Chapter 2

LITERATURE REVIEW

Literature review relevant to the present study has been divided into three parts.
First part contains a comprehensive review of the existing literature on electro
conductive materials and their characteristics. In the second part, literature review
concerning the protective coatings and Cold spray (CS) process has been described.
The studies related to the behavior of the copper coatings have been reviewed in the
third part of this chapter.

PART-I

2.1 ELECTRO-CONDUCTIVE MATERIALS

Electrical resistivity (also known as resistivity, specific electrical resistance, or

volume resistivity) is a measure of how strongly a material opposes the flow of
electric current. A low resistivity indicates a material that readily allows the
movement of electric charge. It is commonly represented by the Greek letter ρ (rho)
and its SI unit is the ohm metre (Ω. m).

Electrical conductivity or specific conductance is the reciprocal quantity, and

measures a material's ability to conduct an electric current. It is commonly
represented by the Greek letter σ, and its SI unit is Siemens per metre (S·m−1):
σ = 1/ ρ

2.1.1 Resistivity of various materials

The conductivity of a solution of water is highly dependent on its concentration of

dissolved salts, and other chemical species that ionize in the solution. Electrical
conductivity of water samples is used as an indicator of how salt-free, ion-free, or
impurity-free the sample is; the purer the water, the lower the conductivity (the higher
the resistivity). Conductivity measurements in water are often reported as specific
conductance, relative to the conductivity of pure water at 25 °C.
13
The effective temperature coefficient varies with temperature and purity level of the
material. The 20 °C value is only an approximation when used at other temperatures.
For example, the coefficient becomes lower at higher temperatures for copper, and the
value 0.00427 is commonly specified at 0 °C.

The extremely low resistivity (high conductivity) of silver is characteristic of metals.

George Gamow (1947) tidily summed up the nature of the metals' dealings with
electrons in his science-popularizing book, One, Two, Three...Infinity: "The metallic
substances differ from all other materials by the fact that the outer shells of their
atoms are bound rather loosely, and often let one of their electrons go free. Thus the
interior of a metal is filled up with a large number of unattached electrons that travel
aimlessly around like a crowd of displaced persons. When a metal wire is subjected to
electric force applied on its opposite ends, these free electrons rush in the direction of
the force, thus forming what we call an electric current". Table 2.1 shows the
resistivity, conductivity and temperature coefficient of various materials at 20 °C (68
°F).

The cold spray process offers advantages to produce, with minimum thermal
exposure, coatings from functional materials such as thermoelectric, magneto-caloric,
photo-voltaic, piezo-electric, super-magnetic, and high-temperature superconductive
formulations.

Electrical resistivity of CS Al was higher than the bulk Al sample measured, and
anisotropic. These characteristics can be attributed to intrinsic defects within Al splats
(e.g. high dislocation density) and oxidized interfaces between splats. Anisotropy, on
initial approximation, could be attributed to the quasi-lamellar microstructure
containing more interfaces per unit length through the thickness of the coating versus
in-plane. Nevertheless, examining the data from Choi (2007) on the thickness
dependence of resistivity (as specimen thickness approaches splat size this effectively
confines electrons to a single interrupted conductive path). This concept was fully
explored in Sharma (2006); the imperfect interfaces between splats (interface normal
perpendicular to substrate normal) have a non-trivial effect on anisotropy. It could
further be argued that these interfaces are less well-bonded than those above and
below particles, but this assertion must be certified with direct measurements.

14
 Table 2.1: Resistivity, conductivity and temperature coefficient of various materials
Temperat
ρ [Ω·m] at σ [S/m] at 20 ure
Material Reference
20 °C °C coefficient1
−1
[K ]
Silver 1.59×10−8 6.30×107 0.0038 Serway, 1998; Griffiths, 1999
Copper 1.68×10−8 5.96 × 107 0.0039 Griffiths, 1999
Annealed Copper2 5.80 × 107
Gold3 2.44×10−8 4.52 × 107 0.0034 Serway, 1998
4 −8
Aluminium 2.82×10 3.5 × 107 0.0039 Serway, 1998
Calcium 3.36x10−8 0.0041
Tungsten 5.60×10−8 0.0045 Serway, 1998
Zinc 5.90×10−8 0.0037 http://physics.mipt.ru/
Nickel 6.99×10−8 0.006
Lithium 9.28×10−8 0.006
Iron 1.0×10−7 0.005 Serway, 1998
Platinum 1.06×10−7 0.00392 Serway, 1998
Tin 1.09×10−7 0.0045
Lead 2.2×10−7 0.0039 Serway, 1998
Titanium 4.20x10−7 X
Manganin 4.82×10−7 0.000002 Giancoli, 1995
Constantan 4.9×10−7 0.000008 O’Malley, 1992
Mercury 9.8×10−7 0.0009 Giancoli, 1995
Nichrome5 1.10×10−6 0.0004 Serway, 1998
Carbon (amorphous) 5-8×10−4 −0.0005 Serway, 1998; Pauleau, 1997
2.5-5.0×10−6
⊥ basal
Carbon (graphite)6 plane Pierson, 1993
3.0×10−3 //
basal plane
Germanium7 4.6×10−1 −0.048 Serway, 1998; Griffiths, 1999
7 −1 Physical properties of sea
Sea water 2×10 4.8
water
Drinking water8 0.0005 to 0.05
Deionized water9 5.5 × 10−6 Pashley et al., 2005
7 2
Silicon 6.40×10 −0.075 Serway, 1998
Glass 1010 to 1014 ? Serway, 1998; Griffiths, 1999
Hard rubber approx. 1013 ? Serway, 1998
Sulphur 1015 ? Serway, 1998
Air 3 to 8 × 10−15 Pawar et al., 2009
Paraffin 1017 ?
Quartz (fused) 7.5×1017 ? Serway, 1998
PET 1020 ?
Teflon 1022 to 1024 ?

1
The numbers in this column increase or decrease the significant portion of the resistivity. For example, at 30 °C (303 K), the
resistivity of silver is 1.65×10−8. This is calculated as Δρ = α ΔT ρo where ρo is the resistivity at 20 °C (in this case) and α is the
temperature coefficient.
2
Referred to as 100% IACS or International Annealed Copper Standard. The unit for expressing the conductivity of nonmagnetic
materials by testing using the eddy-current method. Generally used for temper and alloy verification of aluminium.
3
Gold is commonly used in electrical contacts because it does not easily corrode.
4
Commonly used for high voltage power lines
5
Nickel-Iron-Chromium alloy commonly used in heating elements.
6
Graphite is strongly anisotropic.
7
Corresponds to an average salinity of 35 g/kg at 20 °C.
8
This value range is typical of high quality drinking water and not an indicator of water quality.
9
Conductivity is lowest with mono-atomic gases present; changes to 1.2 × 10-4 upon complete de-gassing or to 7.5 × 10-5 upon
equilibration to the atmosphere due to dissolved CO2.
15
2.1.2 Resistivity density products

In some applications where the weight of an item is very important resistivity density
products are more important than absolute low resistivity- it is often possible to make
the conductor thicker to make up for a higher resistivity; and then a low resistivity
density product material (or equivalently a high conductance to density ratio) is
desirable. For example, for long distance overhead power lines— aluminum is
frequently used rather than copper because it is lighter for the same conductance.

