Dynamic Article LinksC<Chemical Science

Cite this: Chem. Sci., 2012, 3, 3307

www.rsc.org/chemicalscience EDGE ARTICLE

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

28

Febr

uary

201

3Pu

blis

hed

on 3

0 A

ugus

t 201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2SC

2100

5CView Article Online / Journal Homepage / Table of Contents for this issue

Nucleation and growth of metal on nanoelectrodes

Jeyavel Velmurugan, Jean-Marc No€el, Wojciech Nogala and Michael V. Mirkin*

Received 18th July 2012, Accepted 15th August 2012

DOI: 10.1039/c2sc21005c

Three-dimensional nucleation and growth on active surface sites are fundamentally important initial

stages of the electrodeposition of metals. Electrochemical studies of these processes are greatly

complicated by the formation of multiple crystals interacting with each other. We investigated Ag

electrodeposition on the surface of well-characterized, nanometer-sized Pt electrodes and measured the

nucleation/growth kinetics of individual Ag crystals by a combination of nanoelectrochemistry and

atomic force microscopy (AFM). Basic parameters, including the number of surface active sites, the

kinetic time lag and the number of growing nuclei, were directly accessed from current transients and in

situ AFM imaging. The existence of a single nucleation site on the surface of a 50 nm electrode

persisting through several deposition/stripping cycles has been demonstrated.

Introduction

In addition to its fundamental significance, electrodeposition of

metals is at the core of various industrial applications from gold

and chromium plating to fabrication of interconnects in elec-

tronic circuits to preparation of electrocatalysts for energy

storage.1,2 The key to controlling the morphology of deposited

metal is to understand the mechanism of three-dimensional

nucleation and growth—the initial stages of many electrodepo-

sition processes.3 The random nature of nucleation and the

difficulties in quantitative analysis of the signal, which is

produced by a large number of growing crystals interacting with

each other, impeded electrochemical studies of these processes.

Despite the development of approximate analytical models,4–6

more exact numerical simulations7,8 and statistical analysis,9,10

the extraction of numerous kinetic, thermodynamic and trans-

port parameters from experimental data is not straightforward.

Open questions remain about the nature and density of the

incipient nucleation sites on the surface,11 the time lag,10 the

dimensions of an ‘‘exclusion zone’’, i.e., the area around a

growing crystal where no new nuclei can be formed,12 and the

growth kinetics of a nm-sized nucleus.13

An intriguing possibility is to investigate nucleation/growth of

metals at a nanoscopic electrode surface, on which only a single

nucleus can be formed, and to use imaging techniques to facili-

tate the interpretation of the electrochemical data. In earlier

studies, the assumption was that a micrometer-sized electrode is

sufficiently small to observe the formation and growth of a single

nucleus.14,15 More recent studies showed that two nuclei can form

within a submicrometer distance from each other.16 Moreover,

the possibility of multiple nucleation at extremely small (e.g., 5 to

100 nm radius) carbon electrodes was suggested;17 however, with

Department of Chemistry and Biochemistry, Queens College – CUNY,Flushing, NY 11367, USA. E-mail: mmirkin@qc.cuny.edu

This journal is ª The Royal Society of Chemistry 2012

no adequate characterization of such electrodes, this claim is

hard to validate.

Here we employed electrochemical techniques and AFM to

investigate nucleation/growth of silver on well-characterized

nanometer-sized Pt electrodes. AFM was used previously to

image metal nuclei formed on step edges of graphite surfaces,18

monitor the formation of bimetallic micro- and nanoparticles,16

and study the growth of metal nuclei as a function of time and

overpotential.19 The combination of AFM with nanoelectrodes

allowed us to probe single-nucleus formation on individual active

sites.

Results

A cyclic voltammogram (CV) of Ag electrodeposition at the

100 nm-radius polished Pt electrode (Fig. 1A) shows a charac-

teristic hysteresis, appearing after the potential sweep reversal,

and a sharp anodic peak of Ag stripping; both features are

typical of metal nucleation/growth CVs at macroscopic elec-

trodes.20 The potentiostatic transient of Ag deposition (Fig. 1B)

was induced by stepping the potential of the same nanoelectrode

from 300 mV to �120 mV vs. a Ag quasi-reference. After the

initial charging current spike, the current increased proportion-

ally to the square root of time, according to eqn (1) (ref. 22),

I ¼ pzFð2DcÞ3=2VM1=2ðt� t0Þ1=2

"1� exp

�zFh

RT

�#3=2

(1)

where z¼ 1 is the ionic charge, F is the Faraday constant,D and c

are the diffusion coefficient and bulk concentration of Ag+,VM is

the molar volume of silver, h is the overpotential, t is the time and

t0 is the time at which the crystal starts to grow under diffusion

control.

