Martin P.M. Handbook of Deposition Technologies for Films and Coatings, Third Edition: Science, Applications and Technology

Подождите немного. Документ загружается.

Evaporation 221

Figure 4.56: Microstructure of dispersion strengthened Ni–ZrO

alloy bef ore and after exposure

at 1300



Cfor5h[31].

subsequent gas-phase nucleation and condensation.) The negative effects of the presence of a

high gas density on the kinetic energy and the mobility of adatoms on the deposit surface can

be overcome by biasing the substrate [211, 212] and/or heating the substrate to a higher

temperature [213].

222 Chapter 4

Figure 4.57: Structure of TiC deposits at various substrate temperatures [206].

4.9.2 Texture

The texture of evaporated deposits is, in general, dependent on deposition temperature. At low

deposition temperatures, a strong preferred orientation is generally observed: {211} in iron

[203], {220} in TiC [206], and {0002} in Ti [214]. As deposition temperature increases, the

texture tends to become more random. In the case of beryllium [29], the texture changed to a

{110} orientation at high deposition temperatures. The presence of a gas tends to shift the

preferred orientation to higher index planes [215]. For silver, increasing the substrate bias

changes the preferred orientation from {111} to {200} and back to {111} [201]. The

inclination angle affects the texture of the ﬁlm. According to Xu et al., when inclination angle

Evaporation 223

Figure 4.58: Surface morphology of the MgO e-beam evaporated ﬁlms, with different inclination

angles between source and the substrate [207].

increases, out-of-plane texture grows more quickly. However, when the angle is decreased,

in-plane texture shows limited grain alignment, whereas better alignment was observed at

higher angles [207].

4.9.3 Residual Stresses

Residual stresses in deposits are of two types. The ﬁrst kind arises from the imperfections built

in during growth (the so-called growth stresses). An increase in deposition temperature

produces a marked decrease in the magnitude of this stress [206, 216, 217]. The other source

of residual stress is due to the mismatch in the coefﬁcient of thermal expansion between the

substrate and the deposit. Its magnitude and size depend on the values of the thermal

expansion coefﬁcients as well as the thickness and size of the substrate and deposit. In order to

compare residual stresses, the thermal stress component should be subtracted from the

measured values [216]:

thermal

= α · T ·

1 − v

(4.10)

where E

/(1 − V

) is the biaxial modulus for the ﬁlm, α is the difference in the coefﬁcient of

thermal expansion between the substrate and ﬁlm material, and T is the difference in

temperature between deposition temperature and room temperature.

The inﬂuence of a negative bias on the substrate produces a compressive stress in the deposit,

which reaches a maximum value at −200 to −300 V DC bias and then decreases [212].

224 Chapter 4

High residual stresses can cause plastic deformation (buckling or bending), cracking in the

deposit or the substrate, or cracking at the substrate–deposit interface. The latter can be

minimized by grading the interface, i.e. producing the change in material over a ﬁnite distance

instead of producing it abruptly at a sharp interface. A graded interface can be produced by

gradually changing the deposition conditions or by interdiffusion, which is enhanced by higher

substrate temperature or increased kinetic energy of the vapor species.

4.9.4 Defects

Let us next consider the ‘defects’ found in vapor-deposited materials. The ﬁrst one is classiﬁed

as a spit, or small droplet ejected from the molten pool, which lands on the substrate and is

incorporated into the coating [218]. An example is shown in Figure 4.59. The composition of

the droplet is different from that of the coating in the case of an alloy and can therefore be the

site of corrosion initiation. The bond between the droplet and the surrounding material is

usually poor. Hence, corrosion attack can proceed down the boundary to the substrate or

undermine the coating. The spit may also fall out, leaving a pinhole behind which can act as a

stress concentrator and limit the ductility or the uniform elongation of a sheet material. Spits or

pinholes do not affect the yield strength or reduction of area in a ductile material, but they can

be stress raisers and sites for fatigue-crack initiation. Both spits and foreign particles on the

substrate surface induce preferential growth of the deposit in that area because of higher

exposure to the vapor ﬂux than the general growing interface. This region of preferential

growth is termed a ﬂake; typical ﬂakes are shown in Figure 4.60. There is marginal bonding

Figure 4.59: Vapor source droplet (spit) defect in M–Cr–Al–Y coatings. (a, b) Defects overcoated

with additional material; (c) fatigue crack initiated at spit [218]. (Courtesy of American Institute

of Physics.)

Evaporation 225

Figure 4.60: Flake defects in (a) and (b) produced by accelerated coating deposition on foreign

particles. Glass bead peening incorporates ﬂake into the coating (c) or knocks it out and forms a

pit (d) [218]. (Courtesy of American Institute of Physics.)

between the ﬂake and the deposit, which can lead to formation of a pit or crack, or to

nucleation of corrosive attack.

