Advertisement

Nanostructured or Finely Structured Coatings

  • Pierre L. Fauchais
  • Joachim V. R. Heberlein
  • Maher I. Boulos
Chapter

Abstract

According to the definition of the US National Nanotechnology Initiative, USNNI, “nanometer-sized” materials, or more commonly referred to as “nanosized” or “nanostructured” materials, are materials with a particle diameter or an internal structure with at least one dimension smaller than 100 nm. From the pioneering works of McPherson in 1973, who was among the first to identify nano-size features in thermally sprayed alumina coatings, to recent and most advanced work aimed at manufacturing nanostructured coatings from nanosize feedstock particles, the thermal spray community has been actively involved in this area for more than 30 years. Nano-sized materials are at the frontier of materials science due to their remarkable, and in some cases novel, properties [1, 2]. This results in particular from the surface-to-bulk ratio, and interface density between features, that are much higher in nanostructured materials compared to those of coarser particle (e.g., micro-sized) materials. The physical mechanisms involved leading to those remarkable properties lies, at such dimensions, between atom quantum effects (e.g., phonon scattering, surface plasmon, etc.) and bulk behaviors (e.g., cohesion, etc.).

Keywords

Thermal Spray Simulated Body Fluid Plasma Torch Cold Spray Spray Distance 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Abbreviations

APS

Atmospheric plasma spraying

BGCCs

Bioactive glass–ceramic coatings

CGDS

Cold gas dynamic spraying

CVD

Chemical vapor deposition

d.c.

Direct current

EB-PVD

Electron beam-physical vapor deposition

ETAB

Enhanced Taylor’s analogy break-up

FESEM

Field emission scanning electron microscopy

FCT

Furnace cycle test

GDC

Gadolinia-doped ceria

GFA

Glass forming ability

HA

Hydroxyapatite (Cal0(PO4)6(OH))

HCA

Hydroxide carbonate apatite

HPAL

High-pressure acid-leach

HVAF

High velocity air fuel flame

HVOF

High velocity oxy-fuel flame

HVSFS

High velocity suspension flame spraying

i.d.

Internal diameter

IT-SOFCs

Intermediate temperature solid oxide fuel cells

JETS

Jet engine thermal shocks furnace cycle test (FCT)

LCA

Life Assessment Impact

LSCF

Lanthanum strontium cobalt iron oxide (La x Sr1−x Co y Fe1−y O3−δ )

MWCNT

Multiwalled carbon nanometer-sized tubes

Nd-YAG

Neodinium-doped yttrium aluminum garnet

PAA

Ammonium polyacrylic acid

PBTCA

2-Phosphonobutane-1, 2, 4-tricarboxylic acid

PE-CVD

Plasma-enhanced-chemical vapor deposition

PEI

Polyethyleneimine

PGDS

Pulsed gas dynamic spraying

PIV

Particle image velocimetry

PTA

Plasma transferred arc

PVD

Physical vapor deposition

r.f.

Radio frequency

RH

Relative air humidity

SBF

Simulated body fluid

SDC

Samaria-doped ceria

SEM

Scanning electron microscopy

slm

Standard liters per minute

SOFCs

Solid oxide fuel cells

SolPS

r.f. Solution plasma spraying

SPS

Suspension plasma spraying

SPPS

Solution precursor plasma spraying

SPS

Suspension plasma spraying

SPTS

Solution precursor thermal spraying

STS

Suspension thermal spraying

TAB

Taylor’s analogy break-up

TBCs

Thermal barrier coatings

TIG

Tungsten inert gas

We

Weber number

YAG

Yttrium aluminum garnet

YPSZ

Yttria partially stabilized zirconia

YSZ

Yttria-stabilized zirconia

14.1 Introduction

14.1.1 Why Nanostructured Coatings

According to the definition of the US National Nanotechnology Initiative, USNNI, “nanometer-sized” materials, or more commonly referred to as “nanosized” or “nanostructured” materials, are materials with a particle diameter or an internal structure with at least one dimension smaller than 100 nm. From the pioneering works of McPherson in 1973, who was among the first to identify nano-size features in thermally sprayed alumina coatings, to recent and most advanced work aimed at manufacturing nanostructured coatings from nanosize feedstock particles, the thermal spray community has been actively involved in this area for more than 30 years. Nano-sized materials are at the frontier of materials science due to their remarkable, and in some cases novel, properties [1, 2]. This results in particular from the surface-to-bulk ratio, and interface density between features, that are much higher in nanostructured materials compared to those of coarser particle (e.g., micro-sized) materials. The physical mechanisms involved leading to those remarkable properties lies, at such dimensions, between atom quantum effects (e.g., phonon scattering, surface plasmon, etc.) and bulk behaviors (e.g., cohesion, etc.).

Due to the large volume fraction of the internal interfaces, nanostructured coatings should exhibit better properties than conventional microstructured coatings. For example, considering ceramics, they should exhibit in a noticeable increase in dimensional stability, lower thermal diffusivity and hysteresis (due to phonon scattering by boundary defects), higher hardness and toughness (due to small grain sizes), better wear resistance (due to higher hardness, grain-sliding plasticity, and the change of fracture and material-removal mechanism due to ultrafine particle size) [3, 4]. Moreover generally nanostructured coatings are not necessarily harder than the conventional ones, but they tend to be much tougher. This is explained by the crack arresting effect. In conventional thermal-sprayed coatings the crack tends to propagate along splat boundaries due to the weak intersplat bonding. For the nanostructured coating the splat-to-splat cohesion is generally better and the coating is rather homogeneous with splat sizes in the 100 nm range, thus crack propagation is limited. It is also important to point out that most of the research performed over the last decade has been devoted to nanostructured ceramics. To illustrate the interest of these coatings a few examples are briefly described.

1. Spraying of agglomerated nanometer-sized particles [5]. As described in next subsection, among the different processes developed to produce such coatings, spraying of micro-sized particles made of agglomerated nano-sized ones has been the first to be sufficiently developed, well characterized, with resulting coatings tested in service. For example, for a pure ceramic abradable coating design, the concept of the porous nano-zones, if embedded in the coating microstructure in “sufficient numbers,” could be employed to lower the overall stiffness of the coating, allowing the tip of a metallic turbine blade to wear off some of the coating, creating the “seal effect,” without damaging the blade. Al2O3–13 wt% TiO2 HVOF-sprayed ultrafine coating exhibited a 4-fold improvement in wear performance. Evolution of the crack propagation resistance for the alumina–titania coatings showed to be the main reason for the superior wear behavior of the ultrafine coating. Improved bond strength values have been reported for these coatings when compared with their conventional counterparts. It seems that dense nano-zones would tend to impede crack propagation at the interface, which would enhance interfacial toughness and the bond strength levels of the coating. Nanostructured alumina–titania coatings deposited by APS are in use to protect the main propulsion shafts of ships of the US Navy. After 4 years of use in naval applications no significant damage was recorded in these types of coatings. It has been shown that HVOF-sprayed nano-TiO2–hydroxyapatite (HA) composite coatings exhibit bond strength levels of at least 2.5 times that of thermally sprayed conventional HA coatings. Also these coatings exhibit bio-performance levels equivalent or superior to those of an HA coating, which is the current state-of-the-art material.

2. Iron-based coatings with metallic glass formation forming nanostructures upon reheating [6, 7]. Specialized iron-based complex composition with a low critical cooling rate (104 K/s) for metallic glass formation has been developed. Coatings with thicknesses up to 1.5 mm, obtained by almost all spray processes, except cold spray, presented interesting properties, even when sprayed in air. d.c. plasma or HVOF-sprayed powders such as Fe63Cr8Mo2B17C5Si1Al4, resulted in coatings with an exceptional resistance to two-body abrasion, and a higher hardness levels compared to that found in conventional steels. The as-sprayed amorphous coatings showed hardness of 10.2–10.7 GPa and after reheating up to 750 °C or 800 °C for one hour (devitrified nano-composite), their hardness was 11.4–12.8 GPa. With twin wire-arc spraying the SHS8000, iron-based glass forming alloy, coatings were formed primarily of an amorphous matrix with very fine dispersed nano-crystals ferrites and M2(C,B)1 compounds with a hardness in the range 1.05–1.2 GPa. When exposed to 649 °C, the microhardness increased up to ~2 GPa, very significant hardening. The wear resistance to dry sand/rubber wheel abrasion was improved by a factor of 3.5. A similar improvement was also obtained for elevated-temperature erosion resistance. Wire-arc coatings of SHS7170 resist erosion almost independently of contact angle, at temperatures up to 600 °C which is particularly important for real-world boiler applications. The iron alloys developed for PTA spraying had a high hardness, up to Rc 66, due to their fine structure consisting of a high volume fraction of boro-carbides phases. The toughness measured with a Palmqvist method, was up to 74.7 MPa m1/2, which could adequately be described through a crack bridging model.

3.Suspensions or solution-sprayed coatings [5, 8, 9]. Suspension plasma spraying (SPS), high velocity suspension flame spraying (HVSFS), or solution precursor plasma spraying (SPPS) have been used for the preparation of nanostructured thermal barrier coatings with a very fine porosity, high segmentation crack density, and low Young’s modulus. The segmentation cracks survived thermal cycling test. The advantages of SPS were partially lost at very high temperatures (1,400 °C) when YSZ was chosen as coating material. Investigations are ongoing to reduce the amount of overspray embedded in the coatings and to reduce the sintering tendency. SPPS combines the advantages of EB-PVD and plasma or HVOF spraying processes: good strain tolerance and low apparent thermal conductivity with improved bond strength. The spallation life was improved by a factor of 2.5 compared with APS coatings on the same bond coat and substrate and by a factor of 1.5 compared with high-quality EB-PVD coatings. Many studies have also been devoted to solid oxide fuel cells (SOFC) components made by SPS and SPPS and some results are promising. With suspension alumina coatings by HVSFS, low wear rates, nearly one order of magnitude better than for standard micrometer-sized APS and HVOF coatings, were measured using dry sliding tests. The friction coefficient of Al2O3 coatings was decreased by a factor of four (0.2 for SPS coatings sprayed with the 0.4 μm mean diameter feedstock, compared to 0.8–0.9 for the APS one). The wear rate was 30 times lower for SPS coatings compared to that of APS coatings. HVSFS of nano-TiO2–HA composite coatings exhibit bond strength levels of at least 2.5 times higher than those of thermally sprayed conventional HA coatings with equivalent or superior bio-performance levels compared to a standard HA coating. HVSFS was used to deposit bioactive glass coatings. In vitro bioactivity tests indicated that the coatings interacted remarkably with the simulated body fluid (SBF), developing a thick silica-rich layer containing hydroxyapatite crystals.

14.1.2 How to Spray Nanostructure Coatings?

When spraying conventionally micro-sized particles with nanostructures, their melting automatically destroys the nanostructure. While this is not the case in cold spray, but the sprayed nanostructured particles must be ductile enough. Thus processing of nanoparticles is not simple as pointed out in the review of the different processes by Viswanathan et al. [10]. Among the possible routes thermal spray is very promising [11] to achieve nano– or finely structured materials due to a rapid solidification rate. This opens the way to prepare advanced materials with unique properties.

One of the major drawbacks of processing nano-sized particles by thermal spraying is the problem of injecting them into the core of the high enthalpy flow, since the particle injection force has to be of the same order of that imparted to them by the flow [12]. Decreasing the particle average size down to the nano-size requires, besides other technological issues, increasing the particle injection momentum by significantly of increase of cold carrier gas flow rate, this to a point where it disrupts the high enthalpy flow (disruption occurs usually for a gas mass flow rate equal to or higher than one sixth of the high enthalpy flow rate). To circumvent this phenomenon, two routes can be followed: spraying micrometer-sized particles as in conventional thermal spraying, but made of nano-sized agglomerated particles (see Sect.  11.1.2.6), or spraying sub-micrometer or nano-sized particles using a liquid carrier instead of gas (see Sect.  4.5) [13]. For details see the review of Fauchais et al. [5]. In both cases, when spraying materials that are not oxides, care must be taken to avoid the in-flight oxidation of the particles, since the presence of the oxide between layered splats (or particles) diminishes the contacts between them and consequently degrades the coating properties and especially their resistance to corrosion. Compared to conventional coatings (with micrometric structures), the oxide layers are more detrimental with nanostructured coating.

Since the beginning of the mid-1990s, much work have been devoted to the use of the spray technology for the deposition of finely or nanostructured coatings. To produce finely structured coatings by thermal spray techniques, five routes have been suggested:
  1. 1.

    Conventional spraying by plasma, wire arc, PTA, HVOF (with particle or drop sizes in the tens of μm range) of very complex alloys containing multiple elements (7–9 different components) having a low critical cooling rate (104 K/s) for metallic glass formation, in the case when cooled-down, to a undercooling temperature below the glass transition temperature. After deposition, upon subsequent heating in the 500–750 °C range, the metallic glass precursor transforms, through a solid-solid state deglazing, into multiple crystalline phases where the resulting structures are nano-scaled [7, 14].

     
  2. 2.

    Spraying of conventional micro-sized particles, by plasma or HVOF (in the 30–90 μm range), made of agglomerated nano-sized particles (mainly ceramics) [15] and operating the spray systems in relatively narrow spray window [5], in order to have the particles only partially melted [16]. The molten part of the larger particles, or the smaller fully melted particles act as “cement” bonding thereby unmelted nano-sized grains. Coatings exhibit a two-scale architecture, often called bimodal (finding application, for example, in thermal barrier abradable coatings made of partially stabilized zirconia).

     
  3. 3.

    Cold spraying of agglomerated nano or amorphous particles [17].

     
  1. 4.
    When considering particles in the sub-micro- or nano-scale size range, gas can be no longer applied for carrying (see Sect.  4.4.2) and thus the use of a liquid carrier becomes mandatory. With the liquid, injected as jet or drops, two routes can be implemented, the liquid containing:
    1. (a)

      sub-micron- or nano-sized particles via a suspension. Once the liquid has been fragmented and vaporized by the plasma flow or the HVOF flame, particles contained in droplets are heated, accelerated, and sprayed as molten droplets onto the substrate. On a metallic substrate, the first droplets result in the formation of splats, whose equivalent diameters range between 0.1 and 2 μm with an average thicknesses between 20 and 300 nm. However on substrates with poor thermal diffusivity (or on the first deposited layers of poorly conductive materials) and with a high thermal flux from hot gases (up to 20–40 MW/m2), cooling may be delayed. Thus the surface tension may take over and molten particles recover their spherical shape before solidification and forming coatings with granular structures. The stacking of splats or liquid particles forms finely structured coatings, often with a granular structure [5, 18]. It is also possible to spray a suspension of molecularly mixed amorphous powders as feedstock [19]. According to some authors this is an ideal process for the deposition of homogeneously distributed multicomponent ceramics coatings.

       
    2. (b)

      With solutions of precursors of materials to be deposited, as with suspension, the liquid undergoes rapid fragmentation and evaporation once injected in the plasma jet. This is followed by precipitation or gelation, pyrolysis, and melting to result finally in the impact of molten liquid droplets with average diameters ranging from a few tenths to a few micrometers [5, 20]. A sol–gel colloidal solution can also be used instead of the solution [21].

       
     

14.2 Spraying of Complex Alloys Containing Multiple Elements to Form Amorphous Coatings

According to Sergueeva et al. [22]: “Crystallization of metallic glasses has been successfully used as one of the methods for nano-crystalline material production in various alloy systems, e.g., in Fe-, Ni-, and Co-based alloys. This type of transformation involves decomposition of single-phase supersaturated solid solutions into multiphase nano-scale microstructures. To obtain a nano-scale structure, the crystallization process should proceed with the largest nucleation rate possible while suppressing the crystal growth rate as much as possible. Such conditions can be obtained for compositions of some alloys by applying specific methods of heat treatment…. Among metallic glasses those that undergo primary crystallization with a time-dependent, long-range diffusion controlled growth rate are the most suitable candidates for nanometer crystallization".

The methodology involves designing alloys that have low-critical cooling rates for metallic glass formation that result in the formation of amorphous coatings when thermally deposited. The amorphous coatings, when heated (after spraying) above their crystallization temperature are devitrified. Since the diffusion rate in the solid state is very low at the transformation temperature (typically 0.4–0.7 T m for iron alloys), nano-scale microstructures are formed [6, 23, 24, 25].

In the past decade, a series of new bulk amorphous alloys with a multicomponent chemistry and high-glass forming ability (GFA) have been developed in zirconium-, magnesium-, lanthanum-, palladium-, titanium-, and Fe-base systems with various rapid solidification techniques [23, 24, 25]. Compared with other amorphous alloy systems, such as zirconium- and lead-base metallic glasses, the advantages of Fe-base amorphous coatings are lower materials cost, higher strength, and higher wear and corrosion resistance. However to be suitable for thermal spray processes such materials should be produced by gas atomization.

14.2.1 Amorphous Alloys Containing Phosphorus

Complex alloys containing phosphorus are known to present amorphous phases and can be produced by gas atomization. Alloy powders of Fe–10 % Cr–8 % P–2 % C(10 Cr), Fe–20 % Cr–8 % P–2 % C(20 Cr), and Fe–10 % Cr–10 % Mo–8 % P–2 % C(10Mo) compositions (in wt%) were sprayed by the high velocity oxy-fuel (HVOF) process under different conditions where the oxygen and fuel pressures of the HVOF process were changed to vary the flame temperature [26]. The melting point of particles was under 1,273 K and their size distribution between 10 and 45 μm. Amorphous coatings with a small amount of crystalline phases were obtained from the 10Cr and 20Cr alloys and a 100 % amorphous coating was formed from the 10Mo alloy. The volume fraction of the crystalline material increased slightly with the rise of the flame temperature. The coatings appeared very dense, although some pores and oxide films were visible in the coatings. The hardness of the 10Cr and 20Cr coatings was 600–700 HV5N. On the other hand, the 10 Mo coatings composed of a perfect amorphous phase reveal a constant hardness of 560 HV5N independently of the spray condition. The as-sprayed coatings of the 10Cr and 20Cr alloys exhibited the activation–passivation transition. In contrast, the as-sprayed coating of the 10Mo alloy composed of a 100 % amorphous phase structure had excellent corrosion resistance in 1 N H2SO4 and 1 N HCl solutions [26].

14.2.2 NiCrB and FeCrB Alloys

Since the mid-1980s studies were devoted to the production and the properties of air and vacuum plasma or HVOF-sprayed amorphous/nanometer crystalline NiCrB- and FeCrB-based alloys [27, 28, 29, 30, 31]. For example, NiCrMoB alloys [31] were sprayed with a HVOF Miller spray system (Praxair Surface Technologies, Appleton, WI) with hydrogen as a fuel gas and gas and oxygen (H2 to O2 gas ratio 3 to 1). Both powders (two different NiCrMoB alloys: 25–63 μm) and deposits contained substantial quantities of amorphous/nano-crystalline matrices (as shown by the significant broadening of the fcc nickel peak at 2θ = 44°). Hardness values were in excess of 610 ± 10 HV3N. Higher molybdenum and boron levels led to higher hardness values of up to 810 ± 15 HV3N · The deposited layers were found to contain fine Cr5B3 precipitates within the metallic matrix phase. The predominant oxide phase, Cr2O3, occurred principally with a columnar-grained morphology at inter-splat boundaries. Anodic polarization curves showed that in 0.5 M H2SO4 a passive region occurred between approximately +100 and +900 mV (SCE). Increased molybdenum and boron levels modified the pre-passivation corrosion current density but had little effect on either passive current density or the onset of trans-passive corrosion. The potentiodynamic corrosion testing also demonstrated that the corrosion behavior of the experimental coatings A and B was similar to that of HVOF-sprayed Inconel 625. However, the presence of oxides at inter-splat boundaries leads to passive current densities higher than those observed in wrought alloys or melt spun ribbons.

In-flight particle oxidation during atmospheric plasma spraying [APS] and high velocity oxy-fuel [HVOF] spraying is however unavoidable. For example, with NiTiZrSiSn there is a preferential oxidation of Zr and Ti in an in-flight particle [32]. As a matter of fact, oxidation triggers the destabilization of the bulk of metallic glass particles because amorphous phase stability and formability are largely affected by the chemical composition depending on the critical cooling rate. With the HVOF process (H2–O2 mixture) [7] above the 0.20 O2/H2 ratio, severe oxidation was observed, inducing destabilization of the amorphous phase.

14.2.3 Iron-Based Amorphous Alloys

Nano-scale steels have appeared at the end of the past century and the first advances have boosted the field of hard magnets made of neodymium, iron, and boron. Branagan et al. [33] succeeded to synthesize nano-scale metal matrix composite microstructures in a 9-element modified Nd–Fe–B alloy, optimizing the magnetic behavior of permanent magnets

To obtain amorphous and nano-scale composite thermally deposited steel coatings a specialized iron-based composition (Fe63Cr8Mo2B17C5Si1Al4) with a low-critical cooling rate (104 K/s) for metallic glass formation has been developed. The alloy is processed by inert gas atomization to avoid oxidation and to form micron-sized amorphous spherical particles with a size distribution adapted to thermal spraying. The particle size distribution is particularly important because the particles have to melt to form the amorphous coatings. Collecting and quenching sprayed particles was performed to verify the particle characteristics. The results show that a significant fraction of the less-than-50 μm particles are amorphous, while particles greater than 75 μm are crystalline. In Fig.  11.7, backscattered electron micrographs indicate that small particles (<20 μm) have an absence of structure, while larger powder particles (50–75 μm) exhibit a primarily crystalline structure with areas of uncrystallized glass.

A primarily amorphous structure is formed in the as-sprayed coatings (using d.c. plasma or HVOF), independent of coating thickness. After a heat treatment above the crystallization temperature (568 °C which is less than half the melting temperature), the coatings are devitrified into a multiphase nano-composite microstructure with 75–125 nm grains containing a distribution of 20 nm second-phase grain-boundary precipitates [6]. A strong effect of annealing conditions on the nature of crystallization products and, as a result, on the mechanical behavior of the alloy has been revealed [6]. Accordingly, depending on the microstructure produced and the deformation conditions chosen, a wide variety of strength/elongation properties can be generated for various potential applications. To illustrate what happens when spraying such materials and reheating the deposit them the results obtained by Branagan [34] by melt spinning are presented below. He has considered a 5-element glass-forming steel alloy with the following stoichiometry, Fe63.2Cr15.8B17W2C2. The as-cast microstructure was found to be formed by a dendritic solidification morphology, resulting in a multiphase structure with an average phase scale from 1 to 5 μm (see Fig. 14.1a). Phases in the ingot were determined from X-ray diffraction and were found to include α-Fe, M23C6 and M3B. The same 5-element steel was processed by melt spinning at 15 m/s to form a homogeneous glass (glass-to-crystalline peak at 536 °C). After heat-treating the glass above 536 °C, complete deglazing occurred through the solid–solid-state transformation. The nanometer scale structures are shown in the TEM micrographs in Fig. 14.1b–d, respectively, obtained after heat treatment at 700 °C, 750 °C, and 850 °C for 1 h.
Fig. 14.1

Electron microcopy images of a 5-element steel alloy; (a) backscattered electron image of an as-cast ingot, (b) transmission electron image of ribbon heat-treated at 700 °C, (c) transmission electron image of ribbon heat treated at 750 °C, and (d) transmission electron image of ribbon heat treated at 850 °C. Reprinted with kind permission from Elsevier [34]

The phase size is still considered nano-scale up to at least 750 °C, with only a small change from the average size of 25 nm when heat treated from 700 to 750 °C. At the 850 °C heat treatment, extensive grain growth has clearly occurred with phase sizes approximately double those of the lower temperature heat treated samples [34]. When heat-treated at or below 700 °C, the nanometer scale microstructure is found to exhibit a hardness which is approximately 8.83 GPa higher than the micrometer scale microstructure. The nanometer-structured metal hardness exhibits two plateaus with little change in hardness from 600 to 700 °C, followed by a very significant drop in hardness from 700 to 750 °C, and then by a leveling off with finally another significant drop in hardness at 850 °C. Thus in engineering structures to improve properties, microstructural size alone is not the only important facet of the microstructure and other microstructural aspects must be considered. For more details, see the paper of Branagan [34].

Using HVOF deposition, the adhesion strength of the coatings was found to be excellent for a wide variety of metallic substrates. Good wear resistance was obtained in the three-body slurry abrasion tests [6]. Due to the high hardness, the amorphous and nano-composite coatings experienced good wear characteristics. No wear was found in the two-body abrasion tests in the as-sprayed or heat-treated plasma coatings. This was surprising, since the Si3N4 pin was much harder than both the as-sprayed and heat-treated coatings. The NanoSteel Company was recently created to develop and market a range of patented high-performance Super Hard Steel materials. For example, the new Fe-based commercial alloy (Fe52.3Mn2Cr19Mo2.5W1.7B16C4Si2.5) demonstrates strength of more than 6 GPa at room temperature [22], which is much stronger than the commonly used high strength (0.24–0.9 GPa) or even ultrahigh strength steels (0.9–1.5 GPa) of today. Such coatings can be applied with a wide variety of processes: thermal spraying, welding (MIG and PTA), and laser cladding. They are mostly made of Fe–Cr–Mo–W–C–Mn–Si–B or Cr–Mo–B–Si–W–Mn–C–Nb–Fe.

Certain iron-based glass forming materials such as SAM2X5 have an optimized chemistry with a protective adherent oxide layer and in principle a good corrosion resistance. However, Branagan et al. [7] have emphasized that when using optimized HVOF spray parameters to achieve high-density coatings, the development of the glass structure seems to be the dominant factor affecting corrosion performance. All HVOF coatings, regardless of powder lot were found to be primarily amorphous with only subtle differences in crystallinity. However, because the larger particles in the distribution are not completely remelted during spraying, the starting crystallinity in the feedstock powder is found to be an important factor. It is believed that the cores of these larger particles cause the coarsening of the microstructure. These pockets or bands of crystallinity can initiate anodic attack and result in reduced corrosion performance. Thus for a good resistance to corrosion the glass content in the feedstock powder is a paramount factor and must be maximized.

Due to their high abrasion, erosion, and corrosion resistance, hardness and toughness, the main applications of such coatings are boiler tube refurbishment, hard chrome replacement, hard facing, pumps, etc. Generally HVOF spraying is used because of the high particle velocities resulting in dense coatings. The SAM2XS amorphous steel has been used to replace hard chromium [35]. Coatings, sprayed with a JP5000 HVOF gun, were found to be nanometer-structured, hard (about 1,000 HV3N as sprayed and 1,200 HV3N after heat treatment at 700 °C for 10 min, approaching hardness of WC-Co coatings), with an excellent corrosion resistance to sea water and salt fog environment, and a wear resistance far superior to that of stainless steel used for this corrosion resistance. The SHS7574 powder from NanoSteel was successively sprayed with a D-gun and HVOF [36]. In both cases coatings were dense but those sprayed with a D-gun showed better adhesion and presented very good wear resistance.

Branagan et al. [14] have introduced a new iron-base cored wire, SHS7170, which readily forms nanometer composite coatings when sprayed using the wire arc process and exhibits a combination of properties that are superior to the existing available materials for high temperature boiler applications. The oxide content in the coatings is very low and is typically <1 vol.%. The bond strengths are remarkable for wire arc coatings and represent some of the best-published values, along with high toughness and resiliency. The as-sprayed SHS7170 wire arc coatings are found to develop an amorphous matrix structure containing starburst-shaped boride and carbide crystallites with sizes ranging from 60 to 140 nm. After heating to temperatures above the peak crystalline temperature (566 °C), a solid/state transformation occurs that results in the formation of an intimate three-phase matrix structure consisting of the same complex boride and carbide phases, along with α-iron interdispersed on a structural scale from 60 to 110 nm. The nanometer composite microstructure contains clean grain boundaries, which are found to be extremely stable and resist coarsening throughout the range of temperatures found in boilers. The SHS7170 coatings exhibit high bond strength along with high toughness and resiliency. The elevated-temperature erosion resistance of the SHS7170 wire arc coatings was found to be superior based on thickness loss compared with the existing wire arc coatings that have been tested.

Recently a new iron-based alloy SHS7170 has been developed [14]. Upon wire arc spraying, it forms an amorphous matrix with very fine dispersed nanometer-sized crystals that were identified as ferrites and M2(C,B)1 compounds. When exposed to elevated temperatures, the microhardness slightly increases at relatively low temperatures from 316 °C to 482 °C, and it dramatically increases up to ~1.77 MPa when the temperature is raised to 649 °C. The wear resistance was improved by a factor of 3.5 with the significant hardening. The SHS7170 coatings were found to exhibit high bond strength along with high toughness and resiliency. The elevated-temperature erosion resistance of the SHS7170 wire arc coatings was found to be superior based on thickness loss compared with the existing wire-arc coatings that have been tested. SHS7170 coatings resisted erosion almost independently of contact angle and at temperatures at least up to 600 °C. Zhou et al. [37] have shown that the SHS8000 coatings exhibited high bond strength along with high toughness and resiliency. The development of the time–temperature–transformation diagram for the SHS8000 coating allowed predicting the coating performance as a function of temperature and time during elevated temperature exposure.

Fe-base nano-crystalline SHS7172CP1 (particles with diameters 53–150 μm ranging) wear-resistant materials have been cladded on aluminum alloys using a PTA with a negative work piece polarity (see Sect.  10.6.7) process where the heat input into the base material and the formation of spatter can be limited [38]. The dilution is limited to approximately 5 %, while the high conductivity of the base material is responsible for a high cooling rate, which induces nano-crystalline solidification.

14.3 Agglomerated Ceramic Particles Spraying with Hot Gases

14.3.1 Spray Conditions

In order to spray nano-sized particles using regular powder feeders the nano-sized ceramic particles are agglomerated via spray drying (and then sintered) into micro-sized particles. For details about the powders used see Sect. 11.1.2.6. It is important to point out that for these particles, finding the trade off threshold between providing cohesive strength and maintaining the nanostructural character of the feedstock is paramount, i.e., too high heat-treatment temperatures and/or long heat-treatment times may cause the partial or total loss of the nanostructural character of the powder due to particle coarsening and sintering effects [5].

These agglomerated particles can be either dense, or rather porous. Finally [5] not all so-called nanostructured agglomerated powders commercially available are formed via the agglomeration of individual nanostructured particles. For example, Fig. 14.2a shows a spray-dried “nanostructured” agglomerated Al2O3–13 wt% TiO2 (alumina–titania) powder particle (Nanox S2613S, Inframat Corp., Farmington, CT, USA). By viewing this particle at higher magnification (Fig. 14.2b) it is possible to observe that the agglomerate exhibits individual particles varying from about 15 to 300 nm [9]. Therefore, it is suggested that the term “ultrafine” agglomerate is more scientifically rigorous to describe the morphology of these types of powders, which are formed from a mixture of nanometer and sub-micrometer-sized particles.
Fig. 14.2

(a) Spray-dried agglomerated Al2O3–13 wt% TiO2 particle. (b) Higher magnification view showing the “ultrafine” character of the agglomerate. Reprinted with kind permission from IOP organization [5]

Lima and Marple [15] have recently published an excellent overview of the spray process of agglomerated nanometer-structured ceramic particles. The spray processes, except cold spray, are intrinsically associated with the melting of particles, or at least their partial melting. Without some particle melting it is extremely difficult to produce thermal spray coatings, especially ceramic ones. Thermal spraying nanostructured powders is thus a challenge: if all powder particles are fully molten in the thermal spray jet, all the initial nano-structures will disappear. In order to overcome this challenge, it is necessary to carefully control the temperature and size distribution of the particles in the thermal spray jet to keep part of the particles in a semi-molten state. Based on this fact, the expression “nanostructured thermal spray coating” is not the most accurate to designate or represent these types of coatings. The expression “bimodal coating,” representing the mixture of particles in the coating microstructure that were previously semimolten and fully molten in the spray jet, is more scientifically rigorous. However, as the term “nanostructured thermal spray coatings” is the most widely used by the thermal spray community to designate these types of coatings, it will also be employed in this manuscript [5].

Shaw et al. were among the firsts [39] to plasma spray Al2O3–13 wt% TiO2-agglomerated nano-sized particles. They showed that the phase content and grain size of the coating exhibit a strong dependency on the spray temperature. Figure 14.3a shows schematically a micro-sized granule containing hundreds of nano-sized Al2O3 and TiO2 particles. If, during the short exposure to the plasma jet, the temperature to which particles will be submitted is low enough, the TiO2 particles (melting temperatures of 1,854 °C) will be melted, while the Al2O3 nano-grains (melting temperatures 2,040 °C) will remain unmelted (see Fig. 14.3b) [39]. Of course, besides the temperature and size distribution of the particles, the control of the melting also depends on the way particles have been manufactured [40].
Fig. 14.3

Schematic of (a) A reconstituted Al2O3–TiO2 granule and (b) the partly melted granule where solid Al2O3 nanometer grains exist within a liquid TiO2. The viscosity of such a “mushy” granule depends on the volume fraction of the liquid TiO2 and the temperature. Reprinted with kind permission from Elsevier [40]

With zirconia particles the problem is even more complex because particles of the average size must be only partially melted: a molten shell surrounding an unmelted core (see Fig. 14.4). For example Fig.  4.42 of Sect.  4.3.3.4 presents the calculated evolution of the melt front within three agglomerated zirconia particles of different sizes as a function of the axial position in a plasma jet [16]. It allows evaluating the size of the nanometer-structured core, preserved at the end of the heat treatment of zirconia particles (the mass density of porous particles used for calculations was 2,840 kg/m3). The particle, 40 μm in diameter, is totally melted after a 40 mm trajectory, whereas within particles of 50 and 60 μm diameter, nanostructured solid cores remain, respectively, of 27 and 56 μm in diameter corresponding to 54 and 71 % of their initial diameter. This result shows that, for these plasma spray conditions (60 slm, Ar + 25 vol.% H2, nozzle internal diameter 7 mm, effective thermal power 18.24 kW), there is a lower limit to the particle size below which a nanostructured solid core cannot be retained. The dense zirconia particle, 50 μm in diameter, melts much more than the porous particle due to the better heat propagation. Of course, different results are obtained when considering particles with different porosities corresponding to different powder manufacturing processes (see Sect.  11.1.2.6). The particle diameter also varies, controlled by both its surface vaporization and densification due to the porosity loss, as illustrated in Fig. 14.5 for 60 μm zirconia particles with different porosities. As it could be expected, the size of the nanostructured core increases when porosity increases. Figure 14.6 illustrates the evolution of the melting front as a function of the axial position for three zirconia particles 60 μm in diameter but with different porosities.
Fig. 14.4

Mechanism of nanostructured coating building. Reprinted with kind permission from Elsevier [16]

Fig. 14.5

Axial evolution of the radius of agglomerated zirconia particle (d p = 60 μm) with different porosities: 20 %; 30 %; and 40 %. Reprinted with kind permission from Elsevier [16]

Fig. 14.6

Axial evolution of the melting front in agglomerated zirconia particles (d p = 60 μm) with different porosities: 10 %; 30 %; 50 % corresponding respectively to thermal conductivities κ = 1.49; 1.16; 0.83 W/m K. Reprinted with kind permission from Elsevier [16]

To spray these nanostructured agglomerates, the spray parameters must be optimized to produce conditions (particle temperatures and velocities) that result in only partial melting of agglomerates (to avoid the complete loss of the nanostructure). However a sufficiently high degree of melting must be obtained to ensure effective deposition on the substrate and the formation of the so-called nano-zones [15].

