Influence of the chemical structure on the stability and conductance of porphyrin single-molecule junctions

Article abstract of DOI:10.1002/anie.201104757

Different bridging geometries can explain the observation that porphyrin molecules with added thiol end groups and pyridine axial groups form more-stable single-molecule junctions with an increased spread in low-bias conductance. The stability of these ge

Full text of DOI:10.1002/anie.201104757

DOI: 10.1002/anie.201104757
Single-Molecule Electronics
Influence of the Chemical Structure on the Stability and Conductance
of Porphyrin Single-Molecule Junctions**
Mickael L. Perrin, Ferry Prins, Christian A. Martin, Ahson J. Shaikh, Rienk Eelkema,
Jan H. van Esch, Tomas Briza, Robert Kaplanek, Vladimir Kral, Jan M. van Ruitenbeek,
´
Herre S. J. van der Zant, and Diana Dulic*
The use of porphyrin molecules as building blocks of func-
tional molecular devices has been widely investigated.[1,2] The
structural flexibility and well-developed synthetic chemistry
of porphyrins allows their physical and chemical properties to
be tailored by choosing from a wide library of macrocycle
substituents and central metal atoms. Nature itself offers
magnificent examples of processes that utilize porphyrin
derivatives, such as the activation and the transport of
molecular oxygen in mammals and the harvesting of sunlight
in plant photosynthetic systems. In order to exploit the highly
desirable functionality of porphyrins in artificial molecular
devices, it is imperative to understand and control the
interactions that occur at the molecule–substrate interface.
Such interactions largely depend on the electronic and
conformational structures of the adsorbed molecules, which
can be studied using techniques such as scanning tunneling
microscopy,[3–7] UV photoemission spectroscopy,[8] and X-ray
photoemission spectroscopy,[2] and on a theoretical level with
density functional calculations.[9] Recent studies on conju-
gated rod-like molecules have shown that molecular con-
ductance measurements can be significantly affected by the
binding geometry,[10] coupling of the p orbitals to the leads,[11]
or p–p stacking between adjacent molecules.[12] Herein, we
present the results of a study of the interaction of laterally
extended p-conjugated porphyrin molecules with electrodes
by means of time- and stretching-dependent conductance
measurements on molecular junctions. We further investigate
strategies to reduce interactions of the molecular p electrons
with the metal electrodes by modifying the chemical structure
of the porphyrin molecules.
We used the series of molecules represented in Figure 1a–
c to examine the influence of the molecular structure on the
formation of porphyrin single-molecule junctions. Since the
thiol group is most commonly used to contact rod-like
molecules to form straight molecular bridges,[13] we first
compared 5,10,15,20-tetraphenylporphyrin without thiol ter-
mination (H2-TPP; Figure 1a) to a nearly identical molecule
with two thiol groups on opposite sides of the molecule (5,15-
di(p-thiophenyl)-10,20-di(p-tolyl)porphyrin
(H2-TPPdT);
Figure 1b). To investigate the influence of the molecular
backbone geometry on the junction formation we further
studied a thiol-terminated porphyrin molecule with two bulky
pyridine axial groups attached through an octahedral RuII ion
([RuII{5,15-di(p-thiophenyl)-10,20-diphenylporphyrin}(py)2]
(Ru-TPPdT); Figure 1c). As a consequence of steric hin-
drance, the pyridine groups in Ru-TPPdT reduce the direct
interaction of the metal electrodes with the p face of the
porphyrin. A similar strategy was used previously.[4]
Prior to electrical characterization, the molecules were
deposited using self-assembly from solution. To study the
conductance of these molecules we used lithographic mechan-
ically controllable break junctions (MCBJs) in vacuum at
room temperature. The layout of an MCBJ device in a three-
point bending mechanism is shown in Figure 1d. Details
concerning the synthesis of the molecules and the exper-
imental procedures are given in the Supporting Information.
Sets of 1000 consecutive breaking traces from individual
junctions were analyzed numerically to construct “trace
histograms” of the conductance (log10 G versus the electrode
displacement d).[14,15] This statistical method maps the break-
ing dynamics of the junctions beyond the point of rupture of
the last monatomic gold contact (defined as d = 0), which has
a conductance of one quantum unit G0 = 2e2h. Areas of high
counts represent the most typical breaking behavior of the
molecular junctions. Figure 2 presents trace histograms as
well as examples of individual breaking traces for acetone as
reference, H2-TPP, H2-TPPdT, and Ru-TPPdT. For all three
porphyrin molecules as well as for the reference sample
several junctions were measured (see the Supporting Infor-
mation). Herein, we only show a typical set of junctions.
In the junction which was exposed to pure acetone
(Figure 2a), the Au bridge initially gets stretched until a
plateau around the conductance quantum (G ꢀ G0) is
observed (only visible in the individual traces shown in
[*] M. L. Perrin, F. Prins, Dr. C. A. Martin, Prof. Dr. H. S. J. van der Zant,
´
Dr. D. Dulic
Kavli Institute of Nanoscience, Delft University of Technology
Lorentzweg 1, 2628 CJ Delft (The Netherlands)
E-mail: d.dulic@tudelft.nl
A. J. Shaikh, Dr. R. Eelkema, Prof. Dr. J. H. van Esch
Department of Chemical Engineering
Delft University of Technology (The Netherlands)
Dr. T. Briza, Dr. R. Kaplanek, Prof. Dr. V. Kral
Institute of Chemical Technology, Prague (Czech Republic)
Prof. Dr. J. M. van Ruitenbeek
Kamerlingh Onnes Laboratory, Leiden University (The Netherlands)
[**] This research was carried out with financial support from the Dutch
Foundation for Fundamental Research on Matter (FOM) and the
VICI (680-47-305) grant from The Netherlands Organisation for
Scientific Research (NWO). We would like to thank Prof. Dr. Robert
Metzger and Dr. Graeme R. Blake for careful reading of the
manuscript.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201104757.
Angew. Chem. Int. Ed. 2011, 50, 11223 –11226
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
11223