The use of a Multiclamp 700B amplifier instead of a conven-

tional potentiostat allowed us to measure pA-range currents on a

Chem. Sci., 2012, 3, 3307–3314 | 3307

Fig. 1 The cyclic voltammogram (A) and potentiostatic transient (B) of

Ag electrodeposition at a 100 nm-radius Pt electrode. Solution contained

100 mM Ag2SO4 and 0.1 M H2SO4. (A) The potential sweep rate, n ¼50 mV s�1. Arrows show the potential sweep direction. (B) Theoretical

curve (red) was calculated from eqn (1) with D ¼ 1.5 � 10�5 cm2 s�1.21

Fig. 2 The potentiostatic transient of Ag electrocrystallization at (A)

and topographic AFM images of the 160 nm radius Pt electrode before

(B) and after (C) silver deposition. Solution contained 10 mM Ag+ in

0.1 M H2SO4. Theoretical curve (red) in (A) was calculated from eqn (1).

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

28

Febr

uary

201

3Pu

blis

hed

on 3

0 A

ugus

t 201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2SC

2100

5C

View Article Online

microsecond or millisecond time scale and to study nucleation/

growth of smaller Ag crystals. In Fig. 2A, an �15 ms long

experimental current transient was fitted to theory (i.e., eqn (1)).

A higher concentration of Ag+ (10 mM) was used to shorten the

deposition time scale and increase the signal/noise ratio. Two

topographic AFM images were obtained before (Fig. 2B) and

after (Fig. 2C) electrodeposition of Ag. The polished Pt surface

with a radius, a z 160 nm was slightly recessed into glass

(Fig. 2B; the recess depth was d z 8 nm). After the deposition,

one can see silver protruding from the glass sheath by �12 nm

(Fig. 2C). The nucleus in Fig. 2A started to grow �10 ms after

the electrode potential was stepped to �105 mV. This delay is

much shorter than the time lag observed previously in nucleation

experiments at macroscopic (or micrometer-sized) electrodes.10,14

Fig. 3 shows two Ag deposition transients obtained at a

much smaller electrode (a z 20 nm) by stepping its potential to

h ¼ �90 mV. At a higher Ag+ concentration (20 mM; Fig. 3A),

the deposition time for a 20 nm radius nucleus was only�2.5 ms.

The experimental data fitted to eqn (1) corresponds to the

nucleus growth essentially from the moment of its formation

(radius, r < 3 nm (ref. 23)) until its radius becomes larger than

that of the underlying Pt surface. With a lower Ag+ concentra-

tion (0.2 mM; Fig. 3B), the same process was monitored on a

much longer time scale.

The topographic AFM image obtained before Ag deposition

(Fig. 3C) shows that the electrode surface was flat and well-

polished with�1 nm roughness. The Pt surface was flush with the

3308 | Chem. Sci., 2012, 3, 3307–3314

surrounding glass insulator and could not be distinguished from

it.24 To visualize the Pt surface, an �12 nm thick layer of Pt was

removed electrochemically25 after the Ag deposit was dissolved.

The radius of the resulting recessed electrode (Fig. 3D) was

�20 nm in accordance with the value calculated from the steady-

state voltammogram of ferrocene (Fig. 3E).

The comparison of current transients in Fig. 1B, 2A, 3A and B

shows that the extent of the nucleation time lag (s) can vary

significantly. The dependencies of s on the electrode radius,

concentration of Ag+ and deposition overpotential are shown in

Table 1. Overall, the s values in Table 1 are significantly shorter

than the second-scale delay times measured previously at larger

This journal is ª The Royal Society of Chemistry 2012

Fig. 3 Potentiostatic transients of Ag electrodeposition (A and B), topographic AFM images (C and D) and steady-state CV of ferrocene (E) obtained

using the same�20 nm radius Pt electrode. Solution contained 20 mM (A) or 200 mM (B) of Ag+ in 0.1 MH2SO4; 5 mM ferrocene and 0.1 M TBAClO4

in acetonitrile (E). Theoretical curves (red) in (A) and (B) were calculated from eqn (1).

