Interfacial Study of Copper Electrodeposition with the Electrochemical Quartz Crystal Microbalance

INTERFACIAL STUDY OF COPPER ELECTRODEPOSITION WITH THE ELECTROCHEMICAL QUARTZ CRYSTAL MICROBALANCE

Oscar Ulises Ojeda Mota, B.S.

Thesis Prepared for the Degree of

MASTER OF SCIENCE

UNIVERSITY OF NORTH TEXAS

May 2005

APPROVED:

Oliver M. R. Chyan, Major Professor Teresa D. Golden, Committee Member Ruthanne D. Thomas, Chair of the Department of Chemistry Sandra L. Terrell, Dean of the Robert B. Toulouse School of Graduate Studies Ojeda Mota, Oscar Ulises, Interfacial Study of Copper Electrodeposition

with the Electrochemical Quartz Crystal Microbalance (EQCM), Master of

Science (Analytical Chemistry), May 2005, 113 pp., 4 tables, 45 figures,

reference list, 139 references.

The electrochemical quartz crystal microbalance (EQCM) has been

proven an effective mean of monitoring up to nano-scale mass changes related

to electrode potential variations at its surface. The principles of operation are

based on the converse piezoelectric response of quartz crystals to mass

variations on the crystal surface. In this work, principles and operations of the

EQCM and piezo-electrodes are discussed. A conductive oxide, ruthenium oxide

(RuO2) is a promising material to be used as a diffusion barrier for metal

interconnects. Characterization of copper underpotential deposition (UPD) on

ruthenium and RuO2 electrodes by means of electrochemical methods and other

spectroscopic methods is presented.

Copper electrodeposition in platinum and ruthenium substrates is

investigated at pH values higher than zero. In pH=5 solutions, the rise in local pH

caused by the reduction of oxygen leads to the formation of a precipitate, characterized as posnjakite or basic copper sulfate by means of X-ray electron

spectroscopy and X-ray diffraction. The mechanism of formation is studied by means of the EQCM, presenting this technique as a powerful in-situ sensing device.

Oscar Ulises Ojeda Mota

ii ACKNOWLEDGMENTS

I would like to express my earnest appreciation and acknowledgements to all the people that had helped me in completing this goal. To Dr. Oliver M.R.

Chyan who’s many teachings and example I will carry through the rest of my life, thanks for your dedication. My special acknowledgments go to Dr. Teresa

Golden and Dr. Mohamed El-Bouanani, from whom I obtained invaluable patience and knowledge when discussing and learning about their fields of specialization. The financial support form the Welch Foundation and UNT Faculty

Research Grant is greatly appreciated. I am grateful for the support of my fellow group members, Yibin Zhang, Long Huang and Praveen Reddy, especially to Dr.

Tiruchirapalli Arunagiri and Dr. Raymond Chan, pillars of the group, whose enthusiasm and constructive discussions helped in the arrival of the work here presented. The help from the support staff and many graduate and undergraduate students in the chemistry department is highly appreciated. I would like to extend my thanks to Dr. Oscar Mendoza and Dr. Miguel Angel

Valenzuela, whose encouragement was paramount to the continuation of my academic formation. To Dr. Manuel Quevedo, Dr. Rosa Amelia Orozco and Dr.

Alejando Hernández, who showed me the value of good friendship at all times.

Most importantly to my parents, Miguel and Norma, who sparked my interest in science, and supported my decisions with confidence thought the years, and

Sarah Flores; my gratitude goes to you, for giving me understanding, love and support, since we started this plan together.

iii TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS ...... iii

LIST OF TABLES ...... vi

LIST OF FIGURES ...... vii

Chapters

1 INTRODUCTION ...... 1 1.1 Piezoelectricity ...... 1 1.2 The Electrochemical Quartz Crystal Microbalance (EQCM) ...... 5 1.2.1 Frequency-Mass Correlation ...... 6 1.2.2 The Equivalent Circuit ...... 11 1.2.3 Applications of the EQCM ...... 12 1.3 Electrode Shot Preparation ...... 14 1.4 Electrochemistry Fundamentals...... 19 1.5 X-Ray Photoelectron Spectroscopy ...... 25 1.6 Conclusions ...... 27 1.7 References...... 28

2 COPPER UNDERPOTENTIAL DEPOSITION ON RUTHENIUM AND RUTHENIUM OXIDE ...... 33 2.1 Introduction ...... 33 2.2 Experimental ...... 35 2.3 Results and Discussion...... 37 2.3.1 Copper Deposition on Ruthenium ...... 37 2.3.2 Copper Deposition on Ruthenium Oxide ...... 46 2.4 Ongoing Research ...... 53 2.4 Conclusion ...... 57 2.5 References...... 58

3 FORMATION OF BASIC COPPER SULFATES ...... 62 3.1 Introduction ...... 62 3.2 Experimental ...... 64

iv 3.3 Results and Discussion...... 68 3.3.1 Characterization of the Process at Different Stages...... 74 3.3.2 Effect of the Oxygen Reduction Reaction...... 81 3.3.3 Mechanism of Formation...... 86 3.3.4 Mass Dumping ...... 93 3.4 Conclusion ...... 93 3.5 References...... 95

REFERENCE LIST...... 101

v LIST OF TABLES

Page

1-1 Applications of the earlier piezoelectric devices...... 4

2-1 Thermodynamic parameters for Cu UPD on ruthenium...... 44

2-2 Thermodynamic parameters for Cu deposition on ruthenium oxide...... 49

3-1 Mass change efficiencies at anodic peaks “A1” and “A2” ...... 88

vi LIST OF FIGURES

Page

1-1 A quartz crystal ...... 3

1-2 AT type quartz cut...... 5

1-3 Employed EQCM setup ...... 7

1-4 Oscillating crystal...... 10

1-5 Equivalent circuit of the oscillating crystal...... 11

1-6 Mold to make an electrode...... 15

1-7 Soldering the shot and copper wire ...... 16

1-8 Pouring the epoxy into the mold ...... 17

1-9 Steel puck with unpolished electrode...... 18

1-10 Lapping fixture with unpolished electrode inside...... 19

1-11 Cyclic voltammetry potential-time profile...... 23

1-12 Potential-time profile of a CA experiment...... 24

1-13 Photoelectron emission in an XPS analyzer ...... 26

2-1 Inter-grain boundary diffusion prevention by the use of RuOx...... 35

2-2 Ru EQCM electrode in background electrolyte...... 38

2-3 Ru electrode in background electrolyte...... 39

2-4 XRD pattern of a Ru electrode...... 40

2-5 CV response of a Ru shot in a 2 mM CuSO4 solution + 0.5 M H2SO4.... 41

2-6 Potential region of peak “A1” obtained in a 2 mM CuSO4+0.5M H2SO4 solution on a Ru electrode ...... 42

2-7 CV of ruthenium electrode in a higher copper concentration solution.... 45

2-8 Underpotential deposition of Copper from a 10 mM CuSO4+0.5M H2SO4 solution on a Ru electrode ...... 46

2-9 A) Ru surface, B) The oxidized Ru surface (both 5x)...... 47

vii 2-10 Background of oxidized Ru substrate in 0.5 MH2SO4 solution ...... 48

2-11 Ruthenium Oxide CV response in a 2 mM CuSO4+0.5MH2SO4 solution ...... 50

2-12 CV in 2 mM CuSO4+0.5 M H2SO4 solution of a thermally produced ruthenium oxide ...... 51

2-13 Effect of different oxidation times of an oxidized electrode in the 2 mM CuSO4+0.5MH2SO4 solution ...... 52

2-14 CV profile of a Ru film deposited on a Si wafer substrate, in the 2 mM CuSO4+0.5 M H2SO4 solution ...... 54

2-15 EQCM response of a Ru electrode in 2 mM CuSO4+0.5MH2SO4 solution ...... 55

2-16 EQCM response of Ru in a 5 mM CuSO4 + 0.5 M H2SO4 solution ...... 56

2-17 EQCM response of Ru in a 10 mM CuSO4 + 0.5 M H2SO4 solution ...... 57

3-1 Adaptation to the EQCM’s mount for microscope picture ...... 67

3-2 Current/mass profiles of a Pt electrode in 0.5M H2SO4 solution; A) Current density, B) Mass change...... 68

3-3 Current/mass profiles of copper deposition from 2 mM CuSO4 + 0.1M K2SO4 adjusted at different pH values...... 70

3-4 Multiple scanning experiments at pH=5, 2 mMCuSO4+0.1MK2SO4 ...... 72

3-5 EQCM profile of copper deposition in a Pt. electrode at different ...... 73

3-6 XRD spectra of the precipitate formed, as obtained form the top of a Ru electrode P: Posnjakite ...... 75

3-7 Optical pictures of the Pt electrode in a 10 mM Cu2+ concentration solution, taken at different stages of the potential sweep...... 76

3-8 SEM Image of Cu2O crystals on a Ru coated Si wafer ...... 78

3-9 Cu 2p region of XPS spectrum for compounds formed in 2 mM CuSO4 +0.1 K2SO4 pH 5 solution a) holding at “C2” for 3 min; b) holding at “C1” for 3 min; c) “A2” or precipitate ...... 79

3-10 S 2p XPS spectrum of precipitate ...... 80

3-11 Cyclic voltammetry and EQCM corresponding to the CVs on a Pt electrode in 2 mM CuSO4 +0.1M K2SO4 pH 5 solution with Ar purging . 82

viii 3-12 Background of the Pt electrode in 0.1M K2SO4 pH=5 solution, with different gas purging ...... 83

3-13 CV of a Pt electrode in a 5 mM CuSO4+0.1 K2SO4 pH=5 buffer solution ...... 85

3-14 Cycling of a Cu2O loaded platinum surface in the “A2” potential region in 0.1M K2SO4 solution...... 90

3-15 Potential stepping of a cuprite loaded Pt surface to “A2” potential in K2SO4 electrolyte ...... 93

CHAPTER 1

INTRODUCTION

1.1 Piezoelectricity

The first experimental demonstration of a connection between the

appearance of macroscopic piezoelectric phenomena and its relationship with

crystallographic structure was published in 1880 by Pierre and Jacques Curie [1].

Their experiment consisted of measurements of surface charges appearing on

specially prepared crystal [2] (quartz, topaz, sugar cane, Rochelle salt) which

were subjected to mechanical stress. In those years, the results were credited to

the Curies' imagination and perseverance, considering that they were obtained

with very simple and common materials, like tinfoil, glue, wire, magnets and a

jeweler's saw.

While this was a scientific curiosity for the next third of a century, in the

scientific circles of the day, this effect was considered quite a "discovery," and

was quickly dubbed as "piezoelectricity" in order to distinguish it from other areas of materials and electric phenomena such as "contact electricity" (friction generated static electricity) and "pyroelectricity" (electricity generated from crystals by heating).

Analysis of the underlying reason of piezoelectricity is important in understanding the operation of piezoelectric devices. Because some atomic

lattice structures have as an essential unit (or "cell") a cubic or rhomboid cage

made of atoms [3], with no center of symmetry. This cage holds a single semi-

mobile ion which has several stable quantum position states inside the cell. The

ion's position state can be caused to shift by either deforming the cage (applied

strain) or by applying an electric field. The coupling between the central ion and

the cage provides the basis for transformation of mechanical strain to internal electric field shifts and vice versa.

The Curie brothers did not, however, predict that crystals exhibiting the direct piezoelectric effect (electricity from applied stress) would also exhibit the

converse piezoelectric effect (stress in response to applied electric field). This

property was mathematically deduced from fundamental thermodynamic

principles by Lippman [1] in 1881. The Curies immediately confirmed the

existence of the "converse effect" and continued on to obtain quantitative proof of

the complete reversibility of the mechanical deformations in piezoelectric

materials. This reverse piezoelectric effect is the primary principle of many

devices and sensors [4-9]. The first serious applications work on piezoelectric

devices took place during World War I. In 1917, P. Langevin [10] and French co- workers began to perfect an ultrasonic submarine detector. Their transducer was a mosaic of thin quartz crystals glued between two steel plates (the composite having a resonant frequency of about 50 kHz), mounted in a housing suitable for submersion. Working on past the end of the war, they achieved their goal of emitting a high frequency "chirp" underwater and measuring depth by timing the

return echo. The strategic importance of their achievement was not overlooked

by any industrial nation, and since that time the development of sonar

transducers, circuits, systems, and materials has never ceased.

Some of the most commonly used household appliances such as

phonogram cartridges and smoke detectors have evolved from the development

of this type of materials. In account for the evolution in the application of

piezoelectric devices, some of the early applications of these piezoelectric

properties [3] are summarized in the following table (table 1-1).

From all the aforementioned materials, quartz (SiO2) has gained its own

special place as application material for piezo systems, since it has a unique set

of mechanical properties that makes it the most suitable material for precision control devices. Its internal friction is low [3], and its properties can be found repetitively from one specimen to another. The most important characteristic is that its properties are very stable with respect to time. A perfect crystal unit of quartz (crystallographic group no.32) is shown in the next figure [7], showing its lack of symmetry center.

x y

Figure 1-1 A quartz crystal.

Table 1-1 Applications of the earlier piezoelectric devices. TIME PERIOD APPLICATION

• Early sonar

• Development of piezoelectric materials (zirconate 1910 – 1940 titanate).

• Phono cartridge.

• Microphones. 1940 – 1960 • Ceramic audio tone transducers.

• Development of new (Barium Titanate) family of

piezomaterials. 1960 – 1980 • Low fidelity-signal filters.

• Audio buzzers (smoke alarms compatible tone

generators). 1980-1990 • High frequency signal filtering.

When a quartz crystal rod is cut at different angles to its crystallographic axes, different piezoids are obtained [11]. The most commonly used cuts are the

AT and BT cut quartz crystals (figure 1-2). These particular slices are used in

order to minimize the effect of temperature change related to the high frequency

drift. The characteristic frequency vs. temperature of a typical AT –cut quartz

crystal will have a minimum thermal drift at a temperature similar to that of the experimental conditions (room temperature) [12].

z 35º 15’

X y

Figure 1-2 AT type quartz cut.

1.2 The Electrochemical Quartz Crystal Microbalance (EQCM)

One of the most recent developments in the field of electrochemical gravimetric measurements that take full advantage of the piezoelectric and

mechanical properties of quartz is the Electrochemical Quartz Crystal

Microbalance. The microbalance is based on a quartz crystal wafer that is

sandwiched between two electrodes [13]. If one of these AT-Cut quartz slices is

coated with gold or other conductive metal [14], and this metal layers are placed

in contact so as to function as a working electrode of an electrochemical setup,

and connect it to an oscillator circuit design (with a high-gain amplifier in its

feedback loop to allow for oscillation of the crystal in solution), a mass sensing

device based on the reverse piezoelectric effect of the vibrating quartz wafer will

be created [15].

The term “EQCM” refers to the electrochemical application of the quartz

crystal device, with the “M” (as for “micro”) standing for the detection sensitivity

commonly found when performing experiments with the same. A schematic of

EQCM setup issued in the present work is represented in figure 1-3. The two parts of a Teflon®* cell (obtained from CH InstrumentsTM†) compress the crystal

wafer placed between two o-rings in the bottom of the cell. The exposed

electrode area of the crystal consists in part of the metal’s key-hole pattern

deposited on top of the quartz wafer. Platinum, gold and blank crystals are

commercially available from different vendors, including International Crystal

Manufacturing (ICMTM‡).

Until fifteen or so years ago, the general belief was that the application of

a sensing device using quartz in shear mode vibration would be hampered by the

limitation of liquids, causing the cease of oscillation by damping effects of the

fluid in contact having a higher viscosity than air. Bruckenstein and Kanazawa

[16, 17] pioneered the application of the use of a quartz resonator in contact with

liquid, opening the field for new areas of discovery with electrochemical insight.

1.2.1. Frequency-Mass Correlation

* E.I. du Pont de Nemours and Company, www.teflon.com † CH Instruments, www.chinstruments.com ‡ International Crystal Manufacturers Corporation, www.icmfg.com

Sauerbrey [18] first demonstrated the relationship between a crystal vibrating and the corresponding changes in mass, by a series of simple deductions, based on the thickness shear mode of oscillation of a quartz crystal,

depicted as follows [19].

