Large Converse Piezoelectric Effect Measured on a Single Molecule on a Metallic Surface

Article abstract of DOI:10.1021/jacs.7b08729

The converse piezoelectric effect is a phenomenon in which mechanical strain is generated in a material due to an applied electrical field. In this work, we demonstrate the converse piezoelectric effect in single heptahelicene-derived molecules on the Ag(111) surface using atomic force microscopy (AFM) and total energy density functional theory (DFT) calculations. The force-distance spectroscopy acquired over a wide range of bias voltages reveals a linear shift of the tip-sample distance at which the contact between the molecule and tip apex is established. We demonstrate that this effect is caused by the bias-induced deformation of the spring-like scaffold of the helical polyaromatic molecules. We attribute this effect to coupling of a soft vibrational mode of the molecular helix with a vertical electric dipole induced by molecule-substrate charge transfer. In addition, we also performed the same spectroscopic measurements on a more rigid o-carborane dithiol molecule on the Ag(111) surface. In this case, we identify a weaker linear electromechanical response, which underpins the importance of the helical scaffold on the observed piezoelectric response.

Full text of DOI:10.1021/jacs.7b08729

Subscriber access provided by READING UNIV
Article
Large converse piezoelectric effect measured
on a single molecule on a metallic surface
Oleksandr Stetsovych, Pingo Mutombo, Martin Švec, Michal Šámal, Jindrich Nejedly, Ivana Cisarova, Hector
Vazquez, Maria Moro-Lagares, Jan Berger, Jaroslav Vacek, Irena G. Stará, Ivo Stary, and Pavel Jelinek
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b08729 • Publication Date (Web): 23 Dec 2017
Downloaded from http://pubs.acs.org on December 24, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Large converse piezoelectric effect measured on a single molecule on
a metallic surface
Oleksandr Stetsovych†*, Pingo Mutombo†, Martin Švec†,‡, Michal Šámal§, Jindřich Nejedlý§,
Ivana Císařová¥, Héctor Vázquez†, María Moro-Lagares†,‡, Jan Berger†, Jaroslav Vacek§, Irena G.
Stará§, Ivo Starý§*, Pavel Jelínek†,‡*
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†Institute of Physics of the Czech Academy of Sciences, Cukrovarnická 10, Prague 6, Czech Republic.
‡Regional Center of Advanced Technologies and Materials, Palacký University, Olomouc, Czech Republic.
§Institute of Organic Chemistry and Biochemistry, Czech Academy of Sciences, Flemingovo nám. 2, 16610 Prague 6,
Czech Republic.
¥Department of Inorganic Chemistry, Faculty of Science, Charles University in Prague, Hlavova 2030/8, 128 43 Pra-
gue 2, Czech Republic.
ABSTRACT: The converse piezoelectric effect is a phenomenon in which the mechanical strain is generated in a material
due to an applied electrical field. In this work, we demonstrate the converse piezoelectric effect in single heptaheliceneꢀ
derived molecules on the Ag(111) surface using atomic force microscopy (AFM) and total energy density functional (DFT)
calculations. The forceꢀdistance spectroscopy acquired over a wide range of bias voltages reveals a linear shift of the tipꢀ
sample distance at which the contact between the molecule and tip apex is established. We demonstrate that this effect is
caused by the biasꢀinduced deformation of the springꢀlike scaffold of the helical polyaromatic molecules. We attribute this
effect to coupling of a soft vibrational mode of the molecular helix with a vertical electric dipole induced by moleculeꢀ
substrate charge transfer. In addition, we also performed the same spectroscopic measurements on a more rigid oꢀcarborane
dithiol molecule on the Ag(111) surface. In this case, we identify a weaker linear electromechanical response, which underꢀ
pins
the
importance
of
the
helical
scaffold
on
the
observed
piezoelectric
response.
1. INTRODUCTION
Piezoelectric effect (PE) emerges in materials in which the mechanical and electrical degree of freedom are
coupled. Either the electric charge can be reversibly generated if a mechanical stress is applied (direct PE) or,
conversely, the mechanical strain can arise if the electric field is applied (converse PE). Materials endowed with
such a property have to be nonꢀcentrosymmetric, poorly conductive and the magnitude of their dipole moment is
altered under the mechanical stress. Many materials have been demonstrated to exhibit PE such as piezoelectric
ceramics (lead zirconate titanate, PZT), polymers (polyvinylidene difluoride, PVDF) or quartz. They are central
to a large number of applications in the automotive, computer, smartphone, consumer, medical and military inꢀ
dustries1. Nowadays, new challenges have emerged in the context of PE to realize selfꢀpowered devices or draꢀ
matically reduce their size indispensable for a nanoscale use2,3,4,5,6. Also, biological materials such as collagen
fibrils7, proteins8 or bacteriophages9 show piezoelectric properties. In the quest for molecular nanoelectromechanꢀ
ical systems10,11 there is a particular interest in designing singleꢀmolecule piezoelectric devices and demonstrating
their function. Such an achievement has so far remained elusive despite the realization of singleꢀmolecule devices
ranging from the diode12 transistor 13, resonant tunneling junction14, thermoelectric junction15, rotor16, unidirecꢀ
tional motor 17,18,19,20, molecular machine21, artificial muscle22 to nanocars23,24 that have already been reported.
