Comparison of molecular conductance between planar and twisted 4-phenylpyridines by means of two-dimensional phase separation of tetraphenylporphyrin templates at a liquid-HOPG interface

Article abstract of DOI:10.1039/c1cc12041g

Tetraphenylporphyrin (TPP) rhodium chlorides coordinated by planar and twisted 4-phenylpyridine derivatives were synthesized. An STM image was taken by a 2-D phase separation technique and the conductance was evaluated. Difference in apparent height betwe

Full text of DOI:10.1039/c1cc12041g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 8427–8429
www.rsc.org/chemcomm
COMMUNICATION
Comparison of molecular conductance between planar and twisted
4-phenylpyridines by means of two-dimensional phase separation of
tetraphenylporphyrin templates at a liquid–HOPG interfacew
Takeshi Sakano, Kenji Higashiguchi and Kenji Matsuda*
Received 11th April 2011, Accepted 8th June 2011
DOI: 10.1039/c1cc12041g
Tetraphenylporphyrin (TPP) rhodium chlorides coordinated by
planar and twisted 4-phenylpyridine derivatives were synthesized.
An STM image was taken by a 2-D phase separation technique
and the conductance was evaluated. Difference in apparent
height between these phenylpyridines reflects the conductance
ratio of ligands.
The investigation of single molecular conductance is a key
issue in the molecular electronics field and is developing
rapidly using several measurement techniques.1 Although
most experiments on the conductance of a single molecule
Fig. 1 Molecular structures of synthesized 4-phenylpyridines, TPPs,
exhibit considerable variations, the reliability of the measurements
has recently been significantly increased owing to the statistical
and TPP rhodium chlorides.
treatment by the methods of mechanically controllable break
junction (MCBJ)2 and scanning tunneling microscopy break
junction (STM–BJ).3,4
Pyridine-coordinated tetraphenylporphyrin (TPP) Rh(III)
chloride having long alkyl chains forms 2-D lamellar structures
at a liquid–HOPG interface, where a strongly bound axial
ligand of pyridine is placed perpendicular on the TPP.7 Therefore,
TPP rhodium chloride can be used as a molecular template for
the compounds carrying pyridyl groups. 4-Phenylpyridine
derivatives 1 and 2 having the same molecular lengths and
the different torsion angles were selected to investigate the
differences in the apparent height in STM images because they
are expected to have different molecular conductances due to
the different torsion angles (Fig. 1).4
Among several methods for the measurement of the molecular
conductance, apparent height measurement by STM has the
merit of applicability to many samples.5,6 The method often
utilizes the plating of target molecules into the self-assembled
monolayers (SAMs) of alkanethiols on an Au substrate in
order to prevent the intermolecular interaction among target
molecules. However, this usual method precludes the statistical
treatment because only few spots of target molecules are
observable from one STM image. Use of a molecular template
is one candidate to avoid the intermolecular interaction and
STM observation of the SAM of samples on the templates
makes the statistical analysis possible. Herein, we report
on the comparison of molecular conductance between planar
and twisted 4-phenylpyridine derivatives by using porphyrin
templates, and show that a two-dimensional (2-D) phase
separation technique is effective to discriminate these phenyl-
pyridines on the substrate.
Syntheses of 4-phenylpyridine-coordinated TPP rhodium
chlorides (C0-Rh-1, C0-Rh-2, C22-Rh-1, C22-Rh-2, and C30-Rh-1)
were carried out according to the general organic synthesis
procedure described in ESI.w Subscript numbers 0, 22, and 30
show the length of alkyl side chains. Free base TPPs C16-2H,
C22-2H, and C30-2H were also synthesized. X-Ray crystallo-
graphic analysis of C0-Rh-1 and C0-Rh-2 revealed that the
dihedral angles between the phenyl ring and the pyridyl ring of
1 and 2 are 271 and 681, respectively. The ORTEP drawings
are shown in ESI.w
Department of Synthetic Chemistry and Biological Chemistry,
Graduate School of Engineering, Kyoto University, Katsura,
Nishikyo-ku, Kyoto 615-8510, Japan.
STM images for a solution of C22-Rh-1 and for a 1 : 1 mixed
solution of C22-Rh-1 and C22-Rh-2 were obtained at the
1-octanoic acid–HOPG interface in the constant current
mode.8 As shown in Fig. 2a and b, both samples formed
SAMs of characteristic lamellar structures, where TPPs are
aligned side by side in the bright stripes. The stripes are
separated by the alkyl chains from the neighboring TPP
arrays.9 In the dark areas the alkyl chains are interdigitated,
E-mail: kmatsuda@sbchem.kyoto-u.ac.jp
w Electronic supplementary information (ESI) available: Synthetic
procedures of a 4-phenylpyridine derivative, TPP rhodium chlorides,
1H NMR spectra of all compounds, additional STM images, details of
lattice parameters, and crystallographic data of C0-Rh-1 and C0-Rh-2.