Guru and Heberlein [41] have used the triple torch plasma reactor (TTPR), for details see Fig.  7.57, along with a custom-designed shutter mechanism to deposit splat samples of agglomerated nano-sized particles of YPSZ (8 wt% of Y2O3). The TTPR was working with pure argon and the particle size distribution was 15–150 μm. A typical splat, deposited on polished Mo substrate, is presented in Fig. 14.7. The photo shows the features of a semimolten agglomerate particle having the nano-phase structure preserved. A layer of such particles has shown some plastic deformation properties making it suitable as a strain relief intermediate layer between two layers with different thermal expansion characteristics. Many solid particles are present within the melted matrix. The splat shape shown in Fig. 14.7 is conical because only a shell of the particle was fully melted and the impact velocity was relatively low.
Fig. 14.7

Cross-sectional view of a splat resulting of semimolten structure of agglomerated nanometer-sized particles of YPSZ (8 wt% of Y2O3) sprayed by triple torch plasma reactor working with Argon. Reprinted with kind permission from Prof. J. Heberlein [41]

One of the critical goals of parameter optimization is to control the density of the nano-sized zones, i.e., the density of the semimolten nanostructured agglomerates embedded in the coating microstructure [5]. This is achieved by finding the conditions to adjust the amount of the molten part of each semimolten particle that penetrates into the capillaries (i.e., the non-molten particle core) of the agglomerates (Fig. 14.4) during flight within the thermal spray jet and/or at the impact on the substrate surface and subsequent re-solidification. It is important to point out that by introducing porous or dense nano-sized zones throughout the coating microstructure, it is possible to engineer coatings with very different and even opposite properties for a variety of purposes [5]. For example, for thermal barrier and abradable seal application, the presence of porous nanometer-sized zones is paramount. On the other hand, for anti-wear applications, dense nanometer-sized zones are absolutely required to induce high levels of wear resistance [15], as will be discussed later on. A schematic of the embedding of porous and dense nanometer-sized zones throughout the coating microstructure can be found in Fig. 14.8.
Fig. 14.8

Schematic of the embedding of porous and dense nanozones throughout the coating microstructure during thermal spraying. Reprinted with kind permission from IOP organization [5]

As shown in Fig. 14.9 [15] the coating microstructure is formed by semimolten feedstock particles that are spread throughout the coating microstructure and are surrounded by fully molten particles that act as a cement or binder. Such nanostructured thermal spray coatings exhibit significantly higher crack propagation resistance or relative toughness when compared to conventional coatings.
Fig. 14.9

Typical schematic (cross section) of the bimodal microstructure of thermal spray coatings formed by fully molten and semimolten nanostructured agglomerated particles. Reprinted with kind permission from Springer Science Business Media [15], copyright © ASM International

Some authors [42] have described these coatings as exhibiting a “bimodal microstructure.” This is illustrated in Fig. 14.10 for nanostructured YPSZ (ZrO2–7 wt% Y2O3) experimental feedstock Nanox S4007 (Inframat, North Haven, CT) plasma sprayed in air on low-carbon steel substrates. The Metco 3 MB torch (GH nozzle, powder port c2) torch was used with 40 slm Ar, 12 slm H2, 600 A, at 40 kW. The nanostructured coatings show hardness values similar to those of conventional coatings (with the same size distribution) in the low slope regions of the Weibull plots. Since the conventional coating primarily represents the behavior of fully molten particles, this observation indicates that the low slope regions of the Weibull plots reflect microstructural features of the nanostructured feedstock, which are fully molten. On the other hand, the high slope regions at low hardness values represent the non-molten feedstock particles in the coating microstructure [42].
Fig. 14.10

Weibull plot of Knoop microhardness for the nanostructured coating (Nanox S4007) plasma sprayed with the Metco 3MB (40 slm Ar, 12 slm H2, 600 A, 40 kW). Reprinted with kind permission from Elsevier [42]

To achieve this bimodal structure, the temperature of the powder particles should be maintained such that it is not significantly higher than the melting point of the material [5]. The main technique employed for optimizing the spray parameters to engineer coatings produced from nanostructured agglomerated particles is in-flight particle temperature and velocity monitoring. One of the major advantages of in-flight particle monitoring is the fact that it can be employed to optimize spray parameters for plasma spray, flame spray, and high velocity oxy-fuel (HVOF) torches [5]. An essential element of using in-flight particle monitoring to engineer nanostructured coatings is the knowledge of the melting point of the material to be sprayed [15]. Lima et al. [43] have sprayed, using an air plasma spray (APS) torch (F4-MB), the same nano-sized particles (ZrO2–7 wt% Y2O3 Nanox S4007). Figure 14.11 shows the distribution of particle temperatures and velocities of plasma-sprayed YPSZ particles (measured with the DPV2000, Tecnar Automation, Saint-Hubert, QC, Canada). The in-flight particle characteristics were monitored at the same spray distance as used to deposit the coating.
Fig. 14.11

In-flight particle surface temperature and velocity during the plasma spraying of nanostructured agglomerated YSZ particles. Reprinted with kind permission from IOP organization [5]

The melting point of YSZ or YPSZ is around 2,700 °C. By controlling and adjusting the main plasma spray parameters (i.e., torch power, gas flow rates, current, and spray distance), it is possible to control the temperature distribution of the particles. If the majority of the particles in Fig. 14.11 are below the melting point of the YSZ, no effective coating deposition will occur. On the other hand, if the majority of the particles are located above the melting point of the material, although a high deposition efficiency could be achieved, most of the nanostructural character of the feedstock would tend to be destroyed during deposition. It can also induce deeper penetration/infiltration of the molten part, formed near the surface of the agglomerates, into their non-molten cores (capillaries) to form dense nanometer-sized zones. At the conditions depicted in Fig. 14.11 the average particle temperature and velocity are 2,633 ± 174 °C and 213 ± 52 m/s, respectively. Therefore, distribution of particles within the plasma spray jet is around the melting point of YPSZ, i.e., approximately 50 % of the particles are above and 50 % below 2,700 °C. Achieving this type of distribution is an important factor in preserving and embedding part of the original porous nanostructure of the powder throughout the coating microstructure. However, the particle temperature and velocity distributions are not the only important factors in the processing [5]. The particle size distribution is also a key factor. For example, the YPSZ agglomerates used are exhibiting a distribution varying from about 40 to 160 μm [5]. Considering the fact that those typical particle size distributions for thermal spraying are generally found within the range of 10–90 μm, these sprayed particles tend to be quite large. Large particles are necessary in agglomerated nano-sized plasma spraying to avoid a high degree of particle melting caused by the high temperatures of the plasma jet [5]. As described by Fauchais [44], the plasma temperatures a few millimeters downstream of the plasma torch nozzle can be as high as ~15,000 K, i.e., many times higher than the melting point of any material known in nature. In addition, the particle velocities in the plasma jet, in the order of ~150–300 m/s (Fig. 14.11), are considered to be relatively low compared to those possible for some thermal spray processes. Consequently, the dwell time of these particles in the plasma jet is considered to be “sufficient” to induce a high degree of particle melting. The use of large agglomerates is paramount to engineer architectures that exhibit porous nanometer zones embedded in the coating microstructure (Fig. 14.12). This figure shows the microstructure of the coating produced. Lighter colored and darker colored zones can be distinguished in the coating microstructure (Fig. 14.12a), i.e., the bimodal distribution. In the darker colored zones, at higher SEM magnifications (Fig. 14.12b, c), it is possible to recognize the similarities with the morphology of the nano-YPSZ feedstock. The zones like those shown in Fig. 14.12b, c correspond to semimolten nanostructured agglomerated YPSZ particles that became embedded in the coating microstructure. This is achieved when the molten part of each semimolten particle does not exhibit a high level of penetration into the capillaries of the agglomerates during flight and upon impact and resolidification. Nanostructured YSZ coatings exhibiting these porous nano-zones may be very good candidates for thermal barrier coatings (TBCs) and pure ceramic abradable coatings for the hot sections of gas turbine engines [15, 43, 45]. The percentage of semimolten porous agglomerates embedded in the coating microstructure of Fig. 14.12 was estimated via image analysis to be around 35 %.
Fig. 14.12

(a) Microstructure of the nano-structured zirconia–yttria coating made from a nanostructured feedstock. (b) Darker colored region containing the semimolten feedstock particles. (c) Detail of semimolten feedstock contained in dark zone of (b). Reprinted with kind permission from IOP organization [42]

As pointed out by Lima and Marple [15], the key parameter to engineer coatings with very pronounced differences in microstructural characteristics and mechanical performance is the density of the nano-sized zones. This density depends on thermal processing, spraying conditions, and feedstock characteristics (e.g., diameter and density). The nanometer-zones that form the coating can be porous, like the original feedstock (see Fig. 14.12b, c), or can be much denser. The porous nanometer-sized zones are expected to appear when the molten part of the agglomerated semimolten particle does not fully infiltrate its non-molten core during thermal spraying. Using very porous nanostructured agglomerated particles can help obtaining porous nano-sized zones. On the other hand, the dense nano-sized zones probably occur when the molten part of an agglomerated semimolten particle fully or almost fully infiltrates into the small capillaries of its non-molten core, either in the spray jet or at impact with the substrate.

Spraying nanostructured powders with less control of temperatures may become possible if compensated by a significant high particle velocity (e.g., HVOF particle velocities) [15]. The control of particles melting is more difficult with plasma spraying [46] than with HVOF process, in spite of the fact that spraying ceramic materials by HVOF is a challenge. This is due to the high melting point of ceramic materials and the low flame temperatures of HVOF torches (<3,000 °C). HVOF spraying seems to be more adapted to ceramic materials which melting temperature is relatively low such as titania or hydroxyapatite. However controlling the density of the nano-sized zones makes it possible to engineer very distinct properties in thermal spray coatings produced from nanostructured agglomerated powders [5]. Dense nano-sized zones are extremely important if the goal is to produce high performing antiwear coatings. Dense nano-sized zones are obtained when the molten part of each semimolten particle exhibits a high level of penetration/infiltration into the capillaries of the agglomerates (non-molten particle core) before and/or at the impact on the substrate surface and subsequent resolidification [5, 15, 47]. To achieve these dense nano-sized zones or dense ultrafine zones to engineer high performing antiwear coatings, the “ultrafine” alumina–titania ceramic powder of Fig. 14.2 was sprayed via HVOF. The HVOF technique was chosen due to the fact that the powder particles can be accelerated to velocities up to 800–1,000 m/s, thereby producing highly dense coatings, which are paramount for antiwear applications [5]. As previously mentioned, the drawback of this technique is found in its low-thermal spray jet temperature levels. In spite of the fact that HVOF temperature levels are higher than the melting point of the majority of the engineering materials, the high particle velocities shorten considerably the dwell time of the particles in the spray jet. From a certain point of view, it can be stated that HVOF processing (i.e., high velocity and low temperature) is the opposite of plasma spray processing (i.e., low velocity and high temperature). This is the main reason why ceramics are typically sprayed via plasma spray process. The HVOF processing is generally employed to spray materials that are sensitive to high temperatures, such as metals or cermets (e.g., WC–Co and Cr3C2–NiCr). The first challenge to be overcome when spraying pure ceramics via HVOF is found in the particle size distribution [5]. As previously explained, for plasma spraying, to produce and embed porous nano-sized zones in the coating microstructure it is necessary to employ large particles with a wide particle size distribution (40–160 μm). In HVOF, to produce and embed dense nano-sized zones in the coating microstructure, it is necessary to employ small particles with a narrow particle size distribution. The small particles are absolutely necessary to compensate for (1) the low temperature of the HVOF jet and (2) the short dwell time of these particles in the jet. By doing so, the probability of creating an environment where the molten part of each semimolten particle penetrates/infiltrates deeply into the capillaries of the agglomerates (non-molten particle core) before and/or at the impact on the substrate surface is maximized [5]. Figure 14.13 shows the particle size distribution of the spray-dried ultrafine alumina–titania agglomerates depicted in Fig. 14.3. The particle size distribution of the original (i.e., as-purchased) powder varies from ~5 to 90 μm. To produce a powder with a size distribution more appropriate for use with HVOF, this material was sieved to yield a product having particles within the size range of 5–30 μm (Fig. 14.13).
Fig. 14.13

Original and sieved particle size cuts for the ultrafine alumina–titania powder of Fig. 14.2. Reprinted with kind permission from IOP organization [47]

For comparison, a conventional Al2O3–13 wt% TiO2 powder (Amperit 744.0, H. C. Starck, Goslar, Germany) was also sprayed via HVOF process (DJ2700-hybrid, Sulzer Metco, Westbury, NY, USA). The powder was constituted from a blend of fused and crushed alumina and titania particles [47]. The particle size distribution for this powder was ~5–25 μm, i.e., within the typical particle size range for HVOF torches. The melting point of the Al2O3–13 wt% TiO2 composition is ~1,900 °C. The in-flight particle characteristics were monitored and adjusted (using DPV 2000, Tecnar Automation, Saint-Hubert, QC, Canada), and the results are shown in Fig. 14.14. It is possible to observe that the overall temperature distributions of particles from both powders are well above the melting point of the alumina–titania composition. As it was the case for the nanostructured YPSZ coatings, the in-flight particle characteristics were being monitored at the same spray distance used to deposit the coatings.
Fig. 14.14

Particle temperature and velocity levels for the ultrafine and conventional alumina–titania powders. Reprinted with kind permission from IOP organization [5]

It is important to point out that this type of in-flight particle monitoring is based on pyrometer measurements, i.e., the temperature measurements represent the external temperature of the particle. Therefore, even if the temperatures at the outer surface of the particles are above the melting point of the material, it does not necessarily mean that the inner cores of the particles are also fully molten. It has to be stressed that the velocities in the HVOF jet, are in the range of 900–1,100 m/s for the ultrafine agglomerates (Fig. 14.14), which is considered high for thermal spraying. Consequently, the dwell time of these particles in the HVOF jet is considered to be “short” and higher particle temperatures and smaller particle diameters must compensate for these short-heating times.

Lima and Marple [48] have compared coatings sprayed by HVOF (Diamond Jet 2700-hybrid, Sulzer Metco) using nanostructured titania feedstock (Altair VHP-DCS, 5–20 μm) to plasma-sprayed coatings (APS torch SG100 Praxair) using conventional alumina–titania (Al2O3–13 wt% TiO2) feedstock (Metco 130 nominal particle size range from 15 to 53 μm). The particle temperature and velocity of the feedstock particles in-flight in both spray jets were measured using the DPV 2000 at the spray distance at which the substrates would be positioned when depositing a coating (APS system 6.4 cm; HVOF system 20 cm). As it could be expected the air plasma-sprayed alumina–titania coating was 33 % harder than the HVOF-sprayed nanostructured titania. But the nanostructured titania coating exhibited a superior abrasion wear resistance (27 % lower volume loss) when compared with the air plasma-sprayed conventional alumina–titania. It also exhibited a crack propagation resistance almost twice that of the air plasma-sprayed conventional alumina–titania coating (i.e., the nanostructured coating is tougher than the conventional one).

To conclude, spraying agglomerated-nanosized particles, to obtain bimodal structures, is not simple and requires staying within a rather narrow spray window. However, as pointed out in the review [5] Lima and Marple have delineated some general and practical engineering guidelines based on previous experimental results and Table 14.1 summarizes the different conditions to achieve dense nano-sized zones by APS or HVOF or porous nano-sized ones by APS. The first point is the necessity to use a sensor measuring the particle temperatures and velocities distributions (particularly the average temperature \( \overline{T} \) and its standard deviation σ T ), and not ensemble measurements (see Sect.  16.3.4.2). Such measurements should be performed at the same spray distance used to deposit the coating. The particle size distribution is the other key parameter controlling the coating bimodal structure and the nano-zones density.
Table 14.1

Spray parameters to achieve dense nano-sized zones by APS or HVOF or porous nano-sized ones by APS

 

Particle size distribution

Particles surface average temperature

Spray distance

APS (porous nanostructured zones)

Large particles exhibiting a wide particle size distribution, e.g., ~10–150 μm

\( \overline{T} \) and σ T must be adjusted (via spray parameters and particle size distribution) to overlap the melting point of the material

Adjusted to be the initial point at which \( \overline{T} \) and σ T overlap the material melting point

APS (dense nanostructured zones)

The typical particle size distribution for APS systems should be used, e.g., within the range ~5–90 μm

\( \overline{T} \) and σ T must be adjusted (via spray parameters and particle size distribution) to overlap the melting point T M of the material

Compromise between the previous condition and a larger spray distancea such as that used for conventional powders

HVOF (dense nanostructured zones)

Small particles exhibiting a narrow particle size distribution should be used, e.g., ~5–30 μm

\( \overline{T} \) and σ T must be adjusted (via spray parameters and particle size distribution) to be over the melting point T M of the material

Due to the small size of the agglomerates, the spray distance should be around the minimum one required to achieve \( \overline{T}>{T}_{\mathrm{M}} \)

aLarger spray distances yield longer dwell times and induce a higher degree of particle melting and infiltration into the capillaries of the non-molten cores

14.3.2 Applications

The bimodal nanostructured coatings can be engineered to exhibit different properties and microstructures (sometimes with completely opposite performance characteristics such as antiwear and abradable coatings!) by spraying nanostructured ceramic agglomerated powders via air plasma spray (APS) or high velocity oxy-fuel (HVOF) [15].

14.3.2.1 Wear-Resistant Coatings

Compared to titania or alumina–titania conventional coatings, those made with nanostructured agglomerated particles, either plasma [49, 50, 51, 52, 53, 54, 55, 56, 57, 58] or HVOF [47, 48, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69] sprayed, have better abrasion or sliding wear performance [15]. Such coatings exhibit significantly higher crack propagation resistance or relative toughness when compared to conventional ones [15]. The hardness of conventional and nanostructured coatings are generally about the same.

The thermal diffusivity of the nanostructured Al2O3–3 wt% TiO2 coating was higher compared with that of the corresponding conventional coating at temperatures ranging from 200 to 1,000 °C [69]. Zirconia nanostructured plasma-sprayed coatings have also been used against wear. The nanostructured coating reduced the wear rate of this coating against stainless steel [70, 71].

Lima and Marple [5] have compared the abrasion wear behavior of coatings sprayed with conventional Al2O3–13 wt% TiO2 powder (Amperit 744.0, H. C. Starck, Goslar, Germany) and the nanostructured agglomerated powder depicted in Fig. 14.2 with the size distribution presented in Fig. 14.13, both sprayed via HVOF (DJ2700-hybrid, Sulzer Metco, Westbury, NY, USA). The comparison was achieved via the measurement of volume loss after abrasive wear testing. The ultrafine coating exhibited a 4-fold improvement in wear performance. The superior abrasion wear performance of the ultrafine coating could not be explained based on Vickers microhardness values. Both coatings exhibited average Vickers microhardness number values of ~8003N [47]. It has to be pointed out that Turunen et al. [72] compared the abrasion wear resistance of nanostructured and conventional alumina (Al2O3) coatings deposited via HVOF, and they observed superior wear behavior of the nanostructured coating. The evolution of the crack propagation resistance for the alumina–titania coatings turned out to be the main reason for the superior wear behavior of the ultrafine coating (Fig. 14.15) [47]. The evaluation of the crack propagation resistance, which is a measure of relative toughness, is done by indenting the cross section of the coatings using a Vickers indenter (5 N load) with one of its diagonals parallel to the substrate surface to induce crack propagation in between layers. From tip-to-tip, the crack length of the conventional coating (Fig. 14.15a) was found to be ~40 % longer than that of the ultrafine one (Fig. 14.15b). Therefore, the ultrafine coating is significantly tougher than its conventional counterpart.
Fig. 14.15

Crack propagation induced by indention on the cross section of the (a) conventional and (b) ultrafine alumina–titania coatings. Reprinted with kind permission from Elsevier [47]

The discussion below follows that presented in [5]. The question that arises is related to the role of the ultrafine structure of the agglomerated powder in the enhancement of the coating toughness. Gell et al. [51] and Luo et al. [73] observed that the enhanced toughness of these types of coatings was related to crack arresting and deflection when passing through dense nano-zones embedded in the coating microstructure. McPherson has shown [74] that for conventional thermal spray ceramic coatings, cracks will tend to propagate between adjacent layers of the coatings (i.e., the splat boundaries), which are the weakest link of the microstructure. When the microstructure of the ultrafine coating of Fig. 14.15b is analyzed at higher magnifications by using a SEM, it is possible to observe lamellar zones constituted of a finely dispersed material (Fig. 14.16a) distributed evenly within the coating cross-sectional microstructure [47]. By analyzing one of these finely dispersed regions in more detail (Fig. 14.16b) and comparing it with the microstructure of the ultrafine agglomerated powder (Fig. 14.2), one can realize from the similarities that these finely dispersed zones are semimolten ultrafine particles embedded in the coating microstructure. However, due to processing conditions (i.e., particle temperature, velocity, diameter, and dwell time), the molten part from the surface of the agglomerates deeply infiltrated into the capillaries of the non-molten cores, thereby creating dense ultrafine microstructure zones after particle impact and resolidification, as also estimated by Shaw et al. [46]. The percentage of semimolten dense agglomerates embedded in the coating microstructure of Figs. 14.15b and 14.16 was estimated via image analysis to be ~50 % [47]. When the left crack tip of the ultrafine coating of Fig. 14.15b is observed in detail at higher SEM magnifications (Fig. 14.16c), it is noticed that the crack is deflected and arrested when passing thorough a dense ultrafine zone. As in these coatings the microstructure is disrupted periodically by these zones, the crack propagation encounters barriers to propagate through the well-defined layered structure, thereby enhancing coating toughness, as also observed by Gell et al. [51] and Luo et al. [73]. Improved bond strength values have been reported for these coatings when compared to their conventional counterparts [15]. The explanation for this improved bond strength is similar to that used to explain the enhanced toughness. According to Bansal et al. [75], for coatings produced from nanostructured agglomerated powders, the interfaces between the dense nanometer zones and the substrate surface did not exhibit micrometer cracks or gaps, whereas for those of the conventional coatings, micro-sized cracks or gaps were observed. Therefore, the dense nano-zones would tend to impede crack propagation at the interface, which would enhance interfacial toughness and the bond strength levels of the coating. It has to be pointed out that other authors are attributing these higher toughness levels to an enhanced inter-lamellar strength, and not crack path deflection. Ahn et al. [76] observed that in spray-dried nanostructured agglomerated powders, the individual alumina and titania particles are intimately mixed, as readily observed in Fig. 14.2. The same degree of homogeneous mixture is not found in conventional particles. The melting point of pure alumina is 2,050 °C, whereas that of pure titania is 1,855 °C. An addition of 13 wt% of titania into alumina results in a lowering of the melting point of the compound to ~1,900 °C. Therefore, Al2O3–13 wt% TiO2 “well-mixed” agglomerates (Fig. 14.2) would tend to arrive at the substrate exhibiting lower viscosity levels due to the lowering of the melting point, than those of the conventional particles. These lower viscosity levels would translate into an enhanced splat-to-splat cohesion, which would then improve the toughness of the coating. By looking at Fig. 14.15, it is possible to observe a “lower” degree of homogeneity of the coating produced form the conventional powder (Fig. 14.15a) when compared to that produced from the ultrafine powder (Fig. 14.15b). For the conventional coating the alumina-rich (dark) and titania-rich (light) phases are quite well defined, i.e., not well mixed. As titania has lower mechanical strength levels than alumina, it probably creates weak-link zones in between layers of the conventional coating. Therefore, based on these various studies it is believed that for alumina–titania coatings produced from nanostructured or ultrafine powders, two distinct mechanisms are acting together to enhance coating toughness: (1) crack path deflection and (2) enhanced splat-to-splat contact.
Fig. 14.16

Higher magnification views (cross section) of the ultrafine coating of Fig. 14.15b. (a) Overall view of the dense ultrafine zones. (b) Localized view of a dense ultrafine zone. (c) Crack tip arresting via path deflection by passing thorough a dense ultrafine zone. Reprinted with kind permission from Elsevier [47]

Such coatings for antiwear purposes are used in the “real world” applications [5]. Nanostructured alumina–titania coatings deposited by APS are in use to protect the main propulsion shafts of ships of the US Navy. After 4 years of use in naval applications no significant damage was recorded in these types of coatings. This work was based on a research led by the University of Connecticut (USA) [46, 51, 73, 75]. Nanostructured titania thermal spray coatings have been developed and applied with success by Perpetual Technologies (Ile des Soeurs, QC, Canada) in ball valves for autoclaves that operate at high temperature (~260 °C), at high pressure (~5.5 MPa), in corrosive sulphuric acid (>95 %), and at relatively high solids content (>20 wt%) [77]. The in-field performance of this nanostructured coating was compared to that of a conventional one. After 10 months of service the coatings were inspected. The ball valve coated with the conventional coating exhibited delaminating in several areas and had to be replaced, whereas the ball valve coated with the nanostructured coating was nearly intact and put back into service.

14.3.2.2 Abradable Coatings

By embedding a significant amount of semimolten porous particles in the coating microstructure, the ceramic coating becomes friable, producing a nanostructured ceramic abradable coating as shown by Lima et al. [5, 43]. The gap (i.e., clearance) between the tip of the turbine blades and the turbine case should be as tight as possible to improve engine efficiency and performance. It could be stated that abradable coatings are the “opposite” of antiwear coatings, i.e., they must possess a friable structure. It is a challenge to engineer these coatings because they must be at the same time readily abradable and able to withstand the harsh environment of the turbine engine, as described by Ghasripoor et al. [78]. The conventional abradable seal is a composite coating typically made from a metallic-based material, a self-lubricating phase and many pores. Due to environmental and economic issues, there is a significant driving force to increase the combustion temperature of gas turbines. Higher combustion temperatures will generate higher efficiency and less pollution. The abradable coatings have also to follow this trend. To achieve this objective two types of high temperature abradables are currently in use. One is based on a powder composite formed from a metallic superalloy (e.g., CoNiCrAlY), a self-lubricating agent (e.g., BN), and a polymer. The second one is based on a powder composite formed from YSZ, BN, and a polymer. For both cases, the polymer is introduced into the coating microstructure during deposition and burned off in a postdeposition heat treatment to create a large porosity network to make both types of coatings friable. This approach has challenges due to the fact it is difficult to engineer homogeneous microstructures when materials of very distinct physical and chemical properties are sprayed together (e.g., metal + polymer and ceramic + polymer). In addition, the postdeposition heat treatment necessary to burn the polymer off the coating microstructure represents an additional issue of cost and time. Therefore, the use of a pure ceramic abradable material could be a major advance in producing the next generation of high-temperature abradable coatings. However, the lack of plasticity of ceramic materials is a major barrier impeding progress in achieving this goal. For a pure ceramic abradable coating design, the concept of the porous nanometer-sized zones, if embedded in the coating microstructure in “sufficient numbers,” could be employed to lower the overall stiffness of the coating, allowing the tip of a metallic turbine blade to wear off some of the coating, creating the “seal effect,” without damaging the blade [5]. Figure 14.17a shows the wear scar after rub-rig testing for abradables on the top surface of the nanostructured YSZ coating of Fig. 14.12. The rub-rig is a standard test to evaluate the abradability of a coating. Essentially, a metallic blade is placed at the edge of a spinning wheel, which exhibits rotating speeds and incremental incursion rates similar to those of a real turbine engine. By rubbing the blade tip against the top surface of the coating and analyzing the blade tip and the coating wear scar after testing, it is possible to infer the degree of abradability of the material.
Fig. 14.17

(a) Wear scar (top surface) of the nanostructured YSZ coating after rub-rig testing. (b) Wear scar (top surface) of a state-of-the-art high temperature metal-based abradable coating after rub-rig testing. Reprinted with kind permission from Springer Science Business Media [15], copyright © ASM International

By looking at Fig. 14.17a it is possible to observe the absence of macro-cracks or any major damage, delaminating, or debonding in the coating structure after being rubbed off by the tip of a metallic blade. The wear scar is “well-shaped” and smooth, and no wear on the blade was recorded. By comparison, a state-of-the-art metallic-based high temperature abradable (CoNiCrAlY + BN + polyester) was tested under the same conditions (Fig. 14.17b). The volume losses of both coatings were measured and they exhibited an almost identical value [15]. Consequently, these results show that the ceramic abradable exhibited a performance level similar to that of the state-of-the-art metallic-based abradable. Therefore, based on the information and promising results available in this section, it can be envisioned that the use of nanostructured YSZ coatings engineered to contain porous nanometer-sized zones for thermal barriers and abradables may become major applications of this technology for the next decade.

14.3.2.3 Thermal Barrier Coatings

Much work has been performed on plasma-sprayed nanostructured TBC coatings [79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89]. Lima and Marple [5, 15] have extensively studied these coatings. TBCs are made from low-thermal conductivity (<2 W/m/K) ceramics that also exhibit high mechanical and chemical stability at high temperatures. According to Padture et al. [90], they are employed to protect and insulate the metallic components of the hot sections of gas turbine engines (e.g., blades, vanes, and combustion chambers) from the hot gas stream provided by the fuel combustion, allowing higher combustion temperatures and improving engine efficiency. These turbine engines are employed in aerospace propulsion, power generation, and marine propulsion.

Regarding TBC applications, basically the porous nanometer-sized zones will reduce significantly the increase of the elastic modulus and thermal conductivity values of these coatings when exposed at high temperatures by counteracting densification effects via differential sintering, which are highly desired characteristics for TBC applications. In fact, it has been shown that for specific coatings in some studies the growth rate of the thermal conductivity (Fig. 14.18a) and elastic modulus (Fig. 14.18b) values of conventional YPSZ TBCs at 1,400 °C were approximately 5 times higher than those of the nanostructured YPSZ TBCs [91, 92]. By looking at Fig. 14.19 it is possible to understand how the differential sintering mechanism works. The as-sprayed coating (Fig. 14.19a) exhibits the typical porous nanometer-sized zones of Fig. 14.12. The overall coating porosity increased after an exposure at 1,400 °C for 20 h (Fig. 14.19b), which is a counterintuitive effect. Essentially, the nanometer-sized zones sinter at much faster rate than the coating matrix, thereby inducing porosity and reducing the growth rate of thermal conductivity and elastic modulus values, which are governed by sintering. When the conventional YSZ is observed, from the as-sprayed (Fig. 14.20a) to the heat-treated (Fig. 14.20b) coating, the “expected” behavior for a ceramic is noticed, i.e., there is a significant reduction of the overall coating porosity due to sintering effects, which yields higher values of thermal conductivity and elastic modulus.
Fig. 14.18

(a) Thermal conductivity and (b) elastic module values evolution of as-sprayed and heat-treated nanostructured and conventional YSZ coatings at 1,400 °C. Reprinted with kind permission from Springer Science Business Media [91], copyright © ASM International

Fig. 14.19

Cross section of an (a) as-sprayed and (b) heat-treated nanostructured YPSZ coating at 1,400 °C for 20 h. Reprinted with kind permission from Springer Science Business Media [91], copyright © ASM International

Fig. 14.20

Cross section of an (a) as-sprayed and (b) heat-treated conventional YPSZ coating at 1,400 °C for 20 h. Reprinted with kind permission from Springer Science Business Media [91], copyright © ASM International

More information about the mechanism of differential sintering on TBCs can be found elsewhere [45, 91]. In addition to sintering resistance, Liang and Ding [83], Wang et al. [92], and Liu et al. [93] have reported superior thermal shock resistance levels of TBCs produced from nanostructured agglomerated YSZ powders compared to those of TBCs deposited using conventional YSZ powders. Zhou et al. [94] studied the thermal cycling oxidation of nanostructured and conventional YSZ TBCs and reported a superior performance of the nanostructured ones. Therefore, the APS YSZ coatings engineered from nanostructured agglomerated powders have the potential to become the next generation of TBCs. As summarized by Lima and Marple [15] nanostructured TBCs have the following advantages: lower thermal diffusivity, higher (2–4 times) thermal shock resistance, higher creep rate, and healing of the nano-porosity.

As previously discussed, nanostructured YPSZ coatings have demonstrated superior sintering resistance when compared to those of conventional ones [45, 91]. These characteristics are very important because a high increase of elastic modulus and thermal diffusivity/conductivity values caused by sintering hinders the use of these coatings as TBCs. The superior thermal cycle and shock resistances of the nanostructured YSZ coatings [83, 92, 93, 94] show that they may become an important alternative as TBCs in gas turbines for the next decade.

14.3.2.4 Biomedical Applications

For biomedical applications, mainly two materials are used: hydroxyapatite and titania [95, 96, 97, 98, 99, 100]. The calcium phosphate hydroxyapatite (HA) (Ca5(PO4)3(OH)) is the standard thermal spray ceramic coating applied on implants for load bearing applications, such as acetabular cups and hip joints [101]. In spite of this successful application, there are concerns regarding the long-term prognostics of these coatings, which may be hindered by coating dissolution and low mechanical performance [102, 103]. Therefore, it is possible to find nano-sized zones at the coating/substrate interface, in the internal structure of the coating, as well as on its surface. It has to be pointed out that thermal spray coatings produced from conventional powders do not exhibit a significant presence of nano-sized zones on their surfaces. It has been shown that HVOF-sprayed nano-sized TiO2–HA composite coatings exhibit bond strength levels of at least 2.5 times that of thermally sprayed HA coatings. In addition, these coatings exhibited bio-performance levels equivalent or superior to those of HA coatings, which are the current state-of-the-art material. It was hypothesized that one of the reasons for this enhanced behavior was related to the presence of nano-sized zones on the coating surface [104, 105]. Again the comments in the review of Lima and Marple [5, 15] summarize well the results:
  • Osteoblast cell culture (in vitro) shows that the HA coatings with a high content of both crystalline HA and nanostructures are preferred for cell proliferation.

  • HVOF-sprayed nanometer-sized TiO2 coatings have bond strength values at least 2.4 times higher than those of HA coatings on Ti–6Al–4V substrates. In vivo tests (with rabbits) showed that these coatings (deposited on Ti–6Al–4V substrates) have a higher degree of bone apposition when compared to those of uncoated Ti–6Al–4V substrate.

14.3.2.5 Other Applications

The nanostructured titania coatings provided superior resistance against abrasive and erosive wear for ball valves destined for high-pressure acid-leach (HPAL) service [106]. The thermal shock behavior of nanostructured Al2O3/13 wt% TiO2 plasma-sprayed coatings was much higher than that of conventional coatings [107].

14.4 Attrition or Ball Milled Cermets or Alloy Particles Sprayed with Hot Gases

Another possibility to achieve coatings with nano-sized particles is to spray nano-crystalline powders and melt them only partially.
  1. (a)

    During thermal spraying of ceramics, even with adapted spray conditions (discussed below), small particles exceed the melting temperature of the alloy or the binder phase for cermets. In contrast larger particles have the melting temperature exceeded only near the surface, whereas particle cores can remain solid. In volumes that remain solid, micrometer-structural changes are negligible (slow kinetics of solid state diffusion) due to the very short exposure to high temperatures during the spray process [108]. The final crystallite size of the melted alloy or binder phase is determined by the cooling rate during solidification, high cooling rates resulting in much finer crystallites. With cermets, the behavior of the hard phase will strongly depend on the grain size and spatial distributions. In volumes where the binder phase is melted, obvious changes in the microstructure of both the nanostructured and the micro-structured cermet material due to the spraying process will be observed. In contrast to the crystallite size of the binder phase, the diameters and distribution of the hard phases depend on their initial size and solubility in the solid and liquid state of the binder phase.

     
  1. (b)

    To spray nanostructured alloys or cermets, two techniques (HVOF or vacuum plasma spraying) are mainly used to achieve temperatures slightly above the melting temperature of the alloy or cermet binder:

     

With HVOF (see Sect.  5.3.1.4), depending on the selection of fuel gas and proportion of the fuel gas to oxygen, the sprayed powder particles experience much lower temperatures (gas temperatures below 3,000 °C) compared to plasma spraying. Moreover the spray velocity in HVOF spraying (700–1, 400 m/s) is much higher than in plasma spraying, which causes a higher degree of particle impact in the HVOF process. This heavy impact may lead to fragmentation of resolidifying particles and thus formation of large number of nucleation sites, which may cause a higher extent of grain refinement in the deposit structure.

With vacuum plasma spraying (see Sect.  7.9) the plasma jet temperature, except in the few first millimeters of the jet, is relatively low (below 5,000 K) and the Knudsen effect (see Sect.  4.3.2) reduces the heat transfer. Moreover, particle velocities are higher than those obtained with conventional torches used in atmospheric spraying. Of course to limit the particle heating, atmospheric argon plasma spraying could be used, but oxidation would be unavoidable. Probably the most important advantage of vacuum plasma spraying is the low oxygen pressure reducing drastically the particle and coating oxidation phenomena.
  1. (c)

    The last important points are the particle oxidation in flight and the coating oxidation during its deposition. Since the agglomerate size of the nano-crystalline particles is higher than that of the conventional powders, it can be expected that the chemical reactivity of nano-crystalline powders will increase due to the increased surface area. It may result in an increased proportion of oxide phases throughout the grains. For cermets containing materials highly reactive to oxygen, such as WC, some decomposition could occur. It is thus important to control the composition of combustion gases used in HVOF. Vacuum plasma spraying will result in less oxidation than HVOF. This is particularly important for the inter-pass oxidation, which is a consequence of the high temperature that the coating experienced during the spray process. During the time between passes, the coating surface is still at a high temperature and, in the presence of air starts to oxidize (as in HVOF or argon plasma spraying in air).

     

14.4.1 Alloys

Inert gas atomized ASTM F75 Co-based super alloy powders (Cr–28.55 wt%, Mo–5.958, Si–0.53 %, C–0.30 %. Fe–0.27 % Ni–0.13 %, Mn–0.09 %, balance Co, Starmet Corp.) have been mechanically milled in methanol environment for 10 h to produce nano-crystalline powders with an average grain size of 12 nm, determined by TEM dark field imaging [109]. An argon plasma jet (Sulzer Metco 7 M) sprayed these particles onto Ti substrates, and the resultant coating remained nano-crystalline with an average grain size of 21 nm. Furthermore, the micro-hardness and porosity of the nano-crystalline Co–Cr coating are found to be higher than those from the conventional Co–Cr coating, indicating that nano-crystalline coatings may be potential coating candidates for implant applications. It was speculated that the marginal increase in micro-hardness of the nano-crystalline Co–Cr coatings might be due to the presence of an increased proportion of oxide phases throughout the grains.

Cryomilled Ni powders were sprayed with Sulzer Metco Diamond Jet DJ 2600 [110]. The coating was composed of nano-crystalline grains with an average size of 92.5 ± 41.6 nm (the nanostructure was retained) and with extremely fine NiO particles of 5 nm diameter distributed homogeneously inside the grains. Adjusting the spraying conditions, FeNb microcrystalline powders allowed obtaining amorphous coating, while for the FeSi alloy powder, amorphization was obtained with the addition of boron [111]. Ball milled Fe-based powder with additions of Si, B, Nb, and Cu elements were HVOF sprayed (Sulzer Metco CDS with a mixture CH4/O2 (%) = 0.4) [112]. The powders sprayed were Fe93.5–Si6.5, Fe86.5–Si13.5, Fe75–Si6.5–B18.5, Fe75–Si15–B10, and Fe73.5–Si13.5–B9–Nb3–Cu1 (atomic percentages). Amorphous and nanostructured phases were present in the coatings. Coatings produced had a soft ferromagnetic character, the additions of nonmagnetic elements such as boron, niobium, and copper incorporated in the milled powders having little effect on these properties. However the nanostructure phase had a significant effect on modifying the magnetic properties of deposits.

Super alloys were also sprayed with an HVOF processes:
  • Inconel 718 alloy powder, methanol, or cryo-milled producing nano-crystalline coatings exhibiting thermal stability against grain growth up to 1,273 K [113].

  • Oxidation of nano-grain CoNiCrAlY coatings made from cryo-milled powder was studied at 1,000 °C [114, 115].

  • Nano-crystalline NiCrAlY was produced [116] by cryo-milling and sprayed with the Diamond Jet DJ 2700 (Sulzer Metco). The final microstructure of the nanostructured coating had a fine grain structure characterized by a “multimodal structure,” in which both nano-crystalline regions (approximately 75 nm) and submicron grain regions (from 120 to 550 nm) were observed. The oxidation behavior of nanostructured NiCrAlY coatings was investigated after heat treatments in air, at 1,000 °C, for various times. Oxidation led to the formation of a continuous alumina layer, without the presence of other mixed oxides [116].

14.4.2 Cermets

One of the problems when spraying carbide cermets is the decarburization. Moreover, even under HVOF spraying conditions, the extent of decarburization observed in the nano-sized coating material is larger than that in the micro-coating material, because of the high specific surface area of the particles. In particular, heterogeneous melting and localized superheating of the nano-powder, with high surface area is considered to be the controlling factor in decarburization.

