HVOF sprayed Al–Cu–Cr quasicrystalline coatings from coarse feedstock powders

    HVOF sprayed Al-Cu-Cr quasicrystalline coatings from coarse feedstock powders Yingqing Fu, Tianxiang Peng, Deming Yang, Chengqi Sun, Yuzhen Chen, Yang Gao PII: DOI: Reference:

S0257-8972(14)00403-4 doi: 10.1016/j.surfcoat.2014.05.003 SCT 19384

To appear in:

Surface & Coatings Technology

Received date: Revised date: Accepted date:

25 October 2013 21 April 2014 5 May 2014

Please cite this article as: Yingqing Fu, Tianxiang Peng, Deming Yang, Chengqi Sun, Yuzhen Chen, Yang Gao, HVOF sprayed Al-Cu-Cr quasicrystalline coatings from coarse feedstock powders, Surface & Coatings Technology (2014), doi: 10.1016/j.surfcoat.2014.05.003

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT HVOF sprayed Al-Cu-Cr quasicrystalline coatings from coarse feedstock powders Yingqing Fu*, Tianxiang Peng, Deming Yang, Chengqi Sun, Yuzhen Chen, Yang Gao

T

Department of Materials Science and Engineering, Dalian Maritime University, Dalian, 116026, China

RI P

*Corresponding author. Tel: +86-411-84723586; Fax: +86-411-84729611. E-mail: [email protected], [email protected]

Abstract

SC

Al-Cu-Cr quasicrystalline (QC) coatings were prepared from coarse feedstock powders by

NU

a high velocity oxy-fuel (HVOF) system DJ2700. The contrast experiments were performed by low power atmospheric plasma spray (LPAPS) using the same Al65Cu20Cr15 QC powders,

MA

and by HVOF spraying Cu180 powders with the size of 44-61 μm. The phase composition, microstructure and microhardness properties of the QC coatings were investigated. XRD

ED

results showed that the feedstock and coatings contained a predominant phase, icosahedral quasicrystal (IQC) I-Al65Cu24Cr11, and three minor crystalline phases: α-Al69Cu18Cr13, θ-

PT

Al13Cr2 (i.e.Al83Cu4Cr13) and ε-Al2Cu3. A qualitative analysis on the XRD patterns indicated

AC CE

that, a HVOF-sprayed coating contained more IQC and fewer crystalline phases in the one deposited by LPAPS. Moreover, the higher the input heat energy, the fewer IQC and the more crystalline phase the coating contained, for not only HVOF but also LPAPS. As confirmed by experimental results, when the HVOF-sprayed QC particle impacted onto substrate surface, unmelted solid part of particle broke up, and the previously deposited coating portion was deformed and densified and even cracked by impingement of the in-flight particles with high velocity; which minified splats and densified the as-sprayed coating. Thus, the HVOF-sprayed coatings from such coarse (61-74 μm) QC powders had much smaller splats and much denser microstructures with lower porosity and higher microhardness, compared with those deposited by LPAPS using the same feedstock, although the preferential particle size of feedstock powders for HVOF spraying is conventionally 5-45 μm. Furthermore, based on the contrast experimental results, the necessary and sufficient condition for occurrence of the particle -1-

ACCEPTED MANUSCRIPT impact breakage behavior in the thermal spray process are: (1) the high brittleness of feedstock and (2) the high velocity and low melting degree of spray particles, respectively. Keywords: HVOF thermal spray; Quasicrystal; Particle breakage; Microstructure

T

1. Introduction

RI P

Since being firstly reported in 1984 [1, 2], quasicrystalline (QC) materials have become the subject of intense study due to their exceptional structure and properties [3-9]. After 30 years

SC

of quasicrystal studies, the focus of research is currently shifting closer to the reality; much

NU

interest is nowadays concentrated on finding practical production techniques and applications for these materials. The established technology of aluminium fabrication makes the Al-based

MA

QC alloys more attractive than many other quasicrystalline alloys [10]. However, because quasicrystals are brittle in bulk at ambient temperature, most proposed applications employ

ED

QC films/coatings [4-6]. Depositing QC coatings onto metallic substrates allows the advantages of QC alloys, such as their surface properties, to be emphasized, while their

AC CE

materials.

