Impact of molecular symmetry on single-molecule conductance

Article abstract of DOI:10.1021/ja4055367

We have measured the single-molecule conductance of a family of bithiophene derivatives terminated with methyl sulfide gold-binding linkers using a scanning tunneling microscope based break-junction technique. We find a broad distribution in the single-molecule conductance of bithiophene compared with that of a methyl sulfide terminated biphenyl. Using a combination of experiments and calculations, we show that this increased breadth in the conductance distribution is explained by the difference in 5-fold symmetry of thiophene rings as compared to the 6-fold symmetry of benzene rings. The reduced symmetry of thiophene rings results in a restriction on the torsion angle space available to these molecules when bound between two metal electrodes in a junction, causing each molecular junction to sample a different set of conformers in the conductance measurements. In contrast, the rotations of biphenyl are essentially unimpeded by junction binding, allowing each molecular junction to sample similar conformers. This work demonstrates that the conductance of bithiophene displays a strong dependence on the conformational fluctuations accessible within a given junction configuration, and that the symmetry of such small molecules can significantly influence their conductance behaviors.

Full text of DOI:10.1021/ja4055367

Communication
pubs.acs.org/JACS
Impact of Molecular Symmetry on Single-Molecule Conductance
Emma J. Dell,†,§ Brian Capozzi,‡,§ Kateri H. DuBay,† Timothy C. Berkelbach,† Jose Ricardo Moreno,†
David R. Reichman,† Latha Venkataraman,*,‡ and Luis M. Campos*,†
†Department of Chemistry, Columbia University, New York, New York 10027, United States
‡Department of Applied Physics and Mathematics, Columbia University, New York, New York 10027, United States
S
* Supporting Information
electrodes. The rotational barrier for the more symmetric
ABSTRACT: We have measured the single-molecule
conductance of a family of bithiophene derivatives
terminated with methyl sulfide gold-binding linkers using
a scanning tunneling microscope based break-junction
technique. We find a broad distribution in the single-
molecule conductance of bithiophene compared with that
of a methyl sulfide terminated biphenyl. Using a
combination of experiments and calculations, we show
that this increased breadth in the conductance distribution
is explained by the difference in 5-fold symmetry of
thiophene rings as compared to the 6-fold symmetry of
benzene rings. The reduced symmetry of thiophene rings
results in a restriction on the torsion angle space available
to these molecules when bound between two metal
electrodes in a junction, causing each molecular junction
to sample a different set of conformers in the conductance
measurements. In contrast, the rotations of biphenyl are
essentially unimpeded by junction binding, allowing each
molecular junction to sample similar conformers. This
work demonstrates that the conductance of bithiophene
displays a strong dependence on the conformational
fluctuations accessible within a given junction config-
uration, and that the symmetry of such small molecules
can significantly influence their conductance behaviors.
biphenyl is only slightly impacted by the binding event, while
that for the less symmetric bithiophene is significantly altered.
This difference in rotational barriers is manifested in the
difference in conductance distributions measured for the two
molecules.
In order to investigate the role of molecular backbone
symmetry on the charge transport characteristics measured at
the single-molecule level, we first compare the conductance of a
bithiophene derivative (T2) terminated with methyl sulfide end
groups that bind gold electrodes with the biphenyl analog
(Figure 1A).5 Thiophene moieties were synthesized by using
Figure 1. (A) Structure of T2 (blue) and P2 (gray) and a schematic of
the scanning tunneling microscope break-junction (STM-BJ)
technique. (B) Sample conductance traces measured for P2 and T2
at 90 mV.
he drive to miniaturize the active components in
Telectronic devices requires that we understand the
transport characteristics of molecular scale circuits.1 Measure-
ment of the conductance of single metal-molecule-metal
junctions has thus become an important characterization tool
in the sub-10 nm scale.2 In such measurements, it is well-known
that the conductance of single molecule junctions is sensitive to
several experimental parameters such as the electrode structure,
the orientation of the molecule-metal bonds, and the
conformation of the molecular backbone.3 The pervasiveness
of thiophenes in bulk organic electronic and photonic devices4
makes their properties in single molecule devices of particular
interest. In comparing the charge transport of a bithiophene
derivative with that of a biphenyl derivative, we show here that
conductance is also dependent on the degree of symmetry
given by the molecular backbone in a metal-molecule-metal
junction. Specifically, we find that the difference in ring
symmetry, 6-fold for phenyl rings versus 5-fold for thiophene
rings, strongly impacts the rotational barrier that the less
symmetric bithiophene is subject to when bound between two
both Stille and Suzuki cross-coupling reactions.6 The methyl
sulfide end groups were introduced through deprotonation with
n-butyl lithium, followed by a nucleophilic reaction with
dimethyl disulfide. The compounds were purified using column
1
chromatography and characterized by H NMR, 13C NMR,
mass spectrometry, and UV−vis absorption (Supporting
Information (SI)). The biphenyl derivative, 4,4′-bis-
(methylsulfide)biphenyl (P2), was purchased from Sigma
Aldrich and used without purification.
We measured the conductance (current/voltage) of the
molecules using a scanning tunneling microscope break-
junction (STM-BJ) technique under ambient conditions.2a,d
Gold atomic point contacts were repeatedly formed and then
broken by driving the gold tip in and out of contact with a gold-
on-mica substrate. Individual molecular junctions were formed
when the gold point contacts were broken in a solution of the
Received: June 3, 2013
Published: August 1, 2013
© 2013 American Chemical Society
11724
dx.doi.org/10.1021/ja4055367 | J. Am. Chem. Soc. 2013, 135, 11724−11727

Journal of the American Chemical Society
Communication
target molecule (10 mM concentration, 90 mV bias voltage) in
1,2,4-trichlorobenzene (TCB). Thousands of conductance−
displacement traces were collected; each trace displays plateaus
close to integer multiples of the quantum of conductance, G0
(2e2/h), and an additional plateau-like feature at a molecule-
specific conductance range (Figure 1B). These additional
plateaus indicate that a conducting metal-molecule-metal bridge
is formed after the gold point contact is ruptured.
To determine the most frequently measured conductance
value for these two molecules, we compile all measured
conductance traces into logarithmically binned one-dimen-
sional histograms.7 We use logarithmically binned histograms
to highlight the conductance peaks as a linear-binned
histogram, for T2 displays a very broad feature which cannot
be characterized easily; this excessive breadth is unique to this
molecule and can be placed into context by comparing it with
the width of the P2 conductance distribution (Figure 2A and
represent an average conductance through the various con-
formers that are energetically accessible in the junction. If these
freely rotating conformers are locked when bound in a junction,
we would expect to see a difference in conductance
distributions. The rotational degrees of freedom for a 6-fold
phenyl system are different from that of a 5-fold thiophene
system when considering the rotation axis defined by the
attachment points of the linker to the Au electrodes. Rotations
in biphenyl are almost unaffected by containment in a junction,
but for a bithiophene bound in a junction, the barrier to
rotation is increased since changing the torsion angle
necessitates either moving the gold atom to which the molecule
is attached (Figure 2E) or deforming the rest of the structure to
accommodate the change. Thus, in single biphenyl junctions,
conductance will be independent of the molecule’s conforma-
tion, while in bithiophene junctions conductance will be
strongly dependent on the molecule’s conformation imposed
by its binding geometry. This striking difference yields a
conductance distribution for T2 that spans over 2 orders of
magnitude at full width at half max (fwhm), almost twice that of
P2.
Within this reasoning, junction elongation also plays a crucial
role in the measured conductance distributions. From Figure
1B, we see that individual T2 conductance traces display more
variation than the P2 ones. As the junction is elongated, the
molecule generally alters its binding sites on the electrodes,
thus each measured trace consists of conductance measure-
ments of different structures.8 For P2, such changes in
geometry would not be expected to impact the measured
conductance over the course of a single conductance trace. The
molecule’s inter-ring rotations will be unaffected, and the
measured conductance over the course of a trace will continue
to be an average over all such conformations. This situation is
drastically different for T2. Since rotations for this molecule are
restricted based upon its binding geometry, changes in junction
geometry due to elongation will manifest as changes in
conductance, as is seen in the sample traces shown in Figure 1.
In order to experimentally investigate the relation between
conductance and allowed rotations in T2, we synthesized two
other bithiophene derivatives (Figure 2C and 2D): T2-flat,
where the internal rotation between the aromatic rings is locked
by a saturated linker, and T2-twist, where hexyl chains at the
3,3′ positions force the molecule into a twisted conformation
(synthetic details are given in the SI). These derivatives restrict
the rotation around the inter-ring torsion, and therefore
represent “frozen snapshots” of the rotation. The 1D
logarithmically binned conductance histograms of these two
molecules are shown in Figure 2C and 2D. We see that the
conductance distributions for both the twisted and planar
bithiophene derivatives are considerably narrower than that of
T2. In addition, the conductance peak of T2-flat overlaps the
higher-conducting portion of the conductance peak of T2,
which would be expected from its increased conjugation due to
the forced planarity. In contrast, the peak of T2-twist overlaps
the lower-conductance portion of the T2 peak; this would also
be expected from its reduced conjugation. Furthermore, the
conductance distributions for P2 and T2-flat are of similar
width, which is to be expected.2d Although P2 has more
rotational freedom, the time scales of the measurement (10 μs)
are large enough that all of the conformational fluctuations are
averaged, so a narrow conductance distribution is indeed
expected.
Figure 2. Logarithmically binned conductance histograms of (A)
biphenyl (P2); (B) bithiophene (T2); (C) T2-twist; and (D) T2-flat
showing a Gaussian fit with fwhm indicated by the arrows. (E)
Schematic illustrating how ring symmetry impacts rotations for P2 and
T2 when bound in a junction.
2B). For measurements with the STM-BJ technique where
thousands of junction structures are sampled to determine the
most frequently observed conductance values, the width in the
distribution reflects variations in conductances due to variations
in molecular junction structure. Junction structure includes the
electrode structure, the orientation of the Au−S−C donor−
acceptor bonds relative to the molecular backbone,3d and the
average dihedral twist angle between the two rings which is
constrained by the details of the electrode structure.2d Thus in
these two-ring systems, some width to the conductance
distribution is expected. What is surprising, however, is that
the conductance distribution of T2 is much greater than that of
P2, and as both systems include the same electrodes and
methyl sulfide binding groups, we surmise that a factor intrinsic
to the molecular structure is the cause.
At room temperature, when not bound in any junction, both
phenyl rings and thiophene rings are able to freely rotate due to
thermal fluctuations. Since the time scales of the rotations are
significantly smaller than the time scales of our measurements
(100 μs), we expect that the conductance measurements
11725
dx.doi.org/10.1021/ja4055367 | J. Am. Chem. Soc. 2013, 135, 11724−11727

Journal of the American Chemical Society
Communication
To probe the relation between conductance and symmetry,
we calculated the torsional potential energy curves for biphenyl
and the bithiophene derivatives using OPLS-SB-T, a classical
force field that has been specifically designed to correctly
represent the effects of conjugation in such biaryl systems.9 For
each system, the minimum energy unbound structure was
determined while restricting the torsional angle, θ, to a value
between 0° to 180° in steps of 10°. This resulted in the solid
potential energy curves shown in Figure 3A−3D. In order to
which are several times the expected thermal fluctuations of the
system at room temperature, the observed distribution of
torsion angles in bithiophene will be significantly biased by the
binding orientation of each molecule in the break junction.
Since every measurement run has a different electrode structure
(and this structure is changed as the junction is elongated), we
expect the distributions of bithiophene torsional angles to vary
significantly from junction to junction, and possibly also within
each junction as it is elongated. As a result, the conductance
measurements capture variations in junction binding, widening
the measured conductance peak.
To further understand how the trapping of molecules in a
junction might slow or restrict inter-ring rotations, we carried
out room temperature molecular dynamics (MD) simulations
of the four molecules. The details of the calculations are given
in the SI, but, briefly, we ran atomistic gas phase simulations
within the NVT ensemble using the OPLS-SB-T potential.9 In
an initial step, all atoms were allowed to move without
restriction, as would be the case for the molecules prior to their
binding in the junction. Then, in the second step, the distance
between the molecules’ terminal sulfurs was held fixed in order
to simulate the binding event (the pulling forces were not
represented in this model). After some equilibration time, the
average value of the inter-ring dihedral angle, θ, and the average
value of cos2(θ) (which is proportional to the conductance, as
shown in SI Figure S22d) were measured for each run.
Repeating this procedure for 500 different “binding” config-
urations produced significantly broader estimated conductance
histograms for bithiophene than biphenyl, confirming that the
significant differences in the potential energy curves are
manifested in the room temperature dynamics of the different
molecules (Table S1). Figure 3 E−H displays histograms of the
inter-ring torsions angles sampled for the first 10 restricted MD
simulations for all molecules. It is evident that the distributions
are nearly identical for all biphenyl runs, whereas the
distributions vary significantly from run to run for bithiophene.
This further demonstrates that the conductance measured in a
single bithiophene trace is limited to distinct sets of conformers
dependent on the molecule’s various geometries as the junction
extends.
Figure 3. (A−D) Calculated unrestricted (solid) and restricted
(dashed) potential energy curves as a function of internal torsion
angle. (Note: Only one restricted curve is shown for P2 and T2-flat as
detailed in the text.) (E−H) The distributions of torsion angles
sampled during the first 10 restricted MD simulations (out of 500),
where the distance between their Au-binding sulfur atoms is restricted
to its initial value to reproduce the constraints imposed on each
molecule when it binds in the break junction. Each curve represents
the torsion angle distribution from an individual run.
explore the additional barriers to torsional fluctuations that
could be induced upon binding in the junction, the structures
were then minimized with an additional restriction on the
distance between the two sulfur atoms that act as Au-
attachment sites. For P2 and T2-flat, these atoms were frozen
in their optimal positions when θ = 0° (as obtained from the
minimized structure in the first step). For T2 and T2-twist,
these atoms were frozen in their optimal positions both when θ
= 0° and when θ = 180°. The θ = 180° case was not
investigated for P2, as it is identical to the θ = 0° case, or for
T2-flat as it is clearly an inaccessible configuration. The
potential energy curves calculated in the presence of these
additional restrictions are shown in Figure 3A−3D as dashed
lines.
We see that for P2 the restricted and unrestricted geometry
potential energy curves are quite similar (Figure 3A),
suggesting that junction binding should not bias the
distribution of inter-ring torsional angles sampled during a
conductance measurement. In contrast, the torsional potential
of T2 is significantly modified when the Au-binding S atoms are
restricted, as evident in Figure 3B. This is due to the 5-fold
symmetry of the thiophene ring, which imposes a constraint as
discussed above. Given the heights of the barriers introduced
when the distance between the Au-binding S atoms is fixed,
In summary, we have seen that bithiophene exhibits a broad
conductance distribution as compared to biphenyl. By applying
a combination of experiment and theory, we have shown that
the reduced symmetry of bithiophene leads to a restriction of
its inter-ring torsion rotations when confined in a junction. Our
work shows the importance of considering the molecular
backbone symmetry in designing functional conducting
molecular devices.
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthetic, measurement, and computational details. This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
lcampos@columbia.edu; lv2117@columbia.edu
■
Author Contributions
§E.J.D. and B.C. contributed equally.
Notes
The authors declare no competing financial interest.
11726
dx.doi.org/10.1021/ja4055367 | J. Am. Chem. Soc. 2013, 135, 11724−11727

Journal of the American Chemical Society
Communication
ACKNOWLEDGMENTS
■
We thank Dr. Jay Henderson for providing us with a sample of
T2. The experimental portion of this work was supported by
the NSF (DMR-1206202). The computational work was
funded by the Center for Re-Defining Photovoltaic Efficiency
through Molecule Scale Control, an EFRC funded by the U.S.
Department of Energy, Office of Basic Energy Sciences (de-
sc0001085). E.J.D. thanks the HHMI and Dow Chemical
Company for International Research Fellowships. T.C.B. was
supported by a fellowship from the DOE (SCGF) under
Contract No. DE-AC05-06OR23100.
REFERENCES
■
(1) Aviram, A.; Ratner, M. A. Chem. Phys. Lett. 1974, 29, 277.
(2) (a) Xu, B. Q.; Tao, N. J. Science 2003, 301, 1221. (b) Reichert, J.;
Ochs, R.; Beckmann, D.; Weber, H. B.; Mayor, M.; von Lohneysen, H.
Phys. Rev. Lett. 2002, 88, 176804. (c) Zhao, A. D.; Li, Q. X.; Chen, L.;
Xiang, H. J.; Wang, W. H.; Pan, S.; Wang, B.; Xiao, X. D.; Yang, J. L.;
Hou, J. G.; Zhu, Q. S. Science 2005, 309, 1542. (d) Venkataraman, L.;
Klare, J. E.; Nuckolls, C.; Hybertsen, M. S.; Steigerwald, M. L. Nature
2006, 442, 904.
(3) (a) Li, C.; Pobelov, I.; Wandlowski, T.; Bagrets, A.; Arnold, A.;
Evers, F. J. Am. Chem. Soc. 2008, 130, 318. (b) Basch, H.; Cohen, R.;
Ratner, M. A. Nano Lett. 2005, 5, 1668. (c) Quek, S. Y.; Kamenetska,
M.; Steigerwald, M. L.; Choi, H. J.; Louie, S. G.; Hybertsen, M. S.;
Neaton, J. B.; Venkataraman, L. Nat. Nano. 2009, 4, 230. (d) Park, Y.
S.; Widawsky, J. R.; Kamenetska, M.; Steigerwald, M. L.; Hybertsen,
M. S.; Nuckolls, C.; Venkataraman, L. J. Am. Chem. Soc. 2009, 131,
10820. (e) Mishchenko, A.; Vonlanthen, D.; Meded, V.; Burkle, M.; Li,
C.; Pobelov, I. V.; Bagrets, A.; Viljas, J. K.; Pauly, F.; Evers, F.; Mayor,
M.; Wandlowski, T. Nano Lett. 2010, 10, 156.
(4) (a) Perepichka, I. F.; Perepichka, D. F. Handbook of thiophene-
based materials: applications in organic electronics and photonics; Wiley:
Chichester, U.K., 2009. (b) Sirringhaus, H.; Tessler, N.; Friend, R. H.
Science 1998, 280, 1741. (c) Sirringhaus, H.; Brown, P. J.; Friend, R.
H.; Nielsen, M. M.; Bechgaard, K.; Langeveld-Voss, B. M. W.;
Spiering, A. J. H.; Janssen, R. A. J.; Meijer, E. W.; Herwig, P.; de
Leeuw, D. M. Nature 1999, 401, 685. (d) Treat, N. D.; Campos, L. M.;
Dimitriou, M. D.; Ma, B.; Chabinyc, M. L.; Hawker, C. J. Adv. Mater.
2010, 22, 4982.
(5) Park, Y. S.; Whalley, A. C.; Kamenetska, M.; Steigerwald, M. L.;
Hybertsen, M. S.; Nuckolls, C.; Venkataraman, L. J. Am. Chem. Soc.
2007, 129, 15768.
(6) (a) Milstein, D.; Stille, J. K. J. Am. Chem. Soc. 1978, 100, 3636.
(b) Miyaura, N.; Yamada, K.; Suzuki, A. Tetrahedron Lett. 1979, 20,
3437.
(7) Gonzalez, M. T.; Wu, S. M.; Huber, R.; van der Molen, S. J.;
Schonenberger, C.; Calame, M. Nano Lett. 2006, 6, 2238.
(8) Kamenetska, M.; Koentopp, M.; Whalley, A.; Park, Y. S.;
Steigerwald, M.; Nuckolls, C.; Hybertsen, M.; Venkataraman, L. Phys.
Rev. Lett. 2009, 102, 126803.
(9) DuBay, K. H.; Hall, M. L.; Hughes, T. F.; Wu, C.; Reichman, D.
R.; Friesner, R. A. J. Chem. Theory Comput. 2012, 8, 4556.
11727
dx.doi.org/10.1021/ja4055367 | J. Am. Chem. Soc. 2013, 135, 11724−11727