ION-BEAM MODIFICATION PHENOMENA AND APPLICATIONS

 Ion-Beam Mixing

Ion-beam mixing phenomena deal with compositional and structural changes in a two- or multiple-component system under the influence of ion radiation. The effect commonly occurs during sputteríng and results in changes in surface composition during depth profiling anaJysis by SIMS and AES techniques. For example, consider a thin film of A on substrate B bombarded by a beam of inert-gas ions. Typically, the ion range (/?) exceeds the escape depth of the sputtered A atoms. If R does not exceed the thickness of A, then only A atoms sputter. If, after some sputteríng, R now extends into the substrate región, the atomic displacements and interdiffusion that occur within colusión cascades will cause A and B to intermix. The mixing occurs locally at the interface and eventually links with other similarly intermixed zones to créate a continuous ion-beam mixed layer. Now B atoms also enter the stream of sputtered atoms because the combination of continued surface erosión and interfacial broaden-ing, due to ion mixing, has brought them closer to the surface.

Through the use of high-energy ion beams, mixing reactions occur over substantial dimensions. Films can, therefore, be effectively alloyed with sub-strates and layered, but normally immiscible films can be homogenized with the assistance of ion implantation. As an example, consider the multiple-layer structure consisting of alternating Au and Co films. According to the phase diagram, these elements do not dissolve in each other; but they form a uniform metastable solid solution under a flux of 3 x 1015 Xe ions/cm2 at an energy of 300 keV. Colusión cascades, ballistic effects of recoils, and defect migration during room-temperature irradiation all con¬tribute, in a complex way, to the observed mixing.

In suicides, equilibrium as well as metastable phases have been observed after mixing. Experiment has shown that the suicide thickness is both dose-and ion-species-dependent. At the same energy and dose, more mixing occurs the heavier the ion. In metal film systems, extended solubility is virtually always observed, metastable phase formation is a frequent occurrence, and amorphous phases occasionally form at cryogenic temperatures.

 Modification of Mechanically Functional Surfaces

By enhancing the ability of surfaces to resist plástic deformation, the benefits of reduced wear, less tendency to surface cracking, and greater dimensional stability are effected. In recent years there has been considerable research on the use of ion implantation to realize these desirable ends. For the case of steel, the implantation of light interstitial ions, such as nitrogen, boron, and carbón, yields considerable improvement in wear and fatigue resistance. The reason is due to the elastic interaction between dislocations and undersized interstitials; this results in their mutual attractíon and the segregation of the atoms to the defects. Long known in metallurgical circles, the interaction occurs even at room temperature and is quite effective in pinning dislocations, thus restrictíng their motion. In addition, iron nitrides, carbides, borides, etc., form when the limited solubilities of the interstitial atoms are locally exceeded; these precipí-tales are also effective barriers to dislocation movement. Since surface damage processes depend on plástic flow of surface layers, the importance of limiting dislocation motion is apparent.

An altérnate approach to improving the wear resistance of surfaces through the deposition of hard coatings was addressed at length. It is instructive to compare this approach with that of ion implantation. Although CVD deposits manage to conformally coat external as well as internal surfaces, ion implantation is limited by geometry to line-of-sight processing. CVD coatings are bonded to the substrate across an interfacial región, which is frequently a source of adhesión difficulty; on the other hand, ion-implantation modified layers are not subject to adhesión problems because no sharp inter-face exists. Since CVD deposition is conducted at elevated temperatures, the substrate is frequently heat-affected and sometimes softened in the process; ion implantation only modifies a very thin surface layer, leaving the remainder of the substrate unaffected. Lastly, thick CVD coatings several microns thick imply less stringent substrate smoothness requirements than for ion-implanta-tion processing. The latter is the only practical method available for modifying precisión surfaces while preserving extreme dimensional tolerances. Clearly, ion implantation is only cost-effective for high-value added components, such as surgical implants or dies.

The issue of the effective surface depth modified by ion implantation is an interesting one. It has been observed that ion implantation effects persist well beyond the shallow depth of the projected ion range. Implanted atoms are frequently observed considerably deeper within the substrate than can be accounted for by the geometry of wear tracks. The cause has been attributed to the generation of fresh dislocation networks that effectively trap and drag atoms deeper below the damaged surface layers. Frictional wear is also accompanied by temperature increases of as much as 600-700 °C at contacting asperities. Migration of mobile impurities is thus encouraged, especially where the dislocation density is high. Such effects provide an unexpected wear protection bonus for ion-implanted surfaces.

A number of industrial applications involving wear reduction by means of ion implantation methods is Usted in Table 13-2. Hardness and resistance to adhesive and abrasive wear are the attributes required of the assorted cutting, mechanical forming, and molding tools. In steel matrices nitrogen is a favored interstitial ion, and implanted cobalt has been explored as a means of modify-ing tungsten carbide tools. Typical doses are well into the 1017/cm2 range and impart two- to fivefold decreases in wear rate with corresponding increases in tool life. Specific examples of ion-implantation modified tools and components.