Table 1 The effects of the electrode radius (a), overpotential (h) and Agion concentration on the nucleation time lag (s). Each s value is the tableis the average obtained from 20 transients. The shown uncertainties are95% confidence intervals

a (nm) cAg+ (mM)

s (ms)

h¼ �50 mV

h¼ �80 mV h ¼ �100 mV

20 0.2 1731 � 204 717 � 74 173 � 3020 20 221 � 34 36.3 � 7.1 7.2 � 2.9200 0.2 66 � 19 14.5 � 1.7 6.7 � 0.4200 20 68 � 35 4.7 � 1.1 2.0 � 0.9

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

28

Febr

uary

201

3Pu

blis

hed

on 3

0 A

ugus

t 201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2SC

2100

5C

View Article Online

electrodes.10,14,19 The differences can be attributed to much

higher noise and charging current, and equipment limitations

preventing low-current measurements on a short time scale at

macroscopic electrodes.

Fig. 4 shows four pairs of non-contact mode topographic AFM

images obtained in situ before and after four consecutive Ag

electrodeposition experiments. The Pt surface of a�50 nm radius

electrode in Fig. 4A was recessed by �20 nm into the glass insu-

lator. This slight recess facilitated the visualization of the Pt

surface (cf.Fig. 3CandD) and also resulted in a significantlymore

uniform current distribution near the edge of the conductive

surface,26 thus alleviating concerns about a possible edge effect on

This journal is ª The Royal Society of Chemistry 2012

nucleation kinetics. Ag was deposited by stepping the electrode

potential from +200 mV to �100 mV vs. the Ag quasi-reference

for�3 s in solution containing 20 mMAg+. The low concentration

of Ag+ resulted in slow growth of a small (r z 10 nm) nucleus,

which can be seen in Fig. 4B. After stripping the Ag deposit, an

image of the same Pt electrode in Fig. 4C was similar to that in

Fig. 4A. The same deposition protocol was used to form a new

nucleus (Fig. 4D) on the same electrode. This deposition/disso-

lution sequence was repeated several times (Fig. 4A–H).

Fig. 5 shows AFM images representing four consecutive

deposition/stripping cycles conducted at a larger nanoelectrode

(a ¼ 190 nm, d ¼ 12 nm). Similarly to Fig. 4, a nucleus growing

on the same spot can be seen in Fig. 5B, D and H. However, this

nucleus is not present in Fig. 5F and instead a nucleus formed at

a different location, �200 nm away from the first one, can be

seen. Moreover, Fig. 5D shows both nuclei growing

simultaneously.

Discussion

The use of nanometer-sized electrodes in combination with AFM

allowed us to investigate early stages of metal electro-

crystallization at the single nucleus/single nucleation site level. By

eliminating the interactions between multiple growing crystals,

this approach facilitated the extraction of quantitative

Chem. Sci., 2012, 3, 3307–3314 | 3309

Fig. 4 In situ AFM images obtained before and after successive electrodeposition experiments at the same 50 nm radius Pt electrode. Images A, C, E

andGwere recorded before and images B, D, F andH after the 1st, 2nd, 3rd and 4th depositions, respectively. The deposition potential was�100 mV vs.

Ag quasi-reference. Before each deposition, the electrode potential was held at +200 mV to dissolve previously deposited metal.

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

28

Febr

uary

201

3Pu

blis

hed

on 3

0 A

ugus

t 201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2SC

2100

5C

View Article Online

information about nucleation/growth kinetics. Eqn (1) describes

diffusion-controlled growth of a hemispherical nucleus. The

agreement between the experimental transients (black curves in

3310 | Chem. Sci., 2012, 3, 3307–3314

Fig. 1B, 2A and 3A and B) and theoretical (red) curves is

surprisingly good, keeping in mind that a solid Ag nucleus

cannot grow as a perfect hemisphere and the single adjustable

This journal is ª The Royal Society of Chemistry 2012

Fig. 5 In situ AFM images corresponding to four Ag electrodeposition experiments at the same 190 nm radius electrode. Images A, C, E and G were

recorded before and images B, D, F, andH after the 1st, 2nd, 3rd and 4th depositions, respectively. Solution contained 1 mMAg+ and 0.1MH2SO4. The

deposition potential was �85 mV vs. Ag quasi-reference. Before each deposition, the electrode potential was held at +200 mV to dissolve previously

deposited metal.

