Spontaneous solution-phase redox deposition of a dense cobalt(II) phthalocyanine monolayer on gold

Article abstract of DOI:10.1021/jp047752q

A dense monolayer of cobalt(II) phthalocyanine, CoPc, can be formed on a gold surface by spontaneous redox deposition followed by vacuum annealing at about 110?°C. CsCoPc(CN)₂ or KCoPc(CN)₂ in dilute (a??10^-6 M) ethanol solution rapidly forms a dense adlayer that, when washed with ethanol, can be imaged by scanning tunneling microscopy, STM, in air. This adlayer is converted to a monolayer of CoPc by vacuum annealing as confirmed by STM and X-ray photoelectron spectroscopy (XPS). This spontaneous surface redox adsorption process represents a novel method for depositing metallorganic complexes, which are ultimately only physisorbed on gold.

Full text of DOI:10.1021/jp047752q

J. Phys. Chem. B 2004, 108, 17003-17006
17003
Spontaneous Solution-Phase Redox Deposition of a Dense Cobalt(II) Phthalocyanine
Monolayer on Gold
Ursula Mazur, Maya Leonetti, William A. English, and K. W. Hipps*
Department of Chemistry and Materials Science Program, Washington State UniVersity,
Pullman, Washington 99164-4630
ReceiVed: May 25, 2004; In Final Form: June 4, 2004
A dense monolayer of cobalt(II) phthalocyanine, CoPc, can be formed on a gold surface by spontaneous
redox deposition followed by vacuum annealing at about 110 °C. CsCoPc(CN)2 or KCoPc(CN)2 in dilute
(∼10-6 M) ethanol solution rapidly forms a dense adlayer that, when washed with ethanol, can be imaged by
scanning tunneling microscopy, STM, in air. This adlayer is converted to a monolayer of CoPc by vacuum
annealing as confirmed by STM and X-ray photoelectron spectroscopy (XPS). This spontaneous surface redox
adsorption process represents a novel method for depositing metallorganic complexes, which are ultimately
only physisorbed on gold.
Introduction
studied the extensive hydrogen bonding networks formed by
trimesic acid on graphite.8
Phthalocyanine (Pc) and porphyrin-based (P) electrical con-
ductors are used extensively as building blocks for constructing
stable 2D supermolecular electronic assemblies. The architec-
tural forms of these structures include simple clusters, wires,
arrays, and even designer surfaces.1-7 Understanding the
principles of architecture and the ability to control Pc assembly
on surfaces continues to be of critical importance in the rational
design and construction of functional optoelectronic nanoscale
devices derived from phthalocyanines and porphyrins. The most
basic monolayer architectures arise from vapor deposition.
Simple complexes deposited on conductive surfaces in vacuum
or from solution exhibit a close-packed arrangement that results
from van der Waal’s interactions between the molecules.2-4
Hipps and co-workers vapor deposited different 2D metal
phthalocyanine and porphyrin assemblies on gold and imaged
them by UHV scanning tunneling microscopy (STM) with
remarkable submolecular resolution.2,3,6 Bai demonstrated that
monolayer molecular arrays of alkane-derivitized phthalocyanine
can be assembled from solution on straight-chain alkane
templates deposited on graphite.4 Recently, Itaya formed both
a pure MPc and a mixed CoPc and copper tetraphenylporphyrin
(CuTPP) adlayer by immersing a Au single crystal into a
benzene solution containing the parent compounds. By imaging
the adsorbate system in aqueous solution, he demonstrated a
structural dependence of the adlayer on the crystallographic
orientation of the gold surface.5
The 2D self-assembly of adsorbed Pc and P molecules also
can be directed by selectively controlling noncovalent intermo-
lecular interactions such as van der Waals attraction and
hydrogen bonding.6-8 The best-known examples of hydrogen
bonding-driven surface assembly include data reported by Hipps,
Yokoyama, and Griessl. Hipps and co-workers found the
formation of a new well-ordered 2D structure with 1:1 stoichi-
ometry of F16CoPc and NiTPP on gold by vapor-phase deposi-
tion in UHV.6 Yokoyama observed small clusters and chains
of CN-substituted porphyrins vacuum deposited on Au(111) in
a low-temperature STM environment.7 Griessl and co-workers
All of the above methods for monolayer formation share a
common failingsthey produce monolayers that are not substrate
location-selective. Because the final layer is physisorbed rather
than chemisorbed, almost any surface the adsorbate hits will
support it (snow on a field). Given a complex device structure
with exposed surfaces of various metals and semiconductors,
all of the surfaces will be coated.
