Platinum nanofilm formation by EC-ALE via redox replacement of UPD copper: Studies using in-situ scanning tunneling microscopy

Article abstract of DOI:10.1021/jp063766f

The growth of Pt nanofilms on well-defined Au(111) electrode surfaces, using electrochemical atomic layer epitaxy (EC-ALE), is described here. EC-ALE is a deposition method based on surface-limited reactions. This report describes the first use of surface-limited redox replacement reactions (SLR³) in an EC-ALE cycle to form atomically ordered metal nanofilms. The SLR ³ consisted of the underpotential deposition (UPD) of a copper atomic layer, subsequently replaced by Pt at open circuit, in a Pt cation solution. This SLR³ was then used a cycle, repeated to grow thicker Pt films. Deposits were studied using a combination of electrochemistry (EC), in-situ scanning tunneling microscopy (STM) using an electrochemical flow cell, and ultrahigh vacuum (UHV) surface studies combined with electrochemistry (UHV-EC). A single redox replacement of upd Cu from a PtCl₄^2- solution yielded an incomplete monolayer, though no preferential deposition was observed at step edges. Use of an iodine adlayer, as a surfactant, facilitated the growth of uniformed films. In-situ STM images revealed ordered Au(111)-(√3 × √3)R30°-iodine structure, with areas partially distorted by Pt nanoislands. After the second application, an ordered Moire pattern was observed with a spacing consistent with the lattice mismatch between a Pt monolayer and the Au(111) substrate. After application of three or more cycles, a new adlattice, a (3 × 3)-iodine structure, was observed, previously observed for I atoms adsorbed on Pt(111). In addition, five atom adsorbed Pt-I complexes randomly decorated the surface and showed some mobility. These pinwheels, planar PtI₄, complexes, and the ordered (3 × 3)-iodine layer all appeared stable during rinsing with blank solution, free of I^- and the Pt complex (PtCl₄2^-).

Full text of DOI:10.1021/jp063766f

17998
J. Phys. Chem. B 2006, 110, 17998-18006
Platinum Nanofilm Formation by EC-ALE via Redox Replacement of UPD Copper:
Studies Using in-Situ Scanning Tunneling Microscopy
Youn-Geun Kim, Jay Y. Kim, Deepa Vairavapandian, and John L. Stickney*
Department of Chemistry, UniVersity of Georgia, Athens, Georgia 30602
ReceiVed: June 16, 2006; In Final Form: July 24, 2006
The growth of Pt nanofilms on well-defined Au(111) electrode surfaces, using electrochemical atomic layer
epitaxy (EC-ALE), is described here. EC-ALE is a deposition method based on surface-limited reactions.
This report describes the first use of surface-limited redox replacement reactions (SLR3) in an EC-ALE cycle
to form atomically ordered metal nanofilms. The SLR3 consisted of the underpotential deposition (UPD) of
a copper atomic layer, subsequently replaced by Pt at open circuit, in a Pt cation solution. This SLR3 was
then used a cycle, repeated to grow thicker Pt films. Deposits were studied using a combination of
electrochemistry (EC), in-situ scanning tunneling microscopy (STM) using an electrochemical flow cell, and
ultrahigh vacuum (UHV) surface studies combined with electrochemistry (UHV-EC). A single redox
2-
replacement of upd Cu from a PtCl4 solution yielded an incomplete monolayer, though no preferential
deposition was observed at step edges. Use of an iodine adlayer, as a surfactant, facilitated the growth of
uniformed films. In-situ STM images revealed ordered Au(111)-(x3 × x3)R30°-iodine structure, with areas
partially distorted by Pt nanoislands. After the second application, an ordered Moire´ pattern was observed
with a spacing consistent with the lattice mismatch between a Pt monolayer and the Au(111) substrate. After
application of three or more cycles, a new adlattice, a (3 × 3)-iodine structure, was observed, previously
observed for I atoms adsorbed on Pt(111). In addition, five atom adsorbed Pt-I complexes randomly decorated
the surface and showed some mobility. These pinwheels, planar PtI4 complexes, and the ordered (3 × 3)-
iodine layer all appeared stable during rinsing with blank solution, free of I- and the Pt complex (PtCl42-).
Introduction
all known to form uniform epitaxial atomic layers. UPD,
however, is by definition limited to an atomic layer. Au and Pt
cannot be deposited at UPD, due to their slow electrodeposition
kinetics. Atomic level studies of the epitaxial growth of
successive atomic layers of noble metals are rare. Attempts to
grow atomic layers tend to result in formation of clusters, rather
than epitaxial deposits, the result of their high cohesive energy.16
A number of researchers have used well-defined gold
substrates to study the electrodeposition of Pt, which generally
produced platinum particles. Scanning tunneling microscopy
(STM) has been used extensively to characterize the size and
distribution of these nanoclusters.17-19 Although it is not an ideal
layer by layer process, Uosaki et al. have achieved quasi-two-
dimensional layer by layer growth of Pt atoms on Au(111), by
the electrochemical reduction of PtCl62- complexs.17 Recently,
Waibel et al. performed a similar study and showed that ordered
Understanding and control in the electrochemical deposition
of atomic and molecular layers is of fundamental importance
to the electrochemical growth of nanofilms and of great
technological significance. Fields such as nanostructure syn-
thesis, electrocatalysis, magnetic materials, and the formation
of ULSI (ultra large scale integrated circuits), all stand to benefit
if electrochemical methods can be applied with atomic and
molecular level control.