This journal is ª The Royal Society of Chemistry 2012 Chem. Sci., 2012, 3, 3307–3314 | 3311

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

28

Febr

uary

201

3Pu

blis

hed

on 3

0 A

ugus

t 201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2SC

2100

5C

View Article Online

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

28

Febr

uary

201

3Pu

blis

hed

on 3

0 A

ugus

t 201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2SC

2100

5C

View Article Online

parameter in eqn (1), t0, could only be varied within very narrow

limits. A good agreement of our data with the theory for hemi-

spherical nucleus growth may be related to the relatively weak

dependence of the diffusion current on the aspect ratio of the

crystallite, as discussed by Branco et al.27 Deviations in the short-

time region (Fig. 2A and 3B) are most likely due to the electron

transfer limitations at very small Ag nuclei. This effect could be

taken into account by fitting the experimental data to the equa-

tion for the nucleus growth under mixed diffusion/charge

transfer control.28 However, such fitting would involve two

additional adjustable parameters (i.e., the exchange current and

the transfer coefficient for Ag+ reduction at a growing silver

nanocrystal) whose values are difficult to validate.

The variation of silver ion concentration from low

(20–200 mM) to high (10–20 mM) values resulted in a broad

range of deposition times. Overall, Fig. 1–3 show that the growth

of a single Ag nucleus conforms to the classical diffusion-based

theory over four orders of magnitude in time. This finding can be

compared to the results in ref. 13, where TEM was used to

monitor in situ the growth of similarly sized Cu clusters. The

growth rate of an individual crystal was evaluated by measuring

its size as a function of time. However, the overlap of the

diffusion layers of multiple nuclei growing on a macroscopic

electrode surface complicated the data analysis, and discrep-

ancies were found between the growth kinetics measured for

individual nanoclusters and the predictions of conventional

theory.

Table 1 shows that the kinetic time lag for the single nucleus

formation depends strongly on several experimental parameters.

As can be expected from existing theory, s decreases with

increasing electrode radius, metal ion concentration and over-

potential. A very strong effect of the overpotential can be

attributed to the changes in nucleation rate constant (A), active

site density (N0) and surface concentration of Ag adatoms. All of

these quantities are supposed to increase exponentially

with cathodic overpotential, h,3,11,29 and so the change in h

from�50mV to�100 mV resulted in the decrease in s by a factorof 10–30 for all cAg+ and a values in Table 1.

The time lag decrease with increasing cAg+ should be due to the

increasing concentration of Ag adatoms because N0 is indepen-

dent of ion concentration. The effect of cAg+ on s is similar at two

higher overpotentials (i.e., �80 mV and �100 mV) and some-

what lower at h¼�50 mV. Interestingly, this effect is much more

significant at a¼ 20 nm than at the larger (a¼ 200 nm) electrode.

A marked decrease in s at a larger a can be attributed to the

increased number of nucleation sites; the larger this number the

higher the probability of the nucleus formation within a given

time period. Typical values reported in the literature for the

density of latent nucleation sites vary over a wide range, 104 cm�2

< N0 < 1010 cm�2.29 Even for the largest value, N0 ¼ 1010 cm�2,

the expected number of active sites on the surface of a 20 nm

radius electrode is <1. For a 200 nm electrode, the expected

number of sites is >1.