Cell Potentiostat (Top view)

Working Electrode

Oscillator

Counter Electrode Reference Electrode

Computer Cell (Side View) Working Electrode

Figure 1-3 Employed EQCM setup.

If we relate the thickness of the crystal to its resonance frequency by:

= λ tq q / 2 (1)

Where tq is the thickness of the slice, and λq is the wavelength of the

elastic wave propagating at the surface. We can write this relation in terms of the resonant frequency fq and the speed of propagation of the wave as:

v f t = q (2) q q 2

Therefore, by algebraic manipulation, the change in frequency (dfq) caused by an infinitesimal change in the crystal thickness can be represented as:

df q dtq = − (3) f q tq

Where the negative sign indicates that a decrease in the resonance frequency is

caused by an increase in the thickness of the crystal. The relationship between

-3 the mass of the crystal (Mq.) and its density (ρq=2.648 g cm ) is stated as:

M t = q (4) q ρ A q

Where A=surface of the quartz area. Substituting (4) in (3) gives us:

d − dM f = q (5) f q M q

The assumption was made that the acoustic properties of the foreign layer

are identical to those of quartz. Thus the foreign layer is treated as a layer of

quartz itself (dMq), uniformly distributed over the crystal surface. One can define

mf and mq as the mass per unit area of the foreign and quartz crystal layer, correspondingly, and rewrite equation (5) as:

d − m f = f (6) f q mq

And for materials of uniform density, we can rearrange to:

− 2∆mf 2 ∆f = q (7) ρ q vq A

The shear wave velocity(vq) is related to the shear modulus of quartz

11 -1 -2 (µq=2.947x10 g*cm s ) by the relation:

1  µ  2 v =  q  (8) q  ρ   q 

Substituting (8) in (7) will give us the final “Sauerbrey’s equation”, where fq equivalent to f0 is the fundamental frequency of the oscillator (quartz crystal).

− 2∆mf 2 ∆ = 0 f 1 (9) ()µ ρ 2 A q q

This more commonly used variation of the equation is:

= − ∆f Cf ∆m (10)

Where Cf is a constant that can be obtained by using either the

fundamental frequency of oscillation of the quartz, or experimentally, by

measuring the fundamental frequency at the beginning of each experiment. For crystals with a fundamental frequency of oscillation of 7.995 MHz, the calibration value of 1.4156x10-9 ng per hertz change is obtained.

It is assumed that the transverse velocity of the quartz surface is identical

to that of the adjacent liquid layer. Figure 1-4 illustrates the vibration mode of the

EQCM resonator [20]. Its surface is an antinode, which allows the shear wave

displacement in the x direction. This wave radiates from the surface in the z-axis

into the electrolyte. The amplitude of the shear wave is described by an

exponentially damped cosine function, decaying in water to 1/e of its original

amplitude at ca. 0.25 µm from the surface.

Oscillation ( z )

Shear Displacement ( x )

Figure 1-4 Oscillating crystal.

1.2.2. The Equivalent Circuit

For a better understanding of the variables involved in the EQCM

measurements, an equivalent circuit of the quartz resonator is commonly

introduced. From figure 1-5 [21], we can see that the capacitor (Co) represents

the shunt capacitance that includes the metal electrodes evaporated into the

quartz surface, holder, and leads. The rest of the RCA elements are composed

of the motional arm of the crystal (L1), represented as an induction force caused

by the vibrating mass of the arm, the motional capacitance(C1) reflecting the

quartz elasticity, and the bulk losses within the quartz crystal, represented as a

resistance R1 [22]. We can see from these diagrams [23] that the components of

the crystal oscillator can play an important roll in its overall performance.

ww L1 C11 R1

Figure 1-5 Equivalent circuit of the oscillating crystal.

Although it has been shown that thin films, rigidly attached to the resonant

surface, will not cause an increase in the width (impedance) of the oscillation

[24], care should be taken in order to minimize these effects, especially when studying cases of non rigid films.

1.2.3 Applications of the EQCM

The Sauerbrey relationship has been applied so as to measure minute mass changes that accompany many processes, like underpotential deposition

(UPD) or the electrodeposition of metal monolayer(s) on a foreign metal substrate at potentials less negative than that for deposition on the same metal surface [25]) with an observable difference of mass response caused by the

adsorption-desorption process. Also of new interest has been the study of

electrode morphology changes, ion transport during potential changes in redox

and conductive polymer films. The determination of the point of zero charge

(PZC) has been also studied by means of the EQCM [11].

One of the main advantages of the EQCM technique in the study of

adsorption phenomena is the possibility of obtaining simultaneous voltammetric

and gravimetric response. An equivalent molar mass can be determined

according to Faraday’s law, and get direct information about the electrode

reaction species, measured at the same surface [26]. The conventional treatment

of oxidation-reduction cycling to obtain reproducible voltammetric curves should

be treated with caution, since contribution from the electrode roughness (trapped water molecules) could occur and should me minimized.

Another strategy directed toward the taking advantage of the mass

sensing capabilities of the quartz microbalance, is to coat the surface of a QCM

with a polyurethane template, to effectively increase the affinity and detection

limit for patterned molecules. This effect is used in the so-called “artificial noses”,

particularly used in the fragrance and scents industry [27]. Surface imprinting has

been successfully performed on a molecular level for artificial biocatalysts and

protein imprinting [28, 29]. The polymer self-organizes around the living cells and

forms pits with selective recognition capabilities. The imprinting process takes

place on micrometer thick polyurethane coatings directly on microscope slides or

QCM electrodes. Targeted cells are then transferred to a stamp and are

subsequently pressed on pre-coated QCMs. This procedure generates packed honeycomb-like imprinted structures on the polymer surface.

The aforementioned method has also been used to sample E. coli bacteria, tobacco mosaic viruses (TMV) and potassium chloride (KCl) as an inorganic probe molecule [30]. This has shown increased selectivity and sensitivity towards analyte target molecules. These devices have improved time and practical advantages over other silicon chip-based detectors [31], which in turn can help solve the problem of real time sensing for bioterrorist attacks [32].

Among other uses, the crystal microbalance has also been employed as an important tool in studying the controlled release of drugs upon potential

application [7], as immunosensors [9] and to study surface tension and wetting velocities [11, 33]. The concept of “massograms” first described by Snook et al.

[34] in which the rates of changes of mass versus electrode potentials are

plotted, were used to effectively demonstrate the power of the EQCM to trace

non-faradic, non-deposition reactions (gas evolution). Electrochemical studies on

the dissolution mechanism of electroplated films [35-37] as well as selective

adsorption of anions [38-41] have also been reported.

1.3. Electrode Shot Preparation

In order to obtain working electrodes of the different materials used in the

present experiment, a procedure that describes the preparation of the mold and

electrode that will be contained is drafted and presented. Metallic shots from

ESPI®§ were used for the fabrication of the electrodes. For a thorough

description on the materials and polishing procedure, refer to the lab’s manual

[42]:

Procedure for making of the mold:

1. Take the MLA pipette and use the vise to secure the pipette tip.

2. With the hacksaw, cut off the non-tapered end of the pipette.

This non-tapered end of the pipette will serve as the mold to make the

electrode.

Electrode preparation:

§ Electronic Space Products Incorporated, www.espi-metals.com

1. Hold the mold with the vise, and use the 5/64-inch drill bit to drill a hole in the

middle of the mold.

2. Cut a 6-inch piece of copper wire using the wire cutters, and strip the ends of

the wire. Place your metal shot on a clean and hard surface. Using the

soldering iron, heat the metal shot and the copper wire; place the wire on the

metal shot to solder the two together. The risk here is that the metal and

copper wire will become very hot during the soldering step.

Figure 1-6 Mold to make an electrode.

3. Place the mold on a sheet of parafilm®** with the larger uncut opening facing

down, and put the shot inside the center of the mold. Make sure the copper

wire goes through the drilled hole.

4. Prepare the epoxy.

a. Place the petri dish on the analytical balance, and tare it.

Figure 1-7 Soldering the shot and copper wire.

b. In the petri dish, pour 10 grams of resin and 1.4 grams of hardener of the

LECOTM†† epoxy.

c. With the wooden stick, mix the resin and the hardener thoroughly and

gently as you try to minimize the occurrence of bubbles.

** Pechiney Plastic Packaging, Inc. Corporation, www.pechineyplasticpackaging.com †† Leco Corporation, www.leco.com

5. Pour the epoxy slowly into the mold (Fig.1-10), and allow it to cure for 24

hours under a wooden box. This will minimize the effects of air currents.

To test if the epoxy has dried, use a toothpick to touch the epoxy that you did

not use. If the toothpick sticks to the epoxy when attempting to be moved, the

epoxy is dry. If however the toothpick moves, the epoxy has not dried; so do

not touch the mold until further testing.

Figure 1-8 Pouring the epoxy into the mold.

6. After making sure the epoxy has dried, remove the plastic mold from the dried

epoxy by making little cuts on the plastic mold with the wire cutters. The

plastic mold should fracture into pieces, making the removal of the plastic

mold extremely easy. At this step you already have an unpolished electrode.

7. Place the stainless steel puck on the heat plate, and heat it for 5 minutes.

Apply a thin layer of clear hot mounting wax on the surface of the steel puck,

and place the unpolished electrode with the backside on the wax. Let it cool.

Figure 1-9 Steel puck with unpolished electrode.

8. Insert the steel puck with the unpolished electrode into the polishing holder or

lapping fixture. Loosen the side screw on the lapping fixture in order to adjust

and select the amount to be polished off.

Figure 1-10 Lapping fixture with unpolished electrode inside.

The electrode is now ready to be polished to mirror finish, by using increasing grit polishing pads, and poly diamond suspensions up to 1 micron

(from Allied®‡‡).

1.4 Electrochemistry Fundamentals

In order to obtain thermodynamic data of the electrochemical reactions

and of the involved compounds, electrochemical measurements were conducted

in a three-way electrode cell. Information related to the energy and entropy of a

system can only be obtained in systems that are very near or at equilibrium.

Electrochemical reactions always involve the passage of current, so it is

easy to let the reaction proceed near to the equilibrium by limiting the passage of

charge per time (current), which is an expression of the reaction rate [43]. The

relationship between the potential dependence on heterogeneous rate constants

‡‡ Allied High Tech Products Incorporated, www.alliedhightech.com

resides in the fact that the energy of electrons on the electrode’s surface is linearly bound with potential. Negative potential changes will destabilize the energy of electrons, thus increasing the amount of charge available. Positive potential changes will cause an accumulation of a positive excess charge, and thus reduce the energy of the electrons [44].

Data acquisition is accomplished by measuring the electrode’s potential- current proceeding at a working electrode with the use of a reference electrode.

Usually, reference electrodes are chosen to have electroactive species, with a composition that is not affected by current flow, and connected in the low impedance circuit of a potentiostat. A flow of hydrogen gas, covering a Pt foil, or the standard hydrogen electrode (SHE) was among the first reference electrodes to be used, and it electrode potential was chosen to be the zero reference.

The same current that flows in the working electrode must flow in the counter or auxiliary electrode. However, experiments are designed in a way that the processes at the counter electrode are not the rate-determining step. The purpose of the counter-electrode is this to close the circuit which the working electrode.

Thermodynamic equilibrium can de described by the Nernst equation, which make use of the standard potential; an empirical constant that gives information about the redox-active energy levels in oxidized (Ox) and reduced

(R) species[13]:

RT a E = E o + ln Ox (11) nF aR

Although standard potentials are the basic values for all thermodynamic

calculations, in practice, we have to look at potentials with conditional constants.

This means that the constant reflects the specific condition which deviates from standard conditions.

RT [Ox] RT γ [Ox] E 0' = E 0 + ln + ln (12) nF [R] nF γ [R]

According to the Nernst equation, one could measure the potential E=Eº at

25ºC and 1 atmosphere of pressure, when the activities of all species are equal to 1. In solutions with high concentrations, this is not accurate, so the need arises to measure potentials at concentrations orders of magnitude lower than the custom used solutions, and extrapolate the linear dependence to activity units. In this work, solutions in the mili-molar (mM) range were employed.

If we can control the extent and direction of polarization of an electrochemical reaction by the use of an external power supply, like a potentiostat, we can also expect to control the speed and direction of the reaction. These techniques are employed in controlled potential experiments, were the energy of the electrode’s surface, as measured by its potential, is swept or stepped towards a conceited value.

Cyclic voltammetry (figure 1-11) is by far the most widely used technique to obtain qualitative information about electrochemical reactions. It consists of a

linear scanning of the working electrode through triangular potential waveform,

with the current response being observed [45]. The rate at which the potential is

scanned through an experiment, is called the scan rate, and the switching

potential, or apex, can be adjusted by the program before each experiment. In

this work, typical values for scan rate consisted of 20 mV/s.

A variation of the CV experiment is the multiple cycle experiment, in which

the same triangular waveform is repeated for a determined number of times.

Typical values consisted of 10 to 20 cycles.

Among other parameters used to gather thermodynamic and kinetic

information on electron-transfer reactions are the anodic and cathodic peak

currents (ia and ic) and the anodic and cathodic peak potentials (Ea and Ec).

E(λ) Switching potential

Time (λ)

Figure 1-11 Cyclic voltammetry potential-time profile.

The formal redox potential (Eº’) for a reversible process, is given by the

E + E mean ( c a ) of the peak potentials. Another characteristic potential parameter 2

is the separation of the peak potentials ∆Ep. The theoretical value for ∆Ep for a reversible process is 0.057/n V. However, the measured value for a reversible process is generally higher due to the presence of defects on the electrode surface, uncompensated solution resistance and non-linear diffusion and/or a poor cleaning procedure of the surface.

The peak current in a CV experiment for a reversible process is given by the Randles-Sevcik equation [14]:

3 1 1 = 5 2 2 υ 2 ixnADCp (.26910 ) (13)

Where A is the electrode surface area (cm2), D is the diffusion coefficient (cm2/s),

C is the concentration of the electroactive species in the bulk solution (mol/cm3),

n is the number of electrons transferred and υ is the scan rate. Hence, for a

1/2 diffusion limited process, ip is proportional to C and proportional to υ . If the

diffusion coefficient for the redox couple is known ( i.e. potassium ferrous- and

ferri- cyanide), the area can thus be obtained.

When coupled with in-situ EQCM measurements, the potentiostat can be used in such a way so as to hold or “pause” the scanning potential during the regular CV cycle in order to obtain the specific mass change.

Chronoamperometry is another potential step experiment commonly used in electrochemistry. The potential is stepped, from an initial potential, usually the open circuit potential, to a potential desired, for a certain amount of time. This aims to increase or reduce the electron’s energy so a desired electrochemical behavior is obtained. Using this method, the nucleation and growth of different films have been elucidated, and information about the rate of the reaction or mass efficiency of electrochemical depositions, (when used with the EQCM) can also be used (Figure 1.12).

Time

Figure 1-12 Potential-time profile of a C.A. experiment.

The electrochemical experiments were performed using a CHI

InstrumentsTM 400 or 440 model time-resolved EQCM potentiostat and controlled

by a personal computer. Custom software was used to record cyclic

voltammograms (CVs) and to record all data from electrochemical experiments.

A platinum (Pt) foil or coiled rod was used as counter electrode, and the reference electrode consisted of an Ag/AgCl, with a potential difference of

E=0.197 vs. SHE, and a mercury/mercurous sulfate (Hg/HgSO4) reference

electrode (MSE), with a standard potential of 0.64V vs. SHE.

1.5 X-Ray Photoelectron Spectroscopy

Developed in 1940 by Siegban et al. in Uppsala (Sweden), this is a

commonly used technique to elucidate the chemical and structural components

of surfaces. It was originally denominated electron spectroscopy for chemical

analysis (ESCA), and nowadays, is mostly referred as X-Ray photoelectron

spectroscopy (XPS). It uses soft x-ray sources, like Al and Mg Kαhν=1253.6 and

1486.6 respectively) [46] emission, incident to the analyzed surface. Its operation

is based on the photoelectric effect; by energy conservation, the monochromatic

electromagnetic wave will have the effect of kicking out an electron from the

atom’s core energy levels, overcoming its binding energy (figure 1.13).