Inherently chiral helicenes (helical allꢀorthoꢀannulated polyaromatics)25 are promising candidates as piezoelecꢀ
trically active molecules. They lack inversion symmetry, possess a springꢀlike carbon scaffold exhibiting an exꢀ
traordinary deformation capacity (when compressed/stretched along the helix axis)26,27,28,29,30,31 and can be funcꢀ
tionalized by electronically distinguished moieties at their opposite termini to install a properly oriented dipole
moment. Within this scenario, Hutchison et al. theoretically analyzed a set of asymmetrically substituted
[6]helicenes and their analogs equipped with various electron donating and withdrawing groups varying also the
polarizability of their backbones32. Importantly, they identified highly responsive helicene molecules that were
foreseen to yield large piezoelectric coefficients in the range of tens to hundreds of pm/V (surpassing PVDF and
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 10
competing with PZT). Thus, a successful development of the singleꢀmolecule piezoelectrics would enable a botꢀ
tomꢀup approach to nanoelectromechanical systems such as molecular actuators (linear motors), sensors and elecꢀ
tricity generators. A significant milestone on the pathway to realizing singleꢀmolecule piezoelectrics was the
scanning tunneling microscopy (STM) study by Kimura et al. describing a twoꢀstate switching of the conductance
and molecular length of dodecapeptide helix bundles immobilized on gold that was stochastic with time33. A corꢀ
relation between the helix dipole moment orientation, polarity of applied bias and conformational changes of
helices was demonstrated.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Measurements of the piezoelectric effect on the molecule 7HdiET by AFM. (a) Ballꢀandꢀstick model
of the experiment. (b) Measured ∆f(z,VB) for the single molecule 7HdiET with the same tip: the frequency shift ∆f
vs. zꢀdistance spectroscopies acquired for different bias voltages VB.
2. RESULTS AND DISCUSSION
Here, we report large electromechanical response of single molecules on surfaces that was observed on racemic
heptahelicene derivatives (S,S'ꢀheptaheliceneꢀ2,17ꢀdiyl diethanethioate, 7HdiET and heptaheliceneꢀ2,17ꢀdithiol,
7HdiT) deposited on the Ag(111) surface by employing STM and atomic force microscopy (AFM) measureꢀ
ments. Namely, we performed simultaneous STM/AFM spectroscopic measurements over a wide range of the tipꢀ
sample distances and bias voltages, as shown schematically in Figure 1a. The forceꢀdistance spectroscopies (see
Figure 1b) reveal a strong linear shift in the distance at which the molecule is contacted by the metallic tip,
strongly dependent on the applied bias voltage. The experimental data allowed the determination of the singleꢀ
molecule piezoelectric coefficient d33 of 7HdiET and 7HdiT to be ~55 and ~80 pm/V, respectively. We attribute
this effect to the presence of an intrinsic dipole coupled with a soft vibrational mode of the helix scaffold. The
intrinsic dipole moment parallel to the axis of the molecule helix, see Figure 1a, originates from charge transfer
from the molecule to the metallic surface. In addition, we carried out control spectroscopic measurements on 1,2ꢀ
dithiolꢀorthoꢀcarborane (CBdiT) molecules, which have a rigid skeleton, on Ag(111). In this case, we identify a
nonꢀnegligible but much weaker linear electromechanical response, which confirms the determinant influence of
the helical scaffold on the electromechanical response.
Figure 2. Molecules 7HdiET on the Ag(111) surface. (a) STM overview image of selfꢀassemblies of 7HdiET on
Ag(111). Each bright protrusion corresponds to one molecule. Inset: a ballꢀandꢀstick model of 7HdiET. (b) STM
image of a partially reassembled island by scanning probe manipulation. Red arrows highlight the direction of
manipulations. Molecules after manipulations stabilize in two different conformations that can be distinguished
ACS Paragon Plus Environment

Page 3 of 10
Journal of the American Chemical Society
by their different height appearance in STM. (c) STM height profile over two molecules in different conforꢀ
mations. Images taken at 4.4 5 K, (a) It = 10 pA, Vb= 30 mV, image size: 18 × 18 nm2; (b) It = 5 pA, Vb = 1 V,
image size: 15 × 15 nm2; (c) It = 10 pA, Vb = 1.5 V, image size: 7 × 4.3 nm2.