14.4.2.1 WC–Co

He and Schoenung [117] have presented an excellent review of the problems involved with spraying WC–Co. They showed that it was possible to produce nanostructured WC–Co powders without non-WC/Co phases although a high percentage of non-WC/Co phases is frequently reported in conventional WC–Co powders. Nanostructured WC–Co powders experienced higher particle temperatures during spraying than those of the corresponding conventional powder. Therefore, if the same spraying parameters were used, WC particles in nanostructured powders suffered from severe decomposition, degrading the performance of corresponding WC–Co coatings. By controlling agglomerate size of the feedstock powder, fuel chemistry, and fuel–oxygen ratio, near nanostructured WC–Co coatings with a low amount of non-WC/Co phases were successfully deposited. These coatings had increased hardness, toughness, and wear resistance. Qiao et al. [118] showed that when HVOF spraying nanostructured powders having the shape of hollow spheres, particles rapidly reached high temperatures in the various deposition processes and were subject to extensive decarburization. HVOF-sprayed nanostructured WC–12 wt% Co were compared to Cr3C2–NiCr coatings [119]. Results showed that the WC–Co coatings had a nanostructure consisting of nano-sized WC carbide particles in an amorphous matrix phase, while in the nanostructured Cr3C2–NiCr a few elongated amorphous phases discontinuously distributed in the coating were observed.

Much work has been devoted to nanostructured WC–Co spraying mainly by HVOF [120, 121, 122, 123, 124, 125, 126, 127, 128, 129] and much less by d.c. plasma spraying [130]. In nanostructured WC–Co powders manufactured recently, the amount of the non-WC/Co phases was greatly reduced. A grain-growth inhibitor has a strong influence on the microstructure and other properties of the coatings. Preventing WC dissolution in the binder, not only maintains very small grains but also maintains their original shape; it is effective in retarding decarburization but it decreases the cohesion between WC and the binder [125]. Superior sliding wear resistance is obtained from coatings deposited with near nano-powders containing an antigrowth additive [125].

14.4.2.2 Cr3C2–NiCr

Compared to WC–Co system coatings the main shortcomings of Cr3C2–NiCr coatings are lower hardness and subsequent low wear resistance, but they are frequently used as protective coatings for application in corrosive environments at elevated temperatures.

Nanostructured Cr3C2–25(Ni20Cr) coatings are synthesized using mechanical milling and sprayed by HVOF or HVAF. Their decomposition seems to be less than that of WC–Co [131, 132, 133]. Better results are obtained with coatings sprayed by HVAF [132]. After 30 days at 900 °C, the carbide morphology of both HVOF and HVAF coatings was comparable, tending toward an expansive structure of coalesced carbide grains [132]. Nano-crystalline coatings exhibit a 20 % increase in hardness, a 40 % decrease in surface roughness, and comparable fracture toughness and elastic modulus compared to conventional coatings [133].

14.4.2.3 Other Cermets

A typical example is the Y2O3-reinforced milled FeAl sprayed by a Plasma-Technik CDS 100 HVOF torch. The coating displayed the typical dual aspect consisting of fully melted and well-flattened splats together with retained unmelted powder particles, which both contained nano-sized grains: columnar nano-sized grains formed by rapid solidification in the fully melted splats and equiaxed nano-sized grains retained in a deformed matrix at the core of the unmelted powder particles. Oxidation and Al evaporation during thermal spraying led to a modification of the chemistry of the melted zones. The associated Al depletion resulted in the formation of the Fe3Al phase and more complex structures [134].

When cryo-milling Ni powders [135], the presence of ultrafine AlN particles drastically decreased the dimension of the formed agglomerates and increased their surface roughness. The AlN phase was broken down into ultrafine particles of approximately 30 nm in size. These particles were dispersed in the Ni matrix and enhanced the development of a nano-crystalline structure in the Ni matrix during cryo-milling. A Sulzer Metco DJ 2600 HVOF gun was used to spray these particles. AlN nanometer-sized grains were present in the coatings, and they also led to an increase in the amount of the NiO phase, distributed in the coating in the form of ultrafine, round particles. Indentation fracture indicated that the fine, dispersed AlN particles raised the apparent toughness of the Ni coating. The increase in micro-hardness resulted from both grain refinement and the presence of ultrafine particles.

TiC–Ni-based composites [108] and Al2O3 and Al2O3–Ni [72] resulted in high quality coatings with optimized HVOF spray parameters. Introduction of nickel alloying decreased hardness and wear resistance of the coatings, but increased their toughness.

Using plasma spraying, a coating of Al2O3 dispersed in a FeCu or FeCuAl matrix [136] appeared to offer better wear resistance under sliding and abrasion tests than nanostructured Al2O3 coatings, and nanostructured YSZ/NiO anodes provided larger triple phase boundaries for hydrogen oxidation reactions in SOFCs [137].

14.5 Spraying Hypereutectic Alloys with Hot Gases

When melting alloys that exceed the solubility limit, upon solidification a single solid solution is produced. However when alloys continue to cool down, a solid-state reaction occurs, permitting a second solid phase to precipitate from the original one and keeping a nanostructure in the coating, if the cooling is fast enough. This effect has been observed both with plasma and HVOF spraying.

Free-standing ring-shaped Al–21Si nanometer composite structures were fabricated by HVOF spraying of gas atomized hypereutectic Al–21 wt% Si alloy powders (15–45 mm) [138]. Figure 14.21 presents a cross-sectional micrograph of the coating structure showing a homogenous and highly dense splat structure, consisting of ultrafine primary (2–4 μm) Si grains and a network of Si in the eutectic Al–Si matrix. Formation of a thin oxide layer of uneven thickness can also be observed on the splat surfaces. The module of elasticity (138 ±17 GPa) shows an intermediate value between those of pure Al (71.9 GPa) and Si (162.9 GPa). The Vickers microhardness value (441 ± 34 N) was also better than that of a conventionally cast eutectic.
Fig. 14.21

Optical micrograph of the cross section of a HVOF sprayed Al–Si composite structure showing ultrafine primary (2–4 μm) Si and a network of Si in the eutectic Al–Si matrix. Reprinted with kind permission from Springer Science Business Media [138]

A freestanding bulk nano-crystalline structure was vacuum plasma sprayed starting from a hypereutectic Al–21 wt% Si micron size (15–45 μm) gas-atomized powder [139]. The cross sections of the coating exhibited a dense splat structure containing fine eutectic Al–Si grains with homogenously distributed ultrafine primary Si (about 1 μm) particles throughout the eutectic Al–Si matrix (Fig. 14.22). The ultrafine primary Si particles precipitated from the Al–Si alloy due to the rapid solidification in VPS process. Further decrease in size of Si particle was due to fragmentation attributed to high velocity impact of the molten powder.
Fig. 14.22

Cross-sectional optical micrograph of the VPS formed Al–Si deposit showing the presence of eutectic Al–Si grains, ultrafine primary silicon particles and a network of silicon particles. Reprinted with kind permission from Elsevier [139]

Gas atomized hypereutectic Al–21 wt% Si alloy powder was blended and mixed with multiwalled carbon nano-tubes (10 wt%) before being ball milled. The blended powder was plasma sprayed (SG 100 Praxair gun) [140, 141]. The multiwalled carbon nano-tubes (MWCNT) were successfully retained in the spray formed composite structure (Fig. 14.23). The formation of a β-SiC layer, rather than aluminum carbides at the interface of Al–Si matrix and MWCNT reinforcement, has been confirmed. The β-SiC formation occurred in an ultrathin layer (2–5 nm), attributed to the higher rate of reaction at the triple point of MWCNT/Al–Si alloy/vapor [141].
Fig. 14.23

Low magnification SEM image of fractured surface of a composite cone structure, showing the retention of carbon nanotubes. Reprinted with kind permission from Elsevier [140]

14.6 Production of Nanostructured Coatings by Cold Spray

Compared to the other spray processes with hot gases, the only way to achieve nanostructured material by cold spraying is to spray nano-powders or micro-size powders with nano-sized grains or to spray composite agglomerates made either of micron- and nanoparticles or of only nanoparticles. However it must be kept in mind that the sprayed particles must be sufficiently ductile.

14.6.1 Alloys

Cold spray deposition of conventional and nanocrystalline (atomized and cryomilled) (Al–Cu–Mg–Fe–Ni–Sc) coatings was successfully achieved [142]. The conventional cold-sprayed coating showed negligible porosity and an excellent interface with the substrate material. This was not the case for the nano-crystalline coating, in which the porosity level was in the range of 5–10 %. The microstructure of the feedstock powder was retained after the cold spray process. The difference in porosity between the conventional and nano-crystalline coatings can be explained by the hardness and microstructure of the corresponding feedstock powder, the extent of deformation of which is much less, resulting in a less dense coating. The hardness of the nano-crystalline coating was similar to that of the feedstock powder, suggesting that no work hardening took place [142].

Nano-crystalline Al 5083 coatings were cold sprayed with helium without preheating of the carrier gas [17] onto a grit-blasted aluminum substrate. The critical velocity (over 700 m/s) was successfully achieved. The cold-sprayed coating showed negligible porosity and the interface with the substrate material was excellent. Wang et al. [143] produced by ball milling a metastable Fe(Al) alloy powder with morphology and size distribution suitable for cold spraying. The prepared powder exhibited a lamellar microstructure. The metastable microstructure of the milled Fe(Al) feedstock was completely retained in the coating using cold spraying. The FeAl intermetallic phase was formed during the heat treatment of the as-sprayed coatings at a temperature of 500 °C [143]. Nickel powder was mechanically milled in liquid nitrogen to achieve small nonspherical powder particles with a mean size of 15 μm with an average nano-crystalline grain size in the range of 20–30 nm [144]. The cryo-milled powder was sprayed with He (also used as carrier gas) onto grit-blasted aluminum substrates using the cold spray process. The Ni coating showed negligible porosity, and the grain structure of the powder was retained. Microhardness values were comparable to the results obtained by electrodeposition. Nanostructured NiCrAlY starting powders were produced using a high-energy milling process and subsequently a coating was deposited on a IN738 substrate by cold spraying maintaining the feedstock nanostructure [145]. The shot-peened coating presented a rather good resistance to oxidation.

14.6.2 Composites

The difficulty wih cold spraying composites lies in the necessity to spray ductile particles, which is not necessarily the case for WC–Co particles. The WC–12Co powder sprayed by Li et al. [146] exhibited a spherical morphology with nano-sized WC particles partially bonded together and voids appearing within particles Fig. 14.24. According to the XRD pattern the powder consisted of two phases of WC and Co. Critical velocities of about 915 m/s were measured. Coating microhardness was about 1,800 HV0.3N, a value similar to that of sintered bulk material of a similar composition and microstructure. A certain degree of deformation of both impacting particle and deposited layer is required. This is illustrated by Li et al. [146] who cold sprayed with He a powder made from agglomeration of nano-sized WC with cobalt followed by partial sintering. The powder particles had a size range from 5 to 44 μm. The nominal WC grain size was 50–500 nm. As shown in Fig. 14.25 small particles (about 10 μm) are almost completely embedded, into the stainless steel substrate, while large particles penetration is much less. With nanostructured WC–Co, a porous structure, originating from the agglomeration, permits the pseudo-deformation through compaction of particles and the buildup of a thick coating Therefore, it can be considered that a WC–Co powder with WC loosely bonded by the binder is suitable for cold spraying. The annealing treatment at a temperature of 1,000 °C had little influence on the microhardness of the coating. However it seemed that the bonding between deposited WC–Co particles and the toughness of the resulting coatings can be improved by post-annealing treatment [147].
Fig. 14.24

Morphology of nanostructured WC–12Co feedstock (a) with detailed surface structure and (b) at a high magnification. Reprinted with kind permission from Springer Science Business Media [146], copyright © ASM International

Fig. 14.25

WC–Co particles penetrated into the stainless steel substrate. (a) Two particles of different diameters penetrating to different depths; the large particle penetrated less deeply than the small particle and (b) a smaller particle of about 10 μm diameter has almost completely embedded into the substrate. Reprinted with kind permission from Springer Science Business Media [146], copyright © ASM International

The production of cermet coatings with conventional and nano-crystalline feedstock powders by cold gas dynamic spraying (CGDS) and pulsed gas dynamic spraying (PGDS) processes was performed by Yandouzi et al. [147]. No degradation of the phase (phase transformation and/or decarburization of WC) was observed. Dense coatings without major defects were difficult to obtain when using the CGDS process. Coatings with low porosity were obtained when the feedstock powder was preheated above 573 K. The nanostructure of the feedstock powder was preserved.

14.6.3 Amorphous Alloys

Cold-sprayed Fe-base amorphous alloy coatings were successfully produced. The coatings showed a negligible porosity and excellent interfaces with the substrate material [148]. The microstructure of the feedstock powder (gas atomized) was retained after the spray process. The use of an amorphous Al–Co–Ce powder to produce nanostructured coatings with high degree of bonding to aluminum substrates alloys was also demonstrated [149].

14.7 Solutions or Suspensions Spraying

To produce finely structured coatings by thermal spray techniques using liquid injection, two routes have been suggested:
  • Spraying submicron-sized or nano-sized ceramic or cermet particles via a suspension [150, 151, 152]. Once drops have been fragmented and vaporized by the plasma flow or the HVOF flame (see Sect.  4.5), particles contained in the droplets are heated and sprayed onto the substrate. Splats collected on cold substrates have diameters ranging between 0.1 and 2 μm and average thicknesses between 20 and 300 nm. The stacking of splats or grain shaped particles (resulting from flattened liquid droplets recalescence) form finely structured coatings.

  • Spray solutions of final material precursors. As with suspension, the liquid undergoes rapid fragmentation and evaporation once injected in the plasma jet. This is followed by precipitation or gelation, pyrolysis, and melting to result finally in the impact of molten liquid droplets with average diameters ranging from 0.1 to a few micrometers [153, 154, 155, 156].

These new spray techniques differ from the conventional one (solid particles in the tens of micrometers size range) both in measurements and modeling.

(a) Measurements. In contrast to conventional particles in-flight, liquid drops and droplets emit no radiation and thus the only way to see them is to illuminate them with a laser pulse, as described in Sect.  16.5.3. This system also allows accounting for effects of voltage fluctuations, the image being triggered when the voltage reaches a given threshold value [152]. The liquid penetration can be characterized by taking, under the same conditions, several images (during few seconds or more) that are superposed when the plasma high luminosity, masking the low one of the liquid, has been eliminated. A much more sophisticated technique of shadography with a double-pulsed Nd:YAG laser (532 nm wavelength with 8 ns pulse duration) can be used [157, 158, 159] and is described in Sect.  16.5.3.2. It is then possible to discriminate droplets with diameters of about 5 μm at the best and measure their number, velocity and diameter distributions in a relatively small volume (2.25 × 1.8 × 1 mm3). To measure velocities of small droplets (<5 μm) only the particle image velocimetry (PIV) can give accurate results [160], and this method is described in Sect.  16.5.3.1. The plasma jet cooling by the liquid vaporization can be measured by emission spectroscopy [151], but plasma computer tomography (PCT) is mandatory to obtain real temperature distributions [161], as described in section “(b) Emission Spectroscopy of Nonsymmetrical Jets”. Of course, (50–60 mm) downstream of the torch nozzle exit, enthalpy probes can be used [162], also described in Sect.  16.3.3.3, which is a much faster technique than PCT. Soysal and Ansar [163] have proposed to use infrared images emitted by hot liquid vapors to characterize the influence of combustible liquid injection into the atmospheric plasma jet, see Sect.  16.5.3.2. Of course, as in conventional spraying, emission spectroscopy can be used to detect evaporation of micro-particles [162].

(b) Models. Models used for liquid–plasma jet interaction are very complex because they must take into account a two-phase flow, drop and droplet fragmentation, evaporation as discussed in Sect.  4.5.2. In addition, the thermal history and chemistry and eventually coalescence have to be considered, together with the time variation of the plasma flow fields resulting from the anodic arc root movement on the anode wall, the jet turbulence, and also compressibility effects [164, 165, 166, 167, 168]. For fragmentation, two types of models are used. The breakup models, such as TAB or ETAB are used, and Basu and Cetegen [168] were the only ones to integrate the droplet vaporization and the internal precipitation into the calculations. The second route attempts to analyze the injection of a continuous liquid jet into the plasma with regard to its primary and then secondary breakup [165, 166]. This route implies the development of novel numerical multiscale approaches. If the Ohnesorge number [see Eq. (4.69)] Oh < 0.1, breakup modes depend only on the Weber number, We, [see Eq. (4.68)]. According to Eq. (4.68), We is proportional to the drop diameter and inversely proportional to the liquid surface tension, σ l. With drops over 200 μm and ethanol (low σ l) as solvent in an Ar–H2 plasma [165, 166], We values over 350 are easily reached within the plasma jet core, except maybe in the jet fringes, and drops can experience different breakup regimes, the catastrophic one inducing a very rapid breakup into tiny droplets, thus limiting their acceleration. In contrast, when injected drop sizes are below 50–60 μm, and the plasma flow mean velocity is lower, when using larger anode nozzle internal diameter, the We reaches a few tens in the central part of the plasma jet. Of course with big drops, increasing the surface tension of the liquid, σ l, can reduce the Weber number.

14.7.1 Sub-Micrometer and Nanometer-Sized Particles in Plasma or HVOF Jets

14.7.1.1 Particle Inertia

Compared to micro-sized solid particles (generally a few tens of micrometers), sub-micro- and nano-sized ones have a very low mass. Such particles are found either in suspensions or are formed in solutions after pyrolysis, sintering, and melting. Shifting from 10 to 0.1 μm particle corresponds to a mass reduction in a ratio of 10−6! To achieve the same injection force, the acceleration should be increased by a factor of 106, which is impossible with a gas carrier without perturbing drastically the plasma jet. That is why a liquid carrier, about 1,000 times heavier than the gas, is used in suspension spraying.

Moreover the drag coefficient, C D, controlling the particle acceleration by the plasma flow, is drastically reduced by the Knudsen effect (see Sects.  4.2.2.2b and  4.3.3.2, Table  4.3), which is not the case with a HVOF flame. Thus small particles are poorly accelerated by the plasma flow and the characteristic time for the momentum transfer is multiplied by a factor of more than 200 with a 0.1 μm particle compared to a 10 μm one. Of course the problem is even worse when shifting from 0.1 to 0.01 μm. This is illustrated in Fig. 14.26 from Delbos et al. [169] who have calculated the velocity of zirconia particles “deposited” with a zero velocity at the jet axis at the nozzle exit (no penetration problem) where the velocity of the Ar–H2 plasma was 2,500 m/s. The nano-particle with a diameter of 0.1 μm, with its very low inertia, is first very rapidly accelerated and reaches a maximum velocity of 480 m/s after 2 mm trajectory, while the gas velocity is still more than 2,300 m/s. The Knudsen effect takes over even more rapidly (the Drag coefficient is multiplied by 0.046) and farther than 2 mm trajectory the particle velocity decreases regularly. On the contrary, with the 5 μm (Drag coefficient multiplied by 0.26) the particle acceleration is regular and its maximum velocity is reached at about 40 mm downstream of the nozzle exit. With the 1 μm particle (C D multiplied by 0.13) the maximum velocity is reached at 15 mm. This is illustrated in Fig. 14.26. Such values are over estimated because they were calculated for particles “deposited” on the axis at the torch nozze exit, while in spray conditions particles are radially injected and cross-areas where gas velocity is much lower than that along the jet axis. That is why in spray conditions with torches using stick type cathodes the solid particle velocity is that achieved by its carrier droplet just prior to its vaporization. At last, with its low inertia, the nano-particle velocity decreases almost as fast as that of the gas and it is the main reason to use very short spray distances (30–50 mm stand-off distance).
Fig. 14.26

Computed evolutions of particle velocities as a function of the covered distance in a Ar–H2 plasma jet with a 6mm nozzle diameter and a net power of 32 kW supplied to the gas. Reprinted with kind permission from Springer Science Business Media [169]

This problem is of course drastically reduced when longer plasmas are used, such as those produced with the Axial III torch. In this torch three plasma jets, produced by separated cathodes and anodes, are gathered together, constricted in an interchangeable extension nozzle, and the resulting plasma jet flows out the extension and expands in the surrounding atmosphere. Suspensions are injected between the three converging plasma jets, axially before the nozzle extension and accelerated by the long plasma flow within the extension. As pointed out in Sect.  7.8.1, with an extension 50 mm long after the nozzle of a stick type cathode torch the gas velocity is still 1,700 m/s, while at that distance without the extension nozzle the velocity of the expanded plasma jet is only 500 m/s. In these circumstances the gas velocity inside the interchangeable extension is only slightly reduced (about 10–30 % depending on the extension length) compared to that obtained at the nozzle exit of each converging three torches. In such conditions, sub-micro- or nano-sized particles are well accelerated in spite of the Knudsen effect. Of course when the jet expands in air after the interchangeable nozzle, the gas velocity decreases almost as fast as that exiting torches with stick-type cathodes and the spray distance must also be reduced compared to that used for conventional spraying.

With HVOF spraying, gas temperatures are below 2,800–3,200 K and the Knudsen effect is drastically reduced, most authors neglecting it for particles diameters as low as 1 μm and even, which might be more questionable, for particles 0.1 μm in diameter or even less. Anyhow most particles sprayed by HVSFS are bigger than 0.1 μm and the Knudsen effect calculated for a 0.1 μm particle in hydrogen–oxygen flame at 1 MPa corresponds to a CD coefficient multiplied by 0.6 to be compared to the 0.046 in Ar–H2 plasma. To characterize flame flows, small particles (in the μm size range) are added to the atmospheric flame flow and PIV measurements are performed downstream of the injection where it is assumed that particles have reached the same velocity as that of the flow (see Sect.  16.5.3.1).

14.7.1.2 Particle Trajectory

Thermophoresis force, F th, occurs when small particles are subjected to a temperature gradient (see Sect.  4.2.3.2 and Fig.  4.19). This effect means that small particles with trajectories in the plasma core can escape from it when approaching a high temperature gradient zone, and can be ejected into the jet fringes, as shown schematically in Fig.  4.19. Moreover, small particles can follow the gas flow developing parallel to the substrate surface close to it and thus never impact on it, as shown schematically in Fig. 14.27.
Fig. 14.27

Schematic of a sub-micro- or nanometer particle following the hot gas flow because its Stokes number is <1

This situation occurs when their velocity is below that corresponding to a Stokes number, St, below 1 [13]:
$$ St=\frac{\rho_{\mathrm{p}}{d}_{\mathrm{p}}^2{\nu}_{\mathrm{p}}}{\mu_{\mathrm{g}}{\mathcal{l}}_{\mathrm{BL}}} $$
(14.1)

Subscripts p and g are related to particles and gas, respectively, ρ being the specific mass (kg/m3), d the diameter (m), v the velocity (m/s), μ the molecular viscosity (Pa. s), and l BL the thickness of the flow boundary layer, BL, in front of the substrate (m). l BL, varies as the inverse of the square root of the gas velocity close to the substrate and can be lower than 0.1 mm. For example, with an Ar–H2 plasma, produced with a stick type cathode torch, St = 1 for v p = 300 m/s, l BL = 0.1 mm, and d p = 60 nm. The higher the spray distance will be, the lower the particle velocity and the thicker the BL close to substrate, reducing correspondingly the probability to reach the substrate. The entrainment of the nano- or sub-micrometer particles by the gas flow can either eliminate these particles from the coating generation or create defects. Defects are due to the substrate asperities: a roughness with a Ra of 1 μm (relatively smooth surface for conventional particles) corresponds to peaks up to 8 μm which are huge compared to the impacting particles sizes.

The first type of defects have been described by VanEvery et al. [170] who have plasma sprayed coatings from suspensions of nanometer-sized (85 ± 13 nm) YSZ powders in ethanol. They varied the liquid injection pressure and the pass thickness, the standoff distance being 5 cm, the spray conditions being given in Fig. 14.28 caption. Coatings were sprayed directly on 24-grit alumina-blasted copper substrates. Figure 14.28 presents the top surface of the coatings with a cauliflower aspect resulting from particles defection by asperities close to the substrate.
Fig. 14.28

SEM images showing the top surface of a coating sprayed on 24-grit alumina-blasted copper substrate with a standoff distance of 5 cm, mechanical injection with an injection pressure of 45 MPa for an injector nozzle diameter of 230 μm, a suspension flow of 52 ml/min, a coating pass thickness of 1 μm, with a SG-100 plasma torch using 3083-129 cathode, 3083-175 anode, 3083-113 gas injector, and a 6-mm internal diameter nozzle, operated at a current of 1,000 A with 20 slm Ar and 60 slm He, producing a gun voltage of 50 V. Reprinted with kind permission from Elsevier [170]

This type of defect corresponds to the formation of coarse columnar microstructure due to droplet trajectories close to the substrate with shadow effects. From microstructural observations and predictions of droplet flight paths VanEvery et al. [170] suggested that the change in relative influences of drag and inertial forces with YSZ droplet size can result in three types of spray deposition, schematically presented in Fig. 14.29. In the type 1 of spray deposition (1 SD) the plasma drag forces during substrate impingement dominate the droplet inertia and redirect the droplet velocity from normal to along to the substrate surface. Consequently, droplets impact preferentially on asperities, generating deposits that grow to become columnar structures separated by linear porosity bands. With the type 2 deposition (2 SD type), the droplet velocity remains mostly normal to the substrate surface, but the impinging plasma drag influences the droplet trajectories such that surface asperities block deposition in downstream regions and create porosity bands. In the type 3 deposition (3 SD type), the coating microstructure exhibits no signs of droplets being influenced by drag from the plasma impingement. For more details about the different types, see [170]. This has been mainly observed with suspensions, particles formed from solutions being generally bigger in the sub-micrometer size range.
Fig. 14.29

Schematic illustration of the deposition characteristics occurring on and away from substrate surface asperities during (a) 1SD, (b) 2SD, and (c) 3SD. The spray direction is the same in each figure. The black regions denote the substrate. The light gray regions represent material deposited by droplets having a substrate parallel velocity component directed from left to right, and the dark gray regions represent material deposited by droplets having a substrate parallel velocity component directed from right to left. Reprinted with kind permission from Elsevier [170]

The second type of defects that can be observed corresponds to the formation of stacking defects, called speckles, which are very important if the ratio Ra/d 50 > 2 [171, 172]. Ra is the mean roughness, while d 50 is the mass median diameter of particles contained in the suspension or formed in the solutions. Indeed, large columnar stacking defects develop through the coating when the substrate surface roughness is higher than the feedstock particle average diameter. Brousse et al. [171, 172] have studied the influence of substrate roughness on coating structure and the corresponding leakage rate, results being presented in Fig. 14.30a. Reducing the stacking defects, as presented in Fig. 14.30b, to enhance gas tightness, requires spraying onto smooth polished substrates. For example, with a ratio Ra/d 50 = 2 the leakage rate of the YPSZ suspension coating is only 0.02 MPa L/s m [171]. When Ra/d 50 = 50 the leakage rate becomes 0.5. This columnar stacking defect, presented in Fig. 14.30b, can be explained, as for conventional coatings, by the formation of a speckle (see section 13.6.1.1.b). The formation mechanism of this speckle suggested by Racek [173] is presented in Fig. 14.30c.
Fig. 14.30

YPSZ coating sprayed onto a substrate exhibiting a ratio average substrate roughness Ra to the feedstock particle average diameter d 50 of (a) 2 times or (b) 50 times, where columnar stacking defects develop through the coating thickness [171] (c) Suggested mechanism of the speckle formation [173]. Reprinted with kind permission from Dr. E. Brousse [171]

14.7.1.3 Particle Flattening

When collecting on a flat substrate splats resulting from the flattening and subsequent solidification of conventional molten particles, the flattening degree, ξ, (ratio of splat mean diameter to that of impacting particle) is typically in the range of 4–7. In contrast, for submicron- or nanometer-sized particles of the same material and with similar temperature, ξ < 2 [5, 151, 169]. That is due to the much lower impact kinetic energy of the small particle but also to its much smaller diameter (Young-Laplace equation stating that the pressure difference across the fluid interface is inversely proportional to the principal radii of curvature).

14.7.1.4 Particle Heat Transfer

Similar to the effect on the drag coefficient, the Knudsen effect reduces drastically the heat transfer coefficient (see Sect.  4.3.3.2); however, the heating is rather fast due to the low mass of nano-particles. Another interesting phenomenon is observed when the spray distance is short (<50 mm with a PTF4 torch) with heat fluxes imposed by plasma >30 MW/m2 [169, 171] and with a substrate or a previously deposited layer with a low thermal conductivity which is sufficiently insulating. Under these conditions the heat flux delays the solidification of the flattened nano- or sub-micro-sized particle, and the surface tension results in spheroidizing the flattened particles, resulting in a coating with a granular structure [169], as illustrated in the section 14.7.6.3b. “(i) Plasma Spraying” Fig. 14.67.

To conclude this part, the smaller the sprayed particles are, the higher is the probability that they do not reach the substrate with the right impact parameters to form a coating. Sufficiently high impact velocities are only achieved through the velocities of the particle mother droplets.

14.7.2 Liquid Injection

14.7.2.1 Radial Injection into Plasma and HVOF Jets

(i) When spraying solid particles with a size range in the tens of micrometers (conventional spraying), the optimal trajectory is obtained by adjusting the injection force to that imparted to the penetrating particle by the hot gas jet:
$$ {m}_{50}\cdot {\gamma}_{50}\approx {S}_{50}\cdot \rho \cdot {\nu}^2 $$
(14.2)
where m 50 is the mass median corresponding to the log normal distribution of the mass median diameter d 50, γ 50 the corresponding acceleration, S 50 the cross section of the d 50 particle, ρ the mass density of hot gases, and v the hot gas velocity. Of course this expression is oversimplified and in the real world one must take into account the fact that ρ and v vary continuously along the particle trajectory. Moreover, even if vaporization occurs, it happens only when particles have already penetrated into the hot gas jet core, and the trajectory is that determined by the initial particle size. The problem is far more complex with a liquid that is fragmented upon penetration in the high velocity hot gases and then vaporized (the corresponding time t v being 2 to 3 orders of magnitude longer than that of the fragmentation time tf [151]). Thus the mass and correspondingly the cross section of the drops decrease continuously upon their penetration into the hot gases. In order to achieve the drop penetration into the hot gas jet core, the initial velocity must be much higher compared to that which should be given to the initial drop if its diameter was kept constant. Thus the condition for a good radial penetration implies, as explained in Sect.  4.5.2, that
$$ {\rho}_l{v}_1^2>>\rho {v}^2 $$
(14.3)
where the index l corresponds to the liquid while no index is related to the hot gas jet.
The fragmentation and vaporization processes of the liquid jets or drops are described in detail in Sects.  4.5.3 and  4.5.4. Drops or jets will be fragmented first in tiny droplets (few micrometers) before being vaporized. Figure 14.31 illustrates the liquid fragmentation upon its penetration within an Ar–He plasma jet (here a suspension in ethanol of YSZ nanometer-sized particles, 20–300 nm, with a load of 10 wt%). As can be seen, fragmentation starts in the plasma jet core fringes: as soon as the Weber number of the suspension [see Eq. (4.68)] is higher than 12–14, droplets can be fragmented in the jet fringes. Then these droplets traveling in the jet fringes are evaporated, but unfortunately in this zone particles contained in suspension droplets are not melted, but heated enough to stick on the previously deposited layers of the coating, thus creating defects. In case of solutions they are poorly pyrolyzed and also imbedded in the coating being formed as a mud-like film. Fragmentation keeps going with further penetration into the hot plasma core, creating first large drops (the image resolution is about 30 × 30 μm2) and smaller and smaller ones appearing as clouds, with a color, resulting from laser illumination, which is slightly different from that of the plasma jet image. To Illustrate Eq. (14.3), for the conditions of Fig. 14.31, ρ l v l 2 = 0.6 MPa while ρv 2 = 0.015 MPa!
Fig. 14.31

Suspension fragmentation in an Ar–He stable plasma jet (Ar: 30 slm, He: 30 slm, I: 700 A, anode nozzle internal diameter at exit: 6 mm, liquid injection velocity: 26 m/s, average voltage V m = 40 V, relative voltage variation ΔV/V m = 0.3). Reprinted with kind permission from Springer Science Business Media [151], copyright © ASM International

Thus, the liquid should be injected with a device allowing controlled drop dispersion and velocities that are easier to describe than to achieve (see Sect.  4.5.1). According to the complexity of liquid penetration, this process must be controlled on line with appropriate diagnostics (see Sect. 16.7.3).

(ii) For fragmentation of drops and droplets, depending on their Weber number, different breakup mechanisms are considered and classified as: (1) bag breakup (12 < We <100), deformation of the drop as a bag-like structure that is stretched and swept off in the flow direction, (2) stripping breakup (100 < We < 350): thin sheets are drawn from the periphery of the deforming droplets, and (3) catastrophic breakup (We > 350) which is a multistage breaking process [157]. With catastrophic breakup, liquid fragmentation in tiny droplets is very fast, and the acceleration of their solid sub-micro- or nano-sized particles is very poor due to Knudsen effect. When the liquid precursor is injected as liquid jet, it undergoes primary breakup by the plasma flow while blobs and droplets generally undergo secondary breakup in the flow [157].

If the drop diameter depends on the injection setup (see Sect.  4.5), the choice of the solvent to prepare the suspension or solution is also very important. The surface tension of the drop determines the final size of droplets that are finally rapidly vaporized once fragmented (see Fig.  4.75). This final size of droplets before their fast evaporation is very important for the type of coating obtained. With the solutions, if the droplet size is too large (>5–10 μm) the precursor solute concentrates as a shell that pyrolyzes and that can reach the coating being formed incompletely treated. When drops are smaller, pyrolysation covers rapidly the whole volume, get sintered, and then melted forming splats upon impact [20, 155, 174]. Generally a droplet corresponds to one particle well melted or a shell just pyrolyzed or parts of this shell. With suspensions, coating formation depends on the solid particle morphologies, the propensity for forming aggregates being very important for particles obtained by a chemical route, while it is much less with attrition-milled particles. Aggregates either form big particles, as shown by the size of collected splats, or explode resulting in dispersed trajectories and unmelted particles. As with conventional spraying, the size distribution width of particles within suspension must also be limited to reduce the trajectory dispersion. With particles in the sub-micron- or nano-size, splats collected sometimes correspond to bigger particles, due to the coalescence of fully melted, initially smaller particles [5, 151]. This phenomenon is also observed when spraying suspensions of different particles such as zirconia and alumina [175, 176, 177, 178], Co2O3 and Ni2O3 (grain sizes of 30 to 60 nm), or TiO2 and Cr2O3 (median crystallite size of 100 nm) [179].

(iii) Another point that must be emphasized is that, to overcome the Stokes effect [see Sect. 14.7.1.2 and Eq. (14.1)], the substrate must be disposed relatively close to the distance where most droplets have been vaporized. It must correspond to a distance sufficient to melt particles but avoiding they slow down too much. This results in rather short spray distances, for example, between 30 and 50 mm downstream of the nozzle exit when plasma spraying with a stick-type cathode torch. At such distances heat fluxes can reach a few tens of MW/m2 that can modify strongly the heat transferred to the coating under construction and thus its structure.

(iv) The vaporization of the liquid and the heating of the vapor to the temperature of the hot gases, especially the plasma, consumes from 20 to 40 % of the hot gas enthalpy, at least for plasma jets. Section  4.5.5 describes in detail the important cooling of the plasma or HVOF jet by the liquid vaporization. The consequence is that, compared to conventional coatings, torch power levels should be at least 30–40 % higher to compensate for the energy necessary to vaporize the liquid carrier. At last, Sect.  4.5.6 describes the sensitivity of liquid injection to arc root fluctuations for plasma jets (the effect being far more important than that observed when injecting particles in conventional spraying), which requires use of low fluctuating plasma jets when spraying liquids. Unfortunately, the strong fluctuations with H2 in the plasma forming gas require the use of He as secondary gas, which reduces the voltage and thus also the power level. Such fluctuations are also observed when using nitrogen as primary forming gas. For HVOF guns working with C2H2 or liquid fuel the injection of the solution or the suspension should be radially downstream of the nozzle. Using organic solvent does not help, through its combustion, to compensate for the energy lost by its vaporization and transformation into plasma. Indeed combustion can only occur in areas where temperature is below about 3,000 K, and where a sufficient quantity of oxygen has been entrained. See, for example, enthalpy probe measurements presented in Sect.  16.3.1.5. The problem of oxygen entrainement can be partially overcome by using atomization of the suspension by air.

14.7.2.2 Axial Injection

In contrast to radial injection, the injection velocity with axial injection must be sufficient for drop penetration into the high-energy jet, but afterwards the velocity difference between drop and gas flow should assure their fragmentation and subsequent vaporization. However, as in radial injection, the liquid injector must provide a relatively narrow trajectory distribution with controlled drop velocities.

(a) d.c. Plasmas

With a d.c. plasma gun such as the Mettech plasma torch, three plasma jets converge in a nozzle, the liquid drops being injected where the separate jets converge, and they are entrained by the high velocity converging plasma jets which provide a sucking action [5]. The other advantages of this configuration are that fragmentation and vaporization occur in the high velocity and high-temperature plasma jet inside the convergent torch nozzle that can be longer than that of a fee jet flowing in air. The power level can reach up to 2.5–3 times that of the conventional stick-type cathode plasma guns. The length of the common nozzle is a few tens of millimeters, and fragmentation and vaporization occur inside the convergent torch nozzle. In spite of the Knudsen effect, particles are well accelerated in the extension nozzle where the plasma velocity is not drastically reduced (10–30 % depending on the extension length). For example, Oberste-Berghaus et al. [175] for nano- or sub-micro-sized alumina and zirconia mixed particles have obtained ensemble velocities between 500 and 770 m/s.

For solutions, Basu et al. [20] have shown that the quality and property of the coating microstructure depends upon the injection type and droplet sizes. For axial injection of solutions, smaller initial diameter droplets are favorable for getting pyrolyzed. In the axial mode, the droplets are injected right into the high-temperature plasma core. Hence they readily go through the maximum possible heating rate helping the pyrolization process to develop within the whole droplet.