A totally different application which nevertheless exploits the benefits of enhanced hardness and reduced wear involves metallic surgical implants. Tens of thousands of titanium alloy (Ti-6 wt%-Al-4 wt% V) hip and lcnee replace-ment prostheses have already been ion-implanted with nitrogen resulting in improved tribological properties. In service, the implant moves in contact with a high-molecular-weight polyethylene mating socket, so that wear of this couple is of concern. Apparently, the formation of hard TiOz particles results in abrasión of the unimplanted alloy surface during use. Implantation produces a surface containing hard titanium nitride precipitates that effectively resists such wear. Part of the improvement in properties may be attributable to the enhanced corrosión resistance ion-implanted surfaces exhibit.  High defect concentrations promote thickening of air-formed oxide films and enhance chemical homogenization of the underlying metal. The former effect affords an added measure of surface passivation and protection, and the latter helps eliminate localized galvanic corrosión. Lastly, note that there are no practica! alternatives to modifying the surface properties of orthopedic prostheses. Unlike tools whose surfaces can tolérate CVD or PVD coatings, chemical biocompatibility with contac'ting body fluids places severe restrictions on the surface composition of surgical implants.

Ronellys Flores---CRF---libro the materials science of thin films

---

ION-IMPLANTATION EFFECTS IN SOLIDS

Ion-surface interactions have already been discussed within several contexts in this book. Sputtering for film deposition and Rutherford backscattering for microanalysis are the most important examples; typical ion energies involved are 5 keV and 2 MeV, respectively. At ion energies between these extremes, i.e., tens to hundreds of keV, the probability is great that projectile ions will be implanted hundreds to thousands of angstroms deep beneath the surface.

As a surface modification technique, ion implantation has a number of important advantages as well as disadvantages. Among the advan-tages are controllable and reproducible subsurface depth concentrations, no sacrifice of bulk properties, low-temperature processing, no significant dimen¬sional change in implanted objects, extensión of solid solubility limits, forma-tion of metastable phases, and vacuum cleanliness. Significant limitations include line-of-sight processing, shallow penetration of ions, lattice damage, and, of course, very high capital equipment and processing costs. Despite the latter drawbacks, ion implantation is not only indispensible in VLSI process¬ing, but its use has been explored as a means of beneficially modifying virtually every surface property of interest.

Modification of surfaces occurs because the newly implanted distributions of chemical species are accompanied by considerable structural disorder. Some-times it is possible to induce compositional change without appreciable struc-tural modification, e.g., in the case of low-dosage implants followed by thermal annealing. Alternatively, crystalline targets can be disordered struc-turally and even made amorphous by implanting ions that are identical to matrix atoms. In this case no chemical change is effected. Frequently, how-ever, both compositional and structural changes are inseparably linked and serve to broaden the number of possible ways surfaces can be modified. Some choice over the extent of modification can be exercised through control of processing variables.

 Energy Loss and Structural Modification

The collisions of an individual energetic ion in a solid cause the motion of atoms and the excitation of electronic states. At the outset the ions primarily induce "gentle" electronic transitions that cause relatively little structural damage. Nevertheless, the electronic structure is excited by Coulomb interac-tions with moving ions. A relatively narrow lightning-like trail surrounds the ion track and defines a región of intense electronic excitation, e.g., ionization, secondary and Auger electrón production, electron-hole pair formation, lumi-nescence, etc. Through these mechanisms of energy loss the ion slows suffi-ciently until it begins to set in motion violent nuclear collision cascades along its trajectory. These cascades are the result of displaced atoms that dislodge yet other atoms so that a jagged branched trail is produced.

From the standpoint of structural modification, matrix damage and atomic relocation following nuclear collisions are cardinal issues. In addition to the cascades referred to earlier, which can affect a considerable portion of the matrix, depending on the extent of the branch overlap, there are also "spikes." When the density of energy deposition in either electronic or nuclear cascades exceeds a certain level, the result is a thermal spike. These are launched when bombarding particles transfer energies of a few hundred eV to lattice atoms with the virtually instantaneous liberation of heat. In Cu, calculation has shown that the spike will heat a sphere of approximately 20 A in diameter to the melting point within 5 X 10"12 sec. In another 3 X 10"" sec, the temperature decays to 500 °C, quenching the motion of some 1000 atoms in the process (Ref. 16). Small, highly energized cascade regions melt and quench too rapidly to grow epitaxially within the surrounding crystalline material. As a result, amorphous zones form. In silicon these have been found to be more resistant to formation of defect-free material on low-tem-perature annealing than thin amorphous-Si surface layers or films formed by láser melting or by physical vapor deposition. In this sense the amorphous cascade or spike regions have a different defective nature.

Ronellys Flores---CRF---libro the materials science of thin films

---

Láser Surface Alloying (LSA)

When a thin metal film A on a metal substrate B is exposed to láser radiation, the combination can be alloyed through melting to yield a new modified surface layer. This LSA process can be understood with reference to the schematic cross-sectional views. A láser pulse causes film A to melt and the resulting liquid/solid front sweeps past the original A-B interface; interdiffusion of film and substrate atoms occurs as irradiation terminates. The máximum melt depth is reached, where the atomic mixing is quite vigorous. Resolidification then begins and the solid-liquid interfacial velocity, which is initially zero, increases very rapidly. Interdiffusion continúes within the liquid but the resolidified metal cools so rapidly that atoms are immobilized and frozen in place.

 The final result is an alloy of nominal composition A^B,^ which is not necessarily homogeneous. As an example, consider the surface alloying of a thin Au film on a Ni substrate. Typical Q-switched láser pulses genérate total melt times (tM) ranging from 50 to 500 nsec and produce atomic mixing over the diffusion distance ~ 2 \jDtM. The diffusivity of atoms in liquid metáis is rarely outside the range of 10~5 to 10~4 cm2/sec so that compositional change can be expected over a distance of 140-1400 A. The higher estímate roughly agrees with the Au concentration profile data .  In this example the total melt depth is about 4500 A and considerably exceeds the initial Au thickness. If, on the other hand, tM is lengthened to ~ 50 f¿sec, a situation arising when using a cw-C02 láser pulse, then the molten pool is deeper and the diffusional length exceeds 4500 A. Complete homogenization clearly occurs in this case.

