Preparation of high quality electrical insulator self-assembled monolayers on gold. Experimental investigation of the conduction mechanism through organic thin films

Article abstract of DOI:10.1021/ja0548992

Self-assembled monolayers (SAMs) form highly ordered, stable dielectrics on conductive surfaces. Being able to attach larger-area contacts in a MIM (metal-insulator-metal) diode, their electrical properties can be determined. In this paper, the electrical conduction through thiolate SAMs of different alkyl chain lengths formed on gold surfaces were studied and discussed. The influence of the headgroup with respect to the surface quality and prevention of short circuits is investigated. Phenoxy terminated alkanethiols were found to form high quality SAMs with perfect insulating properties. Synthesis of the required terminally substituted long chain thiols have been developed. The I(V) characteristics of MIM structures formed with these SAMs are measured and simulated according to theoretical tunneling models for electrical conductivity through thin organic layers. SAM based electronic devices will become especially important for future nanoscale applications, where they can serve as insulators, gate dielectric of FETs, resistors, and capacitor structures.

Full text of DOI:10.1021/ja0548992

Published on Web 11/18/2005
Preparation of High Quality Electrical Insulator
Self-Assembled Monolayers on Gold. Experimental
Investigation of the Conduction Mechanism through Organic
Thin Films
Steffen Maisch, Frank Buckel, and Franz Effenberger*
Contribution from the Institut fu¨r Organische Chemie, UniVersita¨t Stuttgart, Pfaffenwaldring 55,
D-70569 Stuttgart, Germany
Received July 21, 2005; E-mail: franz.effenberger@oc.uni-stuttgart.de
Abstract: Self-assembled monolayers (SAMs) form highly ordered, stable dielectrics on conductive surfaces.
Being able to attach larger-area contacts in a MIM (metal-insulator-metal) diode, their electrical properties
can be determined. In this paper, the electrical conduction through thiolate SAMs of different alkyl chain
lengths formed on gold surfaces were studied and discussed. The influence of the headgroup with respect
to the surface quality and prevention of short circuits is investigated. Phenoxy terminated alkanethiols were
found to form high quality SAMs with perfect insulating properties. Synthesis of the required terminally
substituted long chain thiols have been developed. The I(V) characteristics of MIM structures formed with
these SAMs are measured and simulated according to theoretical tunneling models for electrical conductivity
through thin organic layers. SAM based electronic devices will become especially important for future
nanoscale applications, where they can serve as insulators, gate dielectric of FETs, resistors, and capacitor
structures.
Introduction
proximately 2.8 and 5.2 eV for fatty acids and perfluorinated
fatty acids, respectively.4
The development of ultrathin insulators with high stability
against electrical breakdown is a decisive problem in nanoelec-
tronics.2 Today, the two electrode surfaces in commercially
available thin film capacitors are normally separated by
polymers of micrometer thickness. Mann and Kuhn3 were the
first who investigated electrical conduction through Langmuir-
Blodgett (LB) monolayers sandwiched between two metal
electrodes. They observed an exponential dependence of the
dc conductivity (σ) on the fatty acid chain length at room
temperature and at -35 °C. From these results, they proposed
a tunneling mechanism through fatty acid monolayers. To gain
better insight into the mechanism of electrical conduction, Sagiv
et al. studied Al-adsorbed monolayer-Al junctions.4 Adsorbed
monolayers of saturated fatty acids (14 to 23 C-atoms), short
chain perfluorinated acids (7 to 10 C-atoms) and n-octadecyl-
trichlorosilane gave the following results: At room temperature
there was no definite relation between the dc conductivity and
the length of the fatty acid. At 77 K and lower temperature,
however, an exponential dependence on the chain length could
be observed. These results suggest that only at lower temperature
the conduction through the monolayer agrees with the tunneling
mechanism. The barrier height was estimated to be ap-
Within the past few years great advances in the preparation
of self-assembled monolayers as well as in the determination
of their structure and their chemical and physical properties have
been achieved.5 From a great number of investigations, it is
known that long-chain alkyl monolayers have excellent insulat-
ing properties. However, due to the difficulties involved in
producing monolayers of good quality and reproducibility
meaningful results concerning electrical conductivity have not
always been achieved.
Since in this paper exclusively investigations on the electrical
conductivity of gold-SAM-metal (electrolyte) junctions are
reported, only the literature of these and closely related systems
will be discussed.
A relatively simple method for the generation of a junction
gold-SAM-metal is to use an STM tip as the counter electrode
of an alkanethiol SAM on gold.6,7 This conductivity system was
applied to mixed layers from alkanethiols and long-chain
conjugated π-systems (e.g., oligophenylethinyl derivatives).
From STM imaging an electron transfer from the gold surface
to the STM tip via the conjugated π-systems was proposed.6a,b,h,7e
Besides the direct application of the STM tip as counter
electrode also gold-SAM-metal cluster-STM tip junctions were
intensively investigated with respect to the electrical properties
(1) (a) Maisch, S. Diplomarbeit, Universita¨t Stuttgart 2000, and Dissertation
in preparation. (b) Buckel, F. Dissertation, Universita¨t Stuttgart 2000.
(2) (a) Kuhn, H. Pure Appl. Chem. 1981, 53, 2105-2122. (b) Tieke, B. AdV.
Mater. 1990, 2, 222-231. (c) Ratner, M. A.; Jortner, J. Molecular
Electronics; Blackwell Science Ltd.: London, 1997.
(3) Mann, B.; Kuhn, H. J. Appl. Phys. 1971, 42, 4398-4405.
(4) Polymeropoulos, E. E.; Sagiv, J. J. Chem. Phys. 1978, 69, 1836-1847.
