Chapter 3

THEORY OF ELECTRODEPOSITION

3.1 Introduction

The art of electroplating metals and metallic alloys has been in

practice for nearly a century and the earlier efforts are well documented.

Most of the development has been more by way of art rather than science,

which started to merge only recently. Further the viability of using the

electrodeposition technique as a tool of material technology is attracting

attention as a means of obtaining films of a wide variety of materials

including semiconductors, superconductors, polymer films, materials for

biostimulation, specific electronic device application materials and others.

Some of the key advantages of the electrodeposition technique are:

1. It is possible to grow films over large areas as well as irregularly

shaped surfaces.

2. Compositionally modulated structures or non-equilibrium alloys can

be electroplated.

3. A wide range of industrial experience can be drawn upon.

4. It is especially attracted in terms of cost, high throughput and

scalability.

Electrodeposition was originally used for the preparation of metallic

mirrors and corrosion resistant surfaces among other things. In its simplest

form electrodeposition consists of an electrolyte containing metal ions, an

electrode or substrate on which the deposition is desired, and a counter

electrode. When a current flows through the electrolyte, the cations and

anions move towards the cathode and anode, respectively, and may

deposit on the electrode after undergoing a charge transfer reaction. The

discovery of electrodeposition can be traced back to Michael Faraday and

his famous laws of electrolysis.

The first law states that the total amount of chemical change

produced by an electric current is proportional to the total charge passing

through the electrolyte. The second law states that the masses of the

different substances liberated in the electrolysis are proportional to their

chemical equivalent weights.

3.2 Electrodeposition System

A simple electrodeposition system consists of the following

components:

3.2.1 Electrolyte:

The electrolyte or bath provides the ions to be electrodeposited. It

has to be electrically conductive, it can be aqueous, non-aqueous or

molten; and it must contain suitable metal salts. Sometimes an additive is

included to improve the quality of the electrodeposits. An ideal additive

should not become incorporated in the film but should lead to improvement

of its adhesion, surface finish, uniformity etc.

3.2.2 Electrodes:

At least two electrodes (cathode and anode) are needed. An applied

electric field across these electrodes provides the main driving force for the
ions. The positive and negative ions deposit at the cathode and anode

respectively. Cathodic deposition is more popular in electroplating because

(1) most metal ions are positive ions and (2) anodic deposition has been

found to give poor stoichiometry and adhesion.

3.2.3 Power Supply:

The power supply can be (1) direct current at constant voltage, which

leads to potentiostatic deposition, (2) direct constant current which leads to

galvanostatic deposition or (3) a current or voltage waveform or pulse.

The process leading to electrodeposition can be outlined in simple

terms as below. On the application of an electric field, Mn+ would move to

the cathode, and the chemical electrodeposition process can be written as

Mn+ + n e -------> M

On the other hand, if the electrolyte contains more than one species

that can be simultaneously deposited, then the electrodeposition process,

for, say, two types of ionic species can be written as

M+ + e -------> M

N+ + e -------> N

Or

M+ + N+ + 2e ---------> MN

Accordingly a compound or alloy of a multicomponent system can be

deposited we can deposit a compound or an alloy of a multi-component

system.
 As the electrodeposition proceeds, the ionic concentration in the bath

is depleted and has to be replenished by adding the salt. Another

alternative is to use one of the electrodes as a sacrificial electrode, one

that corrodes during electrodeposition to maintain the ionic concentration

constant.

The ionic discharge process can be summarized as follows:

1. Ionic species move in the electrolyte bulk toward the electrode on

which the deposition is to occur. This is regarded as the reactant

zone.

2. The moving ionic species approach the electrode (prereaction site)

at, say, the outer Helmholtz plane (OHP) without actually being

discharged. The charge transfer that ultimately leads to

electrodeposition starts at this interface. This is regarded as the

charge transfer zone.

3. The charge transfer is completed at the electrode, and the

electrdeposit is obtained.

According to the Butler-Volmer model, the discharge of the

electroplating ions is assumed to take place only when the latter have

arrived at the outer Helmholtz plane. While calculating the rate of charge

transfer reaction, it assumes that the rate at which the ions are consumed

by the discharge reaction is equal to the rate at which they arrive at the

OHP. In the charge transfer zone or pre-reaction zone the ions are

transported from the electrolyte bulk to be discharged at the electrode. The

concentration term in the Butler Volmer equation should be the ionic

concentration in the charge transfer zone at x = 0, i.e., c x = 0 and not the

bulk concentration c 0 at x = ∞ . The two will be equal only when the ionic

discharge rate is equal to the rate of ionic transport. If the charge transfer

rate is greater than the rate of ionic transport, then the interface region will

become depleted of the discharging ionic species. On the other hand, if the

ionic transport processes are much faster, this will lead to the accumulation

of ions in the interface region. This inequality between the transport flux

and the charge transfer flux gives rise to concentration polarization or

overpotential.

