Dihydroazulene photoswitch operating in sequential tunneling regime: Synthesis and single-molecule junction studies

Article abstract of DOI:10.1002/adfm.201200897

Molecular switches play a central role for the development of molecular electronics. In this work it is demonstrated that the reproducibility and robustness of a single-molecule dihydroazulene (DHA)/vinylheptafulvene (VHF) switch can be remarkably enhance

Full text of DOI:10.1002/adfm.201200897

www.afm-journal.de
www.MaterialsViews.com
Dihydroazulene Photoswitch operating in Sequential
Tunneling Regime: Synthesis and Single-Molecule
Junction Studies
Søren Lindbæk Broman, Samuel Lara-Avila, Christine Lindbjerg Thisted,
Andrew D. Bond, Sergey Kubatkin, Andrey Danilov,* and Mogens Brøndsted Nielsen*
Light provides an attractive external stim-
ulus for triggering molecular conductivity
changes, and one of the long standing
challenges in the field is to build reli-
able single-molecule photoswitch. While
for voltage-controlled molecular switches
durable operation involving more than
105 ON-OFF cycles has already been dem-
onstrated,[2] none of the single-molecule
photoswitches can compete with Com-
plementary Metal–Oxide–Semiconductor
(CMOS) devices. There are several exam-
ples[3] of reversible switching of photo-
sensitive dithienylethene and azobenzene
molecules placed in a solid state device,
like a nanogap between planar electrodes,
but in some cases quenching of the
excited state is observed or the molecule
stopped responding to light after a few
transitions if they could rearrange on the
surface. Theoretical studies have indicated
that a strong coupling is likely responsible
for the quenching of molecular states.[4]
This idea was supported experimentally
by reducing the coupling between the
Molecular switches play a central role for the development of molecular elec-
tronics. In this work it is demonstrated that the reproducibility and robust-
ness of a single-molecule dihydroazulene (DHA)/ vinylheptafulvene (VHF)
switch can be remarkably enhanced if the switching kernel is weakly coupled
to electrodes so that the electron transport goes by sequential tunneling. To
assure weak coupling, the DHA switching kernel is modified by incorporating
p-MeSC6H4 end-groups. Molecules are prepared by Suzuki cross-couplings on
suitable halogenated derivatives of DHA. The synthesis presents an expan-
sion of our previously reported bromination–elimination–cross-coupling
protocol for functionalization of the DHA core. For all new derivatives the
kinetics of DHA/VHF transition has been thoroughly studied in solution. The
kinetics reveals the effect of sulfur end-groups on the thermal ring-closure of
VHF. One derivative, incorporating a p-MeSC6H4 anchoring group in one end,
has been placed in a silver nanogap. Conductance measurements justify that
transport through both DHA (high resistivity) and VHF (low resistivity) forms
goes by sequential tunneling. The switching is fairly reversible and reenter-
able; after more than 20 “ON-OFF” switchings, both DHA and VHF forms are
still recognizable, albeit noticeably different from the original states.
electrode and the molecule, simply by moving the anchoring
group (thiol) to the meta position of the aryl unit linked to the
photoswitch.[3b,g] Along the same line, in our recent study on
the dihydroazulene (DHA)/vinylheptafulvene (VHF) photo-/
thermoswitch,[5] we argued that the operation of a photoswitch
in the coherent transport regime, implying strong coupling of
molecular orbitals to electrodes, should result in a short resi-
dence time for the photo-excited electron. Thus, only a small
fraction of excited electrons reside on the molecule long
enough time to trigger the conversion. Moreover, a tiny vari-
ation in contact geometry can affect coupling and therefore
photosensitivity.
With the objective to expand the number of useful pho-
toswitches for molecular electronics, each with their unique
properties and advantages, we have focused attention on the
DHA/ VHF system. Scheme 1 shows the switching events
of the 2-phenyl-substituted derivatives. DHA 1a undergoes
a light-induced ring-opening reaction to generate VHF 1b,
which, in turn, undergoes a thermally-induced ring-closure to
form DHA.[6] We shall in general use letters a and b to denote
a pair of DHA/VHF isomers. Pioneering work by Daub and
1. Introduction
Molecular electronics aims at using tailor-built molecules as
active device elements to achieve a desired electronic function-
ality.[1] Conventional computing operations can be performed by
a single molecule working as a transistor or as a logical switch.
S. L. Broman, C. L. Thisted, Prof. M. B. Nielsen
Department of Chemistry
University of Copenhagen
Universitetsparken 5, DK-2100 Copenhagen Ø, Denmark
E-mail: mbn@kiku.dk
S. Lara-Avila, Prof. S. Kubatkin, Dr. A. Danilov
Department of Microtechnology and Nanoscience
Chalmers University of Technology
Kemivägen 9, S-41296 Göteborg, Sweden
E-mail: andrey.danilov@chalmers.se
Prof. A. D. Bond
Department of Physics, Chemistry and Pharmacy
University of Southern Denmark
Campusvej 55, DK-5230 Odense M, Denmark
DOI: 10.1002/adfm.201200897
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201200897
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1

www.afm-journal.de
www.MaterialsViews.com
NC
The failure to build reproducible and robust photoswitch
devices operating in a strongly-coupled regime motivated us
to turn to molecular junctions operating in Coulomb blockade
regime. In this regime, transport through the molecule goes
via sequential tunneling:[12] an electron first jumps onto the
molecule, resides there for a time which is long in the scale of
molecular relaxation processes and leaves to the other electrode
by the second jump. In this paper we describe the synthesis,
via Suzuki cross-couplings, of new DHA derivatives 3a and
4a (Figure 1) suitable as molecular kernels for photoswitches
operating in weakly coupled regime, achieved by using meth-
ylsulfides as anchoring groups. Thiols are often chosen as
anchoring groups for adhering the molecule between two
gold or silver electrodes.[13] Usually, the thiol is protected with
a readily cleavable acetyl group, which results in a strong cou-
pling to electrodes. A weaker coupling between the electrodes
and the molecule can be achieved by alkylsulfides,[14,15] which
were therefore chosen as anchoring groups in target mol-
ecules 3a and 4a. Additionally, we present measurements on
a single-molecule switch fabricated with one particular deriva-
tive, compound 4a, where transport, as expected, is in the Cou-
lomb blockade regime for both ON and OFF states. We dem-
onstrate that photosensitivity is at least an order of magnitude
higher than for the non-capped DHA 2a and does not degrade
after many ON-OFF transitions. Moreover, we have performed
detailed switching studies in solution revealing the electronic
effect of the MeS group on the thermal ring-closure reaction.
The experimental technique for fabricating the single-mol-
ecule junction relies upon the evaporation of the molecule in
ultra-high vacuum onto a substrate with a prefabricated nanogap
between silver electrodes. In addition to the recent studies
on DHA 2a,[5] the technique has previously been employed
successfully for studying oligo(phenylenevinylene)s[15] and
fullerene-based[16] and anthraquinone switches.[17] The advan-
tage of the technique is that no solution chemistry is involved
during fabrication; the electrodes are deposited and the mol-
ecule is subsequently sublimed and trapped in the nanogap
in the same vacuum cycle. The need, however, to sublime the
molecule without decomposition imposes some constraints
on its size and structure. Early attempts of subliming a DHA
with an SAc anchoring group were thus not successful due to
decomposition.
