Modulation of Electronic Coupling through Self-Assembled Monolayers via Internal Chemical Modification

Article abstract of DOI:10.1021/ja953210y

Self-assembled monolayers of ω-hydroxyalkanethiols at Au electrodes are used to probe the structural dependence of long-range electron transfer.The electron transfer rates for ferricyanide and osmium(III) tris(bipyridyl) are measured at Au electrodes coated with self-assembled monolayers of HO(CH2)n-X-(CH2)mSH where X denotes an ether, olefin, or alkyne function.Each of these modifications decreases the electronic coupling across the monolayers.Ab initio calculations of the neutral diradical splitting energies for modified and unmodified alkanes correctly predict these decreases in electronic coupling.

Full text of DOI:10.1021/ja953210y

6
80
J. Am. Chem. Soc. 1996, 118, 680-684
Modulation of Electronic Coupling through Self-Assembled
Monolayers via Internal Chemical Modification
Jun Cheng, Gotth a` rd S a` ghi-Szab o´ , John A. Tossell, and Cary J. Miller*
Contribution from the Department of Chemistry and Biochemistry, UniVersity of Maryland,
College Park, Maryland 20742
X
ReceiVed September 20, 1995
Abstract: Self-assembled monolayers of ω-hydroxyalkanethiols at Au electrodes are used to probe the structural
dependence of long-range electron transfer. The electron transfer rates for ferricyanide and osmium(III) tris(bipyridyl)
are measured at Au electrodes coated with self-assembled monolayers of HO(CH2)n-X-(CH2)mSH where X denotes
an ether, olefin, or alkyne function. Each of these modifications decreases the electronic coupling across the
monolayers. Ab initio calculations of the neutral diradical splitting energies for modified and unmodified alkanes
correctly predict these decreases in electronic coupling.
Introduction
structure is enforced by the packing of the alkyl chains.5 The
electronic coupling within these monolayers (particularly the
hydrocarbon chain length dependence of the electronic coupling)
has been determined using electron transfer rate measurements
from the underlying Au electrode to tethered redox sites at the
The role of the intervening medium between an electron donor
and acceptor in controlling the probability of long-range electron
transfer is a question of considerable importance to a number
of diverse scientific and technological research areas. These
areas range from understanding long-range electron transfers
in biochemical systems to designing molecular based electronic
6
monolayer/electrolyte solution interface. These results have
been in very good agreement with similar measurements of
homogeneous electron transfer rates between donor/acceptor
1
devices. A serious deficiency in the current understanding of
7
pairs separated by rigid hydrocarbon spacers.
long-range electron transfer is how each atom and each bond
within the intervening medium affects the long-range electronic
coupling. Advances in this field have been slow due in part to
the difficulty of making experimental measurements of long-
range electron transfer rates in rigid, well-defined structures.
Much of the experimental work in this area has been done using
We have employed ω-hydroxyalkanethiol monolayers at gold
electrodes as electron tunneling barriers to characterize solution
8
redox species. Analysis of the electron transfer data at these
monolayer insulated electrodes provides reorganization energies
and electronic coupling parameters which reveal how the
structure of a redox molecule governs its electron transfer
2
donor/acceptor pairs connected by rigid spacer groups. The
9
reactivity. In this paper we describe another important ap-
required rigidity of these spacers is generally achieved by
including fused ring systems, hydrogen bonding, and other
interactions which necessarily increase the chemical and struc-
tural complexity of the spacer. Apart from the increased
synthetic challenges in making systematic structural modifica-
tions, the increased chemical complexity seriously impedes the
detailed theoretical description of the electronic coupling in these
systems.
plication of these monolayer coated electrodes as a direct probe
of the structural dependence of long-range electronic coupling.
By introducing subtle chemical modifications within the alkyl
segment of the ω-hydroxyalkanethiol monolayer, we probe at
the atom leVel how such modifications affect the long-range
electron transfer. The single ether, olefin, and alkyne modifica-
(3) (a) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481-
An alternate strategy in probing these long-range electron
transfer reactions is to utilize self-assembled thiol monolayers
adsorbed on Au surfaces.3 The structure of these thiol mono-
4483. (b) Sabatani, E.; Rubinstein, I.; Maoz, R.; Sagiv, J. J. Electroanal.
Chem. 1987, 219, 365-371. (c) Porter, M. D.; Bright, T. B.; Allara, D. L.;
Chidsey, C. E. D. J. Am. Chem. Soc. 1987, 109, 3559-3568. (d) Bain, C.
D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R.
G. J. Am. Chem. Soc. 1989, 111, 321-335.
4
layers has been the subject of considerable study in recent years.
For simple alkanethiol monolayers, the crystalline all-trans
(4) (a) Strong, L.; Whitesides, G. M. Langmuir 1988, 4, 546-558. (b)
Chidsey, C. E. D.; Liu, G.-Y.; Rowntree, P.; Scoles, G. J. Chem. Phys.