CCDC 814313–814314. For ESI and crystallographic data in CIF or
other electronic format see DOI: 10.1039/c1cc12041g
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 8427–8429 8427

View Article Online
Fig. 3 STM image at the 1-phenyloctane–HOPG interface in the
constant current mode: (a) a 3 : 1 mixed solution of C16-2H (1.2 ꢀ 10ꢂ6 M)
and C22-2H (4.0 ꢀ 10ꢂ7 M) (50 ꢀ 50 nm2, Iset = 30 pA, Vbias = ꢂ1.0 V).
(b) Section analyses of
C22-2H (red line) and C16-2H (green line)
domains. White parallelograms in the STM images show the unit cells.
Fig. 2 STM images at the 1-octanoic acid–HOPG interface in the
constant current mode: (a) C22-Rh-1 (3.0 ꢀ 10ꢂ6 M) (50 ꢀ 50 nm2,
Iset = 30 pA, Vbias = ꢂ1.0 V); (b) a 1 : 1 mixed solution of C22-Rh-1
(1.5 ꢀ 10ꢂ6 M) and C22-Rh-2 (1.5 ꢀ 10ꢂ6 M) (60 ꢀ 60 nm2, Iset = 30 pA,
Vbias = ꢂ1.0 V). (c) Section analysis for the image (a). Histograms of
apparent height of TPP cores: (d) for the image (a); (e) for the image (b).
although detailed structure of these chains was not observed in
these images. The lattice parameters of the unit cell a ꢀ b and
a were 3.9 ꢁ 0.2 nm ꢀ 1.8 ꢁ 0.2 nm and 711 for C22-Rh-1
and 4.2 ꢁ 0.2 nm ꢀ 2.3 ꢁ 0.2 nm and 821 for a mixture of
C22-Rh-1 and C22-Rh-2. These values are similar to that of
characteristic lamellar structures of C22-2H at the phenyloctane–
HOPG interface (Fig. S3, ESIw, 4.1 ꢁ 0.2 nm ꢀ 1.9 ꢁ 0.2 nm
and 841; also reported in ref. 10 as 4.2 nm ꢀ 2.1 nm and 841).
Contrast of the TPP core shows the apparent height of the
core above the alkyl side chains. Section analysis of Fig. 2a is
shown in Fig. 2c. Statistical analyses of apparent height for
these STM images based on the section analysis are shown in
Fig. 2d and e. Both statistical distributions of apparent heights
were fitted by a single Gaussian function. By mixing C22-Rh-2
into C22-Rh-1, an apparent height distribution got lowered
and broadened from 3.9 ꢁ 0.5 A to 3.6 ꢁ 0.9 A. In terms of
apparent height, absolute values of the height are generally
varied with measurement conditions even in the same setup.
Therefore, the decrease of apparent height by mixing C22-Rh-2
does not give the relationship of apparent height between
C22-Rh-1 and C22-Rh-2. However, the result of broadening
of the distribution suggests that C22-Rh-2 shows the different
apparent height from C22-Rh-1. Since topographic heights are
the same for C22-Rh-1 and C22-Rh-2 at the 1-octanoic
acid–HOPG interface, this difference in apparent height
shows the difference in a tunneling decay constant of axial
ligands of 1 and 2.
Fig. 4 (a) An STM image of a 10 : 1 mixed solution of C22-Rh-2
(1.5 ꢀ 10ꢂ6 M) and C30-Rh-1 (1.5 ꢀ 10ꢂ7 M) (75 ꢀ 75 nm2, Iset = 30 pA,
Vbias = ꢂ1.0 V) at the 1-octanoic acid–HOPG interface in the
constant current mode. Histograms of apparent height in the domain
of (b) C30-Rh-1, (c) C22-Rh-2, and (d) both C30-Rh-1 and C22-Rh-2.
The contribution in counts from each domain was normalized to
1 : 1 in histogram (d).
were 3.4 ꢁ 0.2 nm ꢀ 1.8 ꢁ 0.2 nm and 891, whereas those of
the left one were 4.0 ꢁ 0.2 nm ꢀ 1.6 ꢁ 0.2 nm and 881. These
two parameters are similar to those of C16-2H and C22-2H.
This result shows that the 2-D phase separation of C16-2H and
C22-2H occurred at the solid–liquid interface. Section analysis
of each domain shows that there is no significant difference in
the apparent height, suggesting that the length of alkyl side
chain does not influence the apparent height (Fig. 3b).
