2. vapor deposition technologies

7/27/2019 2. Vapor Deposition Technologies

http://slidepdf.com/reader/full/2-vapor-deposition-technologies 1/73

4 Handbook of Hard Coatings

4

2

Vapor Deposition

Technologies

Rointan F. Bunshah

1.0 SCOPE

This chapter deals with the vapor deposition technologies used

to deposit hard/wear resistant coatings onto various softer substrates.

Companion volumes in this series discuss these vapor deposition

processes in much greater detail.[1][2] Until very recently, these

substrates were metal, or metal alloys (e.g., steels, titanium, alumi-num), or cermets (e.g., W-Co). The hard coating improved the wear

resistance of these materials by diminishing the adhesive, or abrasive

type wear. Currently, polymer materials in the form of long continu-

ous webs or bulk forms such as windows or lenses have become

candidates for wear/water vapor permeation resistant coatings (usu-

ally oxides such as silica, titania, or alumina). This has been a majorchallenge to vapor deposition technology since the polymer materi-

als can withstand very low temperatures (80–300°C) during film

deposition. It is only through the detailed understanding of the

plasma assisted deposition processes that this coating triumph has

been achieved.



Vapor Deposition Technologies 5

2.0 CLASSIFICATION OF VAPOR DEPOSITION

PROCESS

Vapor deposition processes may principally be divided into

two types, (1) physical vapor deposition (PVD) and (2) chemical

vapor deposition (CVD).[1][2]PVD processes involve the creation of

material vapors, (by evaporation, sputtering, or laser ablation) and

their subsequent condensation onto a substrate to form the film.

CVD processes are generally defined as the deposition of a solidmaterial from the vapor phase onto a (usually) heated substrate as a

result of chemical reactions. PVD and CVD processes can be classi-

fied as shown in Table 1.

Table 1. Classification of PVD and CVD Processes

PVD Processes

Metals Compounds

Basic PVD Processes Basic PAPVD Processes

Evaporation Deposition Direct Activated Reactive

Evaporation (ARE)

Evaporation

Sputter Deposition or Reactive Sputtering (RS)

Sputtering

Hybrid PVD Processes Hybrid PAPVD

Processes

“Ion Plating

” “Reactive Ion Plating

”

CVD Processes

Basic CVD Process Basic PACVD Processes

Thermal CVD RF Excitation

Microwave Excitation

Photon Excitation




Table 2 shows the characteristics of basic PVD and CVD

processes. Generally speaking, the CVD process has the advantage

of good throwing power, while the deposition rates in PVD pro-cesses are higher than those in CVD processes at a lower deposition

temperature.

Table 2. Characteristics of Basic PVD and CVD Processes

Evaporation Sputtering Chemical Vapor

Deposition

Mechanism of Thermal energy Momentum Chemical reaction

production of transfer

depositing speciesDeposition rate Can be very high Low except for Moderate

(up to 750,000 pure metals (e.g., (200–2,500

Å /min) Cu-10,000 Å /min) Å /min)

Depositing species Atoms and ions Atoms and ions Atoms

Throwing power Poor line-of-sight Good, but nonuniform Good

coverage except thickness

by gas scattering distributionMetal deposition Yes Yes Yes

Alloy deposition Yes Yes Yes

Refractory compound Yes Yes Yes

deposition

Energy of deposited Low Can be high Can be high with

species (0.1–0.5 eV) (1–100 eV) plasma-aided CVD

Bombardment of Not normally Yes Possiblesubstrate/deposit

Growth interface Not normally Yes Yes (by rubbing)

perturbation

Substrate heating Yes, normally Not generally Yes

(by external means)




As with all processes, one is concerned with the process itself,

as well as the resulting microstructures and properties of the product.

In order to understand various vapor deposition processes, one has tomodel them in terms of three steps[3] illustrated in Fig. 1.

Figure 1. The three steps in film deposition.

Step 1: Creation of Vapor Phase Specie— Material can

be put into the vapor phase by evaporation, sputtering,

laser ablation, gases, vapors, etc.

Step 2: Transport from Source to Substrate— The vapor

species are transported from the source to the substrate

with, or without collisions between atoms and molecules.

During transport, some of the vapor species can be

ionized by creating a plasma in this space.

Step 3: Film Growth on the Substrate— This process

involves the condensation of the vapor species onto the

substrate and subsequent formation of the film by

nucleation and growth processes. The nucleation and

growth processes can be strongly influenced by

bombardment of the growing film by ionic species

resulting in a change in microstructure, composition,

impurities, and residual stress.




The degree of independent control of these three steps deter-

mines the versatility or flexibility of the deposition process. For

example, these three steps can be independently controlled in PVDprocesses, and therefore give greater degree of flexibility in control-

ling the structure, properties, and deposition rate, whereas all of the

three steps take place simultaneously at the substrate, and cannot be

independently controlled in the CVD process. Thus, if a choice is

made for a process parameter such as substrate temperature which

governs deposition rate in CVD, one is stuck with the resultantmicrostructure and properties.

3.0 PHYSICAL VAPOR DEPOSITION (PVD)

PROCESSES

3.1 Introduction

The basic PVD processes fall into two general categories: (1)

sputtering and (2) evaporation. The application of PVD techniques

ranges over a wide variety of applications from decorative, to high

temperature superconducting films. The thickness of the deposits

can vary from angstroms to millimeters. Very high deposition rates(25 µm/sec) have been achieved with the advent of electron beam

heated sources. A very large number of inorganic materials—metals,

alloys, compounds, and mixtures—as well as some organic materials

can be deposited using PVD technologies.

The terms evaporation and sputtering refer to the mechanisms

whereby a condensed species is transferred to the vapor phase,

discussed in detail below. Of major importance is the bombardment

of the growing film by energetic species (ions and energetic neutrals)

from the adjacent plasma. Therefore, the method of plasma genera-

tion, and the ion energies are of crucial importance to the structure,

and properties of the deposited film. Figure 2 illustrates the various

effects on film formation by ions and fast neutrals impinging on the

film as a function of their energy. These phenomena include:




Thermal activation of the condensing species resulting

in an increase in their surface mobility.

Chemical activation to form a compound by reaction

between the depositing species where the dominating

role is played by excited particles.

Desorption of impurity atoms, either originally present

on the substrate or co-deposited.

Creation of activated centers (charge defects, structural

defects, bulk defects).

Sputtering of impurity atoms.

Sputtering of deposited film species followed by re-

condensation, which helps in filling up the gaps between

the film islands thus resulting in a continuous film at

lower total thickness than without particle bombardment.

Implantation, which is not of interest here.

Thus, for thin continuous films, the electron energies should be

in the range of 10 to 103 eV, and the electron density should be in the

range of 1012 to 1013 cm-3. Figure 3 shows the range of plasma

density, and electron energy for different low pressure plasma dis-

charges. Ion Plating is a hybrid PVD process since it is defined as an

atomistic film deposition process in which the substrate surface and/

or the depositing film is subjected to a flux of high energy particles

sufficient to cause changes in the interfacial region between the film

and the substrate, as well as in the properties of the deposited film as

compared to a non-bombarded film. These changes may be in the

adhesion of the film to the substrate, film morphology, density, orstress. The source of the depositing species can be evaporation,

sputtering, gases, or vapors. A classification of PVD processes is

given in Table 1.



thermal activation of condensation,increase in mobility of condensing particles

cb&ifcal activation to form a compound(dominating role of excited parkI&)

creation of activated canters (char edefects, structural defects, matena B defects)

-. ,e

implantation

thermal

- energy + -gf; + - gy=l - - ion guns

and accelerators -11o.2

I I I10 -'

I I I I1 10 lo2 lo3 lo4 lo5

eV

ion energy -

Figure 2. Effect on layer formation by bions and fast neutrals impinging on the layer as a function of their energy.

(Courtesy Dr. S. Schille r, FEP. )




3.2 Sputter Deposition

Sputtering Phenomena. Sputtering is the phenomena of mo-

mentum transfer from an incident energetic projectile to a solid orliquid target resulting in the ejection of surface atoms or molecules.

In the sputter deposition process, the target, (a source of coating

material) and the substrate are placed in the vacuum chamber, and

evacuated to a pressure typically in the rage 10-4 to 10-7 torr. A

schematic diagram of the sputter coating process is shown in Fig. 4.

The target, (also called a cathode) is connected to a negative voltage

supply, and the substrate generally faces the target. A glow dischargeis initiated after an inert gas (usually argon gas) is introduced into the

evacuated chamber. Typical working pressure is in the range of 20 to

150 mtorr. The sputter target erosion rate is given by:

ρ

A JSM R 3.62= Å /min

Figure 3. Range of plasma density and electron energy in different low pressure

plasma discharges. (Courtesy Dr. S. Schiller, FEP.)




where J is the ion current density in mA/cm2, S is the sputtering yield

in atoms/ion, M A is the atomic weight in grams, and ρ is the density

gm/cm3 of the target material. Sputtering yield assuming perpen-dicular ion incidence onto a target consisting of a random array of

atoms can be expressed as:

U

E

M M

M M S i

t i

t i )(

constant

+×= Atom/ion

where: M i: mass of the incident atom

M t : mass of the target atom

E i: kinetic energy of the incident ion

U : heat of sublimation of the target material

Planar Diode Glow Discharge Sputter Deposition. This is

the simplest sputtering system. This configuration consists of the

cathode (the target) and anode facing each other (Fig. 4). The

substrates are placed on the anode. The target, which is usually

water-cooled, performs two functions during the process, one as the

source of coating material, and the other as the source of second-

ary electrons sustaining the glow discharge. The distance between

the cathode and anode is usually about 5 to 10 cm.

Figure 4. Planar diode sputter deposition.