Spits can be suppressed by eliminating porosity, oxide inclusions, and compositional

inhomogeneities in the evaporant source material, since spitting can be caused by included-gas

release or by the release of bound gas through thermal decomposition. In e-beam evaporation,

the beam of electrons dissipates energy over a path extending as much as a mil (25 ␮m) or

more into the melt. If this energy is delivered at a rate faster than the coating material can

accommodate by evaporation, conduction, or radiation, a pocket of vapor forms and spitting

occurs. Spits are also caused by gas pockets included in the evaporant rod that suddenly

226 Chapter 4

expand when rapidly heated by the beam. Non-metallic inclusions can trap pockets of

superheated vapor below them, which can erupt in a shower of molten droplets. Spits can be

avoided by using a high-purity vacuum melted rod as the evaporant. Flake formation can be

avoided by avoiding the presence or impingement of foreign particles on the substrate

(primarily by substrate surface cleaning and good housekeeping of the deposition apparatus).

Deep grooves or ridges on the substrate can also produce ﬂake-type defects by shadowing

adjacent regions of the specimen surface.

Another type of defect occurs in complex alloys [218] such as M–Cr–Al–Y (where M can be

nickel, cobalt, or iron), where even at deposition temperatures of 955

◦

C, the deposit

morphology corresponds to the ﬁbrous transition zone between zones 1 and 2. The grain

boundaries in this morphology are weak, causing intergranular corrosive attack (Figure 4.61).

The problem can be obviated by increasing the adatom mobility through the use of a higher

substrate temperature or specimen bias of about −200 V, or by using a postcoating process that

consists of a room temperature high-intensity glass bead peening followed by a

high-temperature anneal in hydrogen. Compound rotation of the specimen, which exposes

higher surface irregularities to varying angles of impingement of incoming vapor atoms,

produces a signiﬁcant decrease in the number and size of open, columnar defects.

Another problem in deposits of complex alloys is due to the variation in deposit chemistry

attributable to segregation in the ingot and large pool temperature variations caused by the

ﬁnite size of the e-beam [218, 219]. Improved ingot quality, development of improved e-beam

sources, and decrease in the temperature gradient at the crucible walls by using crucible liners

or coolant of lower heat capacity, such as NaK, instead of water would minimize this problem.

Figure 4.61: SEM photomicrograph of impact fracture surface of as-deposited overlay coating.

Fracture is intercolumnar indicating weak boundaries [218]. (Courtesy of American Institute of

Physics.)

Evaporation 227

In a more recent investigation on the origin of defects and continuing on the above [220],it

was found that spits in M–Cr–Al–Y type alloys consist of ejected pool material exhibiting

enrichment in impurity elements of low vapor pressure as a result of superheating of

non-metallic particles (carbides or oxides) in the melt initiating the ejection of pool material.

Flakes, generally cone shaped, were found to originate at non-metallic particles loosely

attached to the surface. Leader formation was found to be weakly dependent on the angle of

incidence of the arriving vapor ﬂux. Both ﬂakes and leaders seem to be enhanced by

preferential growth and shadowing phenomena.

Wolfe et al. deposited titanium carbide/chromium carbide layers by co-evaporation of the

transition metal and carbon simultaneously using e-beam PVD [221]. Figure 4.62 shows the

spit defect on the surface of the TiC/Cr

multilayer coating. According to Figure 4.62, the

obtained microstructure is columnar but the surface of the ﬁlm is not faceted. Porosity has

been observed between the columnar grains. This suggests that the columnar grains are dense

multilayers of chromium and titanium carbides.

Figure 4.62: Surface morphology of titanium carbide/chromium carbide layer showing the spit

defect [221].

228 Chapter 4

4.10 Physical Properties of Thin Films

The Handbook of Thin Film Technology [7] contains an extensive section on the electrical and

electronic conduction, piezoelectric and piezoresistive, dielectric and ferromagnetic properties

of thin ﬁlms. The reader is referred to it.

4.11 Mechanical and Related Properties

4.11.1 Mechanical Properties

4.11.1.1 Mechanical Property Determination

A number of testing techniques have been used to determine the strength properties of thin

ﬁlms. They include the high-speed rotor test [222], the bulge test [223–227], microtensile

testing machines of the soft [228–231] and the hard categories [223–227], and even ﬁxtures

which can be operated in the electron microscope [232, 233]. Hoffman [234, 235] reviewed

the test techniques and the reader could do no better than to read Hoffman’s articles or the

original references. The basic handling problem encountered with the preparation and

mechanical property testing of thin ﬁlm specimens is much less severe with thick ﬁlms for

which many of the standard test specimens, machines, and techniques can be readily used.

Therefore, the spectrum of mechanical properties measured on thick ﬁlms is much broader

than with thin ﬁlms.