CN
Ph
NC
CN
Ph
DHA 1a
VHF 1b (s-cis)
8
4
1
3
7
8a
3a
Ph
6
2
CN
5
CN
VHF 1b (s-trans)
Scheme 1. Dihydroazulene (DHA)/ vinylheptafulvene (VHF) photo-/
thermoswitch.
co-workers together with our recent work have previously estab-
lished a good understanding of how the ring-opening and ring-
closure reactions are influenced by electron withdrawing and
donating groups at specific positions,[7,8] acid/ base,[7] redox-
active groups,[9] and Lewis acids.[10] All these various ways of
tuning the switching events make the DHA/VHF system par-
ticularly attractive for molecular electronics. Our first conduc-
tivity studies[5] on the DHA derivative 2a[7a] (Figure 1) in a junc-
tion were promising, but also showed room for improvement.
The ring-opening time for a related DHA was found to be 1.2
ps,[11] which is much longer than the electron residence time
of about ∼40 fs measured for 2a in the single-molecule junc-
tion.[5] As a result, we found that some DHA junctions of 2a
were not photo-sensitive at all, and even those which originally
responded to light, lost any photosensitivity after a few ON-OFF
cycles. This is likely to occur due to a rearrangement of the mol-
ecule-to-electrode interface which, after a few switching events,
reduces to the most favourable and therefore most strongly-
coupled geometry.
NC
CN
H3CS
2a
NC
CN
2. Results and Discussion
SCH3
3a
2.1. Synthesis
NC
CN
We have previously shown that regioselective functionalization of
the seven-membered ring of DHA is possible via a bromination–
elimination–cross-coupling protocol.[7] This protocol has been
used to synthesize molecules like 5a[7b] (Figure 1) and can be
expanded to functionalization of both rings of the system,
starting from a DHA including an iodophenyl group at the five-
membered ring. This is demonstrated by the synthesis of 3a
as shown in Scheme 2. First, bromination of the known DHA
6a[18] using Br2 in CH2Cl2 at –78 °C, as previously reported
for the related 2-phenyl-DHA 1a,[7] gave the dibromide 7. This
SCH3
4a
NC
CN
5a
Figure 1. DHA derivatives.
©
2
wileyonlinelibrary.com
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201200897

www.afm-journal.de
www.MaterialsViews.com
Br
NC
NC
Br
Br
NC
CN
CN
CN
Br2
LiHMDS
I
I
I
CH2Cl2, -78 °C
1 h; > 99%*
THF, 0 - 20 °C
2 h; > 90%*
6a
7
8a
H3CS
H3CS
9
B(OH)2
Pd(PPh3)4
2 equiv KF 2H2O
Br
NC
CN
NC
.
< 5%**
CN
SCH3
SCH3
PhMe / H2O,
90 °C, 24 h
10a
3a
Scheme 2. Synthesis of 3a via a double Suzuki coupling procedure. ∗Yields based on 1H NMR spectroscopic analysis. ∗∗Yield calculated from 6a.
compound was very unstable, though it was possible, however,
to perform an elimination with 1 molar equivalent of lithium
bis(trimethylsilyl)amide (LiHMDS) in THF to furnish the DHA
8a in >90% purity. This compound could not easily be purified
without significant loss of material, and it was therefore used
crude for further reactions. A sample was, however, purified
for characterization purposes. By a Suzuki coupling in a two-
phase water/toluene mixture, it was coupled with 4-(methylm-
ercapto)phenylboronic acid 9 using KF to activate the boronic
acid (while base was discarded to diminish azulene formation
by elimination of HCN[7,19]). It was found that the aryliodide
was more reactive than the vinylbromide. Thus, by both TLC
and NMR spectroscopy, the presence of the intermediate 10a
was confirmed, and it was also isolated by stopping the reac-
tion half-way to completion. A chemoselective functionalization
could be accomplished by performing the reaction at 65–75 °C
with one molar equivalent of the boronic acid, albeit the reac-
tion did not go to completion. When using two molar equiva-
lents of boronic acid at 90 °C, conversion was also observed at
the bromo-position, but the outcome was unfortunately accom-
panied by many byproducts and significant degradation. After
24 h, a simple TLC inspection showed 4 or 5 different photo-
chromic compounds. These consisted of the aforementioned
intermediate 10a, the doubly coupled product 3a, the iodo-
phenyl-DHA 6a, and possibly two other compounds, of which
one has been isolated and identified as 4a. No clear evidence
of better results was observed by simply adding the boronic
acid or the catalyst gradually in portions. The final low yield
of 3a (at best <5%) suffered from the need of purifying it by
repeated chromatography (performed in the dark to avoid light-
induced ring-opening), and it does not reflect the efficiency of
the reaction.
To improve the yield and ease the purification, we turned
to another strategy, in which the p-MeSC6H4 group was first
attached to the five-membered ring of the DHA. Thus, sub-
jecting 6a and the boronic acid 9 to Suzuki conditions gave 4a
in high yield (Scheme 3). The coupling can be accomplished
at room temperature with a long reaction time (7 days), but a
better result was observed by performing the reaction at 90 °C
for 48 h. The two-step bromination-elimination procedure
was then performed. The dibromide 11 and bromide 10a were
prepared without bromination of the thioether. The bromide
10a was used crude for further reactions, but a sample was
purified for characterization. Its structure was also confirmed
by X-ray crystallographic analysis as shown in Figure 2. The
compound was then subjected to a Suzuki coupling with
the boronic acid 9 by heating at 90 °C overnight, which gave
3a in an acceptable yield. After a tedious purification by dry
column vacuum chromatography followed by a final recrystal-
lization from either toluene/ heptanes or neat acetonitrile, 3a
was obtained pure. Crystals were subjected to X-ray crystallog-
raphy (Figure 2).
In addition, oxidation of 4a was possible in two steps using
either one or two molar equivalents of meta-chloroperoxy-
benzoic acid (m-CPBA) in CH2Cl2, giving the sulfoxide 12a
and sulfone 13a (Figure 3), respectively, in excellent yields
(see SI). The structure of the sulfone 13a was confirmed by
X-ray crystallography as shown in Figure 2. Finally, a deriva-
tive without the methylthio substituent was prepared in order
to elucidate the influence of this group in the thermally
induced ring-closure of the corresponding VHF (vide infra).
Thus, the iodo-DHA 6a was coupled to phenyl boronic acid,
which furnished the biphenyl-DHA 14a (Figure 3; see SI for
synthesis).
NC
NC
CN
9, Pd(PPh3)4
1 equiv KF 2H2O
CN
.