1
2
3
4
5
6
7
8
Figure 2a shows the STM image of the racemic 7HdiET adsorbed on the Ag(111) surface at a temperature of 5
K. The molecules assemble into linear and circular islands or their combinations. From analysis of STM images
and total energy DFT simulations, we identify that the molecules adopt faceꢀon adsorption geometry, as shown on
Figure 1a. We can deliberately disassemble the molecular islands via tip interactions with single molecules, as
shown in Figure 2b. Individual molecules stabilize with two different conformations of the upper acetylsulfanyl
group (distant from the Ag surface) that can be distinguished by the different apparent height of the molecule in
the STM channel (see Figure 2c). We define the two conformers as 'bright' and 'dark' ones according to their apꢀ
pearance in the STM mode.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We can reversibly switch between the 'bright' and 'dark' conformers via a mechanical action with the tip or apꢀ
plying a positive bias voltage, see Figure S3. From DFT calculations we estimated that the 'dark' conformer (Figꢀ
ure 3d) corresponds to an energetically slightly more stable structure (~70 meV, see Figure S2), in which the oxꢀ
ygen atom of the upper acetylsulfanyl group points towards the Ag surface. This 'dark' conformer can be convertꢀ
ed, after the mechanical contact and subsequent tip retraction, to the 'bright' conformer, in which the acetylsulfaꢀ
nyl group is rotated placing the oxygen atom outward the surface (Figures S3 and S4). Conversely, the 'bright'
conformer can be reproducibly switched to the 'dark' conformer by the application of the positive bias voltage
inducing repulsive electrostatic interaction between negatively charged tip and oxygen atom. Details of the
switching mechanism are described in Supporting Information (see Supplementary Discussion).
Both of the molecular conformers can be contacted very reproducibly with a metallic tip. Figure 3a represents a
characteristic evolution of the frequency shift ∆f and simultaneously recorded tunneling current It during the tip
approach over the 'dark' and 'bright' conformers. Such a simultaneous acquisition of the ∆f(z) and It(z) spectrosꢀ
copy allows us to monitor the process of contacting molecules in detail (Figure 3a,b). Qualitatively, it shows very
similar features regardless of the fact whether 'bright' or 'dark' conformers are contacted. At far tipꢀsample disꢀ
tances (regime I), the tunneling current It increases exponentially, while the frequency shift ∆f shows a quadratic
dependence on applied bias voltage (see Figure S7 and related Supplementary Discussion in Supporting Inforꢀ
mation). Regime I corresponds to the tunneling regime where the molecule is not yet contacted. At a certain disꢀ
tance (regime II), we observe a sharp characteristic minimum in the frequency shift ∆f being accompanied by a
jumpꢀlike increase in the tunneling current It. It is worth noting that the width of the regime II is closely related to
the oscillation amplitude (see Figure S8 in Supplementary Discussion). Regime II corresponds to the distance
where the metallic tip apex is continuously establishing and breaking a chemical bond with the molecule in each
oscillation cycle. Thus, we define the contacting distance zc as the tipꢀsample distance, at which the minima of the
frequency shift occur. Approaching the molecule further (regime III), the frequency shift ∆f stabilizes at positive
values, while the tunneling current It returns to the exponential growth (at a lower rate than in the regime I). Conꢀ
sequently, the regime III corresponds to the tipꢀsample distance where the molecule maintains chemical contact
with the tip.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Contacting the molecules 7HdiET in 'bright' and 'dark' conformations. (a) Frequency shift and (b)
simultaneously recorded tunneling current for 'bright' (blue curve) and 'dark' (red curve) conformers. Three
characteristic regimes are marked. Regime I: the approach before the contact; regime II: forming and breaking
the contact every oscillation cycle; regime III: the stable contact. The 'bright' conformer is contacted at the highꢀ
er tip sample distance (about ~0.18 nm) than the 'dark' one. Spectra are taken at 40 pm oscillation amplitude of
the sensor. (c) Calculated 1D plot of the differential electron density (∆ꢀ ꢁ ꢀꢂꢃꢃ ꢄ ꢀꢅꢆꢇꢈ ꢄ ꢀꢉꢊꢃ) transfer beꢀ
tween the molecule 7HdiET and Ag(111) surface for different external electric fields. (d) Calculated force vs. disꢀ
tance curves for the molecules in 'bright' and 'dark' conformations, which reveal a linear regime after the contact.
Numbers correspond to the tipꢀsample separations presented in the ballꢀandꢀstick model insets.
The contacting distance of the molecules 7HdiET occupying the 'bright' conformation is about 0.18 nm higher
than that one in the 'dark' conformer (Figure 3a). This difference can be rationalized by the outward orientation of
the oxygen atom of the acetylsulfanyl group in the 'bright' conformer. To corroborate this scenario, we performed
the total energy DFT simulation of the process of approaching a metallic tip over both the oxygen and sulfur atom
of the acetylsulfanyl group in the 'bright' and 'dark' conformation, respectively. The evolution of the interaction
force between the tip and the molecule during the approach and corresponding structural changes are shown in
Figure 3d. According to the DFT calculations, most of the structural deformation during the contact is related to
vertical changes in the helical pitch but not affecting the AgꢀS bond length (see Figure S13). The contacting ocꢀ
curs at a distance where a sharp decrease of the interaction force appears, see point 3 in Figure 3d. The calculated
height difference of the contact distance zc for the 'bright' or 'dark' conformer is about 0.189 nm, which is in a
good agreement with the experimental value. It is worth noting that the current after the contact is about five
times larger for the 'dark' conformer than for the 'bright' one. This experimental observation matches well with
the calculated transmission spectra34,35 of the both conformers in the contact regime (for more details, see Figures
S9 and S10).