Further insight was obtained by combining nano-

electrochemical nucleation experiments with AFM imaging

(Fig. 4 and 5). The formation of only one nucleus in every

deposition experiment conducted at the 50 nm electrode (Fig. 4)

is in line with existing theory because the probability of multiple

nucleation on the electrode of this size is extremely low.30The size

3312 | Chem. Sci., 2012, 3, 3307–3314

of the grown nucleus varied significantly because of the random

nature of nucleation, however, it always formed on the same spot

(within 5–10 nm uncertainty due to the finite size of the imaged

nucleus and limited spatial resolution of AFM images). This

finding suggests that only one latent nucleation site existed on the

electrode surface. The corresponding effective site density, N0 z1.3 � 1010 cm�2 is higher than the literature values, which is

consistent with the presence of only one active site on the elec-

trode surface. The limited AFM resolution does not allow us to

completely exclude the possibility of more than one nucleation

site located very close (e.g., within 10 nm) to each other.

However, the probability of site clustering seems to be low

because of the absence of latent sites on the rest of the electrode

surface.

Except for the finding that electrochemical nucleation on

graphite surfaces is confined to step edges,18,31 little is known

about the nature of nucleation sites. It is common to assume that

such sites can appear and disappear in the course of an electro-

deposition experiment.11 The data in Fig. 4 shows clear evidence

of a persistent nucleation site that remains active after several

deposition/stripping cycles. Apparently, there were two latent

nucleation sites on the surface of a 190 nm radius electrode

(Fig. 5) and Ag nucleated randomly either on the first, or the

second, or both sites. The effective site density N0 z 8.8 � 108

cm�2 in this case is within the range of literature values.

The presence of two growing nuclei suggests that the radius of

the exclusion zone around each of them was not larger than a few

tens of nanometers. One can also conclude that the difference

between the formation times of these nuclei was not larger than

a2/(pD) z 10 ms.32 Such events are indistinguishable on the time

scale of our experiments.

Comparing the above N0 values (as well as time lag values and

other nucleation parameters) to those obtained at macroscopic

electrodes, one should keep in mind the possibility of ‘‘edge

effects’’ at the nanoelectrode. The fraction of the surface area

adjacent to the Pt/insulator interface is much larger for a nano-

electrode than for a macroelectrode. Although there is no direct

evidence of different nucleation rates at the electrode edge, the

factors that may influence this process include faster bulk and

surface diffusion and possibly a different density of surface

defects at the metal/insulator interface.

Conclusions

The combination of nanoelectrochemistry with AFM allowed us

to observe and quantitatively investigate the formation and

growth of single metal nanocrystals on individual nucleation

sites. The existence of a single nucleation site on the surface of a

50 nm electrode persisting through several deposition/stripping

cycles has been demonstrated. Two active sites can exist on the

surface of a larger (e.g., 200 nm) electrode, so that metal crystals

can form on either of them or on both of them in the same

nucleation experiment. The kinetic time lag at nanoelectrodes

was much shorter than could be expected from previous nucle-

ation experiments at larger electrodes, indicating that the

nucleation rate may be faster than the values reported in earlier

studies.

The growth of a nm-sized Ag nucleus was shown to be diffu-

sion-controlled and to follow the classical theory over four

This journal is ª The Royal Society of Chemistry 2012

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

28

Febr

uary

201

3Pu

blis

hed

on 3

0 A

ugus

t 201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2SC

2100

5C

View Article Online

orders of magnitude in time. The developed approach should

also be useful for studying the effects of deposition conditions on

the geometry of growing metal clusters and nanoparticles,

monitoring the formation of dendrites, and addressing other

practically important issues.

Experimental

Chemicals

Silver sulfate was obtained from Sigma Aldrich. Ferrocene (98%;

Aldrich) was sublimed twice before use. Tetrabutylammonium

perchlorate (Fluka) was used as organic supporting electrolyte.

H2SO4 was from Aldrich. Aqueous solutions were prepared from

deionized water (Milli-Q, Millipore Corp.). 99.95% acetonitrile

(Aldrich) was used to prepare organic solutions.

Preparation of Pt nanoelectrodes and deposition of silver

Disk-type, flat nanoelectrodes were prepared by pulling 25 mm-

diameter annealed Pt wires into borosilicate glass capillaries with

the help of a P-2000 laser pipette puller (Sutter Instrument Co.)

and polished under video microscopic control, as described

previously.33 The electrode radius was evaluated from steady-

state voltammograms of ferrocene in acetonitrile. A two-elec-

trode setup was used for voltammetric experiments with a

nanometer-sized Pt working electrode and a Ag wire reference.

Voltammograms were obtained using a BAS 100B electro-

chemical workstation (Bioanalytical Systems).