By means of the 2-D phase separation method, the apparent
height of 1 and 2 was measured. Fig. 4a shows an STM image
of a mixed solution of C30-Rh-1 and C22-Rh-2. Two domains
having different lattice parameters of the unit cell were observed in
one STM image. A domain on the bottom-left corner had a
lattice spacing corresponding to that of C30-2H (Fig. S3, ESIw)
and another one on the upper-right corner had a similar
spacing of C22-2H lattice. These domains correspond to the
domains of C30-Rh-1 and C22-Rh-2. Histograms of apparent
height were separately created for each domain as shown in
Fig. 4b and c. Apparent heights were obtained as 4.3 ꢁ 0.7 A
To clear up this difference, apparent height distributions for
these two ligands need to be distinctly separated. We focused
our attention on the coadsorption phenomena at a liquid–
solid interface11 and the coadsorption behavior of two free
base TPPs, C16-2H and C22-2H, was examined. Fig. 3a shows
STM images for a mixed solution of C16-2H and C22-2H.
Two domains were observed and each domain has different
lattice parameters. Lattice parameters of the right domain
c
8428 Chem. Commun., 2011, 47, 8427–8429
This journal is The Royal Society of Chemistry 2011

View Article Online
4,40-diaminobiphenyl is proportional to the cos2f, where f is
a dihedral angle.4 Dihedral angles of 1 (271) and 2 (681)
obtained from X-ray crystallographic analysis give conductance
ratio cos2f1/cos2f2 = 5.7, which is in excellent agreement with
experimentally obtained Gmol1/Gmol2 = 5.4. This agreement
warrants that our method is applicable to the comparison of
molecular conductance.
In conclusion, we have succeeded in developing the 2-D
phase separation technique of TPP templates having different
lengths of side chain at a solution–HOPG interface. This
technique was applied to the determination of the ratio of
molecular conductance between planar and twisted phenyl-
pyridines by comparing apparent height in the STM image.
This technique will be the useful method for the determination
of molecular conductance.
Fig. 5 Schematic drawing of the two-layer tunnel junction model for
the STM measurement of a mixed solution of C30-Rh-1 and C22-Rh-2.
for C30-Rh-1 and 2.9 ꢁ 0.6 A for C22-Rh-2. Since geometrical
molecular heights of C30-Rh-1 and C22-Rh-2 are the same,
the difference in apparent height should originate from the
conductance ratio of two ligands 1 and 2. Summation of these
histograms gave a broad distribution (Fig. 4d), which is in
good agreement with the distribution of an apparent height of
1 : 1 mixture of C22-Rh-1 and C22-Rh-2 that has already been
shown in Fig. 2e.
This work was supported by a Grant-in-Aid for Young
Scientist (A) (No. 19685013) and a Grant-in-Aid for Science
Research in Priority Areas ‘‘Photochromism (471)’’ (No.
19050009) from MEXT, Japan and NEXT program (No.
GR062) from JSPS, Japan. T. S. acknowledges JSPS for the
young scientist fellowship.
According to the two-layer tunnel junction model proposed
by Weiss et al.,5 the total conductance (Gtotal) between an STM
tip and a substrate is described by product of the gap
conductance (Ggap = A exp(ꢂad)) and molecular conductance
(Gmol = B exp(ꢂbx)), where A and B are contact conductances,
a and b are decay constants of the gap and the molecule, d is
the gap distance and x is the molecular length. Gtotal is
constant everywhere, therefore, the conductance ratio
(Gmol1/Gmol2) is given by the following eqn (1):
Notes and references
1 C. Joachim, J. K. Gimzewski and A. Aviram, Nature, 2000, 408,
541–548; A. Nitzan and M. A. Ratner, Science, 2003, 300,
1384–1389; A. Salomon, D. Cahen, S. Lindsay, J. Tomfohr,
V. B. Engelkes and C. D. Frisbie, Adv. Mater., 2003, 15,
1881–1890; C. Joachim and M. A. Ratner, Proc. Natl. Acad. Sci.
U. S. A., 2005, 102, 8801–8802; N. J. Tao, Nat. Nanotechnol., 2006,
1, 173–181; P. S. Weiss, Acc. Chem. Res., 2008, 41, 1772–1781;
K. Moth-Poulsen and T. Bjørnholm, Nat. Nanotechnol., 2009, 4,
551–556.