Vacuum hν Eke

Ebe

Core

Figure 1-13 Photoelectron emission in an XPS analyzer.

The photoelectron will enter the analyzer, (assuming vacuum conditions present in the analyzer) with a kinetic energy consisting of:

= υ − Φ − Eke h Eb (14)

Where hν is the energy of the incident ray, Φ is the work function of the apparatus, and Eb represents the electron’s binding energy. [47]. The detected kinetic energy is related to the binding energy of the analyzed species by the use of the work function of the analyzer, and the energy of the monochromatic beam.

The electron binding energy is characteristic of each analyzed sample, and it will reflect the oxidation state or the electron affinity of the chemical environment where the photoelectron was emitted. Higher oxidation states will have a shift towards higher binding energy values.

Since energy levels are quantized, the emitted photoelectrons have a kinetic energy distribution in the form of peaks, associated with electron energy levels from the each ionized orbital. The intensity of the peaks is directly related

to the amount of species found in the detected volume. The beam can penetrate

only to about 100 nm into the analyzed sample. Calibration is done by comparing

a reference value (Au 4F) with the Eb peak obtained by the instrument (4.5 eV).

The results from XPS analysis in the present work were obtained in a

VG®§§ ESCALAB MK II spectrometer, with an aluminum (Al) Kα emission ray as

source of excitation.

1.6 Conclusions

From this chapter, we can observe the historical evolution of the

piezoelectric resonator as a transducer device, arriving at the development in

electrochemical applications of the quartz crystal microbalance. The principles of

operation of the EQCM as a thickness shear mode oscillator were presented. By

means of the Sauerbrey’s equation, the EQCM technique was introduced as a

powerful mean to trace minute mass changes of interfacial reactions, and its

application as a powerful electrochemical in-situ and sensing device tool is

explored. A brief description on the methodology used in for the preparation of electrodes was included.

Finally, some fundamental concepts of the most commonly techniques used in this work, electrochemistry and XPS analysis were described.

§§ Thermo Electron Corportion, www.thermo.com

1.7 References

1. Mason, W.P., The Journal of the Acoustical Society, 1981. 70(6): p. 1561-

1566.

2. Ward, M.D. and Buttry, D.A., Science, 1990. 249(4972): p. 1000-1007.

3. Salt, D., Hy-Q Handbook of Quartz Crystal Devices, 1st edition, ed.

Berkshire, J., 1987. Van Nostrand Reinhold, Amsterdam. 229.

4. Deakin, M.R. and Buttry, D.A., Analytical Chemistry, 1989. 61(20): p.

1147A-1154A.

5. Buttry, D.A., Applications of the Quartz Crystal Microbalance, in

Electroanalytical Chemistry: A Series of Advances, ed. Bard, A.J., 1996.

Marcel Dekker: New York. p. 1-85.

6. Marx, K.A., et al., Biosensors and Bioelectronics, 2001. 16: p. 773-782.

7. O'Sullivan, C.K. and Guilbault, G.G., Biosensors and Bioelectronics, 1999.

14: p. 663-670.

8. Rodhal, M.F., Hook, M.F. and Kasemo, B.,Analytical Chemistry, 1996.

63(12): p. 2219-2227.

9. Sakti, S.P., et al., Sensors and Actuators B, 2001. 78: p. 257-262.

10. Ballentine, J., Acoustic Wave Sensors: Theory, Design and Physico-

Chemical Applications, in Acoustic Sensors: Theory, Design and Physico-

Chemical Applications, ed. J. Ballentine. Vol. 1. 1997, San Diego:

Academic Press. 263.

11. Hepel, M., Electrode-Solution Interface with the Electrochemical Quartz

Crystal Nanobalance, in Interfacial Electrochemistry: Theory, Experiment

and Application, ed. Wieckowski, A., 1999, Marcel Dekker: Postdam. p.

599-629.

12. Lu, C., Applications of Piezoelectric Quartz Crystal Microbalances, in

Methods and Phenomena: Their application in Science and Technology,

ed. C.C. Lu, A.W., 1984, Elsevier Science: Amsterdam. p. 393.

13. Wang, J., Analytical Electrochemistry. 2nd ed. 2000, New York: Wiley-

VCH. 207.

14. Bard, A.J. and Faulkner, L.R., Electrochemical Methods: Fundamentals

and Applications., 2nd ed. 2001, New York: Jonh Wiley and Sons. 833.

15. Buttry, D.A. and Ward, M.D., Chemical Reviews, 1992. 92(6): p. 1355-

1379.

16. Bruckenstein, S. and Swathirajan, S., Electrochimica Acta, 1985. 30(7): p.

851-855.

17. Kanazawa, K.K. and J.G.G. II, Analytical Chemistry, 1985. 57: p. 1770-

1771.

18. Sauerbrey, G., Zeitschrift fuer Physik, 1959. 155: p. 206-222.

19. Zhang, M., Electrochemical Quartz Crystal Microbalance (EQCM) Studies

of Electrocatalytic Reactions, in Chemistry Department. 1995, University

of Ottawa: Ottawa. p. 262.

20. Lee, W.W., White, H.S. and Ward, M.D., Analytical Chemistry, 1993. 65:

p. 3232-3237.

21. Schmitt, R.F., et al., Sensors and Actuators B, 2001. 76: p. 95-102.

22. Tsionsky, V., et al., Looking at the Metal/Solution Interface with the

Electrochemical Quartz Crystal Microbalance: Theory and Experiment., in

Electroanalytical Chemistry, a Series of Advances, ed. A.J. Bard, 2002,

Marcell Dekker: New York.

23. Ehahourn, H., et al., Analytical Chemistry, 2002. 74(5): p. 1119-1127.

24. Tsionsky, V., et al., Journal of Electroanalytical Chemistry, 2002. 525-525:

p. 110-119.

25. Herrero, E., Buller, L.J. and Abruña, H.D., Chemical Reviews, 2001.

101(7): p. 1897-1930.

26. Bard, A.J. and Ru-Ren, F., Journal of Physical Chemistry B, 2002. 106: p.

279-287.

27. Tomic, O., Ulmer, H. and Haugen, J.E., Analytica Chimica Acta, 2002.

220450: p. 1-13.

28. Hayden, O., et al., Sensors and Actuators B, 2003. 91: p. 316-319.

29. Hayden, O. and Dickert, F.L., Advanced Materials, 2001. 13(19): p. 1480-

1483.

30. Dickert, F.L., et al., Sensors and Actuators B, 2001. 76: p. 295-298.

31. Iqbal, S.S., et al., Biosensors and Bioelectronics, 2000. 15: p. 549-578.

32. Dickert, F.L., et al., Sensors and Actuators B, 2003. 95: p. 20-24.

33. Komplin, G.C. and Pietro, W.J., Sensors and Actuators B, 1996. 30: p.

173-178.

34. Snook, G.A., Bond, A.M. and Fletcher, S., Journal of Electroanalytical

Chemistry, 2002. 526: p. 1-9.

35. Itagaki, M., Tagaki, M. and Watanbe, K., Corrosion Science, 1996. 38(7):

p. 1109-1125.

36. Méndez, P.F., et al., Electrochimica Acta, 2004. In press.

37. Uchida, H., Hiei, M. and M. Watanabe, M., Journal of Electroanalytical

Chemistry, 1998. 452: p. 97-106.

38. Galliano, F., Olsson, C.-O.A. and Landolt, D., Journal of the

Electrochemical Society, 2003. 150(11): p. B504-B511.

39. Uchida, H., et al. Quartz Crystal Microbalance Study of the Specific

Adsorption of Halide Ions on Highly Ordered Au(111). in Sixth

International Symposium on Electrodes. 1996: The Electrochemical

Society.

40. Doblhofer, K., et al., Journal of the Electrochemical Society, 2003.

150(10): p. C657-C664.

41. Watanabe, M., Uchida, H. and Ikeda, N. Journal of Electroanalytical

Chemistry, 1995. 380: p. 255-260.

42. Flores, S., Making an Electrode. 2003: Denton, Tx. p. 19.

43. Scholz, F., Electroanalytical Methods. 1st edition. 2002, Berlin, New York:

Springer. 331.

44. Faulkner, L.R., Electrochemical Characterization of Chemical Systems, in

Physical Methods in Modern Chemical Analysis. 1983, San Diego, CA

Academic Press:.

45. Grosser, D.K., Cyclic Voltammetry: Simulation and Analysis of Reaction

Mechanism. 1993, New York: Wiley-VCH.

46. Nascente, P.A.P., Journal of Molecular Catalysis A: Chemical, 2005. 228:

p. 145-150.

47. Ebel, M.F., Journal of Electron Spectroscopy and Related Phenomena,

1976. 8: p. 213-224.

CHAPTER 2

COPPER UNDERPOTENTIAL DEPOSITION ON RUTHENIUM AND

RUTHENIUM OXIDE

2.1. Introduction As second row transition metal, ruthenium is a metal largely immune to

attack of atmosphere. It has a melting point of 2334ºC, and a boiling point of

4150ºC. The density of the solid is 12370 kg·m-3. It has a resistivity value of 6.8

µΩcm, and its hexagonal closed packed structure has a cell size of 449.19 X

310.66 pm [1]. Ruthenium (Ru) has been investigated as a possible catalyst for the oxidation of carbon monoxide [2-4] formic acid and methanol, as a fuel cell catalyst [5-8].

Although ruthenium is insoluble in hot acids, including aqua regia, it reacts readily with oxygen under heating (800ºC) to form Ru2O. Further oxidation will

yield ruthenium (VIII) oxide (RuO4), which is a toxic gas [9]. Ruthenium (IV) oxide

has a rutile structure, with available electrons in the d orbitals that enable it to exhibit metallic properties, (e.g. enhanced conductivity) [10]. Values for the resistivity of RuO2 have been reported to be as low as 35 µΩcm [1]. This material has been used as a dimensionally stable anode for the chlor-alkali industry [11].

Due to its high capacitance (700 F/g), ruthenium (IV) oxide is currently

explored as a promising material for electrode capacitors [12-15], and studied as

a diffusion barrier for oxygen [16-18] and metal gate devices [19].

Under potential deposition (UPD), this is, the electrodeposition of metal

monolayer(s) on a foreign substrate, at potential less negative than that for

deposition on the same metal surface [20], has been the subject of thorough

study, largely driven by the fundamental possibility of understanding the

deposition process in general [21-25], and the observation of different

electrochemical behavior caused by the strong interaction of the adsorbed atoms

and its corresponding substrate [26-30].

No RuOx Proposed Current

Cu Cu ILD ILD RuOx as C C Diffusion Plug Cu Seed Ta (Ru, RuO2) TaN R R

ILD

Figure 2-1 Inter-grain boundary diffusion prevention by the use of RuOx.

Due to its immiscibility with copper, ruthenium and ruthenium (IV) oxide

can be explored as barrier materials for copper (Cu) interconnects in integrated circuits [31, 32]. In the work presented herein, Underpotential deposition of Cu on

Ru and Ru oxide is explored. Ruthenium oxide can be used as a possible “plug”

or intergrain boundary diffusion barrier that can in turn be applied to the sub-65 nm microelectronics industry (figure 2-1).

2.2. Experimental

Hemispherical Ru shots with an average diameter of 4 mm were made into disk electrodes as described earlier (see: Chapter 1). Prior to every data

collection series (electrochemical and x-ray diffraction), the electrodes were

polished to a fine grid, using silicon carbide pads with to a 1µm mirror finish. High

purity copper sulfate (Aldrich®*) and sulfuric acid (Mallinckrodt®†) were used to

make all electrolyte solutions in ultra-pure water (18.2 MΩ, Millipore®‡). All

electrolyte solutions were pre-purged with argon gas and maintained under argon

(Ar) atmosphere.

In the case of reported Ru EQCM current-mass response, deposition of

films were done in a high vacuum chamber (1X10-8 Torr base pressure) at 200W

for 120 minutes, using a ruthenium target (Kurt J. Lesker®§, 99.95% purity) with

Ar as carrier gas.

Electrochemical investigations were performed using CHI 440 (CH

InstrumentsTM**) potentiostat/galvanostat. A conventional three-electrode cell with

a Platinum (Pt) foil as counter electrode and a saturated mercury/mercurous

sulfate (Hg/HgSO4 or MSE) with a potential shift of 0.647 V vs. the standard

hydrogen electrode (SHE), or a silver/silver chloride (Ag/AgCl 3M, 0.197 V vs.

SHE) as reference electrodes were employed. All potentials here referred,

including the potentials presented in the cyclic voltammetry (CV) curves, are

reported versus the MSE reference electrode.

The EQCM cell was placed inside a glass encasement, placed on top of a

granite base. Flowing water from a recirculation pump, connected to the glass

* Sigma Aldrich Corporation, www.sigmaaldrich.com † Mallinkrodt Incorporated, www.mallinkrodt.com ‡ Millipore Corporation, www.millipore.com § Kurt J. Lesker Corporation, www.lesker.com ** Ch Instruments, www.chintrsuments.com

encasement, was kept at 26ºC. This served to maintain a stable temperature throughout each experiment.

X-ray diffraction analysis (XRD) was done using a Siemens®†† model d-

500 diffractometer using CuKα1 (λ=0.15406 nm) and CuKα2 (λ=0.154438nm)

radiation. The diffractometer consists of a Bragg-Brentano [33] configuration

(θ:2θ coupled), 2θ ranging from 30° to 100°. The high voltage generator was set

to a potential of 35 kV and 24mA. Additional x-ray photoelectron spectroscopy

(XPS) characterization was performed on a VG®‡‡ ESCALAB MK II spectrometer, with Al Kα as x-ray source.

2.3 Results and Discussion

2.3.1 Copper Deposition on Ruthenium

Figure 2-2 shows the cyclic voltammetric curves of a ruthenium film in the background electrolyte (05 M H2SO4), studied by means of the electrochemical

quartz crystal microbalance (EQCM). The presence of hydrogen evolution and

desorption of absorbed hydrogen can be observed (region I). The anodic region

of surface oxidation (region II) is overlapped with the former region [34, 35].

At more anodic potentials, the presence of a reversible form of ruthenium

oxide, or early adsorption of oxygen region (region III) is observed, which is

†† Siemens AG, www.siemens.com ‡‡ Thermo Electron Corporation, www.thermo.com

0.5 M H SO / Ru ) 2 4 2 I 600 III 400 II

A/cm 200 µ 0 -200 -400 -600 -800 -1000 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 Current Density ( Density Current 0 -10 -20 -30 -40 -50 II -60 Mass (ng) Mass -70 III ∆ -80 I -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 V vs MSE

Figure 2-2 Ru EQCM electrode in background electrolyte.

suggested by different authors [36-38]. In this work, a clear difference in the Cu plating behavior was found between the freshly polished ruthenium electrodes, and electrodes polarized at higher potentials (0.85 V).

The anodically polarized films had higher capacitance as compared

to the Ru electrode shots, (figure 2-3). XRD characterization of a Ru electrode

0.5M H SO /Ru 2 4

20.0

0.0 ) 2

-20.0 A/cm µ -40.0

-60.0

-80.0 Current Density (

-100.0

-0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 V vs MSE

Figure 2-3 Ru electrode in background electrolyte.

1200 (101)

1000

800

600

400 (002) (102) (103)

Intensity (Ab. Units) (Ab. Intensity 200 (110) (110) (112)(201) (202) (004) 0

20 30 40 50 60 70 80 90 100 110 2θ

Figure 2-4 XRD pattern of a Ru electrode.

(figure 2-4) helped confirm that the 2θ value for the experimental patterns is within the range of 0.01° to 0.04°, as compared to that of the reference pattern of ruthenium (see: card PDF#06-063).

Figure 2-5 shows the result of progressive vertex cycling of a Ru electrode

in 2mM CuSO4 + 0.5 M H2SO4 solution. The voltammetry started from the open

circuit potential value (OCP) close to 0.0 V vs. MSE, and scanned negatively

towards the -0.65V vs. MSE potential limit, at 20mV/s. No electroactivity was

observed when scanning towards positive potentials close to the OCP value.

