3 eq, 3.94 mmol), and tetrakis(triphenylphosphine)palladium (0.09 g,
0.06 eq, 0.079 mmol) were added. The mixture was then refluxed for 24
h at 60 oC, maintaining inert conditions. Then, the reaction was allowed
to cool down to room temperature and diluted with water and ether. The
product was extracted to ether, washed with brine, dried over MgSO4,
and dried in vacuo. Purification by silica gel column chromatography
o
was obtained as a white powder. M.p. 122 C; Rf=0.5 (CH2Cl2/hexane
1
1/1); H NMR (300 MHz, CDCl3): δ 8.16 – 8.02 (m, 1H, Hh), 8.00 – 7.82
(m, 3H, ArH), 7.53 (s, 1H, Hk), 7.48 – 7.36 (m, 4H, ArH), 2.55 (s, 3H,
CH3). 13C NMR (75 MHz, CDCl3): δ 162.79, 147.71, 140.16, 138.62,
135.73, 132.38, 130.93, 130.43, 129.25, 129.21, 127.59, 125.92, 124.72,
124.54, 124.34, 122.65, 12.81 ppm; HRMS (ESI): m/z calcd for
C18H19ClNS2 [M+H]+: 342.0178; found: 342.0175.
Synthesis of 2-(4-chlorophenyl)-5-methyl-4-(2-phenylthiazol-4-yl)-
thiazole (17)
o
(CH2Cl2/hexane 1/1) afforded 0.42 g (96 %) of 13. M.p. 140 C; Rf=0.15
17 was prepared by the same procedure as 13 except that 4-bromo-5-
methyl-2-phenylthiazole was replaced by 11. From 11 (0.4 g, 1.2 eq, 1.39
mmol), n-butyllithium (0.6 mL 2.5 M, 1.3 eq, 1.50 mmol), trimethyl borate
(0.91 g, 7.5 eq, 8.66 mmol), potassium phosphate (0.736 g, 3 eq, 3.47
mmol), tetrakis(triphenylphosphine)palladium (0.08 g, 0.06 eq, 0.069
mmol), 5.8 mL THF, 2.89 mL DME, and 2.89 mL water, 97 mg (23 %) of
16 was obtained. M.p. 188 oC; Rf=0.58 (CH2Cl2/hexane 1/1); 1H NMR
(300 MHz, CDCl3): δ 8.08 – 8.01 (m, 2H, ArH), 7.95 – 7.87 (m, 3H, ArH),
7.51 – 7.39 (m, 5H, ArH), 2.98 (s, 3H,tert-Bu). 13C NMR (75 MHz,
CDCl3): δ 167.68, 162.56, 152.52, 145.96, 135.73, 133.91, 132.32,
131.61, 130.84, 130.14, 129.19, 129.07, 127.63, 126.61, 116.74, 13.29
ppm; HRMS (EI): m/z calcd for C19H13ClN2S2 [M]+: 368.0208; found:
368.0208.
(CH2Cl2/hexane 3/7); 1H NMR (300 MHz, CDCl3): δ 8.08 – 8.01 (m, 2H,
ArH), 8.00 – 7.95 (m, 2H, ArH), 7.92 (s, 1H, m), 7.45 (m, 6H, ArH), 2.99
(s, 3H, CH3); 13C NMR (75 MHz, CDCl3): δ 167.55, 163.94, 152.73,
145.78, 133.96, 133.83, 131.26, 130.08, 129.84, 129.05, 128.98, 126.60,
126.45, 116.64, 13.28 ppm; HRMS (EI): m/z calcd for C19H14N2S2 [M]+:
334.0598; found: 334.0573.