PT

disadvantages, such as room-temperature brittleness, can be compensated for by the substrate

Various techniques are available for QC coating production, such as magnetron sputtering [11], simultaneous vapor deposition [12], laser ablation [13], ion-beam mixing [14], ion implantation [15], thermal spraying [16-21], and so on. Compared with the other techniques, thermal spraying are more versatile in industrialized QC coating production, with a more extensive adaptability; and the coating fabricated by these techniques is more robust [22] and wear-resistant than the one by the others. Moreover, a commercial application of QC coating has emerged, and the product is a cookware surface coating named Cybernox®, deposited by thermal spraying techniques [20]. Thermal spraying techniques are divided, according to the way the energy or heat is provided to melt the material or given it enough plasticity, to allow the formation of the coatings [23]. They usually include flame spraying, High-Velocity Oxy-Fuel (HVOF) -2-

ACCEPTED MANUSCRIPT spraying, detonation spraying, wire arc spraying, and plasma spraying and so on. Widely used for many industrial applications, HVOF spraying is probably the best among thermal spray techniques for some specific needs [24]. It has been well understood that the highest particle

T

velocity can be achieved by spray particles with heat source of the highest velocity; and the

RI P

highest particle temperature is associated with heat source of the highest temperature [25, 26]. Because the maximum temperature of plasma jet reaches over 10,000 K [27], spray particles

SC

can be heated to a high temperature by plasma jet with a medium velocity. Although HVOF is

NU

characterized by high velocity and low flame temperature [27], the flame temperature is significantly influenced by type of fuel gas and flame conditions. As shown in [28], the

MA

HVOF flame temperature using propane as fuel could reach to over 3100 K, and the spray particle velocity in it ranged from 300 to 1200 m/s [29]. HVOF-sprayed coatings are

ED

commonly thick and dense with less oxidation [25, 26], and typically with reduced changes in phase composition, compared with atmospheric plasma-sprayed ones [30]. Due to the above-

PT

mentioned features, HVOF spraying is advantageous for fabrication of QC coatings, and

AC CE

developing HVOF-sprayed QC coatings for industrial applications may be interesting and valuable.

It can be recognized that the heating and accelerating rates of spray particles increase with decreasing particle size [31]. It has also been found that the oxygen content in HVOF-sprayed alloy coatings increases in an exponential fashion with the decrease of spray particle size, and the oxidation of spray alloy particles becomes remarkably severe when the particle size is smaller than 45 μm [32]. Thus, in the current study, the Al65Cu20Cr15 QC powders with a size distribution ranged from 61 to 74 μm, were chosen as feedstock, although the preferential particle size of feedstock powders for HVOF spraying is conventionally 5-45 μm [26]. Such coarse powders, which reduced the vaporization of aluminum and oxidation of spray particles to facilitate the formation of Al-based QC phases, were HVOF and low power atmospheric plasma [33-37] sprayed respectively. Another finer feedstock, the Cu180 powder with the size -3-

ACCEPTED MANUSCRIPT of 44-61 μm, was HVOF-sprayed using identical process parameters. Based on the contrast experiments, the breakage behavior of HVOF-sprayed QC particles after impacting onto the AISI 1045 steel substrate was examined. The influence of spray particle impact breakage on

T

the coating microstructure, the necessary and sufficient condition for its occurrence in the

RI P

thermal spray process, were clarified respectively in this paper (which is extremely different from the previous paper [36]). Furthermore, no study on the impact breakage of HVOF-

SC

sprayed QC particles has been reported.

NU

2. Experimental details

MA

2.1 Spraying materials and process

Scanning electron microscope (SEM) micrographs of the feedstock powders are shown in

ED

Fig. 1. The spherical QC powders (Fig. 1a), produced by gas atomization of a liquid melt with a nominal composition of Al65Cu20Cr15 [36-38], had the size ranged from 61 to 74 μm. Another

PT

water atomized feedstock, Cu180 powders (Fig. 1b) for the contrast experiment with the size of 44-61 μm, had a nominal composition of Cu-5wt.%Ni-10wt.%Al. Both the feedstocks were

AC CE

thermally sprayed onto the previously grit-blasted AISI 1045 steel substrates respectively, with an oxygen-propane HVOF torch DJ2700 (Diamond Jet 2700-hybrid, Sulzer Metco, Westbury, NY, USA), and the HVOF spray parameters are presented in Table 1. The low power atmospheric plasma spray (LPAPS) parameters of the same Al65Cu20Cr15 QC powders have been reported in a previous paper [36]. The reaction equation of the complete combustion of propane is as follow: C3 H 8  5O2  4 H 2 O  3CO2