Two predominant anodic peaks can be observed, one close to the nernstian potential value of -0.4 V vs. MSE, and another one close to the potential value of

-0.26 V vs. MSE.

The relationship between the appearance of the first cathodic wave C1, at

-0.36 V, and the more anodic desorption peak A2, at -0.26V can be established from these experiments. The peak current height of A1 increases as the cathodic

1600 A1 1400 Vertex 1 (-0.35 V) Vertex 2 (-0.4 V) Vertex 3 (-0.45 V) ) 1200 2 Vertex 4 (-0.5 V) 1000 Vertex 5 (-0.55 V) Vertex 6 (-0.65 V) A/cm

µ 800 600 400 200 A2 0 Current Density ( -200 C1 -400 C2 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 V vs MSE

Figure 2-5 CV response of a Ru shot in a 2 mM CuSO4 solution + 0.5 M H2SO4.

vertex is extended to more negative values viz. the behavior observed for the peak “A2”, reaching an almost constant height when scanned beyond -0.4V.

The sharp increase in anodic current with increased cathodic potential related to the peak “A1” is characteristic of bulk electrolysis of the deposited metal, or over-potential deposition (OPD) desorption peak. A more detailed look at the potential region of “A2”, unveils a well-defined diffusion limited peak, with a maximum ca. -0.26 V (figure 2-6).

100.0

50.0

) 0.0 2 A/cm

µ -50.0

-100.0 Current ( Current

-150.0

-200.0 -0.4 -0.2 0.0 0.2 V vs. MSE

Figure 2-6 Potential region of peak “A1”, obtained in a 2 mM CuSO4+0.5M H2SO4

solution on a Ru electrode.

A single crystal Ru (0001) face accommodates ca. 1.65X1015 atoms/cm2.

Nonetheless, assuming that a monolayer of copper, undergoes an epitaxial

growth following the hexagonal structure of the ruthenium substrate at a coverage comprised of 1 X 1015 atoms/cm2 for a polycrystalline ruthenium

surface[39, 40], the copper coverage can be calculated from the measured

charge under the anodic UPD peak, Avogadro’s number (6.022X1023 atoms/mol)

and faraday’s law [41]:

Q = nFN (1)

Where, Q is the measured charge, (C), n is the number of electrons transferred,

F is the Faraday’s constant (assumed as 96484.6 C/mol), and N the number of moles involved in the electrochemical process.

The coverage for the peak in figure 2.6 is close to that of a monolayer

(0.84). Thus assignment of this peak to a UPD process is made. The change in free energy between underpotentially deposited copper and its ruthenium substrate can be calculated from equation (2) [32].

∆ = ∆ G nF E p (2)

Where ∆Ep is the potential difference between bulk and UPD anodic peaks n is

the number of electrons transferred, and F is Faraday’s constant.

The ruthenium electrode scanned to bulk cathodic vertex (-0.65 V), shows

a ∆Ep value of 120 mV. This is an indication of the stronger binding between

copper and ruthenium, as compared to copper deposited into its own lattice (bulk

deposition).

The complete coverage (monolayers) of the copper UPD on ruthenium peak, as function of the cathodic peak potential, is presented in table 2-1, as well as free energy values obtained form OPD-UPD peak to peak separation.

Table 2-1 Thermodynamic parameters for Cu UPD on Ruthenium.

Cathodic Monolayers ∆Ep/V Charge /C ∆G/kJ/mol Vertex/V (ML) -0.35 2.39E-05 0.13 -0.40 1.52E-04 0.84 -0.45 2.12E-04 1.17 -0.50 0.17 1.86E-04 1.03 33.00 -0.55 0.15 1.74E-04 0.96 29.14 -0.65 0.12 1.85E-04 1.02 23.35

An increase in the concentration of copper ions available in solution

shifted the potential more positively, as expected for a nernstian behavior. Figure

2-7 shows the CV for a ruthenium electrode in a 10 mM CuSO4 + 0.5 M H2SO4 solution. Although obscured by the bulk desorption peak of copper, the UPD desorption peak, maintained almost a similar current height as in a 2 mM Cu2+ concentration solution, indicating similar coverage (figure2-8) , and its peak potentials shifted more positively, (ca. 22 mV).

The nature of the surface limited growth of the UPD process is demonstrated by the constant current peak height at the higher copper concentration, indicating a constant copper coverage at this potential.

8.0 10 mM CuSO4 + 0.5 M H SO 2 4

6.0 ) 2 4.0

2.0 Current (mA/cm Current 0.0

-2.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 V vs MSE

Figure 2-7 CV of ruthenium electrode in a higher copper concentration solution.

50.0

0.0 ) 2 A/cm

µ -50.0

-100.0 Current (

-150.0

-0.4 -0.2 0.0 0.2 V vs. MSE

Figure 2-8 Underpotential deposition of copper from a 10 mM CuSO4+0.5M

H2SO4 solution on a Ru electrode.

2.3.2 Copper Deposition on Ruthenium Oxide

Chronoamperometry (see: chapter 1) was used in order to attain oxidation of the Ru substrate. The technique performed involved a potential step program designed to hold the ruthenium electrode at a polarization potential of 0.85 V vs.

MSE, for three minutes. Copious bubbles ascribed to the presence of oxygen evolution were observed during the potential step, and a steep change in color of

the ruthenium electrode, from metallic bright to blue-black color, could be noted

on its surface (figure 2-9), from the optical microscope pictures, taken at 5X magnification. The OCP value increased to 0.3 V, indicating a change in the electrochemical state of the obtained surface oxide.

A B

Figure 2-9 A) Ru surface, B) oxidized Ru surface (both 5x).

A mixed proton-electron conductor, ruthenium oxides obtained by

electrochemical means in aqueous solutions are known to be of the hydrated

form ruthenium oxides, or ruthenium hydroxide [37, 41-43]. The presence of a

larger surface area of hydrous ruthenium oxide is deduced by the increase in the

overall capacitance of the electrode, which can be observed from the CV

recorded in 0.5 M H2SO4 (figure 2-10).

300

200

100 )

2 0

-100 A/cm µ -200

-300 Current ( -400

-500

-0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 V vs MSE

Figure 2-10 Background of oxidized Ru substrate in 0.5 MH2SO4 solution.

Upon potential cycling in the 2 mM CuSO4 +0.5M H2SO4, two anodic desorption peaks are also observed, one at -0.4 V, and the other close to -0.24 V

(figure 2-11). Expanding the cathodic vertex to negative potentials (-0.65 V) demonstrated the presence of a constant peak height of the second anodic peak.

This behavior is similar to the electrochemical behavior of the UPD desorption peak present in ruthenium. The change in coverage, as well as the

thermodynamic relation between the free energy and the OPD-UPD peak to peak

separation is summarized in the following table (table 2.2)

Table 2-2 Thermodynamic parameters for Cu deposition on ruthenium oxide. Cathodic Monolayers Vertex/V ∆Ep/V Charge /C (ML) ∆G/kJ/mol

-0.35 6.65E-05 1.10 -0.4 7.18E-05 1.20 -0.45 8.34E-05 1.38 -0.5 0.206 8.80E-05 1.46 39.76 -0.55 0.204 1.00E-04 1.66 39.37 -0.65 0.177 1.06E-04 1.67 37.73

The peak to peak separation has increased to a value higher than 170

mV. Measurement of the peak to peak separation for the OPD-UPD diffusion

limited processes indicates that the free energy has increased to 14 kJ/mol. This

is due to the stronger affinity of the deposited copper layer to the oxide covered ruthenium substrate than to the bare ruthenium electrode. Usually, the presence of underpotential deposition of a foreign metal adlayer onto a different substrate is explained in the terms of the difference in the workfunction (φ) of the deposited metal [44]. That is; the metal with the lower workfunction number will deposit underpotentially onto a metal with higher workfunction number [45].

In this case, a 2D structure of copper will be formed on an oxidized ruthenium substrate, (higher workfunction value than ruthenium).

A1 Vertex 1 (-0.35 V) 3000 Vertex 2 (-0.4 V) Vertex 3 (-0.45 V) Vertex 4 (-0.5 V) ) 2 2000 Vertex 5 (-0.55 V) Vertex 6 (-0.65 V) A/cm µ 1000 A2

0 Current Density ( Density Current -1000 C2 C1 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 V vs MSE

Figure 2-11 Ruthenium oxide CV response in a 2 mM CuSO4+0.5MH2SO4

solution.

Scan rate dependant studies performed up to 800 mV/s, showed visible

UPD adsorption and desorption peaks and provided additional confirmation of the strong affinity of the sub-monolayer layer copper deposited on top of the ruthenium oxide substrate.

X-ray photoelectron spectroscopy (XPS, see: chapter 1) characterization of a ruthenium oxide electrode, scanned to the UPD potential of -0.45, confirmed the presence of Cu on both ruthenium, and ruthenium oxide electrodes.

The electrodes were analyzed ex-situ after a short time, trying to avoid

any oxidation of the copper layer caused be the atmosphere [32].

40.0

20.0

0.0 A) µ -20.0

Current ( Current -40.0

-60.0

-80.0 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 V vs. MSE

Figure 2-12 CV in 2 mM CuSO4+0.5MH2SO4 solution of a thermally produced

ruthenium oxide.

It is important to remark that the presence of Cu UPD has been observed

only on the electrochemically formed ruthenium oxide; cyclic voltammetry with

copper containing solutions of the same concentration (2 mM), performed on

thermally annealed ruthenium shots under O2, revealed the absence of UPD

cathodic and anodic peaks. Thus, the presence of the hydrated oxide layer is

indicated as an essential requirement for copper UPD deposition on ruthenium

oxide (figure 2-12).

2000

5 seconds oxidation 30 second oxidation 1500 3 minutes oxidation ) 2 1000 A/cm µ 500

Current ( 0

-500

-0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 V vs. MSE

Figure 2-13 Effect of different oxidation times of an oxidized electrode in the

2 mM CuSO4+0.5MH2SO4 solution.

Thermodynamically, the energy required to attain such a behavior is dominated by the polarization potential of the electrode, not by the oxidation time.

This was confirmed by experiments and can be observed by looking at the CV of the ruthenium oxide prepared by doing a potential step to 0.85 V for different

times (5s, 30s and 3 minutes) in figure 2-13. The current vs. potential profile for

the short oxidation pulse (5s.) is similar to that of the longer (i.e. thicker) hydrous

ruthenium oxide, obtained at 3 minutes polarization potential, at the same

potential value.

2.4 Ongoing research

Characterization of Ru films on different electrodes has been a continuous

endeavor. With the aim of attaining a better understanding of the

electrochemistry of copper electroplating on ruthenium substrates, CVs have

been performed on prepared ruthenium films. The used silicon substrates were

hydrogen terminated by HF etching, following the standard (RCA) cleaning

procedure, to assure cleanness of the surface. Figure 2-14 shows the result of

cyclic voltammetry of a Cu containing solution in a ruthenium film deposition on a

silicon wafer via magnetron sputtering at 100 Watts for 2 hr. (120 nm thickness)

under an Ar plasma pressure of 20 mTorr. There is no visible differentiation

between bulk and UPD desorption peaks. Thus an effect on the electrochemical

behavior of the copper deposition process due to morphological effects on the

ruthenium sputtered film’s surface is suspected.

200

100

0 A) µ -100

Current ( -200

-300

-400 -0.8 -0.6 -0.4 -0.2 0.0 0.2 V vs. MSE

Figure 2-14 CV profile of a Ru film deposited on a Si wafer substrate in a 2 mM

CuSO4+0.5MH2SO4 solution.

Experiments were also performed with EQCM crystals, deposited via vacuum sputtering with a ruthenium film of ca. 90 nm thicknesses (2-15). Un- bonded and un-mounted crystals, of 8.0 Mhz base frequency, with no underlying adhesive metal, were employed, in order to avoid any contamination or substrate effect on the electrochemical behavior of the system.

) 2 2000 A/cm

µ 1000 0 -1000 -2000

-0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 Current ( Density 400 300 200 100 Mass (ng)

∆ 0

-100 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 V vs. MSE

Figure 2-15 EQCM response of a Ru electrode in a 2 mM CuSO4+0.5M H2SO4

solution.

This has also been observed at higher copper concentrations cyclic voltammetry experiments on Ru EQCM crystals (figure 2-16). In a 10 mM CuSO4

+ 0.5 M H2SO4 solution potential cycling shows two distinct desorption peaks

(figure 2-17). However, the mass change associated with the desorption peak at

-0.2 V (44 ML) indicates bulk desorption process, and it is comparable to the

coverage obtained by charge measurement (40 ML), therefore ruling out the sole presence of UPD deposition of copper.

) 2 3000 2000 A/cm µ 1000 0 -1000 -2000 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 Current Density ( 600 500 400 300 200 100 Mass (ng) ∆ 0 -100 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 V vs. MSE

Figure 2-16 EQCM response of Ru in a 5 mM CuSO4 + 0.5 M H2SO4 solution.

Since the underpotential deposition of metal adlayers is a process strongly influenced by the chemical and physical structure of the substrate metal, future work can focus on the study of the morphology of the films and its relationship with UPD behavior.

) 2 4000 A/cm

µ 2000

-2000

-4000 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 Current Density ( Density Current 1600 1200 800 400 Mass (ng)

∆ 0

-0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 V vs MSE

Figure 2-17 EQCM response of Ru in a 10 mM CuSO4 + 0.5 M H2SO4 solution.

2.4 Conclusion

The electrochemical characterization of copper underpotential deposition

on ruthenium and ruthenium oxide electrodes by means of electrochemical methods are presented. Spectroscopic (XRD, XPS) and gravimetric (EQCM)

techniques helped characterized different surface states of the presented

ruthenium electrodes.

Distinctive diffusion limited peaks at potentials more positive than the nernstian equilibrium potential are observed in both ruthenium and ruthenium oxide. Increasing the negative potential vertex to -0.65 produced a surface coverage of ca. 1 ML. Annealing of Ru electrodes at high temperature suggested the requirement of a hydrous ruthenium layer for the presence of UPD behavior.

The technological aspects of underpotential deposition are discussed, highlighting the use of ruthenium oxide surface, as a material for applications in the microelectronics industry.

2.5. References

1. Seddon, E.A. and Seddon, K.R. The Chemistry of Ruthenium. in Topics in

Inorganic and General Chemistry, ed. R.J.H. Clark. Vol. 19. 1955,

Amsterdam: Elsevier Science Publishers S.V. 1373.

2. Bracchini, C., et al., Catalysis Today, 2000. 55: p. 45-49.

3. Wang, J.X., et al., Journal of Physical Chemistry B, 2001. 2001: p. 2809-

2814.

4. Lin, W.F., P.A. Christensen, and A. Hamnet, Journal of Physical

Chemistry B, 2000. 104: p. 6642-6652.

5. Waszczuk, P., et al., Electrochimica Acta, 2002. 47: p. 3637-3652.

6. Waszczuk, P., et al., Journal of Catalysis, 2001. 203: p. 1-6.

7. Wieckowski, A. and Kuk, S.T., Journal of Power Sources, 2005. 141: p. 1-

8. Williams, J.O. and Mahmood, T. Application of Surface Sciences, 1980. 6.

9. Trasatti, S. and Buzzanca, G. Journal of Electroanalytical Chemistry and

Interfacial Electrochemistry, 1971. 29(1971): p. A1-A5.

10. Rao, C.N.R. and B. Raveau, Transition Metal Oxides Structure, Properties

and Synthesis of Ceramic Oxides. 2nd. ed. 1998, New York: Wiley-VCH.

11. Trasatti, S., Electrochimica Acta, 1984. 29(11): p. 1503-1512.

12. Kim, H.-K., et al., Thin Solid Films, 2005. 475(54-57).

13. Balek, V., et al., Colloidal and Interface Science, 2003. 260: p. 70-74.

14. Long, J., W., et al., Langmuir, 1999. 15: p. 780-785.

15. Kim, H.W. and C.-J. Kang, Microelectronic Engineering, 2003. 69: p. 89-

96.