(5) (a) Bain, C. D.; Whitesides, G. M. Angew. Chem., Int. Ed. Engl. 1989, 28,
506-512. (b) Ulman, A. Chem. ReV. 1996, 96, 1533-1554. (c) Schreiber,
F. Progress Surf. Sci. 2000, 65, 151-256. (d) McCreery, R. L. Chem.
Mater. 2004, 16, 4477-4496.
9
10.1021/ja0548992 CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 17315-17322
17315

A R T I C L E S
Maisch et al.
of SAMs.7 So, for example, nanometer-sized gold clusters have
been deposited from a cluster beam onto thiol terminated
SAMs.7a,b The deposition of metal clusters on SAMs can,
however, also be performed by chemical reduction of metal
salts7c or by electrodeposition.8 The replacement of the STM
tip as counter electrode could be realized directly with gold-
SAM-metal junctions,9 prepared, for example, by evaporation
of titanium onto a 4-phenylthiophenolate-SAM,9b whereby the
area of contact is restricted to e30 nm diameter. The electrical
properties of these systems have not been reported. A further
metal-insulator-metal junction has been assembled by deposi-
tion of silver paint on top of a gold-octadecanethiol SAM
surface, showing that the SAM increased the resistance of the
device by 10 orders of magnitude compared to devices without
monolayer.10
The most important method for measuring rates of electron
transport across organic thin films has been established by
Whitesides et al. using Hg-SAM||SAM-metal junctions.11a-c
The current that flows across these junctions depends both
on the molecular structure and on the thickness of the SAMs.
The current density decreased with increasing distance between
the electrodes. From the experimental results, it was concluded
that the mechanism of electron transport is super exchange
tunneling. The absolute magnitude of the current density in these
junctions is in agreement with the results of Hg-SAM||SAM-
Hg junctions described by Majda et al.12 These junctions using
a SAM supported on the surface of a drop of mercury provide
an interesting test bed for screening the intrinsic electrical
properties of SAMs, but they probably cannot be developed into
practically useful microelectronic components. Besides the
difficulty in handling electronic devices containing liquid Hg
electrodes, these junctions show a relatively low electrical
breakdown voltage (BDV),11 a severe disadvantage for the
Chart 1
application of SAMs as insulators. The BDV is revealed by an
abrupt increase in current flowing across the junction in response
to increasing the applied potential. The BDV depends expectedly
on the thickness and the tilt angle of SAMs but also on the
type of the metals applied as electrodes (Au, Au/Hg, Cu, Hg,
Ag).11a Interestingly, a dependence of BDV on the chemical
structure of SAMs was not observed so far, SAMs composing
aliphatic or aromatic residues, respectively, do not differ in
BDV.11
The practically most simple method to prepare gold-SAM-
metal junctions appears to be undoubtedly the evaporative
deposition of metals onto gold-SAM surfaces. Penetration of
metal atoms and metal clusters, respectively, into the monolayer,
resulting in electrical short circuits, has probably prevented a
general application of this method.13 Penetration of metals into
SAMs might be hindered by SAMs with functionalized surfaces
(e.g., with COOH end groups), which are capable of reacting
with metals,13 or by decreasing the temperature during the
evaporation process.14 Interestingly, SAM forming molecules
with methyl side groups have fewer defects than SAMs of
comparable n-alkanethiols.13c
In the present publication, we report on the preparation and
the properties of gold-SAM-metal junctions with highly
ordered SAMs of ω-substituted alkanethiols such as 1 and 2
(Chart 1) where the ω-substituents are aromatic or heteroaro-
matic residues capable of giving strong π-electron interactions
(π-stacking).15 These junctions surprisingly show no electrical
short circuits in large-area and have a markedly increased
electrical breakdown voltage. They should therefore have
decisive advantages for various applications in electronic
devices. These stable systems should also enable to determine
the mechanism of electron transport through organic thin films
experimentally without any of the assumptions necessary for
other systems applied until now.11,12
(6) (a) Bumm, L. A.; Arnold, J. J.; Cygan, M. T.; Dunbar, T. D.; Burgin, T.
P.; Jones II, L.; Allara, D. L.; Tour, J. M.; Weiss, P. S. Science 1996, 271,
1705-1707. (b) Cygan, M. T.; Dunbar, T. D.; Arnold, J. J.; Bumm, L. A.;
Shedlock, N. F.; Burgin, T. P.; Jones II, L.; Allara, D. L.; Tour, J. M.;
Weiss, P. S. J. Am. Chem. Soc. 1998, 120, 2721-2732. (c) Datta, S.; Tian,
W.; Hong, S.; Reifenberger, R.; Henderson, J. I.; Kubiak, C. P. Phys. ReV.
Lett. 1997, 79, 2530-2533. (d) Dhirani, A.; Lin, P.-H.; Guyot-Sionnest,
P.; Zehner, R. W.; Sita, L. R. J. Chem. Phys. 1997, 106, 5249-5253. (e)
Bumm, L. A.; Arnold, J. J.; Dunbar, T. D.; Allara, D. L.; Weiss, P. S. J.
Phys. Chem. B 1999, 103, 8122-8127. (f) Fan, Fu-R. F.; Yang, J.; Cai,
L.; Price, D. W., Jr.; Dirk, S. M.; Kosyukin, D. V.; Yao, Y.; Rawlett, A.
M.; Tour, J. M.; Bard, A. J. J. Am. Chem. Soc. 2002, 124, 5550-5560. (g)
James, D. K.; Tour, J. M. Chem. Mater. 2004, 16, 4423-4435. (h)
Wakamatsu, S.; Akiba, U.; Fujihira, M. Jpn. J. Appl. Phys. 2002, 41, 4998-
5002.