The electrodes that exhibited good voltammetric response with

the effective radius value in agreement with that extracted from

AFM images (see below) were employed in silver deposition

experiments. A Multiclamp 700 B amplifier (Molecular Devices

Corporation, CA) was used in the voltage-clamp mode to apply

voltage between the Pt nanoelectrode and Ag wire serving as a

quasi-reference electrode and to measure the resulting current.

The signal was digitized using a Digidata 1440A analog-to-

digital converter (Molecular Device Corporation) at a sampling

frequency of 100–250 kHz. A low pass filter with 1–10 kHz

bandwidth was used, depending on the experimental time scale

and the level of noise. The data were recorded and analyzed using

pClamp 10 (Molecular Device Corporation). Current transients

were recorded following a three-step program. First, a potential

sufficiently positive for Ag oxidation (e.g., 200 mV vs. Ag wire)

was applied for several seconds in order to remove traces of

silver. Then, the potential was switched to 0 mV for 0.1–0.5 s

and stepped to the silver deposition value (between �50 mV

and �100 mV).

AFM imaging

An XE-120 scanning probe microscope (Park Systems) was

employed for imaging nanoelectrodes and electrodeposited

silver. PPP-NCHR AFM probes (Nanosensors) were used for

non-contact imaging in air. Imaging in solution was performed

with PPP-NCH AFM probes (Nanosensors).

The procedures for AFM imaging of nanoelectrodes either in

air or in solution were reported recently.24 Briefly, a nano-

electrode was mounted vertically with its polished surface facing

the AFM probe using a homemade sample holder and the

This journal is ª The Royal Society of Chemistry 2012

cantilever was positioned above it with the help of an optical

microscope. In a non-contact mode, the tip was brought within

close proximity of the sample using the approach function and

then the nanoelectrode was moved laterally in 200 nm steps to

bring the AFM probe to its apex. The travel direction was

selected to effect z-axis retraction of the piezo actuator in a close-

loop mode. This corresponded to sliding of the slanted tip surface

along the edge of the glass insulating sheath of the nanoelectrode.

When the piezo approached its upper limit, the z-stage motor

was retracted by 1 mm to maintain the actuator within its

range (12 mm).

Electrodeposition of a silver nucleus for AFM imaging was

carried out in a commercial liquid cell (Park Systems), which was

mounted on the stage of the XE-120 scanning probe microscope.

A Pt nanoelectrode was inserted through the bottom of the

liquid cell in solution containing Ag2SO4 and 0.1MH2SO4. After

conditioning the working nanoelectrode at +200 mV for

several seconds to remove traces of Ag, its potential was stepped

to �90 � 10 mV vs. Ag reference (slightly different potential

values were used in different experiments, depending on the

electrode size and other conditions) using Multiclamp 700B. The

deposition time was 2–3 s (Fig. 4) and 0.25–0.5 s (Fig. 5).

After each deposition experiment, a non-contact mode topo-

graphic AFM image of the nanoelectrode was obtained in situ.

Then, the silver deposit was stripped by biasing the electrode

at +200 mV vs.Ag wire reference and another AFM image of the

same electrode was obtained. A sequence of deposition/stripping

cycles yielded a series of images like those shown in Fig. 4 and 5.

Acknowledgements

The support of this work by the National Science Foundation

(CHE-1026582) and the Donors of the PetroleumResearch Fund

administrated by the American Chemical Society is gratefully

acknowledged. W. N. thanks for financial support from

European Union 7.FP under grant REGPOT-CT-2011-285949-

NOBLESSE.

Notes and references

1 Y. D. Gamburg and G. Zangari, Theory and Practice of MetalElectrodeposition, Springer, New York, 2011.

2 T. P. Moffat and D. Josell, Isr. J. Chem., 2010, 50, 312.3 A. Milchev, Electrocrystallization: Fundamentals of Nucleation andGrowth, Kluwer, Boston, 2002.

4 B. R. Scharifker and J. Mostany, J. Electroanal. Chem., 1984, 177, 13.5 M. Sluyters-Rehbach, J. H. O. J. Wijenberg, E. Bosco andJ. H. Sluyters, J. Electroanal. Chem., 1987, 236, 1.