16. Giauque, P.H. and Nicolet, M.-A., Journal of Applied Physics, 2003. 93(8):

p. 4576-4583.

17. Burke, L.D. and Healy, J.F., Journal of Electroanalytical Chemistry, 1981.

124: p. 327-332.

18. Chang, C.C. and Wen, T.C., Journal of applied electrochemistry, 1997. 27:

p. 353-363.

19. Zhong, H., Ru-based gate electrodes for advanced dual-metal gate CMOS

devices., in Analytical Chemistry. 2001, North Carolina State University:

Raleigh. p. 257.

20. Herrero, E., Buller, L.J. and Abruña, H.D. , Chemical Reviews, 2001.

101(7): p. 1897-1930.

21. Garcia, S., et al., Electrochimica Acta, 1998. 43(19-20): p. 3007-3019.

22. Mendoza-Huizar, L.H., Robles, J. and Palomar, P., Journal of

Electroanalytical Chemistry, 2003. 545: p. 39-45.

23. Rooryck, V., et al., Journal of Electroanalytical Chemistry, 2000. 482: p.

93-01.

24. Swathirajan, S. and Bruckenstein, S., Electrochimica Acta, 1983. 28(7): p.

865-877.

25. Mendez, P.F., et al., Electrochimica Acta, 2004. In Press.

26. Michalitsch, R., Palmer, B.J. and Laibinis, P.E. , Langmuir, 2000. 16: p.

6533-6540.

27. Krznaric, D. and Goricnik, T., Langmuir, 2001. 17: p. 4347-4351.

28. Santos, M.C. and Machado, S.A.S., Journal of Electroanalytical

Chemistry, 2004. 567: p. 203-210.

29. Santos, M.C. and Machado, S.A.S., Electrochimica Acta, 2004. In Press.

30. Nicic, I., et al., Journal of Physical Chemistry B, 2002. 106: p. 12247-

12252.

31. Chyan, O.A., Tiruchirapalli N.; Ponnuswamy, Thomas., Journal of the

Electrochemical Society, 2003. 150(5): p. C347-C350.

32. Zhang, Y., et al., Electrochemical and Solid-State Letters, 2004. 7(9): p.

C107-C110.

33. Cullity, B.D., Elements of X-Ray Diffraction. 3rd ed. 2001, Upper Saddle

River: Prentice-Hall.

34. Angerstein-Kozlowska, H., et al., Journal of Electroanalytical Chemistry,

1979. 100: p. 417-446.

35. Hadzi-Jordanov, S., Angerstein-Kozlowska, H. and Conway, B.E.,

Electroanalytical Chemistry and Interfacial Electrochemistry, 1975. 60: p.

359-362.

36. Vukovic, M., Valla, T. and Milun, M. Journal of Electroanalytical Chemistry,

1993. 356: p. 81-90.

37. Horvat-Radosevic, V., Kvastek, K. and Vukovic, M. Journal of

Electroanalytical Chemistry, 1999. 463: p. 29-44.

38. Michell, D.R.,and Woods, R., Journal of Electroanalytical Chemistry, 1978.

89: p. 11-27.

39. Quiroz, M.A. and Meas, Y., Journal of Electroanalytical Chemistry, 1983.

157: p. 165-174.

40. Gonzales-Tejera, M.J. and Nguyen Van Huong, C., Journal of

Electroanalytical Chemistry, 1988. 244: p. 249-259.

41. Conway, B.E., Electrochemical Supercapacitors. First. ed. 1999, New

York: Kluwer-Plenum.

42. Prakash, J. and Joachin, H., Electrochimica Acta, 2000. 45: p. 2289-2296.

43. Vukovic, M. and Dunja, H., Journal of Electroanalytical Chemistry, 1999.

474: p. 167-173.

44. Chrzanowski, W. and Wieckowski, A., Langmuir, 1997. 13: p. 5974-5978.

45. Hartmann, A.J., et al., Applied Physics A, 2000. 70(239-242).

CHAPTER 3

FORMATION OF BASIC COPPER SULFATES

3.1 Introduction

The formation of “patinas” or corrosive protective layers [1] is a common

process known to happen after extensive atmospheric exposure of bronzes or

other metallic alloys, usually present in urban areas and historical places, or in

the form of statues, sculptures, façades, and other building objects [2, 3].

Atmospheric corrosion is fostered by the adsorption of water, and in the case of copper, or copper containing alloys, the incorporation of impurities, mainly sulfates, leads to the formation of basic copper sulfates [4, 5], like brochantite (Cu4(OH)6SO4), antlerite (Cu3(OH)4SO4) and posnjakite

. (Cu4(OH)6SO4 2H2O).

The chemistry of formation of the basic copper sulfates has been reviewed

by different authors [4, 6-8]. Experimentally, the processes in which salts are

obtained vary from titration [4] to aging to naturally produced patina examination.

The general conclusion is that a cuprous oxide (Cu2O) or cuprite is the initial

species present in the layer [1].

The study of copper plating is of technological relevance, since its wide

use in power line connections, common household electrical appliances and

more recently, the electrochemical deposition of copper as an important process

for the IC industry [9]. It has a high conductivity (1.7 µΩ.cm) and it has completely

replaced aluminum (Al) as the preferred interconnects material used in microchip manufacturing [10, 11] . At low pH values, copper plating is the most common reaction [9, 11-14]. Electroplating at pH values higher than 6.8, in which Cu2O plating [15-19] is the favored process [8] has also been widely investigated.

Besides the above two common areas of investigation, little is known in the literature about copper electrodeposition in different pH values, hence the importance of this study.

In this chapter, the mechanism of formation of posnjakite, observed by the formation of a precipitate at the end of a cyclic voltammetric cycle is confirmed by the results obtained with the use of the electrochemical quartz crystal microbalance (EQCM), x-ray photoelectron spectroscopy (XPS), x-ray diffraction

(XRD) and optical microscopies.

3.2 Experimental

A 400 series model electrochemical analyzer from CH InstrumentsTM*, was used for the quartz crystal microbalance experimentation. The time resolved instrument can simultaneously read the frequency change of the crystal along

with the surface’s current change. We used platinum coated crystals. In this way,

the instrument is an effective tool for in-situ monitoring of current-mass related

changes of the electrode’s surface. The oscillating frequency of the crystal is

subtracted from a reference oscillator, vibrating at 7.995 MHz, so the difference

is reported. This is called reciprocal method, and it gives the advantage of

* CH Instruments Inc. www.chinstruments.com

precise measurement at a fast rate. In this configuration, the crystal itself,

bonded to a metallic mount, is seated on the bottom of a Teflon®† cell, with its

2 surface facing the body of the solution. The exposed electrode area is 0.206 cm ,

and it is used as the working electrode in a conventional three way electrode cell.

The platinum (Pt) covered electrodes had a thickness of 1000 Ǻ, with a

titanium underlying layer, 100 Ǻ thick, deposited on an AT cut quartz crystal

obtained from International Crystal Manufacturing‡ (ICM®). A Pt rod

(diameter=0.00508 cm) was submerged in the solution’s body to act as a counter

electrode. A mercury/mercury sulfate (Hg/HgSO4) or MSE reference electrode

was employed for all the measurements.

By means of the Sauerbrey’s equation [20], the change in frequency is

related to the change in mass:

∆f = −C∆m (1)

The constant C is determined by various properties of the crystal [21, 22] (i.e. thickness, fundamental frequency, density, etc...). In the case where crystals with a fundamental frequency of oscillation of 7.995 the value of 1.4156x10-9 ng per

hertz change is used, as it is the case of the presented study.

In order to get the cyclic voltammetry profiles (CV), the surface of the electrode was scanned from a value close to the open circuit potential (OCP) of the Pt surface (0.1 V) to the copper plating region (between -.65 and -0.7 V vs.

† E.I. du Pont de Nemours and Company, www.teflon.com ‡ International Crystal Manufacturing, www.icmfg.com

MSE). Potassium sulfate, obtained from Alfa Aesar®§ and copper sulfate

pentahydrate 99.999% pure, from Aldrich®** were used in the solutions, all of

them prepared with ultra pure Millipore®†† water (18.0 MΩ). X-ray diffraction

characterization was done on a Bragg-Brentano Siemens®‡‡ D-500

diffractometer, with a Cu Kα X-ray as source, controlled by a personal computer.

The experiments were run in the 2-68° 2θ angle range, with a step size of

0.05° and 1 second dwell time. XPS data were collected using a VG ESCALAB

MKIITM§§ spectrometer with an Mg Ka X-ray as the excitation source. Optical

images were recorded using a microscope (Nikon® ME600L***)

Acquisition of optical images from the electrode’s surface demanded the

modification of the EQCM crystal mount available from the manufacturer. This

mount impeded the use of the optical microscope’s high magnification objectives,

as it collided against the metallic edge of the lens. In order to overcome this

limitation, two modifications were made; first, the crystal was removed from its original contact pins. For this step, unbounded crystals were purchased so there

was no originally present epoxy material to me removed. The crystal was then

bonded to Teflon® coated stainless steel wire, which was used as extension.

The crystal was protected with a Teflon® tape mask along the process so

its surface remained unscratched. In order to obtain an electrical connection

§ Alfa Aesar, www.alfa.com ** Sigma-Aldrich Corporation, www..sigma-aldrich.com †† Millipore Corporation, www.millipore.com ‡‡ Siemens A.G., www.siemens.com §§ Thermo Electron Corporation, www.thermo.com *** Nikon Corporation, www.nikon.com

between the electrodes deposited on top of the quartz crystal and the stainless steel wire, Tra-Duct 29202, from Tra-Con®††† silver epoxy was used. Curing was

achieved by keeping it for 5 minutes at 160 ˚C. The crystal was then placed on

top of one of the o-rings used in the Teflon® cell, so as to align the central

(electrode) area with the center of the cell.

After this was accomplished, extension wires were fixed into position by using a mass of silicon glue which molded itself inside the cell’s confinement space (Figure 3-1).This allowed the microscope’s lens to be moved freely along the way, without displacing the electrode from the objective.

Electrode Cell Base

Extension

Figure 3-1 Adaptation to the EQCM’s mount for microscope picture.

††† National Starch and Chemicals, www.tra-con.com

3.3 Results and Discussion

Copper electroplating experiments have been performed in Ruthenium and Ruthenium oxide surfaces [23, 24]. In order to systematically study its electroplating behavior at different values of pH, the acidity of the plating bath was increase from its common value of zero (0.5 M H2SO4) to higher pH values.

An electrochemically clean platinum surface was obtained by multiple

cycling, between -0.695V and 0.8V vs. MSE (figure 3-2).

A ) 2 50.0 0.0 Amp/cm

µ -50.0 -100.0 -150.0 -200.0

20.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0

Current Density ( 15.0 10.0 5.0 0.0 B

Mass (ng) -5.0 ∆ -10.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 V vs MSE

Figure 3-2 Current/mass profiles of a Pt electrode in 0.5M H2SO4 solution; A)

current density, B) mass change.

As a way to study this phenomenon by means of the EQCM, parallel experiments were performed using the platinum coated quartz crystal wafers.

Plating baths with pH values of 0, 2, 3 and 4 were prepared by adjusting their acidity with a 3M NaOH solution.

Scanning towards the cathodic apex, towards the hydrogen adsorption and evolution region (-0.4V), a mass loss is observed via EQCM measurement.

This indicates the increasing coverage of a lighter Hydrogen layer. Upon scan reversal, a mass gain is noted, indicating the absorption of water and/or anion molecules, which occupy the sites previously covered by hydrogen. As the potential reaches the anodic vertex, the onset of oxide formation can be observed (0.1V), with a mass gain associated. The return to the original electrode’s mass value, when the scanning returned to the original potential value demonstrated a small amount of mass loss occurring due to cycling between this potential limits [25, 26].

Following the electrochemical cleaning, copper was plated at different pH values, from a 2 mM CuSO4 + 0.1M K2SO4 solution. The CV’s and corresponding

mass responses are shown in figure 3-3.

The positive current or anodic peak, located at the reverse scan, is related

to copper desorption, or copper stripping. The diffusion limited peaks have

potentials of -0.38 to -0.35, 0.34, -0.33 V. along with the increase in pH. The

oxygen reduction peak present at 0V at pH 0 and 2, shifts anodically to -014 and

-0.2 at higher (3, 4) pH.

From the EQCM response, we can see distinct features along the potential scan. Starting from the OCP value of 0.1V, going towards the negative side, the mass accumulation sharply increases after the reaction:

Cu 2+ + 2e − → Cu 0 (2)

has reached its nernstian value of -0.45V vs. MSE. Mass increase continues

even after the potential sweep is reversed, with the negative current indicating a

continuum in the reduction reaction [27]. ) 2 2500 pH 0 2000 pH 2 1500 Amp/cm pH 3 µ 1000 pH 4 500 0 -500 -1000 -0.8 -0.6 -0.4 -0.2 0.0 0.2 600 Current Density ( 500 pH 0 400 pH 2 300 pH 3 200 pH 4

Mass (ng) 100 ∆ 0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 V vs MSE

Figure 3-3 Current/mass profiles of copper deposition from 2 mM CuSO4 + 0.1M

K2SO4 adjusted at different pH values.

The onset of the single anodic peak present in all scans marks the initiation of the reverse reaction, as noted also from the mass decrease. The onset of this peaks moves anodically with increasing pH value. After the anodic current peak is over, the slope of the mass decrease is reduced, to finally end up in a mass value similar to the initial one, indicating that at the pH values below 5, the deposition of copper is a highly reversible process. The associated charge of the anodic peak corresponds well with the value of mass loss, for its potential range. The average efficiency of the deposition process, as a measure of the cathodic charge, versus the mass gain is ca. 60% for all the pH values presented. Nonetheless, the efficiency of amount of mass loss, as read from the

EQCM, compared to the mass loss obtained from integration of the anodic desorption peak, is above 90%.

Figure 3-4 shows the results of a typical CV of a 2mM concentration copper plating solution at pH 5. New features are present in the both the cathodic and anodic scans. Peak labeled “C1” or a “pre-peak” to the copper plating peak

“C2” appears. The peak are related to the anodic peaks “A1” (copper stripping) and “A2”. This is confirmed by the multiple scanning experiments in this solution at the same potential limits. It is interesting to note that the large mass accumulation present at the end of each cycle, triggered by the appearance of

“A2”, is not present at other pH values. This mass accumulation eliminates the presence of active centers in the Pt surface for under potential deposition of

copper, like it has been observed at lower pH values [28]. On the reverse scan, a proportional mass desorption or “dumping” appears at potentials close to -0.42V.

The area of the peak “C1” increases with the number of cycles, and

eventually obscures the peak “C2”. ) 2 1600 A2 1200 A/cm

µ 800 400 A1 0 -400 C2 -800 C1 -0.8 -0.6 -0.4 -0.2 0.0 0.2 Current Density ( 4 3200 3 2400 2 1 1600 800 Mass (ng) ∆ 0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 V vs. MSE

Figure 3-4 Multiple scanning experiments at pH=5, 2mMCuSO4+0.1MK2SO4,

Upon changing the concentration of the Cu2+ ions present in solution, to

5mM and 10 mM, there’s an increase in the total mass accumulated at the

surface up to the peak “A1”, while the amount of mass at “A2” doesn’t increase as should correspond to the increase in copper from the solution (figure 3-5).

) 8000 2 2mM 6000 5mM A/cm

µ 10mM 4000 2000 0 -2000 Current Density ( -4000 -0.8 -0.6 -0.4 -0.2 0.0 0.2 2500

2000

1500

1000

mass (ng) mass 500 ∆ 0

-0.8 -0.6 -0.4 -0.2 0.0 0.2 V Vs MSE

Figure 3-5 EQCM profile of copper deposition in a Pt electrode at different concentrations.

The ratio of cathodic versus anodic charge transfer throughout a complete cycle is always greater than the unity.

Potential holding at the end of the cycle for different times didn’t show any

effect on its dissolution or retention of the precipitate formed. Once the CV

stopped, the mass of the electrode surface, as recorded by the frequency

change, slowly diminished towards the original electrode’s weight, as seen by an

increase in frequency, indicating dissolution of the precipitate formed.