It is of particularly interest to study the atomic layer growth
of nanofilms of noble metals by electrochemical deposition,
given their extreme stability, conductivity, and importance in
catalysis. Electrochemical methodologies are rightly seen as
simple, versatile, and relatively low cost. If atomic level control
can be added to the benefits of electrodeposition, direct
competition with vacuum and gas-phase processes, such as
molecular beam epitaxy (MBE) and chemical vapor deposition
(CVD), will result.
In electrodeposition heteroepitaxial growth of an atomic layer
of one metal on a second has been extensively performed and
is generally referred to as underpotential deposition (UPD).1
More recently, detailed studies of the structures formed during
UPD on single-crystal surfaces have been performed with
increasing frequency.2-6 UPD is a phenomenon where an atomic
layer of one metal deposits on a second, at a potential prior to
that needed to electrodeposit the element on itself, the result of
the free energy of formation of a surface compound or alloy.
Examples of UPD include the deposition of less-noble metals
such as copper,7-10 silver,11-13 or lead14,15 onto gold electrodes,
2-
adlayers of the PtCl4 complex can be formed and that
nucleation of Pt started predominantly at defects, such as step
edges. At low deposition rates, three-dimensional clusters were
formed.18 Nagahara et al. examined Pt nanoparticles produced
from solutions of haloplatinate complexes, PtX42- (X ) Cl, Br,
I), on Au(111). STM images showed well-ordered pinwheel
features for the PtCl42- and PtBr42- solutions, though not with
the PtI42 complex. They were able to reduce these Pt complexes,
forming particles with diameters of 3.0 and 0.46 nm in height.19
The electrodeposition of metal thin films, in general, is an
important commercial process. On the other hand, it is well-
known that depositions of Pt and Au frequently result in less
than optimal deposits, due to the irreversibility of the processes.
The studies described here involve the development of a layer
by layer deposition methodology, via electrochemical atomic
layer epitaxy (EC-ALE), where deposits are formed an atomic
* To whom correspondence should be addressed. E-mail: stickney@
chem.uga.edu.
10.1021/jp063766f CCC: $33.50 © 2006 American Chemical Society
Published on Web 08/22/2006

Platinum Nanofilm Formation by EC-ALE
J. Phys. Chem. B, Vol. 110, No. 36, 2006 17999
layer at a time using surface-limited reactions. EC-ALE is the
electrochemical analogue of atomic layer epitaxy (ALE) and
atomic layer deposition (ALD),20-22 all methods based on the
growth of materials an atomic layer at a time using surface
limited reactions. This article describes the first instance of
atomic layer characterization of the epitaxial growth of multiple
atomic layers of a metal, facilitated by the use of EC-ALE.
EC-ALE was developed by combining the principles of ALE,
specifically, the growth of materials an atomic layer at a time,
with UPD, for the formation of compound semiconductors.23-25
Compounds formed include many of the II-VIs, including Zn-,
Cd-, and Hg-based chalcogenides, as well as some III-Vs,
V-VIs, and IV-VI compounds. In the growth of compounds,
the UPD of the elements making up a compound can be
alternated to grow compound nanofilms. As noted above, UPD
of a metal on itself cannot be performed, by definition, so the
strict application of UPD in a simple cycle will not result in
the growth of nanofilms of pure metal.
Figure 1. Schematics of an EC-ALE cycle for the formation of Pt
nanofilms on iodine-modified Au(111) electrode.
Recently, Adzic and Brankovic et al. illustrated that a pre-
deposited, surface-limited, Cu UPD adlayer can be used as a
sacrificial layer, to control the growth of a submonolayer amount
of Pt on Au(111), by exchanging the Cu in a Pt ion solution at
open circuit (OC). Their results suggested the growth of very
small, monatomically high, nanoclusters randomly distributed
over the surface.26 Such reactions are referred to here as surface-
limited redox replacement reactions (SLR3). Despite the random
distribution of these nanoparticles, the results of Brankovic et
al. suggested it may be possible to form ideal single-crystalline
nanofilms of Pt by repeating this SLR3 in a layer by layer
methodology: EC-ALE. In fact, Weaver et al. attempted to grow
Pt nanofilms by repeated application of the SLR3.27
variation of the Clavilier method.32,33 The flame-annealed bead
electrode was quenched in Milli-Q Plus (Millipore, Houston,
TX) water of a resistivity >18.2 MΩ, previously deaerated with
hydrogen gas. This bead was then quickly transferred through
air, with a protective water film, into the EC-STM cell.