Br2
I
SCH3
PhMe / H2O,
CH2Cl2, -78 °C
90 °C, 48 h; 42%**
1 h; > 99%*
6a
4a
Br
NC
Br
Br
NC
9, Pd(PPh3)4
KF 2H2O
CN
CN
.
LiHMDS
3a
SCH3
SCH3
PhMe / H2O,
90 °C, 16 h; 17%***
THF, 0 - 20 °C
2 h; > 90%*
11
10a
Scheme 3. Synthesis of 3a via a two-step coupling procedure. ∗Yields based on 1H NMR spectroscopic analysis. ∗∗Yield of pure compound. The
compound can be isolated with minor impurities in a yield of 80%. ∗∗∗Yield of pure compound calculated from 4a. The compound can be isolated
with minor impurities in a yield of 85%.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201200897
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3

www.afm-journal.de
www.MaterialsViews.com
NC
CN
CN
O
S
CH3
12a
13a
NC
O
S
CH3
O
NC
CN
14a
Figure 3. DHA reference compounds.
maxima correspond to a difference in the HOMO-LUMO gap
of 0.69 eV.
Kinetics data for the thermally induced VHF to DHA ring-
closures are collected in Table 1 based on Arrhenius plots.
For comparison, the data for the VHF of the parent phenyl-
DHA/VHF 1a/ 1b[20] and for the DHA/VHF 5a/ 5b[7b] are also
listed. The ring-closure reaction was investigated at different
temperatures and the rate constants were obtained by fitting
the decrease in the VHF absorption (and increase of DHA
absorption) over time to first-order kinetics. All measured rate
constants including its standard deviations are listed in SI. It
was previously shown by change of solvent polarity[6,20] and by
Hammett linear free energy correlations[7] that the thermal ring-
closure of VHF occurs via a zwitterionic structure (Scheme 4).
In accordance hereto, a VHF containing a mesomerically elec-
tron-donating para-methoxyphenyl substituent at the heptaful-
vene ring underwent faster ring-closure than VHF 1b, while a
VHF in which the methoxy substituent is moved to the oppo-
site end of the molecule underwent ring-closure (t1/2 230 min)
slower than VHF 1b.[7b] A phenyl group alone only has a small
influence as revealed by comparison of VHFs 1b (t1/2 216 min)
and VHF 5b (t1/2 206 min). Nevertheless, we find a remarkable
increase in the half-life of 49 min when proceeding from com-
pound 4b (t1/2 170 min) to 3b (t1/2 219 min). These two com-
pounds only differ by the presence of a p-MeSC6H4 substituent
in the heptafulvene ring in 3b. Altogether these results show
that the p-MeSC6H4 substituent exerts a retarding influence
when placed at this position; thus, the influence of the MeS
substituent is in that respect opposite to the enhancing effect
exerted by a MeO substituent at the same position. This is inter-
esting in view of the fact that the Hammett σp-value for a MeS
para-substituent is 0.00 (value recommended by McDaniel and
Brown[21]). On the other hand, only a minor effect is observed
when attaching the SMe on the biphenyl group at the opposite
Figure 2. Molecular structures (with displacement ellipsoids at 50%
probability for non-H atoms) of 10a (top, crystals were grown from a
mixture of CHCl3, THF, and heptanes), 3a (middle, crystals were grown
from a mixture of CHCl3 and heptanes), and 13a (bottom, crystals were
grown from a mixture of CH2Cl2/ heptanes).
2.2. Solution Studies
Solution Studies. Switching studies on the DHAs 3a, 4a, 6a,
8a, 10a, 12a, 13a, and 14a were followed by UV-vis absorption
spectroscopy in acetonitrile. All compounds were photochromic
and underwent a light-triggered ring-opening to the corre-
sponding VHFs (as E/Z-isomers for those compounds with a
substituent group in the heptafulvene ring; 8a, 10a, 3a), which
exhibited redshifted absorption maxima. Conversion of the
VHFs (b) to DHAs (a) occurred with isosbestic points in the
absorption spectra, but with generation of isomeric mixtures of
6- and 7-substituted DHA derivatives for 8a, 10a, and 3a after
one light-heat cycle in the polar solvent as observed previously[7]
for related compounds (Scheme 4); this isomerization was
confirmed by NMR spectroscopy. All compounds were stable
enough to survive at least 5 full cycles of light and heat, without
any sign of decomposition. Absorption maxima for the DHAs
and VHFs are listed in Table 1, and in Figure 4 the spectra of
DHAs 3a and 4a and their corresponding VHFs 3b and 4b are
shown. For DHA 4a and VHF 4b, the differences in absorption
R
R
R
NC
R
NC
NC
CN
CN
CN
NC
CN
R
R
R
R
R
NC
R
CN
Scheme 4. Proposed mechanism of a heat induced ring-closure of the substituted DHAs resulting in isomerization of DHAs (in MeCN).
©
4
wileyonlinelibrary.com
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201200897

www.afm-journal.de
www.MaterialsViews.com
T a b le 1 . Measured absorption maxima for the DHAs and VHFs and kinetics data for the thermal back-reaction (VHF → DHA). Molar absorptivities
of DHAs and VHFs are included in the experimental section. Wavelengths in brackets correspond to the absorption maxima of the mixture of 6- and
7-substituted DHAs obtained after one light-heat cycle. The half-life and rate constant are calculated from the Arrhenius plot.
DHA (a)/ VHF (b)
k25 °C
[s−1
t25 °C
[min]
Ea
A
[s−1
λDHA
[nm]
λVHF
[nm]
]
[kJ mol−1
]
]
1a/ 1ba)
6a/6b
353
361
470
476
468
475
462
504
476
490
477
477
216
152
708
170
897
219
182
206
155
119
5.36 × 10−5
7.62 × 10−5
1.63 × 10−5
6.81 × 10−5
1.29 × 10−5
5.26 × 10−5
6.35 × 10−5
5.6 × 10−5
91.8 0.2
91.8 0.5
94.1 0.4
92.5 0.3
95.0 0.4
94.1 0.4
92.2 0.2
89.9 0.2
91.6 0.4
89.8 0.9
6.35 × 1011
9.31 × 1011
5.07 × 1011
11.0 × 1011
5.74 × 1011
16.0 × 1011
8.87 × 1011
3.2 × 1011
8a/8b
362 (362)
376
4a/4b
10a/10b
3a/3b
380 (380)
376 (382)
368
14a/14b
5a/5bb)
13a/13b
12a/12b
355 (364)
369
7.43 × 10−5
9.71 × 10−5
8.38 × 1011
5.30 × 1011
370
a)Data from Ref. [20] From the exponential decay of 1b at 25 °C, k25 °C = 5.31 × 10−5 s−1 (t = 218 min) is obtained; b)Data from Ref. [7b].