Communications
black), which corresponds to a monatomic
contact.[14] Upon further stretching, the
Au–Au contact is broken and the conduc-
tance decreases abruptly to about 10À3 G0.
Beyond this point, electron tunneling
between the electrodes leads to a fast
conductance decay with stretching (visible
as the orange tail), as expected for tunnel-
ing across a vacuum barrier.
In contrast, introducing the porphyrin
molecules on the junctions leads to pro-
nounced plateaus at different conductance
values in the sub-G0 regime. These plateaus
can be flat or sloped.[10,16] The representa-
tive breaking traces that are included in
Figure 2b–d display a set of such plateaus.
Averaging over 1000 traces does not lead to
a narrow region of high counts in the
histograms, in contrast to measurements on
rod-like molecules.[14–17] In the trace histo-
gram of H2-TPP (Figure 2b), however,
there are two distinct regions with high
counts; a high-conductance region (HCR)
around 10À1 G0, and a sloped low-conduc-
tance region (LCR) below 10À3 G0. For H2-
TPPdT (Figure 2c), both the HCR and
LCR are longer than for H2-TPP; thus,
adding the thiol end groups to H2-TPP
increases the plateau length, and it reduces
the slope of the LCR. In Ru-TPPdT (Fig-
ure 2d), the HCR and LCR cannot be
distinguished anymore: only one long
region sloping from 10À1 G0 to 10À5 G0 is
present, which has a shallower slope com-
pared to the histograms of H2-TPP and H2-
TPPdT, and an increased length.
Figure 1. Structures of a) H2-TPP, b) H2-TPPdT, c) Ru-TPPdT. d) Top: layout of an MCBJ.
Bottom: scanning electron micrograph of an MCBJ device (colorized for clarity). From the
scale bar it is clear that the suspended part of the electrode bridge is about 1.5 mm long.
Additional information about molecu-
lar junction configurations can be gained
by measuring the evolution of the junction
conductance over large time intervals at
fixed electrode distances.[18] To obtain such
“time traces”, we opened the junction in
small steps using a servomotor at 77 K. At
this low temperature, the surface diffusion
is reduced, which enhances the stability
without causing extensive changes in the
molecular conductance.[19] We then mea-
sured the conductance in the range from
1 G0 to 10À4 G0 at various fixed electrode
spacings and for time periods exceeding
several hours. As a consequence of the
exceptional control over the electrode
separation in the MCBJ,[18,19] conductance
jumps in these measurements can be
attributed to reconfigurations of the mole-
cule in the junction rather than mechanical
interference of the setup.
Figure 2. Trace histograms constructed from 1000 breaking traces for junctions exposed to
pure acetone, H2-TPP, H2-TPPdT, and Ru-TPPdT. The black curves are examples of individual
breaking traces (offset for clarity). For the construction of the histograms, d=0 for each
curve was set to the point where the conductance drops sharply below 1 G0. All histograms
were taken at a bending speed of 30 mmsÀ1 (on the scale of the electrodes, this bending is
translated into a stretching on the order of 1.8 nmsÀ1) and a bias voltage of 150 mV. No
data selection schemes were utilized. For additional trace histograms see the Supporting
Typical time traces of junctions exposed
to pure acetone, H2-TPP, H2-TPPdT, and
Information. For (a) we found a decay constant of 2 ꢀÀ1
.
11224 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 11223 –11226