(7) (a) Dorogi, M.; Gomez, J.; Osifchin, R.; Andres, R. P.; Reifenberger, R.
Phys. ReV. 1995, B52, 9071-9077. (b) Andres, R. P.; Bein, T.; Dorogi,
M.; Feng, S.; Henderson, J. I.; Kubiak, C. P.; Mahoney, W.; Osifchin, R.
G.; Reifenberger, R. Science 1996, 272, 1323-1325. (c) Grummt, U.-W.;
Geissler, M.; Drechsler, T.; Fuchs, H.; Staub, R. Angew. Chem., Int. Ed.
Engl. 1998, 37, 3286-3289. (d) Beebe, J. M.; Eugelkes, V. B.; Miller, L.
L.; Frisbie, C. D. J. Am. Chem. Soc. 2002, 124, 11268-11269. (e) Engelkes,
B. V.; Beebe, J. M.; Frisbie, C. D. J. Am. Chem. Soc. 2004, 126, 14287-
14296.
(8) Cavalleri, O.; Bittner, A. M.; Kind, H.; Kern, K.; Greber, T. Z. Phys. Chem.
1999, 208, 107-136.
(9) (a) Reed, M. A.; Zhou, C.; Muller, C. J.; Burgin, T. P.; Tour, J. M. Science
1997, 278, 252-254. (b) Zhou, C.; Deshpande, M. R.; Reed, M. A.; Jones,
L., II; Tour, J. M. Appl. Phys. Lett. 1997, 71, 611-613.
(10) Baker, M. V.; Landau, J. Aust. J. Chem. 1995, 48, 1201-1211.
(11) (a) Haag, R.; Rampi, M. A.; Holmlin, R. E.; Whitesides, G. M. J. Am.
Chem. Soc. 1999, 121, 7895-7906. (b) Holmlin, R. E.; Haag, R.; Chabinyc,
M. L.; Ismagilov, R. F.; Cohen, A. E.; Terfort, A.; Rampi, M. A.;
Whitesides, G. M. J. Am. Chem. Soc. 2001, 123, 5075-5085. (c) Holmlin,
R. E.; Ismagilov, R. F.; Haag, R.; Mujica, V.; Ratner, M. A.; Rampi, M.
A.; Whitesides, G. M. Angew. Chem., Int. Ed. 2001, 40, 2316-2320.
(12) (a) Slowinski, K.; Chamberlain II, R. V.; Bilewicz, R.; Majda, M. J. Am.
Chem. Soc. 1996, 118, 4709-4710. (b) Slowinski, K.; Fong, H. K. Y.;
Majda, M. J. Am. Chem. Soc. 1999, 121, 7257-7261. (c) Slowinski, K.;
Chamberlain, R. V.; Miller, C. J.; Majda, M. J. Am. Chem. Soc. 1997,
119, 11910-11919.
Results and Discussion
Synthesis of the Thiols 2a-c. While the synthesis of the
thiols 1a,b has been published already,16a the phenoxy-
terminated thiols 2a-c were prepared as outlined in Scheme 1.
The known bromoalkenes 3a and 3b,c16b have been reacted
with potassium phenolate, prepared from potassium methanolate
(13) (a) Jung, D. R.; Czanderna, A. W. Crit. ReV. Solid State Mater. Sci. 1994,
19, 1-54. (b) Herdt, G. C.; King, D. E.; Czanderna, A. W. Z. Phys. Chem.
1997, 202, 163-196. (c) Braach-Maksvytis, V.; Raguse, B. J. Am. Chem.
Soc. 2000, 122, 9544-9545.
(14) Tarlov, M. J. Langmuir 1992, 8, 80-89.
(15) (a) Hunter, C. A.; Lawson, K. R.; Perkins, J.; Urch, C. J. J. Am. Chem.
Soc. Perkin Trans 2001, 651-669 (b) Whitten, D. G.; Chen, L.; Geiger,
H. C.; Perlstein, J.; Song, X. J. Phys. Chem. B 1998, 10098-10110.
(16) (a) Buckel, F.; Persson, P.; Effenberger, F. Synthesis 1999, 953-958. (b)
Effenberger, F.; Heid, S. Synthesis 1995, 1126-1130. (c) Seifritz, S.
Dissertation, Universita¨t Stuttgart 2001. (d) Aggarwal, V. K.; Angelaud,
R.; Bihan, D.; Blackburn, P.; Fieldhouse, R.; Fonquerna, S. J.; Ford, G.
D.; Hynd, G.; Jones, E.; Jones, R. V. H.; Jubault, P.; Palmer, M. J.; Ratcliffe,
P. D.; Adams, H. J. Chem. Soc., Perkin Trans. 1 2001, 2604-2622.
9
17316 J. AM. CHEM. SOC. VOL. 127, NO. 49, 2005

Electrical Insulator Self-Assembled Monolayers on Gold
A R T I C L E S
Scheme 1. Reflexion Spectrum of the SAM Formed from Compound 2c on Gold.
and phenol, in DMF16c to give the alkenylphenyl ethers 4a-c,
which were isolated after chromatographic purification in 79-
92% yield. The terminal thioacetate function was introduced
by addition of thioacetic acid to the alkenes 4a-c using azo-
isobutyronitrile (AIBN) as radical initiator at 60 °C.16d The
thioacetates 6a-c were obtained in 69-85% yield after chro-
matography on silica gel. An alternative route to the thioacetates
6a-c is shown in Scheme 1 for the preparation of 6c. In a
modified known procedure,16a phenoxy-substituted octadecene
4c was hydroborated with 9-borabicyclo[3.3.1]nonane (9-
BBN)17 and subsequently oxidized with H2O2 to give the
corresponding alcohol 5 in 64% yield. Analogously to a
literature procedure,16a the alcohol 5 was converted in a
Mitsunobu reaction with triphenylphosphine/diethyl azodicar-
boxylate (DEAD) and thioacetic acid to give 6c in an overall
yield of 39%. The thiols 2a-c were eventually obtained by
reduction of the thioacetates 6a-c with LiAlH4.16a
Preparation and Characterization of the Thiol Monolayers
(SAMs) on Gold Surfaces. Commercial gold substrates were
either pretreated according to a known procedure18 or cleaned
with “piranha acid” (33% H2O2/concentrated H2SO4 3:7) and
subsequently tempered on a heating plate for 24 h at 380 °C.