A pathway to the 2D architecture of physisorbed complex
molecules that has not been explored is the formation of self-
organized molecular monolayers by spontaneous redox processes
at the substrate surface. In this report, we present a generalized
approach where a soluble species having a redox reaction
induced by a metal electrode spontaneously decomposes to a
well ordered physisorbed layer of insoluble product upon
electrical contact with that metal. The chemistry described here
is very different from that of the conventional thiol or thiolate
precursor SAM films. In those systems, the final monolayer is
chemisorbed to the substrate surface.9 The procedure presented
here does not require complex vacuum technology. Rather, it
relies on solubility partitioning between solution and adsorbed
phases, and it can create very reproducible well-ordered
submonolayers, monolayers, and multilayers. We note that we
previously observed the conversion of soluble FePcCl to
insoluble FePc by spontaneous reduction on an aluminum
surface.10 In that work, however, we did not determine the
surface structure of the adsorbed layer.
Here we demonstrate the formation of a stable 2D assembly
of CoPc from solution via spontaneous redox reactions between
a biaxially substituted dicyano cobalt phthalocyanine salt,
MCoPc(CN)2 (M ) K, Cs), and a gold substrate. We selected
-1
the CoPc(CN)2 ion (shown in Figure 1) because of its
chemical stability and solubility in simple organic solvents. Its
X-ray structure has been reported.11
The process by which the CoPc adlayer is formed on the
gold electrode can be elucidated in part from the cyclic
voltammetry (CV) studies of MCoPc(CN)2 reported by Hanack
and co-workers.12 On the basis of their voltammetric data,
Hanack proposed that in acetone MCoPc(CN)2 is oxidized to a
* Corresponding author. E-mail: hipps@wsu.edu.
10.1021/jp047752q CCC: $27.50 © 2004 American Chemical Society
Published on Web 07/13/2004

17004 J. Phys. Chem. B, Vol. 108, No. 44, 2004
Letters
Scanning tunneling microscopy (STM) was performed both
in air and in UHV. The ambient STM images were collected
using a Digital Instruments stand alone head and a Nanoscope
E controller, and the UHV STM data were obtained with an
RHK Technology15 microscope (model UHV300) and control
electronics (model SPM100). Images were plane fit and low-
pass filtered. STM tips were prepared from 0.25-mm W by
electrochemical etching or from Pt0.8Ir0.2 wire by mechanical
cutting. W tips always required additional cleaning once in
UHV. XPS analysis was performed on a Kratos Axis-165
electron spectrometer having a base pressure of 5 × 10-10 Torr.
Monochromatic radiation from an Al KR X-ray source was used
for excitation. Samples could be heated and cooled in place in
the XPS system.
Figure 1. Molecular structure of the CoPc(CN)2- ion, which measures
approximately 1.4 nm across and 0.61 nm along the NC-Co-CN
fragment.