3.3 Importance of Diffusion Limited Transport

Diffusion limited transport has been found to be useful in

electrodeposition work. A discussion of the important implications follows.

Diffusion limited transport enables oppositely charged ionic species

to move in the same direction. One can thus perform cathodic

electrodeposition with negatively charged ions also. Diffusion limited

transport enables one to change the equilibrium potential over a range of

values depending upon the ionic activities. According to Nernst equation, a

decrease in ionic activity of the discharging species results in a negative

shift in the deposition potential. This possibility has been used by a number

of workers to electrodeposit two or more components with different

equilibrium potentials.

The current density under the influence of diffusion is given by

   αFη 
j = j 1  1 − exp    ………..(3.1)
  RT 

Suppose in a cathodic electrodeposition experiment the electrode

potential is continuously driven in the negative direction. Initially when the

electrode potential is made more negative than the equilibrium potential of

the ionic species transported by diffusion, the current will rise. The cathodic

current will continue to rise with increasingly negative electrode potential,

and the interfacial ionic activity will decrease continuously. When the rate

of diffusion becomes equal to the rate of discharge, the interfacial ionic

activity will be practically zero and the current density will attain a value j l .

The higher efficiency of the cathodic process results from the higher value

of j l . This value can be increased by (1) increasing the concentration c o ,

(2) increasing the diffusion coefficient by working at a higher temperature,

and (3) lowering the diffusion layer thickness by forced convection

(stirring).

3.4 Transfer Coefficient:

In the derivation of the Butler – Volmer equation, the movement of

ions across the interface was assumed to be the only necessary condition

for the charge transfer reaction. The role of electrons in the metal electrode

as dynamic entities capable of crossing the interfacial barrier was

completely ignored. The electron in general quantum mechanically tunnel

through the interfacial barrier to the ionic species in the electrolyte [1 - 3].

As the electron transfer process is non-radiative, the tunneling should take

place without a change in the energy of the system within the restraints of

the uncertainty principle. This requires sufficient stretching of the bonds

between the reactant and the surrounding ligand to create an acceptor

state of energy equal to the electron energy in the metal.

The transport coefficient (symmetry factor)α for quantum mechanical

tunneling has been defined as the ratio of the energy required to stretch the

ion ligand bond to the critical state and the energy gap between the

electron state in the metal and the solution side of the interface. The

energy gap should be closed to make tunneling possible. It can be seen

that α is correlated to some physically meaningful parameters such as the

ion- solvent interaction and the structure of the double layer.

The charge transfer reaction activated by a bridging ion can be

treated quantum mechanically. In bridge assisted tunneling, the

transmission probability may be large. Also if the bridging ion is of opposite

charge to reacting species, this may lead to a higher concentration of the

latter at the OHP.

3.5 Surface Coverage Factor:

The electrode surface immersed in the electrolyte is generally with

contact adsorbed ions, solute molecules etc., is broadly of two categories:

non blocking and blocking. The effect of blocking species can be

incorporated into the Butler-Volmer equation by considering only the

fraction θ of the electrode surface that is not available for charge transfer

reaction when a current j is flowing. θ is called the surface coverage factor

for all adsorbed species. Frumkin [4] gave a detailed account of the role of

contact adsorption in the electrode process.

The binding energy of the contact adsorbed species is of practical

interest in electrodeposition. For example, during aqueous

electrodeposition, the discharging hydrogen is often strongly bound to the

cathode surface with energy on the order of 50 kcal/ mol. On the other

hand the water molecules are only weakly adsorbed. Interfacial tension has

also been used as a parameter to express the interaction of the solvent-

electrode system.

Many non aqueous solvents interact fairly strong with the electrode

surface and θ become a dominant factor. Apart from solvent adsorption, the

role played by the contact- adsorbed additives is also significant in

controlling θ (and hence the electro-deposition).

Multi step reaction:

For a single step reaction,

r + 2e ----------> p

The above reaction may comprise two steps

r + e ---------> I,

I + e ---------> p

where I refer to an intermediate species formed during the reaction.