3.3.1. Characterization of the Process at Different Stages

XRD and XPS techniques were used to characterize the mechanism of

copper plating from a 2mM solution, at different stages of the CV performed on

Ru electrodes. These characterization methods are still useful, since the features

appearing at the CV of the Pt surface were reproduced in the Ru electrode as

well. The XRD diffractogram (figure 3-6) shows the planes characteristic of

. posnjakite [1, 29] Cu4SO4(OH)6 H2O, a form of basic copper sulfate (PDF#43-

0670, 83-1410). Although this is not the most stable form of the salt [30] it is noted as a hydrated form of brochantite. This undergoes dehydration, and a more stable form brochantite is formed [31]. The latter is known to be the product of corrosion in statues, roof tops and bronzes present in urban areas. Studies of this salt in marine and natural environments have also been reported [32].

800 P (-223) 700 600

500 400 300

Intensity (A.U) Intensity 200 P (001) 100 P (002) Ru (100) P (431) 0 10 20 30 40 50 60 70 2θ

Figure 3-6 XRD spectra of the precipitate formed, as obtained form the top of a

Ru electrode P: posnjakite.

In order to have a visual survey of the surface of the electrode at the different stages of the cyclic voltammetric profile, images from holding at different points were taken. Figure 3-7 shows the resulted optical pictures at 100X magnification taken at different stages of the CV profile of a 10 mM CuSO4 solution with 10 mM Cu2+ concentration. The images are taken in the same place

A ) 2 8000 A1 6000 A/cm

µ 4000 A2 Apex 2000 Val ley B 0 Vertex -2000 C1 C2 -4000 -0.8 -0.6 -0.4 -0.2 0.0 0.2 Current Density ( 2500 A1 C 2000 Vertex A2 Apex 1500 Vall ey 1000 500 Mass (ng) C2 ∆ 0 C1 D -0.8 -0.6 -0.4 -0.2 0.0 0.2 V vs MSE A) Clean Electrode B) 3 min. holding at “C1” C) 3 min. holding at “C1”, stop at Vertex E D) 3 min. holding at “C1”, stop at Valley E) 3 min. holding at “C1”, stop at “A2” Apex

Figure 3-7 Optical pictures of the Pt electrode in a 10 mM Cu2+ concentration

solution, taken at different stages of the potential sweep.

of the electrode’s surface each time, and were obtained after holding at “C1” peak (-0.4 V) for three minutes, and halting the sweep at four different positions, labeled B, C, D, and E, and corresponding to holding at “C1”, the vertex of the

cathodic region (-0.7V), the plateau or valley between the first and second anodic

peaks, and finally, at the apex of the decreased current height peak 2,

respectively.

The image corresponding to the apex of peak ”A2” is displayed in dark

field, to show the features of the blue-white precipitate characterized as

posnjakite [33].

At “C1”, we can observe the number of reddish particles, characteristic of

Cu2O, a brick red electron deficient p-type semiconductor [34-36]. SEM (figure 3-

8) pictures of the film deposited on ruthenium wafer slides unveiled a triangular

and squared shape of the crystals ca. 200 nm wide. This observation of color and

shape, are in accordance to those of Lee et al. [37], who reported the formation

of Cu2O(111) by electroplating from a pH 4.7 solution. Another characteristic of

the deposited particles at “C1” is that their size did not increase with the

concentration of Cu2+ in solution. Only their density (number of particles at a

given surface area) increases.

Figure 3-8 SEM image of Cu2O crystals on a Ru coated Si wafer.

X-ray photoelectron spectroscopy of the Ru electrode held for three

minutes at different potentials is presented in figure 3-9, where the energy region

for electrons emitted from the Cu 2p orbital is shown [38]. This confirms that at

peak “C2”, Cu is electrochemically deposited (A).

Indeed the value of the peak at 932.32 eV corresponds to Cu 2p3/2 photoelectrons. When the electrode is held at C1 for three minutes (B), the peak has shifted to a higher binding energy (932.8 eV), confirming the formation of

Cu2O. Spectra collected from the precipitate indicates the presence of a satellite

952.21 932.32 60

55 941.62 ) 4

50 932.8

45 a

Intensity (A.U.*10 Intensity 40 b

934.29 c 35 970 960 950 940 930 920 Binding Energy (eV)

Figure 3-9 Cu 2p region of XPS spectrum for compounds formed in 2 mM CuSO4

+0.1 K2SO4 pH 5 solution a) holding at “C2” for 3 min; b) holding at “C1” for 3

min; c) “A2” or precipitate.

peak at 941.62 eV, characteristic of CuO corresponding to the formation of Cu2+ oxide and hydroxide species at the surface of the electrode [39]. A peak at

934.29 eV can be assigned to the presence of Cu2+ ions [40] that form upon

dehydration of posnjakite under vacuum. This is an indication of Cu species

under the chemical effect of its sulfate surroundings.

17.4

17.2 ) 4

17.0

16.8 Intensity (A.U.*10

16.6

175 170 165 160 155 150 Binding Energy (eV)

Figure 3-10 S 2p XPS spectrum of precipitate.

Oxygen 1s region spectra, obtained by “C1” holding, with peaks at

530.2eV and 532.2eV confirmed the presence of two chemical states for oxygen species that can be ascribed to surface oxygen adsorbed from the environment, and oxygen present in cuprous oxide. Confirmation of the presence of sulfate, in

. the formation of posnjakite CuSO4(OH)6 H2O can be found in the spectra of the S

2p region of the electrode at the final “A2” state of the potential sweep, by a peak at 168.53eV (figure 3-10).

3.3.2 Effect of the Oxygen Reduction Reaction

The effect of oxygen removal by use of Ar purge was investigated in a set of experiments presented in figure 3-11. In these experiments, an aerated solution was exposed to a flow of Ar from the top of the electrochemical cell, and

CVs were recorded at subsequent time intervals of 30 minutes. By a change in the oxygen partial pressure gradient above the solution, solvated oxygen molecules gradually diffused away from it. The observed amount of mass formed on top of the electrode at the end of each scan decreased with each successive scan, that is, with the increased time of exposure to Ar gas.

The initial mass accumulation of 6ug, at “A2”, as observed from the in-situ

EQCM experiments, gradually reduces to less than 0.5µg. This is an indication of

the correlation of the peak “C1” with “A2” and it also indicates that the presence

of O2 can be the key component in the formation of the precipitate.

Indeed, through the effect of oxygen reduction, with an onset present at -

0.1 V, is present at “C1” potential (figure 3-12), has been known to have a

definite effect on the pH of the solution.

2000 ) 2 1600 1200 1st. CV A/cm

µ 2nd. CV 800 A1 3rd. CV 400 A2 4th. CV 0 -400 -800 C2 C1 Current Density ( -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 700

600 500 400 300 200 100

mass change(ng)mass 0 -100 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 V vs MSE

Figure 3-11 Cyclic voltammetry and EQCM corresponding to the CVs on Pt

electrode in 2 mM CuSO4 +0.1M K2SO4 pH 5 solutions with Ar purging.

0 ) 2

-10 A/cm µ N Purge -20 2 Air Exposed O Purge -30 2

-40 Current Density (

-50

-0.4 -0.2 0.0 0.2 0.4 0.6 0.8 V vs. MSE

Figure 3-12 Background of the Pt electrode in 0.1M K2SO4 pH=5 solution, with

different gas purging.

Oxygen reduction is a cathodic reaction fostering OH- production [4, 41]:

+ + − → − O2( g ) 2H 2O 4e 4OH (3)

In acidic solutions, reaction (2) is the thermodynamically favored reaction.

In order to attain formation of Cu2O [8, 42]:

2+ + + − → + 2Cu 2e 2OH Cu2O H 2O (4)

an increase in the pH values to values higher than 5 should be present. There has been different studies [7, 43-46] indicating that the pH value of a solution can

increase, from neutral to a value of 9 to 12, upon the effect of this reaction. Thus,

with the presence of the oxygen reduction reaction (ORR), an increase in the

local pH value, in a region distancing 25-30 µm from the surface of the electrode can be expected, to values close to pH=13, as reported in the literature [8, 47].

In the presented experiment, ORR gradually decreased as the dissolved

O2 was removed from the solution. Eventually, the trace amounts of oxygen were

not enough to cause an increase in the pH, thus preventing the electrochemical

processes at “C1” and “A2” from occurring. The effect of pH increase by the ORR

could be tested if the OH- ion could be maintained constant by the use of a

solution with a constant pH value.

Figure 3-13 shows the CV in a buffered 5mM CuSO4+0.1MK2SO4 solution,

adjusted at pH=5 by using acetic acid and sodium hydroxide. As expected, the

change in the local pH is not enough for the formation of diffusion limited peaks

“C1” or “A2”. Only current maximums “C2” and “A1” are present. Furthermore, no mass is found to precipitate in the surface of the electrode.

) 2 2500 5mM CuSO + 0.1 M K SO 4 2 4 2000 Buffered Solution A/cm

µ 1500 1000 500 0 -500 -1000 -0.8 -0.6 -0.4 -0.2 0.0 0.2 Current Density ( 500 400 300 200 100 Mass(ng) ∆ 0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 V vs MSE

Figure 3-13 CV of a Pt electrode in a 5 mM CuSO4+0.1 K2SO4 pH=5 buffer

solution.

XPS data (Cu 2p region) obtained from by holding the Ru electrode at

C1(-0.44V) in 2 mM CuSO4 +0.1M K2SO4 pH 5 solution with different gas purging confirmed an increased presence of Cu2O in oxygen purged solutions as

compared to non-purged solutions. The sample obtained during N2 flow

produced the least amount of oxide. In acidic solutions, the concentration

gradient from the bulk solution to the surface of the electrode causes a neutralization of the same, preventing Cu2O to form. This is why the formation of

this oxide is seldom reported [37] in acidic solutions. At pH=5 solutions, the

formation of Cu2O is promoted by the local increase in pH [45] caused by the

ORR. This reaction can also be accounted to the previous observation of the

cathodic versus anodic charge rate, being greater than the unity.

3.3.3. Mechanism of Formation

At this point, firm proposals of the electrochemical mechanism can be

asserted. At the potential point “C1”, formation of Cu2O occurs due to the

availability of Cu2+ species in solution, through equation 4, coupled with an

increase in pH through the ORR (equation 3).

Although it has been suggested that Cu(0) can be formed at “C2” by the

reduction of cuprite (Cu2O) or tenorite (CuO) by different authors [48, 49], this is not likely happening in our case, as an increase in the “C1” (copper oxide formation) should be accompanied by an increase in the peak at “C2”. In our case, a broader and bigger C1 peak virtually obscures the presence of the peak at “C2” in 2 mM Cu2+ solutions, and diminishes the peak height at higher

concentrations; therefore the reduction of Cu from copper ions available in

solution (equation 2) is the most likely process occurring at this stage.

One possible cause for the discrepancy from the results here obtained is

that previous studies focus on a copper electrode in basic solutions viz. Cu

plating on top of a platinum or ruthenium surface, therefore surface changes in a

foreign substrate (Pt/Ru) are not taken into account.

Upon sweep reversal, in the reverse anodic scan peaks, “A1” and “A2” are

observed (figure 3.4). Gewirth and his group [49] have studied the oxide

formation in basic solutions. With the aide of a silicon nitride tip, force microscopy

studies suggested the physisorption of hydroxide species in potentials before the

peak “A1”. Their work established that at “A1” the formation of insoluble Cu2O

2- (via Cu oxidation) in competition with soluble Cu(I) and Cu(II) species (Cu(I)2O2

2- and CuO2 (aq)) is reported:

Cu 0 + OH − → 1 Cu O + 1 H O + e − (5) 2 2 2 2

This implies that a mass increase may be anticipated in the potential

region corresponding to peak A1, due to the change in the formula weight of the

deposited species, Nonetheless, a mass loss, as read by the EQCM is present in

our experimental conditions. Table 3-1 shows the EQCM results obtained from

potential cycling at different Cu2+ ions concentration. To obtain the mass

efficiency, the total charge, measured as the area under each corresponding

peak, was used to obtain the theoretical number of moles for a two electron

process, from the anodic dissolution of copper, was assumed:

Cu 0 → Cu 2+ + 2e − (6)

Thus:

Mt - Theoretical Mass, as obtained from the number of moles, from Faraday’s

Law.

Ma – Actual mass read from the frequency change in the EQCM.

Ef - The efficiency of mass gain or loss: (Ma/Mt)* 100%

Table 3-1 Mass change efficiencies at anodic peaks “A1” and “A2”.

A1 A2

Mt/µg Ma/µg Ef/% Mt/µg Ma/µg Ef/%

2mM CuSO4 0.0308 0.0301 97.7 0.70 1.38 197

5mM CuSO4 0.713 0.582 81.5 0.49 1.56 320

10mM CuSO4 1.68 1.31 78.2 0.49 1.56 320

The value of dissociated mass and the related charge transfer are

comparable for the case of 2 mM CuSO4 solution. So the anodic peak current

“A1” is assigned to the oxidation of Cu species to Cu2+ ions. The difference of the

mass read versus the theoretical mass change is assigned to the presence of

insoluble species, like as reported in literature, and is due to the difference in pH

of our study. We suggested the difference between literature and our study was

caused by pH difference of the solution. It can also be noted (table 3-1) that there

is a marked decrease in the efficiency of this process, with increasing

concentration. It has been reported that Cu oxidation to Cu2+ ions follows a two

step process [50]:

Cu 0 → Cu1+ + e− (7)

Cu1+ → Cu 2+ + e− (8)

At lower concentration (2 mM) Cu ions can easily diffuse away from the electrode surface. As this concentration is increased, the amount of ions will be

increased proportionally, as related to the amount of Cu deposited in peak “C1”.

This will effectively limit the speed at which the oxidized ions diffuse away from

the surface, and the production Cu oxidation will be the Cu+ ion, favored

thermodynamically [50].

The Anodic peak “A2” is associated with the oxidation of Cu ions and

Cu2O to divalent species, Cu(OH)2 being the most probable one, as agreed in

curret literature [14].

+ + → 2+ + + − Cu2O 2H 2Cu H 2O 2e (9)

2+ + − → Cu 2OH Cu(OH)2 (10)

The source of the Cu2O particles comes from the previously discussed

reaction occurring at “C1”. Whenever there’s a high concentration of Cu2O species available on the surface, anodic peak A2 will be present, and it will increase proportionally with “C1” peak area (figure 3-4).

The need for Cu2O species for the formation of peak “A2” was confirmed

by the experiment shown in figure 3-14. Cu2O pre-loading, confirmed by optical microscopy, produced by holding the Pt electrode’s surface at C1 (-0.4V) for three minutes in a 10 mM [Cu2+] ph=5 solution, was oxidized slowly by cycling

through the “A2” potential region, without passing through the A1 peak potential,

in a 0.1 M K2SO4 solution. An observable peak with a potential located at the

potential corresponding to peak’s “A2” location (-0.18 V) is observed, triggering a

mass gain process through the rest of the cycling. As observed from XRD

characterization, posnjakite is formed on top of the electrode at the end of the

potential cycling (table 3.1).

) 2 100.0 80.0 A/cm

µ 60.0 40.0 20.0 0.0 -20.0 -40.0 -60.0 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 Current Density ( Density Current 1200 1000 800 600 400 200 Mass (ng)

∆ 0 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 V vs. MSE

Figure 3-14 Cycling of a Cu2O loaded platinum surface in the “A2” potential

region in 0.1M K2SO4 solution.

The mass accumulation via EQCM sensing at different copper ion concentrations confirms it. Therefore, it is proposed that at “A2”, precipitation of posnjakite resulted from an increase in pH of the solution that can be assigned to the anodic oxidation of cuprite, (equation 9), and the cathodic reduction of

oxygen [51].