4.11.1.2 Tensile Properties Of Thin Films

The tensile properties of thin ﬁlms have been reviewed [234–237]. As Hoffman concludes

[234], the data reported are not very consistent even on the same material. The reader is

advised to consult the references for details. In general, the observed strength of

vapor-deposited metal ﬁlms consists of three parts:

OBS

= σ

Bulk

+ σ

Imperfections

+ σ

Thickness

(4.11)

where σ

OBS

is the inherent strength level of bulk polycrystalline material in the annealed state,

Imperfections

is the contribution due to point defects in excess of those normally found in the

bulk annealed state, and σ

Thickness

is the contribution arising from the smallest dimension of the

ﬁlm and its limiting effect on grain size such that dislocation multiplication and migration are

impeded [230].

Table 4.10 gives the strength properties of thin ﬁlms of some metals and compares them to

bulk values [235]. In many cases, the strengths are about 200 times those of annealed bulk

samples and three to ten times those of hard drawn samples. The tensile strength values are

given numerically as well as by fractions of the shear modulus. The ductility of the

high-strength ﬁlms is very limited, which is similar to the behavior of high-strength ﬁbers or

Evaporation 229

Table 4.10: Strength of properties of thin ﬁlms

Material Structure Maximum

tensile strength

Shear

modulus

Strain at

fracture

Thickness

dependent

Ref.

(kg/mm

) (%)

Au Bulk hard-drawn 28

G/114

––

(111) crystal ﬁlm 81 G/36 1.2 No [79]

(100) crystal ﬁlm 27 G/110 0.5 Yes [141]

(110) (111)

Polycrystalline ﬁlm

49 G/59 1 No [148]

Polycrystalline ﬁlm 55 G/54 0.7 Yes [139]

(100) crystal ﬁlm 26 G/115 3.5 No [151]

Polycrystalline ﬁlm 32 G/92 2.3 No [151]

Ag Bulk hard-drawn 37

G/75 – – –

Polycrystalline ﬁlm 59 G/47 0.7 Yes [128]

Polycrystalline ﬁlm 42 G/68 0.3–0.4 – [30]

Cu Bulk hard-drawn 49

G/98 – – –

Polycrystalline ﬁlm 93 G/51 1.8 Yes [196]

Polycrystalline ﬁlm 88 G/54 – No [145]

Rolled foil 19 G/256 10–15 No [198]

Ni Bulk cold-rolled 125

G/67 – – –

Polycrystalline ﬁlm 210 G/40 1.8 Yes [149]

Al Bulk cold-rolled 16

G/171 – – –

Polycrystalline ﬁlm 42 G/66 0.5–0.8 – [30]

Shear moduli from AIP Handbook, 1957 [237].

Bulk tensile strength from Handbook of Chemistry and Physics, 1961 [238].

Fracture strain not quoted for bulk material.

whiskers. A principal point of contention is whether the ultimate tensile strength is a function

of the ﬁlm thickness or not. The discrepancy also appears to be dependent on the test method

used, i.e. between the bulge test and tensile test. In many cases, it appears that the strength

decreases as the ﬁlm thickness increases from approximately the 200–300

A range to about the

2000–4000

A range. At the greater thickness, the strength is about the same as that of heavily

worked bulk material. There are several papers relating the strength properties of thin ﬁlms to

the ‘crystallite size’ and ‘block structure’ as inﬂuenced by the deposition temperature, stress,

recovery, and recrystallization process [184, 240–248]. One manifestation of this is the

phenomenon of creep or plasticity in room temperature tensile tests as exhibited by an

irreversible initial loading curve but almost reversible unloading and reloading curves as long

as the previous stress level is not exceeded. An example of this is shown in Figure 4.63 from

Neugebauer [229] as the change in slope of the stress–strain curve. The possibility that this

change in slope is related to an elastically soft measurement or to creep in the cementation of

the grips cannot altogether be discarded.

230 Chapter 4

Figure 4.63: Typical stress–strain curve for thin ﬁlm.

Long-term creep rates have been measured and for gold they vary from 10

−7

to 10

−4

min

−1

depending on load, dimensions and the amount of prestrain [229]. The estimates of the relative

elastic and plastic extension at fracture vary from completely elastic to an almost even mixture

of elastic and plastic deformation.

Fracture in ductile gold single crystal ﬁlms [236] results from a localized plastic deformation

with resultant thinning of the ﬁlm and a rise in stress level. Eventually the smaller cracks

formed in this manner join to cause fracture. The dislocations – necessary for the deformation

– are not the grown-in dislocations but those which nucleate and multiply in discontinuous

regions. Most observations show no necking prior to fracture. The maximum stress appears to

correspond to that needed to propagate cracks from ﬂaws existing in the specimen. In

polycrystalline nickel, the fracture is the ‘clean cleavage’ type [230].

4.11.1.3 Mechanical Properties of Thick Condensates and Bulk Deposits

Table 4.11 lists the mechanical properties of thick deposits of metals, alloys, refractory

compounds, and laminated structures. In many cases, the mechanical test data are quite

extensive showing yield strength, ultimate tensile strength, hardness, and ductility as a

function of grain size, deposition temperature, and test temperature. One of the features of the

data is that the properties of thick deposits of metals and alloys are very similar to those of

wrought materials which are produced by the conventional processes of melting, casting,

mechanical working, and heat treatment.