The discharge current increases with the applied voltage, thus,

increasing the sputtering rate. However, the discharge current does

not increase linearly with the applied voltage as the voltage increasesabove 100 eV, since the ionization cross section decreases with

increasing electron energy. The sputtering rate can be increased if the

working gas pressure increases at a given voltage, due to an increase

in ion collection by the cathode.[4] However, the deposition rate

starts decreasing at high gas pressures due to gas scattering of the

sputtered atoms.The deposition rate is mainly determined by the power density

at the target surface, size of the erosion area, source-substrate dis-

tance, source material, and working gas pressure. Some of these

factors are interrelated, such as pressure and power density. There-

fore, the optimum operating condition is obtained by controlling the

parameters to get the maximum power which can be applied to the

target without causing cracking, sublimation or melting. The maxi-mum power limit can be increased if the cooling rate of the target is

increased by designing the coolant flow channels properly, and

improving the thermal conductance between the target and the target

backing plate.

Even though planar diode glow discharge sputter deposition

techniques are widely used due to their simplicity and the relativeease of fabrication of targets for a wide range of materials, they have

several disadvantages, such as low deposition rate, substrate heating

due to the bombardment of high energy particles, and relatively

small deposition surface areas.

Magnetron Sputter Deposition.By employing magnetic fields

to the diode sputtering process, the ionization efficiency near the

target can be greatly increased. In the conventional planar diodeprocess, ions are generated relatively far from the target and the

probability for ions to lose their energy to the chamber walls is large.

Furthermore, the number of primary electrons hitting the anode at

high energies without experiencing collisions is increased as the

pressure decreases, thus reducing ionization efficiency. These elec-

tron losses are not offset by impact-induced secondary emission.Therefore, ionization efficiencies are low and the discharge cannot

be sustained in planar diodes at pressures below about 20 mtorr.




In magnetron sputter deposition process, an applied magnetic

field parallel to the cathode surface forms electron traps and restricts

the primary electron motion to the vicinity of the cathode. Themagnetic field strength is in the range of few hundred gauss and

therefore, it can influence the plasma electrons but not the ions. The

electrons trapped on a given field line can advance across the mag-

netic field to an anode or walls by making collisions (mostly with gas

atoms). Therefore, their chances of being lost to the walls or anode

without collisions are very small. Because of the higher efficiency of this ionization mechanism, the process can by operated at pressures

around 1 mtorr with high current densities at low voltages, thus

providing high sputtering rates.

There are several configurations of magnetron sputter deposi-

tion technologies. Figures 5 and 6 show the cylindrical magnetron

and the planar magnetron respectively. The cylindrical magnetron is

a very useful technique to prepare uniform coatings over large areas,since long cathodes are employed in these techniques. Furthermore,

the cylindrical-hollow magnetron technique is effective for coating

complex-shaped objects. The cylindrical-post magnetron can be

used to substantially decrease substrate bombardment by energetic

particles, thus minimizing the heat of the substrate. Metallic films,

and dielectric films can be deposited with high deposition rates usingplanar magnetron sputtering, as compared to diode sputtering.

Even though magnetron sputtering techniques have the advan-

tages of high sputtering rate and low bombardment rates of energetic

particles onto the substrate, the utilization of this technique is im-

peded by the limitations in choice of target materials, and the diffi-

culties in fabrication of the target. For example, if ferromagnetic

materials are used as the sputtering target, their thickness should bethin enough so as to be saturated by the magnetic field. Since high

power is localized in a selected area in magnetron sputtering process,

targets should be prepared without voids or bubbles to avoid local

melting and spitting.




Figure 5. Cylindrical magnetron (post cathode) sputter deposition.

Figure 6. Planar magnetron sputter deposition.




Balanced and Unbalanced Magnetrons. The magnetic field

in a normal/balanced magnetron is designed to keep the electrons,

and hence, the plasma in the vicinity of the substrate. The balancedmagnetron was developed for microelectronic applications, where

bombardment of the growing film by energetic particles was to be

avoided. For hard coatings, it is necessary to bombard the growing

film with energetic particles. One, then, has the choice of setting up

a secondary plasma near the substrate as a source of energetic

particles, use a separate ion gun, or suitably alter the magnetic fieldsbetween the inner, and the outer sets of magnets in the sputtering

cathode, so as to permit the plasma to extend to the region of the

substrate, resulting in the desired ion bombardment of a positively

biased substrate. This is called the unbalanced magnetron, and ex-

amples are shown in Fig. 7. It may be noted that multiple pairs of

cathodes may be used to ensure that the plasma covers the entire

deposition volume without any dark spots with the south and northpoles alternating from one set to the adjacent set. [5]–[7] This ar-

rangement is now commonly used for deposition of hard nitride/

carbide coatings onto alloy steel and carbide tools.

Figure 7. Balanced and unbalanced sputter deposition.




RF Sputter Deposition. The development of the RF sputter

deposition technique made it possible to deposit films from noncon-

ducting sputtering targets which cannot be sputtered by the DCmethods due to charge accumulation on the target surface.

Most ions are almost immobile, as compared to electrons

which can follow the temporal variations in the applied potential at

the typical RF frequencies used for sputtering (5–30 MHz). When

the electrode is coupled to an RF generator, a negative voltage is

developed on the electrode due to the difference in mobility betweenelectrons and ions. Since the insulating target electrode constitutes a

capacitor in the electrical circuit, there should be no DC component

to the current flow. Therefore, the voltage on the electrode surface is

required to be self-biased negatively to compensate for the differ-

ence in mobility of electrons and ions, and to satisfy the condition of

net zero current (averaged over each cycle). The magnitude of the

resulting negative bias is almost the same as the zero-to-peak voltageof the RF signal. The period for the electrode to act as a anode is very

short and the electrode mostly acts as a cathode during the RF cycle.

Therefore, we can expect the target to be sputtered as in the DC case.

Significant numbers of ions are not accumulated on the target surface

while the electrode acts as a cathode, due to high frequency em-

ployed in RF sputtering (normally 13.56 MHz).Deposition can be performed at considerably lower pressures,

such as 5 to 15 m torr in RF sputtering, as compared to the planar DC

discharge, since electrons oscillating at high frequency can obtain

sufficient energy to cause ionizing collisions, and the number of

electrons lost (without making collisions) can be reduced.

RF sputtering is widely used to deposit various kinds of con-

ducting, semiconducting, as well as insulating coatings despite thecomplexity of the RF power source. This technique can also be

applied to magnetron sputtering sources.

Pulsed Mid-Frequency Sputter (PMS) Deposition Process

for Insulating Compounds. This is a recent development for the

high rate sputter deposition of insulating compounds (see discussion




on reactive sputtering in Sec. 4.3). Professor S. Schiller and his

colleagues at FEP (Fraunhofer Institut Elektronstrahl und

Plasmatechnik) are the pioneering group for the industrialdevelopment.[8a][8b] The stimuli for the development of this technol-

ogy can be gauged from the deficiencies of DC sputter and RF

sputter deposition of insulating compounds, and is illustrated in Fig.

8. For the DC case, one has the undesirable phenomena of arcing

(both macro and micro arcs), see Fig. 9, shifting of electrical poten-

tials by insulating layers, and low plasma density near the substratesurface. For the RF case, one has low deposition rates and high

power losses (reflected power), high expenditure for impedance

matching and associated high equipment cost, limited large area

application, high voltages, and high particle energies. The solution is

mid-frequency sputter deposition process in the frequency range of

10–100 kHz. The pulse width, off-time (if used), and the polarity can

be varied. The pulse can be unipolar, where the voltage is typicallynegative with a no-voltage (off) time or the bipolar, where the

positive and negative voltages are equal (symmetric) or unequal

(asymmetric), perhaps with an off-time. These arrangements are

illustrated in Fig. 10. In the positive bias and off-time, electrons

move to the surface from the plasma to neutralize any charge build

up produced during the negative portion of the cycle, thus preventingarcing over the dielectric surface and through a dielectric film being

deposited on a metallic surface. In the negative portion of the cycle,

energetic ion bombardment can sputter dielectric films that have

been formed on metallic targets during reactive sputtering, thus

keeping the target in the metallic mode. For the asymmetric case, the

negative pulse (-350 V for example) is greater than the positive pulse

(+100 V for example). The negative pulse time is typically 80 to 90%of the voltage cycle, and the positive pulse time is 10 to 20% of the

voltage cycle, resulting in high energy bombardment of the growing

film over a much greater portion of the cycle, as compared to RF

sputtering.




Figure 8. Deficiencies of DC and RF sputter deposition processes. (Courtesy Dr.

S. Schiller, FEP.)

Symmetrical pulsed DC is generally used in a dual magnetronsputtering configuration where each of the magnetrons are alterna-

tively biased positive and negative. This eliminates the “disappear-

ing anode” effect with DC sputtering of insulating films. However,

the dual cathode pulse sputtering arrangement does not sputter dur-

ing as large a portion of the cycle, as compared with asymmetrical

pulsed DC sputtering from a single magnetron.




The material being sputtered influences the frequency, dura-

tion, and pulse height. The frequency is lower for an insulating oxide

material, as compared to a somewhat higher frequency for a poorly

conducting material such as TiC or TiN, or ITO due to the easier

drainage of the charge from the surface of these materials.




Ion Beam Sputter Deposition. A relatively recent variation

called ion beam sputtering (Fig. 11) provides: (i)excellent adhesion,

(ii) a resulting high purity deposit due to a result of low operationalpressures (about 0.1 mtorr), and (iii) very low substrate heating

effects since the substrate is not in contact with the plasma. In this

technique, an ion beam of high energy (hundreds to thousands of

electron volts) extracted from the ion source is directed at a sputter-

ing target of the desired material. An inert, or reactive gas is used for

the ion beam source. The substrate is suitably located to collect thesputtered species from the sputtering target, as shown in Fig. 11.

There are two kinds of ion sources mainly used in practical thin film

deposition, the Kaufman source and the duoplasmatron. No further

discussion of these sources will be presented in this chapter; please

refer to the Handbook of Ion Beam Processing Technology.[9]

Figure 11. Ion beam sputter deposition.