The EC-STM flow cell used for in-situ STM studies allowed
flow and exchange of solutions over the Au bead working elec-
trode. The reference/auxiliary compartment was downstream,
to avoid contamination. Low flow rates in and out of the cell
were equivalent, to maintain the solution level in the cell, using
a single peristaltic pump, flow rate ) 0.9 mL/min, as described
previously.29 In both the EC-STM and voltammetric measure-
ments, the supporting electrolyte consisted of 0.1 M HClO4 or
0.05 M H2SO4 prepared from doubly distilled HClO4 and H2-
SO4 (Aldrich Chemicals, Milwaukee, WI) and Milli-Q Plus
water. Ultrapure-grade K2PtCl4, H2PtCl6, and CuSO4 (Aldrich
Chemicals, Milwaukee, WI) were used. The three-electrode
electrochemical cell included an Au-wire auxiliary electrode and
a 3 M Ag/AgCl reference electrode (BAS); unless otherwise
specified, all potentials are reported with respect to this reference
electrode. EC-STM experiments were performed with a Nano-
scope III (Digital Instruments, Santa Barbara, CA), equipped
with W tips, electrochemically etched (15 VAC in 1 M KOH)
from a 0.25 mm wire. The tip was coated with transparent nail
polish to minimize Faradaic currents. Prior to each experiment,
the presence of a well-ordered Au(111)-(1 × 1) structure in
pure supporting electrolyte was verified by wide-angle and atom-
resolved EC-STM scans, prior to injection of 1 mM Cu2+ or
This article describes studies of the atomic layer growth of
Pt nanofilms, using this SLR3 in an EC-ALE cycle. That is, Cu
2-
UPD layers were replaced by Pt, using a PtCl4 solution at
open circuit, as outlined by Brankovic et al.26 The spontaneous
process was driven by the difference in formal potentials
between Pt metal and UPD Cu, the difference in free energies.
The standard potential for Pt (PtCl42-/Pt), 0.52 V (vs Ag/
AgCl),28 is well positive of the formal potential for Cu2+/Cuupd
,
∼0.2-0.25 V, on gold.
The depositions were followed using an EC-STM flow cell
on Au(111). The electrochemical flow cell for the in-situ STM
studies was developed by this group and allowed solution
exchange while maintaining potential control and while imaging.
This cell was developed for studied of the electrodeposition of
compound semiconductor nanofilms, an atomic layer at a time
using EC-ALE.29
2-
0.1 mM PtCl4 solutions into the EC-STM flow cell.
This paper describes the deposition process in the presence
of an adsorbed layer of I atoms, used as a surfactant. I atoms
are known to protect some metal surface, such as Pt, from
contamination. Adsorbed halides in general are known to
facilitate surface diffusion in electrolyte solutions and thus
promote “electrochemical annealing.”30 In addition, the cor-
rugations of halide atomic layers are strong in STM studies,
making imaging of such surface somewhat easier.31 Finally, it
is well-known that electrodeposition of Cu and Ag on Pt or Au
surfaces coated with an I atom layer occurs, with the depositing
metal inserting between the I atom layer and the metal substrate,
suggesting that the I atoms float on the depositing metal
surface.2,3 Figure 1 is a cartoon of a SLR3 starting with an I
atom layer for the formation of ideal Pt nanofilms.
Low-energy electron diffraction (LEED; Princeton Research
Instruments) and Auger electron spectroscopy (AES; Perkin-
Elmer, Eden-Prairie, MN) were performed in an ultrahigh
vacuum system which allowed sequential electrochemical and
surface analytical measurements, without exposure of the sample
to air. This process of using UHV surface analytical techniques
in studies of electrode surfaces is referred to as UHV-EC
methods.34 The electrode used for the UHV-EC studies was the
Au(111) single crystal slice, 99.999 % pure, commercially
oriented and polished from MaTecK, GmbH. The crystal was
cleaned with multiple cycles of Ar+-ion bombardment and
annealing prior to each experiment. LEED and AES were used
to check surface composition and structure, prior to each
experiment. Voltammetric studies were performed using this 1
cm diameter Au(111) single-crystal slice, 0.1 cm thick. The
polycrystalline sides contributed to the voltammetry, as well as
the (111) faces, as the whole crystal was immersed.
Experimental Section
For the in situ scanning tunneling microscopy experiments
(EC-STM), Au single-crystal surfaces were prepared using a

18000 J. Phys. Chem. B, Vol. 110, No. 36, 2006
Kim et al.
Figure 2. (a) Steady-state cyclic voltammogram for Cu UPD on Au-
(111) in 0.05 M H2SO4 and 1 mM CuSO4 solution. Potential sweep
rate: 5 mV/s. (b) EC-STM image of the honeycomb (x3 × x3)R30°
structure obtained at +0.15 V. Bias voltage: -150 mV. Tunneling
current: 20 nA.
Results and Discussion
The cyclic voltammogram for Cu UPD on Au(111) in a 0.05
M H2SO4 and 1 mM CuSO4 solution (Figure 2a) was very
similar to those reported for studies of gold single-crystal
electrodes previously reported.9,11 The two peaks at +0.23 and
+0.05 V correspond to formation of two distinctly different
ordered Cu UPD adlayers. Cu coverages of 2/3 and 1/3
monolayer (ML), respectively, were deposited in the first and
second deposition peaks, respectively. A ML is defined in this
paper as one adsorbate for every Au surface atom. At +0.15
V, the first Cu UPD peak resulted in the honeycomb (x3 ×
x3)R30° structure at a Cu coverage of 0.67 monolayer (ML),
while, after the second peak, the remaining 0.33 ML of Cu was
deposited at 0.0 V, completing a full Cu ML. In the first UPD
structure, (bi)sulfate anions occupied the centers of Cu hexagons,
the honeycomb structure, with an anion coverage of 1/3 ML,
known from the literature.35-38 This structure (Figure 2b)
remained stable during rinsing with blank at controlled potential
Figure 3. (a) Wide-area EC-STM images of the Pt submonolayer on
Au(111) obtained by replacement of Cu UPD layer in 0.1 M HClO4
solution at OC potential. (b) After I covered Pt submonolayer. Bias
voltage: -350 mV. Tunneling current: 20 nA.