Considerable LSA experimentation has been conducted on assorted binary alloy systems. They include (1) those in which the two components are mutually soluble in both liquid and solid states, e.g., Cr-Fe, Au-Pd, W-V; (2) those where there is appreciable liquid solubility (miscibility) but limited solid solubility, e.g., Cu-Ag, Au-Ni, Cu-Zr; and (3) those that exhibit both liquid and solid phase immiscibility, e.g., Pb-Cu, Ag-Ni, Cu-Mo. Category 1 lends itself to thermodynamically favored interdiffusional mixing. Except for rare  or  expensive  alloying  elements,   however,   láser  processing  in  these systems offers few advantages over conventional bulk alloying processes. Thermodynamic obstacles to mixing in category 3 are not easily overeóme even with LSA methods. 

Only when the films are thin enough and the melt temperature high enough is there the chance that a single-phase liquid will form, which then can be quenched to retain metastable phases. Otherwise, predictable phase separation will oceur. Intermedíate category 2 offers the greatest potential for quenching in metastable and amorphous phases. It is this class of binary systems which had been previously studied by vapor-quenching methods over a decade earlier. The objective was the same as LSA —to extend solubility of terminal phases, freeze in metastable phases and produce amorphous phases by suppressing crystallization. Láser scanning methods offer the best means of achieving these ends over large surface áreas.

Ronellys Flores---CRF---libro the materials science of thin films

---

LÁSER MODIFICATION EFFECTS AND APPLICATIONS

Regrowth Phenomena in Silicon 

In this section attention is directed to the structural and compositional property changes produced in silicon surface layers as a result of láser processing. Silicon has been singled out as the vehicle for discussion because of the large volume of study devoted to this important material. Furthermore, many of the phenomena observed in Si can be readily understood in the context of traditional solidification and recrystallization theories that have evolved over the past four decades.

Impurity-Free Si. Láser melting of single-crystal Si wafer sur-faces results in liguid phase epitaxial (LPE) regrowth. However, when ultrashort picosecond pulses are applied, crystalline -* liquid -► amorphous Si transitions can be sequentially induced.

The phenomenon of solid phase epitaxial (SPE) regrowth of amorphous silicon layers upon surface annealing is worth noting. As we shall see later, ion implantation methods can be used to "amorphize," or make amorphous, surface layers of Si. 

The latter can be recrystallized by láser annealing, in what amounts to a second surface modification treatment. Better control over SPE can be exercised, however, by means of simple furnace annealing. The result is a well-defíned constant planar regrowth velocity with thermally activated kinetics given by where a is the atomic spacing and v is the lattice frequency. The activation energy (E) for impurity-free SPE is 2.35 eV, a valué that corresponds to cooperative bond breaking and rearrangement at the moving amorphous-crys-talline interface. In its wake, dangling bonds are eliminated and interfacial bond straining and distortion are minimized. In contrast to the several m/sec laser-induced solidification rates, SPE regrowth proceeds at a velocity of only ~ 1 A/sec at 500 °C—a difference spanning some 10 orders of magnitude. Epitaxial regrowth rates of Si-implanted amorphous Si are strongly dependent on substrate orientation with (100) and (111) exhibiting the highest and lowest magnitudes, respectively.

 Doped Si

Epitaxial Regrowth. Depending on the type and concentration of impurity atoms, and the thermal processing parameters, a rich assortment of regrowth effects is possible. Consider what happens to an implanted distribution of Bi atoms in Si after irradiation by a 100-nsec ruby láser pulse with an energy density of 2 J/cm2. the distribution, originally centered 1500 A deep, is swept toward the surface during solidification. The reason for this has to do with two facts. The first is that when a liquid and solid are in equilibrium at some temperature, solute generally has greater solubility in the liquid phase; second, solute atoms diffuse rapidly in the liquid phase but are essentially immobile in the solid. When a planar liquid-solid front now passes by the implanted solute, successive partitioning occurs between the two phases in an attempt to maintain a fixed (equilibrium) solute concentration ratio; i.e., k0 = CS/CL, at the interface. Also variably known as the segrega-tion, partition or distribution coefficient, k0 valúes depend on the dopant in question; it can, for example, range from 2.5 X 10~5 for Au to 0.8 for B in Si. A computer simulation of the solidification sequence of events for the case of a solute with k0 = 0.1, where regrowth proceeds at a rate of 2 m/sec. The process resembles zone refining, a technique employed to purify rods of material by directionally sweeping impurities and concentrating them at one end by means of a moving molten zone.

A recent intriguing result is the demonstration of the solid-state analog of zone refining in an amorphous Si layer doped with implanted Au. During epitaxial regrowth of Si, the amorphous Si layer shrinks in extent as the interface moves toward the surface . Simultaneously, Au preferen-tially partitions into the amorphous phase and is concentrated there leaving a virtually Au-free epitaxial Si región behind. Interestingly, Au solubilities in amorphous Si many orders of magnitude greater than in crystalline Si, were measured.