Since the substrate can be isolated from the plasma generation

source, this permits independent control over the substrate tempera-ture, gas pressure, and the type of particle bombardment of the

growing film. Also, it is possible to control the energy and the target

current density independently in this technique, whereas it can be

done only by varying the working gas pressure in glow discharge

sputtering technology. The deposition rate in ion beam sputtering is

lower than in conventional sputtering. The reason for this low rate is

mainly due to the low beam current in the conventional ion beam




sputtering (dual-grid) system. This low-beam current can be greatly

increased by using a single-grid system; Nishimura, et al.,[10] ob-

tained a very high deposition rate over 90 nm/min for variousmaterials, such as Al, Cu, etc., using single-grid ion beam sputtering.

A small coverage of deposition area due to the small ion beam size

(about 1 cm) is another drawback of the ion beam sputtering. The

beam size can be increased to about 10 cm by adaptation of space-

type extraction optics.

3.3 Evaporation Deposition

Introduction. In evaporation processes, vapors are produced

from a material located in a source which is heated by various

methods. A schematic of an evaporation system is shown in Fig. 12.

It consists of an evaporation source to vaporize the desired material,

and the substrates which are located at an appropriate distance facing

the evaporation source. Resistance, induction, arc, electron beam, or

lasers are the possible heat sources for evaporation. The substrate

can be heated and/or biased to the desired potential using a DC/RF

power supply. Evaporation is carried out in vacuum, in a pressure

range of 10-5–10-10 torr. In this pressure range, the mean free path

(MFP) is very large (5×

102–

107

cm) as compared to the source-tosubstrate distance. Hence, the evaporated atoms essentially undergo

a collisionless line of sight transport prior to condensation on the

substrate, thus, leading to thickness build up directly above the

source, and decreasing steeply away from it. Planetary substrate

holders are therefore used in some cases so as to even out the vapor

flux on multiple substrates. In some cases an appropriate gas such as

argon at pressures of 5–200 mtorr is introduced into the chamber to

reduce the MFP so that vapor species undergo multiple collisions

during transport from the source to substrate thus producing reason-

ably uniform thickness coatings on the substrate. The technique is

called gas scattering evaporation or pressure plating.[11]




Figure 12. Evaporation deposition process schematic.

The transition of solids or liquids into the vapor phase is an

atomistic phenomenon. It is based on thermodynamics and results in an

understanding of evaporation rates, source-container reactions, and the

accompanying effect of impurity introduction into the vapor state,

changes in composition during alloy evaporation and stability of

compounds. An excellent detailed treatment of the thermodynamicand kinetic bases of evaporation processes is given by Glang.[12]

The rate of evaporation is given by the well known Hertz-

Knudsen equation,

)()2( 21

p pmkT dt

A

dN

e

e −πα=∗−

υ




where αυ is the evaporation coefficient, dN e /Aedt is the number of

molecules evaporating from a surface area Ae in time dt, p* is the

equilibrium vapor pressure at the evaporate surface, p is the hydro-static pressure acting on the surface, m is the molecular weight, k is

Boltzmann’s constant, and T is the absolute temperature. The evapo-

ration coefficient, αυ, is very dependant on the cleanliness of the

evaporate surface, and can range from very low values for dirty

surfaces, to unity for clean surfaces.

For reasonable deposition rates (100–

1000 nm/min) at a sourceto substrate distance of 20 cm, the vapor pressure should be about

10-2 torr. The source temperature should be adjusted to give this

value of the vapor pressure.

The directionality of evaporating molecules from an evapora-

tion source is given by the Cosine Law. Holland[13] and Graper[14]

have thoroughly discussed the theoretical distribution of vapor from

a point, a wire, a small surface, an extended tip, and from cylindricaland the ring types, or sources. For the ideal case of deposition from a

clean, uniformly emitting point source onto a plane receiver, the rate

of deposition varies as cos θ / r 2 (Knudsen’s cosine law), where r is

the radial distance of the receiver from the source, andθ is the angle

between the radial vector and the normal to the receiver direction.

Evaporation Sources, General Considerations. Evaporationsources are classified by the mode of heating used to convert the

solid or liquid evaporant to the vapor phase. Thus, one talks of

resistance, arc, induction, electron beam, arc imaging, lasers, and

exploding wire types of sources. A very important fact to be noted is

that we cannot evaporate every material from any of the types of

sources listed above for the following reasons:

1. Chemical interaction between the source material

and the evaporant which would lead to impurities in

the deposit. For example, evaporation of titanium

from a MgO source would cause oxygen and

magnesium contamination of the deposit; the

titanium would reduce the MgO. Therefore, for the

evaporation of reactive metals like titanium,zirconium, etc., we use water cooled copper crucibles.




A different type of sublimation source is used for the vaporiza-

tion of thermally stable compounds such as SiO, which are com-

monly obtained as powders or loose chunks. Such source material

would release large quantities of gases upon heating, thus, causing

ejection of particles of the evaporant which may get incorporated

into the film. Figure 15 shows two sources which solve this problemby reflection of the vaporized material.

Figure 14. Chromium sublimation source. (After Robert and Via.) The current

flows through the tantalum cylinder (heavy lines). (From The Handbook of Thin

Film Technology , ©1970, McGraw Hill.)

Figure 15. Optically dense SiO sources; (a) the Drumhellar source; (b) Compart-mentalized source. (After Vergara, Greenhouse, and Nicholas.) (FromThe Hand-

book of Thin Film Technology , ©1970, McGraw Hill.)




Evaporation Source Materials. We have already discussed

the potential problems concerned with the reaction between metal

sources and evaporates. Oxides and other compounds are morestable than metals. There are many metals which can be evaporated

from refractory oxide sources. Note that there is no such thing as an

absolutely stable oxide, nitride, or other compound. Reaction is

controlled by kinetics, i.e., temperature and time.

Oxide crucibles have to be heated by radiation from metal

filaments, or their contents can be heated by induction heating. Thisis illustrated in Figs. 16 and 17 for resistance heated sources.

Figure 16. Oxide crucible with wire coil heater. (From The Handbook of Thin

Film Technology , ©1970, McGraw Hill.)

Figure 17. DaSilva crucible source. (FromThe Handbook of Thin Film Technol-

ogy, ©1970, McGraw Hill.)

Other source materials are nitrides, such as boron nitride. A

50% BN–50% TiB2 is also well established as a crucible material.

This material (HDA composite, Union Carbide) is a fairly good

electrical conductor and hence can be directly heated to evaporate




Electron Beam Heated Sources. Electron beam heated sources

have two major benefits. One, is a very high power density, and

hence, a wide range of control over evaporation rates from very low,to very high. Two, the evaporant is contained in a water-cooled

copper hearth, thus, eliminating the problem of crucible contamina-

tion.

The evaporation rate for pure metals like Al, Au, Ag, which are

good thermal conductors, from water-cooled copper crucibles de-

creases due to heat loss to the crucible walls. In such cases, crucibleliners of carbon, and other refractory materials are used.

Any gun system must consist of at least two elements, a

cathode and an anode. In addition, it is necessary to contain these in

a vacuum chamber in order to produce and control the flow of

electrons, since they are easily scattered by gas molecules. A poten-

tial difference is maintained between the cathode and the anode. This

varies from as little as a few kilovolts to hundreds of kilovolts. Inmelting systems, a normal operational range is of the order of 10–40

kV. In the simple diode system, the cathode emits electrons, which

are then accelerated to the anode across the potential drop. Where the

anode is the workpiece to be heated, this is termed a work-acceler-

ated gun. It is shown schematically in Fig. 20. In a self-accelerated

gun structure, an anode is located fairly close to the cathode, elec-trons leave the cathode surface, are accelerated by the potential

difference between the cathode and anode, pass through the hole in

the anode and continue onward to strike the workpiece. Self-acceler-

ated guns have become the more common type in use and offer more

flexibility than the work-accelerated gun.

Electron beam guns may be further subdivided into two

types depending on the source of electrons: (1) thermionicgun and (2) plasma gun.




Thermionic Gun. In thermionic guns, the source of electrons is

a heated wire, or disc of a high temperature metal or alloy, usually

tungsten or tantalum. Such guns have the limitation of a minimum

operating gas pressure of about 1 × 10-3 torr. Higher pressures causescattering of the electron beam, as well as a pronounced shortening

of the cathode life (if it is a wire or filament) due to erosion by ion

bombardment. Figure 21 shows examples of thermionic electron

beam heated work-accelerated sources. The close cathode gun shown

in Fig. 21 is not a desirable configuration since molten droplet

ejection from the pool impinging on the cathode will terminate thelife of the cathode due to low melting alloy formation. Thus, cath-

Figure 20. Simple electron beam guns, (a) work accelerated gun; (b) self

accelerated gun.




odes are hidden from direct line-of-sight of the molten pool and the

electron beam is bent by electrostatic fields (Figs. 21b and 21c), or

magnetic field (Figs. 22 and 23) generated by electromagnets. Thelatter is a preferred arrangement since variation of the X and Y

components of the magnetic field can be used to scan the position of

the beam on the molten pool surface.

Figure 21. Work accelerated electron beam sources. (a) pendant-drop source, (b)

shielded filament (Unvala), (c) shielded filament. (Chopra and Randlett.) (From

The Handbook of Thin Film Technology, ©1970, McGraw Hill.)

Figure 22. Bent beam electron gun with water cooled evaporant support. (With

permission of Temescal Metallurgical Co., Berkeley, CA.)(From The Handbook

of Thin Film Technology, ©1970, McGraw Hill.)