Generally the step with the lowest exchange current density qualifies

as the rate determining step (RDS). Suppose j is the current density

corresponding to the RDS for an s – step reaction. The total current density
would be sj. The electrode potential will have only one value, but the

overpotentials corresponding to the s different steps may be different. The

over potential is the difference between the electrode potential and the

reverse potential. The different values for overpotential can be attained at

different steps simply because of the differences in the reversible potentials

of each step, which are governed by the respective concentration ratios of

the intermediates.

The exchange current density jo and the transfer coefficients α c

and α a are two important parameters related to the kinetics of the charge

transfer reaction. For a given electrode potential, the net current density

will be higher for the process with the higher exchange current density. The

exchange current density depends on the nature of the reaction, the

electrode material, and the bath composition. The transfer coefficient

describes the effect of the electric field on the charge transfer step and the

symmetry of the cathodic and anodic processes. Its dependence on the

electrode material is usually small.

3.6 Electrodeposition of Semiconductors

In principle the theory of metallic electrodeposition can be applied to

semiconductor film deposition. However, some problems of typical

semiconductors have to be considered.

First, semiconductor films are relatively resistive (in comparison with

metals). As a consequence, the interfacial potential and charge distribution

may drastically change over the first few layers of semiconductor. For

thicker films this may even lead to morphological changes.

Second, the resistivity of a semiconductor is highly sensitive to

defects, orientation, and other factors. As the electrodeposition proceeds,

the semiconductor resistivity may continue to vary. In an extreme situation,

a non degenerate semiconductor may even become a degenerate

semiconductor.

Third, in metal electrode – electrolyte contact, the space charge layer

within the electrode plays an insignificant role, whereas for semiconductor

electrolyte contact the space charge layer is dominant.

Fourth, the presence of the space charge layer, the high density of

surface states or surface defects, and the resistivity of the semiconducting

film can also exercise controlling effects on the charge transfer reactions.

As a result the Tafel plot for semiconductors may not be the same as was

originally envisaged for metallic deposits.

The above remarks are applicable to both elemental and compound

semiconductors. Apart from a few elemental semiconductors (Si, Ge,

Se.Te) most of the semiconductors of interest are compound

semiconductors. The three problems of a typical compound semiconductor

electrodeposition are

1. Most of the compound semiconductors have at least one metallic

(Ga, Cu, In, Cd, Zn, etc.,) and one nonmetallic (S, Se, Te, P, As

etc.,) component as in GaAs, GaP, CdS and CuInSe 2 .

 Unfortunately, the reduction potentials of the nonmetals are widely

different from those of the metal ions. The general condition for co-

depositing the different components of the compound requires that

the constituents have equal reduction potentials. This is difficult to

attain for compound semiconductors.

2. Most compound semiconductors have a very negative Gibbs free

energy of formation. This may shift the deposition potential of the

more noble component to more positive value, helping the co-

deposition.

3. Many compound semiconductors exhibit multiple phases. The

possibility of such multiphase formation may lead to additional

complexities in the activity term controlling the electrodeposition.

3.7 Practical Aspects of Electrodeposition

3.7.1 Choice of substrate:

Substrates in semiconductor electrodeposition are expected to play

an important role. Their characteristics are not restricted to impart certain

morphological characteristics to the growing layer. In a more subtle way

electronic or optical properties may also be affected. Therefore, in choosing

a suitable substrate, in addition to considering the need to provide

mechanical support to the electrodeposits, due consideration must be given

to the possible influence of the substrate on the properties of the deposit.

Broadly speaking the following criteria should be applied for the selection of

the substrate.
1. It should have good conductivity. One can use an insulating

substrate, provided a suitably conducting coating is first applied on

its surface. Good conductivity of the substrate is also beneficial in

improving the carrier collection efficiency.

2. The thermal expansion of the substrate should match well with that of

the electrodeposit. Usually, in semiconductor electrodeposition the

film requires annealing treatments at temperatures that may be fairly

high to improve the grain size or stoichiometry or to fabricate

junctions. A mismatch in the thermal expansion often leads to strains

that result in cracking or peeling of the film.

3. The substrate should have good mechanical strength.

4. In many applications cost is an important consideration.

5. For epitaxial films, it is necessary to match the lattice parameters

between the single crystal substrate and the growing film.

6. The semiconductor metal contacts may be ohmic or rectifying, so

care should be given to the type of interface one desires to obtain.

7. In some cases the atoms of the substrates tend to diffuse inside the

electrodeposited semiconducting film, especially during post

deposition annealing treatments. The film purity, doping

concentration, electronic properties etc., may consequently be

altered. Due care should be given to these considerations. The solid

solubility, diffusion coefficient and other properties of the substrate

 atoms in the semiconductor bulk should be examined to avoid such

problems.