1989, 91, 4421-4423. (c) Fenter, P.; Eberhardt, A.; Eisenberger, P. Science
1994, 266, 1216-1218.
X
Abstract published in AdVance ACS Abstracts, January 1, 1996.
(1) (a) Structure and Bonding: Long Range Electron Transfer in Biology,
Springer-Verlag: New York, 1991; Vol. 75. (b) Metal Ions in Biological
Systems: Electron Transfer Reactions in Metalloproteins, Sigel, H., Sigel,
A., Eds.; Marcel Dekker: New York, 1991; Vol. 27. (c) Newton, M. D.
Chem. ReV . 1991, 91, 767-792. (d) Winkler, J. R.; Gray, H. B. Chem.
ReV. 1992, 92, 369-379. (e) Molecular Electronics-Science and Technol-
ogy: AIP Conference Proceedings; Aviram, A., Ed.; American Institute of
Physics: New York, 1992; Vol. 262.
(5) (a) Nuzzo, R. G.; Fusco, F. A.; Allara, D. L. J. Am. Chem. Soc. 1987,
109, 2358-2368. (b) Ulman, A.; Eilers, J. E.; Tillman, N. Langmuir 1989,
5, 1147-1152.
(6) (a) Li, T. T.-T.; Weaver, M. J. J. Am. Chem. Soc. 1984, 106, 6107-
6108. (b) Chidsey, C. E. D. Science 1991, 251, 919-922. (c) Rowe, G. K.;
Creager, S. E. Langmuir 1991, 7, 2307-2312. (d) Finklea, H. O.; Hanshew,
D. D. J. Am. Chem. Soc. 1992, 114, 3173-3181.
(7) (a) Closs, G. L.; Calcaterra, L. T.; Green, N. J.; Penfield, K. W.;
Miller, J. R. J. Phys. Chem. 1986, 90, 3673-3683. (b) Beratan, D. N. J.
Am. Chem. Soc. 1986, 108, 4321-4326. (c) Oevering, H.; Paddon-Row,
M. N.; Heppener, M.; Oliver, A. M.; Cotsaris, E.; Verhoeven, J. W.; Hush,
N. S. J. Am. Chem. Soc. 1987, 109, 3258-3269. (d) Paddon-Row, M. N.;
Oliver, A. M.; Warman, M.; Smit, K. J.; de Haas, M. P.; Oevering, H.;
Verhoeven, J. W. J. Phys. Chem. 1988, 92, 6958-6962.
(2) (a) Hush, N. S.; Paddon-Row, M. N.; Cotsaris, E.; Overing, H.;
Verhoeven, J. W.; Heppener, M. Chem. Phys. Lett. 1985, 117, 8-11. (b)
Liang, N.; Miller, J. R.; Closs, G. L. J. Am. Chem. Soc. 1990, 112, 5353-
5
354. (c) Vassilian, A.; Wishart, J. F.; Hemelryck, B. V.; Schwarz, H.;
Isied, S. S. J. Am. Chem. Soc. 1990, 112, 7278-7296. (d) Paulson, B.;
Pramod, K.; Eaton, P.; Closs, G. L.; Miller, J. R. J. Phys. Chem. 1993, 97,
1
3042-13045. (e) Ogawa, M. Y.; Wishart, J. F.; Young, Z.; Miller, J. R.;
Isied, S. S. J. Phys. Chem. 1993, 97, 11456-11463. (f) Larson, S. L.;
Hendrickson, S. M.; Ferrere, S.; Derr, D. L.; Elliott, C. M. J. Am. Chem.
Soc. 1995, 117, 5881-5882. (g) Meade, T. J.; Kayyem, J. F. Angew. Chem.,
Int. Ed. Engl. 1995 , 34, 352-354.
(8) (a) Miller, C.; Cuendet, P.; Gr a¨ tzel, M. J. Phys. Chem. 1991, 95,
877-886. (b) Miller, C.; Gr a¨ tzel, M. J. Phys. Chem. 1991, 95, 5225-5233.
(9) Terrettaz, S.; Becka, A. M.; Traub, M. J.; Fettinger, J. C.; Miller, C.
J. J. Phys. Chem. 1995, 99, 11216-11224.
0
002-7863/96/1518-0680$12.00/0 © 1996 American Chemical Society

Structural Dependence of Long-Range Electron Transfer
J. Am. Chem. Soc., Vol. 118, No. 3, 1996 681
Chemicals) with 2,2′-bipyridine as described by Creutz.18 The per-
chlorate salt initially formed was dissolved in CH CN and reprecipitated
3
as the hydrogen sulfate salt via the addition of tetrabutylammonium
hydrogen sulfate. All other chemicals were purchased from Aldrich
and used as received.
tions studied here represent some of the simplest structural
modifications possible. Even so, their affect on the electronic
coupling through the hydrocarbon chain is not easily predicted.