To smooth the surface roughness, the gold wafers were
connected as anode in an electrolyte containing tetrachloroauric-
(III) acid versus platinum as cathode, whereby the gold surface
was partly electrolytically dissolved within 72 h using a sine
alternating voltage (2.4 Vpp). For the monolayer preparation
the gold substrates were immersed in a solution of the thiols 1
and 2, respectively, in ethanol for 24 h. After cleaning and
drying, the samples were characterized by X-ray photoelectron
(XPS) and FTIR spectroscopy as well as by water advancing
contact angle measurement.
from the ν(CdC)aryl vibration band at 1498 cm-1 with an
intensity comparable to the intensity of the symmetric νs(CH2)
band in the reflection absorption IR (RAIR) spectra of SAMs
of 1b.19c In KBr pellets the intensity of the ν(CdC)aryl band is
considerably weaker than the intensity of the νs(CH2) band.19c
Preparation and I(V) Characteristics of Gold-SAM-Al
Junctions (Method A). The gold-SAM-Al junctions were
prepared as follows: aluminum contacts with 0.5 mm diameter
and 250-260 nm height were built by thermal evaporation of
aluminum under vacuum (10-6 mbar) through a mask on top
of the SAMs of the aryl terminated compounds 1a,b on gold,
at a deposition rate of approximately 5 nm s-1. A reproducible
insulator behavior of the roughly 3 nm thick SAMs in the gold-
SAM (1)-Al junctions was unambiguously ascertained for
numerous contacts (about 30-50% of the measured Al dots),
whereby a current only in the range of nA was measured at a
dc voltage of (1 V. On the contrary, in comparable junctions
with methyl terminated SAMs (obtained with eicosanethiol on
gold), already at very low voltage (<(0.1 V) a current in the
range of amperes was measured for all aluminum dots, indicat-
ing many short circuits, and thus these methyl terminated SAMs
show no insulator behavior.
Although the gold-SAM-Al junctions described so far show
the decisive influence of terminal substituent of SAMs on the
I(V) characteristics, this method to build up junctions has
disadvantages. The evaporative aluminum deposition obviously
still allows metal penetration into the monolayer.20 Even in aryl
terminated SAMs, electrical short circuits in roughly 50% of
the samples have been measured as mentioned above. We
therefore were interested in developing other methods to prepare
gold-SAM-metal junctions with perfect insulating behavior in
all cases.
Preparation and Conductivity Measurements of Gold-
SAM-Metal (Electrolyte) Junctions (Method B). To exclude
metal penetration during the evaporation process the metal
deposition onto self-assembled monolayers by means of elec-
trolysis and contacting with electrolyte solution appeared to be
promising, because of the even current distribution over
relatively large areas provided by the bulk electrolyte. Therefore,
we studied the cathodic deposition of thin films of noble metals
(e.g., Au) from the respective electrolyte solution on the
monolayer surface at a very low current which had to pass
through the SAM. Thereby electrochemical redox reactions
which might contribute to damage the SAM had to be avoided.
The SAM coated gold wafers were electrolyzed in the electrolyte
Monolayer assemblies of 1a,b with a film thickness of about
30 Å are bound to the gold surface via thiolate as confirmed
by XPS.19 For the monolayers of both 1a,b19c and 2a-c contact
angles in the range of 90-96° were determined. From the
position of the asymmetric stretching vibration band (νas(CH2))
at 2917-2918 cm-1, densely packed, well ordered monolayers
of thiols 1a-c and 2a,b with estimated tilt angles of 37-40°
were deduced.19c A nearly perpendicular orientation of the
phenyl ring to the gold surface, diminishing the space require-
ment and improving phenyl-phenyl interactions, was concluded
(17) Thomas, N. E. In Organic Synthesis: The Roles of Boron and Silicon;
Davies, S. G., Ed.; Oxford University Press: Oxford, 1991, p 10.
(18) Ron, H.; Mattis, S.; Rubinstein, J. Langmuir 1998, 14, 1116-1121.
(19) (a) Sellers, H.; Ulman, A.; Shnidman, Y.; Eilers, J. E. J. Am. Chem. Soc.
1993, 115, 9389-9401. (b) Castner, D. G.; Hinds, K.; Grainger, D. W.
Langmuir 1996, 12, 5083-5086. (c) Buckel, F.; Effenberger, F.; Yan, C.;
Go¨lzha¨user, A.; Grunze, M. AdV. Mater. 2000, 12, 901-905.
(20) Fisher, G. L.; Walker, A. V.; Hooper, A. E.; Tighe, T. B.; Bahnek, K. B.;
Skriba, H. T.; Reinard, M. D.; Haynie, B. C.; Opila, R. L.; Winograd, N.;
Allara, D. L. J. Am. Chem. Soc. 2002, 124, 5528-5541.