Ronellys Flores---CRF---libro the materials science of thin films

---

LASERS AND THEIR INTERACTIONS WITH SURFACES

Láser Sources

The intense scientific and engineering research associated with the develop-ment of lasers has resulted in much innovation and rapid growth of applica¬tions. Space limitations preclude any discussion of the details of the theory of láser construction, operation, and applications, which are all covered ad-mirably in other textbooks. Suffice it to say, that all lasers contain three essential components: the lasing médium, the means of excitation, and the optical feedback resonator.

The most common lasers employed in materials processing contain either gaseous or solid-state lasing media. Gas lasers include the carbón dioxide (C02:N2:He), argón ion and xenón fluoride excimer types. The solid-state varieties used are primarily the chromium-doped ruby, the neodymium-doped yttrium-aluminum-garnet and neodymium-doped glass láser. These solid-state lasers are excited through pumping by incoherent light derived from flash lamps. Gas lasers, on the other hand, are excited by means of electrical discharges. Láser excitation may be continuous or cw, pulsed, or Q-switched to provide the different output powers shown schemati-cally. 

The distinctions in these power-time characteristics are important in the various materials processing applications. In the welding and drilling of metáis, for example, advantage is taken of the power-time profile in the pulsed and Q-switched lasers. Both the reflectance and the thermal diffusivity of metáis decrease with increasing temperature. Therefore, the high-power leading edge of these lasers is used to preheat the metal and enhance the efficieney of the photon-lattice phonon energy transfer.

the common lasers employed in surface processing together with their pertinent operating characteristics are listed. Among the important láser properties are spatial intensity distribution, the pulse width, and pulse repetition rate. The spatial distribution of emitted light depends on the cavity configuration with Gaussian (TEM^) intensity profiles common. Because a uniform láser flux is desirable in surface processing, methods have been developed to convert emission modes into the "top-hat" spatial profile. The dwell time or pulse length, rp, ranges from less than 10 nsec to 200 nsec for Q-switched lasers, and many orders of magnitude longer for other types of lasers. Repetition rates for pulsed and switched lasers range from one in several seconds to many thousands per second. Although the low repetition rates of Q-switched lasers may not be practical in industrial processing applications because the duty eyele, is low, they are useful for laboratory research.

It is the magnitudes of both the absorbed radiant power and rp that determine the effective depth of the surface layers modified through melting or redistribution of atoms. Generally, the smaller valúes of rp result in submicron melt depths. Melting and extensive interdiffusion over tens to hundreds of microns occur with the longer irradiation times possible with cw lasers. No single láser spans the total range of accessible melt depths. 

 Láser Scanning Methods

Practical modification of large surface áreas with narrowly focused láser beams necessarily implies some sort of scanning operation. For cw lasers the surface generally rotates past the stationary beam in a manner reminiscent of a phonograph record past a needle. Through additional x-y motion, radial positioning and choice of rotational speed, a great latitude in transverse velocities (v) is possible. This also means a wide selection of interaction or dwell times, íd, given by td = dm/v, where dm is the effective melt trail diameter. Typically, td ranges between tens of microseconds to hundreds of milliseconds. In this case the surface-modified región appears to consist of a chain of overlapping elliptical melt puddles.

The experimental arrangement for processing using pulsed or Q-switched lasers. In this case, discrete, overlapping circular-mod-ified (melted) regions are generated by a train of láser pulses. For área coverage larger than the individual melt spots, a mechanism for ráster scanning must be provided. This is usually accomplished by computer-controlled x-y stepping of substrates.

Thermal Analysis of Láser Annealing

The substrate heating caused by an incident láser pulse is due to electronic excitation processes accompanying the absorption of light. Typical pulse durations of 1 nsec or longer far exceed the relaxation time for electronic diffusion length, 2 y/KdTp.

 Therefore, it is permissible to assume that the thermal history of the irradiated sample can be modeled by continuum non-steady-state heat-conduction theory. The fundamental equation for the tempera-ture T(x, t) that has to be solved is where the first two terms representing conventional one-dimensional heat conduction should be familiar to readers. The term A(x, t), in units of W/cm3, is the spatial and time-dependent power density absorbed from the incident láser pulse. Other quantities which appear are p, the density, c, the heat capacity, n, the thermal conductivity, and x and t, the distance measured from the surface into the interior and time, respectively. Depending on the relative valué of the absorption length, a~] cm, of the láser light within the specimen surface, two limiting regimes of thermal response can be distin-guished.

Ronellys Flores---CRF---libro the materials science of thin films

---

Modification of Surfaces and Films

Two main approaches to improving or altering the surface properties of solids have evolved over the years. The more traditional one involves the deposition of films and coatings from solid, liquid, and vapor sources. Processing utilizing these methods has totally dominated our attention in the book until this point. But there is another approach of more recent origin based on modifying existing surfaces through the use of directed-energy sources. These include photon, electrón, and ion beams, and it is their interaction with surfaces.

Coherent (láser) and incoherent light sources, as well as electrón beams, tnodify surface layers by heating them to induce melting, high-temperature solid-state annealing or phase transformations, and, occasionally, vaporization. In the case of lasers, the relation between the required power density and irradiation time is depicted for a number of important commercial processing applications. However, the focus of this book is thin films and in the applications shown much thicker layers of material are modified. These materials processing techniques will, therefore, not be discussed in any detail, ñor will there by any additional mention of electrón beams. 