Availability of sulfate and water molecules in the vicinity of the electrode surface [52] will lead to the following chemical reaction:

+ 2+ + 2− + → • 3Cu(OH )2 Cu SO4 H 2O Cu4SO4 (OH )6 H 2O (11)

The conditions present in the electrode surface had reproduced the

mechanism occurring in nature [4, 51]; upon formation of cuprite in the

atmosphere, the product of its oxidation Cu(OH)2, will react in the presence of trace amounts of sulfates in the available water (rain, moisture). So in our potential sweep experiments, the following consecutive conditions were met that led to the formation of artificial patinas: first, cuprite was formed at C1(-0.4V) in

cathodic range (equation 4). Then, Cu2+ ion species were made available

through the deposition process present at “A1” potential was formed when the

scanning through peak “A2” (equation 6). Oxygen reduction reaction, occurring on the electrode’s surface causing an increase in the available OH- ions leads to

the formation of copper hydroxide Cu(OH)2, on top of the Cu2O layer [53,48],

(equation 10). The precipitate observed at the end of the cycle consisted of

2+ 2- posnjakite formed by the chemical reaction between Cu(OH)2 and Cu /SO4

(equation 11) [54].

The theoretical mass change obtained in table 3.1 is the mass gain of

posnjakite based on the equation 11, the charge calculation using equation 9. In

three reported concentrations of copper (2 mM, 5 mM and 10 mM Cu2+), the

efficiency of posnjakite formation is higher than 100%. This can by explained by

the fact that in the regular CV, the copious amount of copper ions available,

obtained by scanning through peak “A1”, and from the diffusion of copper ions present in the solution, that will be readily available to favor the thermodynamic formation of posnjakite, thus causing an efficiency higher than the theoretical.

The fact that the formation of posnjakite does not increase proportionally

- with the increase in Cu ions concentrations can be expected, since the SO4 ions available in solution from the potassium, sulfate electrolyte K2SO4, and required for the formation of the basic copper sulfate, was kept at a constant 0.1M, pH=5

adjusted. The only increase in sulfate concentration came from the sulfate ions

associated with the copper sulfate pentahydrate salt. An important feature is that

the formation of posnjakite occurs only after peak potential A2 has been reached.

This is because Cu2O oxidation will contribute to the increase in pH of the area

(equation 9).

The importance of Cu2O formation can be depicted by the experiment

presented in figure 3.15. After holding for 3 minutes at “C1” potential, in a 5 mM

solution, after which it was evacuated from the cell. Electrolyte containing

solution 0.1 M K2SO4 was used to cover the surface of the electrode and

potentially stepped form the negative open circuit potential value of -0.34 V vs.

MSE, to the “A2” potential (-0.18 V vs. MSE) through the use of the chronoamperometry technique (CA), in the electrolyte solution. The results of this experiment are in agreement with the proposed mechanism.

) 2 3500 2800 A/cm µ 2100 1400 700 0

0 102030405060

Current Density ( 3500 2800 2100 1400 700 Mass (ng) Mass ∆ 0 0 102030405060 V vs. MSE

Figure 3-15 Potential stepping of a cuprite loaded Pt surface to A2 potential in

K2SO4 electrolyte.

3.3.4. Mass Dumping.

In mass change of EQCM data shown as figure 3-15, a sudden mass loss,

or “dumping” was observed at –0.42V in cathodic region starting from 2nd cycle.

This behavior is also seen in the scans performed in higher copper concentration solutions. A small change in the pH can cause a large effect in the change on the

Cu concentration; Fitzgerald [4] has concluded that this change in local pH,

accompanied by a change in sulfate ions concentration near the electrode surface can cause brochantite to reduce to tenorite. Therefore, two possibilities can led to the desorption; either a sudden dissociation of posnjakite, caused by the departure of surface species present in posnjakite (SO42-, Cu2+, OH-) in the observed potentials, or due to the reduction of posnjakite, which although it has been reported [53, 55], it would imply the reduction of copper oxide species [56,

57], which its formation is known to be highly irreversible [35].

3.4 Conclusion

In this paper, the study of copper electrodeposition on platinum in solutions adjusted to different pH values with 0.1M K2SO4 as supporting electrolyte is reported. In pH=5 solutions, the rise in local pH caused by the reduction of oxygen leads to the formation of posnjakite or basic copper sulfate. This behavior is also observed in solutions with higher copper concentration. The mechanism is understood with characterization made through the use of XPS, XRD and EQCM at different stages of the electrode reaction, and parallel experiments using Ru electrodes. The scope of the in-situ EQCM is proven as a powerful tool to explain this mechanism.

The effect of local pH increase caused by cuprous oxide oxidation and oxygen reduction is presented. It is also noted that the rate of cathodic to anodic rate is always grater than the unity. This emphasizes the role of the ORR reaction in the proposed mechanism. Since some of the OH- ions produced during the scan are neutralized back to water molecules and maintain the original

value of pH at the end of the scan, posnjakite is not observed in solutions with

basicity greater than pH=5.

3.5 References.

1. Jouen, S.J.,and Hannoyer,B ., Surface and Interface Analysis, 2000. 30:

p. 145-148.

2. Alvarez de Buergo, M. and Fort Gonzalez, R., Construction and Building

Materials, 2003. 17: p. 83-89.

3. Cicileo, G. and Rosales, Blanca M., Corrosion Science, 2004. 46(929-

953).

4. Fitzgerald, K.P.N. and Atrens, A., Corrosion Science, 1998. 40(12): p.

2029-2050.

5. Masamitsu, W.T. and Ichino, Toshihiro, Journal of the Electrochemical

Society, 2001. 148(12): p. B522-B528.

6. Schlesinger, R.K. and Bruns,M, Surface and Interface Analysis, 2000(30):

p. 135-139.

7. Schwartz, D.T.M. and Rolf H., Surface Science, 1991(248): p. 349-358.

8. Switzer, J.A.H., Chen-Jen, Huang, Ling-Yuang; Miller, Scott F.; Zhou,

Yanchun, Journal of Materials Research, 1998. 13(4): p. 909-916.

9. Gau, W.C.C.; Lin, Y. S.; Hu, J. C.; Chen, L. J.; Chang, C. Y.; Cheng, C. L.,

Journal of Vacuum Science & Technology A: Vacuum, Surfaces and

Films, 1999. 18(2): p. 656-660.

10. Wu, Y.-L. and Hwang, Y.-C., Thin Solid Fims, 2004. 461: p. 294-300.

11. Grujicic, D.B., Pesic, A., Electrochemica Acta, 2002. 47(18): p. 2901-2912.

12. Härtinger, S.D., Journal of Electroanalytical Chemistry, 1995. 380: p. 185-

191.

13. Zelinsky, A.G.P., Ya, B.; Yujev, O.A., Corrosion Science, 2004. 46: p.

1083-1093.

14. Texier, F.S.; Brueneel, J.L.; Argoul, F., Journal of Electroanalytical

Chemistry, 1998(446): p. 189-203.

15. Stankovic, Z.D.R.; Vukovic, M.; Krcobic, S., Journal of applied

electrochemistry, 1998. 28: p. 1405-1411.

16. Leopold, S.S.; Lu, J.; Toimil Molares, M. E.; Herranen, M.; Carlsson, J.-O.,

Electrochimica Acta, 2002. 47: p. 4393-4397.

17. Huang, L.-Y.B.; Hung, Chen-Jen; Switzer, Jay, Israel Journal of

Chemistry, 1997. 37: p. 297-301.

18. Golden, T.D.S.; Zhou Yanchun; VanderWerf, Rachel A.; Van Leeuwen.;

Switzer, Jay A., Chemistry of Materials, 1996(8): p. 2499-2504.

19. Zhou, Y.C.; Switzer, Jay A., Materials Research Innovation, 1998(2): p.

22-27.

20. Sauerbrey, G., Zeitschrift fuer Physik, 1959. 155: p. 206-222.

21. Hepel, M., Electrode-Solution Interface Studied with Electrochemical

Quartz Crystal Nanobalance, in Interfacial Electrochemistry, ed,

Wieckowski, A., 1999, Marcel Dekker: Postdam. p. 599-629.

22. Lu, C., Applications of Piezoelectric Quartz Crystal Microbalances, in

Methods and Phenomena: Their application in Science and Technology,

ed, Lu, C., 1984, Elsevier Science: Amsterdam. p. 393.

23. Chyan, O.A., Tiruchirapalli N.; Ponnuswamy, Thomas, Journal of the

Electrochemical Society, 2003. 150(5): p. C347-C350.

24. Zhang, Y., et al., Electrochemical and Solid-State Letters, 2004. 7(9): p.

C107-C110.

25. Wilde, P.C., et al., Electrochemica Acta, 2000. 45: p. 3649-3658.

26. Santos, M.C.; Miwa, D.W. and Machado, S.A.S., Electrochemistry

Comunications, 2000. 2: p. 692-696.

27. Jeffrey, C.; Wendy M.; Harrington, David A., Journal of Electroanalytical

Chemistry, 2004. 569(1): p. 61-70.

28. Danilov, A.I.; Polukarov, Yu.M.; Climent, J. M.; Feliu, J. M., Electrochimica

Acta, 2001. 46: p. 3137-3145.

29. Watanabe, M.; Tomita, M. and Ichino, T., Journal of the Electrochemical

Society, 2001. 148(12): p. B522-B528.

30. Noli, F.M.; Hatzidimitriou,A.; Pavlidou, E.; Kokkoris, M., Journal of

Materials Chemistry, 2002(13): p. 114-120.

31. Helliwell, M. and Smith, J.V., Acta Crystallographica, Section C., 1997.

C53: p. 1369-1371.

32. Veleva, L.Q.; Ramanuskas, R.; Pomes,R.; Maldonado, L., Electrochemica

Acta, 1995. 41(10): p. 1641-1646.

33. Barthelmy, D., http://webmineral.com/data/Posnjakite.html. 2004.

34. Zhang, D.K.L.; Liu, Y. L.; Yang, H., Physica B: Condensed Matter, 2004.

351(1-2): p. 178-183.

35. Dai, H.T., Corrine K.; Sarikaya,Mehmet; Baneyex,Francosie; Schwartz,

Daniel T., Langmuir, 2004. 20(8): p. 3483-3486.

36. Switzer, J.A.; Hiten, M.; Bohannan, Eric W., Journal of Physical Chemistry

B, 2002. 106(16): p. 4027-4031.

37. Lee, J.; Yonsung,T., Electrochemical and Solid-State Letters, 1991. 2(11):

p. 559-560.

38. Wagner, C.D., Handbook of X-Ray Spectroscopy.

39. Kautek, W.; Joseph G., Journal of the Electrochemical Society, 1990.

137(9): p. 2672-2677.

40. Borgohain, K.; Murase, N. and Mahamuni, S., Journal of Applied Physics,

2002. 92(3): p. 1292-1297.

41. Park, J.; Huang, Y. H.; Alkire, R. C., Journal of the Electrochemical

Society, 1999. 146(2): p. 517-523.

42. Zhou, Y.; Switzer, Jay A., Scripta Materalia, 1998. 38(11): p. 1731-1738.

43. King, F.; Litke, C.D., Journal of Electroanalytical Chemistry, 1995(385): p.

45-55.

44. Fonseca, I.T.E. and Correia de Sa, A.I., Materials Science Forum, 1995.

192-194: p. 511-524.

45. Vukmirovic, M.B.; Dimitrov, N.; Sieradzki, K., Journal of the

Electrochemical Society, 2003. 150(1): p. B10-B15.

46. Nair, M.; Arenas, Olga L.; Nair, P.K., Applied Surface Science, 1999(150):

p. 143-151.

47. Liu, R.; Switzer, Jay A., Applied Physics Letters, 2003. 83(10): p. 1944-

1946.

48. Cascalheira, A.C. and Abrantes, M.L., Electrochimica Acta, 2004. 49: p.

5023-5028.

49. Kang, M.G. and Andrew, A., Journal of Physical Chemistry B, 2002. 106:

p. 12211-12220.

50. Matos, J.B.; Agostinho,S.M.L.; Barcia,O.E.; Cordeiro,G.G.O.; D'Elia,E.,

Journal of Electroanalytical Chemistry, 2004. 570: p. 91-94.

51. Mendoza, A.; Gomez, Ariel.; Gomez, Jorge., Corrosion Science, 2004. 46:

p. 1189-1200.

52. Maurice, V.K.; Strehblow, Hans-Henning; Marcus, Philippe., Journal of the

Electrochemical Society, 2003. 150(7): p. B316-B324.

53. Zerbino, J.O, Journal of Applied Electrochemistry, 1997. 27: p. 335-344.

54. Christy, A.G., et al., Journal of Applied Electrochemistry, 2004. 34: p. 225-

233.

55. Malvault, J.Y. and Lopitaux, J., Journal of Applied Electrochemistry, 1995.

25: p. 841-845.

56. Lee, S.Y.M., N.; Nguyen,N; Sun, Y.M.; White, J.M., Applied Surface

Science, 2003(206): p. 102-109.

57. Kirsh, P.D, Journal of Applied Physics, 2001. 90(8): p. 4256-4264.

100

REFERENCE LIST

1. Alvarez de Buergo, M. and Fort Gonzalez, R. Construction and Building

Materials, 2003. 17: p. 83-89.

2. Angerstein-Kozlowska, H., et al., Journal of Electroanalytical Chemistry,

1979. 100: p. 417-446.

3. Balek, V., et al., Colloidal and Interface Science, 2003. 260: p. 70-74.

4. Ballentine, J., Acoustic Wave Sensors: Theory, Design and Physico-

Chemical Applications. in Acoustic Sensors: Theory, Design and Physico-

Chemical Applications, ed. J. Ballentine. Vol. 1. 1997, San Diego:

Academic Press. 263.

5. Bard, A.J. and Faulkner, L.R., Electrochemical Methods: Fundamentals

and Applications. Second ed. 2001, New York: Jonh Wiley and Sons.

833.

6. Bard, A.J.; Ru-Ren, F., Journal of Physical Chemistry B, 2002. 106: p.

279-287.

7. Barthelmy, D., http://webmineral.com/data/Posnjakite.html. 2004.

8. Borgohain, K., N. Murase, and S. Mahamuni, Journal of Applied Physics,

2002. 92(3): p. 1292-1297.

9. Bracchini, C., et al., Catalysis Today, 2000. 55: p. 45-49.

10. Bruckenstein, S. and S. Swathirajan, Electrochimica Acta, 1985. 30(7): p.

851-855.

101

11. Burke, L.D. and J.F. Healy, Journal of Electroanalytical Chemistry, 1981.

124: p. 327-332.

12. Buttry, D.A. and M.D. Ward, Chemical Reviews, 1992. 92(6): p. 1355-

1379.

13. Buttry, D.A., Applications of the Quartz Crystal Microbalance, in

Electroanalytical Chemistry: A series of Advances, A.J. Bard, Editor. 1996,

Marcel Dekker: New York. p. 1-85.

14. Cascalheira, A.C. and Abrantes, M.L., Electrochimica Acta, 2004. 49: p.

5023-5028.

15. Chang, C.C. and T.C. Wen, Journal of Applied Electrochemistry, 1997. 27:

p. 353-363.

16. Christy, A.G., et al., Journal of Applied Electrochemistry, 2004. 34: p. 225-

233.

17. Chrzanowski, W. and Wieckowski, A., Langmuir, 1997. 13: p. 5974-5978.

18. Chyan, O.; Tiruchirapalli N.; Ponnuswamy, Thomas, Journal of the

Electrochemical Society, 2003. 150(5): p. C347-C350.

19. Cicileo, G.; Rosales, Blanca M., Corrosion Science, 2004. 46: p. 929-953.

20. Conway, B.E., Electrochemical Supercapacitors. First. ed. 1999, New

York: Kluwer-Plenum.

21. Cullity, B.D., Elements of X-Ray Diffraction. 3rd ed. 2001, Upper Saddle

River: Prentice-Hall.

102

22. Dai, H.T., Corrine, K.; Sarikaya,Mehmet; Baneyex, Francosie; Schwartz,

Daniel T., Langmuir, 2004. 20(8): p. 3483-3486.

23. Danilov, A.I.M., E. B.;Polukarov, Yu. M.; Climent, J. M.; Feliu, J. M.,

Electrochimica Acta, 2001. 46: p. 3137-3145.

24. Deakin, M.R. and Buttry, D.A., Analytical Chemistry, 1989. 61(20): p.

1147A-1154A.