8. The substrate should be stable in the electrolyte bath.

9. The substrate surface should be smooth and one should avoid

surface waviness, porosity, voids, and other irregularities as these

influence the local current distribution. Also the electrodeposits tend

to reproduce the surface morphology of the substrate. An uneven

porous surface with voids will not be useful for any device

application.

Subject to the foregoing considerations one may use single crystal,

polycrystalline or amorphous substrates of metals or semiconductors in the

form of foils, sheets, wafers or thin films. When single crystal

semiconductors are used as substrates a back ohmic contact is normally

formed by using a conducting paint, a suitable solder or an evaporated film.

Metals have been widely used as substrates because of their good

conductivity, easy availability, lower cost and relative ease of handling.

3.7.2 Substrate surface preparation:

The preparation of smooth scratch free substrate surface is

extremely important in semiconductor electrodeposition as surface

inhomogeneities tend to amplify during electrodeposition. On the atomic

scale surface defects may be point defects, dislocations etc., while on the

macroscopic scale they can be scratches left from polishing operations,

grain boundaries etc. Substrate preparation consists of three major steps,

grinding and polishing, cleaning and testing the surface cleanliness.

Grinding and polishing are usually employed to prepare a smooth surface

with a mirror finish. The grinding operation uses abrasives of different grit

sizes. Emery paper can be used for grinding, after grinding the surface is

either polished mechanically or electrochemically, Mechanical polishing is

performed using polishing wheels with abrasive grains glued to them with

the help of adhesives or cement compounds. The glues are available in a

wide range of viscosities and flexibilities. The abrasive most commonly

used in metal polishing is fused alumina grains, which are available in

various grain sizes down to 0.3 µm. Fused alumina is hard, sharp, fast

cutting and long wearing. Silicon carbide can also be employed for certain

special operations. However, it is more difficult to bond it firmly to the

wheel. The polishing surface is lubricated with a lubricating oil or grease

that can be sprayed or friction applied on to the polishing wheel.

Lubrication is desirable to minimize heat and produce a fine polish. Typical

speeds of polishing wheels range from 6000 to 8000 revolutions per

minute.

In some applications electropolishing can also be employed. This is

an electrochemical operation in which the substrate is allowed to dissolve

slowly by applying an anodic bias to it. The microscopic projections are

dissolved away at a greater rate, resulting in smoothing, leveling and/ or

deburring. Brightening of the surface takes place simultaneously. The

surface brightness achieved by electropolishing is different from that

obtained by mechanical means. Electropolishing yields a scratch free,

deformation free surface. Time, temperature and current density are the

critical parameters that control the surface finish. The surface quality and

finish obtained by a prior polishing operation also help to determine the

final finish resulting from electropolishing. One can remove layers of a few

micrometers thick using electropolishing.

3.7.3 Cleaning:

Surface cleaning in thin film technology is an important step prior to

deposition. It is necessary to remove the contaminants that would

otherwise affect the properties of the films. The properties that can be

affected by the presence of the contaminants include adhesion,

morphology, nucleation, electronic properties of the film, and the substrate

film interface. The choice of the cleaning procedure is governed by the

substrate as well as by the type of contaminants that are likely to be

present. The composition, physical properties and chemistry of the

substrate should be carefully considered in designing the cleaning

operation. The cleaning process should be chosen to avoid any undesirable

damage to the substrate surface and yet contaminant is to be removed. It

is easier to select the cleaning procedure when the nature and origin of the

contaminants are known. Some common contaminants are finger tip

grease, glue, dust, leftovers of abrasives, soldering fluxes and similar

materials.
3.7.4 Cleaning by solvents:

Solvent cleaning is employed to dissolve or emulsify the

contamination. Solvent cleaning can be performed by soaking the surface

in petroleum or chlorinated solvents. Some common chlorinated solvents

are trichloroethylene, methylene chloride, perchloroethylene. These

chlorinated solvents may contain hydrochloric acid as a hydrolysis product,

and due care should be exercised to inhibit it. Vapor degreasing units are

also commonly employed for cleaning. Such a unit consists of a chamber in

which a chlorinated solvent is vaporized. Cleaning takes place when the

solvent vapors condense on the colder substrate. Solvent cleaning can also

be accomplished by soaking in an emulsifiable solvent. Detergent cleaning

is useful for metals. Acid cleaning is often employed to remove oxides and

oil from inert glass or metal substrates. An aqueous solution of organic or

inorganic acid is normally employed in soaking, painting, brushing or

spraying applications. The contaminants are first converted into water

soluble compounds that are subsequently removed in a water rinse.

