Electronic decay constant of carotenoid polyenes from single-molecule measurements

Article abstract of DOI:10.1021/ja043279i

The conductance of carotenoid polyenes chemically bound at each end to gold contacts has been measured for single molecules containing 5, 7, 9, and 11 carbon-carbon double bonds in conjugation. The electronic decay constant, β, is determined to be 0.22 ±

Full text of DOI:10.1021/ja043279i

Published on Web 01/11/2005
Electronic Decay Constant of Carotenoid Polyenes from Single-Molecule
Measurements
Jin He,† Fan Chen,† Jun Li,† Otto F. Sankey,† Yuichi Terazono,‡ Christian Herrero,‡ Devens Gust,‡
Thomas A. Moore,‡ Ana L. Moore,‡ and Stuart M. Lindsay*,†,‡,§
Department of Physics and Astronomy, Department of Chemistry and Biochemistry, and Biodesign Institute,
Arizona State UniVersity, Tempe, Arizona 85287
Received November 8, 2004; E-mail: Stuart.Lindsay@asu.edu
Carotenoid polyenes have been used extensively as electron
donors in molecular systems that demonstrate photoinduced electron
transfer1,2 and have recently been found to act as electron shuttles
in photosynthesis.3,4 Direct measurements of their conductance5
show that it is ca. 7 orders of magnitude higher than that of
equivalent-length n-alkanes, as might be expected on theoretical
grounds.6 This high conductivity should correspond to a small value
of â, the electronic decay constant. A lower limit for â has been
inferred from spectroscopic data obtained from intervalence charge-
transfer complexes linked by polyenes,7 but â has not been
measured directly. We have measured the conductance of a series
of carotenoid polyenes as a function of their length, obtaining â )
0.22 ( 0.04 Å-1, in close agreement with the value obtained from
first principles simulations (0.22 ( 0.01 Å-1).
The original measurements of the single-molecule conductance
of a carotenoid polyene were made by inserting the molecule into
an alkanethiol self-assembled monolayer and contacting the inserted
molecules either directly with a conducting atomic force microscope
(AFM) probe8 or by using a gold nanoparticle as an intermediate
contact.5,9 Xu and Tao10 recently introduced a new method for
single-molecule measurements based on repeated formation of
break-junctions. Their method is much easier to use with various
lengths of molecules and yields better data than the conducting
AFM approach.11
The homologous series of carotenoids (illustrated below) was
synthesized as described in the Supporting Information.
Figure 1. (a) Examples of current vs stretching distance data for I
(N ) 5) at a bias of 0.2 V. (b) Histogram of recorded currents shows peaks
at ca. 0.4, 0.8, and 1.2 nA. The slope of a plot of peak value vs peak number
(inset) yields the current per molecule.
in toluene was placed onto a hydrogen-flame-annealed Au(111)
substrate12 in the liquid cell of a scanning tunneling microscope
(PicoSPM, Molecular Imaging, Tempe, AZ). After an hour, the
substrate was rinsed thoroughly and submerged in toluene for
measurement under Ar. Probes were made from freshly cut 0.25
mm Au wire (99.999%). The current was recorded on a digital
oscilloscope as the probe was repeatedly pushed into and retracted
from the substrate, with the tip-sample bias held fixed.10
Examples of current vs stretching distance (calculated from the
known probe velocity) recordings in the presence of I are shown
in Figure 1a. Taken at a fixed tip bias of 0.2 V, the traces show
distinct steps at multiples of ca. 0.4 nA. These features are not
observed without added molecules. A histogram of the recorded
currents for hundreds of break-junctions (Figure 1b) shows a series
of peaks corresponding to one, two, and three molecules in the
gap.10 Fitting these histograms with Gaussians determines the peak
centers and full widths at half-height (fwhh). These peak values
are then plotted vs peak number (Figure 1b, inset), and the slope is
determined using a least-squares fit, with the data points weighted
using the fwhh as error bars. This slope determines the current at
a fixed bias, and plots of current vs bias for all four molecules are
shown in Figure 2. These current-voltage characteristics are linear
over the measured range, and the slopes of these lines yield the
The molecules consist of a polyene backbone of N ) 5 (I), N )
7 (II), N ) 9 (III), and N ) 11 (IV) carbon-carbon double bonds
in conjugation, each terminated with a benzene ring coupled to a
thiol via a methylene linker. The role of the thiols is to bind to
gold electrodes. A 2 µM solution of freshly deprotected molecules
† Department of Physics and Astronomy.
‡ Department of Chemistry and Biochemistry.
§ Biodesign Institute.
9
1384
J. AM. CHEM. SOC. 2005, 127, 1384-1385
10.1021/ja043279i CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
characteristics are listed in Table 1. Values for bonding to an on-
top site (the more appropriate geometry for the break-junction
method) are well-fitted (ø2 ) 0.003) by an exponential decay G )
G0 exp(-âNN) with âN ) 0.539 ( 0.01 and G0 ) 115 ( 6 nS,
yielding â ) 0.22 ( 0.01 Å-1. Similar values are obtained using
the calculated conductances for hollow-site bonding. The computed