Silver, although it is the least resistive metal known, has a high density and does
poorly by this measure. The calcium and the alkali metals have the best products, but
are rarely used for conductors due to their high reactivity with water and oxygen.
Aluminum is far more stable. Table 2.2 shows the resistivity-density products for
various materials.

Table 2.2: Resistivity density product for various materials

Material Resistivity [nΩ·m] Density [g/cm³] Resistivity-density product [nΩ·m·g/cm³]
Sodium 47.7 0.97 46
Lithium 92.8 0.53 49
Calcium 33.6 1.55 52
Potassium 72.0 0.89 64
Aluminum 26.50 2.70 72
Copper 16.78 8.96 150
Silver 15.87 10.49 166

PART-II

2.2 PROTECTIVE COATINGS

In a wide variety of applications, materials have to operate under severe conditions

such as erosion, corrosion and oxidation in hostile chemical environments. Therefore,
surface modification of these components is necessary to protect them against various
types of degradation (Pawlowski, 1995).

Only composite materials are able to meet such a demanding spectrum of

requirements, the base material provides the necessary mechanical strength and
coatings provide a way of extending the limits of use of materials at the higher
16
temperatures (Sidky and Hocking, 1999; Li et al., 2003). Even if the material
withstands environmental degradation without a coating, the coating enhances the
lifetime of the material.

A coating is a layer of material, formed naturally or synthetically or deposited

artificially on the surface of an object made of another material, with the aim of
obtaining required technical or decorative properties. Heath et al. (1997) have
summarized the main advantages of coatings as follows:
 Very high flexibility concerning alloy selection and optimization for specific
resistance to corrosion environments and particle abrasion/erosion. Surface
properties can be separated from required mechanical properties of the structural
component.
 Coating systems (multi-layered or functionally graded) can be used, combining,
for example, good adhesion with optimized corrosion and erosion behavior.
 Unique alloys and microstructures can be obtained with thermal spraying which
are not possible with a wrought material. These include continuously graded
composites and corrosion resistant amorphous phases.
 Costs of a coating solution are normally significantly lower than those of a highly
alloyed bulk material; thermal spray coatings are especially interesting for their
cost/performance ratio.
 Thermal spray coatings additionally offer the possibility of on-site application and
repair of components, given a sufficient accessibility for the sprayer and his
equipment. However, thermal spraying is preferred, whenever possible, to achieve
optimum results.

2.2.1 Use of Coatings for various applications

The growth of power-generation facilities throughout the world has been

unprecedented. With this expansion, tremendous of effort is expended behind the
scenes to keep these power generating facilities running smoothly. These facilities
face numerous corrosion and/or wear issues, and maintenance must regularly be
performed on various machines and systems. In an effort to reduce downtime, thermal

17
spray technology had emerged to extend the life of power-generating components and
systems. Coatings manufactured by this technology are being used throughout the
power-generating industry in applications such as water pumps, conveyor screws,
boiler tubes, coal crushers and contacting elements of electro-technical applications.

2.2.1.1 Pump Repair-Housings, Impeller Fins, Seal Sections, and Wear Rings
Pumps are used in almost every facet of a power-generating facility and must often
endure abrasion as well as cavitations wear. The double suction pump is commonly
used to move river water through a power plant. As river water commonly contains
fine sand and even small stones, several sections of a pump can be attacked including
the impeller fins, the pump housing, the impeller’s seal section, and the wear ring.
The fins of the impeller are abrasively worn by the fine sand and the small stones and
broken down by cavitations. Over time, pump efficiency will be reduced. Similarly,
the housing of the pump also faces wear from the sand and/or rocks as water is
pumped through it. If left unchecked, the housing will eventually wear away to the
point where the pump may rupture. In both cases, the cold spray solution is to apply a
very hard and wear-resistant tungsten carbide coating onto the fins.

The seal section of the impeller shaft undergoes abrasive wear when fine sand slips
into the packing material and scores the journal. If the journal becomes too worn,
water will eventually penetrate beyond the seal section and start corroding the
bearings. An effective solution is to undercut the journal, cold spray a chromium
oxide coating, and finish the journal to size.

Lastly, the wear rings begin to erode as fine sand flows through the gap between the
impeller and the housing. As it wears and widens, the efficiency of the pump is
reduced. These wear rings can be efficiently reconditioned by applying bronze onto
the rings and machining them to size.

2.2.1.2 Coal Crusher Roll Repair-Journals and Seal Sections

In this case, a coal crusher required repair on the seal section and the bearing section.
A quick analysis of the seal section reveals that the coal dust generated during
grinding penetrated the gap between the packing material and the journal, imbedding

18
itself into the packing material and creating abrasion on the journal. The solution was
to undercut the section, cold spray chromium carbide, and then finish the journal to
the size.

2.2.1.3 Reconditioning an abrasively worn conveyor screw

Conveyor screws are used in power plants to transport limestone into the boilers.
Conveyor screw manufactured from carbon steel, needed to be replaced/repaired once
a year due to the abrasion of the limestone. The thermal spray solution was to apply a
thin layer of wear-resistant tungsten carbide on the shaft and both sides of the flights
using the Cold spray system.

Table 2.3 lists the general coating material that are cold sprayed for different
applications.

Table 2.3: General coating material criterion for different applications (Sanderson, 2007)
Application Coating Material Industry Sector
 Cd- plating alternative. Al alloys  Aerospace
 Corrosion mitigation  Oil and gas
 Controlled potential coatings  Petrochemical
 Pb-free bearings e.g. connecting Al, Cu alloys  Automotive
rods, turbochargers  Motorsport
 Aerospace
 Thermal management e.g. power Cu, Al, Cu-W  Electronic
hybrid devices, switchgear
 Conductive tracks  Automotive
 Corrosion mitigation Ti, Ta, Nb, Ni, Cr,  Oil and gas
Fe, Mo  Petrochemical
 Power generation
 High temperature corrosion and Ni alloys,  Aerospace
oxidation mitigation e.g. gas turbines MCrAlYs  Power generation

 Biocompatible coatings for medical Ti  Medical

devices

2.2.1.4 Contacting elements for electro-technical applications

Important structural elements of power engineering systems are tips of connecting
cables and connecting plates. The contact between the copper wire of the transformer
and the aluminum tip of the cable of the electric mains is a typical situation as well.
Under the action of atmospheric moisture and electric current, intense electrochemical
oxidation processes occur in such a contact pair, which increases the resistance of the

19
contact and leads to contact and circuit breakdown. Aqueous solutions of acids, alkali,
and salts are electrolytes, i.e., liquids capable of conducting the electric current
(Papyrin et al., 2008). The problem is also common with the contact of copper wires
and brass terminals in automotive batteries. To prevent oxidation of contacting
elements, it is necessary to avoid the presence of different materials in contacts. The
same may be achieved by coating copper powder on aluminum tips/terminals by cold
spray process.