Synthesis of 4-(benzothiophen-3-yl)-2-(3,5-di-tert-butylphenyl)-5-
methylthiazole (14)
14 was prepared by the same procedure as 12 except that 4-bromo-5-
methyl-phenylthiazole was replaced by 8. From 8 (0.24 g, 1.0 eq, 0.64
mmol), benzothiophene-3-boronic acid (0.17 g, 1.5 eq, 0.96 mmol),
potassium
phosphate
(0.21
g,
1.5
eq,
0.96
mmol),
Synthesis of 2,3-bis(5-methyl-2-phenylthiazol-4-yl)benzothiophene
(1o)
tetrakis(triphenylphosphine)palladium (0.02 g, 0.03 eq, 0.019 mmol,
3mol%) in 12 mL DME and 6 mL water, 0.25 mg (94 % yield) of 13 was
o
obtained as a white powder. M.p. 130 C; Rf=0.63 (CH2Cl2/hexane 1:1);
A
dry Schlenck flask was charged with 4-bromo-5-methyl-2-
1H NMR (300 MHz, CDCl3) δ 8.16 – 8.07 (m, 1H, Hh), 7.94 – 7.88 (m, 1H,
phenylthiazole[6b] (0.099 g, 1.2 eq, 0.3903 mmol), 12 (0.1 g, 1 eq, 0.33
Hj), 7.82 (d, J = 1.8 Hz, 2H, Ha), 7.54 (s, 1H, Hk), 7.51 (t, J = 1.8 Hz, 1H,
mmol), pivalic acid (0.01 g, 0.3 eq, 0.1 mmol), cesium carbonate (0.21 g,
Hb), 7.43 – 7.37 (m, 2H, Hl), 2.55 (s, 3H, CH3), 1.38 (s, 18H, tert-Bu). 13
C
2
eq, 0.65 mmol), di-tert-butylmethylphosphonium tetrafluoroborate
NMR (125 MHz, CDCl3): δ 165.37, 151.63, 147.31, 140.19, 138.83,
133.38, 130.93, 130.13, 130.10, 125.73, 124.62, 124.41, 124.27, 122.60,
120.92, 35.13, 31.57, 12.83 ppm; HRMS (ESI): m/z calcd for C26H30NS2
[M+H]+: 420.1820; found: 420.1808.
(0.008 g, 0.1 eq, 0.03 mmol), palladium acetate (0.004 g, 0.03 mmol,
0.05 eq, 5mol%), and 1 mL mesitylene under Ar atmosphere. The
solution was refluxed at 140 oC overnight and then allowed to cool at
room temperature. The solution was then diluted with dichloromethane
and ethyl acetate and filtered by celite. The solvents were removed and
the crude product was purified by silica gel column chromatography (1:1
CH2Cl2:hexane) twice, affording 0.11 g of product (71 % yield). All
characterization agreed fully with literature.[5b]
Synthesis of 2-(3,5-di-tert-butylphenyl)-5-methyl-4-(2-phenylthiazol-
4-yl)-thiazole (15)
15 was prepared by the same procedure as 13 except that 4-bromo-5-
methyl-2-phenylthiazole was replaced by 8. From 8 (0.262 g, 1 eq, 1.09
mmol), n-butyllithium (0.57 mL 2.5 M, 1.3 eq, 1.42 mmol), trimethyl
borate (0.86 g, 7.5 eq, 8.19 mmol), potassium phosphate (0.695 g, 3eq,
3.28 mmol), tetrakis(triphenylphosphine)palladium (0.076 g, 0.06 eq,
0.066 mmol, 6mol%), 5.5 mL THF, 2.7 mL DME, and 2.7 mL water, 0.267
mg (55 %) of 15 was obtained. M.p. 176 oC; Rf=0.15 (CH2Cl2/hexane
3/7); 1H NMR (300 MHz, CDCl3): δ 8.11 – 8.01 (m, 2H, ArH), 7.97 (s, 1H,
Hm), 7.81 (d, J = 1.8 Hz, 2H, Ha), 7.54 – 7.50 (t, J = 1.8 Hz, 1H, Hb), 7.49
– 7.41 (m, 3H, ArH), 3.00 (s, 3H, CH3), 1.41 (s, 18H,tert-Bu). 13C NMR
(75 MHz, CDCl3:) δ 167.47, 165.22, 159.80, 152.87, 151.55, 145.58,
134.02, 133.22, 130.83, 130.04, 129.04, 126.59, 124.27, 120.93, 120.77,
116.68, 35.11, 31.58, 13.26 ppm; HRMS (EI): m/z calcd for C27H30N2S2
[M]+: 446.1850; found: 446.1853.