ACS Paragon Plus Environment

Page 5 of 10
1
2
3
4
5
6
7
Journal of the American Chemical Society
The recorded ∆f(z) spectra over different molecules with the same tip are well reproducible, as shown in Figure
S12, demonstrating the formation of a wellꢀdefined molecular junction. Interestingly, in the regime III, the freꢀ
quency shift ∆f remains almost constant, especially when acquired with very low oscillation amplitude, see as
shown in Figure S8. This means that the molecular junction acts as a linear spring. Indeed, structural deforꢀ
mations obtained from the total energy DFT calculations mostly consist of linear changes of the helix pitch in this
regime for both conformers. Hence, an effective stiffness kM of the helical spring can be directly estimated from
the ∆f(z) spectroscopy. In the small amplitude regime, the effective stiffness kM of the molecular junction can be
∆ꢍ. 2ꢋ
8
9
calculated as ꢋꢌ
ꢁ
ꢎ , where k=1.08×106 N/m is the sensor stiffness, fo=0.98 MHz its eigenfrequency and
ꢍ
ꢊ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
the ∆f =1.60±0.39 Hz (measured from the ∆f(z) spectra taken from contacting 24 different 'bright' conformers
with the same tip, see Figure S12). Accordingly, we calculate the molecular spring constant kM to be 3.53±0.86
N/m. This value concurs very well with the effective stiffness estimated from the tipꢀmolecule force on contact
from DFT simulations: 3.6±0.1 N/m. In addition, experimental inelastic electron tunneling spectroscopy reveals
the presence of a soft vibrational mode (Figure S11a), which we tentatively relate to the breathing vibrational
mode of the helix (Figure S11b). We should note that the calculated stiffness of AgꢀS bond is 43 N/m, which is
one order of magnitude higher that the stiffness of the helix. Consequently, structural deformation of the 7HdiET
adsorbate by external stimuli takes place predominantly on the helical scaffold (Figure S13).
In the next, we analyze charge transfer between the metal and molecule, which is expected to reflect both the
Pauli repulsion and misalignment of energy bands of silver with energy levels of interacting molecular orbitꢀ
als40,41. Therefore, we performed Kelvin probe force microscopy measurements to determine the electron transfer
from the molecule 7HdiET to the metallic substrate. The ∆f(VB) spectra (Figure S5) reveal a significant variation
between the local contact potential difference Vcpd acquired over the 7HdiET molecule and on the bare Ag(111)
substrate. In addition, we observe a shift of Vcpd above the molecule 7HdiET towards the higher values (from ꢀ
0.70 to ꢀ0.48 V). This shift indicates the presence of an extra dipole moment oriented with its positive pole pointꢀ
ing to the surface as shown in Figure 1a. The origin of this vertical dipole moment can be understood as a conseꢀ
quence of the charge transfer from the molecule to the surface, in good agreement with the total energy calculaꢀ
tions, see Figure 3c. Such a πꢀelectron density rearrangement creates a positive charge at the helix terminus closer
to the surface. Consequently, another dipole moment is created within the molecule helix. This internal molecular
dipole is created by interaction of the positive charge at the helix terminus near the surface with an electron rich
terminal group (acetylsufanyl or thiol group) located at the opposite helix terminus. This dipole pointing downꢀ
ward is mainly sensed by the Kelvin probe in the far distance. Moreover, the presence of the dipole is also supꢀ
ported by the direction of vertical relaxation of the helix in the presence of the external electrostatic field obꢀ
served in DFT calculations (Figure S6).
Figure 4. Piezoelectric response of 7HdiET molecule in 'bright' and 'dark' conformations. ∆f(z,VB) maps recꢀ
orded for the molecule 7HdiET in the (a) 'bright' and (b) 'dark' conformation. Positions of the minima of the freꢀ
quency shift (highlighted with a black line) correspond to the contacting distances between the scanning probe tip
and molecule. The relative contacting distance zc changes with the applied sample bias indicating the piezoelecꢀ
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
tric deformation of the molecule with the applied electric field. The respective ballꢀandꢀstick contacting models
are in the insets.