½
end of the molecule as revealed by comparing compounds 14a
and 4a, which is not unreasonable owing to the non-planarity of
the biphenyl. The kinetics data listed in Table 1 also signal the
influence of an inductively electron-withdrawing bromo-sub-
stituent in the seven-membered ring. Thus, the presence of Br
in VHF 10b results in a significantly retarded ring-closure (t1/2
897 min) relative to VHF 4b (t1/2 170 min). Comparison of the
iodophenyl derivatives 8b and 6b reveals a similar rate-retarding
effect of the bromo-substituent, t1/2 708 min vs. 152 min. This
influence is in agreement with the involvement of a zwitteri-
onic structure in the ring-closure reaction as mentioned above.
The prefabricated nanogap consisted of the source and drain
electrodes of silver deposited onto a planar gate electrode made
of aluminum metal covered with aluminum oxide that was
prepared on a chip of oxidized silicon. DHA 3a with two MeS
anchoring groups could not be sublimed without decomposi-
tion (decomposition point 150 °C) so the studies were focused
instead on 4a. Conductance measurements were performed at
4–25 K as a function of the gate voltage, irradiation, bias poten-
tial between source and drain, and temperature. A schematic
representation of the sample geometry is shown in Figure 5. It
is obviously impossible to say how the molecule exactly binds,
but more importantly, it is found to be weakly coupled to one
electrode (vide infra).
The transport measurements reveal that DHA 4a in a silver
nanogap can exist in two distinct forms–A and B. First of all,
the transport through the original form A was characterized by
measuring the stability plot (a set of differential conductance
curves taken at different gate voltages), presented in Figure 6.
The stability diagram reveals a set of characteristic Coulomb
blockade diamonds, which implies that the molecule is weakly
coupled to electrodes and the transport through the molecule
goes by sequential tunneling.[12a] For low bias voltages (below
100 mV), the stability plot is stable and reproducible. The
2.3. Conductance Measurements
A single-molecule junction based on DHA 4a with one MeS
anchoring group was prepared by evaporating the molecule
under ultra-high vacuum onto a substrate with a prefabricated
nanogap between silver electrodes, following the fabrication
procedure described previously (including detailed nanogap
characterization without the presence of molecules).[15–17,22]
30
25
20
15
10
5
DHA 4a
DHA 3a
VHF 4b
VHF 3b
0
200
300
400
500
600
700
800
Wavelength (nm)
Figure 4. UV-Vis absorption spectra of DHA 3a and its corresponding
VHF 3b (red) and DHA 4a and its corresponding VHFs 4b (black) meas-
ured in MeCN.
Figure 5. A schematic representation of the sample geometry of molecule
4a anchored via the SMe group to one electrode in the silver nanogap.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201200897
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
5

www.afm-journal.de
www.MaterialsViews.com
0.8
0.6
0.4
0.2
0
light
Figure 6. Stability diagram for single-molecule junction taken right after
the molecule was caught in nanogap.
-400
-200
0
200
400
Bias Voltage (mV)
stability diagram taken for bias voltages up to 150 mV reveals
switching from original form A to more conducting (low
resistive) form B, which happens at a gate voltage around 3 V
(Figure 7) .
Figure 8. The effect of shining light on molecular junction (VG = 2.7 V).
Note the reduced transport gap (zero-current area). Blue curve: DHA;
magenta curve: VHF.
0
Such a behavior indicates that below VG ≈ 3 V, the form A
At gate voltages not too far from the critical value VG0, the
junction was found to be light-sensitive; shining light on the
sample for a few minutes (visual light was used) triggered DHA
→ VHF conversion, as shown in Figure 8. Below a gate poten-
tial of 2.6 V, the sample was quenched in DHA form and was
not light-sensitive. The charge distribution within the molecule
in the junction is expected to differ from that of the molecule in
solution and photosensitivity will hence reasonably depend on
correct tuning of the charge distribution by the gate potential.
The future sample evolution after a single light-triggered DHA
→ VHF switching depends on the applied bias voltage: if the
bias voltage is kept low (like in Figure 6) the sample stays in
VHF (low resistivity) form for indefinitely long time (Figure 9).
For higher bias voltages (like in Figure 7), after some time, the
sample spontaneously resets back to DHA (high resistivity). It
was possible to reset the switch by short-time excursion to high
biases, as illustrated in Figure 10. The same light–high bias cycle
was repeated many times and more than 20 ON-OFF cycles
were recorded, some being presented in Figure 11. It should be
noted that all OFF states (DHA form) have similar, though not
exactly identical resistance as evident from Figure 11. The same
applies to all ON states. It was also possible to reset the switch
back to DHA form by increasing the temperature (up to 25 K),
as shown in Figure 12.
0
is the ground state, while above VG it becomes metastable.
To promote switching to a new ground state (form B), high
bias voltage is needed. From our measurements we cannot
deduce whether the switching is actually promoted by the high
bias voltage itself (i.e., by electrostatic field in the junction) or
by high transport current associated with increased bias. The
addition energy EC for form A estimated from the size of the
Coulomb diamonds is about 75 meV (the size of the Coulomb
diamond along the bias axis is 4Ec[23]). Even smaller Coulomb
diamonds (implying smaller HOMO-LUMO gap) for the form
B allow us to identify forms A and B as DHA and VHF respec-
tively.[24] This identification is further supported by an observed
light sensitivity of form A (DHA) to be discussed below.
The above interpretation of junction data as switching
between DHA and VHF forms is naturally based on the ready
light/heat-induced DHA-VHF conversions occurring in solu-
tion. For 1,1-dicyano-2,2-diphenyl-1,2-dihydronaphthalene, a
structurally related photochrome comprised of two six-mem-
bered rings fused together, it was, however, previously shown
that photoinduced C-CN bond dissociation (heterolytic) could
occur in polar solvents.[25] We cannot exclude that such a com-
peting reaction channel could be in play for DHA in the junc-
tion at some gate and bias potentials.
Figure 7. Stability diagram; same data as in Figure 6, but taken for an
increased bias range of 160 mV.
©
6
wileyonlinelibrary.com
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201200897

www.afm-journal.de
www.MaterialsViews.com
Figure 9. Differential resistance dV/dI versus time (VG = 2.67 V). For
each point on the plot a single I(V) curve was taken for bias voltages
V = 100 mV and the differential resistance for the bias voltage voltage
25 mV was computed numerically. The highlighted background (yellow)
indicates the time interval when the sample (DHA) was illuminated with
light. After a single light-triggered switching event the sample stayed in
low-resistive state (VHF) for indefinitely long time.
Figure 11. A series of ON-OFF switchings (VG = 2.67 V). The highlighted
background indicates the time intervals when the sample was illuminated
with light (yellow) or the bias voltage was increased to 80 mV (blue). The
differential resistance dV/dI was taken at the bias voltage V = 25 mV for
both ON (VHF) and OFF (DHA) states.
60
3. Conclusions
In conclusion, we have developed synthetic protocols for func-
tionalizing the DHA/ VHF system with p-MeSC6H4 end-groups,
employing Suzuki cross-coupling reactions. Our regioselective
40
b´
20
0
b
VHF
DHA
a´
a
-20
-40
-60
-400
-200
0
200
400
Bias Voltage (mV)
Figure 10. T w o I(V) curves taken for a bias scan in negative-to-positive
voltage (magenta) and positive-to-negative (blue) directions (VG = 2.7 V).