6 M. V. Mirkin and A. P. Nilov, J. Electroanal. Chem., 1990, 283, 35.7 (a) Y. Cao and A. C. West, J. Electroanal. Chem., 2001, 514, 103; (b)Y. Cao, P. Searson and A. C. West, J. Electrochem. Soc., 2001, 148,C376.

8 L. Guo, A. Radisic and P. C. Searson, J. Phys. Chem. B, 2005, 109,24008.

9 A. Milchev, W. S. Kruijt, M. Sluyters-Rehbach and J. H. Sluyters,J. Electroanal. Chem., 1993, 350, 89.

10 M. Y. Abyaneh, M. Fleischmann, E. Del Giudice and G. Vitiello,Electrochim. Acta, 2009, 54, 879.

11 A. Milchev, J. Electroanal. Chem., 1998, 457, 36.12 M. E. Hyde and R. G. Compton, J. Electroanal. Chem., 2003, 549, 1.13 A. Radisic, P. M. Vereecken, J. B. Hannon, P. C. Searson and

F. M. Ross, Nano Lett., 2006, 6, 238.14 G. Gunawardena, G. Hills and B. Scharifker, J. Electroanal. Chem.,

1981, 130, 99.

Chem. Sci., 2012, 3, 3307–3314 | 3313

Dow

nloa

ded

by U

nive

rsity

of

Ten

ness

ee a

t Kno

xvill

e on

28

Febr

uary

201

3Pu

blis

hed

on 3

0 A

ugus

t 201

2 on

http

://pu

bs.r

sc.o

rg |

doi:1

0.10

39/C

2SC

2100

5C

View Article Online

15 R. L. Deutscher and S. Fletcher, J. Electroanal. Chem., 1988, 239, 17.16 H. Liu and R. M. Penner, J. Phys. Chem. B, 2000, 104, 9131.17 S. Chen and A. Kucernak, J. Phys. Chem. B, 2003, 107, 8392.18 M. P. Zach, K. H. Ng and R. M. Penner, Science, 2000, 290, 2120.19 M. E. Hyde, R. Jacobs and R. G. Compton, J. Phys. Chem. B, 2002,

106, 11075.20 S. Fletcher, C. S. Halliday, D. Gates, M. Westcott, T. Lwin and

G. Nelson, J. Electroanal. Chem., 1983, 159, 267.21 S. Kariuki and H. D. Dewald, Electroanalysis, 1996, 8, 307.22 G. J. Hills, D. J. Schiffrin and J. Thompson, Electrochim. Acta, 1974,

19, 657.23 If the growth rate of a hemispherical nucleus is diffusion-controlled,

its radius can be evaluated as r ¼ [2DVMc(t � t0)]1/2.

24 W. Nogala, J. Velmurugan and M. V. Mirkin, Anal. Chem., 2012, 84,5192.

25 P. Sun and M. V. Mirkin, Anal. Chem., 2007, 79, 5809.

3314 | Chem. Sci., 2012, 3, 3307–3314

26 P. N. Bartlett and S. L. Taylor, J. Electroanal. Chem., 1998, 453,49.

27 D. Branco, J. Mostany, C. Borr�as and B. R. Scharifker, J. Solid StateElectrochem., 2009, 13, 565.

28 (a) S. Fletcher, J. Cryst. Growth, 1983, 62, 505; (b) A. Milchev, Russ.J. Electrochem., 2008, 44, 619.

29 B. R. Scharifker and J. Mostany, in Encyclopedia of Electrochemistry,ed. A. J. Bard and M. Stratmann, Wiley VCH, Weinheim, 2003, vol.2, ch. 5.3.

30 A. Milchev, W. S. Kruijt, M. Sluyters-Rehbach, M. andJ. H. Sluyters, J. Electroanal. Chem., 1993, 362, 21.

31 S. A. Hendricks, Y.-T. Kim and A. J. Bard, J. Electrochem. Soc.,1992, 139, 2818.

32 E. Garc�ıa-Pastoriza, J. Mostany and B. R. Scharifker, J. Electroanal.Chem., 1998, 441, 13.

33 P. Sun and M. V. Mirkin, Anal. Chem., 2006, 78, 6526.

This journal is ª The Royal Society of Chemistry 2012