Alkaline cleaners along with some surface active agents are often

employed after detergent cleaning to remove oil smuts and oxides.

0
Cleaning is generally performed at temperatures of 120 – 200 F. The final

cleaning should be done in flowing deionized water.

Ultrasonic cleaning can also be a useful technique, particularly for

ceramics. The cleaning is assisted by ultrasonic agitation in which jetting,

which accomplishes the collapse of bubbles generated by cavitation near

the surface, increases the rate of solvation and emulsification. The cleaning

liquid can be water based or solvent based chemical in which the

substrates are immersed. The effectiveness of ultrasonic cleaning depends

on the intensity of cavitation achieved. Liquids with higher surface tension

and lower viscosity should be preferred. The dissolved gases in the liquid

also tend to decrease cavitation intensity, whereas increasing temperature

increases it. The ultrasonic frequency and power should also be properly

chosen. Higher frequencies require more power to maintain the same level

and cavitation intensity. Similarly, the ultrasonic power should be matched

to the work desired.

3.7.5 Cleaning by Heating:

Heating the substrate may remove the volatile impurities. The temperature

should be chosen according to the melting point and/ or surface reactivity

of the substrate. Heating is not useful if it causes stresses and cracking

due to non uniform heating or oxidation of the surface.

3.7.6 Cleaning by etching:

In many cases, a suitable etchant can be used as a clean surface. Etching

can be performed in the dark or under illumination. This can be specially

used for semiconductor substrates.

3.8 Other methods:

There are a number of cleaning methods that are more useful than

the foregoing for thin film deposition in vacuum system. Three of these

methods are sputter cleaning, plasma oxidation and glow discharge

cleaning. Cleaning techniques have been discussed by Brown, Holland

[5,6].

3.8.1 Surface Cleanliness Test:

The various tests that can be applied for checking the surface

cleanliness are:

3.8.2 Breath Figure Test:

The substrate is brought near the mouth and a soft breath is blown

on it. A poorly reflecting black specular film indicates a clean surface.

Breath figures of various shapes appear on an uneven surface.

3.8.3 Atomizer Test:

Water is sprayed over the surface of the dried substrate. Surface

cleanliness is indicated by the formation of fine mist, while on an unclean

surface water coalesces into big drops.

3.8.4 Water Break Test:

The substrate is pulled against the surface of water in a beaker. If

the water sticks to the substrate surface as a continuous film, a clean

surface is indicated.

3.8.5 Contact Angle Test:

A contact angle zero between water droplet and the substrate

implies that the substrate is clean.

3.8.6 Coefficient of friction Test:

Resistance encountered when a glass or metal sheet is slid over

the substrate can also be a measure of cleanliness. If the coefficient of

friction approaches unity the surface is clean.

3.8.7 Indium Adhesion Test:

Surface cleanliness is tested by measuring the coefficient of

adhesion between the substrate and a piece of indium. It may range from

zero for a dirty surface to about two for a clean surface.

3.8.8 Fluorescence Dye Test:

Contaminants that can absorb fluorescent dyes can be detected by

illuminating the substrate with UV light.

3.8.9 Edge Lighting Effect:

It is useful to check transparent substrates such as glass. If the

edge of the glass is illuminated, the contaminants present on the surface

become visible as lighted areas against a dark background.

In addition to the foregoing techniques one can also employ

gravimetric or radiotracer methods.

3.9 Electrolytic Bath

The electrolytic bath is the medium that supplies the ions that move

upon application of an electric field. In general ionic transport is facilitated

in aqueous solutions, non aqueous solutions or molten salt bath.

3.9.1 Aqueous and Non-aqueous Electrolytic Solutions:

The choice of solvent depends on primary factors such as solubility

and non-reactivity.

3.9.1.1 Aqueous: These solvents are suitable for a large number of salts,

complexing agents and other compounds. Barring a few hydrolysis

reactions, water is generally a nonreactive solvent. However an aqueous

+ –
solution necessarily contains H and OH ions, which complicate the

electrodeposition process by resulting in the evolution of Hydrogen and / or

oxygen at the electrodes.

3.9.1.2 Non-aqueous: These can be further be classified as protic and

aprotic solvents.

3.9.1.3 Protic Solvents: (Alcohols, formamide etc.,) these solvents are

generally strong hydrogen donors and can exchange protons rapidly. Such

solvents also sometimes lead to hydrogen evolution.