Figures 21, 22, and 23 show linear cathodes (i.e., wires or

rods), and are referred to as transverse linear cathode guns. Figure 24shows a disc cathode which is characteristic of a high power Pierce

type electron beam gun. Low power Pierce type guns may have a hair

pin filament, or a wire loop as the cathode. In either case, the beam

geometry of the Pierce gun is different than that of the transverse

linear cathode guns. In some instances, the electron emitter assembly

is located at a distance from the crucible in a separately pumped

chamber to keep the pressure below 1 × 10-3 torr, with a small orificebetween the emitter chamber and the crucible chamber for the pas-

sage of electrons.

Figure 23. Transverse electron beam gun.




Plasma Electron Beam Gun.A plasma is defined as a region of

high-temperature gas containing large numbers of free electrons and

ions. By a proper application of electrical potential, electrons can be

extracted from the plasma to provide a useful energy beam similar to

that obtained from thermionic guns. There are three types of plasma

e-beam guns:

1. The Cold Cathode Plasma Electron Beam Gun.

The plasma electron beam gun has a cylindrical

cathode cavity made from a metal mesh or sheet,

(Fig. 25) containing the ionized plasma from which

electrons are extracted through a small aperture in

one end. The cathode is maintained at a negative

Figure 24. Schematic representation of a Pierce gun.




formed. Continued ion bombardment of the cathode

results in heating of the cathode and increased

electron emission. Ultimately, a high current “glowdischarge” will occur, analogous to that experienced

in vacuum arc melting at higher pressures. At this

point, the discharge appears as a low power density

beam “flowing” from the cathode aperture and

fanning out in conical shape into the chamber.

However, a parallel axial magnetic field is imposedon the beam (as seen in Fig. 26) which then forms a

high power density, well-collimated beam. The

hollow cathode discharge beam is operationally

stable, and efficient over the pressure range from

10-4 to 10-1 torr. A more detailed description of

physical aspects, operational characteristics, and

cathode design has been given by Morley.[16]

Figure 26. Schematic of the hot hollow cathode electron beam gun.




Wroe[22]in 1958 and Gilmour, et al.,[23]suggested vacuum arcs

as a source for metallic coatings. The U.S. patents to Snaper[24][25] in

1971, and the Russian patents to Sablev[26][27] in 1974, set the stagefor the commercial production of arc coatings, which were achieved

in the USSR around 1977–78. The first commercial use of the arc

evaporation-deposition method was for TiN coatings deposited at

low temperatures, particularly for high-speed steel cutting tools by

arc evaporation of titanium in a nitrogen plasma. This follows on the

heels of the Activated Reactive Evaporation (ARE) process devel-oped in 1971 for deposition of refractory compounds such as TiN

using electron beam evaporation techniques. There is very extensive

Russian literature on vacuum arc coating technology and the reader

can find a convenient source in recent reviews by Sanders[28] and by

Martin,[29] and in a recent companion volume in this series.[30]

There are two types of cathodic arc systems— pulsed and

continuous. In the pulsed devices, the arc is repeatedly ignited, andextinguished using a capacitor blank to supply the arc power. Pulsed

arcs have the advantage of letting the target cool between the pulses.

The disadvantage is the decrease in steady-state coating rates.

The continuous cathodic arc can be random in nature or con-

trolled. By the use of an insulating ring, a random arc source can be

constrained at the edge of the target, but allowed random motionwithin that constraint. Random arc sources have the advantage of

simplicity and excellent target utilization because the entire target

(except near the very edge) is utilized in the arc of very large parts.

The main disadvantage of random arcs is the formation of

macroparticles, which may cause the resulting coating to be unsuit-

able in some applications. Figure 27 shows that macroparticles are

ejected at small angles with respect to the target surface, and cantherefore be minimized using appropriate shielding. Such a strategy

has made possible arc-produced decorative coatings where surface

finish and optical specularity are of concern.




Magnetic fields can be used to control the trajectories of the

arcs. These fields can be used to discourage the arc from leaving the

desired portion of the target surface or can actually be used to define awell controlled path for the arc to follow in the so-called “steered arc”

devices. While the mechanism is still the subject of some debate, it is

clear, at least in the case of ceramic coatings based on refractory

metals, that steered arcs can produce coatings having extremely low,

or no measurable macroparticle component.

Macroparticles can also be removed by the use of suitable

filters as discussed by Sanders[28] and by Martin.[31] This is the so-

called “filtered arc evaporation process.”[30] Other strategies for

macroparticles involve the production of diffuse arcs. In one case,

the cathode is contained in a crucible which is allowed to heat up to

a temperature where the target material has a substantial vapor

pressure.[32] This causes a decrease in the arc voltage and current

density, the discharge becomes diffuse and macroparticles no longer

Figure 27. Phenomena occurring at a discrete cathodic arc spot.




form. The other approach is the so-called “anodic arc,”[33]–[35] (see

Fig. 28). In this process, the cathode initially supplies electronics, as

well as ions until the anode heats up. At this point, with sufficientelectron emission, a diffuse arc forms on the hot anode target mate-

rial which supplies the ions necessary to sustain the discharge. The

cathode material is not evaporated, and the coating material now

emanates from the anode. There are no macroparticles formed. High

deposition rates (several µm per minute) are obtained for a variety of

metals including Al, Ti, V, Ca, Mn, Fe, Ni, Cu, Pd, Ag, Au, andPt.[33] Since the substrate is left relatively cool, the process makes it

possible to produce adherent coatings on plastics at temperatures less

than 70°C, which makes this relatively new process a competitor for

sputter deposition. Alloy coatings such as stainless steel can be

readily deposited with good stoichiometric transfer. For example,

Ni, Al, and stainless steel coatings less than 1 µm thickness impart

excellent corrosion protection to iron.[36]

Figure 28. Schematic of the anodic arc evaporation process.




One of the main advantages of arc deposition processes is the

relatively high level of ionizing atoms in the plasma. This makes it

convenient to extract ion beams from the plasma and depositmacroparticle free coatings entirely from the ion beam.[28][31]

Laser Induced Evaporation/Laser Ablation/Pulsed Laser

Deposition (PLD). This technique with many names was first used

by Smith and Turner[37] in 1965 to deposit thin films in a vacuum

chamber used a pulsed ruby laser. Systematic studies in the 1970s were

performed to provide a better understanding of the physics of laser-solid interactions, and the related issues of deposition mecha-

nisms and film quality. More recently, the process has been extensively

used for growing highly crystalline dielectric films,[38] compound

semiconductor epitaxial layers, layers for bandgap engineering,[39][40]

and very extensively for high T c superconducting films.[41][42] The

reader is referred to an excellent review by Cheung and Sankur.[43]

In this technique, material is vaporized and ejected from thesurface of a target as it irradiated by a laser beam. Films are formed

by condensing the material ablated from the target onto a solid

substrate. Absorption characteristics of the material to be evaporated

determine the laser wavelength to be used. To obtain the high power

density required in many cases, pulsed laser beams are generally

employed. Pulse width, repetition rate, and pulse intensity are se-lected for specific applications.

In some studies on YBCO film deposition, the laser version of

a plasma-assisted reactive evaporation process was used. Oxygen

was bled into the system, and a plasma was created in the target-

substrate space by the use of a positively biased electrode placed

some distance above the target. This is the ARE process geometry

developed earlier and described later.Although laser evaporation is an attractive approach for syn-

thesis of high purity metal alloys and compound films, it suffers from

the following limitations:

• Complex transmitting and focusing systems need

to be employed to direct the beam from the laser

located outside the vacuum system onto the




4.0 PVD TECHNIQUES FOR DEPOSITION OF METALS,

ALLOYS, AND COMPOUNDS

The great versatility of the PVD processes is their ability to

deposit a very large number of materials including metals, alloys,

semiconductors, superconductors, polymers, and fabricate compos-

ites of various types (particulate, fibrous, or laminate).

4.1 Single Element Deposition

This can be carried out by evaporation or sputter deposition

processes. The deposition rate depends on the process and process

parameters.

4.2 Alloy Deposition

Alloys consist of two or more components, which have differ-ent vapor pressures and hence, different evaporation rates. As a

result, the composition of the vapor phase and therefore the deposits

are constantly varying. The following solutions have been used to

obtain alloy films with stoichiometry close to the source composi-

tion by evaporation based techniques.

a. Co-evaporation or co-sputtering using multiple sources.This technique involves simultaneous co-evaporation

of the constitutive elements of the alloy. The

composition of the deposited film is controlled by

adjusting the evaporation/sputtering rate of the

respective elements. In elaborate systems, separate

deposition rate monitors are used with appropriate

feedback networks to control the deposition rate from

each individual source independently. Near

stoichiometric films of many binary alloys have been

successfully deposited using this technique. Dispersion

strengthened alloys such as Ni-ThO2 have also been

successfully deposited by this technique.




b. Evaporation from a single source. This technique

involves evaporation of an alloy using a rod-fed electron

beam source. Evaporation operates under steady-stateconditions where the composition and volume of the

liquid pool on the top of a solid rod are kept constant

(see Fig. 29). A detailed description is given in Ref. 45.

c. Flash evaporation. In this process, pellets of the alloy

are dropped onto a very hot strip, and are vaporized

completely, thus maintaining the composition of thealloy in the deposit. It works very well for elements

with high vapor pressures.

d. Sputter deposition from an alloy target.

e. Sputter deposition from a segmented target where the

segments consist of each of the two components of the

alloy and the ratio of the target sample area of eachelement is inversely proportional to the sputtering yields.

Figure 29. Schematic of direct evaporation of an alloy from a single rod-fed

source.



Figure 30. Magnetron sputter source with double ring plasma on electrically insulated se

Schiller, FEP.)




f. Sputter deposition from a double ring magnetron

target[46] as illustrated in Fig. 30.

g. Laser ablation from an alloy target.