Synthesis of 4,5-bis(5-methyl-2-phenylthiazol-4-yl)-2-phenylthiazole
(2o)
2o was synthesized by the same procedure as 1o to couple 13 and 4-
bromo-5-methyl-2-phenylthiazole[6b] by C-H direct arylation. From 13
(0.18 g, 1 eq, 0.54 mmol), 4-bromo-5-methyl-2-phenylthiazole (0.16 g,
1.2 eq, 0.65 mmol), pivalic acid (0.017 g, 0.3 eq, 0.16 mmol), cesium
carbonate (0.35 g, 2.0 eq, 1.1 mmol), di-tert-butylmethylphosphonium
tetrafluoroborate (0.014 g, 0.1 eq, 0.054 mmol), palladium acetate
(0.006g, 0.05 eq, 0.027 mmol), and 1.7 mL mesitylene, 2o (0.070 g,
26%) was obtained as a white powder. All characterization agreed fully
with literature. [5b]
8
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201601657
European Journal of Organic Chemistry
FULL PAPER
Synthesis of 2,3-bis(5-methyl-2-[3,5-di-tert-butylphenyl]thiazol-4-
yl)benzothiophene (3o)
mmol), cesium carbonate (0.178 g, 2.0 eq, 0.54 mmol), di-tert-
butylmethylphosphonium tetrafluoroborate (0.007 g, 0.1 eq, 0.027 mmol),
palladium acetate (0.003g, 0.05 eq, 0.027 mmol), and 0.9 mL mesitylene,
5o (0.046 g, 31%) was obtained as a white powder. M.p. Rf=0.41
(CH2Cl2/hexane 1/1) 1H NMR (300 MHz, CDCl3): δ 8.08 – 8.03 (m, 2H,
ArH), 7.91 – 7.82 (m, AA’XX’, 2H, ArH), 7.74 – 7.64 (m, AA’XX’, 2H, ArH),
7.50 – 7.45 (m, 3H, ArH), 7.43 – 7.36 (m, AA’XX’, 2H, ArH), 7.33 – 7.27
(m, AA’XX’, 2H, ArH), 2.58 (s, 3H, CH3), 2.13 (s, 3H, CH3); 13C NMR (75
MHz, CDCl3): δ 167.36, 162.94, 162.53, 147.84, 146.55, 144.07, 135.99,
135.66, 133.62, 133.59, 132.83, 132.26, 132.07, 130.42, 129.28, 129.10,
128.82, 128.61, 127.66, 127.53, 126.74, 12.95, 12.50 ppm; HRMS (EI):
m/z calcd for C29H19N3Cl2S3 [M]+: 575.0018; found: 575.0107.
3o was synthesized by the same procedure as 1o to couple 14 and 8 by
C-H direct arylation. From 14 (0.25 g, 1.0 eq, 0.59 mmol), 8 (0.24 g, 1.1
eq, 0.65 mmol), pivalic acid (0.018 g, 0.3 eq, 0.18 mmol), cesium
carbonate (0.39 g, 2 eq, 1.18 mmol), di-tert-butylmethylphosphonium
tetrafluoroborate (0.014 g, 0.1 eq, 0.05 mmol), palladium acetate (0.06 g,
0.03 mmol, 0.05 eq, 5 mol%), and 2 mL mesitylene, 0.384 g of product
(92 % yield) was afforded. Rf=0.35 (CH2Cl2/hexane 1/1) 1H NMR (300
MHz, CDCl3): δ 7.97 – 7.86 (m, 4H, ArH), 7.83 (d, J = 1.8 Hz, 2H, Ha),
7.74 (d, J = 1.8 Hz, 2H, Ha’), 7.51 (t, J = 1.8 Hz, 1H, Hb), 7.49 (t, J = 1.8
Hz, 1H, Hb’), 7.47 – 7.33 (m, 3H, ArH), 2.14 (s, 3H, CH3), 2.06 (s, 3H,
CH3), 1.38 (s, 18H,tert-Bu), 1.36 ppm (s, 18H,tert-Bu); 13C NMR (75 MHz,
CDCl3): δ 165.61, 165.55, 151.64, 151.58, 147.11, 145.80, 140.16,
139.92, 136.12, 133.43, 133.05, 131.54, 131.45, 128.69, 124.95, 124.56,
124.47, 124.36, 124.23, 122.07, 120.90, 120.84, 35.12, 35.08, 31.57,
31.54, 12.43, 12.28 ppm; HRMS (EI): m/z calcd for C44H53N2S3 [M+H]+:
705.3371; found: 705.3370.