Ru-TPPdT are presented in Figure 3. As can be seen in
Figure 3a, a stable contact is formed at 1 G0 in the clean
junction. A slight increase in the electrode spacing leads to
ure 3d shows that a particular junction conductance can be
reached from different starting values (blue and orange traces
in the middle of the graph).
The presence of a large range of molecular adsorption
geometries contrasts with most studies on long rod-like
molecules, which typically assume a straight bridging config-
uration of the molecule with both thiol groups connected to
the electrodes (Figure 4a). For H2-TPPdT, such a configu-
ration could be expected. However, the high affinity of the
porphyrin p cloud for metal surfaces likely stabilizes other
Figure 3. Conductance-versus-time plots for acetone, H2-TPP,
H2-TPPdT, and Ru-TPPdT taken at a bias voltage of 150 mV. The traces
originate from the same opening event and are taken at intervals of
50 pm. Several lines are displayed in orange for clarity.
Figure 4. Illustration of four possible configurations of a porphyrin
molecule in the junction. a) Bridging configuration, b) STM-like config-
uration, and c) intermediate configuration of H2-TPPdT; d) intermedi-
ate configuration of Ru-TPPdT (pyridine groups shown in gray).
abrupt breaking of the contact to a conductance value around
10À4 G0. In contrast, the presence of porphyrin molecules
leads to the formation of junctions with conductances in the
region between 1 G0 and 10À3 G0 (Figure 3b–d). Below
10À3 G0, the H2-TPP molecule (Figure 3b) exhibits a sudden
conductance drop to below 10À5 G0. In contrast, thiol-
terminated H2-TPPdT and Ru-TPPdT (Figure 3c,d) form
molecular junctions over the whole range between 1 G0 and
10À4 G0. The observations support the conclusions drawn
from the trace histograms: adding thiol and pyridine groups
increases the junction stability and leads to a variety of
molecular geometries with different conductance values. To
examine the contribution of p stacking of multiple molecules,
we measured trace histograms on junctions exposed to
5,10,15,20-tetrakis(3,5-di-tert-butylphenyl)porphyrin abbrevi-
ated as TBP. As a consequence of the steric hindrance of the
butyl groups, the possibility of p stacking between two
molecules is reduced. Resulting trace histograms are shown
in the Supporting Information. There is no qualitative differ-
ence between the trace histogram of TBP and TPP molecules,
which demonstrates insignificant influence of p stacking on
the formation of different configurations.
A closer inspection of the time traces reveals interesting
information on the behavior of the molecule in the junction.
Since the electronic noise is much smaller than the width of
the bands of the measured conductance values, we conclude
that what appears to be noise in the time traces is in fact a sign
of small variations in the molecular configuration.[18] Large
jumps in the conductance (Figure 3c,d) indicate that a
molecular junction can spontaneously change its configura-
tion. In Figure 3c, random telegraph noise between two
conductance values is observed indicating the possibility of
forming several metastable configurations. Furthermore, Fig-
configurations, such as sketched in Figure 4b and c. A stable
junction configuration can be formed without thiol groups[20]
and recent break junction experiments have demonstrated
that benzene moieties can bind directly to gold electro-
des.[21,22] Furthermore, it has been reported that the lateral
coupling of p orbitals to electrodes influences the single-
molecule conductance of rod-like molecular wires.[11,23] Such
coupling becomes more likely when laterally extended
molecules are probed. In addition, tetraphenylporphyrin
molecules have internal degrees of freedom, which can
influence the charge transport on a single-molecule level.[24]
STM studies indicate that tetraphenylporphyrins can bind to
Au(111) through an interaction with their phenyl side
groups.[6] On gold surfaces, their conformation can change
through rotations of side groups and buckling of the center.[5]
Such variations in the adsorption geometry, which can be
expected to lead to different conductance values, are likely
the origin of the variety of junction configurations that we
observe.
In summary, we have demonstrated that porphyrin
molecules can form stable bridging molecular junctions
even without thiol anchoring groups. Adding thiol end
groups and pyridine axial groups to the porphyrin backbone,
respectively, increases the stability of the junctions and leads
to an increased spread in conductance. This is a result of the
formation of different junction configurations. To enable the
reliable formation of stable porphyrin junctions, molecules
with well-defined adsorption geometries will be required.
Quantum chemical calculations could yield more insight into
their design as well as the configurations and the conductance
of porphyrin single-molecule junctions. Finally, we expect that
Angew. Chem. Int. Ed. 2011, 50, 11223 –11226
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11225