The sample was illuminated all the time. Transition a → a’ is light-trig-
gered DHA → VHF switching event. An excursion to high bias region
promoted b → b′ transition back to DHA form. Many trace-retrace scans
were taken. Positions for both a → a’ and b → b’ transitions varied from
scan to scan, though the reverse one (b → b’) always happened at bias
voltages |V| > 100 mV. One representative scan with conveniently placed
forward/backward transitions was selected for display.
Figure 12. A series of ON-OFF switchings (VG = 2.7 V). The highlighted
background indicates the time intervals when the sample was illuminated
with light (yellow) or the temperature was increased from 4 K to 25 K
(rose). The differential resistance dV/dI was taken at the bias voltage V =
25 mV for both ON (VHF) and OFF (DHA) states.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201200897
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
7

www.afm-journal.de
www.MaterialsViews.com
and Pd(PPh3)4 (70 mg, 60 μmol, 25 mol%) were added, and the reaction
mixture was stirred at 90 °C overnight. The reaction mixture was diluted
with toluene (100 mL), washed with water (2 × 50 mL) and brine
(50 mL), dried with MgSO4, filtered, and the solvent was evaporated
in vacuo. Purification by dry column vacuum chromatography (SiO2,
12.6 cm2, 0% - 100% CHCl3/ heptanes, 7% steps, 25 mL fractions) gave
3a (106.6 mg, 0.213 mmol, 85%, > 90% purity) as a yellow to red solid.
Recrystallization from acetonitrile (30–50 mL) afforded 3a (21.6 mg,
17%) as yellow needles. Crystals suitable for X-ray crystallography were
grown from acetonitrile. M.p. 150 °C (decomp.). 1H NMR (500 MHz,
CD3CN) δ 7.89 (d, J = 8.7 Hz, 2H), 7.80 (d, J = 8.7 Hz, 2H), 7.67 (d,
J = 8.5 Hz, 2H), 7.37 (d, J = 8.5 Hz, 2H), 7.37 (d, J = 8.6 Hz, 2H);
7.28 (d, J = 8.6 Hz, 2H), 7.21 (s, 1H), 6.87 (dd, J = 11.6, 6.1 Hz, 1H),
6.77 (d, J = 11.6 Hz, 1H), 6.48 (dd, J = 6.1, 1.6 Hz, 1H), 5.96 (d, J =
4.5 Hz, 1H), 3.88 (dd, J = 4.5, 1.6 Hz, 1H), 2.53 (s, 3H), 2.49 (s, 3H)
ppm. 13C NMR (125 MHz, CD3CN) δ 143.1, 142.3, 141.6, 140.6, 140.3,
140.0, 138.5, 137.4, 134.2, 133.9, 132.7, 130.9, 129.4, 128.8, 128.7,
128.3, 127.9, 127.6, 122.2, 117.7, 116.8, 114.8, 52.0, 46.6, 15.9 ppm.
Elem. anal. calc. for C32H24N2S2 (500.68): C 76.76%, H 4.83%, N 5.60%;
found: C 76.63%, H 4.48%, N 5.86%. MS (MALDI-): m/z = 500 [M−].
UV-vis (MeCN): λDHA (ε): 376 nm (21 × 103 M−1 cm−1 ). λVHF (ε): 504 nm
(23 × 103 M−1 cm−1 ).
bromination–elimination–cross-coupling protocol for function-
alizing the DHA at position 7 was hence successfully expanded
to other DHA starting materials via two different routes. While
the reactions seemed to work efficiently, purification of the final
DHA products turned out significantly more tedious than we
have previously experienced with these light-sensitive com-
pounds. From kinetics studies on the thermal ring-closure of
VHF to DHA in solution, we have found that the MeS sub-
stituent exerts a retarding effect when situated as a para-sub-
stituent on a phenyl attached to the seven-membered ring for
this particular example. A DHA with one MeS anchoring group
was successfully placed in the nanogap between silver source
and drain electrodes, and the sample was subjected to a variety
of conductivity measurements. The molecule was weakly cou-
pled to one electrode, and reversible switching between two
forms, most likely DHA (high resistivity, OFF state) and VHF
(low resistivity, ON state), was successfully controlled by gate
voltage, bias voltage, temperature, and light. In previous work
we showed that DHA 2a was well coupled to at least one elec-
trode as no gate-induced conductance nor Coulomb blockade
effects were observed. It seems obvious to grant the MeS sub-
stituent in 4a for the desired weak molecule-electrode coupling,
although we can, of course, only speculate on how the mole-
cule is attached to the electrode. The rich switching behavior
warrants further exploration of this system in molecular
electronics.
2-[4′-Methylthio(1,1′-biphenyl)-4-yl]-1,8a-dihydroazulene-1,1-
dicarbonitrile (4a) : Compound 6a (1.23 mg, 3.17 mmol) was dissolved
in toluene (75 mL) and the solution was flushed with argon for 60 min.
Water (50 mL) was added and the mixture was flushed with argon for an
additional 30 min. The boronic acid 9 (596 mg, 3.55 mmol), KF·2H2O
(365 mg, 3.88 mmol), and Pd(PPh3)4 (652 mg, 0.56 mmol) were added,
and the reaction mixture was stirred for 48 h at 90 °C. The resulting dark
yellow to black reaction mixture was diluted with toluene (200 mL) and
allowed to cool to rt. The solution was washed with sat. aqueous NH4Cl,
dried with MgSO4, and filtered. Any precipitation on the glassware
was dissolved in chloroform, which was then dried with MgSO4. The
combined organic phases were evaporated on Celite. Purification by dry
column vacuum chromatography (SiO2, 12.6 cm2, 0% - 100% CHCl3/
heptanes, 5% steps, 25 mL fractions) gave 4a (960 mg, 80%, >90%
purity) as a yellow powder. Recrystallization from acetonitrile gave
4a (505 mg, 1.34 mmol, 42%) as a bright yellow powder. M.p. 161.8 -
166.5 °C. 1H NMR (500 MHz, CDCl3) δ 7.80 (d, J = 8.6 Hz, 2H), 7.68 (d,
J = 8.6 Hz, 2H), 7.56 (d, J = 8.6 Hz, 2H), 7.35 (d, J = 8.6 Hz, 2H), 6.92 (s,
1H), 6.58 (dd, J = 11.3, 6.4 Hz, 1H), 6.48 (dd, J = 11.3, 6.1 Hz, 1H), 6.35
(dd, J = 6.4, 0.8 Hz, 1H), 6.32 (ddd, J = 10.2, 6.1, 2.1 Hz, 1H), 5.84 (dd,
4. Experimental Section
General Methods: Reactions under argon atmosphere were performed
in a closed vessel equipped with an argon balloon. All reactions using
palladium catalyst were performed in a solvent flushed with argon, by
flushing argon through the solvent for at least 20 minutes while exposing
it to ultrasound. NMR Spectra were acquired on a 300 MHz instrument
or a 500 MHz instrument equipped with a cryoprobe. All chemical shift
values in 1H- and 13C NMR spectra are referenced to the solvent (CDCl3
δH = 7.26 ppm, δC = 77.16 ppm. CD3CN δH = 1.96 ppm, δC = 1.94 ppm.