DFT Calculations
[20]
All DFT calculations were carried out using Gaussian 09.
The
functional B3LYP with basis set 6-31G(d,p) was employed. For the
geometry optimization of 1o, 2o, 4o, and 5o, the reactive anti-parallel
conformation with C2 symmetry around the reactive hexatriene system
was used as initial input for the structure. All geometry optimizations of
the open and closed forms were carried out. TD-DFT calculations set to
solve up to N = 10 states were used to predict the UV-Visible spectrum
both of the open and closed forms of the molecules. Torsion angle scans
around the dihedral angle between the benzothiophene moiety and the
thiazole ring attached to the 3-position were calculated every 20o rotation
from the optimized structure of 1o. Similar scan was done between the
central thiazole ring and the thiazole ring attached to the 5-position for 4o.
Then, the conformations with minimal energy along the scan were further
optimized by the same method as above. All calculations were set in
vacuum.
Synthesis of 4,5-bis(5-methyl-2-[3,5-di-tert-butylphenyl]thiazol-4-yl)-
2-phenylthiazole (4o)
4o was synthesized by the same procedure as 1o to couple 15 and 11 by
C-H direct arylation. From 15 (0.30 g, 1.2 eq, 0.806 mmol), 11 (0.30 g,
1.0 eq, 0.67 mmol), pivalic acid (0.021 g, 0.3 eq, 0.21 mmol), cesium
carbonate (0.44 g, 2.0 eq, 1.3 mmol), di-tert-butylmethylphosphonium
tetrafluoroborate (0.017 g, 0.1 eq, 0.034 mmol), palladium acetate
(0.008g, 0.05 eq, 0.034 mmol), and 2.2 mL mesitylene, 4o (0.359 g,
1
66%) was obtained as a white powder. Rf=0.43 (CH2Cl2/hexane 1/1) H
NMR (300 MHz, CDCl3): δ 8.12 – 8.05 (m, 2H, ArH), 7.72 (d, J = 1.8 Hz,
2H, Ha), 7.59 (d, J = 1.8 Hz, 2H, Ha’), 7.53 – 7.43 (m, 4H), 7.40 (t, J = 1.8
Hz, 1H, Hb’), 2.65 (s, 3H, CH3), 2.17 (s, 3H, CH3), 1.34 (s, 18H,tert-Bu),
1.27 (s, 18H,tert-Bu).; 13C NMR (75 MHz, CDCl3): δ 166.78, 165.28,
164.86, 151.46, 151.32, 148.21, 146.29, 143.70, 133.86, 133.20, 133.01,
132.63, 131.81, 130.19, 129.42, 129.04, 126.71, 124.26, 124.01, 120.91,
120.81, 35.06, 34.99, 31.54, 31.49, 12.97, 12.50 ppm; HRMS (EI): m/z
calcd for C45H53N3S3 [M]+: 731.3402; found: 731.3406.
STM Experiments
STM experiments were performed in an Omicron LT-UHV with a base
pressure of 1.5 x 10-11 mbar. The Ag(111) crystal was prepared by
repeated sputtering of Ar+ ions and annealing at 800 K. Molecules were
first deposited on a 0.15 mm diameter filament and then subjected to
high temperature outgassing before sublimating on the surface. The
molecules were evaporated at ~200°C on the substrate held at 4.5 K.
Imaging conditions were carried out at 4.5 K and constant current mode.