Results and Discussion
neutral radical, CoPc(CN)2. The oxidization potential (E1/2) of
MCoPc(CN)2 in acetone and 0.1 M Bu4NClO4 was measured
at 0.88 V (sce). Applying our orbital-mediated tunneling (OMT)
model to this process, we predict that an E1/2(oxd) of 0.88 V (sce)
is equivalent to 5.29 V relative to the vacuum state.10 That value
is very close to the measured work function for Au(111), 5.3
Solution-grown specimens were first examined in air by STM
and then transferred into the UHV chamber housing either the
RHK STM or the XPS system. Parts a and b of Figure 2 show
typical large-scale STM images of Pc adlayers formed by
exposing Au(111) to KCoPc(CN)2 solution. Samples were
washed with ethanol and dried prior to measurement. Similar
results were obtained from CsCoPc(CN)2 solution. Image 2a
was acquired in air. Micrographs such as 2a were obtained with
a high level of reproducibility, and samples were stable for at
least 1 day when stored in a desiccator. Images b and c of Figure
2 were obtained in an UHV environment after annealing the
solution-grown sample at 110 °C for a few minutes.
Large-scale images show well-ordered 2D assemblies with
distinct molecules. Measurements made in air also show random
groupings of molecules well separated from islands and high
spots associated either with axially substituted phthalocyanines
or adsorbates on the Pc layer. These isolated molecules and
small clusters are not usually observed at room temperature in
CoPc films vapor deposited on Au because the CoPc-Au
interaction is too weak and there is rapid diffusion of molecules
between islands. We were unable to obtain submolecular
resolution images in air. This may be due to the mixed
composition of the initially formed layer (vide infra) or to a
CoPc-O2 adduct that may form under these conditions.16
It is useful to note that the STM image obtained in air from
the as-deposited solution is less densely packed than that of
the annealed monolayer. Figure 2a shows rows with an internal
spacing of about 1.2 nm but a row-row spacing of about 1.7
nm. In contrast, the interatomic spacing in parts b and c of Figure
2 is 1.2 nm in any direction. After annealing, the structure is
defined by a much denser square unit cell. The origin of this
structural change is still unclear but is probably related to the
-1
V. Thus, we expect CoPc(CN)2 in contact with the Au(111)
to oxidize spontaneously to the radical. We have found that the
CoPc(CN)2 radical then follows a complex redox pathway with
the eventual formation of neutral and insoluble CoPc derivative
species adsorbed on the gold surface. Upon low temperature
(110 °C) annealing in vacuum, this monolayer converts com-
pletely to cobalt(II) phthalocyanine. We note here that the
isolated MCoPc(CN)2 salt is stable to 300 °C before the first
cyanide is cleaved from the cobalt ion.13
Experimental Section
The ionic complex, MCoPc(CN)2 (M ) K or Cs), was
prepared and purified by published methods.13 Elemental
1
analysis, H NMR, and infrared spectroscopy confirmed the
purity. Well-organized phthalocyanine monolayers on Au(111)
were formed by exposing freshly annealed Au(111) substrates
to a 10-6 M ethanolic solution of the MCoPc(CN)2 complex
for 10 min, washing with a small amount of ethanol, and drying
under a gentle flow of inert gas for 15 min. A high-vacuum
multipurpose deposition chamber was used for the preparation
of CoPc submonolayers on Au(111)/mica samples by vapor
deposition. CoPc was purchased from Strem Chemicals14 and
twice sublimed before it was deposited from a tantalum box
source. Both solution- and vacuum-deposited samples were
transported in air to the load lock of the XPS where they were
analyzed under UHV.
Figure 2. STM micrographs of Au(111) exposed to 10-6 M ethanolic KCoPc(CN)2 for 10 min and washed with ethanol: (a) 50 × 50 nm2 image
acquired in air with 0.8-V bias voltage and 180-pA tunneling current, (b) 50 × 50 nm2 image, and (c) high-resolution 10 × 10 nm2 image. Both
b and c were obtained under UHV conditions after annealing at 110 °C. Sample bias and tunneling current were +0.32 and +0.75 V and 100 pA,
respectively, for images b and c. A PtIr tip was used for all three images.

Letters
J. Phys. Chem. B, Vol. 108, No. 44, 2004 17005
Figure 3. Correlation-averaged 3D image of a single CoPc molecule
on Au formed from redox adsorption from solution.
mixed composition (vide infra) of the as-formed monolayer.