1
2
Thus, the combination of the vertical dipole moment located within the helix and soft vibrational mode provides
optimal conditions for the converse piezoelectric response. To explore this aspect of the 7HdiET molecule, we
recorded frequency shift maps ∆f(z,VB) as a function of the tipꢀsample distance z and applied sample bias VB (see
Figure 4). Using the contacting distance zc (a distance at which the ∆f(z,VB) minima occur) as a reference, we can
track the mechanical deformation of the molecule induced by an applied external electric field with high preciꢀ
sion. Figure 4 reveals that zc varies linearly with the applied bias voltage. At positive sample bias (i.e. tip has a
negative effective charge) the tip has to approach closer to the molecule in order to contact it (the molecule is
compressed), while at negative sample bias zc is increased (the molecule is stretched). We found a very similar
linear variation of zc with the applied bias independently of the conformers (see Supplemental Table S1). This
suggests that the deformation of the molecule in the electric field is intrinsic to the helicene scaffold rather than to
the movement of the acetylsulfanyl group. Also, the orientation of the molecule dipole discussed explains well
the experimentally observed molecular deformation induced by the electric field, when the molecule contracts at
the positive sample bias (i.e. negatively charged tip apex) and vice versa.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Our ∆f(z,VB) spectra reveal a parabolic dependence of the frequency shift (i.e. force) on applied bias, see Figure
S7, which cannot explain the linear deformation. Thus, we attribute it to a vertical linear relaxation of the helical
scaffold (altering the helix pitch) under the applied bias. Indeed, this is supported by the DFT calculations with
applied external electric field (see Figure S14) showing a dominant vertical relaxation in the helix. We should
note that calculated molecular deformations are one order of magnitude lower than those observed experimentalꢀ
ly. We attribute this discrepancy to inherent problems to describe properly charge transfer at organometallic interꢀ
faces and their polarization. Second, in the experiment the electric field can be enhanced locally by presence of
metallic tip (as in e.g. tipꢀenhanced Raman spectroscopy). Therefore, we cannot expect the perfect agreement
between the magnitude of deformation obtained from DFT calculations and the experiment.
It is also worth noting that we detected a very similar slope zc(VB) for the 'bright' and 'dark' conformers, while the
tunneling current was very different (see Figure 3b). This supports the conclusion that the PE observed is not
affected by any crossꢀtalk between the frequency shift and tunneling current channels36.
The piezoelectric constant d33 of the molecule 7HdiET can be estimated from the linear slope zc(VB) as follows:
∆ꢑ(ꢒ)
ꢑꢊ
1
ꢏꢐꢐ
ꢁ
∗
,
ꢒ
where ∆z is a change in the molecular length for a given applied electric field E and zo represents the molecular
length at the zeroꢀelectric field. In fact, the electric field E is a priori unknown in the experiment. Nevertheless, in
a limiting case, assuming a perfect screening of the applied bias in metals and its linear decay in the molecular
ꢓ
ꢔꢎ
junction, we can estimate the electric field in the junction as ꢒ ꢁ
, where zts is the distance between the
ꢑꢕꢅ
metallic surface and tip apex obtained from the DFT calculations37,38,39 (see Figure S1) and VB is the applied samꢀ
ple bias. Adopting this approximation, we estimate the average piezoelectric constants d33 for the 'bright' and
'dark' conformers as 54 and 53 pm/V, respectively. We obtained very similar results for different molecules during
different sessions (see Figure S15). Lacking experimental values for single molecules, we compare these values
to polymers (e.g. PVDF: 33 pm/V), inorganic materials such as ZnO (e.g. 4ꢀ12 pm/V) or biological materials
such as thin films of bacteriophage (e.g. 8 pm/V). However, we should note, that in the present experiments sinꢀ
gle molecules are exposed to much larger electric field than those typically applied in in traditional mesoscopic
PE materials.
We also measured the electromechanical response of the molecule 7HdiT that shares the heptahelicene scaffold
with the molecule 7HdiET, but is functionalized with the native sulfanyl groups instead of acetylsulfanyl (see
Supplementary Discussion and Figure S16). Similarly to 7HdiET, it is possible to contact the molecule 7HdiT and
measure the corresponding ∆f(z,VB) maps. Even though the molecule 7HdiT interacts differently with the scanꢀ
ning probe tip than 7HdiET, is it still possible to reproducibly contact the molecule 7HdiT within the range of
ACS Paragon Plus Environment

Page 7 of 10
1
2
3
Journal of the American Chemical Society
±0.8 V of the sample bias voltage. The acquired ∆f(z,VB) spectra on the molecules 7HdiT reveal a significant conꢀ
verse piezoelectric effect as well, having a slightly larger value of 80 pm/V of the d33 coefficient than found at the
'bright' and 'dark' conformers of 7HdiET ~ 50 pm/V.
Let us summarize a possible origin of the strong converse piezoelectric response of the racemic heptahelicene
derivatives. The molecule 7HdiET in gas phase does not possess any significant vertical dipole moment aligned
with the helix axis because of its C2 symmetry. Thus, at first glance it seems to be disqualified as a candidate for
singleꢀmolecule PE. However, once deposited on a metallic surface, its C2 symmetry is broken to form a C1 symꢀ
metric adsorbate lacking the inversion symmetry. On top of this, the charge transfer from the molecule to surface
creates the vertical dipole within the molecule helix, which coupled to the helix soft vibrational mode provides
strong PE. This scenario holds for both enantiomeric M and P forms of helicenes. Therefore, the PE effect is inꢀ
dependent of molecular chirality. We should note that changes in the molecular tilt angle induced by the applied
electric field would result in a similar electromechanical deformation as reported here. While we cannot comꢀ
pletely rule this scenario out, there are several arguments based on measurements and simulations (detailed in
Supporting Information) disfavoring this interpretation.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In addition, we performed a control measurement on isolated thiolꢀfunctionalized oꢀcarborane CBdiT molecules.