9
J. AM. CHEM. SOC. VOL. 127, NO. 49, 2005 17317

A R T I C L E S
Maisch et al.
Figure 1. (a) Schematic illustration of the electrochemical preparation and the conductivity measurements of gold-SAM-gold junctions. (b) Schematic
illustration of the measuring principle. Voltage at the sample (Vs), voltage proportional to current (VR), current source (Norton source) generating short
pulses of increasing current.
solution containing tetrachloroauric(III) acid in acetonitrile at
direct current (e0.5 V) for 5 days (Figure 1a). This slow
deposition avoids damage to the SAM, whereas a fast deposition
led to the buildup of Au crystallites resulting in short circuits.
The electrolytic cell, which consists of a carefully designed
Teflon tube and screw, with the gold-SAM-metal junctions
prepared as described, was then carefully rinsed with water and
filled with supporting electrolyte for the measurement of I(V)
characteristics (Figure 2). The gold wafer is contacted by a
platinum wire inside the Teflon screw bottom which touches
its surface only outside the active area to prevent any mechanical
stress or scratches on the measured SAM surface, while the
anode is a gold-plated platinum wire. The I(V) curves for higher
voltages from 0 to 10 V were obtained by applying a Norton
source of high output impedance which programs a constant
current (Figure 1b). The BDV in all cases measured was higher
Figure 2. I(V) characteristics of the gold-SAM 2a-c-gold (electrolyte)
than 15 V which corresponds to an amazingly high electrical
field strength in the order of 50 MV/cm. Short pulses of
increasing current were generated, and both the voltage at the
sample and that proportional to the current flowing through the
system were observed on a digital oscilloscope. At lower
voltages (0 to 1.8 V), where dielectric breakdown is not a major
problem, while low-noise experimental data are required for
the exact evaluation of the conduction mechanism, several (64-
junctions in the range of 0 to 10 V.
256) measurements of the same sample were averaged to
increase the signal-to-noise (Figure 3). These measurements
were carried out by directly contacting the SAM by an
electrolyte without intermediate gold layer.
Discussion of the Influence of the Chemical Structure on
the Insulating Properties and the Electrical Breakdown of
9
17318 J. AM. CHEM. SOC. VOL. 127, NO. 49, 2005

Electrical Insulator Self-Assembled Monolayers on Gold
A R T I C L E S
Figure 3. Averaged I(V) characteristics of the gold-SAM 2a-c-electrolyte
junctions in the range 0 to 1.8 V.
Gold-SAM-Metal Junctions. By applying ω-aryl substituted
SAMs (compounds 1a,b and 2) and electrolytic deposition of
the counter electrode according to method B we succeeded in
the preparation of gold-SAM-metal junctions without any
pinholes, domain formation or impurities in large-area (g30
mm2), and thus resulting in completely reproducible conductivity
measurements through SAMs. Since these gold-SAM-metal
junctions are stable at room temperature for several weeks and
easy to handle, they are best suitable for fundamental investiga-
tions on the mechanism of electrical conduction through thin
organic layers discussed in the next chapter. In this chapter
especially the influence of the chemical structure and the sterical
arrangement of the monolayers on the electrical insulator
behavior is discussed.
In particular Whitesides et al.11 reported comprehensively on
the influence of the chemical structure of SAMs and the
influence of various electrode metals on the BDV. From these
investigations, it was inferred that the thickness of the ordered
hydrocarbon layer is the major factor determining the BDV,
and an exponentially dependence of the current density from
the applied voltage was found. A linear dependence between
thickness and BDV has been found for alkyl monolayers on a
hydrogen terminated Si(111).21 The tilt angle of SAMs versus
the surface normal has also a strong influence. The BDV of
alkanethiol SAMs on gold, with a tilt angle of ∼30°, is lower
by a factor of 2 than that on silver with a tilt angle of ∼0°.11c
The authors note that the value of BDV is surprisingly
indifferent to the detailed structure of the molecules making up
SAMs. There is no evidence that SAMs incorporating simple
aromatic groups differ fundamentally in their electrical behavior
from aliphatic SAMs.11a Because Whitesides et al. have
investigated exclusively Hg-SAM || SAM-metal double
junctions,11 the obtained results cannot be transferred directly
to our results of gold-SAM-metal junctions.
Figure 4. Molecular modeling of the SAM formed from 2c on gold.
rings, elucidated mainly from RAIR spectra. In SAMs of 1a,
the phenyl moiety is nearly perpendicular to the surface
normal,19c whereas in SAMs of 4-eicosylthiophenol, it is tilted.19c
As can be shown by molecular modeling of the lateral SAM
structure of compound 2c on gold, a highly ordered system with
optimal π-stacking (Figure 4) due to the more flexible ether
bridge between the alkyl chain and the aromatic moiety is seen.
This modeling was performed by the Cerius2 software v. 4.2
from MSI using a cff 91 consistent force field and a rigid (1,0,0)
gold surface. The phenoxy-phenoxy distances in both dimen-
sions fit perfectly into the package of the alkyl chains and their
placement on the gold surface.
From the presented structural investigations of SAMs of
compounds 1a,b and 2a-c on gold we conclude that strong
π-interactions of the terminal aromatic rings do not only favor
the perfect order of SAMs but also “densify” their surface,
markedly hindering the penetration of metals. The unusually
high BDV values found for SAMs of compounds 1a,b and 2a-
c, are assumed to arise also from the strong “densification” by
π-stacking, although there is so far no unambiguous explanation
or mechanism for the BDV of organic monolayers.15a
Experimental Evidence for the Tunneling Mechanism of
the Electrical Conduction through Organic Monolayers.