DMSO-d6 δH = 2.50 ppm, δC = 39.52 ppm). Thin layer chromatography
(TLC) was carried out on commercially available, precoated plates
(silica 60) with fluorescence indicator; a color change from yellow to
red or violet upon irradiation with UV-light indicated the presence of
a DHA. Dry column vacuum chromatography is described in Ref. [26]
All melting points are uncorrected. All spectroscopic measurements
(including photolysis) were performed in a 1-cm path length cuvette.
Photoswitching experiments were performed using a 150-W xenon arc
lamp equipped with a monochromator; the DHA absorption maximum
(lowest energy absorption) for each individual species was chosen as
the wavelength of irradiation (line width 2.5 nm). The thermal back-
reaction was performed by heating the sample (cuvette) by a Peltier
unit in the UV-vis spectrophotometer. Light-heat cycles described above
correspond to the following steps: i) irradiation of the DHA species until
no further changes in the absorption spectrum were observed, ii) heating
of the formed VHF species until no further changes in the absorption
spectrum were observed. A detailed description of the experiment
J = 10.2, 3.8 Hz, 1H), 3.81 (dt, J = 3.8, 2.1 Hz, 1H), 2.54 (s, 3H) ppm. 13
C
NMR (125 MHz, CDCl3) δ 142.1, 139.9, 138.9, 138.9, 136.5, 132.2, 131.1,
131.0, 129.3, 127.8, 127.6, 127.4, 127.0, 126.9, 121.1, 119.6, 115.3, 112.9,
51.3, 45.3, 15.8 ppm. Elem. anal. calc. for C25H18N2S (378.49): C 79.33%,
H 4.79%, N 7.40%; found: C 78.71%, H 4.68%, N 7.21%. MS(FAB+):
m/z = 378 [M+]. UV-vis (MeCN): λDHA (ε): 376 nm (28 × 103 M−1 cm−1 ).
λ
VHF (ε): 475 nm (23 × 103 M−1 cm−1 ).
2-(4′-Iodophenyl)-7,8-dibromo-7,8,1,8a-tetrahydroazulene-1,1-
dicarbonitrile (7 ) : Compound 6a (429.3 mg, 1.123 mmol) was dissolved
in CH2Cl2 (15 mL) under an argon atmosphere, and the mixture was
excluded from light. The reaction mixture was cooled to –78 °C and
a solution of Br2 in CH2Cl2 (1.44 mL, 0.78 M, 1.12 mmol) was added
dropwise. The brown mixture was then stirred for 1 h after which
the solution took a yellow color. Evaporation of the solvents gave 7
(608.7 mg, 1.12 mmol, > 99%) as a yellow to brown foam. 1H NMR (300
MHz, CDCl3) δ 7.85–7.79 (m, 2H), 7.50–7.44 (m, 2H), 6.98 (s, 1H), 6.29
(dd, J = 7.5, 2.3 Hz, 1H), 6.09 (dd, J = 12.2, 7.5 Hz, 1H), 5.93 (dd, J =
12.2, 5.5 Hz, 1H), 5.33–5.27 (m, 1H), 5.05–5.01 (m, 1H), 4.65 (br s, 1H)
ppm. 13C NMR (75 MHz, CDCl3) δ 144.3 (two signals), 138.7, 134.7,
129.6, 129.0, 127.8, 125.8, 121.8, 114.5, 111.6, 109.9, 96.8, 53.2, 51.5,
49.1, 44.5 ppm (one signal missing due to overlap). Elem. anal. calc. for
C18H11Br2IN2 (542.01): C 39.98%, H 2.05%, N 5.17%; found: C 39.68%,
H 1.94%, N 5.01%.
and investigation of the accuracy has previously been published.[20]
A
single-molecule junction based on DHA 4a was prepared according to a
previous fabrication method.[15–17,22]
2-[4′-Methylthio(1,1′-biphenyl)-4-yl]-7-[4-(methylthio)phenyl]-1,8a-
dihydroazulene-1,1-dicarbonitrile (3a) : Compound 10a (1/4 of a batch
made in 2 steps from 1.00 mmol 4a) was dissolved in toluene (20 mL)
flushed with argon. Argon-flushed water (10 mL) was added. The
boronic acid (84 mmol, 0.50 mmol), KF·2H2O (23.5 mg, 0.25 mmol),
7-Bromo-2-(4′-iodophenyl)-1,8a-dihydroazulene-1,1-dicarbonitrile (8a) :
Freshly prepared dibromide 7 (609 mg, 1.12 mmol) was dissolved in
dry THF (12 mL, <15 ppm H2O) under argon atmosphere. The solution
©
8
wileyonlinelibrary.com
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201200897

www.afm-journal.de
www.MaterialsViews.com
was cooled to 0 °C and a solution of LiHMDS (1.1 mL, 1.1 mmol,
1.0 M) in THF was added dropwise. The reaction mixture was stirred
for 2 h, while the temperature was allowed to reach rt. The reaction
mixture was diluted with Et2O (100 mL), washed with aqueous NH4Cl
(2 × 50 mL), dried with MgSO4, and filtered. Evaporation of the solvents
gave 8a (516 mg, >90% purity) as a black solid with minor impurities.
An analytically pure sample was obtained by purification by dry column
chromatography (SiO2, 12.6 cm2, 0–30% THF/ heptanes, 5% steps,
40 mL fractions) followed by recrystallization from CH2Cl2/ heptanes,
Acknowledgements
The research leading to these results has received funding from the
European Community’s Seventh Framework Programme (FP7/2007-
2013) under the grant agreement “SINGLE” no 213609, coordinated
by Prof. T. Bjørnholm who is thanked for helpful discussions. In
addition, support from The Danish Council for Independent Research
| Natural Sciences (#10-082088) and the Swedish Research Council is
acknowledged.
1
which gave 8a as discolored yellow needles. M.p. 168°C (decomp.). H
Received: March 29, 2012
Revised: May 17, 2012
Published online:
NMR (500 MHz, CDCl3) δ 7.83 (d, J = 8.7 Hz, 2H), 7.45 (d, J = 8.7 Hz,
2H), 6.90 (s, 1H), 6.58–6.49 (m, 2H), 6.34 (d, J = 5.9 Hz, 1H), 6.11 (d,
J = 4.4 Hz, 1H), 3.79 (dd, J = 4.4, 1.8 Hz, 1H) ppm. 13C NMR (125 MHz,
CDCl3) δ 140.8, 140.7, 138.7, 133.3, 132.3, 132.1, 129.6, 127.9, 120.7,
120.3, 120.1, 114.5, 112.3, 97.1, 51.1, 44.6 ppm. Elem. anal. calc. for
C18H10BrIN2 (461.09): C 46.89%, H 2.19%, N 6.08%; found: C 46.63%,
H 2.05%, N 6.03%. MS (EI+): m/z = 460 [M+]. UV-vis (MeCN): λDHA (ε):
363 nm (21 × 103 M−1 cm−1 ). λVHF (ε): 468 nm (27 × 103 M−1 cm−1 ).