4.3 Deposition of Compounds

Deposition of compounds can be performed in two ways: (1)

Direct evaporation, using conventional heating methods or laser

ablation, where the composition of the evaporant is the same as that

of the compound that is to be deposited and (2) Reactive evapora-

tion, where the elements of the compound are evaporated, and react

with the gas to form the compound. Plasma assisted reactive evapo-

ration process are often used since they activate the reactions leading

to compound formation. Reactive evaporation, and plasma assisted

reactive evaporation will be discussed later.

Direct Evaporation. When a compound is heated, evaporationcan occur with or without dissociation of the compound into frag-

ments. There are a small number of compounds which are evapo-

rated without dissociation, specific examples being SiO2, MgF2,

B2O3, GaF2, and other Group IV divalent oxides.

In the more general case, when a compound is evaporated, the

material is not transformed to the vapor state as compound mol-

ecules but as fragments thereof. Subsequently, the fragments have to

recombine on the substrate to reconstitute the compound. Therefore,

the stoichiometry (anion: cation ratio) of the deposit depends on

several factors including the vaporization rate, the ratios of the

various molecular fragments in the vapor, the impingement of other

gases present in the environment on the film, the surface mobility of

the fragments (which in turn depends on their kinetic energy andsubstrate temperature), the mean residence time of the fragments of

the substrate, the reaction rate of the fragments on the substrate to

reconstitute the compound and the impurities present on the sub-

strate. For example, it was found that direct evaporation of Al2O3

resulted in a deposit which is deficient in oxygen. In other cases, the

deposit from direct evaporation of TiB2 contains both the monoboride,

and diboride phases.[47]




Laser Ablation. Laser ablation deposition techniques have

attracted great interest in recent years for the synthesis of semicon-

ducting, and insulating thin films. Very recently these techniqueshave been applied successfully for the deposition of high T c super-

conducting films.[48][49]

In this technique, material is vaporized and ejected from the

surface of a target as it is irradiated by a laser beam. Films are formed

by condensing material ablated from the target onto a solid substrate.

Absorption characteristics of the material to be evaporated deter-

mine the laser wavelength to be used. To obtain the high power

density required in many cases, pulsed laser beams are generally

employed. Pulse width, repetition rate, and pulse intensity are se-

lected for specific applications.

Although laser ablation is an attractive approach for the syn-

thesis of high purity metal, alloy, and compound films, it suffers

from some of the same limitations as laser evaporation:1. Complex transmitting and focusing systems need

to be employed to direct the beam from the laser

located outside the vacuum chamber onto the

evaporant placed inside the system. This involves

special designs and increases the cost of the setup.

Also, a window material which efficiently transmitsthe wave length band of the laser must be found and

mounted in such a way that it is not rapidly covered

up by the evaporant flux.

2. It is not always possible to find a laser with

wavelength compatible with the absorption

characteristics of the material to be evaporated.

3. Very low energy conversion efficiency.

4. The size of the deposited film is small, 1–2 cm dia.,

resulting from the small size of the laser impact

spot.

Reactive PVD Process. Reactive PVD processes are those in

which at least one of the elements of the coating is introduced in the




gas phase. Various compounds have been synthesized from metal

targets with reactive gases: air, O2, or H2O (oxides), N2 or NH3

(nitrides), O2 + N2 (oxynitrides), H2S (sulfides), etc. There areseveral advantages to these processes: (i) various kinds of com-

pounds can be prepared using relatively easy-to-fabricate metallic

targets, (ii) insulating compound films can be prepared using DC

power supplies (reactive sputtering), and (iii) graded composition

films can be formed.

Reactive Sputtering (RS). Sputter deposition is performed in

the presence of reactive species in the gas phase in reactive sputter-

ing process. Reactions can occur on the cathode surface, at the

substrate, and in the gas phase. However, reactions in the vapor

phase are precluded by considerations of momentum and energy

conservation unless the process is performed at high pressures to

allow multiple many body collisions in the gas phase.

In this process, the target is a nominally pure metal. Thecompound film is synthesized by sputtering in a pure reactive gas or

an inert gas-reactive gas mixture. Usually, the inert-reactive gas

mixture is preferred to the pure reactive gas from considerations of

sputtering rate. In the RS technique, a compound target also can be

used. In this case, the target is chemically decomposed by inert gas

ion bombardment. It is usually necessary to add the reactive gas to

compensate for the loss of reactive component by dissociation.

The main problem in reactive sputtering is target poisoning as

shown in Fig. 31. As the reactive gas partial pressure increases, the

rate of compound formation exceeds the removal rate of compounds

on the target surface, resulting in a decrease in deposition rate due to

the low sputtering yield of compound formed on the cathode surface,

and the fact that compounds have higher secondary electron emis-sion yield than pure metals. The increase in secondary electron

emission results in a reduction in both the discharge voltage, and ion

component in the cathode current at constant voltage. In other words,

more of the energy of incoming ions is consumed to produce and

accelerate secondary electrons.




The poisoning effect has two practical problems. One is the

aforementioned loss in deposition rate. The second relates to the

transition behavior, the material being deposited often passing abruptlyfrom metal to a nearly stoichiometric compound. This transition

behavior has been observed in planar diodes, planar magnetrons, and

cylindrical magnetrons. The transition behavior is most abrupt for

oxide deposition and more gradual for nitrides and carbides.

Much of the analysis of the transition behavior is incomplete

since it considers only the phenomena occurring at the cathode while

ignoring the total system. It is important to realize that reactivesputtering processes are dependent on the total system, i.e., its

geometry, the accumulation of coating on the walls and fixtures, and

the location of the gas injection. All of these have to be carefully

controlled to use reactive sputtering effectively in production. Non-

linear models show that with a sufficiently high overall pumping

speed, there is a smooth transition between metal and compound

Figure 31. Schematic showing target poisoning effect during reactive sputtering.




sputtering modes. This very high pumping speed becomes a practical

concern in the design of the vacuum system and process for reactive

sputtering.Several solutions have been found to reduce the effect of target

poisoning. They are:

1. Very high pumping speed.

2. Feedback control techniques using mass

spectrometric control and varying the reactive gas

pressure to operate in the metal mode.3. Incorporation of getter surfaces on the surfaces

surrounding the target and substrate, and the latter

at the target.

4. Incorporation of a baffle over the target to limit the

impingement of reactive gas molecules on the target.

Such target shields may have a single opening or adistributed opening using a mesh.

5. Creation of a plasma near the target using secondary

electronics from the target and inserting a positively

biased electrode near the substrate.

All of these are discussed in Ref. 50 by Karim, et al.

Reactive Evaporation Process. The difficulties involved indirect evaporation processes due to fragmentation of the vaporized

compounds are overcome in reactive evaporation where a metal is

evaporated in presence of the reactive gas. The compound is formed

by reaction of the evaporating metal species with the molecules of

the reactive gas. Even though this technique has been extensively

used to deposit a variety of oxide films for optical applications, it is

generally observed that the films are deficient in oxygen. It is also

observed in some cases, especially in the synthesis of carbide films,

that the deposition rate becomes a limiting factor governing the

growth of the films. In such cases stoichiometric TiC films could

only be deposited at very low rates (about 1.5 Å /sec). This limitation

of deposition rate in the case of the reactive evaporation process is

due to the reaction kinetics of the compound formation by this




process. The presence of a plasma in the Activated Reactive Evapo-

ration (ARE) process influences the reaction kinetics by providing

activation energy to the reactive species, thereby making it possibleto synthesize compound films at considerably higher rates[51] and

lower temperatures.

Plasma Assisted Reactive Evaporation Processes. The ARE

process which was the earliest of these processes generally involves

evaporation of a metal, or an alloy in the presence of the plasma of a

reactive gas.[52][53] For example, TiC and TiN coatings are deposited

by this process by evaporating Ti in the presence of C2H2 and N2

plasmas respectively. The two basic variants of the ARE process are

shown in Figs. 32 and 33. For more information on the other modifi-

cations of the ARE process, refer to a review by Bunshah and

Desphandey.[54] The role of the plasma is twofold:

1. To enhance the reactions that are necessary for

deposition of compound films.

2. To modify the growth kinetics, and hence, the

structure/morphology of the deposits.

Figure 32. The activated reactive evaporation (ARE) process using an electron

beam evaporator.




In the last few years, new modifications of the plasma assisted

evaporation process have been developed. The key element of theseprocesses is the very high plasma density (much higher than the prior

reactive PVD processes)—see Fig. 3. A comparison of the estab-

lished (older) processes with the new high rate evaporation tech-

niques. It may be noted that with the newer processes (Table 3), there

is almost a factor of 100 improvement in deposition rate, vapor

particle density, plasma density, and ion current density on the

substrate. A dramatic result of this high plasma density is the ability

to deposit oxides such as alumina at high rates (good economics) at

low temperatures so that substrates such as plastic webs can be

continuously coated with alumina.

These three new processes are SAD—Spotless Arc Deposi-

tion, MAD—Magnetron Activated Deposition, and HAD—Hollow

Cathode Activated Deposition.

Figure 33. The activated reactive evaporation (ARE) process using resistance

heated evaporation source.




In the SAD process, shown in Fig. 34, a dense vapor of theevaporant is produced by focusing a high power electron beam (300

kW) onto a liquid metal pool.[55] A positively biased electrode is

located above the pool. The crucible is connected to the negative of

the power supply, and a high-current discharge is initiated. The

substrate is biased to a negative potential. A major advantage of the

SAD process over cathodic arc evaporation is the absence of thediscrete cathode spots, and the corresponding droplets. Figure 35

shows the relationship between the ion current density, deposition

rate, and the impact ratio of ions and condensing vapor particles for

the SAD process, and other PVD process. Results to date show the

considerable improvement of the mechanical properties of refractory

metals.

Table 3. Typical Parameters for the deposition of compounds by

plasma activated high rate evaporation in comparison to current

plasma-activated deposition techniques. (Courtesy Dr. S. Schiller,FEP.)