During the redox replacement, imaging continued in the EC-
STM flow cell, allowing observation of large scale morphology
changes. Figure 3a is a representative STM image of the surface
after the SLR3, a relatively uniform distribution of Pt clusters
over the entire 300 × 300 nm surface, imaged at OC. The image
suggests a continuous layer was not formed nor were large
islands, nothing that suggests a two-dimension epitaxial deposit.
These results show no evidence that the SLR3 resulted in direct
electron transfer from one Cu atom to a Pt ion, direct adatom-
ion interaction, or in surface mediated transfer which should
allow Pt deposition anywhere on the surface, regardless of the
Cu atom site. Pt ions appear to reduce to Pt adatoms and then
diffuse over the surface, creating very small Pt adatom clusters.
The stoichiometry of the SLR3 limits Pt deposition to at most
a submonolayer (2/3 ML) on Au(111). It is entirely possible
that even less Pt was actually deposited, raising the question of
redox replacement efficiency. Determination of the efficiency
was one of the challenges of this work, how to quantify the Pt
deposited. Electrochemical stripping characterization via Pt
2-
in the STM flow cell. A 0.1 mM PtCl4 + 0.1 M HClO4
solution was introduced at open circuit (OC) to replace the Cu
UPD layer. During solution exchange, the OC potential shifted
to 0.6 V, over the course of 3 min, a potential positive enough
to suggest that all Cu had been replaced, subsequently confirmed
via Auger. The cell was then rinsed with blank.

Platinum Nanofilm Formation by EC-ALE
J. Phys. Chem. B, Vol. 110, No. 36, 2006 18001
Figure 4. Height-shaded plot images of the Pt submonolayer on Au(111) in 0.1 M HClO4 solution at OC potential: (a) 100 nm × 100 nm; (d) 15
nm × 15 nm. High-resolution EC-STM images showing the (x3 × x3)R30°-I adlayer consists of hexagonal arrays: (b) 20 nm × 20 nm; (c) 15
nm × 15 nm. Bias voltage: -250 mV. Tunneling current: 20 nA.
coulometry was not possible. In addition, Auger spectra for Au
and Pt are so similar that independent quantification of Pt and
Au was not possible. XPS should allow characterization of
submonolayer amounts of Pt, even in the presences of Au, and
is presently being upgraded on the UHV-EC instrument.
It is well-known that various electrochemical processes, such
as hydrogen and metal UPD, are strongly affected by coadsorbed
electrolyte.3,34,39 In particular, adsorbed iodine enhancing surface
mobility on single-crystal surfaces has been studied in detail
with Cu, Au, and Pd single-crystal electrodes using UHV-EC
and EC-STM techniques.30,40,41 For example, disordered Pd(111)
surfaces can be spontaneously reordered upon exposure, at
ambient conditions, to a dilute aqueous solution of iodide.41 This
process is, in part, responsible for “electrochemical annealing”,
which can also be promoted by the use of relatively positive
potentials, where the exchange current is maximized. In these
studies, iodine adsorbate-induced surface ordering has been ap-
plied to facilitate the growth of more uniformed epitaxial Pt layers.
Figure 3b shows an STM image obtained for an iodine-cover-
ed Pt submonolayer after the first SLR3 cycle (Figure 3a). This
300 nm × 300 nm STM image shows a more uniform distribu-
tion of small nanoparticles, more two-dimensional, than for Pt
deposits without the I atom layer. There is no sign of preferen-
tially deposition along the step edges, a frequent site for three-di-
mensional growths in conventional electrodeposition. Extended
terraces are clearly visible in Figure 3b. Furthermore, a higher
resolution view (100 nm × 100 nm) of the terrace (Figure 4a)
reveals a uniform distribution of 1-3 nm wide and 0.1 nm high
particles over the entire area (Figure 4a). A closer view (Figure
4b) reveals the particles’ distorted hexagonal shapes. These
features display a corrugation of only 0.08 nm. It must be noted
that these images are actually of the I atoms adsorbed on the Pt
particles and the Au(111) surface. From a vast number of halide
adsorption studies on metal surfaces it is known halides tend
to form close-packed hexagonal adlayers, with excellent surface
mobility. In general, a halide atom will form one bond, and
thus they do not tend to bond to each other and are present at
near their van der Waals diameters (0.5 nm). At higher
potentials, I atoms are bonded to each other, resulting in the
formation of I2 molecules which diffuse away. The point is, I
atom layers acts as a blanket over the surface, covering and
conforming to the underlying surface features.
The ordered hexagonal iodine adlayer has close-packed iodine
atomic rows, rotated 30° from the atomic rows in the Au(111).

18002 J. Phys. Chem. B, Vol. 110, No. 36, 2006
Kim et al.