Among the different possible electrical conduction mechanisms
discussed, the tunneling mechanism is most probable for thin
organic films such as self-assembled monolayers. In nearly all
publications dealing with this problem a tunneling mechanism
is postulated for the electrical conductivity based on experi-
mental data.3,4,11,12 However, an unambiguous experimental
evidence for the tunneling mechanism through organic mono-
layers was not yet possible, because up to now large-area
measurements of I(V) characteristics of metal-SAM-metal
junctions could not be performed. The application of an
electrolyte solution for contacting the SAM surface decreases
the risk to measure current flowing through a short path in
imperfections, because small pinhole-like defects have little
influence. This is a result of the very slow diffussion process
in geometrically small structures, whereas these defects give
rise to low-impedance short circuits when metal as electrode is
evaporated on the SAM.9
Recent investigations19c of the influence of aromatic groups
incorporated in long-chain alkanethiol self-assembled mono-
layers on gold have shown that SAMs of both ω-phenyl
eicosanethiol (1a) and 4-eicosylthiophenol consist of densely
packed alkyl chains, independent of the position of the phenyl
ring. Also the tilt angles in both cases are comparable. The
decisive difference, however, is the orientation of the phenyl
(21) Zhao, J.; Uosaki, K.; Appl. Phys. Lett. 2003, 83, 2034-2036.
9
J. AM. CHEM. SOC. VOL. 127, NO. 49, 2005 17319

A R T I C L E S
Maisch et al.
With the assumption of a tunneling mechanism for the
electron transport through an insulator, e.g., thin organic layers,
the tunneling probability and thus the electric current increase
almost exponentially with decreasing thickness of the insulator,20
while showing high dielectric strength.21 Tunneling can occur
through any kind of thin layer and was demonstrated in a heavily
doped germanium diode by Esaki.22c In case of tunneling,
electrons which do not have a potential large enough to
overcome the barrier in a classical model can do so via
tunneling. The tunneling probability for an electron and thus
the current through the SAM depends on the applied voltage,
the thickness of the barrier and its heigth.
The first can be varied with biasing while the properties of
the SAM forming molecules determine the properties of the
barrier. The insulator thickness is determined by the length of
the SAM forming molecules. Under the assumption of a perfect
square well potential with flat top and infinitely steep walls,
the following formula is valid for the tunneling probability T22d
Figure 5. Measurement and comparisation with simulated tunneling through
a square well barrier.
and not along the single molecules. The barrier height is
determined by comparing the experimentally found curves with
the calculated ones: the sum of the residues at discrete
measuring points in the range of 0-1.5 V is minimized. In the
case of SAMs of thiols 2a-c, experimental data and calculation,
however, do not agree as can be demonstrated in Figure 5.
An optimal accordance of experimental results with calcula-
tions can be achieved however, if the potential of each
methylene group in the alkyl chains is additionally modulated
as shown in Figure 6a.23
The Schro¨dinger equation of this model is solved and
optimized to the lowest sum of residues in comparison with
experimental data. In case of a simple square potential, the
calculated curve is obtaind from eq 1, while in case of the
modulated potential (Figure 6), the resulting Schro¨dinger
equations are solved numerically. Table 1 shows the results
together with the modeled molecule lengths (PM3), measured
from the thiol sulfur to the 4′-hydrogen atom of the phenyl rings,
and the relative currents measured at V ) 0.8 V. For the
simulation, not the molecule lenghs l, but the thickness of the
SAMs, d ) l cos(30°), are used. From the obtained values, the
theoretical tunneling curves can be simulated and compared with
the experimental ones. In Figure 6b, the experimental curves
are plotted together with the simulation. It shows considerable
better agreement in the case of the modulated potential than
the simple square well shape (Figure 6a).
1
T )
(1)
2md2(V - E)
V2
1 +
sinh2
4(VE - E2)
p2
x
wherein V is the height of the barrier, m and E mass and energy
of the electrons, respectively, and d the thickness of the layer.
This formula is directly obtained by solving the Schroedinger
equation for the assumed model without any approximations.
In the given case, it can be approximated by an exponential
function.4 Here, (1) was used without further simplification. The
insulator thickness d in self-assembled monolayers is defined
by the chain length and the angle between chain and surface
normal. In most investigations of the electron transport through
organic monolayers this expected exponential dependence of
current on the chain length was found.3,4,11,12,22 The measurement
of I(V) characteristics of alkyl SAMs on silicon, performed by
Vuillaume et al., however, deviate from this finding by showing
a conductivity independent of the chain length, and the authors
therefore exclude a tunneling mechanism.23a
In the preceding chapter, the conductivity through SAMs
depending on the chain length of the SAM forming molecules
and the bias was determined experimentally. On the basis of
the thereby determined I(V) characteristics a tunneling mech-
anism for the electron transport through SAMs will be proved
in the following.
Experimental Section
General Methods. Melting points were determined on a Bu¨chi SMP
20 apparatus and are uncorrected. H NMR spectra were recorded on
The experimentally obtained I(V) characteristics of SAMs of
compounds 2a-c in gold-SAM-gold junctions are compared
with the values expected theoretically for a tunneling mecha-
nism. In the simplest case, a square-well potential is assumed
for the course of the potential in the SAM in the zero-bias case
(Figure 6a). The lengths l of the molecules result from molecular
modeling.22 The thickness of the potential barrier calculated by
multiplying this value by the cosine of the tilt angle of 30° under
the assumption that tunneling occurs through the bulk material
1
a Bruker AC 250 F (250 MHz). Chemical shifts are given with TMS
as an internal standard. Preparative column chromatography was
performed on silica gel S, grain size 0.032-0.063 mm (Riedel-de Haen).