Typical parameters Established High-rate

(near the substrate) technologies evaporation

deposition rate 1–

10 nm/s 100–

1000 nm/svapor particle density* 1011 cm-3 1013 cm-3

plasma density 109–1010 cm-3 1011–1012 cm-3

reactive gas pressure 10-2 Pa 1 Pa

mean free path 300 mm 3 mmbetween reactivegas particles*

ion current density 0.1–5 mA/cm2 5–50 mA/cm2

*Note: High gradient between source and substrate




Figure 34. The spotless arc deposition (SAD) process. (Courtesy Dr. S. Schiller

FEP.)

Figure 35. Relationship between ion current density jbias, deposition rate a D, and

impact ratio (alpha) of ions and condensing vapor particles for various plasma

deposition processes. (Courtesy Dr. S. Schiller, FEP.)




In the MAD process, the metal is evaporated using resistance,

or high voltage electron beam heating. A dense plasma is created

over the molten pool using a pair of opposed pulsed magnetrons.Thus, boat evaporation has been combined with bipolar pulsed

magnetron sputter technology (PMST) to develop a new plasma

activated evaporation process.[56] A schematic of the process is

shown in Fig. 36. A potential application is the deposition of alumina

on plastic webs.

Figure 36. The magnetron activated deposition (MAD) process.(Courtesy Dr. S.

Schiller, FEP.)

In the HAD process, aluminum vapor produced by an electron

beam line evaporator passes through a very dense plasma created by

a LVEB (low voltage electron beam) discharge along with oxygen

gas to deposit an alumina coating on a wide substrate. There are two

variations of the HAD process:




Figure 37. The hollow cathode deposition (HAD) process. (Courtesy Dr. S.

Schiller, FEP.)

1. A liquid vapor source such as aluminum for alumina

coatings (Fig. 37).

2. An electron beam heated silica cylinder for silicacoatings.

From Fig. 3, we note that the HAD process has the highest

plasma density of all the Plasma Assisted Reactive Evaporation

Process.[57]

4.4 Hybrid PVD Processes

Ion Plating. Ion plating is a hybrid PVD or CVD process,[58]

since the coating material is vaporized by thermal energy (i.e.,

evaporation), or momentum transfer (i.e., sputtering), or supplied as

a vapor (very similar to CVD processes). In this technique, the

vaporized (or supplied) coating materials pass through a gaseous




Figure 38. An ion plating configuration using a DC diode discharge and a

thermal vaporization source. (After Mattox.)

glow discharge on their way to the substrate, thus ionizing some of

the vaporized atoms, generally 1% (Fig. 38).

The glow discharge is produced by biasing the substrate to a

high negative potential (-2 to 5 kV) and admitting a gas, usually

argon, at a pressure of 5 to 200 m torr into the chamber, see Fig. 38.In this simple mode, which is known as diode ion plating, the

substrate is bombarded by high energy gas ions which sputter off the

material present on the surface. This results in a constant cleaning of

the substrate (i.e., a removal of surface impurities by sputtering)

which is desirable for producing better adhesion, and lower impurity

content. The ion bombardment also causes a modification in the

microstructure and residual stress in the deposit. On the other hand, it

produces the undesirable effects of decreasing the deposition rate

since some of the deposit is sputtered off, as well as causing a

considerable (and often undesired for microelectronic applications)

heating of the substrate by the intense ion bombardment. The latter

problem can be alleviated by using the supported discharge ion

plating process, where the substrate is no longer at the high negative




Figure 39. Triode ion plating using a DC supported discharge with an electronbeam evaporation source. (After Matthews and Teer.)

potential, the electrons necessary for supporting the discharge come

from an auxiliary heated tungsten filament. The high gas pressure

during deposition causes a reasonably uniform deposition on allsurfaces due to gas scattering (Fig. 39).

Reactive Ion Plating (RIP) Process. Reactive ion plating

(RIP) is very similar to the reactive evaporation process in the metal

atoms, and reactive gases react to form a compound aided by the

presence of a plasma. Since the partial pressure of the gases in

reactive ion plating is much higher (>10-2 torr) than in the ARE

process (>10-4 torr), the deposits can become porous or sooty. The

plasma cannot be supported at lower pressures in the simple diode

ion plating process; therefore, Kobayashi and Doi[59] introduced an




auxiliary electrode biased to a positive low voltage (as originally

conceived for the ARE process) to initiate and sustain the plasma at

lower pressure (about 10-3

torr). This is no different from the AREprocess with a negative bias on the substrate reported[60]much earlier

by Bunshah, which was designated by him as the biased ARE

process (BARE).

Another variation of reactive ion plating using a triode con-

figuration[61] involves injection of electrons into the reaction zone

between the electron-beam-heated evaporation source and the nega-

tively biased substrate from a heated tungsten filament transversely

to the metal vapor path. These low energy electrons are pulled across

the reaction zone by a positively biased anode located opposite to the

cathode. The arrangement is very similar to that shown in Fig. 33

except for the use of an electron-beam-heated evaporation source,

and is also very similar to triode sputtering. This adds versatility, as

well as complexity to the process through the addition of anotherprocess variable.

Murayama[62] used an electron-beam-heated source with a

negatively biased substrate and RF activation of the reactants by

means of a coil electrode of aluminum wire in the reaction zone to

deposit oxide and nitride films.

5.0 CHEMICAL VAPOR DEPOSITION (CVD)

5.1 Introduction

Chemical vapor deposition (CVD) processes are widely used

in industry due to their versatility for depositing a very large varietyof elements and compounds covering a wide range from amor-

phous deposits to epitaxial layers having a high degree of perfection

and purity.

CVD can be defined as a process in which the gaseous

chemical reactants are transported to the reaction chamber, activated

thermally (conventional CVD) or by other than thermal means (plasma




Various heating sources are used in CVD.

1. Hot plate. The substrate is in direct contact with the

hot plate which is either resistively or inductivelyheated.

2. Radiant heat. The substrate is heated by a thermal

radiation technique or optical technique (tungsten

filament lamp or laser).

3. Heating of a conductive substrate. Conductive

substrates can be heated resistively or by RFinduction.

5.3 Classification of CVD Reactions

CVD reactions fall into four general categories.

1. Thermal decomposition reactions (pyrolytic reactions). This

reaction is characterized by

AX(g) → A(s) + X (g)

where AX is a gaseous compound, A a solid material, and B a gaseous

reaction product.

Some examples of these reactions are:

B2H6(g) → B(s) + 3H2(g)

SiH4(g) → Si(s) + 2H2(g)

W(CO)6(g) → W(s) + 6CO(g)

2. Reduction reactions. In this reaction, a gaseous compound is

reduced by a reducing agent (usually hydrogen).




2 AX(g) + H2(g)→ 2A(s) + 2HX(g)

Examples of some reduction reactions are given below:

2BCl3(g) + 3H2(g) → 2B(s) + 6HCl (g)

SiCl4(g) + 2H2(g) → Si(s) + 4HCl (g)

3. Displacement reactions. These reactions are also known as

exchange reactions. In the molecule AX , X is replaced by another

element B.

AX(g) + B(g) → AB(s) + X(g)

Some representative reactions are:

Zn(g) + H2S(g) → ZnS(s) + H2(g)

SiCl4(g) + CH4 → SiC(s) + 4HCl (g)

CrCl2(g) + Fe(s) → Cr-Fe alloy + FeCl2(g)

4. Disproportionation reactions. In these reactions, the oxida-

tion number of an element both increases and decreases through the

formation of two new species. Some typical examples are:

2GeI2(g)→ Ge(s) + GeI4(g)

TiCl2(g) → Ti(s) + TiCl4(g)

Several types of reactions can be involved simultaneously in

some CVD coating processes. An example of these reactions is:

2AlCl3(g) + 3CO2(g) + 3H2(g)→Al2O3(s) + 3CO2(g) + 6HCl(g)

In this reaction, water produced from




Co2(g) + H2(g)→ CO(g) + H2O(g)

is used to form Al2O3 by the following reaction

AlCl3(g) + 3H2O(g) → Al2O3(s) + 6HCl(g)

5.4 Rate-Limiting Steps

The sequence of events in a CVD process as follows:

[63]

1. Diffusion of reactants to the surface.

2. Adsorption of reactants at the surface.

3. Surface events, such as chemical reaction, surface

motion, lattice incorporation, etc.

4. Desorption of products from the surface.

5. Diffusion of products away from the surface.

Among these steps, the slowest one is the rate-determining

step. The rate-limiting step is mainly determined by the process

parameters. The most important rate-limiting steps in the CVD

process are mass transport control, and surface kinetics control. The

latter produces uniform deposits on complex shaped substrates.

5.5 Reactors

There are two kinds of reactors most frequently used in the

CVD processes, hot wall reactor (Fig. 41) and cold wall reactor

(Fig. 42). In the former reactor, the reactor tube is surrounded by a

tube furnace making the substrate, and the reactor wall to be the sametemperature. A large number of substrates can be coated in this type

of reactor. A major drawback of this type of reactor is, deposition on

the reactor wall and possible contamination in the system from

chemical reactions between the reactor wall, and the vapor due to the

high temperature of the reactor wall. Therefore, the hot wall reactor

is ideal for the case where the reaction is exothermic, since the high




Figure 41. Hot wall CVD reactor.

Figure 42. Cold wall CVD reactor.

wall temperature prevents undesirable deposition on the reactor

walls. III–V and II–VI type semiconductors have been successfully

prepared in the hot wall reactor.In the cold wall reactor, only the susceptor where the substrates

are placed is intentionally heated by RF induction, or high radiation

lamps. This type of reactor is predominantly used for the deposition

reaction which is endothermic, such as Si deposition from the ha-

lides. Since the substrates have a higher temperature than the reactor

wall, the reaction will proceed most readily on the hot surface of the

substrate. In this reactor type, contamination due to the interaction

between the reactor wall and the vapor can be greatly reduced. Very

frequently, the walls are water-cooled to further prevent deposition

on the wall or reactions between walls and vapor.