A variety of engineering problems have been solved using thermal spraying

applications. The use of thermal spraying ranges across many manufacturing
processes, from the automotive (Nakagawa et al., 1994; Nicoll, 1994) through to the
space exploration industry (Nguyentat et al., 1992). Metals, carbides and cermets are
the most widely used coating materials; however the spraying of polymers has also
been researched (Varacalle et al., 1996; Kawase and Nakano, 1996). The HVOF
coatings have been widely used in various engineering components for combating
wear and corrosion including propellers, pump impellers and casings, super-heaters
and pre-heaters of boilers, valve bodies/trim and pipe systems (Tan et al., 2005).

Eliminating the deleterious effects of high temperature on coatings and substrates like
oxidation, evaporation, melting, crystallization, residual stresses, debonding, gas
release and other common problems for traditional thermal spray methods, offers
significant advantages and new possibilities for cold spray for thermal and electro-
conductive applications.

2.3 COLD GAS DYNAMIC SPRAY (CGDS)/ COLD SPRAY (CS)

PROCESS

The literature on Cold Spray that deals with strategies and technologies for effectively
optimizing a cold spray process is quite vast. Results of research in the following
basic areas are presented: modeling and analysis of cold spray, effect of process
parameters, interaction of high-speed particle with the substrate, bonding mechanism,
heat treatment effects and technologies and applications.

20
2.3.1 Process modeling of CS process

A detailed analysis shows that particle velocity is known to be one of the most
important parameter affecting the coating properties. Modeling of cold spray process
so as to improve the process capability and reliability for variety of coating
applications has been made. A brief analysis of some research carried out is presented
in this section:
A detailed analysis of the effects that the type of the carrier gas, the inlet gas
temperature and the shape of the cold-spray nozzle have on the impact velocity of the
feed-powder particles is carried out by Dykhuizen and Smith (1998) using an
isentropic, one-dimensional gas-flow model.

In the work by Grujicic et al. (2003), the analysis of Dykhuizen and Smith (1998) was
extended in order to include the effects of finite values of the particle velocity and the
effect of variability of the gas/particle drag coefficient. While the one-dimensional
model of Dykhuizen and Smith (1998) has been found to be quite successful, its
numerical nature does not enable an easy establishment of the relationships between
the gas, process and feed-powder parameters on one side and the gas and the particle
velocities at the nozzle exit and the velocity at which particles impact the substrate
surface, on the other. The research also provides analytical equations that can be used
to estimate the gas dynamics of the cold-spray process. Equations are also presented
that allow calculation of the particle velocity. It is shown how the spray particle
velocity depends on particle size and density, gas stagnation pressure, total gas
temperature, gas molecular weight, and nozzle shape. Use of the equations derived in
this research allows determination of an optimal nozzle shape given the gas
conditions, particle properties, and nozzle length. However, it is shown that the spray
particle velocity is relatively insensitive to the nozzle shape. Thus, a single nozzle can
be used for a variety of operational conditions.

An empirical equation for particle velocity based on the experimental data was
derived by Alkhimov et al. (2001). Numerical and experimental research of wedge-
shaped nozzles shows different types of nozzle dimensions for a given type of

21
particles that produces the maximum possible particle velocity at the moment of
impact on a target surface.

Grujicic et al. (2003) were able to employ a one-dimensional model to predict the
particle velocity at nozzle exit and particle behavior upon impact with a substrate
during CGDS. One-dimensional models generally used to analyze the dynamics of
dilute two-phase (feed-powder particles suspended in a carrier gas) flow during the
cold-gas dynamic-spray process require the use of numerical procedures to obtain
solutions for the governing equations. The researchers also found that a relatively
simple function is defined which relates the gas velocity at the nozzle exit with the
nozzle expansion ratio and the carrier gas stagnation properties.

Numerical analysis for the accelerating behavior of spray particles in cold spraying
using a computational fluid dynamics program, FLUENT was conducted by Li and Li
(2005). The optimal design of the spray gun nozzle is achieved based on simulation
results to solve the problem of coating for the limited inner wall of a small cylinder or
pipe. They found that the nozzle expansion ratio, particle size, accelerating gas type,
operating pressure, and temperature are main factors influencing the accelerating
behavior of spray particles in a limited space. Owing to the axi-symmetrical
characteristic of the flow in this study, a two-dimensional model was used in this
paper.

Jen et al. (2005) reported the effect of friction on the gas dynamic flow through the
Convergent-Divergent (CD) nozzle. The effect of particle size on the acceleration of
the particles was also reported. The acceleration process of micro scale and sub-micro
scale copper (Cu) and platinum (Pt) particles inside and outside De-Laval-type nozzle
is investigated. A numerical simulation is performed for the gas-particle two phase
flows with particle diameter ranging from 100 nm to 50 μm, which are accelerated by
carrier gas nitrogen and helium in a supersonic De-Laval-type nozzle. The carrier gas
velocity and pressure distributions in the nozzle and outside the nozzle are illustrated.
Raletz et al. (2006) began with an overview of critical particle velocity under cold
spray conditions. The system described in study makes it possible to simultaneously
measure particle velocity and number of rebounds. They concluded that the particle
velocities calculated with the one-dimensional isentropic model under the same

22
operating conditions are close to the values found in the literature. The difference
between predicted and experimental values arises from the use of real powders in the
experiments and not ideal ones used in the model.

The comparison between the Convergent-Barrel (CB) and Convergent-Divergent

(CD) nozzle was made by Li (2006). Nitrogen (N2) was employed as driving gas at a
pressure of 1.4 M Pa and a temperature of 460 0C, and the spherical Cu particles of 20
μm in diameter were used. Author found that the main factors influencing
significantly the particle velocity and temperature include the length and diameter of
the barrel section, the nature of the accelerating gas its operating pressure and
temperature, and the particle size when using a CB nozzle. Particles achieve a
relatively lower velocity but a higher temperature using the CB nozzle than a CD
nozzle under the same gas inlet pressure. The gas flow rate, and subsequently the
operating cost for cold spraying, could be reduced.