Their heating effects are basically equivalent to those produced by lasers of comparable power. Furthermore, the great depth of the heat-affected zone is more typical of bulk rather than surface processing. The thin-film or layer-modification regime we shalJ be concerned with is characterized by approximate láser energies of ~ 0.1-2 J/cm2, interaction times of ~ 10~9 to 10~6 sec, and power densities of ~ 106 to 108 W/cm2. These conditions prevail in the indicated región. Surface layers ranging from 0.1 to 10 ¡im in thickness are correspondingly modified by melting under such conditions. The melting-solidificatión cycle frequently does not restore the surface structure and properties to their original states. Rather, interesting irreversible changes may occur. For example, one consequence of láser processing can be an ultrahigh quench rate with the retention of extended solid solutions, metastable crystalline phases, and, in some cases, amorphous materials. Directed thermal energy sources have also been employed to effect annealing, surface alloying, solid-state transformations and homogenization. The controlled epitaxial re-growth of molten Si layers over Si02 or insulators, is an important example of the great potential of such processing.

Like photon and electrón beams, ion beams play an indispensible role in surface analytical methods and have also achieved considerable commercial success in surface processing. In the very important ion-implantation process, ion beams have totally revolutionized the way semiconductors are doped. Depending on the specific ion projectile and matrix combination, dopants can be driven below the semiconductor surface to readily predictable depths through control of the ion energy. Unlike traditional diffusional doping where the highest concentration always occurs at the surface, ion-implanted distribu-tions peak beneath it. The reduction of the threshold voltage required to trigger Láser processing regimes illustrating relationships between power den-sity, interaction times and specific energy. D-drilling; SH-shock hardening; LG-laser glazing; DPW-deep penetration welding; TH-transformation hardening.

current flow in MOS transistors, by means of ion implantation, ushered in battery-operated, handheld calculators and digital watches. Today ion-implan-tation doping is practiced in MOS as well as bipolar transistors, diodes, high-frequency devices, optoelectronic devices, etc., fabricated from silicon and compound semiconductors. Achievements in microelectronics encouraged broader use of ion implantation to harden mechanically functional surfaces, improve their wear and fatigue resistance, and make them more corrosión resistant. Critical components such as aircraft bearings and surgical implant prostheses have been given added valué by these treatments. In addition, there are other novel ion-beam-induced surface-modification phenomena such as ion-beam mixing, or subsurface epitaxial growth, that may emerge from their current research status into future commercial processes.

The purpose of this chapter is to present the underlying principies of the interaction of directed-energy beams with surfaces, together with a description of the changes which occur and why they occur. Accordingly, the subject matter is broadly subdivided into the following sections:

Lasers and Their Interaction with Surfaces

Láser Modification Effects and Applications

Ion-Implantation Effects in Solids

Ion-Beam Modification Phenomena and Applications

Ronellys Flores---CRF---libro the materials science of thin films

---

Oxidation and Oxide Films

The universal response of metal surfaces exposed to oxygen-bearing atmo-spheres is to oxidize. The oxidation product may be a thin adherent film that protects the underlying metal from further attack, or a thicker porous layer that may flake off and offer no protection. In this section, discussion is limited to oxidation via high-temperature exposure; aqueous corrosión oxidation phenom-ena are already the subject of a broad and accessible literature. From the standpoint of thermodynamics all of the structural metáis exhibit a tendency to oxidize. As noted in Chapter 1, the driving forcé for oxidation of a given metal depends on the free-energy change for oxide formation. What thickness of oxide will form and at what rate are questions dependent on complex Icinetics and microstructural considerations, and not on thermodynamics. two simultaneous processes occur during oxidation.

 At the metal-oxide interface neutral metal atoms lose electrons and become ions that migrate through the oxide to the oxide-ambient interface. The released elec-trons also travel through the oxide and serve to reduce oxygen molecules to oxygen ions at the surface. If metal cations migrate more rapidly than oxygen anions (e.g., Fe, Cu, Cr, Co), oxide grows at the oxide-ambient interface. On the other hand, oxide forms at the metal-oxide interface when metal ions diffuse more slowly than oxygen ions (e.g., Ti, Zr, Si). An important implication is that highly insulating oxides, such as A1203, Si02, do not grow readily because electrón mobility, so central to the process, is low. This is what limits their growth and results in ultrathin protective native oxide films.

The model of growth Icinetics developed for oxidation of Si, is applicable to other systems. Both parabolic oxide growth under diffusion-controlled conditions, as well as linear oxide growth when interfaciai reactions limit oxidation, are frequently observed. However, not all oxidation processes fit the aforementioned categories, and other growth rate laws have been experimentally observed in various temperature and oxygen pressure regimes. 

 Thermal Coatings 

Ever-increasing demands for improved fuel effíciency in both civilian and military jet aircraft has continually raised operating temperatures of turbine engine components. Among those requiring protection are turbine blades, stators, and gas seáis. The metáis employed for these critical applications are Co-, Ni-, and Fe-base superalloys, which possess excellent bulk strength and ductility properties at elevated temperatures. A widely used cost-effective way to achieve yet higher temperature resistance to degradation in the hot gas environment is to employ an additional thermal barrier coating (TBC) system. This consists of a metaUic bond coat and a top layer composed primarily of Zr02. The bond coating, as the ñame implies, is the glue layer between the base metal and the outer protective oxide. Its function is not unlike that of a bond or primer coating used to prepare surfaces for paintíng. Typical bond coatings consist of MCrAlY or MCrAlYb, where M = Ni, Co, Fe. 