3.9.1.4 Aprotic Solvents: They contain hydrogen bonded only with the

carbon (propylene carbonate, DMF, DMSO, acetonitrile, tetrahydrofuran

etc.,).

The nonaqueous solvents have attracted attention for semiconductor

electrodeposition work due to the greater flexibility they afford in choosing

dopants, solutes, complexants, temperature range and working electrode

potentials and the absence of hydrogen evolution reaction, among other

reasons.
3.9.2 Preparation of Electrolytic Bath:

3.9.2.1 Selection of Solvent: The first step is to choose aqueous or non-

aqueous solvents depending on the material to be deposited.

Electrodeposition solvents are stable in only a limited potential range,

beyond which reduction- oxidation take place. This range is called the

working potential range or the window. The potential at which the

electrodeposition is to be carried out should be within this range. The

morphology and rate of growth depends upon the temperature at which

electrodeposition is being carried out. The solvent should remain liquid at

the desired temperature. Therefore solvents with a large range are

preferred to provide greater flexibility.

High vapour pressure solvents are preferred because the electrolytic

concentration remains more constant. This is particularly true when the

solvent is being purged with nitrogen or an inert gas to drive out the

dissolved oxygen.

The dielectric constant should be more than 10. Too low a dielectric

constant facilitates ion pair formation, giving poor conductivity and

ionization in the solution.

Lower viscosity is always preferred because of the better conductivity

and diffusion. Furthermore, many irreversible electrochemical reactions

limited by mass transport become reversible in a medium with lower

viscosity.
3.9.2.2 Selection of Supporting Electrolyte:

The supporting electrolyte performs several functions in the

electrochemical process. (1) It increases the conductivity of the electrolyte.

This minimizes Joule heating and provides more uniform current

distribution and iR compensation.(2) It reduces the electrode double layer

thickness and also influences ion pairing and adsorption.(3) It effectively

eliminates the effect of migration in the mass transport. The criteria for

selection of a supporting electrolyte are its solubility in the solvent and a

dissociation constant sufficiently high to yield good conductivity and the

electrochemical oxidation of anion and electroreduction of cation at more

anionic or cationic potentials, respectively in the electrochemical process

under investigation.

In aqueous media, KCl, HCl have been commonly used, In organic

solvents lithium perchlorate, lithium tetrafluoborate, trifluoroacetate and

quaternary ammonium salts have been employed as supporting

electrolytes.

3.9.2.3 Additives in Electrolytes:

Additives (brightening agents, surfactants, complexants etc.) are

often added to the plating bath to obtain a brighter and smoother deposit,

controllable reaction rate, better adhesion and also a better texture. The

role of additives in the electrodeposition process is used either (1) to

control the rate of electrodeposition process or to (2) influence deposit

morphology.

3.9.2.4 Solvent Purity:

Impurities in the solvent interfere with the electrochemical processes.

In many cases electrodeposition may not be possible or the deposit

morphologies may be affected by the impurities. The electronic properties

of the semiconductor are also extremely sensitive to impurities that co-

deposit. It is therefore essential to purify the solvent. Some of the

commonly used methods employ deionization using ion exchange resin or

distillation from an alkaline permanganate solution. However this is not a

very satisfactory method and many organic contaminants may still be

present. Further purification by passing the water vapours through a

column of platinum gauge heated to 750 – 800ºC in a stream of oxygen is

necessary to remove these contaminants. In some case purification by

flushing an inert gas to remove oxygen followed by pre electrolysis at a

constant potential may also be employed to remove heavy metal ions.

Non aqueous solvents also need to be purified by refluxing with

strong oxidizing or reducing agents, distillation under reduced pressure, or

passage through molecular sieves. A common impurity in non-aqueous

solvent is oxygen, which can be removed by purging with nitrogen.

3.10 Molten Salt Electrolysis Bath:

The term molten salt includes molten media that may be wholly ionic

or derived from simple salts. The conductivity and ionicity of these systems
are generally sensitive to temperature, pressure and composition. Molten

salts exhibit a wide range of electrochemical stability high ionic

conductivity, high heat capacity and good thermal conductivity. They also

exhibit good electrochemical reaction rates. In contrast to aqueous

electrodeposition molten salt electrodeposition can be carried out without

the annoying intervention of hydrogen evolution, oxide formation, hydride

formation and similar problems.

Some of the commonly used molten salt electrolytes are, NaCl + KCl + Na 3

PO 4, NaF + KF + NaPO 3 + In 2 O 3, etc. Molten salt baths suffer from certain

disadvantages like high operating temperatures, lack of sufficient

thermodynamical and kinetic data.