Jodoin et al. (2006) described a framework for cold spray modeling and validation
using an optical diagnostic method. In this study, an axi-symmetric two-dimensional
mathematical model is presented which was used to predict the flow inside a cold
spray nozzle as well as the particle velocity in the vicinity of the nozzle exit. The
model results are compared with those obtained using the one-dimensional isentropic
theory and with particle velocity measurements (using laser diagnostic techniques)
made on a commercial cold spray system. The study has allowed coming to the
following conclusions: The measured nozzle mass flow rate follows closely the
overall behavior predicted by the one-dimensional isentropic theory. The predictions
of the axi-symmetric two-dimensional mathematical model are more accurate than the
one-dimensional approach due to the inclusion of viscous effects in the model. The
proposed model is more accurate than the one dimensional theory in predicting the
particle velocity inside the nozzle and in the free exiting jet. It is concluded that this is
due to the proper treatment of the shock waves in the nozzle by the proposed model.
For the two propellant gases used (helium and nitrogen), the proposed model predicts
oblique shock waves in the diverging section of the nozzle. The shocks reduce
drastically the gas velocity and Mach number while the gas temperature and pressure
are increased sharply. The proposed model results reveal that the maximum gas

23
velocity is always reached just in front of the first oblique shock wave. With the use
of nitrogen as the propellant gas, the jet has a lower gas velocity and takes longer to
reach the aerodynamic equilibrium state (jet pressure equal to external pressure). The
experimental study reveals that increasing the stagnation temperature or pressure
leads to an increase in particle velocity. The values predicted by the proposed model
are close to the measured ones, confirming that the mathematical model is accurate
even in the presence of shock waves.

Li (2007) carried out simulations using FLUENT to study the effect of throat & exit
area and diverging section length on the particle flow and to optimize the nozzle
design for different inlet conditions. Numerical modeling was performed by using
commercial software FLUENT (Ver. 6.1) to determine the flow field of driving gas
inside and outside the nozzle, and subsequently the acceleration of particles in cold
spraying. It was concluded that the optimization of nozzle exit diameter is influenced
by the gas conditions, particle size, and nozzle divergent section length and throat
diameter.

A one-dimensional isentropic gas flow model in conjunction with a particle

acceleration model to calculate particle velocities was employed by Pardhasaradhi et
al. (2008). A laser illumination-based optical diagnostic system is used for validation
studies to determine the particle velocity at the nozzle exit for a wide range of process
and feedstock parameters such as stagnation temperature, stagnation pressure, powder
feed rate, particle size and density. The relative influence of process and feedstock
parameters on particle velocity is presented in this work. It was found that the
stagnation temperature have greater influence on particle velocity in comparison to
stagnation pressure and powder feed rate (particle loading) have negligible effect on
particle velocity for the range employed in the present study due to the lower particle
concentrations in the present study.

Lupoi and O’Neill (2010) analyzed the powder stream characteristics in CGS
supersonic nozzles. The powder injection location was varied within the carrier gas
flow, along with the geometry of the powder injector, in order to identify their relation
with particles trajectories. Computational Fluid Dynamic (CFD) results are presented
along with experimental observations. Three nozzles configurations were examined,

24
characterized by a different acceleration channel lengths and powder injector’s
geometry and locations. When powder is released axially and upstream the nozzle
throat, particles trajectories do not stay close to the centre line, but tend to spread over
the entire volume of the channel. The particle stream diameter at impact with
substrate is comparable with the nozzle exit cross-section diameter. A theoretical
(CFD) analysis has shown that one of the reasons for this effect is a relatively high
gas turbulence level generated at the vicinity of the nozzle throat. The beam geometry
by the CFD results compares well against experimental observations. On the other
hand, when powder is released axially and straight into the supersonic region of the
nozzle, a more focused stream can be achieved. In this case, CFD results were in
accordance with the experiments, but failed to provide an accurate prediction of the
powder beam geometry.

2.3.2 Effect of CS process parameters

2.3.2.1 Gas temperature effects

Lee et al. (2007) found that increasing the gas pressure caused an increase in particle
velocity, while increasing the gas temperature not only affected the particle velocity
but also the particle temperature. Increasing the particle temperature could enhance
thermal softening, which is important for bonding. The deposition efficiency at the
same particle velocity became higher at higher process gas temperatures.

2.3.2.2 Powder properties

The irregular shape particle presents higher in-flight velocity than the spherical shape
particle under the same condition as showed by Ning et al. (2007). It was also found
that the larger size powder presents a lower critical velocity in the study.

2.3.2.3 Spray parameters

VanSteenkiste et al. (1999) found that kinetic sprayed coatings have relatively low
porosity values, hardness comparable with the corresponding bulk materials, adhesive
strengths as high as 68-82 M Pa, and oxide contents essentially the same as in the
powders. They observed that powders essentially do not stick to the substrates below
25
a certain inlet air temperature range. Substrate effects appeared to be relatively weak
in this experiment.

2.3.2.4 Incidence angle

The contact area between the deformed particle and substrate decreases and the crater
depth in the substrate reduces with increasing the tilting angle at the same impact
velocity was found by Gang et al. (2007).

2.3.2.5 Stand-off distance

Li et al. (2008) found that the deposition efficiency was decreased with the increase of
standoff distance from 10 mm to 110 mm for both Al and Ti powders used in study.
However, for Cu powders, the maximum deposition efficiency was obtained at the
standoff distance of 30 mm, and then the deposition efficiency decreased with further
increasing the standoff distance to 110 mm.

Pattison et al. (2008) found that the bow shock formed at the impingement zone not
only reduces the velocity of the gas, but also that of the entrained particles. Therefore
at small standoff distances, when the strength of the bow shock is high, deposition
performance is reduced. While at large standoff distances, when the bow shock has
disappeared, deposition can continue unhindered. Three distinct standoff regions were
identified that affect deposition performance: the small standoff region, where the
presence of the bow shock adversely affects deposition performance, and is limited by
the length of the nozzle's supersonic potential core; the medium standoff region,
where the bow shock has disappeared and, if the gas velocity remains above the
particle velocity (positive drag force), the deposition efficiency continues to increase;
and the large standoff region, where the gas velocity has fallen below the particle
velocity (negative drag force), and the particles begin to decelerate. For optimal
deposition performance, the standoff distance should be set within Region 2.

2.3.2.6 Impact velocity analysis

A simple function, which relates the gas velocity at the nozzle exit with the nozzle
expansion ratio and the carrier gas stagnation properties, was defined by Grujicic et
al. (2004).
26
Wu et al. (2005) showed that particle velocity increases with increasing the process
gas pressure and temperature. At higher temperature (or pressure), gas pressure (or
temperature) do more effect particle velocities. It is believed that gas pressure and
temperature have little effect on the flux distribution of flying particles. Further, it
was concluded that with increasing particle velocity, two critical velocities are
observed: one is for particle deposition onto the substrate (Vcr1) and the other for
particle–particle (Vcr2) bonding.