Figure 5. High-resolution EC-STM image showing the isolated Pt
atoms as well as Pt dimers and trimers, with in the close-packed Au-
(111)-(1 × 1) substrate surface in 0.1 M HClO4 solution obtained when
the tunneling current was abruptly changed from 20 to 50 nA during
the scan. Bias voltage: -150 mV. Tunneling current: 50 nA.
Figure 6. Wide-area EC-STM images of the Pt layer after second Pt
metal EC-ALE cycle in 0.1 M HClO4 solution at OC potential. Bias
voltage: -340 mV. Tunneling current: 25 nA.
Arrows drawn in Figure 4c,d denote the directions of the x3
unit vectors ([121]), characterizing this structure as a Au(111)-
(x3 × x3)R30°-I.42 The surface appears to be a similar
hexagonal array of I atoms but with a series of hexagonally
shaped clusters, varying between 1.0 and 3.5 nm, and presum-
ably associated with the deposited Pt atoms. The hexagonal
features suggest significant order to the deposited Pt atoms.
To learn more about these features and the nucleation of Pt
on the I-coated Au(111) surface, EC-STM experiments were
performed using increased tunneling currents, to try and image
the underlying Pt particles. Figure 5 is a “dynamic tunneling
current” STM image, where the tunneling current was abruptly
changed from 20 to 50 nA, at OC, revealing a surface decorated
with isolated Pt atoms as well as Pt dimers and trimers. These
atoms appear to exist within the close-packed Au(111)-(1 × 1)
substrate surface structure. Note in Figure 4a the monatomic
step edge, in the lower left corner, which shows the corrugation
due to the Pt clusters is much less than an atom in height, only
about 0.08 nm. This strongly suggests that the first partial
monolayer of Pt atoms deposited with the SLR3 are not present
as adatoms but incorporated into the Au(111) surface to form
a surface alloy. Some Pt atoms may go subsurface as well. The
image also shows that the Pt atoms are almost perfectly incor-
porated into the parallel atomic rows of the Au(111) surface.
The first assumption would be that the Pt atoms are present
in hexagonal islands, or rings, from images of the I atom layer
(Figure 4). However, given the apparent distribution of, and
presence in the surface of, the Pt atoms (Figure 5), it is more
likely that the dispersed Pt atoms, dimmers, and trimers only
cause a limited perturbation in the overlying I layer. This dis-
turbs the (x3 × x3)R30°-I hexagonal array structure and pro-
duces hexagonal rings of I atoms pushed from their ideal, low,
3-fold sites, accounting for the observed images (Figure 4).
In this study, after one SLR3 cycle, it appears the Pt atoms
incorporate into the Au layer, rather than on top, while, without
the I atom layer, the Pt nanoclusters appear on top.43 The Pt
atom incorporation appears random, not preferentially near or
Figure 7. Auger electron spectrums of (a) clean Au(111), (b) after two
cycles of Pt replacements, and (c) after four cycles of Pt replacements.
at step edges, and shows little tendency to segregate forming
larger, pure, Pt islands. Theoretically, the Pt coverage should
2-
be 0.66 ML, given one for one exchange of Cu for PtCl4
.
Such a high coverage is not consistent with the image shown
in Figure 5, where high tunneling currents were used. It is
probable that the exchange was not 100% efficient, and studies
are underway to better quantify the exchange process.
The deposition began with exposure of the clean annealed
Au(111) surface to the KI solution and formation of the (x3 ×
x3)R30°-I adlattice. The solution was then exchanged for blank,
to flush I- ions from the cell, and followed by introduction of
the Cu2+ ion solution at +0.1 V, where it was held for 15 s or
2-
more. After Cu UPD, the deposit went OC while the PtCl4
solution was introduced, flushing Cu2+ ions from the cell and
allowing exchange of the Cu UPD layer for Pt. Finally, the cell
was rinsed again with the blank solution to remove excess
PtCl42- ions. Given that I atoms strongly adsorb to Pt, Cu, and
Au, significant loss of I atoms during a cycle was not expected.
However, an extra step, exposure of the surface to a 0.1 mM
KI solution, could be added to renew the I atom layer.
Alternatively, the cycle can simply begin again with Cu UPD
at +0.1 V.

Platinum Nanofilm Formation by EC-ALE
J. Phys. Chem. B, Vol. 110, No. 36, 2006 18003
Figure 9. High-resolution EC-STM images showing the formation of
a Moire´ pattern on the Au(111) surface after second Pt metal EC-ALE
cycle. Two images, taken 1 min apart, show the formation of a Moire´
pattern on the surface consisting of five atom clusters and I in 0.1 M
HClO4 solution at OC potential. Bias voltage: -300 mV. Tunneling
current: 25 nA.
Figure 8. High-resolution EC-STM images of the Pt layer after second
Pt metal EC-ALE cycle. Two images, taken 1 min apart, show the
arrangements of adlayers consisting of five atom clusters and I in 0.1
M HClO4 solution at OC potential. Bias voltage: -250 mV. Tunneling
current: 20 nA.
Au(111) substrate and a monolayer of Pt atoms. Figure 10
presents a structure proposed to account for Figure 9. The
proposed structure shows a (x37 × x37)R25.29° match
between the Au(111) surface and a Pt(111) monolayer, with
the I atoms adopting a (x3 × x3)R30° arrangement with
respect to the Pt monolayer. The period and rotation of the Moire´
pattern are consistent with the lattice mismatch between metallic
Au (2.88 nm) and Pt (2.78 nm).