Original bond coating compositions such as Ni-26Cr-6Al-0.15Y (in wt%) have been continually modifíed in an effort to squeeze more performance from them. The role of Y or other rare earth substitutes is critical. These elements apparently protect the bond coat from oxidation and shift the site of failure from the base metal and coat interface to within the outer thermal barrier oxide. Just why is not known with certainty; it appears that these reactive metáis easily diffuse along the boundaries of the plasma-sprayed particles of the bond coating, oxidize there, and limit further oxygen penetration.

The use of Zr02 is based on a desirable combination of properties: melting point = 2710 °C, thermal conductivity = 1.7 W/m-K, and thermal expansión coefficient = 9 X 10~6 K~'. However, the crystal structure under-goes transformation—from monoclinic to tetragonal to cubic—as the tempera-ture increases, and vice versa, as the temperature decreases. A rapid, diffu-sionless martensitic transformation of the structure occurs in the temperature range of 950-1400 °C accompanied by a volume contraction of 3-12%. The thermal stresses so generated lead to fatigue cracking, which signifies that Zr02 alone is unsuitable as a TBC. 

The ZrOz overlayers are generally stabilized with 2-15 wt% CaO, MgO, and Y203. Through alloying with these oxides, a partially stable cubic structure is maintained from 25 °C to 2000 °C. Actually the tetragonal and monoclinic phases coexist together with the cubic phase, whose stabilization depends on the amount of added oxide. Cubic phase stabilization results in stress-induced transformation toughening, which can be understood as follows. If a crack front meets a tetragonal particle, the latter will transform to the monoclinic phase a process that results in a volume increase. The resultant compressive stresses blunt the advance of cracks, toughening the matrix.

Ronellys Flores---CRF---libro the materials science of thin films

---

TRIBOLOGY OF FILMS ANO COATINGS

Tribology is the study of phenomena associated with interacting surfaces in relative motion. Adhesión, hardness, friction, wear, erosión, and lubrication are among some of the scientific and technological concerns of this important subject. From an engineering standpoint three sepárate categories of behavior can be distinguished based on the relative magnitudes of friction and wear :

1. Friction and wear are both low. This is the case in bearings, gears, cams, and slideways.

2. Friction is high but wear is low. This combination of properties is desired in devices that use friction to transmit power, e.g., clutches, belt drives, tires.

3. Friction is low and wear of one contacting body is high. This is the situation that prevails during material removal processes such as machining, cutting, and grinding. In these operations plástic deformation is involved, whereas in the first two categories the contacting bodies are generally elastically stressed.

We now explore the concepts of friction and wear in more detail.

 Friction

AJÍ engineering surfaces are rough and characterized by a density of projec-tions or asperities with some distribution of heights. Surfaces may be thought of as producing contact with each other at the summits of the asperities. The sum of these local contact áreas represents the real área of contact, which may only be but a fraction of the apparent geometric contact área. In the majority of contacts surface films are present with properties different from the underlying bulk material. Surface films may be gases, fluids (e.g., oil), or deposited solid layers such as graphite, metáis, or ceramic coatings. In we consider three cases of contact between rough surfaces with an intervening film layer . In the first, the film thickness is less than the roughness so that some asperities pierce the film enabling contact between bulk bodies . the film (2) is still thin but of the order of roughness dimensions. All contacts are now between the upper body and the film . the film is thicker, and all contacts are also of 1/2 type. Surface roughness is not expected to play a marked role here.

The dependence of frictional effects on the film thickness is of considerable technological importance. Consider a soft solid film such as lead on steel. As the film thickness increases, / decreases from fa (steel-steel) to fb (lead-steel), reflecting the changing nature of contact. Beyond this thickness, T essentially stays the same but P, which is related to the effective film hardness changes. Thin lead films are harder than thicker ones because of the support provided by the steel. Therefore, P decreases as the film thickness increases, leading to an increase in fb or /. This causes a friction variation in film thickness that passes through a mínimum. The same effect occurs with thin fluid films used to lubrícate metal surfaces in contact. The well-known Stribeck curve then displays a minimum in the coefficient of friction as a function of thickness or equivalent quantity (viscosity • speed/pressure). As the film thick¬ness increases, boundary, quasi-hydrodynamic and hydrodynamic lubrication regimes are successively operative.

An important set of applications involves the dry lubrication of moving parts where fluid lubrication is not possible, at high and low temperaturas or pressures. Thin, low-friction solid films composed of chalcogenides, oxides, fluorides, or carbón are used instead. In particular, MoS2 is a favored lubricant and has been deposited by reactively sputtering MoS2 targets in an H2S ambient.

 Wear Mechanisms

Wear may be defined as the progressive removal of material from surfaces that are under load in relative motion. Several different mechanisms have been identified to characterize contact wear, and these are briefly described below.

Adhesive Wear.

Adhesive wear occurs when applied tangential forces cause fracture between surfaces bonded at asperities. One option is for the fracture path to follow the original microwelded interface. Other paths lie above or below the interface when the strength of the bonds between asperities exceeds the cohesive strength of the bodies in contact. The result is material transfer, usually from the softer to harder body. During subsequent surface motion cycles, these particles may eventually be removed by fatigue fracture. In more severe cases of adhesive wear, smearing, galling, and seizure of surfaces may occur.

Abrasive Wear.

Abrasive wear is a form of cutting wear where the material is removed by hard wear particles, by hard asperities, or by hard particles entering the interface from the environment.

Fatigue Wear.