Alternatively, the coadsorption of solvent may also influence
the as-grown structure.
Details of the internal structure, orientation, and packing of
the adsorbed complex are seen in the high-resolution image
(Figure 2c) obtained in UHV after sample annealing. Individual
phthalocyanine molecules can be recognized by their clover-
leaf shapes with a central peak and four additional spots at the
corners. To better define the single-molecule constant current
contour, we performed a correlation average on the data in
Figure 2c, and a highly accurate image results (Figure 3). In
this low-noise, high-resolution image, one may observe that the
benzene rings of the Pc ring are more effective conductors than
the five-membered rings. The large characteristic increase in
the tunneling probability at the center of the molecule has been
Figure 4. XPS of the Co 2p region of phthalocyanine complexes on
Au(111).
Our proposed mechanism begins with the spontaneous
oxidation of MCoPc(CN)2 by gold, driven essentially by its
rather large work function. To test this, we have performed XPS
studies of the solution-phase adsorption on both gold and
aluminum. (Al has a work function about 1 eV less than that of
Au.) We find that, as expected, the initial reaction occurs rapidly
with gold metal, but there is no apparent reaction with
aluminum. Thus, one may envisage the selective deposition of
CoPc upon gold contacts in the presence of a number of other
materials (including copper, aluminum, and silicon) that have
significantly smaller work functions than does gold. In fact, by
tailoring the redox potential of the initial reaction and the relative
solubility of the initial and final species, one could tailor the
selective deposition of materials where the final product is only
physisorbed to the surface. For example, a species that was
reduced at about -0.4 V (sce) and became insoluble would be
selectively deposited on aluminum but not on gold or silicon.
In contrast, either vapor- or solution-phase deposition of the
desired material would coat the entire surface.
2
attributed to the half-filled dz orbital acting as an atomic wire.
The apparent height at the molecular center is lower than
expected for the case of cyano substitution, but this is NOT a
definitive test. Note that although alkali ions can be observed
in STM17 no features attributable to either potassium or cesium
ions have been seen in our STM studies.
XPS analysis of both the unannealed and annealed CsCoPc-
(CN)2 ethanolic solution samples (after washing) shows that
there is less than 5% of the expected Cs/Co ratio. That is, there
was no detectable Cs in the XPS of the adsorbed film. Thus,
the surface species is clearly a neutral complex. The Co 2p
region of the XPS is also very informative. Initially (lower curve
of Figure 4) there are both Co2+ and Co3+ species present on
the surface, with peaks near 778.2 and 780.2 eV assigned to
the 2p3/2 transition in Co2+ and those near 779.7 and 781.1 eV
being due to Co3+. With annealing at 100 °C for 10 min, most
of the surface layer has been converted to Co2+. After annealing
at 110 °C, the Co 2p XPS becomes identical to that of authentic
CoPc (top curve of Figure 4).
Our interpretation of these results is that the initial monolayer
is a mixture of CoPc and a monoaxially substituted cobalt(III)
phthalocyanine. Upon annealing, the axial ligand is removed
with the donation of one electron to form the final CoPc
monolayer. This model is supported by the fact that the N 1s
peak in the XPS shifts by only 0.2 eV throughout the entire
conversion process. Thus, the Pc ring is most likely in the -2
state throughout. That the final (annealed) species is unsubsti-
tuted CoPc is born out by the following: (1) the XPS spectral
positions and the C/N ratio, which is identical to that of authentic
CoPc, (2) the fact that the species is neutral (no alkali metal
ions present), and (3) the STM images of both the individual
molecules and of the overall lattice are identical to those of
vapor-deposited CoPc on Au(111).
Acknowledgment. We thank the National Science founda-
tion for support in the form of grants CHE 0138409 and CHE
0234726. Acknowledgment also is made to the donors of the
Petroleum Research Fund, administered by the American
Chemical Society. We also thank Dr. Louis Scudiero for his
assistance in acquiring the XPS data.