Figure 6 was taken after a second SLR3, equivalent to that
shown in Figure 1. Instead of the hexagons (Figure 4) observed
after one cycle, the bright features on this surface appear more
like chains on a hexagonal background of I atoms. Figure 7
displays Auger spectra taken after two and four SLR3 of Cu
replacement by Pt. Although the Pt signals are not differentiable
from those for Au, the presence of the I atom layer is clearly
evident by the doublet at 500 eV. Closer inspection (Figure 8)
suggests the chains were composed of five atom clusters
(discussed below). The two images in Figure 8 were taken 1
min apart, indicating that some clusters have moved and some
have remained stationary while others have disappeared. There
are indications that some of the clusters are disappearing, while
pits in the surface are filling. Figure 9 shows the formation of
a Moire´ pattern on the surface, suggesting that the Pt layer has
been completed, and involves a lattice mismatch between the
Pt and Au. As noted, the I atoms act as a cover, indicating that
the Moire´ pattern results from the lattice mismatch between the
Figure 11 shows an image of the surface after a third SLR3
deposition cycle, displaying a new I atom structure, a (3 × 3)-
I, previously observed on Pt(111) surfaces.31,44,45 This is a strong
indication that a Pt atomic layer has been formed. Also evident
in Figure 11 is a significant disorder and an increased coverage
of the five atom clusters.
That some islands were formed on the surface after three
SLR3 cycles is consistent with the fact that the coverage of Cu
deposited each cycle is less than a monolayer. Studies of Cu
UPD on I-coated Au have suggested a coverage of only ∼0.44
ML, forming a (3 × 3) structure,46 while Cu deposition on Pt-

18004 J. Phys. Chem. B, Vol. 110, No. 36, 2006
Kim et al.
Figure 10. Structure proposed to account for the Moire´ pattern
observed in Figure 9. The structure shows a (x37 × x37)R25.29°
match between the Au(111) surface and a Pt(111) monolayer, with the
I atoms adopting a (x3 × x3)R30° with respect to the Pt monolayer.
The period and rotation of the Moire pattern are consistent with the
lattice mismatch between metallic Au (2.88 nm) and Pt (2.78 nm).
Figure 12. High-resolution image of the internal features of five atom
clusters proposed to be one central discharged Pt atom in the center of
four of the surface I atoms. Two images, taken 1 min apart, show the
discharged Pt atom is commensurate with filling in underlying Pt atomic
layer. Bias voltage: -300 mV. Tunneling current: 30 nA.
simply Pt adatoms that have not incorporated into a second layer
of Pt yet. Some island like features appears to be composed of
groups of these clusters. It is proposed here that the clusters
are composed of a discharged Pt atom in the center of four
surface I atoms. Again, repetitive scans show that these Pt atoms
have mobility on the surface, facilitating electrochemical
annealing.47 Real time EC-STM images clearly shows that the
number of 5 atom clusters goes down with time, as they diffuse,
forming second layer Pt islands or domains.
Figure 11. High-resolution EC-STM image of the Pt layer after a third
Pt metal EC-ALE cycle. This image shows a new I atom structure
developing on the surface, a (3 × 3)-I on deposited Pt layer. Bias
voltage: -250 mV. Tunneling current: 20 nA.
(111) surface coated with I atoms results in a coverage of 0.44
ML, as well.3 Thus after two cycles, a little under a ML would
be expected if the conversion efficiency was 100%. The relative
homogeneity and the presence of a moire´ pattern after two cycles
suggest that a near full monolayer was formed after two cycles.
This would then suggest that the coverage after three cycles
would be closer to 1.5 ML, rather than two, and would explain
the presences of some island like features on the surface.
Figure 12 shows a high-resolution image of the five atom
clusters. Similar features have been seen in a number of studies
of PtX42- (X ) Cl, Br, I).18,19 It is proposed here that these are
Some care must be taken when comparing previous EC-ALE
works where UPD was used to deposit each atomic layer of
the elements composing a compound, with the present use of a
SLR3. UPD is a thermodynamic process, producing homoge-
neous equilibrium atomic layers. In a SLR3 the process starts
with UPD, but the atomic layer is then exchanged for a more
noble atomic layer at OC. From these studies, well-ordered
atomic layers result, using the I atom layer as a surfactant to
promote annealing.
The alternated UPD of two elements to form a compound
using EC-ALE does have an advantage, in that the chemistry

Platinum Nanofilm Formation by EC-ALE
J. Phys. Chem. B, Vol. 110, No. 36, 2006 18005
of forming a compound promotes deposition of a UPD layer of
one element only where an atomic layer of the other exists,
mandating layer by layer growth. In the present case, no such
driving force exists; deposition of an atomic layer can deposit
anywhere on the surface where the sacrificial metal exists,
including on islands of Pt already formed. This does not suggest
that the surface will roughen dramatically, only that it may show
more tendency to roughen than a strictly layer by layer process.
Atomic resolution imaging during growth will be required better
understand any tendency.