Mass spectrometric data were obtained with a Finnigan MAT 711 or
Finnigan MAT 95 mass spectrometer. Spectroscopic and analytical data
are offered as Supporting Information. I(V) characteristics were
measured using a Tektronix Digital TDS 210 Oscilloscope.
Gold substrates were purchased from Metallhandel Schro¨er GmbH.
On a 1 cm2 glass slide a chromium layer was deposited and
subsequently the gold layer prepared by high vacuum evaporation.
General Procedure for the Preparation of the Alkenylphenyl
Ethers 4a-c. To a solution of phenol (1.30 g, 13.80 mmol) in dry
dichloromethane (10 mL) under N2 atmosphere was added potassium
methanolate (0.90 g, 12.81 mmol). The reaction mixture was stirred at
room temperature for 30 min and concentrated under vacuum. The
remaining potassium phenolate was dissolved in dry DMF (6 mL), and
(22) (a) Mikkelsen, K. V.; Ratner, M. A. Chem. ReV. 1987, 87, 113-153 (b)
York, R. L.; Nguyen, Ph. T.; Slowinski, K. J. Am. Chem. Soc. 2003, 125,
5948-5953. (c) Esaki, L. Phys. ReV. 1958, 109, 603. (d) Schwabl, F.
Quantum Mechanics; Springer: New York 1992, 59-65.
(23) (a) Boulas, C.; Davidovits, J. V.; Rondelez, F.; Vuillaume, D. Phys. ReV.
Lett. 1996, 76, 4797-4800. (b) Vuillaume, D.; Boulas, C.; Collet, J.; Allan,
G.; Delevue, C. Phys. ReV B 1998, 58, 16491-16498. (c) Gu, Y.;
Akhremitcher, B.; Walker, G. C.; Waldeck, D. H. J. Phys. Chem. B 1999,
103, 5220-5226. (d) Selzer, Y.; Salomon, A.; Cahen, D. J. Am. Chem.
Soc. 2002, 124, 2886-2887.
9
17320 J. AM. CHEM. SOC. VOL. 127, NO. 49, 2005

Electrical Insulator Self-Assembled Monolayers on Gold
A R T I C L E S
Figure 6. (a) Square well potential and potential modulated by addition of the dielectrics of the single methylene groups of the alkyl chains. (b) Measurement
and comparisation with simulated tunneling through the barrier of part a.
Table 1. Comparisation of the Results of Tunneling Evaluation
with the Potential Models According to Figure 6a
(9-BBN) (20.0 mmol) in THF (40 mL) was refluxed under N2
atmosphere for 7 d. Then a solution of KOH in H2O2 (33%), prepared
from KOH (2.81 g, 50.0 mmol) and 10 mL H2O2 at 0 °C, was added
at room temperature. The reaction mixture was stirred at 50 °C for 40
min. After being cooled (ice bath), 40 mL water were added followed
by very slow addition of a solution of CoCl2‚6H2O (0.3 g) in 2 mL
water (O2 generation). The reaction mixture was then acidified with
H2SO4 (30%) and a solution of FeSO4‚7H2O (0.5 g) in water (5 mL)
was added. After being stirred for 12 h at room temperature, water
was added, and the reaction mixture extracted with tert-butylmethyl
ether in a perforator. The extract was dried (Na2SO4), concentrated under
vacuum and chromatographed on silica gel with dichloromethane/
petroleum ether (1:3) to give 2.32 g (64%) 5.
SAM
molecule length l by
molecular modeling
(PM3) (nm)
relative
error of simulation
(sum of residues)
case 5
error of simulation
(sum of residues)
case 6
forming
current at
compound
V ) 0.8V
2a
2b
2c
1.63
2.42
2.67
1.00
0.31
0.12
1.68
0.22
0.02
0.340
0.060
0.003
the respective bromide 3a-c (9.86 mmol) was added by syringe. After
being stirred at 65 °C for 3 h and at room temperature for a further 12
h, to the reaction mixture was added dichloromethane and water (60
mL each) to solve precipitated potassium bromide. The phases were
separated and the organic layer was washed five times with water (60
mL), dried (Na2SO4), filtered through a silica gel column (×5 cm) and
eluted with dichloromethane/petroleum ether (1:1) (TLC. control).
Evaporation gave the products 4a-c.
(b) Thioacetate 6c was prepared analogously to a known procedure15a
from 5 (0.18 g, 0.50 mmol); yield: 0.128 g (61%).
General Procedure for the Preparation of Phenoxy-1-alkanethiols
2a-c. To a solution of the respective compounds 6a-c (0.37 mmol)
in diethyl ether/THF (1:1) (6 mL) was added solid LiAlH4 (0.79 mmol,
30 mg) under N2 atmosphere, and the reaction mixture stirred at room
temperature for 1 h. The reaction mixture was first hydrolyzed with 1
mL of degassed water, and then with HCl (25%, 2 mL). Then 10 mL
water and 15 mL dichloromethane were added. The phases were
separated and the organic layer was washed with water (10 mL), and
dried (Na2SO4). Concentration under vacuum gave products 2a-c.
4a: yield 2.16 g (89%) as a colorless liquid; 4b: yield 2.47 g (79%)
as a white solid; mp 34 °C; 4c: yield 3.13 g (92%) as a white solid;
mp 44 °C.
General Procedure for the Preparation of the Phenoxyalkyl
Thioacetates 6a-c. To a solution of the respective alkene 4a-c (0.38
mmol) in chloroforme (4 mL) under N2 atmosphere was added
thioacetic acid (2.67 g, 35.1 mmol) and 4 mg R,R′-azo-isobutyronitrile
(AIBN). After being heated at 60 °C for 24 h, the reaction mixture
was concentrated under vacuum, and the light yellow residue was
chromatographed on silica gel with dichloromethane/petroleum ether
(1:1) to give the products 6a-c.
6a: yield 85 mg (69%) as a yellow oil; 6b: yield 127 mg (85%) as
a white solid; 6c: yield 120 mg (75%) as a white solid; mp 65 °C.
Alternative Preparation of 6c via 18-Phenoxyoctadecan-1-ol (5).
(a) A solution of 4c (3.45 g, 10.0 mmol) and 9-borabicyclononane
2a: yield 70 mg (61%) as a yellow oil; 2b: yield 93 mg (72%) as
light yellow solid; mp 55 °C; 2c: yield 98 mg (70%) as light yellow
solid; mp 60 °C.
Purification and Surface Treatment of Gold Substrates. Com-
mercial gold substrates (12 × 12 × 1.1 mm borosilicate glass with
chromium layer of 1-4 nm thickness as adhesion promoter and
vaporcoated gold layer of 200-300 nm thickness) were cleaned with
“piranha” acid (33% H2O2/concentrated H2SO4 3:7) for 15 min at 65
9
J. AM. CHEM. SOC. VOL. 127, NO. 49, 2005 17321

A R T I C L E S
Maisch et al.
Conclusion
Terminally aryl substituted long chain alkanethiols form
highly ordered monolayers (SAMs) on gold surfaces. Mainly
from RAIR-investigations and supported by molecular modeling
it can be shown that the lateral SAM structure of the ω-phenoxy
alkanethiols 2 on gold is perfectly ordered with optimal
π-stacking of the aromatic rings and phenoxy-phenoxy distances
in both dimensions which fit completely into the package of
the alkyl chains and their placement on the gold surface. The
strong π-interactions of the terminal aromatic rings do not only
favor the perfect order of the SAMs but also “densify” their
surface, markedly hindering the penetration of metals. Since
phenoxy terminated SAMs were found to be perfect insulators,
they can be applied for construction of stable and easy to handle
metal-insulator-metal (MIM) junctions. The I(V) characteristics
of gold-SAM-gold junctions have been measured depending
on the chain length of compounds 2. From the measured I(V)
characteristics and simulation with theoretical tunneling models,
a tunneling mechanism of the electrical conduction through thin
organic layers is unequivocally confirmed experimentally.
The outstanding insulator properties of SAMs with terminal
aryl substituents are not limited to gold as metal surface.
Comparable results were obtained also with silicon surfaces by
applying ω-phenoxy alkyl trichlorosilanes as SAM forming
compounds.16c,24 It was possible for example to prepare organic
thin film transistors (TFTs) with the corresponding SAMs as
gate dielectric and a high mobility organic semiconductor
(pentacene).25 These TFTs operate with supply voltages of less
than 2 V, yet have gate currents that are lower than that of
advanced silicon field-effect transistors with SiO2 dielectrics.25
Figure 7. FT-IR reflexion spectrum of the SAM formed from compound
2c on gold.
°C. The gold substrates were then tempered at 650 K for 24 h on a
heating plate. To reduce the surface roughness, the substrates were
treated in an electrolyte from tetrachloroauric(III) acid (0.08 g),
NBu4BF4 (9.5 g), Mg(NO3)2 (1.2 g) in degassed acetonitrile (1 L) and
connected as anode versus platinum at sine alternating voltage with
interfered direct voltage for 72 h.
Monolayer Preparation. The cleaned gold substrates were immersed
in a stirred solution of the respective thiol in absolute ethanol, degassed
by three freeze-pump-thaw cycles (1 mM for 1a,b, 0.5 mM for 2a-
c) under inert gas atmosphere for 24 h (1a,b) or 48 h (2a-c). Samples
were rinsed with absolute dichloromethane and blown dry with nitrogen.
Spectroscopic Characterization. Figure 7 shows an FT-IR-spectrum
of the SAM formed from compound 2c.
Both the symmetric and asymmetric CH2 valence vibrations at ν˜s )
2850 cm-1 and ν˜as ) 2916 cm-1 are visible. They correspond to the
values of the same compound in crystalline state, rather than in liquid.
In the range from 1496-1604 cm-1 aromatic ring vibrations can be
seen. The spectrum was taken by averaging 2000 scans of the same
sample in an FT-IR spectrometer.
Cathodic Metal Deposition. The coated gold substrates in an
electrolyte from tetrachloroauric(III) acid (0.9 g), Mg(NO3)2 (0.5 g),
NBu4BF4 (7.5 g) or tetrachloroplatinic(IV) acid (0.45 g), KCN (5.0 g),
NBu4BF4 (3.5 g) in degassed acetonitrile (1 L each) were electrolyzed
at direct current (20 nA) for 5 d. The cell was carefully rinsed twice
with water and filled with a solution of NBu4BF4 (12.0 g) and Mg-
(NO3)2 (1.2 g) in degassed acetonitrile (1 L) for I(V) characteristic
measurement.
Supporting Information Available: Spectroscopic and ana-
lytical data. This material is available free of charge via the
Internet at http://pubs.acs.org.
JA0548992
(24) (a) Schu¨tz, M. Dissertation, Universita¨t Stuttgart 2002. (b) Rittner, M.
Dissertation, Universita¨t Stuttgart 2002.
(25) Halik, M.; Klauk, H.; Zschieschang, U.; Schmid, G.; Dehm, Ch.; Schu¨tz,
M.; Maisch, S.; Effenberger, F.; Brunnbauer, M.; Stellacci, F. Nature 2004,
431, 963-966.
9
17322 J. AM. CHEM. SOC. VOL. 127, NO. 49, 2005