3.11 Classification of Electrodeposition Techniques

Electrodeposition techniques can be classified according to the

nature of the electric field applied across the electrolysis cell:

Deposition at constant dc potential (Potentiostatic)

Deposition at constant dc current (Galvanostatic)

Deposition using a periodic or pulse source.

3.11.1 Potentiostaic Deposition:

In this technique different charge transfer reactions proceed under a

steady state condition at rates appropriate to the steady state interfacial

overpotential and exchange current density. The choice of the overpotential

is dictated by the composition of the bath, the substrate and the reversible

potential of the species to be deposited. The table of standard electrode

potentials serves as an approximate guide in finding the potentials at which

electrodeposition of particular species will be possible, but in practice the

actual deposition depends on a number of factors such as substrate deposit

interaction of the hydrogen overvoltage, the interaction between the

components during compound electrodeposition and the polarization

characteristics of the bath. Potentiostatic deposition is carried out under

pure activation, diffusion or mixed control depending on the choice of the

deposition potential. Potentiostatic deposition has been used to grow

layers of elemental, binary and ternary semiconductors. In the case of

semiconductors it is desirable to obtain large grained polycrystalline

deposits. However, very low overpotential may not be suitable, as they may

lead to a spongy or porous deposit due to low nucleation rates. As the

overpotential is increased, the supersaturation increases and a large

number of nuclei are formed. The deposit thus acquires a fine grained

morphology. Very high overpotentials may, however lead to the growth of

dendrites or whiskers. In the case of compound semiconductors, more

stringent control of the deposition potential is required to maintain the

deposit stoichiometry.

3.11.2 Deposition at Constant Direct Current:

The constant current between the working and counter electrodes

required for electrodeposition at a constant direct current can be obtained

from a galvanostat.
 The initial guideline for choosing the required deposition current

density can be obtained from the knowledge of (1) the Faradaic efficiency

and (2) the maximum permissible growth rate for a good crystal. The former

can be obtained by a series of preliminary deposition experiments. The

latter is to be obtained from the available crystal growth data.

The galvanostatic electrodeposition of compound semiconductors is

more complex and a straight forward answer for the proper choice of

deposition current density is not possible. As discussed earlier, compound

semiconductors are usually deposited by the transport of at least one of the

constituents under diffusion control. The simplest situation is when the

deposition of all the components is diffusion controlled. The rates of

deposition will then be directly proportional to the corresponding limiting

current densities. As an example, consider the cathodic electrodeposition of

a binary compound semiconductor A x B y . If the electrodeposition of B is

under diffusion control, then the maximum current distribution due to

species B cannot exceed the corresponding limiting current density.

Further, if the deposition of A is under pure activation control, then the

current distribution due to A for galvanostatic deposition will depend on the

value of exchange current density of A and the steady state overpotential.

For the compound electrodeposition to be favoured, the value of the total

current density should be chosen so that the steady state overpotential

satisfies the thermodynamic condition for codeposition of A and B.

 Galvanostatic electrodeposition is therefore still an art as it depends

on the manipulative experience and ability of the individual worker. Initial

trials may be necessary to obtain the final optimum values of the

electrodeposition current density.

3.11.3 Electrodeposition from a Periodic or Pulsed Source:

The use of non dc signals for electrodeposition is known to improve

the deposit quality. However, only a few reports have appeared on its

application to semiconductor electrodeposition. Non dc electrodeposition

can be carried out by either varying the current or the overpotential.

Current variation can be accomplished by using a periodic reverse current,

a pulsating current or alternating current without or superimposed over a dc

current.

3.12 Morphology of The Electrodeposits

The nature and magnitude of the applied electric field across the

electrolysis decides (1) grain size (2) surface roughness (3) dendritic

growth and (4) spongy and powdery deposit formation.

3.12.1 Grain Size:

The grain size depends upon the overpotential which in turn controls

whether the deposition is controlled by diffusion, activation or both. The

relevant conclusions can be briefly stated as follows. At low overpotentials,

initially a small number of nuclei grow independently. It is therefore

expected that a large grained deposit will be obtained under these

conditions. As the overpotential is increased, a large number of nuclei may

be formed leading to a decrease in grain size.

3.12.2 Surface Roughness:

At an ideal surface, the value of the diffusion layer thickness and the

limiting current are constant throughout, which leads to an uniform growth.