2-[4′-Methylthio(1,1′-biphenyl)-4-yl]-7-bromo-1,8a-dihydroazulene-1,1-
dicarbonitrile (10a) : Freshly prepared dibromide 11 (from 1.00 mmol
4a) was dissolved dry THF (20 mL) and cooled to 0 °C. A solution of
LiHMDS in THF (1.00 mL, 1.00 mmol, 1.0 M) was added dropwise. After
addition of 0.1 mL the solution changed color from dark green to yellow.
Addition of the rest of the base caused a color change to black. The
reaction mixture was stirred for 1 h, while the temperature was allowed
to reach rt. The reaction mixture was then diluted with Et2O (100 mL),
washed with brine (100 mL), dried with MgSO4, and filtered. Evaporation
of the solvents gave 10a (>90% purity) as a black solid. An analytically
pure sample was obtained by dry column vacuum chromatography (SiO2,
12.6 cm2, 0% - 22% THF/ heptanes, 1% steps, 40 mL fractions), which
gave 10a (74% recovered, > 95% purity) as a yellow solid. A subsequent
recrystallization from acetonitrile gave 10a (12% recovered) as yellow
crystals. M.p. 154°C (decomp.). 1H NMR (500 MHz, CDCl3) δ 7.80 (d,
J = 8.7 Hz, 2H), 7.69 (d, J = 8.7 Hz, 2H), 7.56 (d, J = 8.6 Hz, 2H), 7.35 (d,
J = 8.6 Hz, 2H), 6.92 (s, 1H), 6.54 (m, 2H), 6.33 (dt, J = 4.1, 1.8 Hz, 1H),
6.14 (d, J = 4.4 Hz, 1H), 3.82 (dd, J = 4.4, 1.8 Hz, 1H), 2.54 (s, 3H) ppm.
13C NMR (125 MHz, CDCl3) δ 142.7, 141.4, 141.2, 139.2, 136.3, 132.9,
132.2, 131.3, 128.8, 127.6, 127.5, 127.1, 127.0, 120.3, 120.0, 120.0, 114.8,
112.5, 51.2, 44.7, 15.8 ppm. Elem. anal. calc. for C25H17BrN2S (457.38):
C 65.65%, H 3.75%, N 6.12%; found: C 65.24%, H 3.65%, N 6.40%. MS
(EI+): m/z = 456/ 458 [M+, 79/81Br]. UV-vis (MeCN): λDHA (ε): 380 nm
(45 × 103 M−1 cm−1 ). λVHF (ε): 462 nm (37 × 103 M−1 cm−1 ).
[1] a) L. Carroll, C. B. Gorman, Angew. Chem. Int. Ed. 2002, 41, 4378;
b) N. Robertson, C. A. McGowan, Chem. Soc. Rev. 2003, 32, 96;
c) A. C. Benniston, Chem. Soc. Rev. 2004, 33, 573; d) A. Troisi,
M. A. Ratner, Small 2006, 2, 172; e) N. Weibel, S. Grunder, M. Mayor,
Org. Biomol. Chem. 2007, 5, 2343; f) R. Klajn, J. F. Stoddart,
B. A. Grzybowski, Chem. Soc. Rev. 2010, 39, 2203;
g) S. J. van der Molen, P. Liljeroth, J. Phys.: Condens. Matter 2010,
22, 133001.
[2] A. V. Danilov, P. Hedegård, D. S. Golubev, T. Bjørnholm,
S. E. Kubatkin, Nano Lett. 2008, 8, 2393.
[3] a) D. Dulic, S. J. van der Molen, T. Kudernac, H. T. Jonkman,
J. J. D. de Jong, T. N. Bowden, J. van Esch, B. L. Feringa,
B. J. van Wees, Phys. Rev. Lett. 2003, 91, 207402; b) N. Katsonis,
T. Kudernac, M. Walko, S. J. van der Molen, B. J. van Wees,
B. L. Feringa, Adv. Mater. 2006, 18, 1397; c) C. Zhang, Y. He,
H.-P. Cheng, Y. Xue, M. A. Ratner, X.-G. Zhang, P. Krstic, Phys. Rev.
B 2006, 73, 125445; d) A. C. Whalley, M. L. Steigerwald, X. Guo,
C. Nuckolls, J. Am. Chem. Soc. 2007, 129, 12590; e) A. S. Kumar,
T. Ye, T. Takami, B.-C. Yu, A. K. Flatt, J. M. Tour, P. S. Weiss, Nano
Lett. 2008, 8, 1644; f) V. Ferri, M. Elbing, G. Pace, M. D. Dickey,
M. Zharnikov, P. Samorì, M. Mayor, M. A. Rampi, Angew. Chem.
Int. Ed. 2008, 47, 3407; g) A. J. Kronemeijer, H. B. Akkerman,
T. Kudernac, B. J. van Wees, B. L. Feringa, P. W. M. Blom,
B. de Boer, Adv. Mater. 2008, 20, 1467; h) S. J. van der Molen,
J. Liao, T. Kudernac, J. S. Augustsson, L. Bernard, M. Calame,
B. J. van Wees, B. L. Feringa, C. Schönenberger, Nano Lett. 2009,
9, 76; i) W. R. Browne, B. Feringa, Annu. Rev. Phys. Chem. 2009,
60, 407; j) T. Kudernac, N. Katsonis, W. R. Browne, B. L. Feringa, J.
Mater. Chem. 2009, 19, 7168; k) T. Tsujioka, M. Irie, J. Photochem. &
Photobiol. C: Photochem. Rev. 2010, 11, 1.
[4] a) M. Zhuang, M. Ernzerhof, Phys. Rev. B 2005, 72, 073104;
b) M. Zhuang, M. Ernzerhof, J. Chem. Phys. 2009, 130, 114704.
[5] S. Lara-Avila, A. V. Danilov, S. E. Kubatkin, S. L. Broman, C. R. Parker,
M. B. Nielsen, J. Phys. Chem. C 2011, 115, 18372.
[6] J. Daub, T. Knöchel, A. Mannschreck, Angew. Chem. Int. Ed. Engl.
1984, 23, 960.
[7] a) S. L. Broman, M. Å. Petersen, C. G. Tortzen, A. Kadziola, K. Kilså,
M. B. Nielsen, J. Am. Chem. Soc. 2010, 132, 9165; b) M. Å. Petersen,
S. L. Broman, K. Kilså, A. Kadziola, M. B. Nielsen, Eur. J. Org. Chem.
2011, 1033.