5.6 Low Pressure CVD (LPCVD)

Low pressure CVD technology is widely used in the semicon-ductor industry due to several advantages over conventional atmo-

spheric-pressure CVD technique. By operating at lower pressure to

increase diffusibility in the gas phase, and increasingly subject the

system to surface kinetics control, uniformity of deposition is en-

hanced. The mass transfer rate and the surface reaction rate are

generally of the same order of magnitude at normal atmosphere

pressure (CCVD), while the mass transfer rate is much higher than

the surface reaction rate at lower pressure, i.e., 0.5–1 torr (LPCVD),

and thereby the rate-determining step is the surface reaction. The

transfer rates of gaseous reactants and reaction products are in-

versely proportional to pressure. If the pressure is reduced from 760

torr to 0.5–1 torr, diffusibility increases by a factor of 1000 which is

only partially offset by the increase in thickness of the boundarylayer (by the square root of pressure).[64] At this low pressure, mass

transfer cannot be the rate limiting step, and the deposition rate is

mainly controlled by surface reactions, resulting in uniform film

thickness, and properties over extended surfaces with better step

coverage and conformity, and good structural integrity with fewer

pinholes.

Another advantage of LPCVD occurs because the mean free

path is very large at lower pressures; thus wafers can be stacked on

edge instead of lying flat, and thereby a large number of wafers can

be loaded and deposited in the same run.

Polycrystalline Si films from SiH4, Si3N4 films from SiH2Cl2and NH3, and SiO2 films from SiH2Cl2 and N2O have been success-

fully prepared using this technique.[65][66]

5.7 Plasma Assisted CVD (PACVD)/Plasma Enhanced CVD

(PECVD)

Plasma assisted chemical vapor deposition (PACVD) can be

defined as a process in which the constituents of the vapor phasereact to form a solid film assisted by an electric discharge. In the




PACVD technique, the gas molecules are mainly dissociated by

electron impact generating very reactive neutral, radical, and ion

species. These reactive species arrive on a surface and react witheach other via an ionic or free radical mechanism in the film forming

process. Since the gas molecules are activated by the energetic

electrons instead of thermal energy, the reaction temperature can be

easily reduced. Films can be deposited at temperatures typically less

than 300°C. Furthermore, the inherent limitations of conventional

thermodynamics and of chemical availability in CCVD are elimi-

nated in plasma activation[67] due to the non-equilibrium nature of

the glow discharge plasma. A schematic of a microwave plasma

CVD apparatus is shown in Fig. 43.

Figure 43. Schematic of a radial-flow plasma enhanced CVD reactor.

Neutral radicals are believed to be the major deposition agents

among the reactive species generated in the plasma due to the

following two effects. First, the dissociation energy is usually lower

than the ionization energy for many gas molecules and therefore the

generation rate for radicals is generally greater than that for ions.

Second, positive ions may drift toward any surface and recombine

with electrons while neutral radicals may have more chance to stay inthe plasma and have a longer lifetime.




The number of neutral radicals is determined by a set of

variables, which are the glow discharge power, electrode spacing,

gas collision mean free path, and ion diffusibility. These variablesdetermine the electron temperature (energy states of the electrons in

the plasma), and thereby the radical generation rate. The radical

generation rate is roughly proportional to e-∆ E/kT e, where ∆ E is the

dissociation energy of gas molecules, T e is the electron temperature,

and k is the Boltzmann constant. This relation indicates that the

number of gas radicals which have higher dissociation energy will

change more with the electron temperature than those of the species

with lower dissociation energy.[66] These changes of radical ratio is

the important factor determining the film stoichiometry.

Even though ions do not contribute greatly to the film-forming

process,[67][68] the ions impinging on the film during growth signifi-

cantly affects the physical properties of the film.[69] The ionic bom-

bardment energy is determined by several factors, such as dischargepower, gas type and pressure, target bias, and frequency (RF). At

lower pressure (less than 0.05 torr), and low frequency (<< 1 MHz),

the ions can bombard onto the surface with the full energy available

to them, since the ions can follow the RF field and do not experience

appreciable scattering. The ionic bombardment energy is an average

plasma potential as the pressure and the frequency increase.

The substrate temperature still plays an important role in the

PACVD process, through the activation energy for the chemical

reaction is provided mainly by the glow discharge. During film

growth, the absorbed radical has to diffuse to a stable site to become

part of the growing film. This radical mobility on the surface of the

substrate is strongly affected by substrate temperature. The radicals

on the surface obtained more energy and diffuse to the stable siteeasily at high substrate temperature producing a denser film, while,

at low temperatures the diffusion of adatoms on the surface is much

retarded and thereby the film has more defects and a lower density.

Furthermore, the stoichiometry of the film can be influenced by

substrate temperature. In the case of plasma CVD silicon nitride, the




content of hydrogen in the grown film can be reduced greatly by

increasing the substrate temperature.

5.8 Advantages and Disadvantages of PACVD

There are several advantages of PACVD processes over con-

ventional CVD processes:

1. Ability to deposit films at much lower temperature.

2. Almost unique method to prepare the heavily

hydrogenated amorphous silicon films for solar cells.

3. Good adhesion of the films to their substrates and

the bond strength.

4. Higher deposition rates than by the CCVD

technique.

The present limitations of PACVD processes are:

1. More defects in the films and a lower density of the

film compared to a high temperature deposited film.

2. Difficulties of deposition of pure materials.

3. Extreme difficulty in controlling stoichiometry.

5.9 Advanced CVD Techniques

Hot-Filament CVD. Yasui, et al.,[70] prepared high-quality

silicon nitride films with a low concentration of hydrogen at high

deposition rates by the hot-filament CVD technique. In this tech-

nique, the hot filament (W filament), which is heated to very hightemperatures (such as 2400°C) is placed close to the substrate (8 cm).

They deposited films at the deposition rate which was one order of

magnitude higher with a smaller activation energy of the growth rate

than that of CCVD technique using the same gases. The same

techniques are now used to grow diamond films.




Laser Induced CVD (LCVD). Since the substrate is bom-

barded by charged and energetic particles, as well as by high energy

radiation in the PACVVD process, this technique can introducedeleterious effects for radiation sensitive electronic devices. Re-

cently, laser induced CVD processes have been attracting much

attention. In this process the reaction energy is selectively provided

by photons.

There are two types of processes in LCVD, pyrolytic LCVD

and photolytic LCVD. In pyrolytic LCVD, polyatomic gas molecules

are dissociated near a gas-substrate interface by localized heating of

the substrate which is exposed to the laser beam. This technique is

limited by the choice of the laser/source gas/substrate. The gas vapor

sources are required to be relatively transparent at the exciting laser

wavelength and the substrates strongly absorbing. In photolytic

LCVD, a molecule near the substrate is decomposed by means of a

photochemical reaction. Specific chemical bonds in polyatomic mol-ecules can be broken selectively through the choice of the laser

wavelength. In this technique, photodissociation of the vapor source

fixes the maximum allowable wavelength of the laser, since only

radiation which is absorbed by the reactants can lead to a photo-

chemical reaction. LCVD techniques have been successfully em-

ployed to synthesize polycrystalline Si, and Ge,[71] aluminum ox-

ide,[72] and silicon.

6.0 MATERIALS DEPOSITED BY PVD AND CVD

TECHNIQUES

In addition to the deposition of simple element species, e.g.

refractory metals by CVD for large scale integrated circuits as well

as for chemical applications, Al for metallization and decorative

coatings, etc., a number of more complex materials are deposited

using PVD and CVD techniques particularly the plasma assisted

versions as illustrated in Table 4.[74]




Table 4. Examples of some compounds synthesized and deposition

rates obtained by ARE, reactive sputtering, and PACVD processes.

ARE Reactive PACVD

Compound (Å min-1) (Å min-1) (Å min-1)

Carbides 2000–3000 400–500 150–400

TiC, HfC, ZrC, VC

Nitrides 2000–

3000 300–400 60

–150TiN, HfN, ZrN

Oxides 1000–2000 200–800 200–300

TiO2, ZrO2, Al2O3, SiO2

Sulfides 1000–2000

TiS2, MoS2, MoS3

Novel Materials

Superconducting materials 1000–1500

Nb3Ge, CuMo6S8

Photovoltaic materials 1500–2000 50–200

a-SiH, CuInS2

Optoelectronic materials 500–1000

Indium-tin-oxide,

zinc

Cubic BN 1000–1500Diamond 1000 Å h-1

Carbon 300 200

SUGGESTED READING

1. Hurkmans, T., Hauzer, F., Buil, B., Engel, K., Tietema, R., “A New

Large Volume PVD Coating System Using Advanced Controlled

Arc and Combined Arc/Unbalanced Magnetron ABStm Deposition

Techniques,” Surface and Coatings Technology, 92:62–68 (1997)




2. Hogmark, S., Hedenquist, P., Jacoson, S., “Tribological Properties

of Thin Hard Coatings: Demands and Evaluation,” Surface and

Coatings Technology, 90:247–257 (1997)

3. Stals, L. M. M., Nesladek, M., Quaeyhaegens, C., “Current Industrial

Practice—Critical Issues in Hard PVD and PA-CVD Coatings,”

Surface and Coatings Technology, 91:230–239 (1997)

4. Pochet, L. F., Howard, P., and Saeed, S., “CVD Coatings: From

Cutting Tools to Aerospace Applications And its Future Potential,”

Surface and Coatings Technology, 94/95:70–75 (1997)

5. Strafford, K. N., “Tribological Properties of Coatings—Expectations,Performance and the Design Dilemma,” Surface and Coatings

Technology, 81:106–117 (1996)

6. Prengel, H. G., Pfouts, W. R., and Santhanam, A. T., “State of The

Art in Hard Coatings For Carbide Cutting Tools,” Surface and

Coatings Technology, 102:83–190 (1998)