Fatigue wear occurs in situations where there is repeated loading and unloading of surfaces in contact. Failure may initiate at both surface flaws or cracks or at subsurface inhomogeneities. Crack growth eventually results in detached wear particles.

Fretting Wear.

Fretting wear may be viewed as a type of fatigue wear that occurs under conditions of oscillatory movement of small amplitude (in the range of 1-200 ¿un), but relatively high frequency. Many sequential damage processes occur during fretting, including breakup of protective films, adhesión and transfer of material, oxidation of metal wear particles, and nucleation of surface cracks.

Delamination Wear.

Delamination wear takes the form of regu¬lar detachments of thin platelike particles from wearing surfaces due to the influence of high tangential (friction) forces in the surface contact zone. During cyclic loading the cracks that develop propágate parallel to the surface at a depth governed by the material properties and coefficient of friction.

 Oxidation Wear.

Oxidation wear arises from the continuous rub-bing and removal of surface films produced by reaction with the environment. Wear damage is modest in this case because oxide regrows soon after it ís lost. However, at high temperatures where chemical reactions are accelerated oxidation wear is aggravated.

Ronellys Flores---CRF---libro the materials science of thin films

---

HARDNESS AND FRACTURE

 Hardness

Hardness is an important material property of concern in films utilized for electronic and optical as well as mechanically functional applications. It affects wear resistance and plays an important role in the friction and lubrication of surface films in contact. Hardness, a complex property related to the strength of interatomic forces, apparently depends on more than one variable. Hard materials are generally modeled by a deep potential energy well with steep walls. These characteristics imply a combination of high cohesive energy and short bond length. Measures of the cohesive energy are the heat of sublimation and enthalpy of compound formation (A//298). If some account is taken of the bond length by dividing the cohesive energy by the molar volume of the material, then the correlations with hardness shown in result. It is apparent that the hardest materials are covalently bonded, and that increasing the ionic character of the bond leads to reduced hardness. At a microscopic level, directional bonds more readily resist distortion and rupture by concentrated loads than do ionic bonds.

The hardness of a material is usually defined as its resistance to local plástic deformation. Since the hardness test consists of pressing a hard indenter into the surface, it is equivalent to performing a highly localized compression test. A correlation between hardness H and yield strength ay is therefore expected. Typically H « 3ay in bulk metáis, but the correlation has not been directly verified in films or hard coatings. It is well known that ay for bulk materials decreases with temperature, a fact that facilitates hot mechanical forming operations. Therefore, it can be anticipated that the hot hardness of coatings will be lower than that at ambient temperatures for which valúes are usually quoted. It is these hot hardness valúes that are importam in thermal coatings and in applications such as machining. 

 Hardness Testing

Hardness testing of films and coatings is a relatively simple (though decep-tively so) measurement to perform. The most frequently employed methods are modifications of techniques having long standing in the metallurgical commu-nity. The Vickers hardness test, also known as the diamond pyramid hardness (DPH) test, employs an indenter consisting of a square-based diamond pyramid ground to have a face angle of 136°. Hv, the Vickers hardness number, is obtained as the ratio of the applied load L to the surface área of the resulting indentation.

 Effect of Microstructure on Hardness

A fundamental cornerstone of materials science is the relationship between microstructure and properties. Following Sundgren and Hentzell connections between the hardness of films and coatings and various microstruc-tural characteristics will be discussed.

What does apparently affect the hardness and strength of refractory com-pounds is the perfection of the grain boundaries. Porosity and fine microcracks are very deleterious to such coatings and lower their strength and hardness significantly. Metal films are, however, somewhat more tolerant of such defects, whose influence can be blunted by plástic deformation effects. Increas-ing the substrate temperature is the simplest and most common way to reduce the grain-boundary defect structure and enhance the hardness of these com-pounds. The effect can be rather dramatic. For example, an increase in substrate temperature from 100 to 600 °C raised the hardness of magnetron-sputtered TiN from 1300 to 3500 kg/mm2. Elevated temperatures eliminate void networks and evidently promote the strengthening of grain boundaries. The grain size increases as does hardness, contrary to the Hall-Petch prediction. 

Metastable Structures. 

Metastable phases are frequently ob-

served in refractory compound films. As in the case with metastable metal

alloy films, high deposition rates and low substrate temperatures are conducive

to the formation of nonequilibrium structures and a fine grain size. Manifesta-

tions of the metastability are the incorporation of C and N in interstitial lattice

sites and the generation of supersaturated solid solutions. This generally occurs

during PVD rather than CVD, which is usually carried out under conditions

closer to thermodynamic equilibrium. The incorporated interstitials tend to

distort the lattice and the subsequent difficulty in initiating dislocation motion

is reflected in increased hardness. The effect can be quite large. Hardnesses of

2500 to 3500 kg/mm2 are generally found in reactively sputtered HfN films

compared to 1600 in CVD-grown films. Apparently the bombardment of the

growing HfN film by the sputtering gas forces N into tetrahedral interstitial

positions. This creates a high compressive stress in the plañe of the film and

thus a higher hardness.

In general, metastable phases and structures can be frozen in up to tempera-tures of 0.37^ (TM is the melting point). As a consequence, metastable hard coating systems can be used at temperatures up to 550-800 °C (for TM = 2500-3300 °C) 

Impurities.