References and Notes
(1) Hecht, S. Angew. Chem., Int. Ed. 2003, 42, 24.
(2) Scudiero, L.; Barlow, D. E.; Hipps, K. W. J. Phys. Chem. B 2000,
104, 11899. (b) Scudiero, L.; Barlow, Dan E.; Hipps, K. W. J. Phys. Chem.
B 2002, 106, 996. (c) Hipps, K. W.; Barlow, Dan E.; Mazur, U. J. Phys.
Chem. B 2000, 104, 2444. (d) Hipps, K. W.; Barlow, Dan E.; Mazur, U. J.
Phys. Chem. B 2000, 104, 5993.
(3) Scudiero, L.; Barlow, D. E.; Mazur, U.; Hipps, K. W. J. Am. Chem.
Soc. 2001, 123, 4073.
(4) Lei, S.; Wang, J.; Dong, Y.; Wang, C.; Wan, L.; Bai, C. Surf.
Interface Anal. 2002, 34, 767. (b) Lei, S.; Wang, C.; Yin, S.; Bai, C. J.
Phys. Chem. B 2001, 105, 12272. (c) Lei, S.; Wang, J.; Dong, Y.; Wang,
C.; Wan, L.; Bai, C. Surf. Interface Anal. 2002, 34, 767.
(5) Suto, K.; Yoshimoto, S.; Itaya, K. J. Am. Chem. Soc. 2003, 125,
14976. Yoshimoto, S.; Tada, A.; Suto, K.; Itaya, K. J. Phys. Chem. B 2003,
107, 5836-5843.
(6) Hipps, K. W.; Scudiero, L.; Barlow, D. E.; Cooke, M. P., Jr. J.
Am. Chem. Soc. 2002, 124, 2126. (b) Scudiero, L.; Hipps, K. W.; Barlow,
D. E. J. Phys. Chem. B 2003, 107, 2903.

17006 J. Phys. Chem. B, Vol. 108, No. 44, 2004
Letters
(7) Yokoyama T.; Yokoyama, S.; Kamikado, T.; Okuno, Y.; Mashiko,
S. Nature 2001, 413, 619.
(8) Griessl, S.; Lackinger, M.; Edelwirth, M.; Hietschold, M.; Heckl,
W. M. Single Mol. 2002, 3, 25.
(9) An Introduction to Ultrathin Organic Films; Ulman, A., Ed.;
Academic Press: San Diego, CA, 1991.
(10) Mazur, U.; Hipps, K. W. J. Phys. Chem. B 1999, 103, 9721.
(11) Inabe, T.; Maruyama, Y. Bull. Chem. Soc. Jpn. 1990, 63, 2277.
(12) Leverenz, A.; Speiser, B.; Hanack, M. J. Electroanal. Chem. 1992,
232, 273. (b) Bchnisch, B.; Leverentz, K.; Hanack, M. Synth. Met. 1991,
41-43, 267.
(13) Metz J.; Hanack, M. J. Am. Chem Soc. 1983, 105, 828. (b) Orihashi,
Y.; Nishikawa, M.; Ohno, H.; Tsuchida, E.; Matsuda, H.; Nkanishi, H.;
Kato, M. Bull. Chem. Soc. Jpn. 1987, 60, 3731.
(14) Strem Chemicals, Inc. Newburyport, MA.
(15) RHK Technology, Inc. Troy, MI.
(16) Yoshimoto, S.; Tada, A.; Suto, K.; Narita, R.; Itaya, K. Langmuir
2003, 19, 672.
(17) Kim, Y. G.; Yau, S. L. Itaya, K. J. Am. Chem. Soc. 1996, 118,
393. (b) Stuhlmann, C.; Villegas, I.; Weaver, M. J. Chem. Phys. Lett. 1994,
219, 319.