If the exchange of Pt for Cu were direct, the electrons of one
Cu atom transferring to one Pt ion, then layer by layer deposition
should be promoted. However, if the surface mediates electron
transfer, more roughening would be expected. Recent studies
of an SLR3 where Pb UPD was replaced by Cu have shown
that if Pb UPD was performed on half the electrode, followed
by exposure of the whole electrode to a Cu ion solution, from
Auger, Cu was observed to only deposit where the Pb UPD
was present, suggesting that direct transfer may be occurring.
If electrode-mediated electron transfer was predominant, Cu
would be expected to deposit on areas devoid of Pb UPD, as
well.
acknowledged, as well as the Department of Energy (NNSA)
under award No. DE-FG36-05GO85012.A000.
References and Notes
(1) Kolb, D. M. AdVances in Electrochemistry and Electrochemical
Engineering; Gerischer, H., Tobias, C. W., Eds.; John Wiley: New York,
1978; Vol. 11, p 125.
(2) Stickney, J. L.; Rosasco, S. D.; Song, D.; Soriaga, M. P.; Hubbard,
A. T. Surf. Sci. 1983, 130, 326-347.
(3) Stickney, J. L.; Rosasco, S. D.; Hubbard, A. T. J. Electrochem.
Soc. 1983, 131, 260-267.
(4) Aldaz, A.; Clavilier, J.; Feliu, J. M. J. Phys. IV 1994, 4, 75-93.
(5) Gewirth, A. A.; Niece, B. K. Chem. ReV. 1997, 97, 1129-1162.
(6) Herrero, E.; Buller, L. J.; Abruna, H. D. Chem. ReV. 2001, 101,
1897-1930.
(7) Magnussen, O. M.; Hotlos, J.; Nichols, R. J.; Kolb, D. M.; Behm,
R. J. Phys. ReV. Lett. 1990, 64, 2929-2932.
(8) Magnussen, O. M.; Hotlos, J.; Beitel, G.; Kolb, D. M.; Behm, R.
J. J. Vac. Sci. Technol., B 1991, 9, 969-975.
(9) Manne, S.; Hansma, P. K.; Massie, J.; Elings, V. B.; Gewirth, A.
A. Science 1991, 251, 183-186.
(10) Hachya, T.; Honbo H.; Itaya, K. J. Electroanal. Chem. 1991, 315,
275-291.
(11) Kimizuka, N.; Itaya, K. Faraday Discuss. 1992, 94, 117-126.
(12) Corcoran, S. G.; Chakarova, G. S.; Sieradzki, K. Phys. ReV. Lett.
1993, 71, 1585-1588.
(13) Ogaki, K.; Itaya, K. Electrochimi. Acta 1995, 40, 1249-1257.
(14) Tao, N. J.; Pan, J.; Li, Y.; Oden, P. I.; DeRosa, J. A.; Lindsay, S.
M. Surf. Sci. 1992, 271, L338-L344.
Conclusion
(15) Hsieh, S. J.; Gewirth, A. A. Surf. Sci. 2002, 498, 147-160.
(16) Bauer, E.; van der Merwe, J. H. Phys. ReV. B 1986, 33, 3657-
3671.
This paper represents the first atomic level study of the
repeated application of an SLR3 in an EC-ALE. Use of an I
atom atomic layer as a surfactant served to promote layer by
layer growth and the formation of well-ordered deposits. It
appears that the first atomic layer of Pt deposited resulted in a
surface alloy, while deposition of a second Pt atomic layer
resulted in formation of a well-ordered Pt adlayer, producing a
Moire´ pattern in the I atom layer. The Moire´ pattern is easily
explained by the lattice mismatch between Pt and Au. The
pattern appears to correspond to a (x37 × x37)R25.29° unit
cell between the Au and Pt and a (x3 × x3)R30° between the
Pt and I atoms.
(17) Uosaki, K.; Ye, S.; Naohara, H.; Oda, Y.; Haba, T.; Kondo, T. J.
Phys. Chem. B 1997, 101, 7566-7572.
(18) Waibel, H. F.; Kleinert, M.; Kibler, L. A.; Kolb, D. M. Electrochim.
Acta 2002, 47 (9), 1461-1467.
(19) Nagahara, Y. H., M.; Yoshimoto, S.; Inukai, J.; Yau, S.-L.; Itaya,
K. J. Phys. Chem. B 2004, 108, 3224-3230.
(20) Goodman, C. H. L.; Pessa, M. V. J. Appl. Phys. 1986, 60, R65-
R82.
(21) Kuech, T. F.; Dapkus, P. D.; Aoyagi, Y. Atomic Layer Growth
and Processing; Materials Research Society: Pittsburgh, PA, 1991; Vol.
222, p 360.
(22) Bedair, S. Atomic Layer Epitaxy; Elsevier: Amsterdam, 1993; p
304.
(23) Gregory, B. W.; Suggs, D. W.; Stickney, J. L. J. Electrochem. Soc.
1991, 138, 1279-1284.
Along with formation of the Moire´ pattern, evidence of five
atom clusters was seen distributed across the surface. Those
clusters appear mobile and disappear with time, commensurate
with filling in pits in the surface. It is proposed that the clusters
are neutral Pt atoms surrounded by four I atoms. When the
cluster finds a site, the Pt atom inserts, filling pits in the surface.