It is true that small particles exit the nozzle at high velocity; their velocity at impact
can be significantly lower because of the bow shock wave. It was shown by Helfritch
and Champagne (2008) that impact velocity increases as the particle diameter
decreases until a diameter of 4 to 8 μm is reached. Impact velocity then decreases as
the particle diameter is further reduced.

Critical velocities decrease with increasing particle size, which can be attributed to
effects by heat conduction or strain rate hardening, respectively. By defining the
influences of particle temperature and velocity on bonding, a parameter window for
cold spray deposition can be developed, which for successful spray experiments must
be met by the respective particle impact conditions was demonstrated by Schmidt et
al. (2006).

2.3.2.7 Impact phenomenon

Klinkov et al. (2005) showed that results of particle impacts depend not only on
impact velocity but also on particle size. Impacts result in two contrasting
phenomena: destruction (also called erosion, cratering etc.) and adhesion (also called
attachment, sticking). At hyper-velocity impacts, cratering and destruction are typical
features which exhibit minor scale effects. At low impact velocities, there is a
transition from erosion to adhesion (sticking) when the particle size decreases. Here,
the nature of adhesion is due to van der Waals and electrostatic forces. At high
velocities, the results of impact depend not only on size and velocity but also on other
parameters (e.g. plasticity, particle flux concentration, etc.).

Wu et al. (2006) concluded that a rebound phenomenon was observed in which a high
particle velocity caused a high fraction of rebound particles. A maximum impact

27
velocity was found for the particle deposition onto the substrate. The deposition of
individual particles was controlled by the adhesion energy and the rebound (elastic
recovering) energy. The impacting particles could only be attached to the substrate
when the adhesion energy was higher than the rebound energy.

2.3.3 Bonding mechanism of CS process

The two main factors contributing to the observed higher deposition efficiency in the
case of copper deposition on aluminum are larger particle/substrate interfacial area
and higher contact pressures as indicated in literature (Grujicic et al., 2003). Both of
these are the result of a larger kinetic energy associated with a heavier copper feed-
powder particle. They concluded that the critical velocity, above which cold-spray
deposition takes place, is associated with the attainment of a condition for the
formation of a particle/substrate interfacial jet composed of both the particle material
and the substrate material. An interfacial instability due to differing viscosities and the
resulting interfacial roll-ups and vortices may promote interfacial bonding by
increasing the interfacial area, giving rise to material mixing at the interface and by
providing mechanical interlocking between the two materials.

Assadi et al. (2003) found that the bonding of a particle can be attributed to adiabatic
shear instabilities which occur at the particle/substrate or particle/particle interfaces at
high velocities. The modeling also shows a very non-uniform development of strain
and temperature at the interface, suggesting that this bonding is confined to a fraction
of the interacting surfaces. The analysis also suggests that density and particle
temperature have significant effects on the critical velocity and are thus two of the
most influential parameters in cold gas spraying.

Li et al. (2006) indicated that both the flattening ratio and compression ratio of
particles increase with the increase in particle velocity. The compression ratio is more
convenient for characterizing the extent of particle deformation owing to the easy
estimation and its independency on meshing size.

28
2.3.4 Adiabatic shear instability in CGDS process

The minimal impact particles velocity needed to produce shear localization at the
particles/substrate interface correlates quite well with the critical velocity for particles
deposition by the cold-gas dynamic-spray process in a number of metallic materials
was shown by Grujicic et al. (2004). This finding suggests that the onset of adiabatic
shear instability in the particles/substrate interfacial region plays an important role in
promoting particle/substrate adhesion and, thus, particles/substrate bonding during the
cold-gas dynamic-spray process.

2.3.5 Properties of Cold Sprayed coatings

Stoltenhoff et al. (2006) demonstrate that at negligible porosities, microstructures of

cold-sprayed and thermally sprayed coatings are mainly determined by high degrees
of deformation or oxidation. Here, cold-sprayed coatings show a more uniform
appearance and respectively higher electrical conductivity. Overall, a solid-state
deformation results in higher hardness of cold-sprayed and HVOF-sprayed coatings,
similar to that of cold rolled bulk material after 90% reduction in thickness.
Subsequent annealing experiments and respective analyses demonstrated that in cold-
sprayed coatings mainly recovery and recrystallization determine further micro-
structural developments.

2.3.6 Effect of heat treatment

Gartner et al. (2006) demonstrate that cold-sprayed coatings, which are processed
with helium, show a similar performance as highly deformed bulk material. Also after
subsequent annealing, strength and elongation to failure develop in a similar manner
as for cold rolled sheets. In the as sprayed state, cold-sprayed coatings, processed with
nitrogen, and thermal spray coatings show brittle failure already under comparatively
low tensile stress. Whereas the performance of cold-sprayed coatings, processed with
nitrogen, can be substantially improved by annealing, mechanical properties of

29
thermal spray coatings are influenced only to a minor extend by following heat
treatments.

Copper alumina coatings are found to be resistant to grain growth and softening even
upon heat treatment at 950 0C which is close to the melting point of copper (1083 0C)
owing to the presence of fine alumina particles. Grain size is found to be the most
dominant factor affecting the electrical conductivity of the coatings. The present study
by Phani et al. (2007) clearly indicates the potential of cold sprayed copper alumina
coatings in high-strength, high conductivity applications.

PART III

2.4 ROLE OF THE COATINGS

The thermal spray coatings used for corrosion resistance must be dense enough so that
the protective oxides can form within and fill voids, and thick enough to postpone the
diffusion of corrosive species to the substrate material before the protective oxide can
form within the coating (Bluni and Marder, 1996). Guilemany et al. (2002) have also
reported that a thicker coating provides better resistance against corrosion.

Gurrappa (2003) suggested that the coating used for corrosion resistance should have
a composition that will react with the environment to produce the most protective
scale possible, provide corrosion resistance with long term stability and have
resistance to cracking or spallation under mechanical and thermal stresses induced
during operation of the components. Further, the protective oxide scale should not
react with the corrosive environment and at the same time, it should not allow the
corrosive species to diffuse through the coating.

Alloys and coatings designed to resist oxidizing environments for corrosion resistance
should be able to develop a surface oxide layer, which is thermodynamically stable,
slowly growing and adherent (Brandl et al., 2004).

Table 2.4 lists the benefits of using thermal spray process.