REFERENCES

1. Handbook on Deposition Technologies for Films and Coatings, 2nd

Ed., (R. F. Bunshah, ed.), Noyes Publications, Park Ridge, NJ

(1994)2. Pierson, H. O., Handbook of Chemical Vapor Deposition, Noyes

Publications, Park Ridge, NJ (1992)

3. Deposition Technologies for Films and Coatings, 2nd Ed., (R. F.

Bunshah, ed.), p. 463, Noyes Publications, Park Ridge, NJ

4. Schiller, S., Heisig, U., and Goedicke, K., Thin Solid Films, 40:327

(1997)

5. Teer, D. B., Surface and Coatings Technology, 36:901 (1988)

6. Musil, J., et al.,Surface and Coatings Technology, 39/40:270 (1990)

7. Sproul, W. D., et al., Surface and Coatings Technology, 43/44:270

(1990)

8. (a) Kirchoff, V., et al., German Patent No. DD252, 205, (September

1986; (b) Schiller, et al., Surface and Coatings Technology, 61:331

(1993)




9. Wilson, R. G., and Brewer, G. R., Ion Beams with Applications to

Ion Implantation, John Wiley and Sons, New York (1973)

10. Nishimura, C., et al., J. Vac. Sci. Tech. A, 5(3):363 (1987)11. Beale, H. A., and Bunshah, R. F., Proc. 4th Intl. Conf. on Vac. Met.,

p. 238, Iron and Steel Company of Japan, Tokyo, Japan (1973)

12. Glang, R., Handbook of Thin Film Technology, (L. Maissel and R.

Glang, eds.), McGraw-Hill, New York (1970)

13. Holland, L., Vacuum Deposition of Thin Films, Chapman and Hall,

London (1957)

14. Graper, E. P., J. Vac. Sci. Technol., 10:100 (1973)

15. Cocca, M. A., and Stouffer, L. H., Trans. Vac. Met. Conf., Am. Vac.

Soc., 203 (1963)

16. Morley, J. R., Trans. Vac. Met. Conf., Am. Vac. Soc., 186 (1963)

17. Bardos, L., Surface and Coatings Technology, 86/89:648 (1996)

18. Berghaus, B., German Patent No. 683,614 (1939)

19. Sabalev, L. P., et al., U.S. Patent No. 3,783,231 (January 1, 1974);

3,793,179 (February 19, 1974)

20. Dorodnov, A. M., Soviet Phys. Tech. Phys., 23:1058 (1978)

21. Osipov, V. A., et al., Soviet Rev. Sci. Inst., 21:1651 (1978)

22. Wroe, H., Br. J. Appl. Phys., 9:488–491 (1958)

23. Gilmour, A. S., Jr., Lockwood, D. L., Proc. IEEE , 60(8):977–991

(1972)

24. Snaper, A. A., Arc Deposition and Apparatus, U.S. Patent No.

3,625,848 (1971)

25. Snaper, A. A., Arc Deposition and Apparatus, U.S. Patent No.

3,836,451 (1974)

26. Sablev, L. P., Apparatus for Vacuum Evaporation of Metals Under

the Action of an Electric Arc, U.S. Patent No. 3,783,231 (1974)27. Sablev, L. P., Apparatus for Metal Evaporation Coating, U.S.

Patent No. 3,793,179 (1974)

28. Sanders, D. M., Handbook of Plasma Processing Technology, (S.

Rosnagel, J. J. Cuomo, W. D. Westwoods, eds.), p. 419, Noyes

Publications, Park Ridge, NJ (1990)

29. Martin, P. J., et al., Thin Solid Films, 153:91 (1987)




30. Boxman, R. L., Martin, P. J., Sanders, D. M., Handbook of Vacuum

Arc Science and Technology, Noyes Publications, Park Ridge, NJ

(1995)

31. Martin, P. J., Netterfield, R. P., Kinder, T. J., Thin Solid Films, 193/

194:77 (1990)

32. Vasin, A. I., Dorodnov, A. M., et al.,Sov. Tech. Phys. Lett., (English

Translation of Pis’ma Zh. T. Fiz.), 5:23–24 (1979)

33. Ehrich, H., Hasse, B., et al., Proc. 8th Intl Conf. On Discharge

Appl., pp. 591–592, 596, Essen Univ. (1985)

34. Dorodnov, A. M., Kunetsov, A. N., et al., Sov. Tech. Phys. Lett.,(English Translation of Pis’ma Zh. T. Fiz.), 5:418–419 (1979)

35. Ehrich, H., J. Vac. Sci. Technol. A, 6:143–138 (1988)

36. Meassick, S., Chan, C., and Allen, R., Thin Film Deposition

Techniques Using the Anodic Arc, to be published

37. Smith, H. M., and Turner, A. F., Appl. Opt., 4:147 (1965)

38. Sankur, H., DeNatale, J., Gunning, W., and Nelson, J. G., J. Vac.Sci. Technol. A, 5:2869 (1987)

39. Cheung, J. T., and Madden, J., J. Vac. Sci. Technol. B, 5:705 (1987)

40. Cheung, J. T., Chen, J. S., and Otsuka, N., Proc. IRIS IR Detector

Specialty Meeting, Seattle, WA, (August 1987); This work was

followed by several other similar investigations presented at the

34th National Symp. of Am. Vac. Soc., Anaheim, CA (November

1987)

41. Dijkkamp, D., Venkatesan, T., Wu, S. D., Shaheen, S. A., Jisrawi,

N., Min-Lee, Y. H., McLean, W. L., and Croft, M., Appl. Phys. Lett.,

51:619 (1987)

42. Wu, X. D., Dijkkamp, D., Olgale, S. B., Ina, A., Chase, E. W.,

Miceli, P. F., Chang, C. C., Tarascon, J. M., and Venkatesan, T.,

Appl. Phys. Lett., 51:861 (1987)

43. Cheung, J. T., and Sankur, H., Solid State and Materials Sciences,

16:63 (1988)

44. Greer, J. A., J. Vac. Sci. Technol., 10(4):1821 (1992)




45. Bunshah, R. F., Handbook of Deposition Technologies for Films

and Coatings, 2nd Ed., p. 179, Noyes Publications, Park Ridge, NJ

(1994)

46. Frach, P., Goedicke, K., Gottfried, Chr., Walde, H., and Hentsch,

W., Surface and Coatings Technology, 74/75:85 (1995)

47. Bunshah, R. F., Nimmagadda, R., Dunford, W., Movchan, B. A.,

Dernchishin, A. V., and Chusanov, N. A., Thin Solid Films, 54:85

(1978)

48. Rosa, B., and Schultz, L., Appl. Phys. Lett., 53(16):1557 (1988)

49. Humphreys, R. G., Satchell, J. S., Chew, N. G., and Edwards, J. A., Appl. Phys. Lett., 54(1):75 (1989)

50. Karim, A. A., Deshpandey, C., Doerr, H. J., and Bunshah, R. F.,

Thin Solid Films, 172:111–112 (1989)

51. Bunshah, R. F., Thin Solid Films, 107:21 (1983)

52. Bunshah, R. F., U.S. Patent No. 37,791,852 (1972)

53. Bunshah, R. F., and Raghuram, A. C., J. Vac. Sci. Technol., 9:1385(1972)

54. Bunshah, R. F., and Deshpandey, C. V.,Physics of Thin Films, (J. L.

Vossen and H. Francombe, eds.), (1987)

55. Goedicke, K., Scheffel, B., and Schiller, S., Surface and Coatings

Technology, 68/69:799 (1994)

56. Reschke, J., Goedicke, K., Schiller, Surface and Coatings

Technology, 76/77:763 (1995)

57. Neumann, M., Morgner, H., Staach, S., SVC Conf., (1996)

58. Mattox, D. M., Deposition Technologies for Films and Coatings,

(R. F., Bunshah, ed.), p. 320, Noyes Publications, Park Ridge, NJ

(1994)

59. Kobayashi, M., and Doi, Y., Thin Solid Films, 54:57 (1978)

60. Bunshah, R. F., New Trends Mater . Process, Am. Soc. Met., p. 200,(1978)

61. Matthews, A., and Teer, D. G., Thin Solid Films, 80:41 (1981)

62. Murayama, Y., J. Vac. Sci. Technol., 12:818 (1975)




63. Kern, W., and Ban, V. S.,Thin Film Processes,(J. L. Vossen and W.

Kern, eds.), p. 267, Academic Press, New York (1978)

64. Kern, W., and Ban, V. S.,Thin Film Processes, (J. L. Vossen and W.Kern, eds.), p. 274, Academic Press, New York (1978)

65. Yee, K. K., Intl. Metals Rev., 1:19 (1978)

66. Dun, H., Pan, P., White, F. R., and Douse, R. W., J. Electrochem.

Soc., 178(7):1555 (1981)

67. Coburn, J. W., and Chen, M., J. Applied Physics, 51:3134 (1980)

68. Zesch, J. C., Lujan, R. A., and Deline, V. R., J. Non-Cryst. Solids,

35–36:273 (1980)

69. Jansen, F., Plasma Deposited Thin Films, (J. Mort and F. Jansen,

eds.), p. 4, CRC Press Inc., Boca Raton, FL (1985)

70. Yasui, K., Katoh, H., Komaki, K., and Kaneda, S., Appl. Phys. Lett.,

56(10):898 (1990)

71. Andreata, R. W., Abel, C. C., Osmundsen, J. F., Eden, J. G., Lubben,

D., and Greene, J. E., Appl. Phys. Lett., 40(2):183 (1982)72. Minakata, M., and Furukawa, Y., J. Electron. Matter., 15(3):159

(1986)

73. Peters, J. W., Gebhart, F. L., and Hall, T. C., Solid State Technol.,

p. 121 (1980)

74. Bunshah, R. F., IEEE Transactions on Plasma Science, 18:846

(1990)

2. vapor deposition technologies

Documents