 Since hard PVD coatings are generally grown in médium vacuum or under even higher pressure ambients, the incorporation of noble gases, C, N, and O from residual gases, and impurities from chamber hardware and walls is not uncommon. The deposit impurities are located in both substitutional, interstitial as well as grain-boundary sites at total levéis up to a few atomic percent. Even at such low concentrations, the effect on hardening can be pronounced. The mechanism of hardening due to impurities apparently involves the electrostatic attachment of the latter to charged disloca-tions in ionic materials and to dangling dislocation bonds in covalent com-pounds. Such interactions limit dislocation mobility by pinning effects.

Film Texture. 

By texture we mean the preponderance of one (or more) crystallographic planes oriented parallel to the film surface compared with the case of randomly oriented planes. In the latter, isotropic behavior may be expected. Films grown by PVD and CVD techniques usually display a preferred oríentation, however, with low índex planes lying parallel to the substrate surface, creating a texture that is strongly dependent on virtually all deposition and process variables. Factors of as much as 2 in hardness have been observed in cubic coatings (e.g., TiC, TiN, ZrC) as a function of the preferred oríentation plañe [e.g., (111), (100) and (110)]; similar hardness anisotropy on basal (0001) and prísmatic (1100) planes of hexagonal materials (e.g., WC, SiC) has been reponed.

 Fracture

It is a fact of nature that materials that are extremely hard are simultaneously brittle and prone to fracture. The phenomenon of fracture is of cardinal concern in a great many materials engineering applications; once fracture of a component or structure occurs, other issues quickly assume secondary impor-tance. Films and coatings are no exception. For example, with the exception of oxidation wear, all wear mechanisms are based on some sort of crack development that creates new surfaces, from which particles can be detached by fracture processes. In a similar vein, high-temperature fracture or spalling of coatings leaves the underlying substrate unprotected and at the mercy of harsh corrosive atmospheres.

At the outset it is important to distinguish between brittle and ductile fracture. The latter occurs after the material has undergone some plástic deformation. Metáis tend to undergo ductile fracture upon overloading. Brittle fracture, on the other hand, occurs rapidly, wíthout warning and in such a way that the broken pieces can usually be neatly fitted together. Generally, brittle fracture occurs more readily in materials having small tensile strengths com¬pared with their compressive strengths. Thus, glasses, ceramic oxides, and covalent, as well as hard metal compounds, to a lesser extent, are particularly prone to brittle fracture.

Two approaches to fracture of coatings will be presented next. The paramount assumption is that the materials involved possess an intrinsic collection of structural flaws distributed laterally as well as through the coating thickness. Voids, porosity, interconnected voids or porosity, voided or grooved grain boundaries, local regions of de-adhesion, etc. may be viewed, in a broader context, as flaws and even incipient cracks. Under either external or residual internal stressing, each flaw will locally concéntrate stress and the surrounding material will tend to deform. If the stresses are sufficiently large, they can ultimately destroy the coating by crack propagation if tensile, and by wrinkling or buckling if compressive.

Ronellys Flores---CRF---libro the materials science of thin films

---

HARD COATING MATERIALS

 Compounds and Properties

Hard coating materials can be divided into three categories, depending on the nature of the bonding. The first includes the ionic hard oxides of Al, Zr, Ti, etc. Next are the covalent hard materials exemplified by the borides, carbides, and nitrides of Al, Si, and B, as well as diamond. Finally, there are the metallic hard compounds consisting of the transition metal borides, carbides, and nitrides. Typical mechanical and thermal property valúes for important representatives of these three groups of hard materials are Usted in Table 12-1. The reader should be aware that these data were gathered from many sources (Refs. 1-6) and that there is wide scatter in virtually all reported property valúes. Differences in processing (e.g., CVD, PVD, and sintering of powders), variations in structure (e.g., grain size, porosity, density, defects) and composition (e.g., metal-nonmetal ratio, purity), to-gether with statistical error in measurement, contribute to the uncertainties. Perusal of this tabulated information leads to the following broad conclusions:

1. All of these compounds have extremely high hardnesses. This can be appreciated by noting that heat-treated tool steel has a hardness of about Hu = 850. Hardness is the most often quoted material property of hard coatings. Therefore, Section 12.3 has been specially reserved for an extensive discussion of the concept of hardness, the technique of measure-ment, and the significance of its magnitude in coatings.

2. These compounds have very high melting points and decomposition temper-atures. For example, the decomposition temperatures of TaC, HfC, and diamond exceed the melting point for tungsten (MP = 3410 °C).

3. The modulus of elasticity is lowest for the ionic solids. In comparison, only the stiffest metáis have modulus valúes overlapping those of the oxides Usted.

4. The linear thermal expansión coefficient generally increases in going from the covalent to metallic to ionic hard compounds. Metáis tend to have thermal expansión coefficients that are higher by approximately a factor of two or more than these hard compounds.

5. The thermal conductivity of the hard metallic and covalent compounds is comparable to that of the transition metáis and their alloys. Good metallic electrical conductors have proportionally higher thermal conductivities. Ceramic oxides are the poorest thermal conductors.

The last two properties have important implications for the properties and performance of coatings. An important source of coating residual stress is the thermal contribution generated by the difference in expansión between coating and substrate. The illustrative problem dealing with TiC on steel is worth reviewing and indicates the magnitude of possible effects.

The susceptability to cracking of coatings subjected to varying temperature histories is an important limitation to the performance of thermal coatings. To see how thermal cracking can occur, consider the rapid cooling of a high-tem-perature component. The surface coating contracts more than the interior, which is still relatively hot. As a result, the surface forces the interior into compression and is itself stretched in tensión. 

Ronellys Flores---CRF---libro the materials science of thin films

---