30
 Table 2.4: Benefits of using thermal spray process (Sidhu, 2006)
Coating benefit
Higher density (lower porosity)
Improved corrosion barrier
Higher hardness ratings
Improved wear resistance
Higher bond and cohesive strengths
Lower oxide content
Fewer unmelted particle content
Greater chemistry and phase retention
Thicker coatings
Smoother as-sprayed surfaces
Main reason for the benefit
Higher impact energy
Less porosity
Better bonding, less degradation
Harder, tougher coating
Improved particle bonding
Less in-flight exposure time to air
Better particle heating
Reduced time at higher temperatures
Less residual stress
Higher impact energy

2.4.1 Behavior of copper coatings

Pratt & Whitney used deposition of a copper coating by the cold spray to improve the
structure and functioning of a new aircraft engine (Haynes and Karthikeyan, 2003;
Cooley et al., 2002). A thick layer of the copper coating was necessary to improve
heat removal from the combustion-chamber zone. If the coating thicker than 1 mm is
applied by the traditional electrolytic technology, the adhesion is very low; in
addition, the process takes a long time (two weeks). Application of a copper coating
applied by the cold spray allows one to significantly reduce the time of the process,
improve adhesion, and avoid the use of hazardous chemicals necessary in the
conventional electrolytic method.

The requirements to the coating are rather severe: it has to withstand cyclic thermal
loads, including connection (welding) of tubes, etc. The highest thermal load is the
heating up to 1200 K in vacuum, and the coating has to be still operational under
these conditions (have no bubbles and exfoliation). High temperature gradients arise
when the engine is started and shut down; therefore, the coating has to withstand high
thermal impacts without exfoliation. To check the coating operation in this regime,
the coated samples were cooled in liquid nitrogen and water.

An analysis of various methods for coating application showed that it is possible to

obtain a copper coating with prescribed parameters by the electrolytic method,
vacuum plasma deposition, and cold spray. The researchers from Pratt & Whitney
first decided to use the electrolytic technology to obtain the required coating. It turned

31
out, however, that this method requires several processes of treatment to obtain a
high-quality coating and that this method should be significantly upgraded.

The plasma coating satisfied the tests (Hickman and Mckechnie, 2001), but the
process was too long and too expensive. Moreover, masking and mechanical post-
processing (improvement of accuracy to a necessary level) are rather labor-
consuming, because the article to be coated has a very intricate shape, which makes it
difficult to reach the necessary dimensions within admissible margins.

A preliminary analysis of the cold spray showed that it is possible to obtain dense
coatings without inclusions and with a low content of oxides (in some cases, even
lower than in the initial material) (Smith et al., 1999). Moreover, obtaining very thick
coatings with good adhesion during a short time (Karthikeyan, 2002) and with
acceptable expenses is the advantage of the cold spray, which evoked the interest in
this method.

A series of experiments (Haynes and Karthikeyan, 2003) were performed with copper
spraying onto stainless steel under different conditions: the varied parameters were
the surface roughness, spraying parameters (pressure and temperature), etc. Initially,
the coatings exfoliated from most samples. A micro structural study showed that the
coating behavior during the tests is affected, in addition to spraying parameters, by the
powder purity and the state of the surface. Therefore, an attempt was made to obtain
coatings from powders of higher purity and with precise control of the spraying
parameter (stagnation pressure and temperature of the gas). Testing of the samples
obtained showed that they can withstand thermal cycles without exfoliation of
coatings.

A metallographic analysis of coatings applied with optimized parameters revealed

good adhesion and low porosity. After this encouraging result was obtained, an engine
mock-up was fabricated, and a copper coating more than 2.5 mm thick was applied.
Then the coating was treated by grinding and drilling holes to demonstrate that the
coating can withstand stresses caused by machining. Thus, it was demonstrated that
the cold spray can be used to obtain dense, phase-pure, and thick coatings with good
adhesion, which can withstand significant thermal loads. In addition, the cold spray
was shown to be promising for production of Hi-Tech articles/elements. In the latter
32
activity, the cold spray advantages are the high production efficiency, acceptable cost,
environmental safety, etc. (Haynes and Karthikeyan, 2003) over other methods.

Obtaining an aluminum-bronze coating on a stainless-steel substrate from aluminum-

bronze powder with a particle size smaller than 40µm was considered in Xiong et al.
(2004). Detailed inspection of the element composition in the vicinity of the
coating/substrate interface showed that copper and aluminum in the coating migrate to
the substrate material. At the same time, iron and chromium from the substrate
migrate to the coating material. Moreover, the farther from the interface, the lower the
diffusion of copper and aluminum to the substrate and the lower the diffusion of iron
and chromium to the coating. The width of the diffusion zone is several microns.

The effect of the spraying parameters (stagnation pressure and temperature, velocity
of substrate motion) on the properties of copper coatings (from a copper powder with
a particle size of −22+5 µm) on an aluminum substrate was examined in Galla et al.
(2004). Prior to spraying, the substrate was cleansed by ethyl alcohol denatured by
methyl alcohol. Spraying was performed with the use of helium. An increase in
deposition efficiency with increasing stagnation pressure was noted, as well as a
certain (from 90 to 85%) decrease in deposition efficiency with increasing flow rate
of the powder from 45 g/min (21 wt% of the gas flow rate) to 76 g/min (36 wt% from
the gas flow rate). It was noted that the coating starts exfoliating from the surface after
it reaches a certain thickness owing to compressive stresses. The tendency to
exfoliation can be prevented or significantly reduced in many cases by means of gas
heating. The coating hardness increases with increasing pressure from 1.1 to 2.2 M Pa
and then stabilizes with a further increase in pressure to 2.9 M Pa. An increase in
temperature (from 300 to 700 K) leads to a decrease in hardness.

Obtaining coatings from a copper powder (5–25 µm) on aluminum, copper, steel, and
stainless steel pre-processed by sandblasting was considered in Hukanuma and Ohno,
(2004). Coatings approximately 600 µm thick were applied onto samples 20 mm in
diameter; before the tensile tests by the glue method, however, the coating was
polished off to a thickness of 300 µm. The experiments showed that the adhesion of a
copper coating applied with the use of nitrogen was higher on the aluminum substrate
than on the copper substrate. In spraying the particles onto the carbon steel and

33
stainless steel substrates, the coating exfoliated from the substrate in the course of
spraying, as soon as the coating thickness reached approximately 500 µm, whereas no
exfoliation occurred if helium was used. In addition, the adhesion was much higher
than that in the case of spraying by nitrogen. The authors also noted that the adhesion
force depends on the working gas pressure (in the range of 1–3 M Pa): the higher the
pressure, the higher the adhesion. If the threshold pressure is exceeded, the coating
starts separating along the epoxy resin/coating interface rather than along the
coating/substrate interface. Depending on the substrate material, the adhesion force is
distributed as follows (in increasing order): stainless steel, steel, aluminum, and
copper.

The influence of the spraying angle on application of coatings of copper particles was
considered in Li et al. (2003). The deposition efficiency increases with increasing
spraying angle from zero at a critical angle to 100% at 900. The size of this region
depends on the particle-velocity distribution. The most important parameter, however,
is the velocity of the impact onto the substrate. The critical velocities for copper are
given in Alkhimov et al. (1991): approximately 560–580 m/s. The critical velocity is
affected by the particle size, the particle-size distribution (Van Steenkiste et al. 2002),
and the substrate material. On the other hand, the particle velocity for a certain
material is determined by the type of the accelerating gas, its pressure and
temperature, and the nozzle structure. The particle properties (density, size, and
shape) also affect the acceleration of particles and, correspondingly, the spraying
process (Van Steenkiste et al. 1999).