Indeed, a large motivation for these measurements is to serve
as reference points for ab initio calculations of long-range
electronic coupling.1 Following previous ab initio approaches,
we calculate diradical splitting energies from which relative
tunneling rates are determined.11 These ab initio predictions
are compared with the measured electron tunneling rates.
Electrode Fabrication. Au electrodes were fabricated by radio
frequency sputtering ca. 3000 Å from a 99.99% Au target onto
microscope slides through a special mask. A ca. 500 Å chromium
layer was sputtered first to promote adhesion of the gold films. The
Au electrodes were cleaned through successive exposures to chromic
acid and aqueous HF, rinsed with water, and immediately placed into
ethanolic thiol solutions as described previously.12 The Au electrodes
were kept overnight in the ca. 30 mM solution of the corresponding
ω-hydroxyalkanethiols prior to their use in the electrochemical studies.
0
Experimental Section
Syntheses. (a) 14-Hydroxytetradecane-1-thiol (C14) was synthe-
1
2
1
sized from 1,14-tetradecanediol as previously described.
200 MHz, CDCl ) δ 3.56 (t, 2H, -CH -OH), 2.46 (q, 2H, -CH
SH); C NMR (200 MHz, CDCl ) δ 62.7 (-CH -OH), 33.9 (-CH
H NMR
2
The geometric area of the electrodes was 0.13 cm .
(
3
2
2
-
-
Electrochemical Measurements. All electrochemical measure-
ments were made using a BAS-100A electrochemical analyzer in
13
3
2
2
-
1
SH); IR (KBr pellets) 3294 (broad O-H stretch), 2919, 2844 cm
solutions which had been purged with N
jacketed electrochemical cell. The Os(bpy)
2
and were held at 0 °C in a
(
C-H stretches).
2+
3
was oxidized in situ
(
b) 14-Hydroxy-x-oxotetradecane-1-thiol (E14x, x ) 6, 7, 8). All
using ammonium cerium(IV) nitrate prior to its voltammetric charac-
terization. All kinetic measurements were made in aqueous solutions
x
of E14 were synthesized from the corresponding alkanediols and
dibromides as previously described.
E
3
1
3
Spectral characterization of
) δ 3.39 (t, 4H, -CH -O-CH -),
-SH); C NMR (200
-OH), 34.2
containing 0.25 M CF COONa and 3 mM of the redox molecule. All
3
7
1
14
:
H NMR (200 MHz, CDCl
3
2
3
2
potentials were measured and are reported versus a saturated calomel
electrode.
1
.58 (t, 2H, -CH
2
-OH), 2.48 (q, 2H, -CH
2
MHz, CDCl
3
) δ 70.8, 71.0 (-CH
2
-O-CH
2
-), 62.8 (-CH
2
Ab Initio Calculations. Ab initio computations were performed
(
(
-CH
broad O-H stretch), 2938, 2856 cm (C-H stretches).
c) 14-Hydroxy-7-tetradecyne-1-thiol (T147). The synthesis of T147
2
-SH); IR (KBr pellets) 1119 (C-O-C, C-O stretch), 3425
19
11
using GAMESS. Relying on previous theoretical approaches, we
have calculated R,ω-diradical splitting energies for the neutral triplet
diradical using an unrestricted Hartree-Fock SCF calculation. The
geometries of all the trans-alkane and modified alkanes were optimized
at the 3-21G basis set level prior to the introduction of the radical
reporter groups.20
-
1
(
was achieved by the following sequence. 1,6-Hexanediol was converted
to 6-iodo-1-hexanol via refluxing the diol in a mixture of octane and
5
1% aqueous HI. After purification, the alcohol was protected via
1
4
reaction with dihydropyran and reacted with acetylene according to
the method of Beckmann et al. to form 1-tetrahydropyranyloxy-7-
1
5
Results and Discussion
octyne. This terminal alkyne was alkylated with 6-chloro-1-iodohexane
1
6
17
as described by Schwarz et al. and Gensler et al. to give
-tetrahydropyranyloxy-14-chloro-7-tetradecyne. After deprotection of
Heterogeneous electron transfer rate measurements made at
thiol monolayer modified electrodes give a surprisingly easy
and accurate way of probing structural influences on long-range
electronic coupling. Figure 1 shows a schematic representation
of the thiol monolayers investigated. The single internal
chemical modifications were centrally located within the
hydrocarbon chain of the HO(CH2)14SH parent in order to
minimize their interactions with either the Au surface or the
electrolyte solution. The oxygen, trans-olefin, and alkyne
groups were chosen because of their compatibility with the all-
trans ω-hydroxyalkanethiol monolayer structure.