Gmol1 Ggap2 A2
¼
¼
expfaðd1 ꢂ d2Þg
ð1Þ
Gmol2 Ggap1 A1
2 M. A. Reed, C. Zhou, C. J. Muller, T. P. Burgin and J. M. Tour,
Science, 1997, 278, 252–254; M. T. Gonzalez, S. M. Wu, R. Huber,
S. J. van der Molen, C. Schonenberger and M. Calame, Nano Lett.,
2006, 6, 2238–2242; M. Tsutsui, M. Taniguchi, K. Yokota and
T. Kawai, Nat. Nanotechnol., 2010, 5, 286–290.
3 B. Xu and N. J. Tao, Science, 2003, 301, 1221–1223.
4 L. Venkataraman, J. E. Klare, C. Nuckolls, M. S. Hybertsen and
M. L. Steigerwald, Nature, 2006, 442, 904–907; D. Vonlanthen,
A. Mishchenko, M. Elbing, M. Neuburger, T. Wandlowski and
M. Mayor, Angew. Chem., Int. Ed., 2009, 48, 8886–8890;
A. Mishchenko, D. Vonlanthen, V. Meded, M. Burkle, C. Li,
I. V. Pobelov, A. Bagrets, J. K. Viljas, F. Pauly, F. Evers,
M. Mayor and T. Wandlowski, Nano Lett., 2010, 10, 156–163.
5 L. A. Bumm, J. J. Arnold, T. D. Dunbar, D. L. Allara and
P. S. Weiss, J. Phys. Chem. B, 1999, 103, 8122–8127.
6 K. Moth-Poulsen, L. Patrone, N. Stuhr-Hansen, J. B. Christensen,
J.-P. Bourgoin and T. Bjørnholm, Nano Lett., 2005, 5, 783–785;
S. Wakamatsu, S. Fujii, U. Akiba and M. Fujihira, Jpn. J. Appl.
Phys., 2006, 45, 2736–2742.
7 J. Otsuki, Coord. Chem. Rev., 2010, 254, 2311–2341; T. Ikeda,
M. Asakawa, M. Goto, K. Miyake, T. Ishida and T. Shimizu,
Langmuir, 2004, 20, 5454–5459.
8 UV-vis spectrum of the ligand-coordinated porphyrin in octanoic
acid showed 3 nm of red-shift compared with the ligand-free
porphyrin, suggesting that the pyridyl ligand is bound to the
porphyrin template. See M. Hoshino, H. Seki, K. Yasufuku and
H. Shizuka, J. Phys. Chem., 1986, 90, 5149–5153.
9 X. Qiu, C. Wang, Q. Zeng, B. Xu, S. Yin, H. Wang, S. Xu and
C. Bai, J. Am. Chem. Soc., 2000, 122, 5550–5556; H. Wang,
C. Wang, Q. Zeng, S. Xu, S. Yin, B. Xu and C. Bai, Surf. Interface
Anal., 2001, 32, 266–270.
10 J. Otsuki, S. Kawaguchi, T. Yamakawa, M. Asakawa and
K. Miyake, Langmuir, 2006, 22, 5708–5715.
11 F. Tao, J. Goswami and S. L. Bernasek, J. Phys. Chem. B, 2006,
110, 19562–19569.
This equation means that the ratio of A2/A1, decay constant of
the gap (a), and difference in gap distance (d1–d2) give the
conductance ratio. The measurement condition is identical
because STM measurement was carried out for structurally
similar phenylpyridines C30-Rh-1 and C22-Rh-2 and both
molecules were observed in the same STM image. Therefore,
contact-dependent terms A1 and A2 are assumed to be equal.
Additionally, because x1 and x2 are the same, the term (d1–d2)
is equal to the difference in apparent height DhSTM (Fig. 5).
Then, eqn (1) is transformed to the following simple form:
Gmol1
¼ expfaDhSTM
g
ð2Þ
Gmol2
This eqn (2) means that decay constant a of the gap and
experimentally obtained DhSTM give the conductance ratio
between C30-Rh-1 and C22-Rh-2. Since the STM measurement
was conducted at the 1-octanoic acid–HOPG interface, a value
of vacuum cannot be applied. We adopted the decay constant
of a methylene unit (b = 1.2 Aꢂ1) reported by the measurement
of a series of alkanethiols as a substitute for 1-octanoic acid.5
By introducing the measurement result DhSTM = 1.4 A,
the conductance ratio between C30-Rh-1 and C22-Rh-2 is
finally obtained to be Gmol1/Gmol2 = 5.4. Since these phenyl-
pyridines 1 and 2 are supported by cognate templates, this
conductance ratio originated from the twisting effect of the
ligands. This result was compared to the cos2f law proposed
by Venkataraman et al., in which molecular conductance of
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 8427–8429 8429