If the particles hit the substrate surface at an angle other than 900, the normal
component of the particle velocity is smaller than that in the normal impact. As the
particle deformation depends on the normal velocity, the angle can be assumed to
affect the spraying process and the coating microstructure. Though the influence of
the angle on the microstructure and properties of coatings obtained by thermal
spraying are known, such effects have different features in the cold spray, because the
sprayed particles are in a non-melted state. There are only a few papers where the
influence of the spraying angle in cold gas-dynamic spraying was considered
(Gilmore et al., 1999; Li et al., 2003).

34
Commercial copper powders (with a particle size of 15–37µm) were used in
experiments. These copper powders were obtained by atomization and contained
spherical particles. Stainless steel was used as a substrate material. The substrate was
subjected to sandblasting by a corundum powder to study the process of copper
deposition and was polished to study the copper-coating microstructure. The nozzle
had a throat diameter of 2 mm, an exit diameter of 6 mm, and a length of the
supersonic section of 100 mm; the powder was injected along the nozzle centerline.
Nitrogen with a pressure of 2.0 M Pa and a temperature of 220 0C was used to
accelerate copper particles. The distance between the nozzle exit and the substrate
was 15 mm, and the velocity of substrate motion was 80 mm/s. The microstructure
was analyzed with the use of a scanning electron microscope (JEOL, JSM5800). It
was demonstrated that the dimensionless deposition efficiency has a maximum at
angles of 80-900. This means that the spraying angle has an insignificant effect within
these limits. With a further decrease in spraying angle, however, the dimensionless
deposition efficiency rapidly decreases and vanishes at an angle of 400. Thus, there is
an angle below which no deposition occurs. As the particles in the normal impact are
attached at a velocity higher than the critical value, the normal component of velocity
can be assumed to be the main factor. As the spraying angle decreases, the normal
component of the impact velocity also decreases; when the normal velocity
component becomes lower than the critical velocity, the particle is not attached.

The cold spray method simultaneously involves deposition of particles and erosion of
the substrate. Adhesion of particles occurs when they reach a velocity higher than the
critical value, while erosion occurs because there are particles whose velocity is lower
than the critical value. As there is some distribution of particles in terms of their
velocity, the critical velocity can turn out to be inside this distribution and only some
portion of particles attach to the substrate. On the other hand, particles with lower
velocities rebound and destroy the coating. If the particle hits the substrate at a certain
angle, the impact velocity of the particle can be divided into the normal component
and the tangential component with respect to the substrate surface. If the effect of the
tangential component is assumed to be small, the deposition is mainly determined by
the normal component of velocity. Correspondingly, the dimensionless deposition
efficiency changes insignificantly as the angle decreases from 900 to an angle at
35
which the normal component reaches the critical velocity. With a further decrease in
spraying angle, the dimensionless deposition efficiency decreases from 100 to 0%.
The experimental results obtained support the validity of this model. Following this
model, we can assume that the range of transitional angles depends on the particle-
velocity distribution. The transitional region in terms of the spraying angle is wide for
particles with a wide velocity distribution and narrow for particles with a narrow
velocity distribution. The transitional region for copper is approximately 400. A wide
distribution of the particle size leads to a wide distribution of the particle velocity.

The effect of various parameters of the substrate, such as the substrate thickness,
surface roughness, substrate temperature, number of passes, and velocity of substrate
motion, was considered in Sakaki et al. (2004) by an example of copper coatings
(mean particle size of 8 µm) on soft steel. Nitrogen with a pressure of 3 M Pa and a
temperature of 623 K was used as a working gas. The deposition efficiency for copper
increases (from 60 to 70%) with increasing substrate thickness (from 6 to 32 mm).
The deposition efficiency also slightly increases with increasing substrate roughness
(from 0.2 to 7 µm Ra). Pre-heating of the substrate from room temperature to 450 K
and an increase in the number of passes (from 1 to 3) exert the same effect as an
increase in the substrate thickness. The deposition efficiency significantly decreases
(from 70 to 55%) with increasing velocity of substrate motion (from 20 to 100 mm/s).
Unfortunately, the authors did not give any comments on these effects.

The copper coating on an aluminum substrate (polished or etched) obtained in Voyer

et al. (2003) displays a low level of porosity. Moreover, the content of oxygen in the
coating is approximately 0.1%, which coincides with the content in the initial material
of the powder. It follows from here that cold spray does not oxidize the initial
material. The electrical conductivity of coatings is approximately 90% of the
electrical conductivity of copper proper, which is much higher than the electrical
conductivity of coatings obtained by gas-thermal methods, such as the gas-plasma and
the electric-arc methods (only 30–40% of the electrical conductivity of copper
proper). The Vickers 0.3 hardness of the cold spray coating is 150, and the
coating/substrate interface does not contain defects. Mechanical tests were performed
for two types of coatings: immediately after spraying and after annealing at 400 0C

36
during 1 hour to examine the influence of annealing on mechanical properties. After
annealing, the coating becomes more plastic. The coating sample without annealing
after spraying broke under a load of 66 M Pa and a strain of 0.06%, whereas the
annealed sample could withstand a load of 195 M Pa with a total strain of 1.04%. A
comparison with characteristics of copper shows that Young’s moduli differ
insignificantly, but the coating strength is lower. Composite coatings from copper and
aluminum were also dense and, which is especially important, had a uniform
distribution of materials. Thus, the cold spray allows obtaining composite coatings
consisting of mixtures of materials (Voyer et al., 2003). The properties of the copper
coating on an aluminum substrate and the degree of the influence of spraying
parameters on the coating density were considered in Xiong et al. (2005). In the
examined range, the greatest effect on the coating quality is exerted by the standoff
distance, and then there follow the gas temperature and pressure. The influence of
voltage on the powder feeder (of the drum type), which occupies the last position, is
also noted. Yet, the authors commented that the degree of the influence of
temperature and pressure could be prevailing if the range of parameters is expanded.
Thus, the tests performed are primarily useful for optimization of a particular cold
spray setup rather than for optimization of the cold spray process proper. A decrease
in micro hardness with increasing coating annealing temperature from 160 to 90
(Hv0.2) is noted. Coating adhesion turned out to be 18 MPa. The authors assumed that
the basic mechanisms of adhesion are sub melting and mechanical adhesion. The
experiments aimed at determining the thermal and electrical conductivity of coatings
showed that these quantities reach approximately 70% of the corresponding values for
the cast material. This ratio is almost unaffected by annealing.