1
1
4
the alcohol, the chloride was converted to the thiol using alcoholic
thiourea followed by treatment in base as described previously.12
NMR (200 MHz, CDCl ) δ 2.12 (m, 4H, -CH -CCtCH -), 3.61 (t,
H, -CH -OH), 2.50 (q, 2H, -CH -SH); C NMR (200 MHz,
CDCl ) δ 80.1, 80.3 (-CtC-), 62.9 (-CH -OH), 33.9 (-CH -SH);
1
H
3
2
2
1
3
2
2
2
3
2
2
-
1
IR (KBr pellets) 3425 (broad O-H stretch), 2931, 2856 cm (C-H
stretches).
(
d) 14-Hydroxy-(E)-7-tetradecene-1-thiol (D147) was synthesized
from 14-hydroxy-7-tetradecyne-1-thiol via Na reduction in liquid NH
3
1
6
1
using the method of Schwarz et al.
H NMR (200 MHz, CDCl
-CHdCH-CH
3
) δ
-),
5
.40 (m, 2H, -CHdCH-), 1.96 (m, 4H, -CH
.60 (t, 2H, -CH -OH), 2.48 (q, 2H, -CH -SH); C NMR (200
) δ 130.2, 130.5 (-CHdCH-), 63.0 (-CH -OH), 34.0
-SH); IR (KBr pellets) 969 (trans-CHdCH, C-H stretch), 3423
2
2
The self-assembly of these modified ω-hydroxyalkanethiols
onto Au electrodes was monitored via surface wetting and
capacitance measurements. We find that each of the modified
monolayers is completely wetted by water, precluding extensive
disruption of the monolayer packing which would expose the
1
3
3
2
2
MHz, CDCl
3
2
(
(
-CH
broad O-H stretch), 2924, 2854 cm (C-H stretches).
e) Tris(2,2′-bipyridyl)osmium(II) hydrogen sulfate (Os(bpy)
was synthesized by the reductive ligand substitution of K OsCl (Strem
2
-
1
(
3
2+)
21
hydrophobic alkyl chains. The capacitances of the ω-hy-
2
6
droxyalkanethiol monolayer coated electrodes listed in Table 1
are also consistent with a close-packed thiol monolayer. As
expected due to the increased polarity or polarizability of the
modifying groups, we find higher capacitances for each of the
ether, alkene, and alkyne modified monolayers. Additional
structural characterization for the ether modified monolayer
(10) (a) Paddon-Row, M. N.; Wong, S. S.; Jordan, K. D. J. Am. Chem.
Soc. 1990, 112, 1710-1722. (b) Evenson, J. W.; Karplus, M. Science 1993,
2
62, 1247-1249.
11) (a) Naleway, C. A.; Curtiss, L. A.; Miller, J. R. J. Phys. Chem.
(
1
9
1
9
991, 95, 8434-8437. (b) Liang, C.; Newton, M. D. J. Phys. Chem. 1992,
6, 2855-2866. (c) Jordan, K. D.; Paddon-Row, M. N. J. Phys. Chem.
992, 96, 1188-1196. (d) Liang, C.; Newton, M. D. J. Phys. Chem. 1993,
7, 3199-3211. (e) Curtiss, L. A.; Naleway, C. A.; Miller, J. R. J. Phys.
(18) Creutz, C.; Chou, M.; Netzel, T. L.; Okumura, M.; Sutin, N. J. Am.
Chem. Soc. 1980, 102, 1309-1319.
(19) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.;
Gordon, M. S.; Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.;
Su, S. J.; Windus, T. L.; Dupuis, M.; Montgomery, J. A. J. Comput. Chem.,
1993, 14, 1347-1363.
(20) (a) Binkley, J. S.; Pople, J. A.; Hehre, W. J. J. Am. Chem. Soc.
1980, 102, 939-947. (b) Gordon, M. S.; Binkley, J. S.; Pople, J. A.; Pietro,
W. J.; Hehre, W. J. J. Am. Chem. Soc. 1982, 104, 2797-2803. (c) Pietro,
W. J.; Francl, M. M.; Hehre, W. J.; Defrees, D. J.; Pople, J. A.; Binkley,
J. S. J. Am. Chem. Soc. 1982, 104, 5039-5048.
(21) The complete wetting of these modified ω-hydroxyalkanethiol
monolayer coated electrodes was inferred from the observation of interfer-
ence fringes as the surface of the electrode was dried in a stream of air.
Chem. 1993, 97, 4050-4058.
(12) (a) Becka, A. M.; Miller, C. J. J. Phys. Chem. 1992, 96, 2657-
2
6
668. (b) Becka, A. M.; Miller, C. J. J. Phys. Chem. 1993, 97, 6233-
239.
(
13) Sinniah, K.; Cheng, J.; Terrettaz, S.; Reutt-Robey, J. E.; Miller, C.