Communications
[11] I. Diez-Perez, J. Hihath, T. Hines, Z.-S. Wang, G. Zhou, K.
Mꢀllen, N. Tao, Nat. Nanotechnol. 2011, 6, 226.
[12] S. Wu, M. T. Gonzꢁlez, R. Huber, S. Grunder, M. Mayor, C.
Schçnenberger, M. Calame, Nat. Nanotechnol. 2008, 3, 569 – 574.
[13] N. J. Tao, Nat. Nanotechnol. 2006, 1, 173 – 181.
[14] C. A. Martin, D. Ding, J. K. Sørensen, T. Bjørnholm, J. M.
van Ruitenbeek, H. S. J. van der Zant, J. Am. Chem. Soc. 2008,
130, 13198 – 13199.
[15] M. Kamenetska, M. Koentopp, A. C. Whalley, Y. S. Park, M. L.
Steigerwald, C. Nuckolls, M. S. Hybertsen, L. Venkataraman,
Phys. Rev. Lett. 2009, 102, 126803.
multiple junction configurations with considerable variation
in the measured conductance values can also be observed for
other non-rod-like molecules.
Received: July 8, 2011
Published online: September 28, 2011
Keywords: break junctions · molecular electronics ·
.
porphyrinoids · single-molecule transport
[16] A. Mishchenko, D. Vonlanthen, V. Meded, M. Bꢀrkle, C. Li, I. V.
Pobelov, A. Bagrets, J. K. Viljas, F. Pauly, F. Evers, M. Mayor, T.
Wandlowski, Nano Lett. 2010, 10, 156 – 163.
[1] M. Jurow, A. E. Schuckman, J. D. Batteas, C. M. Drain, Coord.
Chem. Rev. 2010, 254, 2297 – 2310.
[17] M. T. Gonzꢁlez, S. Wu, R. Huber, S. J. van der Molen, C.
Schçnenberger, M. Calame, Nano Lett. 2006, 6, 2238 – 2242.
[2] S. Mohnani, D. Bonifazi, Coord. Chem. Rev. 2010, 254, 2342 –
2360.
[3] V. Iancu, A. Deshpande, S.-W. Hla, Nano Lett. 2006, 6, 820 – 823.
[4] G. Sedghi, K. Sawada, L. J. Esdaile, M. Hoffmann, H. L.
Anderson, D. Bethell, W. Haiss, S. J. Higgins, R. J. Nichols, J.
Am. Chem. Soc. 2008, 130, 8582 – 8583.
[5] G. V. Nazin, X. H. Qiu, W. Ho, Science 2003, 302, 77 – 81.
[6] X. H. Qiu, G. V. Nazin, W. Ho, Phys. Rev. Lett. 2004, 93, 196806.
[7] J. Brede, M. Linares, S. Kuck, J. Schwçbel, A. Scarfato, S.-H.
Chang, G. Hoffmann, R. Wiesendanger, R. Lensen, P. H. J.
Kouwer, J. Hoogboom, A. E. Rowan, M. Brçring, M. Funk, S.
Stafstrçm, F. Zerbetto, R. Lazzaroni, Nanotechnology 2009, 20,
275602.
´
[18] D. Dulic, F. Pump, S. Campidelli, P. Lavie, G. Cuniberti, A.
Filoramo, Angew. Chem. 2009, 121, 8423 – 8426; Angew. Chem.
Int. Ed. 2009, 48, 8273 – 8276.
[19] C. A. Martin, D. Ding, H. S. J. van der Zant, J. M. van Ruiten-
beek, New J. Phys. 2008, 10, 065008.
[20] E. A. Osorio, M. Ruben, J. S. Seldenthuis, J.-M. Lehn, H. S. J.
van der Zant, Small 2010, 6, 174 – 178.
[21] S. T. Schneebeli, M. Kamenetska, Z. Cheng, R. Skouta, R. A.
Friesner, L. Venkataraman, R. Breslow, J. Am. Chem. Soc. 2011,
133, 2136 – 2139.
[22] S. Martın, I. Grace, M. R. Bryce, C. Wang, R. Jitchati, A. S.
Batsanov, S. J. Higgins, C. J. Lambert, R. J. Nichols, J. Am. Chem.
Soc. 2010, 132, 9157 – 9164.
[8] L. Scudiero, D. E. Barlow, U. Mazur, K. W. Hipps, J. Am. Chem.
Soc. 2001, 123, 4073 – 4080.
[23] W. Haiss, C. Wang, I. Grace, A. S. Batsanov, D. J. Schiffrin, S. J.
Higgins, M. R. Bryce, C. J. Lambert, R. J. Nichols, Nat. Mater.
2006, 5, 995 – 1002.
[24] M. Ruben, A. Landa, E. Lçrtscher, H. Riel, M. Mayor, H. Gçrls,
H. B. Weber, A. Arnold, F. Evers, Small 2008, 4, 2229 – 2235.
[9] Z.-L. Cai, M. J. Crossley, J. R. Reimers, R. Kobayashi, R. D.
Amos, J. Phys. Chem. B 2006, 110, 15624 – 15632.
[10] S. Y. Quek, M. Kamenetska, M. L. Steigerwald, H. J. Choi, S. G.
Louie, M. S. Hybertsen, J. B. Neaton, L. Venkataraman, Nat.
Nanotechnol. 2009, 4, 230.
11226 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 11223 –11226