(b) HVOF

Any liquid, being water or organic, starts to evaporate rapidly when introduced into the flame of the combustion chamber causing two important effects: (1) a significant cooling of the flame and (2) strong disturbance of the free expanding hot gas stream due to the evaporation process caused by the expanding vapor [180]. Vaporization and pressure increase can modify the jet temperature and velocity distribution in the expanding nozzle and hence produce different resulting particle morphologies [181]. An HVOF process using micro-sized particles (a process called high velocity suspension flame spraying: HVSFS [180]), according to Domongo et al. [181], comprises the following series of complex processes: (1) transformation of chemical energy into thermal energy of the gas; (2) conversion of thermal energy into kinetic energy of the gas by expansion through the nozzle, including energy transfer from the gas to particles during this expansion process; (3) free jet flow field with flow patterns strongly depending on the pressure difference between nozzle outlet and the atmospheric pressure; and (4) conversion of the particles’ kinetic and thermal energies into viscous deformation work and surface energy during coating deposition. First results were obtained with propane as fuel gas (fuel/oxygen ratio of 0.9) and ethanol as suspension medium. The evolution of the mass fractions of propane, ethanol, oxygen, and the exhaust gases along the torch centerline shows a clear correlation with the temperature of the gases. Ethanol evaporates no earlier than inside the expansion nozzle due to the injection conditions. Also, the increase in pressure in the combustion chamber due to ethanol evaporation leads to delayed evaporation. Ethanol evaporation results in a rapid decrease in flame temperature, which is not compensated by gaseous ethanol combustion. At the same time, oxygen is not completely consumed by the ethanol combustion. The diffusion process (non-premixed combustion) controls reaction kinetics, under the boundary condition of a high gas velocity. This also means that the gas temperature in the nozzle cannot be influenced by the oxygen content in the premixed oxy/fuel mixture. As a consequence, the spray particles, which are heated only after release from the suspension droplets, are heated at lower gas temperatures and with lower dwell time compared to the conventional HVOF process. The only possibility to change the heating characteristics of the spray particles is to move the position of ethanol evaporation inside the combustion chamber, for example, by modifying the combustion chamber (mainly through the increase of its length) [182, 183]. It is also possible to modify the composition of the liquid phase, in particular the ratio between water and organic liquid phase (isopropanol), which is an important parameter for controlling the flame temperature in the HVSFS process. Indeed, water absorbs energy from the flame as it is vaporized and heated, whereas the organic liquid phase (isopropanol) contributes to the heat generation process by burning after vaporization [183, 184].

(c) r.f. Plasmas

With r.f. plasmas, flow velocities being below 100 m/s, fragmentation does not occur. With no fragmentation three cases can be encountered, of course for all of them the solvent is progressively vaporized, then (1) for suspensions made of particles agglomerating easily, such as those prepared by chemical routes, the solid particles sinter and then melt forming material drops in the tens of micro-size range, the molten drop impacting onto the substrate to form coatings as in conventional spraying, (2) for suspensions made of particles hardly agglomerating, each solid particle is melted separately before impacting the substrate forming splats with diameters in the few micrometer range or below, (3) solutions are completely vaporized in the plasma, and then nanostructured particles are synthesized in flight by homogeneous nucleation from the supersaturated precursor vapor, to subsequently build up a coating with high porosity.

1. A good example is the production of hydroxyapatite (HA: Cal0(PO4)6(OH)) coatings or micrometer-sized powders described by Bouyer et al. [185]. HA material in the form of an atomized colloidal water suspension was fed into the center of an inductively coupled r.f. argon plasma discharge. The HA suspension was brought into the plasma discharge core via a gas-atomizing probe fed with a peristaltic pump. As mentioned previously the relative gas-drop velocity is too small (<50 m/s) to fragment the drops (<100 μm) that are only successively flash-dried and melted, then deposited on the substrate to be coated or collected in-flight as spherical particles. The induction plasma allows sufficient time for the droplet drying and melting steps. The first part of the drop trajectories are within a pure Ar plasma with a rather low heat transfer coefficient and it is only downwards that this coefficient increases with the diffusion of the diatomic sheath gas (oxygen) into the torch central part. Figure  11.29a presents the suspension with needles of HA. Figure  11.30 shows the three main steps for HA deposit preparation. Finally Fig.  11.29b presents the mostly spherical HA particles obtained. The authors have shown the feasibility of SPS for HA deposition with high deposition rates (>150 μm/min.) and also for spheroidized powder production. Oxygen as sheath gas helps in limiting the decomposition of HA during plasma treatment. As a general rule, this decomposition can be either avoided or at least minimized by using an appropriate choice of plasma gas, namely, a plasma–sheath gas mixture with moderate enthalpy and high oxidizing potential. For example, hydrogen acts as a harmful gas for the HA stabilization, while oxygen is beneficial. In addition, the presence of water in the suspension contributes to the resistance of the apatite structure to decomposition during the high-temperature treatment by maintaining a high partial pressure of water vapor in the deposition reactor. Varying both the atomization parameters and the suspension properties can control HA SPS powder.

2. When r.f. spraying gadolinia-doped ceria (GDC) with particle mean sizes of 0.6 μm (particles with no agglomerating tendency) in suspension in distilled water, the resulting coating, as shown in Fig. 14.32, consists of layered splats with typical diameters from 0.2 to 2 μm corresponding to a flattening ratio below 2.7 [186]. First the solvent is vaporized and then the solid particles melted.
Fig. 14.32

FESEM micrographs of a r.f. suspension plasma-sprayed GDC coating top surface. Reprinted with kind permission from Springer Science Business Media [186], copyright © ASM International

3. In the r.f. Solution Plasma spraying, called SolPS process, described by Jia and Gitzhofer [186], the GDC solutions of nitrate hexahydrate were completely vaporized in the plasma and then nanostructured GDC particles were synthesized in flight by homogeneous nucleation from the supersaturated precursor vapor, to subsequently build up a coating with high porosity, as illustrated in Fig. 14.33. The formation of globular GDC coatings was observed over the entire experimental range of this study. Similarly Shen et al. [187] have r.f. sprayed solutions of Lanthanum strontium cobalt iron oxide (La x Sr 1x Co y Fe1−y O3−δ ) and GDC to produce a composite nanopowder, which was used in a next step to produce cathode coatings for intermediate temperature solid oxide fuel cells (IT-SOFCs) by suspension plasma spray (SPS), It has been possible to achieve a homogeneously mixed nanometer-sized composite GDC/LSCF powder without using a prolonged period of mechanical mixing. The nanometer-sized powders exhibited a perovskite structure and a fluorite structure as well as separated GDC and LSCF phases. By applying a low feeding rate, it was easy to synthesize nanometer-sized powders with different compositions by only adjusting the metal nitrate concentrations in the precursor solution. All synthesized powders were spherical with a diameter between 10 and 60 nm regardless of their composition. The deposited coating had a homogeneous nano-cauliflower structure, similar to that presented in Fig. 14.33, with an average porosity of 51 %.
Fig. 14.33

SEM micrograph of the surface of SPS cathode coating. Reprinted with kind permission from Springer Science Business Media [187], copyright © ASM International

14.7.3 Spray Torches Used

14.7.3.1 General Remarks

Radial liquid injection results in drop fragmentation before the vaporization of the resulting droplets [13]. In plasma flows, this vaporization followed by the transformation of vapor into plasma, drastically cool down the plasma flows even if the liquid is combustible. The combustion can occur only far downstream of the nozzle exit where the air entrainment is sufficiently important, i.e., at distances longer than the standoff distances with which suspensions or solutions are sprayed. However, when using a gas-atomized liquid, with air as atomization gas, combustion of the combustible solvent (ethanol in the experiment performed with a Triplex I torch by Vassen et al. [188]) can occur rapidly in the plasma jet and increase the gas temperature and velocity, as measured with an enthalpy probe 5 cm downstream of the nozzle exit (see Figs.  16.9 and  16.10).

With HVSFS, as previously indicated, almost no fragmentation occurs in the combustion chamber [183]. The liquid drops are vaporized cooling the flame and increasing significantly the combustion chamber pressure. With HVOF guns, designed for spraying conventional particles (in the tens of micrometer range), organic solvent is even not completely evaporated inside the expansion nozzle. Thus the decrease in flame temperature is not compensated by gaseous organic solvent combustion. The spray particles (mainly suspensions are sprayed) are released and heated only downstream of the nozzle throat.

Another very important point is that plasma spray torches must achieve high flow velocities in order to accelerate sufficiently the mother droplets, the particles contained in the droplets or formed being otherwise poorly accelerated due to the Knudsen effect.

14.7.3.2 d.c. Plasma Torches Used

To spray solutions or suspensions different types of torches are used:
  • The conventional ones with a stick-type cathode (30–50 kW power levels) are working with Ar–H2, Ar–He, or Ar–H2–He plasma-forming gas mixtures. If hydrogen provides excellent heat transfer, it also induces very high arc root fluctuations resulting in important voltage fluctuations having a mean amplitude ΔV relative to the time averaged voltage \( \overline{V} \), in the order of \( \Delta V/\overline{V}<1.5 \). In contrast, Ar–He mixtures result in \( \Delta V/\overline{V}<0.25 \) [18]. The role of these fluctuations can be illustrated with the following example. When using a PTF4 (Plasma Technik) torch with a 6 mm internal diameter nozzle, an arc current of 500 A with 45 slm Ar and 15 slm H2, the mean voltage is 60 V with a minimum value of 40 V and a maximum one of 80 V. Figure 14.34 presents two pictures of the plasma jet taken, respectively, at the minimum and the maximum voltages. The velocity difference between both jets is about 800 m/s, which corresponds to Δ(ρv 2) ~ 320 %! For more details about these fluctuations see Sect.  7.3.5. Such voltage variations modify deeply the liquid penetration [151]. This is illustrated in Fig.  4.79 for the Ar/H2 (45/15 slm) d.c. plasma jet, which plasma jet length fluctuations with voltage are illustrated in Fig. 14.34. On the contrary with the more stable Ar–He plasma presented in Fig.  4.73 the dispersion is drastically reduced. For more details about the drastic dispersion occurring in Ar–H2 plasma jets, see Sect.  4.5.6. With these plasma torches the plasma jet core, corresponding to the high gas velocity, is short (a few tens of mm) and the gas velocity with the jet expansion in the surrounding atmosphere decreases very fast. Thus, with this gas velocity fast decrease and the Knudsen effect small particle velocities are low.
    Fig. 14.34

    Pictures of an Ar–H2 plasma jet (PTF4 plasma torch, 6 mm i.d. anode nozzle, 45 slm Ar, 15 slm H2, 500 A, ‾V = 60 V) taken at the maximum (80 V) and minimum (40 V) voltages. Reprinted with kind permission from Springer Science Business Media [169]

  • The Triplex torch (Sulzer Metco, Wohlen, Switzerland) with three cathodes and three independent electric sources results in three arcs attaching at a single anode, with insulating rings between the cathode and the anode permitting the generation of longer electric arcs compared to conventional rod-type cathode plasma torches. Mean voltages can reach here 100–120 V with Ar–He plasma gas mixture, instead of less than 40 V with conventional torches. Thus, if ΔV is still about 10 V, the ΔV/V ratio is below 0.1. This system has been used for example to spray suspensions [190].

The Axial III torch (Mettech, Vancouver, Canada) gun contains three cathodes and three anodes and is operated by three power supplies. The three plasma jets converge within an interchangeable nozzle, and liquid feedstock is injected axially between the three plasma jets [175, 191]. This configuration minimizes drastically the Knudsen effect problem because longer plasmas are obtained. They comprise the interchangeable water-cooled exit nozzle with a length of a few tens of mm where the gas velocity is not very much reduced, followed by the plasma jet, expanding in the surrounding atmosphere where, as with stick type cathodes gun, the gas velocity decreases rapidly. With this torch, where power levels can reach 150 kW, both the evaporation of the solvent does not reduce too much the gas temperature and the nano-sized particle velocities reach 800 m/s resulting in better flattening on the substrate [191]. Also, particle temperatures at impact are higher (e.g., about 2,200 °C for WC–Co particles, which is not necessarily the optimum for such particles since decarburization occurs!). With this technique, the spray distance is also shorter, due to the low inertia of particles, than when spraying micro-sized particles: about 50–60 mm against 150 mm. Wang et al. [192] have shown that when spraying the YSZ electrolyte of half cells of an SOFC, the coating density increased as plasma torch input power increased, in good correlation with the higher particle velocities and their higher degree of melting.

14.7.3.3 r.f. Torches Used

Torches with power levels between 35 and 50 kW, see Boulos [193] and Chapt. 8, are generally used [186, 187], with or without a supersonic nozzle (diameter = 24.2 mm), in order to increase the plasma velocity.

14.7.3.4 HVOF Torches Used

Torches with axial injection such as DJ2700 (Sulzer Metco, Wohlen, Switzerland) [191, 194] or Top-Gun (GTV, Düsseldorf, Germany) [9, 181, 183, 184] are used in the HVSFS process. When the suspension is made of water, poor coatings are obtained due to insufficient flame enthalpy. Much better results are obtained with ethanol, but even a low percentage of water in ethanol drastically cools the flame. Injecting combustible liquids raises the combustion chamber pressure resulting in instabilities in the acetylene flow (acetylene–oxygen giving the highest combustion temperature). Thus propane and ethane are used [181, 182] to achieve stable flames with the Top-Gun system, while with the DJ2700 one, Oberste-Berghauss et al. have chosen propylene [175]. Spray guns have been modified especially for the injection system and also to obtain optimal combustion with longer combustion chambers. As with plasma torches, spray distances are significantly lower (60–110 mm) than when spraying particles in the tens of micrometers range (150–200 mm). These spray distances, significantly higher than those with conventional plasma torches, are possible; thanks to the much higher gas velocities and lower temperatures decreasing the gas mean free path and thus the Knudsen effect.

14.7.4 Solutions or Suspensions Preparation

Either for solutions or suspensions, the main constituent is the solvent which is either water or an organic liquid, the most frequently used one being ethanol. As it can be expected, fragmentation of the liquid depends mainly upon its surface tension, σ s, and also, but to a lesser extent, on its viscosity, μ s, while its vaporization is linked to its thermal properties. Table 14.2 presents some values for five liquids often used as solvents.
Table 14.2

Main characteristics of solvents used in suspension or solution spraying

Liquid

Boiling temperature (°C)

Viscosity (mPa. s)

Surface tension (mN/m)

Vaporization energy (kJ/mole)

Energy of combustion (MJ/kg)

Water

99.98

1.74

71.97

40.67

 

Ethanol

78

1.0

22.27

38.6

29

Triethanolamine

335

45.95

67.7

34.2

Pentanole

137–139

4.0

25

44.6

35

Isopropanol

82.5

2.37

23

44.0

30.5

Water requires more energy than ethanol but less than pentanol and triethanolamine to get vaporized. The lowest surface tension is that of ethanol. Nevertheless, all, except water, contain carbon, which pollute the coatings. Moreover, for an easier liquid injection, the viscosity of the suspension or the solution must remain as close as possible to that of the solvent. At last it must be pointed out that combustion of ethanol, pentanol, and triethanolamine, reheating the hot gases, can occur only if oxygen is present and gas temperatures are below the combustion temperature. It means that combustion is possible with HVOF, once the liquids are vaporized, and in the plume of plasma jets run in air but at spray distances higher than those used to spray suspensions or solutions, or when the drops are injected with a gas containing oxygen.

14.7.4.1 Solutions

Compared to other spray techniques, solutions with a mixing at the molecular level of the constituent chemicals allow an excellent chemical homogeneity [195]. Several liquid precursors, such as solution/sol/polymeric complexes, have been evaluated for different oxide systems [196]. The success in forming the phase required for a given system depends on the decomposition characteristics of the different precursors. They include (a) mixture of nitrates in water/ethanol solution, (b) mixtures of nitrates and metal-organics in isopropanol (hybrid sol), (c) mixed citrate/nitrate solution (polymeric complex), and (d) co-precipitation followed by peptization (gel dispersion in water/ethanol). It is important to mention that the molecular level mixing of constituents results in chemical homogeneity. It also allows creating metastable coating phases due to rapid cooling of ultrafine splats during deposition [197]. For example, alumina, YSZ, and zirconia were produced by aluminum isopropoxides, zirconium butoxides, zirconium acetate, and yttrium acetate in isopropanol with n-butanol and distilled water as solvents [198, 199, 201]. Other researchers have used aqueous solutions of zirconium, yttrium, and aluminum salts [199, 200]. Aqueous solutions allow higher precursor concentrations than organic ones that are less expensive to produce and are safer to store and handle. According to work related to spray pyrolysis, volumetric precipitation, when heating droplets within the plasma jet, depends on the solute initial concentration and its value relative to the equilibrium saturation. The equilibrium saturation concentration is determined by concentrating the precursor in an evaporator at room temperature until precipitation occurs [200]. This permits then to prepare several solutions with different concentrations. This results in different viscosities, surface tensions, and specific masses: viscosity increases (by a factor of 5) and surface tension decreases (by a factor of 1.23) when increasing the solution concentration in a ratio of 4. It must be moreover emphasized that the solutions behave as the solvent; i.e., as a Newtonian fluid.

The precursor concentration in solutions can be varied up to the equilibrium saturation [202]. In their study of 7YSZ solutions, Chen et al. [200] have considered two different precursor concentrations: one of high molar concentration (2.4 M) that is 4 times higher than that of the low concentration (0.6 M). When the initial 7YSZ precursor is concentrated four times in water, the solution viscosity increases from 1.4 × 10−3 to 7.0 × 10−3 Pa s, and the surface tension decreases from 4.82 × 10−2 to 4.663 × 10−2 N/m. Both precursors pyrolyze below 450 °C and crystallize at ~500 °C [201]. This indicates that the solution precursor concentration has little effect on the precursor pyrolysis and crystallization temperatures. The concentration of precursors produces almost no variation in the solution specific mass and surface tension, but large variations in the solution viscosity. Chen et al. [203] studied also the influence of the properties of the liquid phase. Of course results are quite similar to those observed with suspensions: droplets with a high surface tension and also high boiling point of the liquid phase experience incomplete liquid phase evaporation in the plasma jet, while droplets created from a low surface tension and low-boiling point liquid phase undergo rapid liquid phase evaporation.

As with suspensions, liquid vaporization and then gaseous phase heating in HVSFS or transformation into plasma in plasma spraying, consumes energy. Thus it cools the flame or plasma jet downstream, where higher energy would be necessary to pyrolyze solutions or melt particles contained in suspension droplets. Moreover, with solutions often precursors undergo significant endothermic processes in the early stages of heating [204]. As pointed out by Muoto et al. [204], this endothermic behavior can be significant in solution precursor plasma spraying (SPPS) because the material is traveling in a plasma jet, the temperature of which is dropping rapidly. Endothermic behavior at low temperatures will delay particle heating and pushing the location where the melting point is reached further downstream. It appears, from the work of Muoto et al. [204], that the two characteristics that distinguish the desirable YSZ precursor from others is the propensity for the YSZ precursor to form dense oxide particles and the absence of endothermic events during pyrolysis. Dense particles are better at making dense coatings because they will impinge on the substrate at near normal incidence (Stokes number effect), and therefore do not introduce porosity associated with the arriving particle. These authors have plasma sprayed binary mixtures of yttrium nitrate or yttrium acetate, combined with magnesium nitrate or magnesium acetate (the 4 combinations being tested), with water as the solvent to produce Y2O3–MgO nano-composites. Moreover, they have added ammonium acetate (CH3COONH4) to their solution of nitrates to increase the exothermicity of the decomposition process. Acetates can oxidize in the air that is present during spraying, and the nitrate can act as an oxidizer for the acetate in the acetate–nitrate combinations. Coatings obtained with the ammonium acetate were denser than the four other ones. This relatively high density was attributed to a combination of two factors: (1) the exothermic decomposition characteristics of the precursor and (2) the formation of oxide particles which are not too flakey or fluffy to remain entrained in the core of the plasma jet. However, if the processes are too highly exothermic then the oxide particles may shatter in the plasma, causing them to travel along the colder periphery of the jet; also, the increased heat resulting from excessive exothermic reactions may overheat the substrate [204].

14.7.4.2 Suspensions

The easiest way to produce a suspension is to make a simple slurry with particles and a solvent, particle sizes varying from a few tens of nanometers to micrometers. The most frequently used solvents are alcohols (ethanol or isopropanol) or water or a mixture of both [205, 206]. After stirring, the suspension stability can be tested by sedimentation. Typical values of slurry stability are a few tens of minutes, the stability increasing with the powder load [207]. Slurries with TiO2 and ZrO2 have been prepared that way [205, 206, 207, 208] as well as with Al2O3 and ZrO2–Al2O3 mixtures [209]. For example, with zirconia, a phosphate ester has been used [151] that provides a combination of electrostatic and steric repulsions. The pH adjustment is also an important factor to be taken into consideration. However, nano-sized particles of oxides have the tendency to agglomerate or aggregate, even when stirring the suspension. The stability problem can be overcome by using a suitable dispersant, which adsorbs on the particle surface and allows an effective dispersion of particles by electrostatic, steric, or electro-steric repulsions. The percentage of dispersant must be adjusted in such a way that it displays the minimum viscosity of the suspension with a shear-thinning behavior [151]. This behavior means that when the shear stress imposed by the plasma flow is low, the suspension viscosity is high, and it decreases drastically when the shear stress increases as the drop penetrates more deeply into the plasma flow. For example, with zirconia, a phosphate ester that provides a combination of electrostatic and steric repulsion has been used [152]. With a mixture of WC–Co [191], the problem is more complex due to the different acid/base properties of both components: WC or, more precisely, WO3 at its surface is a Lewis acid, while CoO is basic. Thus, a complex equilibrium between the dispersing agent and the suspension pH must be found; for example, the latter must be adjusted to less basic conditions, but avoiding the cobalt dissolution. Similar problems have been observed with Ni.

At last, it must be pointed out that when the wt% of powder increases the viscosity of the suspension increases too. Different products can be added to the liquid phase to modify its surface tension and/or its viscosity [210, 211]. It is also possible to modify the suspension by adding viscous ethylene glycol (the boiling point of which is 200 °C) at the expense of an additional thermal load on the plasma [175]. The addition of binders also controls the suspension viscosity almost independently of the dispersion. For internal feeding against the pressure in axial injection, especially in combustion chambers (HVSFS), the choice of appropriate rheological properties may be critical in terms of a constant suspension feeding and transport, because any sedimentation of the solid phase will cause clogging in the feeding line [192]. Arevalo-Quintero et al. [212] have investigated the behavior of YSZ, samaria-doped ceria (SDC), and mixtures of YSZ and SDC powders in aqueous suspensions. They have determined the optimum dispersant concentrations of three potential dispersant materials (ammonium polyacrylic acid, PAA, polyethyleneimine PEI, and 2-phosphonobutane-1, 2, 4-tricarboxylic acid PBTCA) on the suspension stability.

A key point to achieve good spray conditions is to adapt the size distribution of particles within the suspension to the heat transfer from the hot gases and to limit the width of the particle size distribution [8] (as in conventional spraying). This will allow reducing the trajectory dispersion. At last, powders that have the tendency to agglomerate or aggregate, which is often the case with nano-sized particles, especially oxides, when prepared by chemical routes, must be avoided.

14.7.4.3 Sol Preparation: A New Process Called “PROSOL”

A new process called “PROSOL” [213] has recently been developed by the CEA Le Ripault to avoid the drawbacks of suspensions with solid particles. It consists of injecting a stabilized suspension of nanometer-sized particles (1–100 nm). This kind of sol–gel solution, called colloidal, is prepared by the hydrolysis and condensation of metallic precursors: the inorganic polymerization reaction (nucleation and growth) is controlled by varying the chemical conditions (pH, hydrolysis ratio, etc.) and allows preparing colloidal particles (1–100-nm particles) directly dispersed in the liquid medium [21]. The method of stabilization permits avoiding the use of additives such as dispersant or adding ultrasound during the deposition phase. Consequently, the main advantages are the purity of the deposited material, a very low level of agglomeration and aggregation, well-structured nanometer coatings featuring sizes below 100 nm, simplification of the process, and the capability of reaching sub-micrometric thickness with high deposition efficiency.

14.7.4.4 Suspensions of Amorphous Powders

Amorphous powders can be used in suspensions [19]. Chen et al. [19] have prepared Al2O3–ZrO2 amorphous powders by a chemical solution process. Aluminum nitrate (Al(NO3)∙9H2O)3, and zirconium acetate (ZrO(OOCCH3)3, were dissolved in deionized water based on molar volumes to produce a ceramic composition of Al2O3–40 wt% ZrO2. The obtained solution was heated at 80 °C and stirred continually to get the sol transformed into a dried gel. The dried gel powders were heated to 750 °C at a heating rate of 10 °C/min and then held for 2 h. The phase composition and microstructure of the as-prepared Al2O3–ZrO2 powders heat treated at 750 °C for 2 h was investigated. In the XRD patterns no crystalline peaks appeared (amorphous powders). To prepare the suspension, the powder, with an average particle size of ~5 μm, was mixed and ball-milled in ethanol using ZrO2 balls with a loading rate of 50 wt% for 24 h [19].

14.7.5 Liquid Stream: Hot Flow Interactions

14.7.5.1 General Remarks

Liquid stream or drop fragmentation depends strongly upon the dimensionless Weber number, We, and Ohnesorge number, Oh, (see Sects.  4.5.3 and 14.7.2). When considering a plasma jet, three major zones can be identified [189, 192] (Fig. 14.35):
Fig. 14.35

Schematic of the three major zones of a stationary d.c. plasma flow. Reprinted with kind permission from Springer Science Business Media, copyright © ASM International [196]

  • The plasma jet core (zone 1 in Fig. 14.35)

  • The plasma plume (zone 2 in Fig. 14.35) where the heat and momentum capabilities from the plasma are drastically reduced compared to those in zone 1

  • The plasma fringes (zone 3 in Fig. 14.35) where fragmentation can occur but where the droplet heat treatment will be insufficient

It will be thus of prime importance to inject the liquid as close as possible to the torch nozzle exit and have it penetrating into zone 1 without being fragmented in zone 3. For HVSFS where injection is mainly axial, it is necessary to treat completely the suspension solvent in the combustion chamber (vaporization followed for combustible solvent by full combustion). It explains the use of combustion chambers longer than those used for conventional HVOF [181, 182, 183].

Once the drops or the stream are/is fragmented into droplets of a few micrometers, the vaporization becomes very fast. Indeed the fragmentation time is about two to three orders of magnitude shorter than the vaporization time [151, 152]. The fragmentation depends strongly upon the Weber number (see (2) in Sect. 14.7.2.1) and, for example, if We > 350 (catastrophic breakup) in less than 10 mm trajectory in the plasma jet the liquid is fragmented in droplets smaller than a few μm, droplet sizes that are poorly accelerated due to Knudsen effect (see Sect. 14.7.1.1). In most cases of solution or suspension plasma spraying, the whole mechanisms (fragmentation + vaporization) is achieved about 20–40 mm downstream of the injection location [152], depending on the solvent used, plasma torch power level, and plasma flow velocity. Once fragmentation in droplets below a few μm is achieved, the fast vaporization of liquid and the heating of the resulting gases cool down drastically the hot gas flow (20–30 % of the energy available) [152]. That means that because of this hot gas cooling, the low inertia of the sub-micro- or nano-sized particles [5] the Knudsen effect, not very important in HVSFS, and the Stokes effect [see Eq. (14.1)], the spray distance has to be shorter than that in conventional spraying: 30–50 mm against 100–120 mm for conventional plasma spraying [189] and 60–90 mm with axial injection HVOF against 150–300 mm for conventional HVOF spraying [180]. Correspondingly, the heat flux imposed by the plasma to the coating and the substrate can be up to 40 MW m−2, as illustrated in Fig. 14.36, while in conventional spraying typical heat fluxes are about 2 MW m−2 [189].
Fig. 14.36

Dependence on the spray distance of the heat flux imparted by a stick-type cathode torch (6 mm i.d. anode nozzle) working with Ar–H2, Ar–He, and Ar–He–H2 plasma-forming gas mixtures. Reprinted with kind permission from Dr. E. Brousse [171]

At last, according to the drastic importance of the liquid momentum density [see Eq. (14.3)] for radial injection, the atomization process when resulting in drops of different sizes and velocities does not appear to be necessarily well adapted to achieve the optimum penetration of the liquid, even with improved atomizers [160, 214] (see also Sect.  4.5.1.1). However it must be pointed out that, if originally only laboratories have designed the liquid injectors, in 2011 an industrial injector, based on atomization studies, has been offered by Sulzer Metco [215].

Undoubtedly, besides the solvent, the precursors in solutions and the particles and their dispersant in suspensions, the liquid injection, penetration, dispersion, and fragmentation in droplets below 10 μm within the hot gases is one of the key issues for nanostructured coatings [158, 159]. To understand better the phenomena involved, models of the liquid–plasma interactions have been developed (see “(b) Models” in Sect. 14.7.a). These models rely on the We and Oh numbers as well as the constants and parameters of previous models that have been established for conditions rather different from those prevailing in liquid plasma spraying, and their extension to these conditions should be validated. The model validation through experiments is also difficult (see Sect. 14.7.a). To illustrate the droplet formation, examples of images obtained by shadowgraphy are presented in Fig. 14.37. Figure 14.37a presents one image taken with laser illumination and pulse duration of 8 ns, while Fig. 14.37b presents 50 images recorded by the shadowgraphy system and averaged. The image acquisition at a frequency of 6 Hz allowed taking into account the plasma jet instabilities. And the superposition of images, triggered at the same voltage level, allowed visualizing the average trajectories of the drops.
Fig. 14.37

Liquid feedstock interaction with the plasma flow and average trajectories of the drops downstream of the torch outlet (Ar–H2: 33–10 slm). Reprinted with kind permission from Springer Science Business Media [159], copyright © ASM International

Of course the primary fragmentation depends strongly on the Weber number resulting from the plasma jet, the velocity and specific mass of which depend on the drop or jet trajectory through the radial and axial positions where these numbers are calculated. However it is possible to calculate rapidly the main value of the Weber number of the plasma jet, assuming that all the energy supplied is distributed uniformly in the plasma jet section at the nozzle exit. Such calculations have been performed by D. Damiani [216] and Meillot et al. [217] and results are shown in Fig. 14.38 for a water jet injected within two different plasmas: Ar We = 41 and Ar–He–H2 We = 495. It can be readily seen that the primary fragmentation is by far more important with the mixture than with pure Ar. The drops formed are then fragmented farther downstream, more tiny droplets being formed 15 mm downstream with the mixture compared to the Ar plasma. It can be seen in Fig. 14.38b that fragmentation increases from the jet fringes to the jet axis and farther.
Fig. 14.38

Visualization of a water jet injection within two plasma jets with different Weber numbers: (a) Ar We = 41, (b) Ar–He–H2 We = 495. Horizontal lines correspond to the nozzle axis and limits (torch i.d. 6 mm). Reprinted with kind permission from Dr. D. Damiani [216]

For example, Fig.  16.57 from Bertolissi et al. [159] show the detected number of droplets of a solution in ethanol 15 mm downstream of their injection location (the liquid jet of mechanical injection being focused to the axis of the anode nozzle exit) for 3 spray conditions (Ar, Ar–H2, Ar–He–H2). The liquid injector nozzle was 200 μm in i.d. corresponding about to about injected liquid jet diameter of 400 μm. This figure shows clearly that fragmentation increases with plasma jet velocity and power dissipated in the torch, the highest values being obtained with the Ar–H2–He mixture. When injecting water or ethanol, Fig.  16.58 [158] shows that at 15 mm most of the ethanol droplets are fragmented below 5 μm, while this is far from being the case for water droplets. At 40 mm, ethanol droplets are necessarily below 5 μm because none is seen, which is not the case with water (see Fig.  16.58) and with these spray conditions 20 mm downstream of the nozzle exit no ethanol droplet can be observed. Once droplets cannot be measured with this device, pictures such as that presented in Fig.  16.55b show a “mist” of particles. According to calculations such as those presented in Fig.  4.75, it can be inferred that once droplets have reached a size below a few μm they vaporize and free the solid particles contained for suspensions and pyrolyze, then sinter, and at last melt for solutions. However, according to the drastic Knudsen effect for particles around or below 1 μm, the velocity of the resulting melted or partially melted particles will be mainly that imparted to their mother droplet.

14.7.5.2 Solutions

As described by the model of Ozturk and Cetegen [218] and Basu et al. [174], besides the momentum and heat transfer equations for the interactions between drops or droplets and plasma, the conservation of solute mass and energy equation have to be considered. As for the suspension drops, solution drops are first fragmented, the heating of the droplets precipitating the solute as a shell. Then, compared to the suspension droplets, where the diameter is controlled by the solvent evaporation, the solution droplet size is now fixed by the outer diameter of the precipitate shell formed through the droplet heating. The interior of the droplet is divided into three zones: solid shell, liquid core, and a vapor annulus between them. Depending on the solute and solvent characteristics and thermophysical conditions controlling the precipitation, the shell will be more or less porous. The pressure of the vapor inside it, depending on the rate of vapor leaving it, that is, the shell porosity, can be calculated as well as the critical pressure over which the shell will fracture (if its failure stress is known). The precipitate formation depends on droplet sizes and solute initial mass fraction. In small droplets (5 μm in diameter), precipitation, whatever may be the initial solute level, encompass the whole droplet, thus creating solid particles [174]. For larger particles (only particles up to 40 μm were modeled) a shell precipitate is formed and subsequently fragments, depending on the vapor pressure within particle. To summarize, Saha et al. [219] show that different particle morphologies result from the different processes undergone by drops or droplets, processes linked to their trajectories in hot gases. Finally these morphologies include solid particles, hollow shells, and fragmented shells as shown in Fig. 14.39a–c. Small droplets with high solute diffusivity exhibit a propensity to precipitate volumetrically to form solid particles as shown in Fig. 14.39a. Rapid vaporization combined with low solute diffusivity and large droplet sizes can lead to significant increase in solute concentration near the droplet surface resulting in surface precipitation to form a crust around the liquid core of the droplet. The crust/shell may have varied levels of porosity. Shells having low porosity usually rupture due to internal pressurization to form shell fragments (path I in Fig. 14.39b). Shells that are completely impervious rupture and secondary atomization having a trapped liquid core may be observed (path III in Fig. 14.39b). For shells with a high level of porosity, the internal pressure rise is counterbalanced by the vapor venting through the pores resulting in hollow shells (path II in Fig. 14.39b). For particular precursors, elastic inflation and subsequent collapse and rupture of the shell can also be observed (Fig. 14.39c). The particle morphology resulting from droplet processing is hence sensitive to the solute chemistry, mass diffusivity, solute solubility, droplet size, thermal history, injection type, and velocity [219]. In summary, according to the calculations of Saha et al. [219], the final coating microstructure depends upon the size of the droplet, not on directly whether they follow a trajectory on the torch axis or a deviated path after primary precipitation. Globally, droplets in the size range 5 or 10 μm get pyrolyzed completely before reaching the substrate, while the 20 μm and larger droplets remain partially pyrolyzed. Ozturk and Cetegen [218] presented similar results.
Fig. 14.39

Different routes for droplet vaporization and solid precipitations: (a) Uniform concentration leading to solid particles by volume precipitation. (b) Super-saturation near the surface: (i) low permeable shell leading to fragmented shell formation, (ii) high permeable shell leading to unfragmented shell formation, (iii) impermeable shell leading to droplet internal heating, pressurization and droplet breakup, secondary atomization. (c) Elastic shell formation causing inflation and deflation by solid consolidation. Reprinted with kind permission from Elsevier [219]

Thus, the drop fragmentation and the resulting droplet penetration into the plasma jet core or its fringes again play a key role as pointed out by Jordan et al. [155] and Gell et al. [196]. Figure 14.40 from [155] illustrates the different mechanisms including precursor solute precipitation, pyrolysis, sintering, melting, and crystallization. According to Gell et al. [167] and Jordan et al. [155], particles traveling in the hot core of the plasma jet exhibit mechanisms A, B, and C of Fig. 14.40. If the standoff distance increases molten particles may resolidify and crystallize before impact (D mechanism). When droplets travel in the jet fringes but still in the hot area (T < 4,000–5,000 K), A and B (Fig. 14.40) mechanisms are the most probable. At last, for droplets traveling in the low temperature regions of the jet, some precursor solution can reach the substrate in liquid form.
Fig. 14.40

Schematic illustration of the different treatments of solution droplets within a d.c. plasma jet. Reprinted with kind permission from Springer Science Business Media [155], copyright © ASM International

Works have been devoted to the different parameters controlling the solution spraying [155, 200, 219, 220, 221]. To elucidate the difference in deposition mechanisms of precursor droplets, two different substrate arrangements were employed with a stationary plasma torch [155]. In order to investigate the spray patterns produced under various processing conditions, Xie et al. [220], used the spray experiment shown schematically in Fig. 14.41. In this experiment, the plasma torch moves only in one direction and the substrate is held stationary to obtain a bead, which thickness can be amplified by multiple passes. Due to the vast differences in the temperature histories experienced by the precursor droplets traveling in the core of the plasma jet on the one hand and in its fringes on the other, the central part of the bead and its fringes are rather different. To study only the central part of the bead, they placed a ceramic plate with a 20-mm diameter hole in its center, in the front of the plasma torch, and aligned with its centerline. Thus the coating material traveling in the outer region of the plasma jet was shielded from the substrate. Xie et al. [221] sprayed an aqueous chemical precursor feedstock, resulting in a ZrO2–7 wt% Y2O3 ceramic solid solution, with a Metco 9MB plasma torch (35–45 kW) working with a mixture of Ar and H2. Coatings were collected on the substrate preheated to 500 °C. It was found that the deposits from the precursor injected into the hot region of the plasma jet are crystalline 7-YSZ, while those from the precursor droplets riding along the cold region of the plasma jet are amorphous and contain significant amount of water. The morphology of the deposits that contain significant amount of water is illustrated in Fig. 14.42a. They were collected on the substrate preheated to 500 °C. Figure 14.42b–d, at higher magnification, shows the individual features of the deposit. Figure 14.42b shows a hollow shell, while Fig. 14.41c shows a partially broken shell. Figure 14.42d shows a completely broken shell. During evaporation of the remaining solvent, the deposits appear to inflate (Fig. 14.42b), fracture (Fig. 14.41c), and rupture (Fig. 14.42d). These types of deposits were rarely observed on substrate with no-preheat, reinforcing the notion that evaporation of the wet-precursor occurs after the precursor has landed on the hot substrate.
Fig. 14.41

Schematic of the window-shield fixed scan spray set-up. Reprinted with kind permission from Elsevier [220]

Fig. 14.42

SEM micrographs of deposit from wet-precursor (30 mm offset and 500 °C substrate): (a) overview; (b–d) detailed views of the individual deposit. Reprinted with kind permission from Elsevier [221]

Coatings (made with the ceramic shield of Fig. 14.41) from the jet hot region are different [221]. The first morphology comprises splat-like features, as shown in Fig. 14.43a. The diameters of the “splats” in the SPPS coating (Fig. 14.43a) range from 0.5 to 5 μm. The second morphology is the fine spherical particles, as indicated in Fig. 14.43a. The third morphology is shown in Fig. 14.43b, which also appears to be “splat” like. A higher magnification image (Fig. 14.43c) shows that this deposit is rough and consists of angular grains of sizes ranging from tens to hundreds of nanometers.
Fig. 14.43

SEM micrograph of the top view of the deposit collected on window-shielded substrate: (a) showing “splats,” (b) showing a “splat”-like large deposit, and (c) a higher magnification SEM image showing fine, equiaxed grain structure. Reprinted with kind permission from Elsevier [221]

The initial solution concentration also plays a key role in the droplet pyrolysis. Once the solvent is selected (water, ethanol, isopropanol, etc.), the solution concentration can be varied up to the equilibrium saturation concentration. A high concentration close to the equilibrium saturation concentration tends to produce volume precipitation [155, 200, 219, 220, 221]. Collecting beads, or more precisely overlapped beads obtained by passing the torch a given number of times at the same position, gives relevant information on the heat treatment of the particles [206] by considering the central part and edges of the beads. Figure 14.44 from Chen et al. [211] shows typical SEM microstructures of the collected solution coatings of Y-PSZ on substrates at room temperature from low (A) and high concentration (B) solutions, respectively. No splats are observed in the central zone of the deposited bead with the low concentration solution precursor, mainly composed of ruptured bubbles and a small volume fraction of solid spheres (<0.5 μm) (Fig. 14.44a). The deposited bead central zone made from the high concentration solution is mainly composed of overlapping splats, with an average diameter ranging from 0.5 to 2 μm, and a small amount of unmelted solid spheres (< 0.5 μm) (Fig. 14.43c). The deposited bead edges from both dilute and concentrated solutions (Fig. 14.43b, d) are made of un-pyrolyzed precursor containing significant amounts of water. The mud-like cracks presented at the edges are the result of shrinkage due to liquid phase evaporation.
Fig. 14.44

Deposited beads collected on room temperature substrate by single scan experiment. (A) From diluted precursor (a) deposited bead central zone, (b) deposited bead edge. (B) From high concentration precursor: (c) deposited bead central zone, (d) deposited bead edge. Reprinted with kind permission from Elsevier [200]

At last it must be recalled that by adding to the solution a component to increase the exothermal potential of the decomposition process helps to achieve denser coatings (see Sect. 14.7.2.1).