J. J. Phys. Chem. 1995, 99, 14500-14505.
14) Miyashita, M.; Yoshikoshi, A.; Grieco, P. J. Org. Chem. 1977, 42,
772-3774.
15) Beckmann, W.; Doerjer, G.; Logemann, E.; Merkel, C.; Schill, G.;
(
3
(
Z u¨ rcher, C. Synthesis, 1975, 423-425.
(
(
16) Schwarz, M.; Waters, R. M. Synthesis 1972, 567-568.
17) Gensler, W. J.; Prasad, R. S.; Chaudhuri, A. P.; Alam, I. J. Org.
Chem. 1979, 44, 3643-3652.

6
82 J. Am. Chem. Soc., Vol. 118, No. 3, 1996
Cheng et al.
Figure 1. Schematic representation of the self-assembled ω-hydroxyalkanethiol monolayers adsorbed on the Au surface: HS(CH
HS(CH O(CH OH (B); trans-HS(CH CHdCH(CH OH, (C); HS(CH CtC(CH OH (D).
2
)14OH, (A);
2
)
6
2
)
7
2
)
6
2
)
6
)
2 6
2 6
)
Table 1. Capacitances and Relative Electron Tunneling Rates for the ω-Hydroxyalkanethiol Monolayersa
m
k /kum
2
b
3-
(mb)
3+
3
(mb)
molecular structure
capacitance (µF/cm ) (m )
Fe(CN)
6
Os(bpy)
HO(CH
HO(CH
HO(CH
HO(CH
HO(CH
HO(CH
2
)
2
)
2
)
2
)
2
)
2
)
14SH
1.60 ( 0.07 (27)
1.84 ( 0.07 (15)
1.91 ( 0.12 (10)
1.85 ( 0.06 (7)
1.78 ( 0.09 (15)
1.92 ( 0.16 (14)
1
1
7
O(CH)
6
O(CH)
8
O(CH)
6
6
6
7
5
SH
SH
SH
0.68 ( 0.08 (12)
0.70 ( 0.11 (12)
0.77 ( 0.16 (14)
0.28 ( 0.08 (21)
0.38 ( 0.05 (22)
0.66 ( 0.14 (6)
0.57 ( 0.09 (6)
0.50 ( 0.06 (4)
0.41 ( 0.07 (10)
0.77 ( 0.15 (8)
CHdCH(CH)
6
SH
CtC(CH) SH
6
a
The average ratios of the heterogeneous electron transfer rates for the Fe(CN) 3-
relative to that measured at the unmodified HO(CH
independent measurements used to determine the value.
3+
6
and Os(bpy)
3
couples measured at the modified monolayers
b
2
)14SH monolayer are listed along with the standard deviations. m denotes the number of
films was supplied by infrared reflection-adsorption measure-
ments which indicate that the introduction of the ether function
results in no measurable change in the monolayer tilt angle and
a slight, 10°, reduction in the monolayer twist angle.13 Initial
attempts at characterizing the structure of the alkyne modified
monolayers via infrared measurements have been hampered by
significant broadening of the CH2 bands caused by the alkyne
group. This broadening along with a significant decrease in
the CH2 band intensities has impeded the accurate integration
of the infrared bands required for the orientational analysis.
Broadening of the CH2 bands was also observed for the ether
modification. In that case the broadening was ascribed to the
presence of the ether group within ω-hydroxyalkanethiol rather
than disorder caused by the modifying group.
Figure 2. Heterogeneous electron transfer rates versus the formal
Heterogeneous electron transfer rate measurements of solution
redox probes made at these ω-hydroxyalkanethiol monolayers
are also consistent with the formation of low defect density
3
-
overpotential for Fe(CN)
ω-hydroxyalkanethiol monolayers: HS(CH
OH (O); HS(CH CtC(CH OH (0); trans-HS(CH
OH (2). These data were obtained from cyclic voltammograms
6
measured at Au electrodes coated with
14OH (b); HS(CH O(CH
2 6
CHdCH(CH ) -
2
)
2
)
6
2 7
) -
)
2 6
)
2 6
)
2 6
3-
monolayers. The electron transfer rates for Fe(CN)6 and Os-
+
(
bpy)33 were obtained from the linear sweep voltammograms
3
measured at 5.12 V/s in aqueous solutions of 0.25 M CF COONa at 0
after correction for both diffusion limitations and the double
layer effect as described previously,9, 12 (Figures 2 and 3). In
every instance, the internally modified monolayers give smaller
electron transfer rates, indicating that these modified ω-hy-
droxyalkanethiol monolayers block the long-range electron
transfer more efficiently than the parent ω-hydroxyalkanethiol
°C.
the electronic coupling across the monolayers.23 By comparing
the heterogeneous electron transfer rate constants, one can gauge
how these subtle chemical modifications affect the long-range
electronic coupling through the organized monolayers (Table
2
2
monolayer.