They consist of a compact core of covalently linked boron and carbon atoms (Figure S17). Therefore, the elecꢀ
tromechanical deformation may only take place in thiol bonds on the interface between the molecule and surface.
The compact character of the molecular core and presence of two thiol bonds also minimize possibility of any
molecular tilt or rotation. The acquired ∆f(z) curves show different character than those acquired on helicene
molecules. Namely, we do not observe the characteristic jump to the contact observed in helicenes, but a gradual
decrease of the frequency shift upon tip approach. This behavior is very similar to that observed in single atomic
metallic contacts (see e.g.46). This striking difference in the ∆f(z) spectra indicates a distinct mechanical response
of oꢀcarborane and helicene molecules. The rigid internal structure of the oꢀcarborane do not allow any sudden
jump to the contact, while for helicenes the presence of the soft helix mode does.
The measured electromechanical linear response of CBdiT was ~10 pm/V. Though not negligible, this value is at
least three times lower than for helicenes (≳ 30 pm/V). These results reveal, on one hand, that thiolate bonds may
experience some electromechanical deformation too but, on the other hand, that the helical scaffold plays an imꢀ
portant role in enhancing the molecular piezoelectric response. This helical structure brings soft vibrational
modes where the molecule is deformed along the axis of the helix.
Importantly, there is significant room for the further enhancement of the singleꢀmolecule (converse) PE either
by increasing the piezoelectric coefficient of the molecule (cf. ref.29) or by varying the surface workfunction to
increase the charge transfer between the molecule and metal (i.e. the vertical dipole). The existence of the singleꢀ
molecule converse PE indicates that the singleꢀmolecule direct PE effect could also be realized as these effects
are complementary and reversible. The first experimental observation of the converse singleꢀmolecule PE introꢀ
duces a new concept into the bottomꢀup design of new molecular nanoelectromechanical devices such as actuaꢀ
tors, sensors or electricity generators. Not only polarized molecules on the metallic surface reported here but also
isolated dipolar molecules29, as analyzed by Hutchison32, may constitute diverse PE materials at nanoꢀ, microꢀ or
macroscale. In particular, the so far neglected PE moleculeꢀelastomer hybrid materials are envisaged to find
meaningful applications in downsized soft robotics42,43complementing, for instance, dielectric electroactive elasꢀ
tomers44 but enabling much lower driving voltages45.
3. CONCLUSIONS
In this work, we investigate the piezoelectric effect in single helicene molecules on Ag(111) surface, using simulꢀ
taneous nonꢀcontact AFM and STM measurements. The helicene on the substrate forms circular and linear isꢀ
lands, which can be effectively disassembled to individual molecules by the AFM tip, to exclude collective influꢀ
ence on the piezoelectric measurements. We performed the biasꢀdependent ∆f(z,VB) spectroscopy, which reveals
the strong linear shift of the contact distances with applied bias. We attributed the linear biasꢀinduced deforꢀ
mations of the helicene molecules to the piezoelectric response of the molecule. In addition, we found that the
piezoelectric response is independent of the acetylsulfanyl group conformation and that it is free from crossꢀtalk
artifacts. We corroborated the experimental evidence by the total energy DFT simulations. The calculations reꢀ
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 10
veal strong charge transfer between the helicene molecule and the metallic substrate, which gives rise to a vertical
electric dipole across the molecule. The presence of the electric dipole coupled with a soft vibrational mode inꢀ
herent to the helical structure of the molecule originates the strong piezoelectric effect. The role played by the
helical structure in amplifying the electromechanical response was evidenced from control experiments on rigid
thiolꢀfunctionalized oꢀcarborane molecules. The presence a smaller but still nonꢀnegligible electromechanical
response in the CBdiT molecules points to importance of the electromechanical phenomenon in molecular juncꢀ
tions. We believe that the presented data can stimulate further investigation of the electromechanical response in
the molecular junction and an exploitation of this effect for design of new molecular machines.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4. METHODS
Heptahelicene derivatives (S,S'ꢀheptaheliceneꢀ2,17ꢀdiyl diethanethioate, 7HdiET and heptaheliceneꢀ2,17ꢀ
dithiol, 7HdiT) were synthesized using the procedure described in Supporting Information. The molecules were
evaporated under ultrahigh vacuum (UHV, 3 x 10ꢀ10 mbar) from a tantalum crucible at ~350 K (heptahelicene
derivatives) or ~400 K (oꢀcarboranes) onto the atomically clean Ag(111) singleꢀcrystal surface at ~350 K. The
imaging and spectroscopies of the individual molecules and were performed by highꢀresolution nonꢀcontact atomꢀ
ic force microscopy (ncꢀAFM) and scanning tunneling microscopy (STM) under UHV at low temperature (4.4 K
5 K) using a commercial microscope (SPECS Surface Nano Analysis GmbH) with simultaneous force/current
detection capabilities. Prior to the experiments, the probe tip was repeatedly poked into the Ag(111) surface (usuꢀ
ally at least 10 nm away from any closest molecular islands) applying 0ꢀ3V bias pulses until a good tip terminaꢀ
tion capable of molecule contacting was obtained. The optimized structures of molecules on Ag(111), and their
interaction with metallic probe were calculated by the FHIꢀAIMS program packages35 based on ab initio density
functional theory (DFT)36,37. Electron transport calculations were carried out using the TransSiesta code32,33 (for
details, see Supplementary Methods).