25. Dickert, F.L., et al., Sensors and Actuators B, 2001. 76: p. 295-298.

26. Dickert, F.L., et al., Sensors and Actuators B, 2003. 95: p. 20-24.

27. Doblhofer, K., et al., Journal of the Electrochemical Society, 2003.

150(10): p. C657-C664.

28. Ebel, M.F., Journal of Electron Spectroscopy and Related Phenomena,

1976. 8: p. 213-224.

29. Ehahourn, H., et al., Analytical Chemistry, 2002. 74(5): p. 1119-1127.

30. Faulkner, L.R., Electrochemical Characterization of Chemical Systems, in

Physical Methods in Modern Chemical Analysis. 1983, Academic Press:

San Diego, CA.

31. Fitzgerald, K.P.; Atrens, A., Corrosion Science, 1998. 40(12): p. 2029-

2050.

32. Flores, S., Making an Electrode. 2003: Denton, Tx. p. 19.

33. Fonseca, I.T.E. and Correia de Sa, A.I., Materials Science Forum, 1995.

192-194: p. 511-524.

103

34. Galliano, F.; Olsson, .A.; D. Landolt, Journal of the Electrochemical

Society, 2003. 150(11): p. B504-B511.

35. Garcia, S., et al., Electrochimica Acta, 1998. 43(19-20): p. 3007-3019.

36. Gau, W.C.; Lin, Y. S.; Hu, J. C.; Chen, L. J.; Chang, C. Y.; Cheng, C. L.,

Journal of Vacuum Science & Technology A: Vacuum, Surfaces and

Films, 1999. 18(2): p. 656-660.

37. Giauque, P.H. and Nicolet, M.A., Journal of Applied Physics, 2003. 93(8):

p. 4576-4583.

38. Golden, T.D.; Zhou Yanchun; VanderWerf, Rachel A.; Van Leeuwen;

Switzer, Jay A., Chemistry of Materials, 1996(8): p. 2499-2504.

39. Gonzales-Tejera, M.J. and Van Huong, C. Nguyen, Journal of

Electroanalytical Chemistry, 1988. 244: p. 249-259.

40. Grosser, D.K., Cyclic Voltammetry: Simulation and Analysis of Reaction

Mechanism. 1993, New York: Wiley-VCH.

41. Grujicic, D. and Pesic, B., Electrochemica Acta, 2002. 47(18): p. 2901-

2912.

42. Hadzi-Jordanov, S., Angerstein-Kozlowska, H. and Conway, B.E.

Electroanalytical Chemistry and Interfacial Electrochemistry, 1975. 60: p.

359-362.

43. Hartmann, A.J., et al., Applied Physics A, 2000. 70(239-242).

44. Hayden, O. and Dickert, F.L., Advanced Materials, 2001. 13(19): p. 1480-

1483.

104

45. Hayden, O., et al., Sensors and Actuators B, 2003. 91: p. 316-319.

46. Härtinger, S.D., Journal of Electroanalytical Chemistry, 1995. 380: p. 185-

191.

47. Helliwell, M. and Smith, J.V., Acta Crystallographica, Section C, 1997.

C53: p. 1369-1371.

48. Hepel, M., Electrode-Solution Interface Studied with Electrochemical

Quartz Crystal Nanobalance, in Interfacial Electrochemistry, ed,

Wieckowski, A., 1999, Marcel Dekker: Postdam. p. 599-629.

49. Herrero, E.; Buller, L.J.; Abruña, H.D., Chemical Reviews, 2001. 101(7): p.

1897-1930.

50. Horvat-Radosevic, V.; Kvastek, K.; Vukovic, M. Journal of

Electroanalytical Chemistry, 1999. 463: p. 29-44.

51. Huang, L.; Hung, Chen-Jen; Switzer, Jay, Israel Journal of Chemistry,

1997. 37: p. 297-301.

52. Iqbal, S.S., et al., Biosensors and Bioelectronics, 2000. 15: p. 549-578.

53. Itagaki, M.; Tagaki, M.; Watanbe, K., Corrosion Science, 1996. 38(7): p.

1109-1125.

54. Jeffrey, C.; Wendy, M.; Harrington, David A., Journal of Electroanalytical

Chemistry, 2004. 569(1): p. 61-70.

55. Jouen, S. ; Hannoyer, B., Surface and Interface Analysis, 2000. 30: p.

145-148.

105

56. Kanazawa, K.K. and Gordon, J.G. II, Analytical Chemistry, 1985. 57: p.

1770-1771.

57. Kang, M.G. and Andrew, A., Journal of Physical Chemistry B, 2002. 106:

p. 12211-12220.

58. Kautek, W.; Joseph, G., Journal of the Electrochemical Society, 1990.

137(9): p. 2672-2677.

59. Kim, H.W. and Kang, C.-J., Microelectronic Engineering, 2003. 69: p. 89-

96.

60. Kim, H.-K., et al., Thin Solid Films, 2005. 475(54-57).

61. King, F.Q.; Litke, C.D., Journal of Electroanalytical Chemistry, 1995(385):

p. 45-55.

62. Kirsh, P.D., Journal of Applied Physics, 2001. 90(8): p. 4256-4264.

63. Komplin, G.C. and Pietro, W.J., Sensors and Actuators B, 1996. 30: p.

173-178.

64. Krznaric, D. and Goricnik, T., Langmuir, 2001. 17: p. 4347-4351.

65. Lee, J. and Yonsung, T. Electrochemical and Solid-State Letters, 1991.

2(11): p. 559-560.

66. Lee, W.W.; White, H.S.; Ward, M.D., Analytical Chemistry, 1993. 65: p.

3232-3237.

67. Lee, S.Y.; Nguyen, N; Sun, Y.M.; White, J.M., Applied Surface Science,

2003(206): p. 102-109.

106

68. Leopold, S.S.; Lu, J.; Molares, Toimil M. E.; Herranen, M.; Carlsson, J.-O.,

Electrochimica Acta, 2002. 47: p. 4393-4397.

69. Lin, W.F.; Christensen, P.A.; Hamnet, A., Journal of Physical Chemistry B,

2000. 104: p. 6642-6652.

70. Liu, R.B.; Switzer, Jay A., Applied Physics Letters, 2003. 83(10): p. 1944-

1946.

71. Long, J., W., et al., Langmuir, 1999. 15: p. 780-785.

72. Lu, C., Applications of Piezoelectric Quartz Crystal Microbalances, in

Methods and Phenomena: Their application in Science and Technology,

ed, Lu, C.C. 1984, Elsevier Science: Amsterdam. p. 393.

73. Malvault, J.Y. and Lopitaux, J. Journal of Applied Electrochemistry, 1995.

25: p. 841-845.

74. Marx, K.A., et al., Biosensors and Bioelectronics, 2001. 16: p. 773-782.

75. Masamitsu, W.; Masato, T.; Ichino, Toshihiro, Journal of the

Electrochemical Society, 2001. 148(12): p. B522-B528.

76. Mason, W.P., The Journal of the Acoustical Society, 1981. 70(6): p. 1561-

1566.

77. Matos, J.B.; Agostinho,S.M.L.; Barcia,O.E.; Cordeiro,G.G.O.; D'Elia,E.,

Journal of Electroanalytical Chemistry, 2004. 570: p. 91-94.

78. Maurice, V.K.; Lorena, H.; Strehblow, Hans-Henning.; Marcus, Philippe.,

Journal of the Electrochemical Society, 2003. 150(7): p. B316-B324.

79. Mendez, P.F., et al., Electrochimica Acta, 2004. In Press.

107

80. Mendoza, A.R.C., Francisco.; Gomez, Ariel.; Gomez, Jorge, Corrosion

Science, 2004. 46: p. 1189-1200.

81. Mendoza-Huizar, L.H.; Robles, J.; Palomar, P. Journal of Electroanalytical

Chemistry, 2003. 545: p. 39-45.

82. Michalitsch, R., B.J. Palmer, and P.E. Laibinis, Langmuir, 2000. 16: p.

6533-6540.

83. Michell, D.R., D.A.J.;Woods, R., Journal of Electroanalytical Chemistry,

1978. 89: p. 11-27.

84. Nair, M.T.; Arenas, Olga, L.; Nair, P.K., Applied Surface Science,

1999(150): p. 143-151.

85. Nascente, P.A.P., Journal of Molecular Catalysis A: Chemical, 2005. 228:

p. 145-150.

86. Nicic, I., et al., Journal of Physical Chemistry B, 2002. 106: p. 12247-

12252.

87. Noli, F.M.; Hatzidimitriou, A.; Pavlidou, E.; Kokkoris, M., Journal of

Materials Chemistry, 2002(13): p. 114-120.

88. O'Sullivan, C.K. and Guilbault, G.G., Biosensors and Bioelectronics, 1999.

14: p. 663-670.

89. Park, J.O.; Huang, Y. H.; Alkire, R. C., Journal of the Electrochemical

Society, 1999. 146(2): p. 517-523.

90. Prakash, J. and Joachin, H. Electrochimica Acta, 2000. 45: p. 2289-2296.

108

91. Quiroz, M.A. and Meas, Y. Journal of Electroanalytical Chemistry, 1983.

157: p. 165-174.

92. Rao, C.N.R. and Raveau, B. Transition Metal Oxides Structure, Properties

and Synthesis of Ceramic Oxides. 2nd. ed. 1998, New York: Wiley-VCH.

93. Rodhal, M.F.; Hook, M.F.; Kasemo, B. Analytical Chemistry, 1996. 63(12):

p. 2219-2227.

94. Rooryck, V., et al., Journal of Electroanalytical Chemistry, 2000. 482: p.

93-01.

95. Sakti, S.P., et al., Sensors and Actuators B, 2001. 78: p. 257-262.

96. Salt, D., Hy-Q Handbook of Quartz Crystal Devices. First Edition, ed, Salt,

D. Berkshire. 1987: Van Nostrand Reinhold. 229.

97. Santos, M.C.; Miwa, D.W.; Machado, S.A.S. Electrochemistry

Comunications, 2000. 2: p. 692-696.

98. Santos, M.C. and Machado, S.A.S. Journal of Electroanalytical Chemistry,

2004. 567: p. 203-210.

99. Santos, M.C. and Machado, S.A.S., Electrochimica Acta, 2004. In Press.

100. Sauerbrey, G., Zeitschrift fuer Physik, 1959. 155: p. 206-222.

101. Schlesinger, R. and Bruns, M., Surface and Interface Analysis, 2000(30):

p. 135-139.

102. Schmitt, R.F., et al., Sensors and Actuators B, 2001. 76: p. 95-102.

103. Scholz, F., Electroanalytical Methods. 1st. ed. 2002, Berlin, New York:

Springer. 331.

109

104. Schwartz, D. and Rolf, H., Surface Science, 1991(248): p. 349-358.

105. Seddon, E.A. and K.R. Seddon, The chemistry of Ruthenium. in Topics in

Inorganic and General Chemistry, ed, Clark. R.J.H. Vol. 19. 1955,

Amsterdam: Elsevier Science Publishers S.V. 1373.

106. Snook, G.; Alan, B.; Fletcher, Stephen, Journal of Electroanalytical

Chemistry, 2002. 526: p. 1-9.

107. Stankovic, Z.; Vukovic, M.; Krcobic, S., Journal of applied

Electrochemistry, 1998. 28: p. 1405-1411.

108. Swathirajan, S. and Bruckenstein, S., Electrochimica Acta, 1983. 28(7): p.

865-877.

109. Switzer, J.A.; Chen-Jen, H.; Huang, Ling-Yuang; Miller, Scott F.; Zhou,

Yanchun, Journal of Materials Research, 1998. 13(4): p. 909-916.

110. Switzer, J.A.; Hiten, K. M.; Bohannan, Eric W, Journal of Physical

Chemistry B, 2002. 106(16): p. 4027-4031.

111. Texier, F.S., L.; Brueneel, J.L.; Argoul, F., Journal of Electroanalytical

Chemistry, 1998(446): p. 189-203.

112. Tomic, O.; H. Ulmer, and J.E. Haugen, Analytical Chimica Acta, 2002.

220450: p. 1-13.

113. Trasatti, S. and Buzzanca, G., Journal of Electroanalytical Chemistry and

Interfacial Electrochemistry, 1971. 29(1971): p. A1-A5.

114. Trasatti, S., Electrochimica Acta, 1984. 29(11): p. 1503-1512.

110

115. Tsionsky, V., et al., Looking at the Metal/Solution Interface with the

Electrochemical Quartz Crystal Microbalance: Theory and Experiment., in

Electroanalytical Chemistry, a Series of Advances, ed, Bard, A.J. 2002,

Marcell Dekker: New York.

116. Tsionsky, V., et al., Journal of Electroanalytical Chemistry, 2002. 525-525:

p. 110-119.

117. Uchida, H., et al. Quartz Crystal Microbalance Study of the Specific

Adsorption of Halide Ions on Highly Ordered Au(111). in Sixth

International Symposium on Electrodes. 1996: The electrochemical

Society.

118. Uchida, H.; Hiei, M.; Watanabe, M. Journal of Electroanalytical Chemistry,

1998. 452: p. 97-106.

119. Veleva, L.; Ramanuskas, R.; Pomes,R.; Maldonado, L., Electrochemica

Acta, 1995. 41(10): p. 1641-1646.

120. Vukmirovic, M.; Dimitrov, N.; Sieradzki K., Journal of the Electrochemical

Society, 2003. 150(1): p. B10-B15.

121. Vukovic, M.; Valla, T.; Milun, M., Journal of Electroanalytical Chemistry,

1993. 356: p. 81-90.

122. Vukovic, M. and Dunja, C., Journal of Electroanalytical Chemistry, 1999.

474: p. 167-173.

123. Wagner, C.D., Handbook of X-Ray Spectroscopy.

111

124. Wang, J., Analytical Electrochemistry. 2nd ed. 2000, New York: Wiley-

VCH. 207.

125. Wang, J.X., et al., Journal of Physical Chemistry B, 2001. 2001: p. 2809-

2814.

126. Ward, M.D. and Buttry, D.A., Science, 1990. 249(4972): p. 1000-1007.

127. Waszczuk, P., et al., Journal of Catalysis, 2001. 203: p. 1-6.

128. Waszczuk, P., et al., Electrochimica Acta, 2002. 47: p. 3637-3652.

129. Watanabe, M.; Uchida, H.; Ikeda, N., Journal of Electroanalytical

Chemistry, 1995. 380: p. 255-260.

130. Watanabe, M.; Tomita, M.; Ichino, T., Journal of the Electrochemical

Society, 2001. 148(12): p. B522-B528.

131. Wieckowski, A. and Kuk, S.T., Journal of Power Sources, 2005. 141: p. 1-

132. Wilde, P.C., et al., Electrochemica Acta, 2000. 45: p. 3649-3658.

133. Williams, J.O. and Mahmood, T., Application of Surface Sciences, 1980.

134. Wu, Y.-L. and Hwang, Y.-C., Thin Solid Fims, 2004. 461: p. 294-300.

135. Zelinsky, A.; Ya, B.; Yujev, O.A., Corrosion Science, 2004. 46: p. 1083-

1093.

136. Zerbino, J., et al., Journal of Applied Electrochemistry, 1997. 27: p. 335-

344.

112

137. Zhang, M., Electrochemical Quartz Crystal Microbalance (EQCM) Studies

of Electrocatalytic Reactions, in Chemistry Department. 1995, University

of Ottawa: Ottawa. p. 262.

138. Zhang, Y., et al., Electrochemical and Solid-State Letters, 2004. 7(9): p.

C107-C110.

139. Zhang, D.; Liu, Y. L.; Yang, H., Physica B: Condensed Matter, 2004.

351(1-2): p. 178-183.

140. Zhong, H., Ru-based gate electrodes for advanced dual-metal gate CMOS

devices., in Analytical Chemistry. 2001, North Carolina State University:

Raleigh. p. 257.

141. Zhou, Y. and Switzer, Jay A., Materials Research Innovation, 1998(2): p.

22-27.

142. Zhou, Y. and Switzer, Jay A., Scripta Materalia, 1998. 38(11): p. 1731-

1738.

113