(
23) The heterogeneous electron transfer rate measured at these thiol
The differences in the electron transfer rates measured at these
monolayer coated electrodes are due primarily to a change in
modified electrodes depends on both the electronic coupling between the
electrode and redox probe and the reorganization energy of the redox probe.
A potential problem with comparing heterogeneous electron transfer rates
between different monolayer coated electrodes is that the reorganization
energy of a redox probe may vary at each of the electrodes due to varying
degrees of electrostatic screening of the redox probe by the electrode image
charge. For the monolayers studied in this work, these image charge effects
are anticipated to be negligible. (See Liu, Y.-P.; Newton, M. D. J. Phys.
Chem. 1994, 98, 7162-7169.)
(22) If these modifications resulted in more highly defective monolayers,
one might expect an increase in the electron transfer rate for a solution
species. Because a HO(CH2)14SH monolayer decreases the heterogeneous
electron transfer rate of solution species by over six orders of magnitude,
even a small number of defects allowing a closer approach of the redox
molecule to the electrode surface could result in significantly larger currents.

Structural Dependence of Long-Range Electron Transfer
J. Am. Chem. Soc., Vol. 118, No. 3, 1996 683
Table 2. Comparison between ab Initio Electronic Coupling
Calculations and Experimentsa
∆
Ean
∆Ecat
(mH)
molecular structure
(mH)
kthb
kexb
•
•
•
•
•
•
CH2(CH2)3CH2(CH2)3CH2
CH2(CH2)3O(CH2)3CH2
4.158
3.389
-2.691
-1.174
8.730
10.330
6.214
-7.718
-4.186
15.696
1
1
•
0.44 0.67
1
0.27 0.35
5.5 0.38
1
•
CH2(CH2)3CH2CH2(CH2)3CH2
CH2(CH2)3CHdCH(CH2)3CH2
CH2(CH2)3CtC(CH2)3CH2
1
•
•
c
c
Li CH3(CH2)3CH2(CH2)3CH3 Li
Li CH3(CH2)3O(CH2)3CH3 Li
0.01340
0.01096
0.01185
0.00870 0.61 0.67
1
Li CH3(CH2)3CH2CH2(CH2)3CH3 Li -0.00771 -0.00703
1
1
Li CH3(CH2)3CHdCH(CH2)3CH3 Li -0.00567 -0.00472 0.50 0.35
Li CH3(CH2)3CtC(CH2)3CH3 Li
0.00547
0.00468 0.47 0.38
a
The theoretical relative rates were calculated from the sum of the
anion and cation splitting energies. (See text.) The experimental
Figure 3. Heterogeneous electron transfer rates versus the formal
relative rates were taken as the average relative rates from Table 1.
3
+
overpotential for Os(bpy)
ω-hydroxyalkanethiol monolayers: HS(CH
OH (O); HS(CH CtC(CH OH (0); trans-HS(CH
OH (2). These data were obtained from cyclic voltammograms
measured at 5.12 V/s in aqueous solutions of 0.25 M CF COONa at 0
3
measured at Au electrodes coated with
14OH (b); HS(CH O(CH
CHdCH(CH
b
k
th and kex denote the theoretical and experimental relative rates,
2
)
2
)
6
2
)
7
-
-
c
respectively. This relative ratio was taken as being equal to that
)
2 6
)
2 6
)
2 6
2
)
6
obtained from the Fe(CN)63- data in Table 1.
3
This direct interaction is not an effective electronic coupling
pathway for the electron through the monolayer. Indeed, several
groups have presented theoretical analyses of the electronic
coupling through saturated hydrocarbon spacers and concluded
that the electronic coupling in these systems should be domi-
°
C.
1
). The relative heterogeneous electron transfer rates are not a
strong function of the applied potential so that a single average
relative rate is given for each modification. The electronic
coupling is highest for the unmodified ω-hydroxyalkanethiol.
The electronic coupling of the ether modified monolayer is
smaller and independent of whether the oxygen is positioned
at the 6, 7, or 8 position along the alkyl chain.24 The electronic
coupling decreases by approximately a factor of 3 when a single
trans-olefin is introduced. This is a slightly greater decrease
than would be caused by increasing the monolayer length by
nated by the overlap and energy matching between orbitals
11
located on nearby atoms.
The direct electronic coupling
beyond these 1st-, 2nd-, and 3rd-nearest-neighbor interactions
becomes insignificant to the overall electronic coupling through
the hydrocarbon spacer. Therefore, the effect of these modifica-
tions is to disrupt the electronic coupling through the hydro-
carbon network by decreasing these near-neighbor interactions
at the modification. In this light one could predict that any
single chemical modification which permits close packing of
the thiol monolayer should decrease the electronic coupling
through the monolayer. The extent of the decrease would
depend on the energy and spatial overlap mismatch between
the modification and the polymethylene chain.
1
2
one methylene unit. The single alkyne modification reduces
the electronic coupling less than that observed for the olefin.