Synthesis
of
2,3-bis(5-methyl-2-[4-chlorophenyl]thiazol-4-
yl)benzothiophene (5o)
5o was synthesized by the same procedure as 1o to couple 13 and 8 by
C-H direct arylation. From 13 (0.20 g, 1.0 eq, 0.69 mmol), 8 (0.24 g, 1.0
eq, 0.69 mmol), pivalic acid (0.021 g, 0.3 eq, 0.21 mmol), cesium
carbonate (0.46 g, 2.0 eq, 1.39 mmol), di-tert-butylmethylphosphonium
tetrafluoroborate (0.018 g, 0.1 eq, 0.069 mmol), palladium acetate (0.008
g, 0.05 eq, 0.035 mmol), and 2.31 mL mesitylene, 2o (0.080 g, 20 %
Acknowledgements
This work was supported by the MEXT Program for Promoting
the Enhancement of Research Universities in NAIST, the CNRS
the University Paul Sabatier (Toulouse) and the Programme
Investissements d’Avenir ANR-11-IDEX-0002-02, reference
ANR-10-LABX-0037-NEX. We would like to thank Dr Sébastien
Gauthier for fruitful discussions. This research was also partly
supported by the JSPS KAKENHI Grant Number JP26107006 in
Scientific Research on Innovative Areas “Photosynergetics”.
yield) was obtained as
a white powder. M.p. 171°C; Rf=0.50
(CHCl3/hexane 1/1) 1H NMR (300 MHz, CDCl3): δ = 7.93-7.82 (m, 6 H,
ArH), 7.43-7.37 (m, 6H, ArH), 2.07 (s, 3H, CH3), 2.03 (s, 3H, CH3); 13C
NMR (75 MHz, CDCl3): δ 163.33, 163.31, 147.42, 146.07, 139.93,
139.88, 135.99, 135.87, 135.86, 132.39, 132.38, 132.04, 129.28, 129.24,
128.40, 127.65, 127.61, 125.25, 124.74, 124.28, 122.21, 12.37, 12.24
ppm; HRMS (EI): m/z calcd for C28H18Cl2N2S3 [M]+: 548.0008; found:
548.0003.
Keywords: • photochromism • terarylenes • molecule-surface
interactions • DFT • STM
Synthesis
of
4,5-bis(5-methyl-2-[4-chlorophenyl]thiazol-4-yl)-2-
phenylthiazole (6o)
[1] a) S. Marqués-González, P.J. Low, Aust. J. Chem. 2008, 3, 244-253; b)
J.M. van Ruitenbeek in Single-Molecule Electronics: An Introduction to
Synthesis, Measurement and Theory (Ed.: M. Kiguchi), Springer,
Singapore, 2016, pp. 1-23; c) M.S. Dresselhaus, Nat. Rev. Mater., 2016,
1, 1-2.
6o was synthesized by the same procedure as 1o to couple 16 and 8
were coupled by C-H direct arylation. From 16 (0.12 g, 1.5 eq, 0.41
mmol), 8 (0.1 g, 1.0 eq, 0.27 mmol), pivalic acid (0.008 g, 0.3 eq, 0.081
9
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201601657
European Journal of Organic Chemistry
FULL PAPER
[2] a) Molecular Switches (Ed. B. L. Feringa), Wiley-VCH, Weinheim, 2001;
b) S. Shinkai, Pure Appl. Chem., 1987, 59, 425-430; c) Y. Hasegawa, T.
Nakagawa, T. Kawai., Coord. Chem. Rev. 2010, 254, 2643-2651.
[3] a) M.Irie, M. Mohri, J. Org. Chem., 1988, 53, 803-808; b) M. Irie, Chem.
Rev., 2000, 100, 1685–1716; c) M. Irie, T. Fukaminato, K. Matsuda, S.
Kobatake, Chem. Rev. 2014, 114, 12174-12277; d) New Frontiers in
Photochromism (Eds. M. Irie, Y. Yokoyama, T. Seki), Springer, Tokyo,
2013; e) G. Szaloki, J.L.Pozzo, Chem. Eur. J. 2013, 19, 11124-11132
[4] a) M. M. Krayushkin, S. N. Ivanov, A. Yu. Martynkin, B. V. Lichitsky, A. A.
Dudinov and B. M. Uzhinov, Russ. Chem. Bull., Int. Ed. 2001, 50, 116-
122; b) H. Tian, B. Chen, H. Tu, K. Müllen, Adv. Mater. 2002, 14, 918-
923; c) V. W.-W. Yam, C.-C. Ko, N. Zhu, J. Am. Chem. Soc. 2004, 126,
12734-12735. d) M. M. Krayushkin, M. A. Kalik, V. A. Migulin, Russ.
Chem. Rev. 2009, 78, 329-336.