After a third SLR3, the I atom layer changed from a (x3 ×
x3)R30°-I, generally observed on Au(111), to a (3 × 3)-I
structure, observed on Pt. Thus, there appears to be a transition
from the Au to the Pt that starts with simply alloying the Pt
into the Au(111) surface (first cycle), followed by formation of
a full Pt atomic layer, where a lattice mismatch results in a Moire´
pattern (second cycle). The Pt monolayer producing the Moire´
pattern still supports a (x3 × x3)R30°-I adlattice. However,
after the third cycle, initiation of a second monolayer of Pt
resulted and the Moire´ pattern was gone, as was the (x3 ×
x3)R30°-I structure. Apparently, the second Pt atomic layer
supports a structure more similar to bulk Pt(111) than to Au-
(111).
(24) Stickney, J. L. Electroanal. Chem. 1999, 21, 75-209.
(25) Stickney, J. L. In AdVances in Electrochemical Science and
Engineering; Alkire, R. C., Kolb, D. M., Eds.; Wiley-VCH: Weinheim,
Germany, 2001; Vol. 7, pp 1-105.
(26) Brankovic, S. R.; Wang, J. X.; Adzic, R. R. Surf. Sci. 2001, 474,
L173-L179.
(27) Mrozek, M. F.; Xie, Y.; Weaver, M. J. Anal. Chem. 2001, 73,
5953-5960.
(28) Colum, F. Standard Potentials in Aquous Solution; Bard, A. J.,
Parsons, R., Jordan, J., Eds.; Marcel Dekker: New York, 1985; Chapter
16.
(29) Lay, M. D.; Sorenson, T. A.; Stickney, J. L. J. Phys. Chem. B 2003,
107, 10598-10602.
(30) Villegas, I.; Ehlers, C. B.; Stickney, J. L. J. Electrochem. Soc. 1990,
137, 3143-3148.
(31) Schardt, B. C.; Yau, S. L.; Rinaldi, F. Science 1989, 243, 1050-
1053.
(32) Clavilier, J.; Faure, R.; Guinet, G.; Durand, R. J. Electroanal. Chem.
Interfacial Electrochem. 1980, 107, 205-209.
(33) Soriaga, M. P.; Kim, Y.-G.; Soto, J. E. Interfacial Electrochemistry;
Wieckowski, A., Ed.; Marcel Dekker: New York, 1999; p 249.
(34) Soriaga, M. P.; Stickney, J. L. Modern Techniques in Electroanaly-
sis; Vanysek, Ed.; John Wiley & Sons: New York, 1996; Vol. 139, pp
1-58.
Electrochemical annealing is promoted by the presence of
the I atom layer. Some mobility was evident in the five atom
clusters, where the central atom was a zerovalent Pt, surrounded
by four I atoms. Those clusters appear to diffuse across the
surface, until the Pt atom is inserted into the growing Pt atomic
layer. Alternatively, it may only be the Pt atom that diffuses,
from 4-fold site in the I atom layer to the next I atom site.
(35) Zei, M. S. Q., G.; Lehmpfuhl, G.; Kolb, D. M. Ber. Bunsen-Ges.
Phys. Chem. 1987, 91, 349.
(36) Shi, Z. L. J. J. Electroanal. Chem. 1994, 364, 289-294.
(37) Shi, Z.; Lipkowski, J. J. Electroanal. Chem. 1994, 365, 303-309.
(38) Toney, M. F.; Howard, J. N.; Richer, J.; Borges, G. L.; Gordon, J.
G.; Melroy, O. R. Phys. ReV. Lett. 1995, 75, 4472-4475.
(39) Hubbard, A. T., Chem. ReV. 1988, 88, 633-656.
(40) Goetting, L. B.; Huang, B. M.; Lister, T. E.; Stickney, J. L.
Electrochim. Acta 1995, 40, 143-158.
Acknowledgment. Support from the National Science
Foundation, Divisions of Materials and Chemistry, is gratefully
(41) Kim, Y.-G.; Baricuatro, J. H.; Soriaga, M. P.; Suggs, D. W. J.
Electroanal. Chem. 2001, 509, 170-174.

18006 J. Phys. Chem. B, Vol. 110, No. 36, 2006
Kim et al.
(42) Bravo, B. G.; Michelhaugh, S. L.; Soriaga, M. P.; Villegas, I.;
Suggs, D. W.; Stickney, J. L. J. Phys. Chem. 1991, 95, 5245-5249.
(43) Brankovic, S. R.; Wang, J. X.; Adzic, R. R. J. Serb. Chem. Soc.
2001, 66, 887-898.
(44) Felter, T. E.; Hubbard, A. T. J. Electroanal. Chem. 1979, 100, 473-
491.
(45) Inukai, J.; Osawa, Y.; Wakisaka, M.; Sashikata, K.; Kim, Y. G.;
Itaya, K. J. Phys. Chem. B 1998, 102, 3498-3505.
(46) Martinez-Ruiz, A.; Valenzuela-Benavides, J.; de la Garza, L. M.;
Batina, N. Surf. Sci. 2001, 476, 139-151.
(47) Stickney, J. L.; Villegas, I.; Ehlers, C. B. J. Am. Chem. Soc. 1989,
111, 6473-6474.