Except for the alkyne modified monolayer, the relative ratios
determined using Fe(CN)63 and Os(bpy)3 are in reasonable
agreement. The discrepancy in the electronic coupling ratio
for the alkyne using the two redox couples is most likely due
to some penetration of the more hydrophobic Os complex within
the alkyne monolayer. In order to probe for the penetration of
the redox complexes at monolayer defect sites, we have repeated
these voltammetric experiments in the presence of 1.5 mM
-
3+
In order to predict quantitatively the effect of these modifica-
tions on the long-range electronic coupling, we have first
calculated the neutral triplet diradical splitting energies for
•
•
CH -(CH ) -X-(CH ) -CH , where X is the centrally
2
2 n
2 n
2
located chemical modification. For these unrestricted Hartree-
Fock calculations, we have employed a 3-21G basis set which
has been shown to be the simplest basis set which produced
1
-octanol. This amphiphile was previously found to act as a
12a
competitive binder for these defect sites.
Except for the
reduction rate for Os(bpy)33 measured at the alkyne modified
monolayer, we observe little change in these electron transfer
rates measured in the presence of 1-octanol. The alkyne
modified monolayer coated electrode rates decrease by ap-
proximately 40%, lowering the relative electronic coupling to
a value close to that measured using ferricyanide.
+
11
reasonable electronic coupling estimates. A more flexible
+
+
6-31G basis set was found to produce results within 20% of
those obtained using the 3-21G basis set. For each of the
structures shown in Table 2, the energy differences between
the two highest occupied radical orbitals, ∆Ecat, and the two
lowest unoccupied radical orbitals, ∆Ean, are averaged to give
a measure of the electronic coupling through the hydrocarbon
The decreases in the electronic coupling caused by these
internal modifications may seem surprising especially for the
olefin and alkyne modifications. One may have expected the
π and π* orbitals to couple the electron through the thiol
monolayer more efficiently due to their smaller HOMO/LUMO
energy gap when compared with the σ and σ* orbitals available
in the unmodified monolayer. However, because these modi-
fications are buried within the monolayer, the electronic coupling
of the electrode and redox orbitals by the orbitals of the
modification would involve two long through-space interactions.
25
chains. These calculations were repeated changing the donor/
acceptor reporter orbitals from carbon centered radicals to a pair
of nonbonded Li atom radicals separated by 4 Å from the
terminal hydrogens of the CH3(CH2)n-X-(CH2)nCH3 molecule.
The 4 Å separation was chosen to minimize bonding interactions
between the Li atoms and the hydrocarbon orbitals. At smaller
separation distances, the energy levels for the hydrocarbon
orbitals change significantly upon addition of the Li atoms. The
purpose of this change in the calculation was to mimic more
closely the donor/acceptor structure of the monolayer experi-
ments.
(24) Whether the ether function is positioned at an even or odd site along
the hydrocarbon chain does affect the measured electron transfer rates for
charged redox probes via a change in the electrode’s potential of zero charge
These ab initio electronic coupling estimates are compared
with the experimental results in Table 2. In the Fermi’s golden
(
pzc). We observe a ca. (130 mV shift in the pzc for the ether modified
thiol monolayers relative to the unmodified ω-hydroxyalkanethiol. (Please
see ref 13.) The systematic differences in the measured electron transfer
rates caused by this electrostatic effect are largely eliminated by the double
layer correction.
(25) This is the Koopman’s theorem approach. It should be noted that
only the relative magnitude of this coupling energy versus the unmodified
hydrocarbon chain is required in this analysis.

6
84 J. Am. Chem. Soc., Vol. 118, No. 3, 1996
Cheng et al.
rule limit, the square of the electronic coupling energy is
proportional to the probability of long-range electron transfer
through the hydrocarbon spacer.26 The squared ratio of the
electronic coupling energies between the modified and unmodi-
fied hydrocarbons is compared to the average ratio of the
heterogeneous electron transfer rate constants measured in
highlight the importance of selecting appropriate model struc-
tures for the estimation of the electronic coupling.
Conclusions
Some of the most fundamental aspects of long-range electron
transfer can be studied with surprising experimental ease using
these self assembled ω-hydroxyalkanethiol monolayers. Fea-
tures advantageous for the study of long-range electron coupling
in this electrochemical system include the simplicity of intro-
ducing single or multiple point modifications to the alkyl chains
and the ability to measure both oxidative and reductive electron
27
Figures 2 and 3. For the ether and olefin modified monolayers,
one can observe that the two ab initio estimates correctly predict
a decrease in the electronic coupling. The size of these predicted
decreases are also in reasonable agreement with the experimen-
tally observed values. The agreement between theory and
experiment for the alkyne modification depends on the details
of the ab initio calculation. Using covalently bonded methylene
radical reporter groups, the electronic coupling through the
alkyne modified hydrocarbon is predicted to be greatly enhanced
relative to the unmodified hydrocarbon. In contrast, when the
nonbonded Li atoms are used for the donor and acceptor, the
calculation is in better agreement with the experimental findings.