5. ASSOCIATED CONTENT
Supporting Information and chemical compound information are available in the online version of the paper. This
material is available free of charge via the Internet at http://pubs.acs.org. Correspondence and requests for materiꢀ
als should be addressed to P.J. and I.S.
Figures S1ꢀS18 with additional discussion, Table S1 and synthesis of molecules 7HdiET and 7HdiT.
6. AUTHOR INFORMATION
Corresponding Author
*pavel.jelinek@fzu.cz
*stary@uochb.cas.cz
Notes
The authors declare no competing financial interest.
7. ACKNOWLEDGMENT
We thank T. Baše for providing with us the oꢀcarborane CBdiT molecules. This work was financially supported
by a Czech Science Foundation grants (17ꢀ24210Y, 16ꢀ08327S, 15ꢀ19672S and 14ꢀ374527G) and by the Institute
of Organic Chemistry and Biochemistry, Czech Academy of Sciences (RVO: 61388963). H. V. gratefully
acknowledges financial support from the Purkyně Fellowship program of the Czech Academy of Sciences. We
thank the National Grid Infrastructure MetaCentrum and the "Projects of Large Research, Development, and Inꢀ
novations Infrastructures" CESNET LM2015042 and LM2015087. P. J. acknowledges support of the Czech
Academy of Sciences through Praemium Academiae award. The authors declare no conflicts of interest.
8. REFERENCES
(1) Advanced Piezoelectric Materials, Science and Technology; Uchino, K., Ed.; Woodhead Publishing: Camꢀ
bridge, 2017.
ACS Paragon Plus Environment

Page 9 of 10
1
Journal of the American Chemical Society
(2) Zhu, H.; Wang, Y.; Xiao, J.; Liu, M.; Xiong, S.; Wong, Z. J.; Ye, Z.; Ye, Y.; Yin, X.; Zhang, X. Nat. Nanoꢀ
technol. 2015, 10, 151–155.
2
3
4
5
6
7
8
(3) Wu, W.; Wang, L.; Li, Y.; Zhang, F.; Lin, L.; Niu, S.; Chenet, D.; Zhang, X.; Hao, Y.; Heinz, T. F.; Hone, J.;
Wang, Z. L. Nature 2014, 514, 470–474.
(4) Nguyen ,T. D.; Nguyen, T. D.; Deshmukh, N.; Nagarah, J. M.; Kramer, T.; Purohit, P. K.; Berry, M. J.;
McAlpine, M. C. Nat. Nanotechnol. 2012, 7, 587–593.
(5) MinaryꢀJolandan, M.; Bernal, R. A.; Kuljanishvili, I.; Parpoil, V.; Espinosa, H. D. Nano Lett. 2012, 12, 970–
976.
(6) Wang, Z. L.; Song, J. Science 2006, 312, 242–246.
(7) Fukada, E. Q. Rev. Biophys. 1983, 16, 59–87.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(8) Farrar, D.; Farrar, D.; Ren, K.; Cheng, D.; Kim, S.; Moon, W.; Wilson, W. L.; West, J. E.; Yu, S. M. Adv.
Mater. 2011, 23, 3954–3958.
(9) Lee, B.Y.; Zhang, J.; Zueger, C.; Chung, W. J.; Yoo, S. Y.; Wang, E.; Meyer, J.; Ramesh, R.; Lee, S. W. Nat.
Nanotechnol. 2012, 7, 351–356.
(10) Masmanidis, S. C.; Masmanidis, S. C.; Karabalin, R. B.; De Vlaminck, I.; Borghs, G.; Freeman, M. R.;
Roukes, M. L. Science 2007, 317, 780–783.
(11) Craighead , H. G. Science 2000, 290, 1532–1535.
(12) Elbing, M.; Ochs, R.; Koentopp, M.; Fischer, M.; von Hänisch, C.; Weigend, F.; Evers, F.; Weber, H. B.;
Mayor, M. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 8815–882.
(13) Park, H.; Park, J.; Lim, A. K.; Anderson, E. H.; Alivisatos, A. P.; McEuen, P. L. Nature 2000, 407, 57–60.
(14) Perrin, M. L.; Frisenda, R. Koole, M.; Seldenthuis, J. S.; Celis Gil, J. A.; Valkenier, H.; Hummelen, J. C.;
Renaud, N.; C. Grozema, F.; Thijssen, J.M.; Dulić, D.; van der Zant, H. S. J. Nat. Nanotechnol. 2014, 9, 830–834.
(15) Reddy, P.; Jang, S. Y.; Segalman, R. A.; Majumdar, A. Science 2007, 315, 1568–1571.
(16) Zheng, X.; Mulcahy, M. E.; Horinek, D.; Galeotti, F.; Magnera, T. F.; Michl, J. J. Am. Chem. Soc. 2004,
126, 4540–4542.