14.7.5.3 Suspensions

The selection of the particles (morphology and size distributions) and their mass load in the suspension are key parameters. Typical mass loads vary between 5 and 20 wt%. However, with a plasma spray torch with a rod-type cathode, coatings manufactured with a suspension containing particle mass load over 10 % are less cohesive [13] (due very likely to the loading effect on the plasma flow), unless the hot gas enthalpy is increased. The size distribution and particle morphologies are also important parameters to consider. Results will be drastically different depending on the ability of solid particles contained in the liquid to agglomerate, or worse, to aggregate. For example, particles (yttria partially stabilized zirconia: YPSZ) made by attrition milling, after the solvent evaporation, are accelerated (their initial velocities being those gained by droplets before their complete vaporization), heated, and melted by the plasma as shown schematically in Fig. 14.45.
Fig. 14.45

Schematic of suspension droplet treatment when containing attrition-milled particles. Reprinted with kind permission from Springer Science Business Media [169]

When using nano-sized particles obtained by a chemical process (commercial YPSZ Tosoh powder), where the size distribution is between 0.1 and 3 μm, due to the agglomeration and aggregation of the initial grains (mean size of 25 nm), the behavior upon solvent vaporization is different. It seems that the bigger agglomerates or aggregates explode upon solvent evaporation resulting in molten particles and smaller particle evaporation as illustrated in Fig. 14.46.
Fig. 14.46

Scheme of the suspension droplet treatment in the plasma jet for the Tosoh nanometric powder containing aggregates and agglomerates resulting in a size distribution between 0.1 and 3 μm. Reprinted with kind permission from Springer Science Business Media [169]

14.7.5.4 Conclusions

Small particles contained in suspensions or formed in solutions are poorly accelerated by the plasma flow. That is why finally the solid particle velocity contained in suspensions is that achieved by its carrier droplet just prior its vaporization. For solutions, droplets vaporization is stopped by pyrolization followed by sintering and thus the final size of particles formed is bigger than those contained in suspensions (typically a few μm).

When the liquid jet or drops penetrate radially the plasma jet, they are either progressively fragmented in droplets if the Weber number is below 200–300 or very rapidly if We > 350. Thus their volumes and apparent surfaces become smaller. The later will be the fragmentation, the higher will be the solid particle velocities. The different values of the Weber numbers along drop and droplet trajectories will condition their fragmentation. To delay fragmentation without modifying the spray and injection conditions, the Weber number can be reduced when shifting from ethanol to water or when reducing the size of initial drops.

With axial injection in HVOF guns, the solvent is mainly evaporated in the combustion chamber, if it is long enough, the solid particles contained in suspensions, being accelerated by the flow downstream of the nozzle throat. Compared to plasma jets, their acceleration is much better than in conventional plasma jets because, trajectories in hot gases are longer, the Knudsen effect is much lower and the gas mass density higher. To our best knowledge no solutions have been sprayed that way. For axial injection with the Mettech Axial torch no clear explanation of the phenomena can be given because no measurements can be performed where drops are injected. However drops are probably fragmented when entrained in the convergin three plasma jets and it can be supposed that drops vaporization plays a key role. As the high temperature plasma core is much longer than with conventional plasma jets, particles reach higher velocities.

14.7.6 Coating Manufacturing Mechanisms

14.7.6.1 Splats

To collect solid sub-micrometric particles, after their plasma treatment, one way is to intercept the particles on a polished stainless steel or glass substrate positioned at the extremity of a pendulum and traversing the jet at about 1 m/s at different distances from the nozzle exit [151, 169, 222, 223]. Of course, the melted particles flatten and form splats, while those completely resolidified or not melted rebound or form some sort of gel phase for poorly treated droplets. Similar results are be obtained by moving rapidly (1 m/s) the torch relatively to the substrate positioned at different distances. Of course, to collect such tiny splats the substrate surface has to be polished. Besides, as in conventional plasma spraying [224], preheating the substrate over the transition temperature (see Sect.  13.4) allows obtaining close to disk-shaped splats while on cold substrates splats with irregular shape are collected, as shown in Fig. 14.47.
Fig. 14.47

YSZ splats collected on mirror-polished stainless steel substrates (a) cold and (b) preheated. Spray conditions identical to those of Fig.  13.15 ethanol suspension with 10 wt% Tososh powder. Reprinted with kind permission from Springer Science Business Media [13], copyright © ASM International

(a) Solutions
The surface of spray beads can be divided into adherent deposits (bead central part) and powdery deposits (bead edges) that correspond to the hot and cold regions of the plasma jet, respectively. Four deposition mechanisms, corresponding to splat formation or particle deposition have been identified [196, 221]:
  1. 1.
    Smaller droplets that undergo further heating to a fully molten state and crystallize upon impact to form ultrafine (0.5–2 μm average diameter) splats, as illustrated in Fig. 14.48a.
    Fig. 14.48

    Several deposition modes of SPPS. (a) Fine splats, (b) crystallized spheres, (c) ruptured shell, and (d) vapor deposited film. Reprinted with kind permission from Springer Science Business Media [196], copyright © ASM International

     
  2. 2.

    At certain spray distances, droplets undergo resolidification and crystallization before impact upon the substrate to form fine crystallized spheres, as illustrated in Fig. 14.48b.

     
  3. 3.

    Droplets entrained in the cold regions of the thermal jet where they experience sufficient heating to cause solute evaporation leading to the formation of a gel phase, deposited on the substrate. Some droplets also form a pyrolyzed shell containing un-pyrolyzed solution that fracture during deposition, as illustrated in Fig. 14.48c.

     
  4. 4.

    Some precursor solution droplets can reach the substrate in liquid form, having undergone none of the aforementioned processes, as illustrated in Fig. 14.48d.

     
(b) Suspensions
Splats, obtained when spraying a suspension containing attrition-milled particles, are presented in Fig. 14.49 from Delbos et al. [169]. On the whole the flattening degree is below 2 and the splat distribution follows rather well the particle size distribution. However some splats are bigger than those that should be observed with such flattening degree. They probably correspond to droplets that have contained a few particles that have melted together. This is confirmed when spraying suspensions of two different powders [178]. However when spraying a solution of alumina and YPSZ particles, splats of alumina and YSZ were obtained and the resulting coating was made of layered alumina and YPSZ splats [222]. However, due to the good heat transfer of hydrogen (see spray conditions in Fig. 14.49 caption), particles below 0.4 μm are either vaporized and afterwards recondensed or, for those only melted in-flight, resolidified at the spray distance.
Fig. 14.49

Splats collected on a glass beam when spraying an ethanol suspension of attrition-milled YSZ particles with a distribution of 0.2–3 μm and sprayed with an Ar–H2 plasma (45–15 slm, h = 17.9 MJ/kg). Reprinted with kind permission from Springer Science Business Media [169]

For YPSZ powder made of nanometer-sized particles, which agglomerate easily, even with a dispersant, splats collected have a rather wide distribution, as shown in Fig. 14.50. Most splats collected are below 1μm, but few of them are collected with diameters up to 2 μm.
Fig. 14.50

Splats collected on a glass beam when spraying an ethanol suspension of attrition-milled YSZ particles with a distribution of 0.2–3 μm and sprayed with an Ar–H2–He plasma (40–10–30 slm, h = 22.3 MJ/kg). Reprinted with kind permission from Springer Science Business Media [169]

Thus, agglomerations of such particles must be avoided or at least limited [151], which can be achieved, for example, by attrition or ball milling the suspension after its preparation.

It is also possible to collect splats and particles on a substrate protected by a water-cooled shield, positioned at the end of an actuator and intercepting particles and plasma fluxes. The substrate is exposed to the particle flux only during a few tenths of seconds. To enlarge the collecting zone of splats and particles in one direction, the line-scan test, as developed by Bianchi et al. [225] for conventional spraying, can also be used. Siegert et al. [226] have also used a similar device. Such systems allow collecting splats and almost spherical particles distributed as shown schematically in Fig. 14.51. Particles that have traveled in the hot zones of the jet (zone 1 of Fig. 14.35) are well melted and form splats, while those that have traveled in the jet fringes (Zone 2 and 3 in Fig. 14.35) form tiny spheres sticking all around the central zone (see Fig. 14.51).
Fig. 14.51

Schematic distribution of splats and particles collected on a disk-shaped substrate: PT-F4 torch nozzle 6 mm i.d., Ar–H2 (45–15 slm), I = 500 A, zirconia suspension 7 wt%, particle size (0.1–3 μm). Reprinted with kind permission from Dr. Etchart-Salas [189]

This figure illustrates well that the particles treated in zone 1 of Fig. 14.35 form splats, the layering of which will result in dense coatings, while those traveling in zones 2 and 3 of Fig.14.35 will result in the powdery part of coatings. With solutions similar results are obtained with a mud-like cracked film in the periphery (see Fig. 14.42). Under the repeated scan of the high temperature plasma jet, the mud-like film in situ evaporates, pyrolyzes, and crystallizes on the substrate and results in a porous coating.

14.7.6.2 Spray Beads

The degree of melting of particles can be evaluated with a line-scan-spray experiment considering either a simple bead in one passage or overlapped beads with successive passes [218, 221]. The bead thickness depends upon the relative torch/substrate velocity, the number of passes, the suspension or solution flow rates and the injection parameters, the mass load of powder particles in suspension or the solution concentration and, of course, the torch-operating conditions. A typical bead, manufactured by suspension plasma spraying, is presented in Fig. 14.52 together with the schematic of the torch movement. It can be seen that the bead can be fit rather well with a Gaussian profile. Xie et al. [220] have obtained similar beads with solutions.
Fig. 14.52

Typical spray bead manufactured by suspension plasma spraying with the schematic of the corresponding torch movement

(a) Solutions
Xie et al. [220, 221] and Chen et al. [200] have studied beads obtained with YSZ solutions and a conventional plasma spray torch under various operating conditions. The beads can be divided into adherent deposit and powdery deposit corresponding respectively to particles traveling in warm or cold regions of the plasma jet. Adherent deposits seem to result from precursors traveling in both the warm and cold regions, the poorly treated precursor issued from cold regions being incorporated into the adherent deposit. Powdery deposits in the edges of the spray bead correspond to precursors traveling in the low temperature regions (fringes). Figure 14.53a is a representative macro-photo of samples produced in the fixed scan spray experiments, and Fig. 14.53b–d is the top view of the illustrated regions of the sample. As shown in Fig. 14.53b and d, the powdery deposits consist of loosely connected powders, while the structure of the adherent deposits is dense as shown in Fig. 14.53c.
Fig. 14.53

Top view of the samples deposited using SPPS. (a) Macrophoto, (b and d) micrograph of the powdery deposits, and (c) micrograph of the adherent deposits. Reprinted with kind permission from Elsevier [220]

Chen et al. [203] have studied the influence of the precursor concentrations. The central part of the spray bead deposited on a cold substrate consists of semi-pyrolyzed material that will form soft and porous coatings. Upon impacting the substrate at room temperature, the shells on the droplet surfaces fracture and form bubbles. However, bubbles evolve to spongy deposits when the substrate temperature is raised to 450 °C. In contrast, the high concentration precursor experiences volume precipitation leading to a fully melted lamella microstructure and resulting in a dense coating. In both cases, non-pyrolyzed material will in situ pyrolyze and form aggregates when the substrate temperature is above the precursor pyrolysis temperature. This pyrolysis also occurs when reheating the coating by the successive torch passes [196].

(b) Suspensions
A similar observation can be made when considering, the spray beads obtained with suspensions, see Fig. 14.54 [227, 228]. The central part of the bead is relatively dense and mainly consists of mainly a few splats, spherical particles, and unmelted ones.
Fig. 14.54

Structure of an alumina suspension sprayed bead (particles size 0.02–1.0 μm, spray conditions see Fig. 14.26 caption). Reprinted with kind permission from Elsevier [228]

This central part corresponds to particles that have been well treated in the plasma core plus particles that have traveled in colder zones. In the bead at mid-height, the coating is less dense with many unmelted particles. At last, in the edges of the bead, the coating is fully powdery. To our knowledge, no bead was studied in experiments with axial injection (Mettech plasma torch or HVOF torches). It is also interesting to study the influence of the spray parameters on the bead morphology [227, 228]. Figure 14.55, depicting the influence of the spray distance and the mass percentage of the particles, illustrates the influence of spray parameters. The densest coating is obtained at a spray distance of 30 mm (when the plasma heat flux reaches 15 MW/m2). The density seems to be slightly lower with the 10 wt% mass load compared to the 5 wt% one. The coating thickness, when the spray time and the mass load are doubled, is also slightly diminished, which is probably due to the loading effect. When spraying at 40 mm, beads are thicker than at 30 mm but less dense (due to the incorporation of more untreated particles) and it is worst at 50 mm.
Fig. 14.55

Alumina suspension plasma-sprayed microstructures obtained with the spray conditions depicted in Fig. 14.31 for three different spray distances (30, 40, 50 mm) and two powder mass loads (5 and 10 wt%). Reprinted with kind permission from Springer Science Business Media [227], copyright © ASM International

14.7.6.3 Coatings

As discussed in the previous section, when superposing beads during spraying, more or less non-pyrolyzed particles (solution spraying) or poorly treated particles (suspension spraying) are embedded in the coating during deposition. The key parameters on which coating properties depend are the following [5]:
  1. 1.

    The momentum of drops/momentum density of the liquid stream that controls the drop penetration into the plasma jet hot core. When considering atomized drops, it is thus of primary importance to generate a pattern of small dimensions and with a narrow range of trajectories, a narrow drop size distribution (see Fig.  4.71) together with a narrow drop velocity distribution [214].

     
  2. 2.

    For solutions, a high concentration precursor (close to equilibrium saturation) to promote volume precipitation leading to a fully melted splat microstructure and a high-density coating [200].

     
  3. 3.

    For suspensions, the size distribution and the particle morphology (linked to their manufacturing process) [5].

     
  4. 4.

    The type of liquid phase [203]: droplets with a liquid phase with high surface tension and high boiling point experience incomplete liquid phase evaporation in the thermal jet. For solutions, the result is a mud-like cracked film on the substrate leading to a porous coating. Droplets created from a low surface tension and low-boiling point liquid phase undergo rapid liquid phase evaporation, solute precipitation, pyrolysis, melting process in the plasma jet and form splats upon impact on the substrate, the stacking of which result in a dense coating. For suspensions the solid particles contained in the liquid droplets that are incompletely evaporated are poorly treated [229].

     
  5. 5.

    The substrate and coating temperature during spraying. For suspensions and solutions, substrates must be preheated over the transition temperature to get rid of adsorbates and condensates. For solutions, non-pyrolyzed materials are pyrolyzed at the substrate surface and form aggregates when the substrate temperature is above the precursor pyrolysis temperature. This pyrolysis also occurs when reheating the coating by the successive torch passes [196, 203]. Under the repeated processing by the high-temperature thermal jet, the mud-like film (for example, formed when spraying droplets with a high surface tension and high-boiling point liquid phase) in situ evaporates, pyrolyzes, and crystallizes on the substrate and results in porous coatings. For suspensions, the high temperature substrate and coating, as well as the hot gas heat flux can delay the splat cooling, resulting in the recoil of the liquid flattened drop before solidification starts.

     
  6. 6.

    At last the surface roughness Ra must be adapted to the size of particles d 50 with a ratio Ra/d 50 < 1–2 (see Sect. 14.7.1.2 and) to avoid speckles [173] and cauliflower [170]-type coatings.

     
(a) Solutions
In solution spraying the amount of non-decomposed precursor can be controlled by spray parameters, primarily droplet injection momentum density, spray droplet sizes, and trajectories dispersions, as well as precursor concentration [196]. Figure 14.56 illustrates the necessity to adjust the liquid droplet trajectories to achieve a distribution as narrow as possible in the hot core of the plasma jet and obtain uniform dynamic and thermal treatments.
Fig. 14.56

Possible liquid droplet trajectory distributions in a d.c. plasma jet. Reprinted with kind permission from ASM International [214]

When overlapping beads contain un-pyrolyzed particles (mainly those in the jet fringes) embedded within the layers during manufacturing (see section “(a) Solutions”), upon heat exposure either by subsequent plasma torch passes or postdeposition treatment, their pyrolysis generates tensile stress driving the formation of vertical cracks [214, 229], as illustrated in Fig. 14.57a.
Fig. 14.57

Intra-lamellar cracked zirconia coatings deposited by solution plasma spraying with an Ar–H2 plasma produced by a 9-MB Sulzer Metco d.c. plasma torch [214] (a) general view (b) detail. Reprinted with kind permission from ASM International [214]

Coatings are made of micron-sized lamellae, which have grain sizes smaller than 30 nm, and they exhibit a porous architecture at two scales (nano- and micro-scales) made of voids and through-thickness stress relieving intra-lamellar cracks. The unique intra-lamellar cracks could result from pyrolyzation shrinkage. Porosities, as those shown in Fig. 14.57b, result also from the entrapment of poorly treated droplets in the coating. The fraction of un-pyrolyzed material has hence to be carefully controlled through the drop atomization (size and trajectory distributions), the spray pattern, and the solution concentration. To manufacture a dense (and hard) coating under the same operating conditions, the fraction of poorly treated material in the jet fringes needs to be significantly reduced, if not suppressed. Besides the optimization of the drop injection, dense coatings are linked to solutions with high concentrations [214]. Figure 14.58 shows that many more splats are observed at the surface of a coating sprayed with a high solution concentration than with a low concentration. With optimized conditions, coating architectures as those presented in Fig. 14.59 are obtained. Those layers are reasonably dense (88 % total pore level), hard (1023 VHN, average value), and exhibit no apparent intra-lamellar cracking.
Fig. 14.58

Coating surface morphologies from solution precursors. (a) Low concentration. (b) High concentration. Reprinted with kind permission from Elsevier [200]

Fig. 14.59

Dense and hard zirconia coatings deposited by solution plasma spraying with an Ar–H2 plasma using a capillary atomizer. Reprinted with kind permission from ASM International [214]

(b) Suspensions

In HVOF or plasma suspension spraying, it is crucial to achieve particle melting, or particles close to the melting state, at impact. In spite of their much lower temperature, HVOF jets provide good heat transfer, with particles in the sub-micron or few micron size, compared to that of plasma jets where the Knudsen effect is drastic (see Sect.  4.5.4). Moreover, when using water instead of ethanol as suspension solvent, the resulting coatings are generally more porous than those achieved with ethanol [5, 13]. What concern the online monitoring of suspension spraying (as well as that of solution spraying) most studies have been parametric ones, as illustrated in [170, 230, 231] for conventional d.c. plasma torches or for the Mettech Axial III torch [198]. In the following, to illustrate the influence of the different parameters for plasma spraying, only results will be presented of plasma-sprayed suspensions of partially yttria-stabilized zirconia.

(i) Plasma Spraying

Influence of Particle Morphologies and Size Distributions

The effects of powders are also linked to the choice of the plasma-forming gas adapted to particles to be treated. In the following it is illustrated with different yttria (8 mol %) partially stabilized zirconia particles, the size distributions of which are summarized in Table 14.3 [171, 189, 232]. All suspensions were sprayed with either Ar–H2 or Ar–H2 or Ar–He plasmas produced with a stick cathode torch at the same spray distance of 40 mm [189, 232].
Table 14.3

Powders used in suspensions by Delbos [232] and by Etchart-Salas [189] Delbos: Tosoh-TZ-8Y and Medipur attrition milled) (Etchart-Salas: Marion powder resulting from soft chemistry, UC-001h and UC-002h both attrition milled)

 

d 0.1 μm

d 0.5 μm

d 0.9 μm

Mean grain size, nm

Powder wt%

Tosoh-TZ-8Y

0.1

1.5

3.5

30

7

Medipur attrition milled

0.15

1

3

>150

7

Marion

0.03

0.06

0.76

20–40

20

UC-001h attrition milled

0.03

0.05

0.29

30

20

UC-002h attrition milled

0.26

0.39

0.7

260

20

First of all, when using nanometer-sized particles, which have the tendency to agglomerate and aggregate, such as the Tosoh TZ-8Y, resulting in a broad particle size distribution between 0.1 and 3.5 μm, the coating is very porous as illustrated in Fig. 14.59a.

Fig. 14.60a presents the Tosoh powder, the size grading of which is large (see Table 14.3 and Fig. 14.60b) with nanoparticles, and their agglomerates or aggregates, the typical specific surface area being 13 m2.g-1. A typical coating obtained with this powder onto a 316L stainless steel substrate with a Ra of 0,05 μm is presented in Fig. 14.61a. It was sprayed at 40 mm standoff distance with Ar–H2 plasma (45-15 slm) working at 500 A and an anode nozzle 6 mm in i.d. (enthalpy of 17.9 MJ/kg). The liquid was injected as a jet (aimed at the nozzle exit axis without plasma), which injection pressure corresponded to the optimized penetration of the suspension (see Fig.  4.79). Figure 14.61a shows the very porous coating with a columnar structure. Thin coating cross sections (below a few μm) have been analyzed to compare the initial particle distribution to those of the splats and spheres within the coating and results are presented in Fig. 14.61b. All splats have sizes over 0.2 μm, which means that the smallest fully melted particles impacting the substrate are about 0.1 μm in diameter. The spheres correspond probably to small particles, bigger than 0.1 μm, which were melted rapidly but have started to solidify before reaching the substrate. It means that coatings are mostly formed; thanks to particles over 0.1 μm. Due to Knudsen effect, temperatures of the smaller particles are below the melting point and/or their velocities not high enough to achieve Stokes numbers over one. The columnar structure is essentially due to the high voltage fluctuations (between 40 and 80 V for a mean voltage of 60 V), the impact velocity of particles injected at low voltage being too low to achieve Stokes numbers > 1, thus promoting the shadow effect described in Sect. 14.7.1.2. Moreover the agglomerate fragmentation creates a large dispersion of trajectories and nonuniform treatment of particles resulting in coatings with poor cohesion. Tiny particles were also probably evaporated and then recondensed on the coating during its formation creating defects. The agglomerate fragmentation creates a large dispersion of trajectories and nonuniform treatment of particles resulting in coatings with poor cohesion. To clarify the role of the different particle sizes, the small particles were separated from the big ones (aggregates and agglomerates in the few micrometer-sized range). It was achieved by forming an ethanol suspension with no dispersant and by its sedimentation during a sufficient time (42 h 25 min) to achieve particles separation. Figure 14.62 represents the powder distribution of the micron-sized agglomerates and aggregates sprayed with the same spray conditions together with resulting splats and spheres. These two distributions are rather similar to those presented in Fig. 14.61 obtained with the initial powder, confirming that coating result mainly of particles with sizes over a few tenths of microns. At last the small particles (< 0.3 μm) have been sprayed in the same conditions, and Fig. 14.63 shows that the splat distributions of both, the sub-micron- and micro-sized particles, are quite similar, coatings formed with splats result mainly of micro-sized particles.
Fig. 14.60

(a) SEM view of Tosoh powder. (b) Size grading of the powder by number [232]

Fig. 14.61

(a) Coating cross section obtained with the suspension of Tosoh powder, ethanol 10 wt% YSZ, 2 vol.% dispersing agent, sprayed with Ar–H2 plasma : 45–15 slm, h = 17.9 MJ/kg. (b) Size distributions of initial powder, splats, and spheroidized particles. Reprinted with kind permission from Springer Science Business Media [169]

Fig. 14.62

Size distributions of initial powder, splats, and spheroidized particles obtained when spraying suspension of agglomerates and aggregates with the spray conditions depicted in the caption of Fig. 14.61. Reprinted with kind permission from Springer Science Business Media [169]

Fig. 14.63

Size distributions of splats obtained when spraying suspensions on the one hand of agglomerates and aggregates (micron particles) and on the other of the Tosoh powder with the spray conditions depicted in the caption of Fig. 14.61. Reprinted with kind permission from Dr. Delbos [232]

A suspension of Tosoh powder, where most aggregates and agglomerates were taken away, was also sprayed with the same torch but using Ar–He (40-80 slm) plasma with a current of 500 A corresponding to an enthalpy of 13 MJ/kg and a mean gas velocity of 1,425 m/s, against 17.9 MJ/kg and 1,312 m s−1 for Ar–H2 plasma. The coating cross section and its fracture are presented in Fig. 14.64a, b. With this more stable plasma (V m =62 V and ΔV = 15 V) no more columnar structure, as that shown in Fig.14.61a, is achieved. Figure 14.64a shows that the coating is relatively dense with big pores and the fractured coating, shown in Fig. 14.64b shows that it is mostly granular (the explanation is given farther in this section) with most particles with sizes over 0.05 μm. It must also be noted that the deposition of the same coating thickness is about twice longer than that obtained with initial Tosoh powder sprayed with the Ar–H2 plasma. It seems that with this faster plasma jet with low voltage fluctuations the small particles reach sufficient velocities and temperatures to be imbedded within the coating under formation.
Fig. 14.64

Coating achieved with Ar–He plasma (40–80 slm), anode nozzle i.d. 6 mm, I = 500A, 17.9 MJ/kg: (a) Cross section. (b) Fractured cross-section. Reprinted with kind permission from Springer Science Business Media [169]

Then, Delbos [232] has sprayed attrition milled (during 13 h), fused, and crushed particles of Medipur (initial size distribution 5–25 μm). Figure 14.65a presents a SEM view of these particles and Fig. 14.65b their size distribution.
Fig. 14.65

(a) SEM view of attrition milled Medipur powder. (b) Powder size distribution. Reprinted with kind permission from Dr. Delbos [232]

The suspension of attrition-milled Medipur particles was sprayed first with the same conditions as those used for Tosoh powder suspension (Ar–H2 plasma) and the corresponding coating cross section is presented Fig. 14.66a. For this coating, many tiny particles, almost spherical, in the range of 0.05–0.3 μm, are contained in the pores. As with Tosoh powder most splats (studied when spraying layers a few μm thick) have sizes over 0.2 μm (see Fig. 14.67). The coating is denser and less porous (6 %) than those obtained with Tosoh powder.
Fig. 14.66

Attrition-milled Medipur powder suspension sprayed: (a) with the Ar–H2 plasma jet used to spray Tosoh powder suspension (see Fig. 14.61 caption), (b) with a more powerful jet: Ar–H2–He (40–10–50 slm), anode nozzle i.d. 6 mm, h = 22.3 MJ/kg. Reprinted with kind permission from Dr. Delbos [232]

Fig. 14.67

Size distributions of initial powder, splats, and spheroidized particles obtained when spraying suspension of attrition-milled Medipur powder with the spray conditions depicted in the caption of Fig. 14.61. Reprinted with kind permission from Dr. Delbos [232]

In order to improve the heat transfer, this suspension was also sprayed with Ar–H2–He plasma, which enthalpy was 22.3 MJ/kg against 17.9 for the Ar–H2 one, the corresponding coating cross section being presented in Fig. 14.66b. In spite of the fact that, with this plasma forming gases containing 10 vol.% of H2 the voltage fluctuations are not negligible (V m = 75.1 V and ΔV = 30 V), the coating presented in Fig. 14.66b is rather dense (about 7 % porosity). Compared to the coating presented in Fig. 14.66a, with the same suspension flow rate (which injection was optimized for these spray conditions) and pass number, the coating thickness is increased by almost 50 % due probably to the better melting of particles.

Etchart-Salas [189] has studied 3 powders (see Table 14.3): Marion powder made by a “soft chemistry process” and two Unitec Ceramic powders fused and crushed and then milled: UC-001h and UC-002h. The size distributions are presented in Table 14.3. In all cases the injection with liquid jets of suspensions, all containing 20 wt% of particles, has been optimized. The Marion powder, presented in Fig. 14.68, is made of nanometer-sized grains (20–40 nm) and presents much less agglomerates and especially aggregates than the Tosoh powder, and its size distribution is between 0.03 and 0.9 Δm. When it is sprayed (see Fig. 14.69) with Ar–He plasma (30–30 slm with low fluctuations: V m = 50.9 V and ΔV = 18 V), the coating is rather dense (porosity of 4.7 %) and its morphology is close to that of the coating obtained with Medipur attrition-milled powder and the Ar–H2–He plasma (see Fig. 14.66b).
Fig. 14.68

SEM picture of the Marion powder [189]

Fig. 14.69

YSZ coating obtained with a suspension with 20 wt% of Marion powder (nanograins) (Table 14.3) in ethanol injected in Ar–He (30–30 slpm) plasma jet (700 A) where the enthalpy is 17.9 MJ/kg and the anode nozzle 6 mm in i.d. Reprinted with kind permission from Springer Science Business Media [13], copyright © ASM International

When spraying suspension of the attrition-milled powder (UC.001h), shown in Fig. 14.70a, b at two different scales, and, the size distribution (see Fig. 14.70c) of which is between 0.03 and 0.29 μm, with few agglomerates between 0.3 and 1 μm, slightly denser coatings (porosity 4 %) than those achieved with Marion powder are obtained, as shown in Fig. 14.71, [189].
Fig. 14.70

Unitec powder 001h attrition milled: (a, b) SEM pictures at two different scales (c) size distribution by number. Reprinted with kind permission from Springer Science Business Media [151]

Fig. 14.71

YSZ coating obtained with a suspension with 20 wt% of UC-001 attrition-milled powder (Table 1) in ethanol injected into a Ar–He (30–30 slpm) plasma jet where the enthalpy is 17.9 MJ/kg and the anode nozzle is 6 mm in i.d. Reprinted with kind permission from Springer Science Business Media [13], copyright © ASM International

Similar results are obtained when bigger particles with sizes between 0.26 and 0.7 μm (Unitec UC-002h), also attrition milled, are used.

Unmolten Tiny Particles

These tiny particles have a strong tendency to stick to hot (T > 1,000 °C) zirconia layers and create defects, especially between successive passes. When injecting the suspension at the end of each pass when starting to spray the next one, particles traveling in jet fringes stick on the previously deposited and still very hot pass. This is illustrated in Fig. 14.72 showing the deposition of successive pass with the defects created between them by this deposition of unmelted tiny particles.
Fig. 14.72

Schematic of successive pass deposition. (a) Substrate and particle distribution (well heated in the central part, poorly heated in the periphery). (b) Spray pattern. (c) Torch movement with the poorly heated particles in the jet fringes. Reprinted with kind permission from Dr. Delbos [232]

The result is illustrated in Fig. 14.73 with coatings obtained with Unitec UC-002h powder (attrition milled with a rather narrow size distribution, see Fig. 14.70c). In Fig. 14.73a the fluctuating Ar–H2 plasma jet was used with the injection conditions optimized. The formation of columnar structure can be observed, but smoothed compared to that obtained when spraying with the same conditions Tosoh powder (obtained by a chemical route and with a broad size distribution, as shown in Fig. 14.61b). When reducing the suspension penetration in the plasma jet (lower injection pressure) with particles traveling more in the jet fringes, where the plasma jet fluctuations have more influence on particles with lower temperatures and velocities, and the columnar structure is emphasized, as shown in Fig. 14.73b. At last when using Ar–He plasma much more stable than the Ar–H2 one, the deposition of the unmolten particles between passes is very clearly seen, as shown in Fig. 14.73c. On the cold stainless steel substrate only few unmelted particles have been deposited and the first pass is dense even close to the substrate. It is no more the case for the new successive passes which are more porous close to the previously deposited one. During pass deposition in an x–y pattern with beads overlapping by 1/20, the phenomenon also occurs but the tiny particles are included within beads and less perturb the coating cohesion.
Fig. 14.73

Coatings formed with a PTF4 torch, with a 6mm i.d. anode nozzle, onto a stainless steel smooth substrate (Ra = 0.1 Δm) with a standoff distance of 40 mm, using a suspension (20 wt% of powder) of Unitec UC-002h: (a) Ar–H2 (45–8 slm), I = 500 A, h = 14.5 MJ/kg, with optimized injection, (b) same spray condition as with a, but with a reduced injection velocity (less penetration of the suspension within the plasma jet core), (c) Ar/He (30/30 slm) plasma, I = 700 A, h = 19.6 MJ/kg, optimized injection. Reprinted with kind permission from IOP organization [5]

This powdery layer between successive passes can be avoided by changing the spray pattern and the suspension mass load, as is illustrated in Fig. 14.74 representing the same coating as that of Fig. 14.73c. The spray pattern must be such that beads overlapping limit as much as possible the inclusion of particles traveling in jet fringes and also limit the overheating of successive passes also to limit poorly heated particles sticking.
Fig. 14.74

Cross section of the same coating as that presented in Fig. 14.73c but with an optimized spray pattern. Reprinted with kind permission from IOP organization [5]

Lamellar or Granular Structures

When collecting ceramic sub-micron- or nano-sized molten particles onto a smooth substrate preheated above the transition temperature (∼500 K for YSZ on stainless steel, alumina, superalloys, see Sect.  13.3.1.2), splats are identical to those obtained with micro-sized sprayed particles.

However, for d.c. stick-type cathode spray torches [151], their diameters are generally between 3 and 0.2 μm with thicknesses between 300 and 60 nm, corresponding to a flattening degree of about 2 at the maximum. This low flattening degree results from lower impact velocities together with larger surface tension due to the small size of the particles. Moreover with these splat sizes, for ceramic materials, no cracks during quenching appear.

When using high enthalpy jets (Ar–H2–He, 22.3 MJ/kg, with a heat flux of 18 MW/m2), the structure close to the stainless steel substrate is columnar, resulting from splats layering. The columns develop through a thickness of about 4 μm corresponding to the layering of 5 passes, each of them being 0.8 μm thick. It seems that, after a few layers (a few μm thick), the thermal insulation together with the drastic heat flux from the plasma delays the splat solidification, allowing the surface tension force to take over resulting in quasi-spherical particles in a molten state sticking together before their solidification to achieve the granular structure observed in Fig. 14.75.
Fig. 14.75

Fractured cross section of a suspension coating sprayed on a 316L stainless steel substrate 40 mm downstream of the nozzle exit with a plasma torch 6 mm in i.d., an attrition milled powder and an Ar–H2–He: 40–10–30 slm plasma 22.3 MJ/kg, with a heat flux of 18 MW/m2. Reprinted with kind permission from Springer Science Business Media [169]

Spraying Mixtures of Powders

Delbos et al. [222] have used YSZ Tosoh powder, with a specific surface area of 13 m2/g and alumina powder (Baïkowski CR125) with a specific area of 105 m2/g to prepare a suspension. They were mixed with the same weight and the suspension solvent was ethanol and the spray parameters were those for the Tosoh powder, except that the nozzle i.d. was 5 mm to increase the plasma jet velocity [222]. Figure 14.76 shows that the structure of the coating is made up of YSZ and alumina splats alternatively layered (there is a difference of contrast between both oxides: gray for alumina and white for YSZ) with thickness lower than 100 nm, as observed with AFM. Moreover, it could be noticed that the contact between splats seems to be excellent. Oberste-Berghaus et al. [208] obtained similar results with an Axial III torch of Mettech. A detailed study of alumina–yttria-stabilized zirconia composite coatings was performed by Tarasi et al. [178] using also Axial III torch of Mettech. They used either agglomerates of nano-particles or suspensions made either with loose nano-powders mixture or loose micro-powders mixture.
Fig. 14.76

Cross section of an alumina–YSZ coating obtained with suspension spraying with the spray conditions of Fig. 14.75 caption, but with a torch nozzle i.d. of 5 mm instead of 6 mm. Reprinted with kind permission of Dr. Delbos [222]

Through the study of the in-flight collected particles and coatings produced from the two processes, the comparison of fragmentation, melting and mixing phenomena was possible. They showed that both methods can be used in the production of high amorphous coatings provided that the appropriate parameters for each process are utilized. The components in the composite materials sprayed by plasma processes may appear in different forms. They may form crystalline structure of alumina or YSZ with no additional solute atoms. They also can dissolve the solute atoms of the second component and form crystalline solid solutions (even to exceptionally high levels of solubility), and/or form amorphous phase [178]. In-flight melting followed by mixing are crucial processes in amorphous formation. This melting is strongly controlled by the particle velocity, a lower one resulting in higher amorphous content, in spite of the lower cooling rates.

(ii) HVOF Spraying (HVSFS)
An example of HVSFS spraying results is presented with the spraying of TiO2 nanometer-sized (grain size 12nm) particles onto aluminum substrates [180]. In SEM, the titania nanometer-sized powder appears agglomerated with a primary grain size of approximately 10 nm with agglomerates of about 100 nm. In XRD both the rutile and the anatase phase are detected. The suspension contained 10wt% TiO2 and the solvent was isopropanol/water with a volume ratio of 90/10, DOLAPIX being the additive. Water (10wt%) was added to facilitate the chemical stabilization of the suspension (but it cools the flame when vaporizing). The HVSFS torch used was based on a Top-Gun-G system (manufactured by GTV) modified to meet suspension spray requirements. Propane and ethene were preferentially chosen as combustion fuels as they allow stable operation of the HVSFS process. The HVSFS torch was operated on a six-axes robot system using a simple meander kinematics with a 2-mm offset. Sample cooling was performed using two air nozzles mounted on the torch. The suspension was fed at a rate of about 8 g/min using a conically shaped 0.3 mm nozzle. The thickness of coating, deposited on aluminum substrate, was around 30 μm, the deposition rate being around 4 μm/cycle. Considering its cross section, the TiO2 coating appears totally dense (Fig. 14.77), homogeneous, no internal structure is visible, and no unmolten particles are detected. The SEM surface image (Fig. 14.78) reveals fully molten splats ranging from a few tenths to about 10 μm. From the XRD data, the main fraction of the coating consists of anatase (approx. 75 %), the rest of the material being rutile. Traces of a sub-stoichiometric species, Ti3O5, could also be identified in the coating.
Fig. 14.77

Light microscope image of HVSFS-sprayed TiO2 coating. Reprinted with kind permission from Elsevier [180]

Fig. 14.78

SEM image of HVSFS-sprayed TiO2 coating surface. Reprinted with kind permission from Elsevier [180]

Toma et al. [233] compared alumina coatings HVSFS sprayed by suspension with particles in the range 0.4–3.5 μm to those obtained with HVOF and particles in the size range 5–25 μm. They used for HVOF spraying a TopGun working with ethylene–oxygen (90–270 slm) with a spray distance of 150 mm and for HVSFS the same gun, which chamber was modified, but working with same gases (75–230 slm) and a standoff distance of 80 mm. The HVSFS coating presents a specific microstructure: partially melted fine particles with elongated shapes (of about 5 μm in length and 150 nm in thickness), nearly spherical μm-sized particles and clusters of agglomerated tiny sub-μm-sized grains embedded in the matrix of well-melted ones (Fig. 14.79). The main crystalline phase is the α-phase representing 60 %. The HVOF coating is more conventional (Fig. 14.80) with layered splats, γ-alumina phase being the main one. The suspension-sprayed Al2O3 coatings present the better electrical resistance stability that can be explained by their specific microstructure and retention of a higher content of α-Al2O3.
Fig. 14.79

SEM micrographs of cross sections of thermally sprayed alumina coatings. (a) Conventional coatings HVOF-sprayed C2H4/O2 (75/230 slm) with particles in the range 5–25 μm. (b) Suspension HVOF sprayed C2H4/O2 (90/270 slm) with particles in the range 0.4–3.5 μm. Reprinted with kind permission from Springer Science Business Media [233], copyright © ASM International

Fig. 14.80

SEM micrographs with higher magnifications of cross sections of HVOF sprayed coatings depicted in Fig. 14.79. (a) Conventional and (b) Suspension. Reprinted with kind permission from Springer Science Business Media [233], copyright © ASM International

Bolelli et al. [184] also HFSFS-sprayed alumina coatings, using seven different Al2O3-based suspensions prepared by dispersing two nano-sized Al2O3 powders (having analogous size distribution and chemical composition but different surface chemistry), one micron-sized powder and their mixtures in a water + isopropanol solution. HVSFS coatings were deposited using these suspensions as feedstock and adopting two different sets of spray parameters (propane–oxygen either 55–325 slm or 60–350 slm). As with plasma spraying, the characteristics of the suspension, particularly its agglomeration behavior, have a significant influence on the coating deposition mechanism and, hence, on its properties [184]. Dense and very smooth (Ra ~ 1.3 μm) coatings, consisting of well-flattened lamellae having a homogeneous size distribution, were obtained with micro-sized (~ 1–2 μm) powders. Spray parameters favoring the breakup of the few agglomerates present in the suspension enhance the deposition efficiency (up to 50 %), as no particle or agglomerate larger than ~2.5 μm can be fully melted. Nano-sized powders, by contrast, generally form stronger agglomerates, which cannot be significantly disrupted by adjusting the spray parameters. When the chosen nanopowder forms small agglomerates (up to a few microns), the deposition efficiency is satisfactory and the coating porosity is limited. If the nanopowder forms large agglomerates (on account of its surface chemistry), poor deposition efficiencies and porous layers are obtained [184].

Müller et al. [234] have used two different spray processes, suspension plasma spraying (SPS) and HVSFS. They used alumina particles: MR52 (0.33–4.07 μm) and APA-0.5 (0.26–1.19 μm) and suspensions were made either with water or isopropanol. For HVSFS they used a TopGun system equipped with a 22-mm long combustion chamber and working with ethene–oxygen (410 slm) with different oxygen to fuel ratios and a spray distance of 120 mm. For SPS they used a F6-Gun of GTV working with Ar–H2 (55–12 slm) and standoff distances of 55 mm with water and 75 mm with isopropanol (to benefit from its combustion).

Figure 14.81 represents particle distributions before spraying and collected in water after HVSFS spraying. Collected spray particles produced by HVSFS, exhibited a spherical shape, corresponding to their good melting during their in-flight phase. Those from the coarser MR52 powder exhibit the same size distribution after the spray process as those in the used suspension. The finer APA-0.5 powder has its initial d 50 value increased from 0.55 μm to a value of 1.23 μm for the collected particles. The size distribution also broadened from a narrow size distribution in the suspension to a broader distribution of the collected particles after spraying. This could be explained by the agglomeration phenomenon. Using water for the suspension results in higher amount of α-alumina phase.
Fig. 14.81

Size distribution of the particles in the used suspensions and of the collected particles from HVSFS spray process for the two used powders. Reprinted with kind permission from Springer Science Business Media [234], copyright © ASM International

With the SPS process, particle melting was better with isopropanol than with water due to the combustion phenomenon and compared to HVSFS process splat sizes were smaller. When comparing coatings obtained by HVSFS and SPS, α-alumina phase is dominant in the HVSFS coatings, which confirms the better particles melting. Using the SPS process, high coating porosities, up to values of 40 %, were achieved, while with HVSFS low coating porosities (down to 2 %) were obtained.

14.7.6.4 Environmental Impact (Life Assessment Impact: LCA)

The LCA of plasma spray processes using powder, suspension, or solution as feedstock to manufacture zirconia coatings was studied by Moign et al. [235]. The authors pointed out the following points:

– The comparison of the manufacturing of solution for SPPS, suspension for SPS, and powder for APS showed that solution had a lower environmental impact than powder and suspension.

– Different results are obtained when considering the spray process because the amount of electricity used is responsible for about 70–80 % of the environmental impact, and the deposition time linked to deposition efficiency mainly controls the environmental impact. With spraying efficiency of SPS and SPPS of about 30 % at the best against 60 % for the APS process, these processes appear to be much less friendly from an environmental point of view.

14.7.7 Applications

In the following suspension plasma spraying is named: SPS and solution precursor plasma spraying: SPPS. Compared to conventional coatings, those manufactured from solutions or suspensions exhibit quite interesting features such as the absence of lamella boundaries and cracks and porous microstructures with nano-sized grains. Many works have been devoted to many potential applications, as described in the following, but unfortunately, to the best of our knowledge, no industrial developments of these techniques have been mentioned yet.

14.7.7.1 Thermal Barrier Coatings

In the following suspension plasma spraying is named: SPS and solution precursor plasma spraying: SPPS. Compared to conventional coatings, those manufactured from solutions or suspensions exhibit quite interesting features such as the absence of lamella boundaries and cracks and porous microstructures with nano-sized grains. Many works have been devoted to many potential applications, as described in the following, but unfortunately, to the best of our knowledge, no industrial developments of these techniques have been mentioned yet.

(a) Solutions

TBCs were mainly manufactured by solution plasma spraying [155, 156, 196, 221, 229, 236, 237, 238, 239]. As explained previously, the coatings contain uniformly distributed cracks, providing a high degree of strain balance to the ceramic topcoat. Compared to APS conventional TBCs, coatings produced by Solution Precursor Plasma Spray (SPPS) demonstrate [154, 196] interesting properties related to their specific features, including ultrafine splats (which form dense coating regions), through thickness vertical cracks, embedded un-pyrolyzed particles, and voids, as shown in Fig. 14.48. The adhesive bond strength, measured according to ASTM C633-79 standard, was 24.2 MPa, while that of conventional APS-deposited coatings was found to be 19.9 MPa [154]. The improved bond strength probably results not only from the finer splat size but also may be due to an interfacial thin oxide layer that would form on the substrate and would promote adhesion. The vertical crack density increases the in-plane TBC fracture toughness, improving the coatings cyclic durability in the thermal cycling [196]. Gell et al. [196] have shown that, during thermal cycling, the spallation life was improved by a factor of 2.5 compared to APS coatings on the same bond coat and substrate, and by a factor of 1.5 compared to high-properties EB-PVD coatings. The apparent thermal conductivity, as measured by laser-flash technique from 100 to 1,000 °C, has been found to be approximately 1.0–1.2 W m−1 K−1, a value lower than that of EB-PVD coatings, but higher than that of conventional APS coatings. Probably this high conductivity (relative to APS coatings) is due to the increased internal contact area of solution-sprayed coatings. The thermal cyclic stability of these coatings was found to show no significant microstructural or phase changes during 1,090 cycles of 1 h each at 1,121 °C. The critical features of solution coatings, vertical cracks, and ultrafine splats, remained stable throughout the test. Vertical cracks were retained or reformed, even after exposure to 1,500 °C. APS and solution-sprayed TBCs exhibited similar grain growths, density and hardness increase between 1,200 and 1,400 °C. Above 1,400 °C, APS TBCs exhibited a faster rate of grain growth and transformation into monoclinic phase. The microstructural features of these coatings are retained in very thick coatings (up to 4 mm), the spallation lives being significantly less thickness sensitive and demonstrating superior lives.

(b) Suspensions

Work has also been performed with suspensions (suspension plasma spraying, SPS, and high velocity suspension flame spraying, HVSFS) [9, 183, 190, 210]. For example, Kassner et al. [190] have obtained segmented coatings (7 cracks.mm-1) for 300-μm thick coatings obtained by plasma spraying. Ben-Ettouil et al. [240] nevertheless showed that resistances to furnace cycle test (FCT) of YPSZ SPS coatings were higher for coatings exhibiting lower crack density. Besides, the thermal diffusivity of such as-sprayed coatings has been measured at atmospheric pressure to change from 0.015 to 0.025 mm2/s for a temperature rise from room temperature to 250 °C. Such low values, about 10 times lower than values commonly measured on coatings exhibiting dual architectures (nano- and micro-sized) manufactured with micro-sized agglomerates made of nano-sized particles, can be explained in a first approximation by the peculiar void network architecture (80 % of voids smaller than 30 nm) of such coatings. This architecture induces, besides phonon scattering phenomena in the smallest grains of the coating structure, numerous thermal resistances in the structure, each thermal resistance being increased by rarefaction effects due to small grain volume [239]. Besides, such nano-sized coatings exhibit enhanced optical absorption and strong IR scattering compared to conventional micro-sized YPSZ coatings, due to the number and size of the voids, and due to the density of interfaces in between grains and lamellae forming such coatings [239].

14.7.7.2 Solid Oxy-Fuel Cell Components

Numerous works are in progress in the world for the SOFC [172, 180, 188, 192, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254] For the cathode and electrolyte, Kassner et al. [190] have plasma sprayed, using two injectors and a Triplex torch, LSM ((La0.65Sr0.3)MnO3) and YPSZ to produce a functional layer for SOFC cathodes. Works have also been devoted to deposition of LaMnO3 perovskite thin films by suspension plasma spraying [247].

According to the small particle size, such coatings should provide a higher amount of triple phase boundaries. The cathode manufacturing by thermal spraying is challenging because LaMnO3 doped with SrO, CaO, etc., is prone to decompose easily above the melting temperature. Monterubio et al. [247] have shown that pure LaMnO3 (the easiest to decompose) can be sprayed, with less than 10 % of La2O3 formation, if the feedstock particles were doped with 10 mol.% of MnO2. In this condition, particles with sizes between 1 (d10) and 5 (d90) μm (mean size d 50 of 3 μm) were sprayed with a pure Ar d.c. plasma jet to limit, as much as possible, the heat transfer, and consequently the particle temperature. A La0.8Sr0.2MnO3 (LSM) [250] cathode was prepared and its properties were strongly influenced by the spray distance.

Thin (~ 10–20 μm) electrolyte coatings of Y-PSZ by SPS are supposed to reduce Ohmic losses, but these layers must be impervious to gases. Efforts were mainly devoted to the study of the influence of various solid loadings in suspensions together with their state of dispersion, the stability of the suspension (by adjusting the amount of dispersant), the suspension tailored viscosity, the atomization effect (gas-to-liquid mass ratio (ALR), influence of Weber number), the effect of mechanical injection, the solid particle size distribution, and the manufacturing process [152, 171, 188, 190, 222, 231, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255]. By adapting the spray conditions, it became possible to achieve a leakage rate of 0.5 MPa L/s m2 for a 15μm thick electrolyte, which is not yet the requirement of 0.08 MPa L/s m2 [171].

With the Prosol process [21] (sol suspension), the structure and the proportion of the crystalline phases in the coatings are typically the same as those encountered in the injected sol. Coatings are dense when the sol is processed with high enthalpy plasma flows (20 MJ/kg) and consist mainly of very fine particles. Bouaricha et al. [249] sprayed, with the Axial II torch, suspension of nano-sized particles of samarium-doped ceria (SDC) in order to produce thin electrolyte layers for SOFC applications with reduced both Ohmic losses and operating temperature. Original phase composition was retained and, independently of thermal spray conditions, the SDC coatings retained the nanostructure of the sprayed feedstock. Depending upon plasma spray parameters and substrate temperature, the process can produce a wide variety of structures ranging from porous to dense coatings, with very low open void content.

A method for manufacturing metal-supported SOFCs with atmospheric plasma spraying (APS) is presented in [253], making use of aqueous suspension feedstock for the electrolyte layer and dry powder feedstock for the anode and cathode layers. Even with optimized spray parameters, three main defect types were detected in such suspension plasma sprayed electrolyte layers: (1) through-thickness cracks (perpendicular to the substrate surface) caused by thermal stresses developing during coating deposition, (2) voids or defects caused by unmelted particles embedded within the coating, and (3) small inter-splat voids. Three different types of dispersant were also considered for ensuring feedstock dispersion and the effects of solid particles loading, dispersant type, and dispersant concentration on suspension properties such as viscosity and feed ability, and layer characteristics such as microstructure and deposited thickness, were also examined [253].

Ni–YSZ anode and YSZ electrolyte half cells were successfully deposited by SPS with the Axial III torch from Mettech on porous Hastelloy X substrates by Wang et al. [192]. The thickness of the anode and electrolyte layers was ~40 μm and 10–20 μm, respectively. The NiO–YSZ coating deposition process was optimized by a design of experiment approach. The YSZ electrolyte spray process was examined by changing one parameter at a time. The maximum power density reached 0.610 W/cm2 at 750 °C.

14.7.7.3 Wear-Resistant Coatings

(a) WC–Co

WC–Co coatings were deposited using an axial injection plasma torch and an Et-OH suspension of sub-micrometer-sized feedstock powders [256]. With a proper selection of spray parameters, porosities below 0.2 % were achieved with hardness over 700 HV3N. The best coatings corresponded to the highest particle velocities (~800 m/s) and the lowest particle temperatures (at the minimum, 2,200 °C). It seems that the measured carbon loss in the carbide is principally associated to deposition of layers of highly oxidized overspray. The properties of the feedstock powder and feed suspension play a critical role in the resulting coatings quality.

(b) Alumina

Toma et al. [257] provided an overview of suspension-sprayed Al2O3 coatings. They sprayed alumina coatings by using SPS and HVSFS (axial injection) spraying [258]. The initial mean particle diameter was 0.3 μm and either water or Et-OH was used for the suspension. Et-OH base suspension spraying with Ar–H2 plasma-forming gas mixture resulted in 18.9 ± 2.5 % porosity against 11.6 % with an Ar–He one. With HVSFS (acetylene–oxygen) and aqueous suspension, the porosity was 7.6 ± 2.8 %. Toma et al. in another work [233] made a comparative study of the characteristics (microstructure, phase composition, and mechanical properties) and electrical properties (insulating resistance, electrical resistivity, and dielectric strength) of Al2O3 coatings produced by conventional HVOF (fed with micrometer-sized particles) and HVFSF fed with a suspension of particles, which size distribution was between 0.4 and 3.5 μm. At low levels of relative humidity, RH < 40 %, the electrical resistivity was on the order of 1011 Ω.m for both coatings. At a very high humidity (97 % RH) the electrical resistivity values for the HVFSF coatings were in the range 107–1011 Ω.m, up to five orders of magnitude higher than those recorded for the HVOF coating (106 Ω m). The specific microstructure of the suspension-sprayed coating is probably due to its much lower pore sizes. However, at RH = 97 % the electrical properties were found to degrade significantly (by up to 4–5 orders of magnitude) in these coatings after long-term exposure. The dielectric strength of suspension-sprayed coatings was found to be 19.5–26.8 kV/mm for coating thicknesses ranging from 60 to 200 μm, which is probably due to its microstructure.

Darut et al. [259] compared the tribological properties of alumina coatings structured at two different scales, a micrometer one (d 50 of particle size distribution of 36 μm) manufactured by APS and a sub-micrometer one (d 50 of particle size distribution of 0.4 μm) manufactured by SPS. Coating architectures were analyzed and their friction coefficient and wear rate in dry sliding mode measured in ball-on-disk friction conditions (α-Al2O3 ball of 6 mm in diameter). The friction coefficient of Al2O3 coatings was decreased by two (0.4 for SPS layers to be compared to 0.8–0.9 for APS ones) when the characteristic scale was divided by two orders of magnitude. Moreover, wear rate was 30 times lower for SPS layers compared to the one of APS layers, as shown in Fig. 14.82.
Fig. 14.82

Wear profiles for α-Al2O3 micrometer-sized (APS) and sub-micrometer-sized (SPS, spray distance = 30 and 40 mm) structural scales (ball-on-disk friction test, α-Al2O3 ball of 6 mm in diameter, 2 N load, relative speed of 0.1 m/s and sliding distance of 1,500 m. The test was operated in the dry mode and the wear scraps were constantly removed by an air jet located behind the contact point). Reprinted with kind permission from Springer Science Business Media [259], copyright © ASM International

Qiu and Chen [260] manufactured by suspension plasma spraying nanostructured fine alumina coating using nano-scale alumina powders (d ~ 20 nm), ethanol being used as liquid phase. The quality of the coatings strongly depended upon the plasma power, spray distance, liquid phase, solid phase concentration, inner diameter of the suspension injector, and suspension flow rate.

Nanostructured Al2O3–13 wt% TiO2 coatings were manufactured [261, 262] by plasma spraying with nano-crystalline powders. The TEM analysis revealed that partially melted Al2O3 particles, in the size range of 20–70 nm, were embedded in a TiO2-rich matrix. The mechanism of the substructure formation was explained in terms of melting and flattening behaviors of the powder particles during plasma spray processing.

Oliker et al. [263] found that SPS alumina coatings exhibited better hardness and mainly toughness than conventionally sprayed coatings.

(c) Zirconia–Alumina

Nanometer-sized composites made of alumina–zirconia can potentially exhibit higher hardness, fracture toughness, slower crack growth, and lower thermal conductivity than alumina or zirconia alone. Both solution [199] and suspension spraying [11, 175, 264, 265, 266] were used. Gadow et al. [11] have sprayed by HVSFS a mixture of Al2O3 (150 nm) and Y-PSZ (12 nm) dispersed in 30/70wt% isopropanol/water mixtures with a 20wt% solid content. The coatings reveal that nanometer-sized particles of zirconia are embedded in an alumina matrix. The spray conditions were tailored to achieve complete alumina melting with the zirconia particles remaining solid. As the time between melting and reconsolidation is extremely short, the two components are almost immiscible. Oberste-Berghaus et al. [175] have also deposited Al2O3–ZrO2 coatings by SPS and HVSFS, both with axial injection. With optimized spray conditions, the smaller nano-sized particles create fine laminates of alumina and zirconia layers, while slightly larger particles promote pseudo-alloyed alumina–zirconia amorphous phase components. A large variety of microstructures and properties of suspension-sprayed Al2O3–ZrO2 composites was introduced by the different feedstock, spray systems and operating conditions.

Also an alumina/8 wt% YPSZ was deposited by axial injection SPS process by Tarasi et al. [177]. The effects of main deposition operating parameters on the microstructural features were evaluated using the Taguchi design of experiment. The results indicate that thermal diffusivity of the coatings, an important property for potential thermal barrier applications, is barely affected by the changes in porosity content during annealing treatments. At last, Chen et al. [264] have sprayed with a rod-type cathode d.c. plama torch an Al2O3–ZrO2 amorphous powder feedstock [19] the preparation of which was described in Sect. 14.7.4.4.

Solution plasma spraying was used to spray 75 % dense coatings of metastable ceramics in the ZrO2–10 mol.% Al2O3 binary system [199]. The coating microstructure contains a rich variety of features and phases. Coatings were predominantly nanostructured (grain size < 100 nm) and made of tetragonal phase of ZrO2 with Al3+ in solid solution.

Tingaud et al. [265] manufactured by suspension plasma spraying (SPS) several alumina (d 50 = 0.6 μm) and alumina–zirconia (YSZ d 50 = 0.4 μm) composite coatings, varying the operating conditions to achieve dense and cohesive structures. For alumina they measured particle temperatures around 2000 °C and velocities as high as 450 m/s, resulting in coatings with high amount of α-phase. The preliminary results of the SPS wear behavior show rather low friction coefficient and good wear resistance, which is even better with the addition of ZrO2 in the alumina matrix.

Darut et al. [259] showed that the Al2O3-ZrO2 composite coatings showed higher wear resistance than the Al2O3 coatings.

(d) Alumina–Zirconia–Yttria

Solution plasma spraying [201] was used to process the ternary systems (mol%) 10Al2O3–86.4ZrO2–3.6Y2O3 and 20Al2O3–76.4ZrO2–3.6Y2O3. The nanostructures (10–40 nm) in both coatings were made predominantly of tetragonal ZrO2 phase with some cubic phase. The sub-micrometer-sized regions contain, in addition to t-ZrO2, small amounts of crystalline Al2O3 phases. Thus solution plasma spraying allows producing metastable ceramics with unusual structures.

(e) TiO2–TiC Coatings

Manzat et al. [266, 267] have used HVSFS to spray TiO2/TiC coatings for automotive engine applications. The suspension was composed of a nano-sized titania and a sub-micron-sized TiC dispersed in isopropanol. They succeeded to achieve spraying onto the inner surfaces of light metal and gray cast cylinder liners. Tests performed under dry sliding and lubricated conditions demonstrated an improved wear resistance compared to gray cast iron.

14.7.7.4 Titania Photo-Catalytic Coatings

Titania photo-catalytic layers were sprayed by suspension plasma spraying [268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282] or HVSFS [183, 283, 284, 285]. TiO2 coatings are mostly designed for photo-catalytic applications. As pointed out by Toma et al. [207, 272, 273, 280], it is generally assumed that the metastable phase, anatase, presents higher photo-catalytic activity than the stable one, rutile. Coatings sprayed with Et-OH base suspensions contained only 23 % of anatase ratio (against 82 vol.% in the feedstock powder) and ensured a very low photo-catalytic decomposition of nitrogen oxides. In contrast, with water base suspensions, the anatase phase and crystallites size were preserved and the conversion rate of pollutant reached 40 % against 32 % for the starting powders [274].

Besides, Tomaszek et al. [275] measured the field emission characteristics of the obtained TiO2 deposits. The influence of conditioning on emission properties was observed. The lighting segment excited by electrons from the field titania cathode has also been studied by Znamirowski et al. [276]. The titania cathode of the segment was manufactured with suspension plasma spraying. Operating tests reveal effective lighting properties of the segment, with very intense light emitted.

Kozerski et al. [277] studied the influence of the suspension solvent and injection on photo catalytic properties of plasma-sprayed titania suspension. The main conclusion of the study is that the coatings containing mainly rutile are also photo catalytically active for the degradation of aqueous solution of dye methylene blue. The photo-catalytic activity was not found to be correlated with the anatase content; thus further studies should clarify the role of suspension plasma spraying processing parameters, substrate material, suspension formulation, and coating microstructure (porosity, content of hydroxyl groups on the coatings surface, gap energy of the coatings, etc.) on the photo-catalytic activity. Bannier et al. [278] prepared TiO2 coatings by suspension plasma spraying from a TiO2 nanoparticle suspension using two different substrates (a standard stainless steel and a Pyrex glass). They studied the influence of spray parameters and plasma torch design. A large amount of anatase was obtained, ranging from 32 to 72 wt%. Coatings displayed a bimodal microstructure with completely fused areas and non-molten ones with agglomerated anatase nanoparticles. The main parameters controlling this bimodal structure were the spray distance and the cooling. Yuan et al. [279] flame sprayed mixtures of suspensions and powder feedstock. They tailored over a wide range the anatase–rutile ratio and the proportion of nanostructures in the coatings, and up to 30 wt% of anatase in the coatings was obtained. The hybrid micro-/nanostructured TiO2 coatings exhibited super-hydrophilic performances (0° contact angle). Results were closely linked to the concentration of nano TiO2 particles (10–30 nm in size) in the starting suspension.

The HVSFS process leaves a fairly large degree of freedom to tailor TiO2 coating characteristics (thickness, porosity, anatase content, hardness, etc.) according to the required functional properties by adjusting operating spray parameters [283]. For example, coatings with higher anatase content and higher specific surface can be produced to achieve higher photo catalytic efficiency, better than conventional APS and HVOF TiO2.

Porous titania coatings were plasma sprayed from an aqueous solution containing titanium iso-propoxide [286]. It was shown that when increasing the power level, the anatase phase content decreased and the rutile one increased. In fact, the formation of crystalline anatase and rutile in the as-sprayed coatings was the result of plasma flow heat treatment on the gel like deposit (many non-pyrolized droplets).

14.7.7.5 Coatings for Medical Applications

Bouyer et al. [185] were the firsts to produce by r.f. plasma suspension spraying micrometer-sized Hydroxyapatite (HA) particles that were easy to spray by conventional means. Since then, many efforts were devoted to suspension or solution spraying of HA [287, 288, 289, 290, 291, 292, 293]. The development of multilayer coatings of hydroxyapatite (HA) and TiO2 onto titanium substrates was later studied using suspension plasma spraying [287]. Jaworski et al. [271] have presented the recent developments in suspension d.c. plasma-sprayed HA coatings with rather low thickness (10–50 μm). Two types of coatings were tested: duplex and gradient coatings. The density of coatings was improved as well as their cohesion when increasing the torch power level, but unfortunately the HA was partially decomposed.

Łatka et al. [288] plasma sprayed homemade HA powder. The coatings presented two types of zones: (i) dense ones corresponding to splats, as in conventionally sprayed coatings; (ii) sintered ones containing fine hydroxyapatite grains, corresponding to the fine solids from the initial suspension. The latter disappeared after soaking the coating in simulated body fluid (SBF), and the pores were filled by the re-precipitated calcium phosphates.

An ethylene glycol-based suspension was HVSFS processed [180] with 13wt% of HA powder, resulting in a dense coating structure with a porosity of about 1 % (coating thickness 50 μm). The coatings consisted of about 90 % of HAP with a small amount of tricalcium phosphate.

Stiegler et al. [293] used the HVSFS technique to deposit HA coatings onto titanium plates. The use of different dispersion media to prepare the HA suspensions affected the microstructure and mechanical properties of the resulting coatings more than the process parameters. However, coatings containing substantial amounts of amorphous phase tended to dissolve very rapidly in the SBF solution and to be replaced by a precipitation layer consisting of crystalline HA.

A phosphorous containing biocompatible glass was also HVSFS processed [294]: coatings displayed a certain porosity that seemed to consist mainly of closed pores with some transverse micro-cracks.

The high-velocity suspension flame spraying technique (HVSFS) was used by Altomare et al. [295] to deposit 45S5 bioactive glass coatings onto titanium substrates, using a suspension of micron-sized glass powders dispersed in a water–isopropanol mixture as feedstock. Depending on the spray parameters, coatings with different porosities and thicknesses were obtained. The sprayed coatings were entirely glassy but exhibited a through-thickness microstructural gradient and the glass network structure was analogous to that of bulk annealed 45S5 bioglass. The interaction mechanisms between coatings and the simulated body fluid (SBF) were also analogous to those of the bulk bioglass. They involved controlled dissolution of the glass, polymerization of an amorphous silica layer on the glass surface itself, and growth of a carbonated hydroxyapatite layer on top of the silica layer. Thus it seems that bioglass poses no trouble related to decomposition and phase alteration and the reactivity of the HVSFS-deposited bioglass coatings seems to be particularly fast, most of the original coating being replaced by the products of the interaction with the simulated body fluid (SBF) in only one week. Xiao et al. [296] have plasma sprayed P2O5–Na2O–CaO–SiO2 bioactive glass–ceramic coatings (BGCCs), where sol and suspension were used as feedstock. Results indicated that coatings with higher crystallinity were obtained using the sol precursor, while nanostructured coatings predominantly consisting of amorphous phase were synthesized using the suspension precursor. Overall, the liquid precursor plasma spraying process seems to be a promising technique to synthesize nanostructured BGCCs with good in vitro bioactivities. The same authors [297] plasma sprayed a solution to produce the same BGCC on a Ti substrate. The as-deposited coating consisted of two phases, with a predominant amorphous phase and a crystalline phase of Na2Ca2Si3O9. The in vitro bioactivity of the as-deposited coating was studied by SBF soaking. Results suggest that this coating has quick hydroxide carbonate apatite (HCA) formation capability in vitro, likely due to the existence of nanostructure in the as-sprayed coatings.

14.7.7.6 Anti-Corrosion Coatings

A CaO–ZrO2–SiO2 glass frit was processed by HVSFS [298]. Such glass, if dense, should have excellent anticorrosion properties on metal or ceramic substrates. The coatings obtained were much denser than those formerly investigated and plasma sprayed. Some problems were connected to the formation of molten glass deposits in the combustion chamber and their subsequent embedment in the coating. Work is in progress to overcome this problem and to achieve dense coatings with strong inter-lamellar cohesion.

Waltz et al. [299] have developed a synthesis for the preparation of nano-crystalline magnesium fluoride suspensions, which delivers nearly mono-disperse nanoparticles. These suspensions were deposited with a plasma torch onto TIG welded seams of the magnesium alloy AZ31, thus producing a protective magnesium fluoride layer. The corrosion properties of the magnesium fluoride layer on the welded seams were studied by means of potentiodynamic potential measurements and showed a significant lower corrosion rate.

14.7.7.7 SnO2 Layers

An aqueous SnCl4 solution feedstock was used to integrate the spray pyrolysis and sintering of nanometer-sized SnO2 particles in nanometer-sized porous coating production [300]. The coating consisted of interpenetrating pores separating agglomerated particles which themselves and exhibited a layered structure with micropores between the layers. Each layer was comprised of SnO2 nanometer-sized particles. The structure was apparently formed from overlapping viscous hollow spheres of Sn solution, which were pyrolyzed and annealed in situ. Sensors constructed directly from the porous coatings have good ethanol gas sensitivity at 200 °C.

14.7.7.8 Inconel Coatings

A HVSFS process has been developed to conduct spraying of small Inconel alloy 625 particles in suspension [301]. The coatings exhibited several interesting characteristics for potential applications, including full density, uniform microstructure, and high bond strength. Unfortunately, the preliminary erosion tests, carried out at 90° impact angle, indicated that the conventional HVOF coating was superior to the HVSFS coatings.

14.7.7.9 Antierosion Coatings

Kitamura et al. [302] have sprayed yttrium oxide (Y2O3) coatings by axial suspension plasma spraying with fine powders (d 50 = 1 μm). Coatings had high hardness (640 HV1.96N), low porosity (0.3-0.4 %), high erosion resistance (about 1.5 times that of bulk yttria) against CF4-containing plasma and retention of smooth eroded surfaces. This suggested that the axial suspension plasma spraying of Y2O3 is applicable to fabricating equipment for electronic devices, such as dry etching.

14.7.7.10 Adhesive Layer on Smooth and Thin Substrate

Vert et al. [303] have studied the use of suspension plasma spraying to manufacture 1-mm thick zirconia coatings on smooth and thin (1 mm) substrates. The spraying conditions were optimized for the deposition of coatings on stainless steel AISI 304L substrates, and the process then adapted for Haynes 230 nickel-based alloy substrates. Special attention was paid to coating adhesion that was investigated by using a Vickers indentation cracking method. The adhesion of various coatings was studied and the effect of the substrate temperature on adhesion confirmed.

14.8 Summary and Conclusions

For about a decade, the interest for manufacturing on large surfaces “thick” finely structured or nanostructured layers have been increasingly growing. If gas condensation routes (CVD, PE-CVD, PVD, EB-PVD, etc.), allow manufacturing of nanostructured architectures, their thicknesses can hardly be higher than a few micrometers, except for EB-PVD. In contrast, conventional plasma spray coating thicknesses between 50 μm to a few millimeters are easily achieved but with no real nanostructured architectures after particle melting.

This explains the interest for the different techniques developed this last decade:
  • Spraying with hot gases of amorphous metals, agglomerated ceramic particles, attrition or ball milled cermet or alloy particles, hypereutectic alloys

  • Production of nanostructured coatings by cold spray

  • Suspension thermal spraying (STS) and solution precursor thermal spraying (SPTS), both allowing achieving finely structured layers of thicknesses varying between a few micrometers up to a few hundreds of micrometers

These techniques seem promising to manufacture dense or porosity controlled coatings and functionally graded layers.

Some industrial applications are already using coatings made from powders with a size range in the tens of micrometers, especially those from amorphous powders. Nevertheless, coatings made from suspensions or solutions injected into hot gases, are by far more complex because the chemistry (choice of solvent, dispersant, precursors,…), the liquid fragmentation and vaporization control the processes. This is probably why publications on first commercial applications are still missing. Moreover, presently diagnostic techniques are not available to measure in-flight droplets below 5 μm, which would permit getting a much better understanding of the phenomena involved. According to all papers published during the last decade, and also the beginning of the new one, the interaction plasma jet and liquid jet or drops is crucial for their fragmentation to sizes below around 10 μm and velocities imparted to particles contained in droplet suspensions or formed from solutions. However to optimize the process first the injection means of the liquid must be improved and new plasma torches developed. Injection setups producing liquid jets or drops with controlled narrow trajectories dispersion, narrow distributions of sizes (if possible around or below 50 μm), and velocities, both varying independently must be developed. To spray them, plasma torches with higher power levels (more than 100 kW), longer plasmas and low voltage fluctuations must be designed. The only one available now in the Axial III of Mettech. At last, the deposition rate, which is about 10–30 % that of conventional coatings, should also be improved. For suspensions HVOF spraying guns, called HVSFS, with longer combustion chambers have been designed to achieve combustion of both combustion gases and organic solvent. Dense and very smooth (Ra ~ 1.3 μm) coatings, consisting of well-flattened lamellae having a homogeneous size distribution, were obtained when micron-sized (~1–2 μm) powders. Both for plasma or HVSFS suspension spraying, the size distribution of particles used in suspension should be as narrow as possible. Mixtures in suspensions with a wide size distribution because of agglomerates and aggregates result in poor coatings. Thus particles having the tendency to agglomerate or aggregate must be avoided.

Thus it could be said that SPS and SPPS are in the same state as conventional spraying was in the 1980s and the beginning of the 1990s. As conventional spraying in the nineties, SPS and SPPS, for the last decade are the subjects of numerous papers in international reviews and conference proceedings. Coatings obtained (mostly ceramics) present very interesting features, especially a much better toughness than conventional coatings.

Numerous studies are still necessary to reach a better understanding of the involved phenomena and for that, the development of diagnostic techniques to either quantify temperature and velocity of the small (between a few tens of nanometers and a few tenths of micrometers) droplets or particles in flight, or to visualize their interactions with hot gases must be improved or newly developed.

No standards have been yet defined both for the spray process and solution or suspension preparation. It is not surprising that, to our best knowledge, no industrial developments of these techniques have been yet mentioned. If many potential applications have been described it is very difficult to predict which one will be the first industrially developed. However, a spray equipment manufacturer has recently proposed an industrial liquid-based feedstock feeder prototype that might be an indication of the potential interest of industry.

As pointed out by Killinger et al. [9]: “to promote industrial implementation, an urgent concern of future research work is to qualify liquid feedstock thermal spray coatings in a more application-oriented way and to communicate the results convincingly. As suspension and solution spray are to be clearly differentiated from classical thermal spray processes, also new industrial standards have to be defined concerning suspension and solution preparation, storage, feeding as well as qualification and handling of raw materials (i.e., ultra-fine powders, solvents, additives, etc.)”.

References

  1. 1.
    McPherson R (1973) Formation of metastable phases in flame and plasma-prepared alumina. J Mater Sci 8:851–858Google Scholar
  2. 2.
    Ashby MF, Ferreira PJ, Schodek DL (2009) Nanomaterials properties. In: Nanomaterials, nanotechnologies and design. Elsevier, Kidlington, pp 199–255Google Scholar
  3. 3.
    Dahotre NB, Nayak S (2005) Nanocoatings for engine application. Surf Coat Technol 194(1):58–67Google Scholar
  4. 4.
    Lu Y, Liaw PK (2001) The mechanical properties of nanostructured materials. J Metals 53(3):31–35Google Scholar
  5. 5.
    Fauchais P, Montavon G, Lima RS, Marple BR (2011) Engineering a new class of thermal spray nano-based microstructures from agglomerated nanostructured particles, suspensions and solutions: an invited review. J Phys D 44:093001Google Scholar
  6. 6.
    Branagan DJ, Swank WD, Haggard DC, Fincke JR (2001) Wear-resistant amorphous and nanocomposite steel coatings. Metall Mater Trans A 32A:2615–2621Google Scholar
  7. 7.
    Branagan DJ, Swank WD, Meacham BE (2009) Maximizing the glass fraction in iron-based high velocity oxy-fuel coatings. Metall Mater Trans A 40A:1306–1313Google Scholar
  8. 8.
    Guignard A, Mauer G, Vaßen R, Stöver D (2012) Deposition and characteristics of submicrometer-structured thermal barrier coatings by suspension plasma spraying. J Therm Spray Technol 21(3–4):416–424Google Scholar
  9. 9.
    Killinger A, Gadow R, Mauer G, Guignard A, Vaßen R, Stöver D (2011) Review of new developments in suspension and solution precursor thermal spray processes. J Therm Spray Technol 20(4):677–695Google Scholar
  10. 10.
    Viswanathan V, Laha T, Balani K, Agarwal A, Seal S (2006) Challenges and advances in nanocomposite processing techniques. Mat Sci Eng R54:121–285Google Scholar
  11. 11.
    Gadow R, Kern F, Killinger A (2008) Manufacturing technologies for nanocomposite ceramic structural materials and coatings. Mat Sci Eng B148:58–64Google Scholar
  12. 12.
    Fauchais P, Montavon G, Vardelle M, Cedelle J (2006) Developments in direct current plasma spraying. Surf Coat Technol 201(5):1908–1921Google Scholar
  13. 13.
    Fauchais P, Etchart-Salas R, Rat V, Coudert J-F, Car N, Wittmann-Ténèze K (2008) Parameters controlling liquid plasma spraying: solutions, sols or suspensions. J Therm Spray Technol 17(1):31–59Google Scholar
  14. 14.
    Branagan DJ, Breitsameter M, Meacham BE, Belashchenko V (2005) High-performance nanoscale composite coatings for boiler applications. J Therm Spray Technol 14(2):196–204Google Scholar
  15. 15.
    Lima RS, Marple BR (2007) Thermal spray coatings engineered from nanostructured ceramic agglomerated powders for structural, thermal barrier and biomedical applications: a review. J Therm Spray Technol 16(1):40–63Google Scholar
  16. 16.
    Ben-Ettouil F, Mazhorova O, Pateyron B, Ageorges H, El-Ganaoui M, Fauchais P (2008) Predicting dynamic and thermal histories of agglomerated particles injected within a d.c. plasma jet. Surf Coat Technol 202:4491–4495Google Scholar
  17. 17.
    Ajdelsztajn L, Jodoin B, Kim GE, Schoenung JM (2005) Cold spray deposition of nanocrystalline aluminum alloys. Metall Mater Trans A 36A:657–666Google Scholar
  18. 18.
    Fauchais P, Rat V, Coudert J-F, Etchart-Salas R, Montavon G (2008) Operating parameters for suspension and solution plasma-spray coatings. Surf Coat Technol 202:4309–4317Google Scholar
  19. 19.
    Chen D, Jordan EH, Gell M (2009) Microstructure of suspension plasma spray and air plasma spray Al2O3-ZrO2 composite coatings. J Therm Spray Technol 18(3):421–426Google Scholar
  20. 20.
    Basu S, Jordan EH, Cetegen BM (2008) Fluid mechanics and heat transfer of liquid precursor droplets injected into high-temperature plasmas. J Therm Spray Technol 17(1):60–72Google Scholar
  21. 21.
    Wittmann-Ténèze K, Vallé K, Bianchi L, Belleville P, Caron N (2008) Nanostructured zirconia coatings processed by PROSOL deposition. Surf Coat Technol 202:4349–4354Google Scholar
  22. 22.
    Sergueeva AV, Branagan DJ, Mukherjee AK (2008) Microstructure/properties relationship in Fe-based nanomaterials. Mater Sci Eng A 493:237–240Google Scholar
  23. 23.
    Peker A, Johnson WL (1993) A highly processable metallic-glass Zr41.2Ti13.8Cu12.5Ni10.0Be22.5. Appl Phys Lett 63:2342–2344Google Scholar
  24. 24.
    Inoue A (1995) Thermal properties of Zr-TM-B and Zr-TM-Ga (TM = Co, Ni, Cu) amorphous alloys with wide range of supercooling. Mater Trans JIM 36:1411–1419Google Scholar
  25. 25.
    Sergueeva AV, Mara NA, Kuntz JD, Branagan DJ, Mukherjee AK (2004) Shear band formation and ductility of metallic glasses. Mater Sci Eng A 383(2):219–223Google Scholar
  26. 26.
    Otsubo F, Era H, Kishitake K (2000) Formation of amorphous Fe-Cr-Mo-8P-2C coatings by the high velocity oxy-fuel process. J Therm Spray Technol 9(4):494–498Google Scholar
  27. 27.
    Das SK, Norin EM, Bye RL (1984) Ni-Mo-Cr-B alloys: corrosion resistant amorphous hardfacing coatings. In: Kear BH, Cohen M (eds) Materials research society symposium proceedings. Elsevier Science, Amsterdam, pp 233–237Google Scholar
  28. 28.
    Sampath S, Neiser RA, Herman H, Kirkland JP, Elam WT (1993) A structural investigation of a plasma sprayed Ni-Cr based alloy coating. J Mater Res 1:78–86Google Scholar
  29. 29.
    Kishitake K, Era H, Otsubo F (1996) Characterization of plasma sprayed Fe-10Cr-10Mo-(C, B) amorphous coatings. J Therm Spray Technol 5(2):145–153Google Scholar
  30. 30.
    Kishitake K, Era H, Otsubo F (1996) Characterization of plasma sprayed Fe-17Cr-38Mo-4C amorphous coatings crystallizing at extremely high temperature. J Therm Spray Technol 5(3):283–288Google Scholar
  31. 31.
    Dent AH, Horlock AJ, McCartney DG, Harris SJ (1999) The corrosion behavior and microstructure of high-velocity oxy-fuel sprayed nickel-base amorphous/nanocrystalline coatings. J Therm Spray Technol 8(3):399–404Google Scholar
  32. 32.
    Choi H, Kim J, Lee C, Lee KH (2005) Critical factors affecting the amorphous phase formation of NiTiZrSiSn bulk amorphous feedstock in vacuum plasma spray. J Mater Sci 40:3873–3875Google Scholar
  33. 33.
    Branagan DJ, Kramer MJ, Tang Y, McCallum RW, Crew DC, Lewis LH (2000) Engineering magnetic nanocomposite microstructures. J Mater Sci 35:3459–3466Google Scholar
  34. 34.
    Branagan DJ (2007) Engineering structures to achieve targeted properties in steels on a nanoscale level. Comput Coupling Phase Dia Thermochem 31:343–350Google Scholar
  35. 35.
    Branagan DJ, Marshall MC, Meacham BE, Aprigliano LF, Bayles R, Lemieux EJ, Newbauer T, Martin FJ, Farmer JC, Haslam JJ, Day SD (2006) Wear and corrosion resistant amorphous/nanostructured steel coatings for replacement of electrolytic hard chromium. In: Marple B (ed) Proceedings of the 2006 international thermal spray conference, e-procceedings. ASM International, Materials Park, OHGoogle Scholar
  36. 36.
    Parco M, Fagoaga I, Bobzin K, Lugsheider E, Zwick J, Hildago G. Developement and characterization of nanostructured iron-based coatings by HFPD and HVOF. In: Marple B (ed) Proceedings of the 2006 international thermal spray conference, e-proceedings. ASM International, Materials Park, OHGoogle Scholar
  37. 37.
    Zhou J, Walleser JK, Meacham BE, Branagan DJ (2010) Novel in situ transformable coating for elevated-temperature applications. J Therm Spray Technol 19(5):950–957Google Scholar
  38. 38.
    Wilden J, Bergmann JP, Reich S, Schlichting S, Schnick T (2007) Cladding of aluminum substrates with nano crystalline solidifying wear resistant iron-based materials. In: Marple BR, Hyland MM, Lau Y-C, Li C-J, Lima RS, Montavon G (eds) Thermal spray 2007: global coating solutions. ASM International, Materials Park, OH, e-proceedingsGoogle Scholar
  39. 39.
    Shaw LL, Goberman D, Ren R, Gell M, Jiang S, Wang Y, Xiao TD, Strutt PR (2000) The dependency of microstructure and properties of nanostructured coatings on plasma spray conditions. Surf Coat Technol 130:1–8Google Scholar
  40. 40.
    Wang M, Shaw LL (2007) Effects of the powder manufacturing method on microstructure and wear performance of plasma sprayed alumina–titania coatings. Surf Coat Technol 202:34–44Google Scholar
  41. 41.
    Guru D, Heberlein J (2008) Plasma spray processing of nanostructured partially stabilized zirconia for a strain accommodating inter-layer – splat characteristics. In: Lugsheider E (ed) ITSC 2008: thermal spray crossing borders. DVS, Düsseldorf, Germany, e-proceedingsGoogle Scholar
  42. 42.
    Lima RS, Kucuk A, Berndt CC (2002) Bimodal distribution of mechanical properties on plasma sprayed nanostructured partially stabilized zirconia. Mater Sci Eng A A327:224–232Google Scholar
  43. 43.
    Lima RS, Marple BR, Dadouche A, Dmochowski W, Liko B (2006) Nanostructured abradable coatings for high temperature applications. In: Marple BR, Hyland MM, Lau Y-C, Lima RS, Voyer J (eds) Building on 100 years of success: proceedings of the international thermal spray conference 2006. ASM International, Materials Park, OH, e-proceedingsGoogle Scholar
  44. 44.
    Fauchais P (1991) Molten particle deposition. In: Schneider SJ Jr (ed) Engineered materials handbook, vol 4 – ceramics and glasses. ASM International, Materials Park, OHGoogle Scholar
  45. 45.
    Lima RS, Marple BR (2008) Nanostructured YSZ thermal barrier coatings engineered to counteract sintering effects. Mater Sci Eng A 485:182–193Google Scholar
  46. 46.
    Shaw LL, Goberman D, Ren R, Gell M, Jiang S, Wang Y, Xiao TD, Strutt PR (2003) The dependency of microstructure and properties of nanostructured coatings on plasma spray conditions. Surf Coat Technol 130:1–8Google Scholar
  47. 47.
    Lima RS, Marple BR (2006) From APS to HVOF spraying of conventional and nanostructured titania feedstock powders: a study on the enhancement of the mechanical properties. Surf Coat Technol 200:3428–3437Google Scholar
  48. 48.
    Lima RS, Marple BR (2005) Superior performance of high-velocity oxyfuel-sprayed nanostructured TiO2 in comparison to air plasma-sprayed conventional Al2O3-13TiO2. J Therm Spray Technol 14(3):397–404Google Scholar
  49. 49.
    Ahmed I, Bergman TL (1999) Thermal modeling of plasma spray deposition of nanostructured ceramics. J Therm Spray Technol 8(2):315–322Google Scholar
  50. 50.
    Zhu Y, Huang M, Huang J, Ding C (1999) Vacuum-plasma sprayed nanostructured titanium oxide films. J Therm Spray Technol 8(2):219–222Google Scholar
  51. 51.
    Gell M, Jordan EH, Sohn YH, Goberman D, Shaw L, Xiao TD (2001) Development and implementation of plasma sprayed nanostructured ceramic coatings. Surf Coat Technol 146–147:48–54Google Scholar
  52. 52.
    Ok Chwa S, Klein D, Toma FL, Bertrand G, Liao H, Coddet C, Ohmori A (2005) Microstructure and mechanical properties of plasma sprayed nanostructured TiO2–Al composite coatings. Surf Coat Technol 194:215–224Google Scholar
  53. 53.
    Wang Y, Jiang S, Wang M, Wang Xiao STD, Strutt PR (2000) Abrasive wear characteristics of plasma sprayed nanostructured alumina/titania coatings. Wear 237:176–185Google Scholar
  54. 54.
    Jiang X-L, Jordan E, Shaw L, Gell M (2003) Mechanism of deformation of nanostructured ceramic coatings. Mater Sci 39(2):305–306Google Scholar
  55. 55.
    Song EP, Ahn J, Lee S, Kim NJ (2006) Microstructure and wear resistance of nanostructured Al2O3–8wt.%TiO2 coatings plasma-sprayed with nanopowders. Surf Coat Technol 201:1309–1315Google Scholar
  56. 56.
    Ctibor P, Neufuss K, Chraska P (2006) Microstructure and abrasion resistance of plasma sprayed titania coatings. J Therm Spray Technol 15(4):689–694Google Scholar
  57. 57.
    Ahn Jeehoon, Byoungchul Hwang, Eun Pil Song, Sunghak Lee, Nack J Kim (2006) Correlation of microstructure and wear resistance of Al2O3-TiO2 coatings plasma sprayed with nanopowders. Metall Mater Trans A 37A:1851–1861Google Scholar
  58. 58.
    Song Eun Pil, Jeehoon Ahn, Sunghak Lee, Nack J Kim (2008) Effects of critical plasma spray parameter and spray distance on wear resistance of Al2O3–8 wt.%TiO2 coatings plasma-sprayed with nanopowders. Surf Coat Technol 202(15): 3625–3632Google Scholar
  59. 59.
    Lima RS, Marple BR (2005) Enhanced ductility in thermally sprayed titania coating synthesized using a nanostructured feedstock. Mater Sci Eng A 395:269–280Google Scholar
  60. 60.
    Lima RS, Marple BR (2000) From APS to HVOF spraying of conventional and nanostructured titania feedstock powders: a study on the enhancement of the mechanical properties. Surf Coat Technol 200:3428–3437Google Scholar
  61. 61.
    Liu Y, Fischer TE, Dent A (2003) Comparison of HVOF and plasma-sprayed alumina/titania coatings—microstructure, mechanical properties and abrasion behavior. Surf Coat Technol 167:68–76Google Scholar
  62. 62.
    Lima RS, Marple BR (2003) High weibull modulus HVOF titania coatings. J Therm Spray Technol 12(2):240–249Google Scholar
  63. 63.
    Lima RS, Marple BR (2003) Optimized HVOF titania coatings. J Therm Spray Technol 12(3):360–369Google Scholar
  64. 64.
    Turunen E, Varis T, Hannula S-P, Vaidya A, Kulkarni A, Gutleber J, Sampath S, Herman H (2006) On the role of particle state and deposition procedure on mechanical, tribological and dielectric response of high velocity oxy-fuel sprayed alumina coatings. Mater Sci Eng A 415:1–11Google Scholar
  65. 65.
    Lima RS, Moreau C, Marple BR (2007) HVOF-sprayed coatings engineered from mixtures of nanostructured and submicron Al2O3-TiO2 powders: an enhanced wear performance. J Therm Spray Technol 16(5–6):866–872Google Scholar
  66. 66.
    Varis T, Knuuttila J, Turunen E, Leivo J, Silvonen J, Oksa M (2007) Improved protection properties by using nanostructured ceramic powders for HVOF coatings. J Therm Spray Technol 16(4):524–532Google Scholar
  67. 67.
    Ibrahim A, Lima RS, Berndt CC, Marple BR (2007) Fatigue and mechanical properties of nanostructured and conventional titania (TiO2) thermal spray coatings. Surf Coat Technol 201:7589–7596Google Scholar
  68. 68.
    Lima RS, Kruger SE, Marple BR (2008) Towards engineering isotropic behaviour of mechanical properties in thermally sprayed ceramic coatings. Surf Coat Technol 202(15):3643–3652Google Scholar
  69. 69.
    Lin Xinhua, Yi Zeng, Xuebin Zheng, Chuanxian Ding (2005) Thermal diffusivity of plasma sprayed monolithic coating of alumina–3 wt.% titania produced with nanostructured powder. Surf Coat Technol 195:85–90Google Scholar
  70. 70.
    Chen Huang, Chuanxian Ding, Pingyu Zhang, Peiqing La, Soo Wohn Lee (2003) Wear of plasma-sprayed nanostructured zirconia coatings against stainless steel under distilled-water conditions. Surf Coat Technol 173:144–149Google Scholar
  71. 71.
    Shunyan T, Liang B, Ding C, Liao H, Coddet C (2005) Wear characteristics of plasma-sprayed nanostructured yttria partially stabilized zirconia coatings. J Therm Spray Technol 14(4):518–523Google Scholar
  72. 72.
    Turunen E, Varis T, Gustafsson TE, Keskinen J, Falt T, Hannula S-P (2006) Parameter optimization of HVOF sprayed nanostructured alumina and alumina-nickel composite coatings. Surf Coat Technol 200:4987–4994Google Scholar
  73. 73.
    Luo H, Goberman D, Shaw L, Gell M (2003) Indentation fracture behavior of plasma-sprayed nanostructured Al2O3-13wt%TiO2 coatings. Mater Sci Eng A 346:237–245Google Scholar
  74. 74.
    McPherson R (1989) A review of microstructure and properties of plasma sprayed ceramic coatings. Surf Coat Technol 39–40(1):173–181Google Scholar
  75. 75.
    Bansal P, Padture NP, Vasiliev A (2003) Improved interfacial mechanical properties of Al2O3-13wt%TiO2 plasma-sprayed coatings derived from nanocrystalline powders. Acta Mater 51:2959–2970Google Scholar
  76. 76.
    Ahn J, Hwang B, Song EP, Lee S, Kim NJ (2006) Correlation of microstructure and wear resistance of Al2O3-TiO2 coatings plasma sprayed with nano-powders. Metall Mater Trans A 37A:1851–1861Google Scholar
  77. 77.
    Goberman D, Sohn YH, Shaw L, Jordan EH, Gell M (2002) Microstructured development of Al2O3-13wt%TiO2 plasma sprayed coatings derived from nanocrystalline powders. Acta Mater 50:1141–1152Google Scholar
  78. 78.
    Ghasripoor F, Schmid R, Dorfman MR (1997) Abradable coatings increase gas turbine engine efficiency. Mater World 5(6):328–330Google Scholar
  79. 79.
    Lima RS, Kucuk A, Berndt CC (2001) Evaluation of microhardness and elastic modulus of thermally sprayed nanostructured zirconia coatings. Surf Coat Technol 135:166–172Google Scholar
  80. 80.
    Chen H, Ding CX (2002) Nanostructured zirconia coating prepared by atmospheric plasma spraying. Surf Coat Technol 150:31–36Google Scholar
  81. 81.
    Chen H, Zhou X, Ding C (2003) Investigation of the thermomechanical properties of a plasma-sprayed nanostructured zirconia coating. J Eur Ceram Soc 23:1449–1455Google Scholar
  82. 82.
    Chen H, Lee SW, Choi CH, Hur BY, Zeng Y, Zheng XB, Ding CX (2004) Plasma sprayed nanostrucutred zirconia coatings deposited from different powders with nano-scale substructure. J Mater Res 39:4701–4703Google Scholar
  83. 83.
    Liang B, Ding C (2005) Thermal shock resistances of nanostructured and conventional zirconia coatings deposited by atmospheric plasma spraying. Surf Coat Technol 197:185–192Google Scholar
  84. 84.
    Liang B (2005) Chuanxian ding, phase composition of nanostructured zirconia coatings deposited by air plasma spraying. Surf Coat Technol 191:267–273Google Scholar
  85. 85.
    Liang B, Ding C, Liao H, Coddet C (2006) Phase composition and stability of nanostructured 4.7wt.% yttria-stabilized zirconia coatings deposited by atmospheric plasma spraying. Surf Coat Technol 200:4549–4556Google Scholar
  86. 86.
    Gong WB, Sha CK, Sun DQ, Wang WQ (2006) Microstructures and thermal insulation capability of plasma-sprayed nanostructured ceria stabilized zirconia coatings. Surf Coat Technol 201:3109–3115Google Scholar
  87. 87.
    Racek O, Berndt CC, Guru DN, Heberlein J (2006) Nanostructured and conventional YSZ coatings deposited using APS and TTPR techniques. Surf Coat Technol 201:338–346Google Scholar
  88. 88.
    Soltani R, Garcia E, Coyle TW, Mostaghimi J, Lima RS, Marple BR, Moreau C (2006) Thermomechanical behavior of nanostructured plasma sprayed zirconia coatings. J Therm Spray Technol 15(4):657–662Google Scholar
  89. 89.
    Racek O, Berndt CC (2007) Mechanical property variations within thermal barrier coatings. Surf Coat Technol 202:362–369Google Scholar
  90. 90.
    Padture NP, Gell M, Jordan EH (2002) Thermal barrier coatings for gas turbine engine applications. Science 296:280–284Google Scholar
  91. 91.
    Lima RS, Marple BR (2008) Toward highly sintering-resistant nanostructured ZrO2-7wt%Y2O3 coatings for TBC applications by employing differential sintering. J Therm Spray Technol 17(5–6):846–852Google Scholar
  92. 92.
    Wang WQ, Sha CK, Sun DQ, Gu XY (2006) Microstructural feature, thermal shock resistance and isothermal oxidation resistance of nanostructured zirconia coating. Mater Sci Eng A 424:1–5Google Scholar
  93. 93.
    Liu C-B, Zhang Z-M, Jiang X-L, Liu M, Zhu Z-H (2009) Comparison of thermal shock behaviors between plasma-sprayed nanostructured and conventional zirconia thermal barrier coatings. Trans Nonferrous Metals Soc China 19:99–107Google Scholar
  94. 94.
    Zhou C, Wang N, Xu H (2007) Comparison of thermal cycling behavior of plasma-sprayed nanostructured and traditional thermal barrier coatings. Mater Sci Eng A 452–453:569–574Google Scholar
  95. 95.
    Lima RS, Khor KA, Li H, Cheang P, Marple BR (2005) HVOF spraying of nanostructured hydroxyapatite for biomedical applications. Mater Sci Eng A 396:181–187Google Scholar
  96. 96.
    Li H, Khor KA, Cheang P (2000) Effect of the powders’ melting state on the properties of HVOF sprayed hydroxyapatite coatings. Mater Sci Eng A 293:1–80Google Scholar
  97. 97.
    Li H, Khor KA (2006) Characteristics of the nanostructures in thermal sprayed hydroxyapatite coatings and their influence on coating properties. Surf Coat Technol 201:2147–2154Google Scholar
  98. 98.
    Yang YC, Chang C (2003) The bonding of plasma-sprayed hydroxyapatite coatings to titanium: effect of processing, porosity and residual stress. Thin Solid Films 444:260–275Google Scholar
  99. 99.
    Li H, Khor KA, Kumar R, Cheang P (2004) Characterization of hydroxyapatite-nano-zirconia composite coatings deposited by high velocity oxy-fuel (HVOF) spray process. Surf Coat Technol 182:227–236Google Scholar
  100. 100.
    Webster TJ, Siegel RW, Bizios R (1999) Osteoblast adhesion on nanophase ceramics. Biomaterials 20:1221–1227Google Scholar
  101. 101.
    Sun L, Berndt CC, Gross KA, Kucuk A (2001) Materials fundamentals and clinical performance of plasma-sprayed hydroxyapatite coatings: a review. J Biomed Mater Res 58:570–592Google Scholar
  102. 102.
    Lai KA (2002) Failure of hydroxyapatite-coated acetabular cups. J Bone Surg (Br) 84-B(5):64646Google Scholar
  103. 103.
    Yang Y-C, Yang C-Y (2013) Mechanical and histological evaluation of a plasma sprayed hydroxyapatite coating on a titanium bond coat. Ceram Int 39: 6509–6516Google Scholar
  104. 104.
    Gutwein LG, Webster TJ (2004) Increased viable osteoblast density in the presence of nanophase compared to conventional alumina and titania particles. Biomaterials 25:4175–4183Google Scholar
  105. 105.
    Lima RS, Li H, Khor KA, Marple BR (2006) Biocompatible nanostructured high-velocity oxyfuel sprayed titania coating: deposition, characterization, and mechanical properties. J Therm Spray Technol 15(4):623–627Google Scholar
  106. 106.
    Kim GE, Walker J (2007) Successful application of nanostructured titanium dioxide coating for high-pressure acid-leach application. J Therm Spray Technol 16(1):34–39Google Scholar
  107. 107.
    Wang Y, Tian W, Yang Y (2007) Thermal shock behaviour of nanostructured and conventional Al2O3/13 wt% TiO2 coatings fabricated by plasma spraying. Surf Coat Technol 201:7746–7754Google Scholar
  108. 108.
    Eigen N, Gartner F, Klassen T, Aust E, Bormann R, Kreye H (2005) Microstructures and properties of nanostructured thermal sprayed coatings using high-energy milled cermet powders. Surf Coat Technol 195:344–357Google Scholar
  109. 109.
    Lau ML, Strock E, Fabel A, Lavernia CJ, Lavernia EJ (1998) Synthesis and characterization of nanocrystalline Co43 coatings by plasma spraying. NanoStructured Mater 10(5):723–730Google Scholar
  110. 110.
    Ajdelsztajn L, Lee J, Chung K, Bastian FL, Lavernia EJ (2002) Synthesis and nanoindentation study of high-velocity oxygen fuel thermal-sprayed nanocrystalline and near-nanocrystalline Ni coatings. Metall Mater Trans A 33A:647–655Google Scholar
  111. 111.
    Cherigui M, Fenineche NE, Coddet C (2005) Structural study of iron-based microstructured and nanostructured powders sprayed by HVOF thermal spraying. Surf Coat Technol 192:19–26Google Scholar
  112. 112.
    Cherigui M, Fenineche NE, Gupta A, Zhang G, Coddet C (2006) Magnetic properties of HVOF thermally sprayed coatings obtained from nanostructured powders. Surf Coat Technol 201:1805–1813Google Scholar
  113. 113.
    Jing HG, Lau ML, Lavernia EJ (1998) Grain growth behavior of nanocrystalline Inonel 718 and Ni powders and coatings. NanoStructured Mater 10(2):169–178Google Scholar
  114. 114.
    Tang F, Ajdelsztajn L, Kim GE, Provenzano V, Schoenung JM (2004) Effects of surface oxidation during HVOF processing on the primary stage oxidation of a CoNiCrAlY coating. Surf Coat Technol 185:228–233Google Scholar
  115. 115.
    Tang F, Ajdelsztajn L, Schoenung JM (2004) Influence of cryomilling on the morphology and composition of the oxide scales formed on HVOF CoNiCrAlY coatings. Oxidation Metals 61(3/4):219–238Google Scholar
  116. 116.
    Ajdelsztajn L, Tang F, Kim GE, Provenzano V, Schoenung JM (2005) Synthesis and oxidation behavior of nanocrystalline MCrAlY bond coatings. J Therm Spray Technol 14(1):23–30Google Scholar
  117. 117.
    He J, Schoenung JM (2002) A review on nanostructured WC–Co coatings. Surf Coat Technol 157:72–79Google Scholar
  118. 118.
    Qiao Y, Liu Y, Fischer TE (2000) Sliding and abrasive wear resistance of thermal-sprayed WC-Co coatings. J Therm Spray Technol 10(1):118–125Google Scholar
  119. 119.
    He J, Ice M, Lavernia EJ (2000) Synthesis of nanostructured Cr3C2-25(Ni20Cr) coatings. Metall Mater Trans A 31A:555–564Google Scholar
  120. 120.
    Kear BH, Sadangi RK, Jain M, Yao R, Kalman Z, Skandan G, Mayo WE (1999) Thermal sprayed nanostructured WC/Co hardcoatings. J Therm Spray Technol 9(3):399–406Google Scholar
  121. 121.
    He J, Ice M, Dallek S, Lavernia EJ (2000) Synthesis of nanostructured WC-12 wt. % Co coating using mechanical milling and high velocity oxygen fuel thermal spraying. Metall Mater Trans A 31A:541–553Google Scholar
  122. 122.
    Skandan G, Yao R, Sadangi R, Kear BH, Qiao Y, Liu L, Fischer TE (1999) Multimodal coatings: a new concept in thermal spraying. J Therm Spray Technol 9(3):329–331Google Scholar
  123. 123.
    Ban Z-G, Shaw LL (2002) Synthesis and processing of nanostructured WC-Co materials. J Mater Sci 37:3397–3403Google Scholar
  124. 124.
    Dent AH, DePalo S, Samph S (2001) Examination of the wear properties of HVOF sprayed nanostructured and conventional WC-Co cermets with different binder phase contents. J Therm Spray Technol 11(4):551–558Google Scholar
  125. 125.
    Qiao Y, Fischer TE, Andrew Dent (2003) The effects of fuel chemistry and feedstock powder structure on the mechanical and tribological properties of HVOF thermal-sprayed WC–Co coatings with very fine structures. Surf Coat Technol 172:24–41Google Scholar
  126. 126.
    Ban ZG, Shaw LL (2003) Characterization of thermal sprayed nanostructured WC-Co coatings derived from nanocrystalline WC-18wt.%Co powders. J Therm Spray Technol 12(1):112–119Google Scholar
  127. 127.
    Marple BR, Lima RS (2005) Process temperature/velocity-hardness-wear relationships for high-velocity oxyfuel sprayed nanostructured and conventional cermet coatings. J Therm Spray Technol 14(1):67–76Google Scholar
  128. 128.
    Bartuli C, Valente T, Cipri F, Bemporad E, Tului M (2005) Parametric study of an HVOF process for the deposition of nanostructured WC-Co coatings. J Therm Spray Technol 14(2):187–195Google Scholar
  129. 129.
    Guilemany JM, Dosta S, Miguel JR (2006) The enhancement of the properties of WC-Co HVOF coatings through the use of nanostructured and microstructured feedstock powders. Surf Coat Technol 201:1180–1190Google Scholar
  130. 130.
    Siegmann S, Brandt O, Dvorak M (2004) Thermally sprayed wear resistant coatings with nanostructured hard phases. J Therm Spray Technol 13(1):37–43Google Scholar
  131. 131.
    He J, Ice M, Schoenung JM, Shin DH, Lavernia EJ (2001) Thermal stability of nanostructured Cr3C2-NiCr coatings. J Thermal Spray Technol 10(2):293–300Google Scholar
  132. 132.
    Matthews S, Hyland M, James B (2004) Long-term carbide development in high-velocity oxygen fuel/high-velocity air fuel Cr3C2-NiCr coatings heat treated at 900 °C. J Therm Spray Technol 13(4):526–536Google Scholar
  133. 133.
    Roy M, Pauschitz A, Bernardi J, Koch T, Franek F (2006) Microstructure and mechanical properties of HVOF sprayed nanocrystalline Cr3C2-25(Ni20Cr) coating. J Therm Spray Technol 15(3):372–381Google Scholar
  134. 134.
    Ji G, Grosdidier T, Morniroli J-P (2007) Microstructure of a high-velocity oxy-fuel thermal-sprayed nanostructured coating obtained from milled powder. Metall Mater Trans A 38A:2455–2463Google Scholar
  135. 135.
    He J, Schoenung JM (2003) Nanocrystalline Ni coatings strengthened with ultrafine particles Ni-AlN HVOF. Metall Mater Trans A 34A:673–683Google Scholar
  136. 136.
    Basak AK, Achanta S, Celis J-P, Vardavoulias M, Matteazzi P (2008) Structure and mechanical properties of plasma sprayed nanostructure ed alumina and FeCuAl–alumina cermet coatings. Surf Coat Technol 202:2368–2373Google Scholar
  137. 137.
    Hwang C, Chia-ho Y (2007) Formation of nanostructured YSZ/Ni anode with pore channels by plasma spraying. Surf Coat Technol 201:5954–5959Google Scholar
  138. 138.
    Laha T, Agarwal A, McKechnie T (2004) Forming nanostructured hypereutectic aluminum via high-velocity oxyfuel spray deposition. JOM J Min Met Mater Soc 56(1):54–56Google Scholar
  139. 139.
    Laha T, Agarwal A, McKechnie T, Rea K, Seal S (2005) Synthesis of bulk nanostructured aluminum alloy component through vacuum plasma spray technique. Acta Mater 53:5429–5438Google Scholar
  140. 140.
    Laha T, Agarwal A, McKechnie T, Seal S (2004) Synthesis and characterization of plasma spray formed carbon nanotube reinforced aluminum composite. Mater Sci Eng A 381:249–258Google Scholar
  141. 141.
    Laha T, Kuchibhatla S, Seal S, Li W, Agarwal A (2007) Interfacial phenomena in thermally sprayed multiwalled carbon nanotube reinforced aluminum nanocomposite. Acta Mater 55:1059–1066Google Scholar
  142. 142.
    Ajdelsztajn L, Zúñiga A, Jodoin B, Lavernia EJ (2006) Cold-spray processing of a nanocrystalline Al-Cu-Mg-Fe-Ni alloy with Sc. J Therm Spray Technol 15(2):184–190Google Scholar
  143. 143.
    Wang H-T, Li C-J, Yang G-J, Li C-X, Zhang Q, Li W-Y (2007) Microstructural characterization of cold-sprayed nanostructured FeAl intermetallic compound coating and its ball-milled feedstock powders. J Therm Spray Technol 16(5–6):669–676Google Scholar
  144. 144.
    Ajdelsztajn L, Jodoin B, Schoenung JM (2006) Synthesis and mechanical properties of nanocrystalline Ni coatings produced by cold gas dynamic spraying. Surf Coat Technol 201:1166–1172Google Scholar
  145. 145.
    Zhang Q, Li C-J, Li C-X, Yang G-J, Lui S-C (2008) Study of oxidation behavior of nanostructured NiCrAlY bond coatings deposited by cold spraying. Surf Coat Technol 202(14):3378–3384Google Scholar
  146. 146.
    Li C-J, Yang G-J, Gao P-H, Ma J, Wang Y-Y, Li C-X (2007) Characterization of nanostructured WC-Co, deposited by cold spraying. J Therm Spray Technol 16(5–6):1011–1020Google Scholar
  147. 147.
    Yandouzi M, Sansoucy E, Ajdelsztajn L, Jodoin B (2007) WC-based cermet coatings produced by cold gas dynamic and pulsed gas dynamic spraying processes. Surf Coat Technol 202:382–390Google Scholar
  148. 148.
    Ajdelsztajn L, Jodoin B, Richer P, Sansoucy E, Lavernia EJ (2006) Cold gas dynamic spraying of iron-base amorphous alloy. J Therm Spray Technol 15(4):495–500Google Scholar
  149. 149.
    Sansoucy E, Kim GE, Moran AL, Jodoin B (2007) Mechanical characteristics of Al-Co-Ce coatings produced by the cold spray process. J Therm Spray Technol 16(5–6):651–660Google Scholar
  150. 150.
    Fauchais P, Rat V, Delbos C, Coudert J-F, Chartier T, Bianchi L (2005) Understanding of suspension D.C. plasma spraying of finely structured coatings for SOFC. IEEE Trans Plasma Sci 33(2):920–930Google Scholar
  151. 151.
    Fazilleau J, Delbos C, Rat V, Coudert J-F, Fauchais P, Pateyron B (2006) Phenomena involved in suspension plasma spraying part 1: suspension injection and behaviour. Plasma Chem Plasma Process 26(4):371–391Google Scholar
  152. 152.
    Etchart-Salas R, Rat V, Coudert JF, Fauchais P, Caron N, Wittman K, Alexandre S (2007) Influence of plasma instabilities in ceramic suspension plasma spraying. J Therm Spray Technol 16(5–6):857–865Google Scholar
  153. 153.
    Bhatia T, Ozturk A, Xie LD, Jordan EH, Cetegen BM, Gell M, Ma XC, Padture NP (2001) Mechanisms of ceramic coating deposition in solution precursor spray. J Mater Res 17(9):2363–2372Google Scholar
  154. 154.
    Gell M, Xie LD, Ma XC, Jordan EH, Padture NP (2004) Highly durable thermal barrier coatings made by the solution precursor plasma spray process. Surf Coat Technol 13(1):97–102Google Scholar
  155. 155.
    Jordan EH, Xie LD, Ma XC, Gell M, Padture NP, Cetegem B, Roth J, Xiao TD, Bryant PEL, Roth J, Xiao TD, Bryant PEC (2004) Superior thermal barrier coatings using solution precursor plasma spray. J Therm Spray Technol 13(1):57–65Google Scholar
  156. 156.
    Padture NP, Schlichting KW, Bhatia T, Ozturk A, Cetegem B, Jordan EH, Gell M, Jiang S, Xiao TD, Strutt PR, Garcia E, Miranzo P, Osendi MI (2001) Towards durable thermal barrier coatings with novel microstructures deposited by solution precursor plasma spray. Acta Mater 49:2251–2257Google Scholar
  157. 157.
    Marchand C, Vardelle A, Mariaux G, Lefort P (2008) Modelling of the plasma spray process with liquid feedstock injection. Surf Coat Technol 202:4458–4464Google Scholar
  158. 158.
    Fauchais P, Vardelle A (2011) Innovative and emerging processes in plasma spraying: from micro- to nanostructured coatings. J Phys D Appl Phys 44:194011 (14pp)Google Scholar
  159. 159.
    Bertolissi G, Chazelas C, Bolelli G, Lusvarghi L, Vardelle M, Vardelle A (2012) Engineering the microstructure of solution precursor plasma-sprayed coatings. J Therm Spray Technol 21(6): 1148–1158Google Scholar
  160. 160.
    Marchand O, Girardot L, Planche MP, Bertrand P, Bailly Y, Bertrand G (2011) An insight into suspension plasma spray: injection of the suspension and its interaction with the plasma flow. J Therm Spray Technol 20(6):1310–1320Google Scholar
  161. 161.
    Landes K (2006) Diagnostics in plasma spraying techniques. Surf Coat Technol 201:1948–1954Google Scholar
  162. 162.
    Mauer G, Guignard A, Vaßen R, Stöver D (2010) Process diagnostics in suspension plasma spraying. Surf Coat Technol 205:961–966Google Scholar
  163. 163.
    Soysal D, Ansar A (2013) A new approach to understand liquid injection into atmospheric plasma jets. Surf Coat Technol 220:187–190Google Scholar
  164. 164.
    Shan Y, Coyle TW, Mostaghimi J (2007) Numerical simulation of droplet break-up and collision in solution precursor plasma spraying. J Therm Spray Technol 16:698–704Google Scholar
  165. 165.
    Vincent S, Balmigere G, Caruyer C, Meillot E, Caltagirone J-P (2009) Contribution to the modeling of the interaction between a plasma flow and a liquid jet. Surf Coat Technol 203:2162–2171Google Scholar
  166. 166.
    Meillot E, Vincent S, Caruyer C, Caltagirone J-P, Damiani D (2009) From DC time-dependent thermal plasma generation to suspension plasma-spraying interactions. J Therm Spray Technol 18:875–86Google Scholar
  167. 167.
    Caruyer C, Vincent S, Meillot E, Caltagirone J-P (2010) Modeling the first instant of the interaction between a liquid and a plasma jet with a compressible approach. Surf Coat Technol 205:974–979Google Scholar
  168. 168.
    Basu S, Cetegen BM (2008) Modeling of liquid ceramic precursor droplets in a high velocity oxy-fuel flame jet. Acta Mater 56:2750–2759Google Scholar
  169. 169.
    Delbos C, Fazilleau J, Rat V, Coudert JF, Fauchais P, Pateyron B (2006) Phenomena involved in suspension plasma spraying part 2: zirconia particle treatment and coating formation. Plasma Chem Plasma Process 26:393–414Google Scholar
  170. 170.
    VanEvery K, Krane MJM, Trice RW (2012) Parametric study of suspension plasma spray processing parameters on coating microstructures manufactured from nanoscale yttria-stabilized zirconia. Surf Coat Technol 206:2464–2473Google Scholar
  171. 171.
    Brousse-Pereira E (2010) Elaboration by thermal spray of finely structured elements of a high temperature electolyser for hydrogen production: processes, structures and characteristics. Ph.D. thesis, University of Limoges, France, 21 Sept 2010Google Scholar
  172. 172.
    Brousse E, Montavon G, Fauchais P, Denoirjean A, Rat V, Coudert J-F, Ageorges H (2008) Thin and dense yttria-partially stabilized zirconia electrolytes for IT-SOFC manufactured by suspension plasma spraying. In: Lugscheider E (ed) Thermal spray crossing borders. DVS, Düsseldorf, Germany, pp 547–552Google Scholar
  173. 173.
    Racek O (2010) The effect of HVOF particle-substrate interactions on local variations in the coating microstructure and the corrosion resistance. J Therm Spray Technol 19(5):841–851Google Scholar
  174. 174.
    Basu S, Jordan EH, Cetegen BM (2006) Fluid mechanics and heat transfer of liquid precursor droplets injected into high temperature plasmas. J Therm Spray Technol 15(4):576–581Google Scholar
  175. 175.
    Oberste-Berghaus J, Legoux J-G, Moreau C, Tarasi F, Chraska T (2008) Mechanical and thermal transport properties of suspension thermal-sprayed alumina-zirconia composite coatings. J Therm Spray Technol 17(1):91–104Google Scholar
  176. 176.
    Oberste Berghaus J, Legoux J-G, Moreau C, Tarasi F, Chráska T (2007) Mechanical and thermal transport properties of suspension thermal sprayed alumina-zirconia composite coatings. In: Marple BR, Hyland MM, Lau Y-C, Li C-J, Lima RS, Montavon G (eds) Thermal spray 2007: global coating solutions. ASM International, Materials Park, OH, e-proceedingsGoogle Scholar
  177. 177.
    Tarasi F, Medraj M, Dolatabadi A, Oberste-Berghaus J, Moreau C (2008) Effective parameters in axial injection suspension plasma spray process of alumina-zirconia ceramics. J Therm Spray Technol 17(5–6):685–691Google Scholar
  178. 178.
    Tarasi F, Medraj M, Dolatabadi A, Oberste-Berghaus J, Moreau C (2011) Amorphous and crystalline phase formation during suspension plasma spraying of the alumina–zirconia composite. J Eur Ceram Soc 31:2903–2913Google Scholar
  179. 179.
    Erne M, Kolar D, Hübsch C, Möhwald M, Bach Fr-W (2012) Synthesis of tribologically favorable coatings for hot extrusion tools by suspension plasma spraying. J Therm Spray Technol 21(3–4):668–675Google Scholar
  180. 180.
    Gadow R, Killinger A, Rauch J (2008) New results in high velocity suspension flame spraying (HVSFS). Surf Coat Technol 202:4329–4336Google Scholar
  181. 181.
    Dongmo E, Gadow R, Killinger A, Wenzelburger M (2009) Modeling of combustion as well as heat, mass, and momentum transfer during thermal spraying by HVOF and HVSFS. J Therm Spray Technol 18(5–6):896–908Google Scholar
  182. 182.
    Dongmo E, Wenzelburger M, Gadow R (2008) Analysis and optimization of the HVOF process by combined experimental and numerical approaches. Surf Coat Technol 202:4470–4478Google Scholar
  183. 183.
    Killinger A, Kuhn M, Gadow R (2006) High-velocity suspension flame spraying (HVSFS), a new approach for spraying nanoparticles with hypersonic speed. Surf Coat Technol 201:1922–1929Google Scholar
  184. 184.
    Bolelli G, Cannillo V, Gadow R, Killinger A, Lusvarghi L, Rauch J, Romagnoli M (2010) Effect of the suspension composition on the microstructural properties of high velocity suspension flame sprayed (HVSFS) Al2O3 coatings. Surf Coat Technol 204:1163–1179Google Scholar
  185. 185.
    Bouyer E, Gitzhofer F, Boulos MI (1997) The suspension plasma spraying of bioceramics by induction plasma. JOM February:58–62Google Scholar
  186. 186.
    Jia L, Gitzhofer F (2010) Induction plasma synthesis of nanostructured SOFCs electrolyte using solution and suspension plasma spraying: a comparative study. J Therm Spray Technol 19(3):566–574Google Scholar
  187. 187.
    Shen Y, Almeida VAB, Gitzhofer F (2011) Preparation of nano-composite GDC/LSCF cathode material for IT-SOFC by induction plasma spraying. J Therm Spray Technol 20(1–2):145–153Google Scholar
  188. 188.
    Vaßen R, Kaßner H, Mauer G, Stöver D (2010) Suspension plasma spraying: process characteristics and applications. J Therm Spray Technol 19(1–2):219–225Google Scholar
  189. 189.
    Etchart-Salas R (2007) D.C. plasma spraying of suspensions of submicronic particles. Experimental and analytic approaches of phenomena implied in the reproducibility and quality of coatings (in French). Ph.D. thesis, University of Limoges, FranceGoogle Scholar
  190. 190.
    Kassner H, Siegert R, Hathiramani D, Vassen R, Stoever D (2008) Application of suspension plasma spraying (SPS) for manufacture of ceramic coatings. J Therm Spray Technol 17(1):115–123Google Scholar
  191. 191.
    Oberste-Berghaus J, Marple B, Moreau C (2006) Suspension plasma spraying of nanostructured WC-12Co coatings. J Therm Spray Technol 15(4):676–681Google Scholar
  192. 192.
    Wang Y, Legoux J-G, Neagu R, Hui S, Marple BR (2012) Suspension plasma spray and performance characterization of half cells with NiO/YSZ anode and YSZ electrolyte. J Therm Spray Technol 21(1):7–16Google Scholar
  193. 193.
    Boulos MI (1992) RF induction plasma spraying: state-of-the-art review. J Therm Spray Technol 1(1):33–40Google Scholar
  194. 194.
    Hui R, Oberste-Berghaus J, Decès-Petit C, Qu W, Yick S, Legoux J-G, Moreau C (2009) High performance metal-supported solid oxide fuel cells fabricated by thermal spray. J Power Source 191:371–376Google Scholar
  195. 195.
    Ravi BG, Sampath S, Gambino R, Devi PS, Parise JB (2006) Plasma spray synthesis from precursors: progress, issues and considerations. J Therm Spray Technol 15(4):701–707Google Scholar
  196. 196.
    Gell M, Jordan EH, Teicholz M, Cetegen BM, Padture N, Xie L, Chen D, Ma X, Roth J (2008) Thermal barrier coatings made by the solution precursor plasma spray process. J Therm Spray Technol 17(1):124–135Google Scholar
  197. 197.
    Karthikeyan J, Berndt CC, Tikkanen J, Reddy S, Herman H (1997) Plasma spray synthesis of nanomaterial powders and deposits. Surf Coat Technol 238(2):275–286Google Scholar
  198. 198.
    Jadhav AD, Padture NP, Jordan EH, Gell M, Miranzo P, Fullu ER (2006) Low thermal conductivity plasma sprayed thermal barrier coatings with engineered microstructure. Acta Mater 54(12):3343–3349Google Scholar
  199. 199.
    Vasiliev AL, Padture NP, Ma X (2006) Coatings of metastable ceramics deposited by solution precursor plasma spray: I-binary ZrO2-Al2O3 system. Acta Mater 54(19):4913–4920Google Scholar
  200. 200.
    Chen D, Jordan EH, Gell M (2008) Effect of solution concentration on splat formation and coating microstructure using the solution precursor plasma spray process. Surf Coat Technol 202:2132–2138Google Scholar
  201. 201.
    Vasiliev AL, Padture NP, Ma XC (2006) Coatings of metastable ceramics deposited by solution-precursor plasma spray: II. Ternary ZrO2–Y2O3–Al2O3 system. Acta Mater 54(19):4921–4936Google Scholar
  202. 202.
    Chen D, Jordan E, Gell M (2007) Thermal and crystallization behavior of zirconia precursor used in the solution precursor plasma spray process. J Mater Sci 42(14):5576–5580Google Scholar
  203. 203.
    Chen D, Jordan EH, Gell M (2010) The solution precursor plasma spray coatings: influence of solvent type. Plasma Chem Plasma Process 30:111–119Google Scholar
  204. 204.
    Muoto CK, Jordan EH, Gell M, Aindow M (2011) Identification of desirable precursor properties for solution precursor plasma spray. J Therm Spray Technol 20(4):802–816Google Scholar
  205. 205.
    Toma FL, Bertrand G, Rampon R, Klein D, Coddet C (2006) Relationship between the suspension properties and liquid plasma sprayed coating characteristics. In: ITSC 2006. ASM International, Materials Park, OH, e-proceedingsGoogle Scholar
  206. 206.
    Rampon R, Bertrand G, Toma FL, Coddet C (2006) Liquid plasma sprayed coatings of yttria stabilized for SOFC electrolyte. In: ITSC 2006. ASM International, Materials Park, OH, e-proceedingsGoogle Scholar
  207. 207.
    Toma FL, Sokolov D, Bertrand G, Klein D, Coddet C, Meunier C (2006) Comparison of the photocatalytic behavior of TiO2 coatings elaborated by different thermal spray processes. J Therm Spray Technol 15(4):576–581Google Scholar
  208. 208.
    Oberste-Berghaus J, Boccaricha S, Legoux JG, Moreau C, Chraska T (2005) Suspension plasma spraying of nanoceramics using an axial injection torch. In: ITSC 2005, DVS, Dusselörf, Germany, e-proceedingsGoogle Scholar
  209. 209.
    Fazilleau J, Delbos C, Violier M, Coudert JF, Fauchais P, Bianchi L, Wittmann-Teneze K (2003) Influence of substrate temperature on formation of micrometric splats obtained by plasma spraying liquid suspension. In: Marple BR, Moreau C (eds) ITSC 2003. ASM International, Materials Park, OH, pp 889–895Google Scholar
  210. 210.
    Rampon R, Filiatre C, Bertrand G (2008) Suspension plasma spraying of YPSZ coatings: suspension atomization and injection. J Therm Spray Technol 17(1):105–114Google Scholar
  211. 211.
    Rampon R, Toma F-L, Bertrand G, Coddet C (2006) Liquid plasma sprayed coatings of yttria-stabilized zirconia for SOFC electrolytes. J Therm Spray Technol 15(4):682–688Google Scholar
  212. 212.
    Arevalo-Quintero O, Waldbillig D, Kesler O (2011) An investigation of the dispersion of YSZ, SDC, and mixtures of YSZ/SDC powders in aqueous suspensions for application in suspension plasma spraying. Surf Coat Technol 205:5218–5227Google Scholar
  213. 213.
    Patent FR0453390, Revêtement nanostructuré et procédé de revêtement, CEA Le Ripault, Monts, 37, France (in French)Google Scholar
  214. 214.
    Jordan EH, Gell M, Benzani P, Chen D, Basu S, Cetegem B, Wa F, Ma XC (2007) Making dense coatings with the solution precursor plasma spray process. In: Marple BR et al (eds) Thermal spraying 2007:global coating solutions. ASM International, Materials Park, OH, e-proceedingsGoogle Scholar
  215. 215.
    Cotler EM, Chen D, Molz RJ (2011) Pressure-based liquid feed system for suspension plasma spray coatings. J Therm Spray Technol 20(4):967–973Google Scholar
  216. 216.
    Damiani D. Elaboration, fitting in shape and characterization of materials for energy by the interaction radiation-matter. HdR University of Limoges, France, 13 Dec 2012Google Scholar
  217. 217.
    Meillot E, Vert R, Caruyer C, Damiani D, Vardelle M (2011) Manufacturing nanostructured YSZ coatings by suspension plasma spraying (SPS): effect of injection parameters. J Phys D Appl Phys 44:194008 (8pp)Google Scholar
  218. 218.
    Ozturkand A, Cetegen B (2005) Modeling of axially and transversely injected precursor droplets into a plasma environment. Int J Heat Mass Transfer 48(21–22):4367–4383Google Scholar
  219. 219.
    Saha A, Seal S, Cetegen B, Jordan E, Ozturk A, Basu S (2009) Thermo-physical processes in cerium nitrate precursor droplets injected into high temperature plasma. Surf Coat Technol 203:2081–2091Google Scholar
  220. 220.
    Xie L, Ma X, Ozturk A, Jordan EH, Padture NP, Cetegen BM, Xiao DT, Gell M (2004) Processing parameter effects on solution precursor plasma spray process spray patterns. Surf Coat Technol 183(1):51–61Google Scholar
  221. 221.
    Xie L, Ma XC, Jordan EH, Padture NP, Xiao DT, Gell M (2004) Deposition mechanisms of thermal barrier coatings in the solution precursor plasma spray process. Surf Coat Technol 177–178:103–107Google Scholar
  222. 222.
    Delbos C, Fazilleau J, Coudert JF, Fauchais P, Bianchi L (2004) Finely structured ceramic coatings elaborated by liquid suspension injection in a DC plasma jet. In: Thermal spray solutions : advances in technology and application. DVS, Düsseldorf, GermanyGoogle Scholar
  223. 223.
    Wittmann K, Blein F, Fazilleau J, Coudert JF, Fauchais P (2002) A new process to deposit thin coatings by injecting nanoparticles suspensions in a d.c. plasma jet. In: Lugsheider E (ed) ITSC 2002. DVS, Düsseldorf, Germany, e-proceedingsGoogle Scholar
  224. 224.
    Cedelle J, Vardelle M, Fauchais P (2006) Influence of stainless steel substrate preheating on surface topography and on millimeter and micrometer-sized splat formation. Surf Coat Technol 201(3–4):1378–1382Google Scholar
  225. 225.
    Bianchi L, Denoirjean A, Blein F, Fauchais P (1997) Microstructural investigation of plasma sprayed ceramic splats. Thin Solid Films 299:125–135Google Scholar
  226. 226.
    Siegert R, Doring JE, Marque’s JL, Vassen R, Sebold D, Stöver D (2005) Influence of injection parameters on the suspension plasma spraying coating properties. In: Lugsheider E (ed) ITSC 2005. DVS, Düsseldorf, Germany, e-proceedingsGoogle Scholar
  227. 227.
    Tingaud O, Grimaud A, Denoirjean A, Montavon G, Rat V, Coudert JF, Fauchais P, Chartier T (2008) Suspension plasma-sprayed alumina coating structures: operating parameters versus coating architecture. J Therm Spray Technol 17(5–6):662–670Google Scholar
  228. 228.
    Tingaud O, Bacciochini A, Montavon G, Denoirjean NA, Fauchais P (2009) Suspension DC plasma spraying of thick finely-structured ceramic coatings: process manufacturing mechanisms. Surf Coat Technol 203:2157–2161Google Scholar
  229. 229.
    Xie L, Chen D, Jordan EH, Padture NP, Ozturk A, Wu F, Ma XC, Cetegem BM, Gell M (2006) Formation of vertical cracks in solution-precursor plasma-sprayed thermal barrier coatings. Surf Coat Technol 201:1058–1064Google Scholar
  230. 230.
    Kozerski S, Łatka L, Pawlowski L, Cernuschi F, Petit F, Pierlot C, Podlesak H, Laval JP (2011) Preliminary study on suspension plasma sprayed ZrO2 + 8 wt.% Y2O3 coatings. J Eur Ceram Soc 31:2089–2098Google Scholar
  231. 231.
    Waldbillig D, Kesler O (2011) Effect of suspension plasma spraying process parameters on YSZ coating microstructure and permeability. Surf Coat Technol 205:5483–5492Google Scholar
  232. 232.
    Delbos C (2004) Contribution to the understanding of liquid injection of ceramics (Y.S.Z., Perovskite…) or metals (Ni,…) in blown arc plasma jets to achieve finely structured coatings for SOFCs (in French). Ph.D. thesis (in French), University of Limoges, FranceGoogle Scholar
  233. 233.
    Toma F-L, Berger L-M, Scheitz S, Langner S, Rödel C, Potthoff A, Sauchuk V, Kusnezoff M (2012) Comparison of the microstructural characteristics and electrical properties of thermally sprayed Al2O3 coatings from aqueous suspensions and feedstock powders. J Therm Spray Technol 21(3–4):480–488Google Scholar
  234. 234.
    Müller P, Killinger A, Gadow R (2012) Comparison between high-velocity suspension flame spraying and suspension plasma spraying of alumina. J Therm Spray Technol 21(6):1120–1128Google Scholar
  235. 235.
    Moign A, Vardelle A, Themelis NJ, Legoux JG (2010) Life cycle assessment of using powder and liquid precursors in plasma spraying: the case of yttria-stabilized zirconia. Surf Coat Technol 205:668–673Google Scholar
  236. 236.
    Xie L, Jordan EH, Padture NP, Gell M (2004) Phase and microstructural stability of solution precursor plasma sprayed thermal barrier coatings. Mater Sci Eng A 381:189–195Google Scholar
  237. 237.
    Madhwal M, Jordan EH, Gell M (2004) Failure mechanisms of dense vertically-cracked thermal barrier coatings. Mater Sci Eng A 384:151–161Google Scholar
  238. 238.
    Low AD, Jadhav NP, Padture EH, Jordan M, Gell P, Miranzo ER (2006) Fuller Jr., Low-thermal-conductivity plasma-sprayed thermal barrier coatings with engineered microstructures. Acta Mater 54:3343–3349Google Scholar
  239. 239.
    Vaßen R, Stuke A, Stöver D (2009) Recent developments in the field of thermal barrier coatings. J Therm Spray Technol 18(2):181–186Google Scholar
  240. 240.
    Ben-Ettouil F, Denoirjean A, Grimaud A, Montavon G, Fauchais P (2009) Sub-micrometer-sized Y-PSZ thermal barrier coatings manufactured by suspension plasma spraying: process, structure and some functional properties. In: ITSC 2009, ASM, Materials Park, OHGoogle Scholar
  241. 241.
    Bacciochini A, Ben-Ettouil F, Brousse E, Ilavsky J, Montavon G, Denoirjean A, Valette S, Fauchais P (2010) Quantification of void network architectures of as-sprayed and aged nanostructured yttria-stabilized zirconia (YSZ) deposits manufactured by suspension plasma spraying. Surf Coat Technol 205:683–689Google Scholar
  242. 242.
    Fauchais P, Etchart-Salas R, Delbos C, Tognovi M, Rat V, Coudert J-F, Chartier T (2007) Suspension and solution plasma spraying of finely structured layers for SOFCs. J Phys D Appl Phys 40(8):2394–2406Google Scholar
  243. 243.
    Waldbilling D, Kesler O, Tang Z, Burgess A (2007) Suspension plasma spraying of solid oxide fuel cell electrolytes. In: Marple BR et al (eds) Thermal spraying 2007:global coating solutions. ASM International, Materials Park, OH, e-proceedingsGoogle Scholar
  244. 244.
    Stöver D, Hathiramani D, Vaßen R, Damani RJ (2006) Plasma_sprayed components for SOFC applications. Surf Coat Technol 201:2002–2005Google Scholar
  245. 245.
    Brinley E, Babu KS, Seal S (2007) The solution precursor plasma spray processing of nanomaterials. JOM July:54–59Google Scholar
  246. 246.
    Oberste Berghaus J, Legoux J-G, Moreau C, Hui R, Deces-Petit C, Qu W, Yick S, Wang Z, Maric R, Ghosh D (2008) Suspension HVOF spraying of reduced temperature solid oxide fuel cell electrolytes. J Therm Spray Technol 17(5–6):700–707Google Scholar
  247. 247.
    Monterrubio-Badillo C, Ageorges H, Chartier T, Coudert J-F, Fauchais P (2006) Preparation of LaMnO3 perovskite thin films by suspension plasma spraying for SOFC cathodes. Surf Coat Technol 200:3743–3756Google Scholar
  248. 248.
    Wang X-M, Li C-X, Li C-J, Yang G-J (2010) Effect of microstructures on electrochemical behavior of La0.8Sr0.2MnO3 deposited by suspension plasma spraying. Int J Hydrogen Energy. To be publishedGoogle Scholar
  249. 249.
    Bouaricha S, Oberste-Berghaus J, Legoux J-G, Moreau C, Ghosh D (2005) Production of doped-ceria plasma sprayed nano-coatings using an internal injection of a suspension containing nanoparticles. In: Lugscheider E (ed) Thermal spray connects: explore its surfacing potential! DVS-Verlag GmbH, Düsseldorf, Germany, e-proceedingsGoogle Scholar
  250. 250.
    Waldbillig D, Kesler O (2009) Characterization of metal-supported axial injection plasma sprayed solid oxide fuel cells with aqueous suspension plasma sprayed electrolyte layers. J Power Sources 191:320–329Google Scholar
  251. 251.
    Marchand O, Bertrand P, Mougin J, Comminges C, Planche M-P, Bertrand G (2010) Characterization of suspension plasma-sprayed solid oxide fuel cell electrodes. Surf Coat Technol 205(4):993–998Google Scholar
  252. 252.
    Michaux P, Montavon G, Grimaud A, Denoirjean A, Fauchais P (2010) Elaboration of porous NiO/8YSZ layers by several SPS and SPPS routes. J Therm Spray Technol 19(1–2):317–327Google Scholar
  253. 253.
    Waldbillig D, Kesler O (2009) The effect of solids and dispersant loadings on the suspension viscosities and deposition rates of suspension plasma sprayed YSZ coatings. Surf Coat Technol 203:2098–2101Google Scholar
  254. 254.
    McCoppin J, Young D, Reitz T, Maleszewski A, Mukhopadhyay S (2011) Solid oxide fuel cell with compositionally graded cathode functional layer deposited by pressure assisted dual-suspension spraying. J Power Sources 196:3761–3765Google Scholar
  255. 255.
    Xiaoming W, Chengxin L, Changjiu L, Lihui T, Guanjun Y (2011) Microstructure and electrochemical behavior of La0.8Sr0.2MnO3 deposited by solution precursor plasma spraying. Rare Metal Mater Eng 40(11):1881–1886Google Scholar
  256. 256.
    Oberste-Berghaus J, Marple B, Moreau C (2006) Suspension plasma spraying of nanostructured WC-12Co coatings. In: Marple BR et al (eds) Thermal spray conference 2006. ASM International, Materials Park, OH, e-proceedingsGoogle Scholar
  257. 257.
    Toma F-L, Berger L-M, Stahr CC, Naumann T, Langner S (2010) Microstructures and functional properties of suspension-sprayed Al2O3 and TiO2 coatings: an overview. J Therm Spray Technol 19(1–2):262–274Google Scholar
  258. 258.
    Toma F-L, Berger L-M, Naumann T, Langner S (2008) Microstructures of nanostructured ceramic coatings obtained by suspension thermal spraying. Surf Coat Technol 202:4343–4348Google Scholar
  259. 259.
    Darut G, Ageorges H, Denoirjean A, Montavon G, Fauchais P (2008) Effect of the structural scale of plasma-sprayed alumina coatings on their friction coefficients. J Therm Spray Technol 17(5–6):788–797Google Scholar
  260. 260.
    Qiu C, Chen Y (2009) Manufacturing process of nanostructured alumina coatings by suspension plasma spraying. J Therm Spray Technol 18(2):272–283Google Scholar
  261. 261.
    Zhang J, He J, Dong Y, Li X, Yan D (2008) Microstructure characteristics of Al2O3-13wt.%TiO2 coating plasma spray deposited with nanocrystalline powders. J Mater Process Technol 197(1–3):31–35Google Scholar
  262. 262.
    Sodeoka S, Suzaki M, Inone T (2006) Mechanical properties of plasma sprayed alumina-zirconia nano-composite film. In: Marple BR et al (eds) Thermal spray conference 2006. ASM International, Materials Park, OH, e-proceedingsGoogle Scholar
  263. 263.
    Oliker VE, Terentev AE, Shvedova LK, Martsenyuk IS (2009) Use of aqueous suspensions in plasma spraying of alumina coatings. Powder Metall Metal Ceram 48(1–2):115–120Google Scholar
  264. 264.
    Chen D, Jordan EH, Gell M (2009) Suspension plasma sprayed composite coating using amorphous powder feedstock. Appl Surf Sci 255:5935–5938Google Scholar
  265. 265.
    Tingaud O, Bertrand P, Bertrand G (2010) Microstructure and tribological behavior of suspension plasma sprayed Al2O3 and Al2O3–YSZ composite coatings. Surf Coat Technol 205:1004–1008Google Scholar
  266. 266.
    Manzat A, Killinger A, Gadow R (2010) Application of supersonic flame spraying for next generation cylinder liner coatings. In: Wellnitz J (ed) Sustainable automotive technologies. Springer, Berlin, pp 175–181Google Scholar
  267. 267.
    Killinger A, Gadow R, Rempp A, Manzat A (2010) Advanced ceramic tribological layers by thermal spray routes. Adv Sci Technol 66:106–119Google Scholar
  268. 268.
    Tomaszek R, Znamirovski Z, Pawlowski L, Wojnakowski A (2006) Impedance spectroscopy of suspension plasma sprayed titania coatings. Surf Coat Technol 201(5):2099–2102Google Scholar
  269. 269.
    Garcia E, Zhang ZB, Coyle TW, Hao SE, Ma SL (2007) Thermal spraying 2007:global coating solutions. In: Marple BR et al (eds) ASM International, Materials Park, OHGoogle Scholar
  270. 270.
    Yang G-J, Li C-J, Huang X-C, Wang Y-Y, Li C-X (2007) Influence of silver doping on photocatalytic activity of liquid flame sprayed nanostructured TiO2 coating. In: Marple BR et al (eds) Thermal spray 2007: global coating solutions. ASM International, Materials Park, OH, e-proceedingsGoogle Scholar
  271. 271.
    Jaworski R, Pawlowski L, Roudet F, Kozerski S, Le Maguer A (2008) Recent developments in suspension plasma sprayed titanium oxide and hydroxyapatite coatings influence of suspension plasma spraying process parameters on TiO2 coatings microstructure. J Therm Spray Technol 17(1):240–247Google Scholar
  272. 272.
    Toma F-L, Bertrand G, Begin S, Meunier C, Barres O, Klein D, Coddet C (2006) Microstructure and environmental functionalities of TiO2-supported photocatalysts obtained by suspension plasma spraying. Appl Catalysist B 68:74–84Google Scholar
  273. 273.
    Toma F-L, Bertrand G, Chwa SO, Meunier C, Klein D, Coddet C (2006) Comparative study on the photocatalytic decomposition of nitrogen oxides using TiO2 coatings prepared by conventional plasma spraying and suspension plasma spraying. Surf Coat Technol 200:5855–5862Google Scholar
  274. 274.
    Podlesak H, Pawlowski L, Laureyns J, Jaworski R, Lampke T (2008) Advanced microstructural study of suspension plasma sprayed titanium oxide coatings. Surf Coat Technol 202:3723–3731Google Scholar
  275. 275.
    Tomaszek R, Znamirowski Z, Pawlowski L, Zdanowski J (2007) Effect of conditioning on field electron emission of suspension plasma sprayed TiO2 coatings. Vacuum 81:1278–1282Google Scholar
  276. 276.
    Znamirowski Z, Ladaczek M (2008) Lighting segment with field electron titania cathode made using suspension plasma spraying. Surf Coat Technol 202:4449–4452Google Scholar
  277. 277.
    Kozerski S, Toma F-L, Pawlowski L, Leupolt B, Latka L, Berger L-M (2010) Suspension plasma sprayed TiO2 coatings using different injectors and their photocatalytic properties. Surf Coat Technol 205:980–986Google Scholar
  278. 278.
    Bannier E, Darut G, Sánchez E, Denoirjean A, Bordes MC, Salvador MD, Rayón E, Ageorges H (2011) Microstructure and photocatalytic activity of suspension plasma sprayed TiO2 coatings on steel and glass substrates. Surf Coat Technol 206:378–386Google Scholar
  279. 279.
    Yuan J, Zhan Q, Lei Q, Ding S, Li H (2012) Fabrication and characterization of hybrid micro/nanostructured hydrophilic titania coatings deposited by suspension flame spraying. Appl Surf Sci 258:6672–6678Google Scholar
  280. 280.
    Toma F-L, Berger L-M, Jacquet D, Wicky D, Villaluenga I, de Miguel YR, Lindeløv JS (2009) Comparative study on the photocatalytic behaviour of titanium oxide thermal sprayed coatings from powders and suspensions. Surf Coat Technol 203(15):2150–2156Google Scholar
  281. 281.
    Vaßen R, Yi Z, Kaßner H, Stöver D (2009) Suspension plasma spraying of TiO2 for the manufacture of photovoltaic cells. Surf Coat Technol 203(15):2146–2149Google Scholar
  282. 282.
    Moroz NA, Umapathy H, Mohanty P (2010) Synthesis and microstructure evolution of nano-titania doped silicon coatings. J Therm Spray Technol 19(1–2):294–302Google Scholar
  283. 283.
    Bolelli G, Cannillo V, Gadow R, Killinger A, Lusvarghi L, Rauch J (2009) Properties of high velocity suspension flame sprayed (HVSFS) TiO2 coatings. Surf Coat Technol 203:1722–1732Google Scholar
  284. 284.
    Gutzmann H, Kliemann J-O, Albrecht R, Gärtner F, Klassen T, Toma F-L, Berger L-M, Leupolt B (2010) Evaluation of the photocatalytic activity of TiO2-coatings prepared by different thermal spray techniques. In: Thermal spray: global solutions for future application. DVS-Berichte, Düsseldorf, Germany, pp 182–186Google Scholar
  285. 285.
    Rauch J, Stiegler N, Killinger A, Gadow R (2009) High velocity suspension flame spraying (HVSFS): process development and industrial applications. In: Marple BR et al (eds) Thermal spray 2009: expanding thermal spray performance to new markets and applications. ASM International, Materials Park, OH, pp 150–155Google Scholar
  286. 286.
    Chen D, Jordan EH, Gell M (2008) Porous TiO2 coating using the solution precursor plasma spray process. Surf Coat Technol 202:6113–6119Google Scholar
  287. 287.
    Tomaszek R, Pawlowski L, Gengembre L, Laureyns J, Le Maguer A (2007) Microstructure of suspension plasma sprayed multilayer coatings of hydroxyapatite and titanium oxide. Surf Coat Technol 201:7432–7440Google Scholar
  288. 288.
    Łatka L, Pawlowski L, Chicot D, Pierlot C, Petit F (2010) Mechanical properties of suspension plasma sprayed hydroxyapatite coatings submitted to simulated body fluid. Surf Coat Technol 205:954–960Google Scholar
  289. 289.
    D’Haese R, Pawlowski L, Bigan M, Jaworski R, Martel M (2010) Phase evolution of hydroxyapatite coatings suspension plasma sprayed using variable parameters in simulated body fluid. Surf Coat Technol 204(8):1236–1246Google Scholar
  290. 290.
    Kozerski S, Pawlowski L, Jaworski R, Roudet F, Petit F (2010) Two zones microstructure of suspension plasma sprayed hydroxyapatite coatings. Surf Coat Technol 204(9–10):1380–1387Google Scholar
  291. 291.
    Podlesak H, Pawlowski L, dHaese R, Laureyns J, Lampke T, Bellayer S (2010) Advanced microstructural study of suspension plasma sprayed hydroxyapatite coatings. J Therm Spray Technol 19(3):657–664Google Scholar
  292. 292.
    Stiegler N, Killinger A, Gadow R (2010) Hydroxyapatite coatings for biomedical applications deposited by different thermal spray techniques. Surf Coat Technol 205:1157–1164Google Scholar
  293. 293.
    Stiegler N, Bellucci D, Bolelli G, Cannillo V, Gadow R, Killinger A, Lusvarghi L, Sola A (2012) High-velocity suspension flame sprayed (HVSFS) hydroxyapatite coatings for biomedical applications. J Therm Spray Technol 21(2):275–287Google Scholar
  294. 294.
    Bolelli G, Cannillo V, Gadow R, Killinger A, Lusvarghi L, Rauch J (2009) Microstructural and in vitro characterization of high-velocity suspension flame sprayed (HVSFS) bioactive glass coatings. J Eur Ceram Soc 29:2249–2257Google Scholar
  295. 295.
    Altomare L, Bellucci D, Bolelli G, Bonferroni B, Cannillo V, De Nardo L, Gadow R, Killinger A, Lusvarghi L, Sola A, Stiegler N (2011) Microstructure and in vitro behaviour of 45S5 bioglass coatings deposited by high velocity suspension flame spraying (HVSFS). J Mater Sci Mater Med 22:1303–1319Google Scholar
  296. 296.
    Xiao Y, Song L, Liu X, Huang Y, Huang T, Chen J, Wu Y, Wu F (2011) Bioactive glass-ceramic coatings synthesized by the liquid precursor plasma spraying process. J Therm Spray Technol 20(3):560–568Google Scholar
  297. 297.
    Xiao Y, Song L, Liu X, Yi H, Huang T, Wu Y, Chen J, Wu F (2011) Nanostructured bioactive glass–ceramic coatings deposited by the liquid precursor plasma spraying process. Appl Surf Sci 257:1898–1905Google Scholar
  298. 298.
    Bolelli G, Rauch J, Cannillo V, Killinger A, Lusvarghi L, Gadow R (2008) Investigation of high-velocity suspension flame sprayed (HVSFS) glass coatings. Mater Lett 62:2772–2775Google Scholar
  299. 299.
    Waltz F, Swider MA, Hoyer P, Hassel T, Erne M, Möhwald K, Adlung M, Feldhoff A, Wickleder C, Bach F-W, Behrens P (2012) Synthesis of highly stable magnesium fluoride suspensions and their application in the corrosion protection of a magnesium alloy. J Mater Sci 47:176–183Google Scholar
  300. 300.
    Chien K, Coyle TW (2007) Rapid and continuous deposition of porous nanocrystalline sno2 coating with interpenetrating pores for gas sensor applications. J Therm Spray Technol 16(5–6):886–892Google Scholar
  301. 301.
    Ma XQ, Roth J, Gandy DW, Frederick GJ (2006) A new high-velocity oxygen fuel process for making finely structured and highly bonded Inconel alloy layers from liquid feedstock. J Therm Spray Technol 15(4):670–675Google Scholar
  302. 302.
    Kitamura J, Tang Z, Mizuno H, Sato K, Burgess A (2011) Structural, mechanical and erosion properties of yttrium oxide coatings by axial suspension plasma spraying for electronics applications. J Therm Spray Technol 20(1–2):170–185Google Scholar
  303. 303.
    Vert R, Chicot D, Dublanche-Tixier C, Meillot E, Vardelle A, Mariaux G (2010) Adhesion of YSZ suspension plasma-sprayed coating on smooth and thin substrates. Surf Coat Technol 205:999–1003Google Scholar

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  • Pierre L. Fauchais
    • 1
  • Joachim V. R. Heberlein
    • 2
  • Maher I. Boulos
    • 3
  1. 1.Sciences des Procédés Céramiques et de Traitements de Surface (SPCTS)Université de LimogesLimogesFrance
  2. 2.Department of Mechanical EngineeringUniversity of MinnesotaMinneapolisUSA
  3. 3.Department of Chemical EngineeringUniversity of SherbrookeSherbrookeCanada

Personalised recommendations