Bonacchi, M. El Garah, A. Ciesielski, M. Herder, S. Conti, M. Cecchini, S.
Hecht, P. Samori, Angew. Chem. Int. Ed. 2015, 54, 4865-4869; f) N.
Maeda, T. Hirose, S. Yokoyama, K. M. Matsuda, J. Phys. Chem. C, 2016,
120, 9317-9325.
[14] a) S. V. Snegir, A. Marchenko, P. Yu, F. Maurel, O. Kapitanchuk, S.
Mazerat, M. Lepeltier, A. Léaustic, E. Lacaze, J. Phys. Chem. Lett. 2011,
2, 2433-2436; b) S. V. Snegir, P. Yu, F. Maurel, O. Kapitanchuk, A.
Marchenko, E. Lacaze, Langmuir 2014, 30, 13556-13563.
[15] a) J. K. Gimzewski, C. Joachim, R. R. Schlittler, V. Langlais, H. Tang, I.
Johannsen, Science 1998, 281, 531-533; b) M. Schunack, L. Petersen, A.
Kuhnle, E. Laegsgaard, I. Stensgaard, I. Johannsen, F. Besenbacher,
Phys. Rev. Lett. 2001, 86, 456-459; c) F. Chiaravalloti, L. Gross, K. H.
Rieder, S. Stojkovic, A. Gourdon, C. Joachim, F. Moresco, Nat. Mater.
2007, 6, 30-33; d) H.P. Jacquot de Rouville, R. Garbage, R.E. Cook, A.R.
Pujol, A.M. Sirven, G. Rapenne, Chem. Eur. J. 2012, 18, 3023-3031; e)
M. Alemani, M. V. Peters, S. Hecht, K. H. Rieder, F. Moresco, L. Grill, J.
Am. Chem. Soc. 2006, 128, 14446-14447; f) M. Alemani, S. Selvanathan,
F. Ample, M. V. Peters, K. H. Rieder, F. Moresco, C. Joachim, S. Hecht, L.
Grill, J. Phys. Chem. C 2008, 112, 10509-10514.
[5] a) T. Kawai, T, Iseda, M. Irie, Chem. Commun. 2004, 72-73; b) T.
Nakashima, K. Atsumi, S. Kawai, T. Nakagawa, Y. Hasegawa, T. Kawai,
Eur. J. Org. Chem. 2007, 19, 3212-3218.
[6] a) S. Fukumoto, T. Nakashima, T. Kawai, Eur. J. Org. Chem. 2011, 26,
5047-5053; b) S. Fukumoto, T. Nakashima, T. Kawai, Angew. Chem. Int.
Ed. 2011, 50, 1565-1568; c) R. Li, T. Nakashima, O. Galangau, S. Iijima,
R. Kanazawa, T. Kawai, Chem. Asian J. 2015, 10, 1725-1730.
[7] a) T. Nakashima, Y. Kajiki, S. Fukumoto, M. Taguchi, S. Nagao, S. Hirota,
T. Kawai, J. Am. Chem. Soc. 2012, 134, 19877-19883; b) J.P. Calupitan,
T. Nakashima, Y. Hashimoto, T. Kawai., Chem. Eur. J. 2016, 22, 10002-
10008.
[16] a) B. Baris, V. Luzet, E. Duverger, P. Sonnet, F. Palmino, F. Cherioux,
Angew. Chem. Int. Ed. 2011, 50, 4094-4098; b) Y. Makoudi, B. Baris, J.
Jeannoutot, F. Palmino, B. Grandidier, F. Cherioux, Chem. Phys. Phys.
Chem. 2013, 14, 900-904.
[17] a) M. Begtrup, L. B. L. Hansen, Acta Chem. Scand. 1992, 46, 372-383;
b) G. Gavrel, P. Yu, A. Leaustic, R. Guillot, R. Metivier, K. Nakatani,
Chem. Commun. 2012, 48, 10111-10113.
[8] R. Kanazawa, M. Taguchi, T. Nakashima, T. Kawai, New J. Chem., 2015,
39, 7397-7402.
[18] a) T. Nakashima, K. Imamura, K. Yamamoto, Y. Kimura, S. Katao, Y.