[8] H. Görner, C. Fischer, S. Gierisch, J. Daub, J. Phys. Chem. 1993, 97,
4110.
2-[4′-Methylthio(1,1′-biphenyl)-4-yl]-7,8-dibromo-1,7,8,8a-
tetrahydroazulene-1,1-dicarbonitrile (11) : Compound 4a (378.5 mg,
1.00 mmol) was dissolved in CH2Cl2 (20 mL). The solution was cooled
to –78 °C and stirred for 20 min. A solution of Br2 in CH2Cl2 (1.28 mL,
1.00 mmol, 0.78 M) was added dropwise, and the reaction mixture
was stirred for 1 h. The dry ice/ acetone bath was removed and the
temperature was allowed to rise to rt. Evaporation of the solvent gave
11 (538.3 mg, 1.00 mmol, >99%) as a green solid/ foam. M.p. 135 °C
1
(decomp.). H NMR (500 MHz, CDCl3) δ 7.82 (d, J = 8.6 Hz, 2H), 7.69
(d, J = 8.6 Hz, 2H), 7.56 (d, J = 8.4 Hz, 2H), 7.35 (br d, J = 8.4 Hz, 2H),
7.01 (s, 1H), 6.28 (dd, J = 7.6, 2.2 Hz, 1H), 6.10 (dd, J = 12.2, 7.6, Hz,
1H), 5.92 (dd, J = 12.2, 5.5 Hz, 1H), 5.35–5.31 (m, 1H), 5.07 (dt, J =
3.0, 1.5 Hz, 1H), 4.68 (br s, 1H), 2.54 (s, 3H) ppm. Elem. anal. calc. for
C25H18Br2N2S (538.30): C 55.78%, H 3.37%, N 5.20%; found: C 55.54%,
H 3.24%, N 5.16%.
Crystallographic data (excluding structure factors) for the structure(s)
reported in this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication no. CCDC-
857551 (10a), CCDC-857550 (3a), CCDC-857552 (13a).
[9] a) J. Daub, S. Gierisch, J. Salbeck, Tetrahedron Lett. 1990, 31, 3113;
b) J. Achatz, C. Fischer, J. Salbeck, J. Daub, J. Chem. Soc., Chem.
Commun. 1991, 504; c) M. Å. Petersen, L. Zhu, S. H. Jensen,
A. S. Andersson, A. Kadziola, K. Kilså, M. B. Nielsen, Adv. Funct.
Mater. 2007, 17, 797.
Supporting Information
[10] C. R. Parker, C. G. Tortzen, S. L. Broman, M. Schau-Magnussen,
K. Kilså, M. B. Nielsen, Chem. Commun. 2011, 47, 6102.
Supporting Information is available from the Wiley Online Library or
from the author.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201200897
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
9

www.afm-journal.de
www.MaterialsViews.com
[11] V. De Waele, M. Beutter, U. Schmidhammer, E. Riedle, J. Daub,
Chem. Phys. Lett. 2004, 390, 328.
[17] S. Lara-Avila, A. Danilov, V. Geskin, S. Bouzakraoui, S. Kubatkin,
J. Cornil, T. Bjørnholm, J. Phys. Chem. C 2010, 114, 20686.
[18] L. Gobbi, P. Seiler, F. Diederich, V. Gramlich, C. Boudon,
J.-P. Gisselbrecht, M. Gross, Helv. Chim. Acta 2001, 84, 743.
[19] M. Å. Petersen, S. L. Broman, A. Kadziola, K. Kilså, M. B. Nielsen,
Eur. J. Org. Chem. 2009, 2733.
[20] S. L. Broman, S. L. Brand, C. R. Parker, M. Å. Petersen, C. G. Tortzen,
A. Kadziola, K. Kilså, M. B. Nielsen, Arkivoc 2011, ix, 51.
[21] D. H. McDaniel, H. C. Brown, J. Org. Chem. 1958, 23, 420.
[22] a) S. E. Kubatkin, A. V. Danilov, H. Olin, T. Claeson, Appl. Phys. Lett.
1998, 73, 3604; b) S. E. Kubatkin, A. V. Danilov, H. Olin, T. Claeson,
J. Low Temp. Phys. 2000, 118, 307; c) A. V. Danilov, D. S. Golubev,
S. E. Kubatkin, Phys. Rev. B 2002, 65, 1253121.
[12] a) H. van Houten, C. W. J. Beenakker, A. A. M. Staring, in Single
Charge Tunnelling and Coulomb Blockade Phenomena in Nanostruc-
tures, Plenum, New York 1992, Ch. 5; b) H. Park, J. Park, A. K. L. Lim,
E. H. Anderson, A. P. Alivisatos, P. McEuen, Nature 2000, 407, 57.
[13] K. Nørgaard, M. B. Nielsen, T. Bjørnholm, in Functional Organic
Materials (Eds: T. J. J. Müller, U. H. F. Bunz), VCH-Wiley, Weinheim
2007, p. 353–392.
[14] a) Y. J. Kwon, S. B. Lee, K. Kim, M. S. Kim, J. Mol. Struct. 1994, 318,
25; b) J. Hu, D. L. Mattern, J. Org. Chem. 2000, 65, 2277; c) L. Zobbi,
M. Mannini, M. Pacchioni, G. Chastanet, D. Bonacchi, C. Zanardi,
R. Biagi, U. Del Pennino, D. Gatteschi, A. Cornia, R. Sessoli, Chem.
Commun. 2005, 1640; d) Y. S. Park, A. C. Whalley, M. Kamenetska,
M. L. Steigerwald, M. S. Hybertsen, C. Nuckolls, L. Venkataraman, J.
Am. Chem. Soc. 2007, 129, 15768.
[23] L. P. Kouwenhoven, D. G. Austing, S. Tarucha, Rep. Prog. Phys. 2001
64, 701.
[24] The junction properties (conductance and the phase of Coulomb
blockade oscillations) for form B were not as stable as for form
A. Therefore we could not extract from our data the exact value
for the addition energy for form B, but qualitatively, it is quite clear
from Figure 7 that the addition energy is reduced after A → B
switching.
[15] a) S. Kubatkin, A. Danilov, M. Hjort, J. Cornil, J.-L. Brédas,
N. Stuhr-Hansen, P. Hedegård, T. Bjørnholm, Nature 2003, 425,
698; b) S. Kubatkin, A. Danilov, M. Hjort, J. Cornil, J.-L. Brédas,
N. Stuhr-Hansen, P. Hedegård, T. Bjørnholm, Current Appl. Phys.
2004, 4, 554.
[16] A. V. Danilov, S. E. Kubatkin, S. G. Kafanov, T. Bjørnholm, Faraday
Discuss. 2006, 131, 337.
[25] H. Görner, T. Mrozek, J. Daub, Chem. Eur. J. 2002, 8, 4008.
[26] D. S. Pedersen, C. Rosenbohm, Synthesis 2001, 2431.
©
10 wileyonlinelibrary.com
2012 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2012,
DOI: 10.1002/adfm.201200897