value of â is in better agreement with experiment than is the
prefactor, G0. A partial explanation of this is that â is almost
exclusively a property of the molecule and can be estimated even
without having the molecule in contact with the metal.6 G0 is
sensitive to details of the molecule-metal interface, and this is
probably not well-described by the flat surface used in our
calculations.
Our previous measurement5 of the conductance of III (N ) 9)
yielded G ) 0.2 nS, within 30% of the present measurement,
justifying the interpretation of the break-junction histogram peaks
in terms of single-molecule conduction.10 The somewhat lower
value of current obtained with a nanoparticle contact9 is consistent
with the trends observed in measurements of n-alkane conductance11
caused by suppression of transmission owing to the electronic
structure of the nanoparticle.
Figure 2. Current-voltage characteristics for the four carotenoids. Lines
are linear fits, and error bars (not visible on most points) are (1 SE.
Table 1. Measured and Calculated Conductances (Hollow-Site
Values Are in Parentheses)
N
G (measured) (nS)
G (calculated) (nS)
5
7
9
11
2.06 ( 0.05
0.96 ( 0.07
0.28 ( 0.02
0.11 ( 0.07
7.84 (9.71)
2.60 (3.55)
0.89 (1.01)
0.31 (0.30)
In summary, we have measured the electronic decay constant
for carotenoid polyenes, finding a value in good agreement with
the results of electronic structure calculations.
Acknowledgment. We thank Nongjian Tao for helping us to
set up the break-junction apparatus. This work was supported by a
NIRT grant (ECS 01101175) from the National Science Foundation.
Supporting Information Available: Sample preparation details.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Moore, T. A.; Gust, D.; Mathis, P.; Mialocq, J.-C.; Chachaty, C.;
Bensasson, R. V.; Land, E. J.; Doizi, D.; Liddell, P. A. Nature 1984,
307, 630-632.
Figure 3. Conductance vs number of double bonds in conjugation on linear
(circles) and log (squares) scales. Repeated data points are shown for
N ) 7 and N ) 9 samples. Open symbols are data from ref 5.
(2) Osuka, A.; Yamada, H.; Shinoda, T.; Nozaki, K.; Ohno, T. C. P. L. Chem.
Phys. Lett. 1995, 238, 37-41.
(3) Vretos, J. S.; Steward, D. H.; de Paula, J. D.; Brudvig, G. W. J. Phys.
Chem. B 1999, 103, 6403-6406.
(4) Faller, P.; Pascal, A.; Rutherford, A. W. Biochemistry 2001, 40, 6431-
molecular conductance data listed in Table 1 and plotted vs N in
Figure 3. These data are well-fitted by an exponential decay, G )
G0 exp(-âNN) (lines in Figure 3), with âN ) 0.556 ( 0.09 and
G0 ) 37 ( 18 nS. The measured value for âN, taken together with
a linear distance between pairs of carbon atoms of 2.48 Å (produced
by the zigzag structure of a double bond of length 1.34 Å and a
6440.
(5) Ramachandran, G. K.; Tomfohr, J. K.; Li, J.; Sankey, O. F.; Zarate, X.;
Primak, A.; Terazano, Y.; Moore, T. A.; Moore, A. L.; Gust, D.; Nagahara,
L. A.; Lindsay, S. M. J. Phys. Chem. B 2003, 107, 6162-6169.
(6) Tomfohr, J.; Sankey, O. F. Phys. ReV. B 2002, 65, 245105-245116.
(7) Woitellier, S.; Launay, J. P.; Spangler, C. W. Inorg. Chem. 1989, 28,
758-762.
(8) Leatherman, G.; Durantini, E. N.; Gust, D.; Moore, T. A.; Moore, A. L.;
Stone, S.; Zhou, Z.; Rez, P.; Li, Y. Z.; Lindsay, S. M. J. Phys. Chem. B
1999, 103, 4006-4010.
(9) Cui, X. D.; Primak, A.; Zarate, X.; Tomfohr, J.; Sankey, O. F.; Moore,
A. L.; Moore, T. A.; Gust, D.; Harris, G.; Lindsay, S. M. Science 2001,
294, 571-574.
(10) Xu, B.; Tao, N. J. Science 2003, 301, 1221-1223.
(11) Tomfohr, J.; Ramachandran, G.; Sankey, O. F.; Lindsay, S. M. Making
Contacts to Single Molecules: Are We Nearly There Yet? In Introducing
Molecular Electronics; Fagas, G., Richter, K., Eds.; Springer: Berlin,
2005.
(12) DeRose, J. A.; Thundat, T.; Nagahara, L. A.; Lindsay, S. M. Surf. Sci.
1991, 256, 102-108.
(13) Bart, J. C.; MacGillavry, C. H. Acta Crystallogr. 1968, B24, 1587-1606.
(14) An STO basis set is used in the GAMESS program suite. Schmidt, M.
W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M. S.; Jensen,
J. J.; Koseki, S.; Matsunaga, M.; Nguyen, K. A.; Su, S.; Windus, T. L.;
Dupuis, M.; Montgomery, J. A. J. Comput. Chem. 1993, 14, 1347-1363.
(15) Tomfohr, J. K.; Sankey, O. F. J. Chem. Phys. 2004, 120, 1542-15554.
single bond length of 1.45 Å5,13), leads to â ) 0.22 ( 0.04 Å-1
.
The structural geometry of each molecule was determined by
energy minimization using Hartree-Fock theory14 for determination
of bond lengths that are critical to the determination of electronic
transmission.15 The molecule is linked via sulfur to ideal (111) Au
surfaces at either the “hollow” site above a three Au atom junction
or an “on-top” site directly above a single Au atom (as might be
expected to occur for a molecule under tension as in the break-
junction experiments10). The two approaches give some indication
of the sensitivity of the result to local geometry. The currents were
calculated as a function of applied bias as previously described.5
These current-voltage characteristics were linear in the range
0 < V < 0.6 V, in good agreement with the experimental results,
and the molecular conductances obtained from the slopes of these
JA043279I
9
J. AM. CHEM. SOC. VOL. 127, NO. 5, 2005 1385