Some minor variations in thickness may arise due to convective effect,

which is neglected here. However the situation is complex for a real surface

which is rough consisting of elevation and recesses. The rate of deposition

at the elevations may be higher due to shorter diffusional path between the

outer plane of the diffusion layer and the elevations. At the tips of the

elevations the diffusion conditions may approach those of spherical

diffusion. Spherical diffusion is faster than the linear diffusion because of

wider diffusional field in the former case. As a result surface roughness

gets amplified during electrodeposition.

3.12.3 Dendritic Growth:

In the case of dc electrodeposition, surface roughness can lead to

the initiation of preferential growth at protrusions or dendrites. The

mechanism of dendritic growth has been studied by Diggle et al and Popov

et al [7,8].

3.12.4 Formation of Powdery or Spongy Deposit:

Other than dendritic growth another undesirable type of

electrodeposit is a powdery or spongy deposit. Dendritic growth is primarily

controlled by overpotential or critical current density. However, powdery or

spongy growth is interplay of many factors other than the overpotential,

such as viscosity and temperature. Powdery deposits are classified by their

small particle size and poor adhesion to the electrode surface. They are

generally obtained when the deposition is carried out under diffusion limited

current conditions or close to transition time. The amorphous nature of the

powdery deposit is due to the high nucleation rate. Powder formation is

enhanced when the concentration of the depositing species is decreased,

the supporting electrolyte concentration is increased, solution viscosity is

decreased, the temperature is decreased or the stirring rate is decreased.

The mechanism of formation of a spongy deposit has been discussed by

Popov et al [9]. It was shown that a spongy deposit is formed if the radius

of the growing grain exceeds a critical value.

3.13 Basics of pulse plating

Generally in the electrodeposition technique for producing a metal or

compound, a driving force (i.e., the free energy) in the form of a potential or

current is applied to the electrode. Either of them can be used as a variable as in

the case of continuous electrodeposition. Modern electronics allows one to make

use of these parameters as a function of time. This permits a number of possible

ways of varying the conditions.

Four variable parameters are of primary importance in pulse plating. They

are: peak current density, ip, average current density, ia, ON time and OFF time.

The sum of the ON and OFF times constitute one pulse cycle. The duty cycle is

defined as follows:
Duty Cycle = ______ON time _____ x 100 % ------------ (1)

ON time + OFF time

A duty cycle of 100% corresponds to conventional plating because OFF

time is zero.

In practice, pulse plating usually involves a duty cycle of 5% or

greater.

The average current density(Ia) under pulse plating conditions is defined as

Ia = peak current density x duty cycle

= ip x duty cycle ---------------- (2)

During the ON time the concentration of the metal ions to be

deposited is reduced within a certain distance from the cathode surface.

This so-called diffusion layer pulsates with the same frequency as the

applied pulse current. Its thickness is also related to ip but reaches a

limiting value governed primarily by the diffusion coefficient of the metal

ions. During the OFF time the concentration of the metal ions build up

again by diffusion from the bulk electrolyte and will reach the equilibrium

concentration of the bulk electrolyte if enough time is allowed.

These variables result in two important characteristic features of pulse

plating which make it useful for alloy plating as well as property changes

as mentioned earlier.

(i) Very high instantaneous current densities and hence very high

negative potentials can be reached. The high over potential causes a shift

in the ratio of the rates of reactions with different kinetics. This high over
potential associated with the high pulse current density greatly influences

the nucleation rate because a high energy is available for the formation of

new nuclei.

(ii) The second characteristic feature is the influence of the OFF time

during which important adsorption and desorption phenomena as well as

recrystallization of the deposit occur.

3.14 References

[1]. H.Gerischer, Recent Advance in Electrochemistry,Vol.I, Wiley

Interscience, New York,1962.

[2]. V.G.Levich, Advances in Electrochemical and Electrochemical

Engg,vol-4,!966.

[3]. J.O.M.Bockris and A.K.N.Reddy, Modern Electrochemistry, Vol-2,

Plenum Press, New York, 1973.

[4]. N.Frumkin, Advances in Electrochemistry and Electrochemical

Engineering, vol.3, Intersceince, New York,1963.

[5]. R.Brown, Handbook of Thin film Technology,McGraw Hill, New

York,1970.

[6]. L.Holland, Vacuum Deposition of thin films, Chapman and Hall,

London, 1955.

[7]. J.W.Diggle, A.R.Despic and J.O.M.Bockris, J.Electrochem.Soc,

116(1969)1503.

[8]. K.I.Popov, M.C.Maksimovic and Djokic, Surf.Tech,14(1981)423.

[9]. K.I.Popov, M.D.Maksimovic, S.K.Zecevik and M.R.Stojik,

Surf.Technol, 27(1986)117.