Hashimoto, T. Kawai, Chem. Eur. J. 2014, 20, 13722-13729; b) O.
Galangau, T. Nakashima, F. Maurel, T. Kawai, Chem. Eur. J. 2015, 21,
8471-8482.
[9] Y. Hashimoto, T. Nakashima, D. Shimizu, T. Kawai, Chem. Commun,
2016, 52, 5171-5174.
[10] a) H. Nakagawa, S. Kawai, T. Nakashima, T. Kawai., Org. Lett. 2008, 11,
1475-1478. b) T. Nakashima, K. Tsuchie, R. Kanazawa, R. Li, S. Iijima, O.
Galangau, H. Nakagawa, K. Mutoh, Y. Kobayashi, J. Abe, T. Kawai, J.
Am. Chem. Soc. 2015, 137, 7023-7026; c) R. Li, T. Nakashima, R.
Kanazawa, O. Galangau, T. Kawai, Chem. Eur. J. 2016, 22, 16250-
16257; d) O. Galangau, S. Delbaere, N. Ratel-Ramond, G. Rapenne, R.
Li, J.P.D.C. Calupitan, T. Nakashima, T. Kawai, J. Org. Chem. 2016, 81,
11282-11290.
[19] a) M. Takeshita, C. N. Choi, M. Irie, Chem. Commun. 1997, 2265-2266; b)
K. Matsuda, Y. Shinkai, T. Yamaguchi, K. Nomiyama, M. Isayama, M. Irie,
Chem. Lett. 2003, 32, 1178-1179; c) M. Takeshita, M. Nagai, T. Yamato,
Chem. Commun. 2003, 1496-1497; d) T. Yamaguchi, M. Irie, J. Photoch.
Photobio. A 2006, 178, 162-169.
[20] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J.
R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H.
Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G.
Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T.
Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J.
Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J.
Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J.
Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V.
Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09 Revision A.02, Gaussian Inc.
[11] N. Battaglini, H. Klein, C. Hortholary, C. Courdret, F. Maurel, P. Dumas,
Ultramicrosocpy, 2007, 107, 958-962.
[12] a) C. Coudret, G. Guirado, N. Estrampes, R. Coratger, Phys. Chem.
Chem. Phys. 2011, 13, 20946-20953; b) Arramel, T.C. Pijper, T.
Kudermac, M. van der Maas, B.L. Feringa, B.J. van Wees, J. Appl. Phys.,
2012, 111, 083716; c) T. K. Shimizu, J. Jung, H. Imada, Y. Kim, Angew.
Chem. Int. Ed. 2014, 53, 13729-13733; d) J. Wirth, N. Hatter, R. Drost, T.
R. Umbach, S. Barja, M. Zastrow, K. Rꢀck-Braun, J. I. Pascual, P.
Saalfrank, K. J. Franke, J. Phys. Chem. C, 2015, 119, 4874-4883.
[13] a) D. Frath, T. Sakano, Y. Imaizumi, S. Yokoyama, T. Hirose, K. Matsuda,
Chem. Eur. J. 2015, 21, 11350-11358; b) S. Yokoyama, T. Hirose, K.
Matsuda, Chem. Eur. J. 2015, 21, 13569-13576; c) S. Yokoyama, T.
Hirose, K. Matsuda, Chem. Comm. 2014, 50, 5964-5966; d) T. Sakano, Y.
Imaizumi, T. Hirose, K. Matsuda, Chem. Lett. 2013, 42, 1537-1539; e) S.
Wallingford
CT,
2009.
10
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201601657
European Journal of Organic Chemistry
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Without tert-butyl
With tert-butyl
Key Topic: terarylenes photochromes
J.P.D.C. Calupitan, O. Galangau, O.
Guillermet, R. Coratger, T. Nakashima, G.
Rapenne*, T. Kawai*
Page No. – Page No.
Tert-butyl functionalized terarylenes have been synthesized to improve the contrast
in the STM images as well as to minimize aggregation and preserve photochromic
properties. Analogues with peripheric chlorine centers has also been synthesized to
study such systems on NaCl or KCl / metal surfaces.
Synthesis and photochromism of
chloro and tert-butyl terarylene
derivatives for surface deposition
11
This article is protected by copyright. All rights reserved.