The effect of thermal treatment of copper coatings on their structure and properties
was considered in Calla et al. (2005). The grain size increases from 60 to 120 nm with
increasing annealing temperature from 25 to 200 0C. The micro hardness of copper
coatings remains roughly unchanged. As the annealing temperature increases to 300
0
C and higher, recrystallization with a further increase in the grain size occurs
(dislocations related to micro strains form new strain-free grains), which impairs the
hardness of coatings. It was assumed in Meyers et al. (1992) that grains of an
approximately identical size are formed under conditions of dynamic recrystallization
37
(i.e., under high strains and strain rates), but the recrystallization itself is initiated at
temperatures above 0.4Tm. Dynamic recrystallization can occur in the course of
deposition because of the formation of a high density of dislocations and a local
increase in temperature owing to adiabatic shear and heating of the substrate. The
grain size in the deposited coating is too small for optical microscopy; annealing at
200 0C reveals several very small grains, and annealing at 500 0C leads to noticeable
recrystallization with a grain size from 1 to 5 µm. The feature most frequently
encountered in tensile tests of coatings is the plastic fracture in the case of annealing
at 600 0C. This indicates that there exists a clear boundary with good adhesion
between the particles already during the spraying process, and the microstructure is
improved during annealing. Directly after deposition, the coatings display fracture
close to brittle fracture. The tensile force, however, is very high, which again
confirms the hypothesis about high-quality adhesion of particles.

The influence of thermal treatment on electrical resistance and hardness of copper

coatings (spherical particles 24 µm in diameter on the average) on tough copper
substrates, which were obtained with the use of nitrogen at a temperature of 150 0C
and a pressure of 2 M Pa, was considered in Li and Li (2005). It turned out that the
coating resistance immediately after deposition in the direction parallel to the
substrate surface is approximately half its value in the perpendicular direction (47%
against 81% of the value for the cast material). After annealing, however, they
become very close and reach 95–96% of the value for the cast material. The coating
hardness decreases with increasing annealing temperature.

The effect of thermal treatment of coatings obtained from copper powders (10–36
µm) on electrical resistance and coating hardness was examined in Lagerbom et al.
(2005). It was noted that annealing of the copper coating at a temperature of 300 0C or
higher decreases the coating hardness (from 140 to 80 Hv0.1). The electrical resistance
of the entire composition (steel substrate and copper coating) also decreases with
increasing annealing temperature. At an annealing temperature of approximately 200
0
C and higher, the resistance reaches the value of 1.8×10−8Ωm (the resistance of the
cast material is 1.6–1.7×10−8Ωm) against 2.3×10−8Ωm in the coating without thermal
treatment. The resistance of the coating proper without the substrate is 3.7×10−8Ωm.

38
The influence of the flow rate of the powder was examined in Taylor et al. (2005).
Significant concentrations of the powder in the gas flow reduce the impact velocity of
particles. If the powder concentrations are insufficient, the coating formed has some
holes. In both extreme cases, the quality of coatings is not as good as that desired. The
effect of the copper-powder concentration in a helium flow with deposition onto an
aluminum substrate on the coating thickness and microstructure and on deposition
efficiency was considered. A copper powder with angular particles (similar to crushed
stone) with a mean size of 25–30µm was used in the experiments. An axi-symmetric
nozzle with an exit diameter of 6.3 mm was used. The stagnation temperature of
helium was equal to room temperature, its stagnation pressure was 1.7 M Pa, and the
difference in pressure between the feeder and the pre-chamber was 35 k Pa. During
the deposition, the flow rate of the powder was changed (from 0.9 to 5.0 g/min),
whereas the substrate velocity was constant; one pass was made. The modeling
showed that the particle velocity should be higher than the critical velocity (equal to
450 m/s for copper); therefore, the deposition efficiency was expected to be fairly
high. In addition, the influence of particles on the gas can be neglected even for the
highest concentrations of the powder. Interesting results were obtained. The coating
thickness and mass (per unit length of the deposition band) increase if the flow rate of
the powder is low, but their values decrease if the flow rate of particles exceeds 4.1
g/min. Moreover, the coating starts exfoliating. The deposition efficiency thereby
remains approximately unchanged (about 85–90%). If the substrate velocity is
increased, however, no exfoliation occurs, and the coating is as dense as that obtained
with lower flow rates of the powder. The authors attribute exfoliation to more intense
cold working owing to enhanced bombardment of the surface, i.e., residual stresses in
the coating become higher, which is the reason for exfoliation.

The influence of the temperature and impact angle of copper particles with a mean
particle size of 56µm on the critical velocity of adhesion on a copper substrate was
studied experimentally and theoretically studied in Li and Li (2005). Nitrogen and
helium was used for acceleration in an axi-symmetric nozzle with the supersonic
section 100 mm long. The motion of the gas and particles is modeled by the FLUENT
+ DPM software (LS-DYNA, 1998), which takes into account the effects of
hardening and thermal softening but ignores heat transfer. The theoretical value of the
39
critical deposition velocity is related to the emergence of adiabatic shear instability.
Experimentally, the critical deposition velocity is determined on the basis of the
measured deposition efficiency. The values obtained, however, were slightly lower
(from 295 to 355 m/s) than those mentioned in other publications (about 410 m/s).
This difference was attributed to the presence of an oxide film, but this phenomenon
was not modeled.

The properties of cold spray produced copper coatings were examined in Borchers et
al. (2003) and compared with copper coatings obtained by various gas-thermal
methods. It was found that the electrical resistance of the coatings is commensurable
with that of the cold-drawn cast material (1.7µΩcm), which turned out to be lower
than the resistance of gas-thermal coatings. The hardness of cold spray produced
copper coatings was estimated at 140–160 Hv0.3, which coincides with the hardness of
cold-drawn copper; the adhesion of cold spray produced copper coatings was
estimated at 30–40 M Pa, which coincides with the adhesion of coatings obtained by
gas-thermal methods (HVOF, HVCW (combustion wire)).

Copper coatings obtained by the cold spray and plasma methods were compared in
Barradas et al. (2005): porosity 0.5% against 9%; surface roughness Ra 8µm against
13µm; roughness of the coating/substrate boundary Ra 1.3µm against 13µm (the first
and second values in each pair refer to cold spray and plasma-produced coatings,
respectively). The coating was produced by the cold spray with the use of nitrogen
with stagnation parameters of 2.8 M Pa and 823 K and spheroidized copper powder
with a particle size of −22+5µm on a 2017 Al substrate.

Restrictions of thermal spraying in the automobile industry are often related to a

comparatively low quality of electro conducting elements with coatings applied by
conventional methods. The cold spray method can find its application for contact
joints, etc. The study of the microstructure of copper coatings (McCune et al. 2000 a;
McCune et al. 2000 b) showed that their properties depend on the powder
characteristics and on the spraying regime and that the coating becomes more plastic
after annealing.

40