The difference in the two ab initio estimates for the alkyne
electronic coupling is most likely due to the difference in the
symmetry of the reporter donor and acceptor groups. The π
radical orbitals are more directional than the 2s orbitals of the
Li atoms. In the calculation using the π radical orbitals, the
orientation of the π radical orbitals with respect to each other
and the central modification was found to affect dramatically
the electronic coupling estimates in agreement with previous
30
transfers over a continuous range of voltages. In this work
we find that single atom or bond changes within a hydrocarbon
spacer lead to significant decreases in the long-range electronic
coupling. The modulation in the electronic coupling occurs even
though the number of covalent bonds has not been changed.
The approximation commonly used in theoretical studies of
electronic coupling through proteins that all covalent bonds
31
should display the same electronic coupling, while reasonable
in a global treatment of protein electron transfer, breaks down
at the local level. The electronic coupling is observed to be a
complex function of the structure of the intervening medium at
the atom level.
We are continuing these studies to probe the range of
modifications which can be introduced within these monolayers
without destroying their ability to act as tunneling barriers.
Because the ability to place even a few modifications at specific
sites within these monolayers opens up a very large number of
possible monolayer structures, the development of predictive
theories is clearly needed to aid in designing combinations of
modifications which impart useful electronic properties to these
monolayers. Important future directions include probing quan-
tum interference effects in monolayers composed of two or more
compatible thiols and designing molecule-based current recti-
fication and optical transducers based on these thiol tunneling
barriers.
1
1
studies. There appears to be a long-range coupling between
parallel π orbitals which greatly enhances the overall electronic
coupling estimate.28 This is essentially an artifact of the
calculation which we have tried to eliminate. We have varied
the length of the hydrocarbon chain from 5 to 14 carbons to
determine if a greater separation between the π orbitals would
eliminate this effect. We see the same anomalously high
coupling estimate from the diradical calculation. A second
change made was to increase the flexibility of the basis set used
to perform the calculation. We find that using either a 4-31G*
+
+
29
or a 6-31G
basis set does not markedly affect the high
relative electronic coupling prediction. These calculations
Acknowledgment. The authors gratefully acknowledge
financial support from the National Association of Corrosion
Engineers, the Petroleum Research Fund, and the National
Science Foundation (CHE-9417357, CHE-9503348).
(
26) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265-
22.
27) We are comparing the electronic coupling calculated for a single
3
(
hydrocarbon chain with the measured coupling through an assembly of
ω-hydroxyalkanethiol monolayers. Given that the intermolecular electronic
coupling is likely to be much weaker than the intramolecular electronic
coupling, this simplification is a reasonable starting point.
JA953210Y
(30) The potential control afforded by this electrochemical system allows
one to probe how electric fields within the tunneling barrier affect the long-
range coupling. Such control should allow one to probe the relative
importance of electron and hole tunneling in the long-range electron transfer.
For these modifications, we observe little potential dependence of the
electronic coupling suggesting that both electron and hole mechanisms are
equally operative.
(31) (a) Beratan, D. N.; Onuchic, J. N.; Hopfield, J. J. J. Chem. Phys.
1987, 86, 4488-4498. (b) Beratan, D. N.; Betts, J. N.; Onuchic, J. N. Science
1991, 253, 1285-1288. (c) Wuttke, D. S.; Bjerrum, M. J.; Winkler, J. R.;
Gray, H. B. Science 1992, 256, 1007-1009. (d) Regan, J. J.; Risser, S. M.;
Beratan, D. N.; Onuchic, J. N. J. Phys. Chem. 1993, 97, 13083-13088. (e)
Ullmann, G. M.; Kostic′, N. M. J. Am. Chem. Soc. 1995, 117, 4766-4774.
(28) It is interesting to note that for the olefin calculation, the reporter
orbitals were held perpendicular to the olefin π orbitals which eliminates
this coupling pathway accounting for the good agreement with the observed
relative electronic coupling.
(
29) (a) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971,
5
1
2
4, 724-728. (b) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys.
972, 56, 2257-2261. (c) Hariharan, P. C.; Pople, J. A. Mol. Phys. 1974,
7, 209-214. (d) Gordon, M. S. Chem. Phys. Lett. 1980, 76, 163-168. (e)
Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213-222. (f)
Petersson, G. A.; Bennett, A.; Tensfeldt, T. G.; Al-Laham, M. A.; Shirley,
W. A.; Mantzards, J. J. Chem. Phys. 1988, 89, 2193-2218. (g) Petersson,
G. A.; Al-Laham, M. A. J. Chem. Phys. 1991, 94, 6081-6090.