(17) Tierney, H. L.; Murphy, C. J.; Jewell, A. D.; Baber, A. E.; Iski, E. V.; Khodaverdian, H. Y.; McGuire, A. F.;
Klebanov, N.; Sykes, E. C. H. Nat. Nanotechnol. 2011, 6, 625–629.
(18) Leigh, D. A.; Wong, J. K.; Dehez, F.; Zerbetto, F. Nature 2003, 424, 174–179.
(19) Kelly, T. R.; Silva, R. A.; Silva, H. D.; Jasmin, S.; Zhao, Y. A. J. Am. Chem. Soc. 2000, 122, 6935–6949.
(20) Koumura, N.; Zijlstra, R. W.; van Delden, R. A.; Harada, N.; Feringa, B. L. Nature 1999, 401, 152–155.
(21) ErbasꢀCakmak, S.; Leigh, D. A.; McTernan, C. T.; Nussbaumer, A. L. Chem. Rev. 2015, 115, 10081–
10206.
(22) Bruns, C. J.; Stoddart, J. F. Acc. Chem. Res. 2014, 47, 2186–2199.
(23) Kudernac, T.; Ruangsupapichat, N.; Parschau, M.; Maciá, B.; Katsonis, N.; Harutyunyan, S. R.; Ernst, K.
H.; Feringa, B. L. Nature 2011, 479, 208–211.
(24) Shirai, Y.; Osgood, A. J.; Zhao, Y.; Kelly, K. F.; Tour, J. M. Nano Lett. 2005, 5, 2330–2334.
(25) Chen, C. F.; Shen, Y. Helicene Chemistry: From Synthesis to Applications; SpringerꢀVerlag: Berlin, 2017.
(26) Šesták, P.; Wu, J.; He, J.; Pokluda, J.; Zhang, Z. Phys. Chem. Chem. Phys. 2015, 17, 18684–18690.
(27) Guo, Y. D.; Yan, X. H.; Xiao, Y.; Liu, C. S. Sci. Rep. 2015, 5, 16731.
(28) Vacek, J.; Chocholoušová, J. V.; Stará, I. G.; Starý, I.; Dubi, Y. Nanoscale 2015, 7, 8793–8802.
(29) Rulíšek, L.; Exner, O.; Cwiklik, L.; Jungwirth, P.; Starý, I.; Pospíšil, L.; Havlas, Z. J. Phys. Chem. C 2007,
111, 14948–14955.
(30) Rempala, P.; King, B. T. J. Chem. Theory Comput. 2006, 2, 1112–1118.
(31) Jalaie, M.; Weatherhead, S.; Lipkowitz, K. B.; Robertson, D. Electron. J. Theor. Chem. 1997, 2, 268–272.
(32) Quan, X.; Marvin, C. W.; Seebald, L.; Hutchison, G. R. J. Phys. Chem. C 2013, 117, 16783–16790.
(33) Kitagawa, K.; Morita, T.; Kimura, S. A. Angew. Chem. Int. Ed. 2005, 44, 6330–6333.
(34) Soler, J. M.; Artacho, E.; Gale, J. D.; García, A.; Junquera, J.; Ordejón, P.; SánchezꢀPortal, D. J. Phys.
Condens. Matter 2002, 14, 2745.
(35) Brandbyge, M.; Mozos, J. L.; Ordejón, P.; Taylor, J.; Stokbro, K. Phys. Rev. B: Condens. Matter Mater.
Phys. 2002, 65, 165401.
(36) Majzik, Z.; Setvín, M.; Bettac, A.; Feltz, A.; Cháb, V.; Jelínek, P. Beilstein J. Nanotech. 2012, 3, 249.
(37) Blum, V.; Gehrke, R.; Hanke, F.; Havu, P.; Havu, V.; Ren, X.; Reuter, K.; Scheffler, M. Comput. Phys.
Commun. 2009, 180, 2175–2196.
(38) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865.
(39) Tkatchenko, A.; Scheffler, M. Phys. Rev. Lett. 2009, 102, 073005.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
(40) Koch, N. ChemPhysChem 2007, 8, 1438–1455.
(41) Ishii, H.; Sugiyama, K.; Ito, E.; Seki, K. Adv. Mater. 1999, 11, 605–625.
(42) Rus, D.; Tolley, M. T. Nature 2015, 521, 467–475.
(43) Ilievski, F.; Mazzeo, A. D.; Shepherd, R. F.; Che, X.; Whitesides, G. M. Angew. Chem. Int. Ed. 2011, 50,
1890–1895.
(44) Keplinger, C.; Sun, J. Y.; Foo, C. C.; Rothemund, P.; Whitesides, G. M.; Suo, Z. Science 2013, 341, 984–
987.
(45) Carpi, F.; Bauer, S.; De Rossi, D. Science 2010, 330, 1759–1761.
(46) Ternes, M.; Gonzalez, C.; Lutz, C. P.; Hapala, P.; Giessibl, F. J.; Jelinek P.; Heinrich A. J.; Phys. Rev. Lett.
2011, 106, 016802.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment