Aromaticity decreases single-molecule junction conductance

Article abstract of DOI:10.1021/ja411143s

We have measured the conductance of single-molecule junctions created with three different molecular wires using the scanning tunneling microscope-based break-junction technique. Each wire contains one of three different cyclic five-membered rings: cyclopentadiene, furan, or thiophene. We find that the single-molecule conductance of these three wires correlates negatively with the resonance energy of the five-membered ring; the nonaromatic cyclopentadiene derivative has the highest conductance, while the most aromatic of this series, thiophene, has the lowest. Furthermore, we show for another wire structure that the conductance of furan-based wires is consistently higher than for analogous thiophene systems, indicating that the negative correlation between conductance and aromaticity is robust. The best conductance would be for a quinoid structure that diminishes aromaticity. The energy penalty for partly adopting the quinoid structure is less with compounds having lower initial aromatic stabilization. An additional effect may reflect the lower HOMOs of aromatic compounds.

Full text of DOI:10.1021/ja411143s

Communication
pubs.acs.org/JACS
Aromaticity Decreases Single-Molecule Junction Conductance.
Wenbo Chen,† Haixing Li,‡ Jonathan R. Widawsky,‡ Chandrakumar Appayee,† Latha Venkataraman,*,‡
and Ronald Breslow*,†
†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
cyclic units and use resonance energy as a measure of its
ABSTRACT: We have measured the conductance of
single-molecule junctions created with three different
molecular wires using the scanning tunneling micro-
scope-based break-junction technique. Each wire contains
one of three different cyclic five-membered rings: cyclo-
pentadiene, furan, or thiophene. We find that the single-
molecule conductance of these three wires correlates
negatively with the resonance energy of the five-membered
ring; the nonaromatic cyclopentadiene derivative has the
highest conductance, while the most aromatic of this
series, thiophene, has the lowest. Furthermore, we show
for another wire structure that the conductance of furan-
based wires is consistently higher than for analogous
thiophene systems, indicating that the negative correlation
between conductance and aromaticity is robust. The best
conductance would be for a quinoid structure that
diminishes aromaticity. The energy penalty for partly
adopting the quinoid structure is less with compounds
having lower initial aromatic stabilization. An additional
effect may reflect the lower HOMOs of aromatic
compounds.
aromatic character. We compare conductance through a
thiophene, furan, and cyclopentadiene species. Cyclopentadiene
has 4π electrons in a conjugated system and is nonaromatic.
Furan and thiophene both have 6π electrons cyclically
delocalized. Because the cyclopentadiene does not have a
cyclic conjugated system, it is nonaromatic, while the furan and
thiophene are both aromatic with resonance energies of 16 and
29 kcal/mol, respectively.3,5 We expect that the conductance
will be greatest for the five-membered unit that is least aromatic
as it can most easily acquire a quinoidal structure, as illustrated
in Figure 1A.6 In this form, a better coupling between the Au
Figure 1. (A) Illustration of the quinoidal structure that maximizes
conductance across a five-membered ring. (B) Structure of the three
molecular wires investigated. (C) Sample conductance traces
measured for 1, 2, and 3 at a bias voltage of 225 mV. Inset: Schematic
of a single-molecule junction.
onjugated organic oligomer wires have formed the
1
Cbuilding blocks of organic electronics and photovoltaics.
One basic component in these wires is an aromatic unit, such as
a benzene or thiophene unit.2 From a chemical perspective,
these ring systems are more stable than noncyclic analogues
due to the delocalization of electrons within the ring. This
additional stability of the aromatic system is measured by
resonance energy or aromatic stabilization energy. The greater
the resonance energy for a unit, the more stable the system is.
For example, a benzene ring gains a (negative) resonance
energy of 36 kcal/mol, as the electron cloud of benzene is more
delocalized than that of a noncyclic system with three single
and double bonds.3 Previous measurements with acenes had
shown that the central ring of anthracene was a better
conductor than one ring of naphthalene, which was a better
conductor than a benzene ring.4 One explanation for this
finding was that the central ring of anthracene was the least
aromatic, while benzene was the most aromatic. However, these
acenes also had different numbers of electrons in the π-systems,
so a direct correlation between conductance and aromatic
character of the molecular wires was not unambiguous.
electrodes and the cyclic unit is achieved (the dashed double
C−C bond is shortened as shown in Figure 1A). With the
nonaromatic cyclopentadiene, achieving the quinoidal structure
does not cost any stabilization energy, in contrast to furan or
thiophene where such increased external π-character decreases
aromaticity.
We use the scanning tunneling microscope-based break-
junction (STM-BJ) technique7 to measure the conductance of a
series of three molecular wires (Figure 1B) that have a cyclic
five-member unit built into backbones. These three amine-
terminated molecular wires (Figure 1B): 1,4-bis(4-amino-
phenyl-ethinyl)-5,5-dimethylcyclopenta-1,3-diene (1), 2,5-bis-
(4-aminophenylethinyl)furan (2), and 2,5-bis(4-
aminophenylethinyl)thiophene (3) are synthesized following
procedures detailed in the Supporting Information document
Here we study the correlation between conductance and
aromaticity in a single-molecule junction containing five-atom
Received: October 31, 2013
Published: January 7, 2014
© 2014 American Chemical Society
918
dx.doi.org/10.1021/ja411143s | J. Am. Chem. Soc. 2014, 136, 918−920

Journal of the American Chemical Society
Communication
Figure 2. Log-binned one-dimensional conductance histograms for molecules 1, 2, and 3 generated without any data selection, from 10000 traces,
using a bin size of 100/decade. B Two-dimensional histograms for 1, 2, and 3 showing conductance peaks extending over a distance of ∼1 nm
relative to the break of the G0 contact.
(SI). These wires vary only in their aromatic character; they
have the same length and internal torsion angle, due to the
triple bonds and linker groups, which allows us to compare
conductance directly from aromaticity (or specifically reso-
nance energies). The amine terminations bind selectively to
under-coordinate gold in order to form a single-molecule
junction between two gold electrodes, thus enabling the STM-
BJ conductance measurements.7,8
across the entire molecular backbone, as the lengths of the
features seen in these 2D histograms are roughly the same for
all three.
To show that the correlation between the aromaticity and
conductance is robust, we also synthesized and measured
another simpler series of three molecules: 1,4-bis(4-amino-
phenyl)-5,5-dimethylcyclopenta-1,3-diene (4), 2,5-bis(4-
aminophenyl)furan (5), and 2, 5-bis(4-aminophenyl)thiophene
(6). Figure 3A shows the structures of these molecules, and
We measure the molecular conductance of these molecules
by repeatedly forming and breaking Au point contacts in the
presence of molecules using a modified STM-BJ setup that has
been described in detail previously.9 We use an Au/mica
substrate with an Au wire tip (Alfa-Aesar, 99.998% purity) for
these measurements. The STM operates in ambient conditions
at room temperature and the junctions are broken in a 0.2 mM
solution of the molecules in 1,2,4-trichlorobenzene (Alfa-Aesar
99% purity). Each conductance measurement starts by moving
the tip into the substrate to create a metal point-contact with a
conductance of at least 5G0.10 This ensures that a new electrode
structure is created for each measurement. The tip is then
withdrawn from the substrate at a speed of about 16 nm/s
while the current is recorded at a fixed applied bias voltage of
225 mV at a 40 kHz data acquisition rate. This yields a
conductance (current/voltage) versus displacement trace. In all
measurements reported here, 10,000 traces were collected with
each molecule to allow for detailed statistical analysis. Figure
1C shows three sample traces measured with molecules 1−3.
These show conductance plateaus at integer multiples of G0 =
2e2/h, the quantum of conductance, as well as a molecular
dependent conductance feature between 10−3G0 and 10−5G0.
These illustrative traces show that the conductance of 1 is the
largest, while that of 3 is the smallest.
To determine the conductance of each molecule in a
statistically significant manner, we create one-dimensional and
two-dimensional histograms of all 10000 traces measured. One-
dimensional histograms show the frequency of conductance
values measured averaged over all traces while two-dimensional
histograms reveal the junction conductance evolution as a
function of elongation.11 Figure 2A shows one-dimensional
conductance histograms of molecules 1−3 created using
logarithm bins. We find that the peak of these histograms
shifts systematically to lower conductance as the aromaticity of
the central unit is enhanced. Specifically, the cyclopentadiene
derivative has the highest conductance, while the thiophene
derivative has the lowest. We show, in Figure 2B, two-
dimensional conductance histograms generated from the same
traces used to create the histograms in Figure 2A. These
illustrate clearly that in all cases we are measuring conductance
Figure 3. (A) Structures of molecules 4, 5, and 6. (B) Log-binned one-
dimensional conductance histograms for molecules 4, 5, and 6
generated from 10000 conductance traces using a bin size of 100/
decade.
Figure 3B shows one-dimensional conductance histograms for
this series (see SI for 2D histograms). We again see a
decreasing conductance due to enhanced aromaticity; however,
for molecule 4 (the cyclopentadiene derivative), we do not see
a clear increase in conductance when compared with that for
the furan derivative (5). We attribute this to the internal torsion
angle of 4, determined from its X-ray structure, which is ∼16°
(see SI Figure S1). Without the acetylene bridging within the
molecular wire, the two methyls on the cyclopentadiene ring
interfere with the ortho hydrogens on the aniline rings,
resulting in larger internal torsion angles and a lower
conductance.8,12
We plot, in Figure 4, the conductance values for molecules
1−3 obtained by fitting a Gaussian function to the histogram
peaks shown in Figure 2A against the resonance energy of the
five-membered ring units. We see that conductance correlates
negatively with resonance energy, i.e. the more aromatic the
molecule is, the lower the conductance. The best structures for
conductivity would have π bonds external to the rings, quinoid
structures, while the best structures for the aromatic rings
themselves would not have such external π bonds. The
compromise structures adopted would have a larger energy
919
dx.doi.org/10.1021/ja411143s | J. Am. Chem. Soc. 2014, 136, 918−920

Journal of the American Chemical Society
Communication
(4) Quinn, J. R.; Foss, F. W.; Venkataraman, L.; Hybertsen, M. S.;
Breslow, R. J. Am. Chem. Soc. 2007, 129, 6714.
(5) Wheland, G. W. The Theory of Resonance and its Application to
Organic Chemistry; John Wiley and Sons: New York, 1944.
(6) Breslow, R.; Foss, F. J. Phys. Condens. Matter 2008, 20, 4104−
4111.
(7) Xu, B. Q.; Tao, N. J. Science 2003, 301, 1221.
(8) Venkataraman, L.; Klare, J. E.; Nuckolls, C.; Hybertsen, M. S.;
Steigerwald, M. L. Nature 2006, 442, 904.
(9) Venkataraman, L.; Klare, J. E.; Tam, I. W.; Nuckolls, C.;
Hybertsen, M. S.; Steigerwald, M. L. Nano Lett. 2006, 6, 458.
(10) (a) Trouwborst, M. L.; Huisman, E. H.; Bakker, F. L.; van der
Molen, S. J.; van Wees, B. J. Phys. Rev. Lett. 2008, 100, 175502.
(b) Sabater, C.; Untiedt, C.; Palacios, J. J.; Caturla, M. J. Phys. Rev. Lett.
2012, 108, 205502.
(11) (a) Kamenetska, M.; Koentopp, M.; Whalley, A.; Park, Y. S.;
Steigerwald, M.; Nuckolls, C.; Hybertsen, M.; Venkataraman, L. Phys.
Rev. Lett. 2009, 102, 126803. (b) Martin, C. A.; Ding, D.; Sorensen, J.
K.; Bjornholm, T.; van Ruitenbeek, J. M.; van der Zantt, H. S. J. J. Am.
Chem. Soc. 2008, 130, 13198.
(12) 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.
Figure 4. Conductance values for molecules 1, 2, and 3 as a function
of resonance energy of the aromatic component in the molecular wire.
penalty for systems with larger initial aromatic stabilization.
Another consideration is that the HOMOs for aromatic
compounds are lower in energy than those for nonaromatic
compounds, and conductivity in these electron-rich amine-
terminated systems depends on the location of the HOMOs
relative to the gold electrode Fermi energy.13
(13) Quek, S. Y.; Venkataraman, L.; Choi, H. J.; Louie, S. G.;
Hybertsen, M. S.; Neaton, J. B. Nano Lett. 2007, 7, 3477.
In conclusion, we have studied the relationship between
aromaticity and conductance of a single-molecule junction in
ambient conditions. We are able to attain a series of molecular
structures where the change in conductance would only come
from the different aromatic components that we are studying.
This work thus provides an experimental guide to building
active elements for organic photovoltaics using nonaromatic
components, as our results show that stronger aromaticity leads
to lower conductance at the single-molecule level.
ASSOCIATED CONTENT
* Supporting Information
Synthetic and measurement details and the crystal structure of
4. This material is available free of charge via the Internet at
http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Authors
rb33@columbia.edu
■
lv2117@columbia.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was funded primarily by the Center for Re-Defining
Photovoltaic Efficiency through Molecule Scale Control, an
EFRC funded by the US Department of Energy, Office of Basic
Energy Sciences (DE-sc0001085). L.V. thanks the Packard
Foundation for support. H.L. is supported by the Semi-
conductor Research Corporation and New York CAIST
program. We thank Professor Ged Parkin and his co-worker
Joshua Palmer for the crystal structure determination.
REFERENCES
■
(1) Scharber, M. C.; Muhlbacher, D.; Koppe, M.; Denk, P.; Waldauf,
C.; Heeger, A. J.; Brabec, C. J. Adv. Mater. 2006, 18, 789.
̈
(2) Roncali, J. Chem. Rev. 1992, 92, 711.
(3) Smith, M. B.; March, J. March’s Advanced Organic Chemistry:
Reactions, Mechanisms, and Structures, 5th ed.; John Wiley & Sons, Inc.:
New York, 2001.
920
dx.doi.org/10.1021/ja411143s | J. Am. Chem. Soc. 2014, 136, 918−920