(1)

According to this equation, the stoichiometric propane-oxygen ratio is 1/5 = 0.2, where the flame enthalpy is maximum. Two spray conditions, reducing (R) and oxidizing (O), were used. Reducing condition was propane rich and oxidizing was oxygen rich. Since 20% oxygen is present in the high-pressure air used to transport the feedstock powders, the fuel-oxygen -4-

ACCEPTED MANUSCRIPT ratios of the reducing and oxidizing flame were set at 0.204 and 0.196 respectively, as listed in Table 1. The thickness of the as-sprayed coatings ranged from 200 to 900 μm.

T

2.2 Composition and structure examination

RI P

X-ray diffraction (XRD) was done using an X-ray diffractometer (Rigaku D/MAXUltima+/PC) with a Cu anode (λ=0.15406 nm). A Philips XL30 scanning electron microscope

SC

(SEM) was used to characterize the topographic morphologies of the feedstock and coatings, and the Al and O content in the coatings were estimated by its energy dispersive X-ray

NU

spectroscopy (EDS). The coating cross sections were observed and photographed by an Olympus GX51F computerized light microscope, whose Olycia m3 software was used to

MA

evaluate the coating porosity. A MH-6 hardness tester was used to measure the microhardness at a load of 100 g and a dwell time of 5 s, and Vickers indentation marks were performed on

PT

3. Results and Discussion

ED

10 different locations for each sample.

3.1 Phase composition

AC CE

There are two types of QC phases in Al-Cu-Cr system, icosahedral quasicrystal (IQC) and decagonal quasicrystal (DQC). IQC forms in rapidly solidified alloys with a nominal composition of Al65Cu20Cr15 [38-40]. Simple cubic Al65Cu24Cr11 IQC and face-centered cubic Al69Cu21Cr10 IQC were observed in [38]. DQCs with two different periodicities along the tenfold axis, 1.24 and 3.72 nm were found in rapidly solidified Al67Cu18Cr15 [41]. The DQC composition in [42, 43] varies in the range of 71-73 at.% Al, 11-12 at.% Cu and 15-18 at.% Cr, and differs from that in Al70.5Cu18Cr11.5 [44], Al71.4Cu19.2Cr9.4 [45] and Al78Cu10Cr12 [38]. Friedel [46] pointed out that the QC and approximant phases are Hume-Rothery phases, so they arise preferentially in the characteristic ranges of the valence concentration e/a, known as the Hume-Rothery rule [44-46]. Although the DQCs mentioned above form over a wide compositional range, they exist on or near to a line at constant valence concentration e/a=2.09, -5-

ACCEPTED MANUSCRIPT called the “decagonal QC line” in the Al-Cu-Cr composition-structure diagram [38, 39]. Similarly, Al-Cu-Cr IQCs are formed by conforming to the “icosahedral QC line” with e/a=1.86 [38, 39].

T

As mentioned in 2.1, the spraying feedstock was fabricated by gas atomization, and had a

RI P

nominal composition of Al65Cu20Cr15 [36-38], which determined that the QC phase in the feedstock and thermal sprayed coatings should be Al65Cu24Cr11 IQC, instead of a DQC.

SC

Besides this main phase I-Al65Cu24Cr11 (I), three minor phases are also found in the phase

NU

assemblage. As the approximants of Al65Cu24Cr11 IQC, α-Al69Cu18Cr13 [38] and ε-Al2Cu3 [47] are 1/1-type and B2-based respectively. θ-Al13Cr2 (i.e.Al83Cu4Cr13) has a monoclinic structure

MA

with lattice parameter a=2.52 nm, b=0.76 nm, c=1.09 nm, β=128.7° [38, 39]. Imitating the previous paper [36], XRD peak intensity ratios (PIR) of α/I, ε/I and θ/I [48,

ED

49], shown in Table 2 and Fig. 2 were calculated, based upon a qualitative analysis of XRD patterns of the Al65Cu20Cr15 QC feedstock and as-sprayed coatings. And the value of PIR had

AC CE

[48, 49].

PT

been confirmed to roughly equal the volume fraction of each crystalline phase in the sample

In the present work, the formation of IQC follows two mechanisms expatiated in [36] , which are unnecessary to repeat here. Compared with the LPAPS jet [36], the HVOF flame has a relatively low temperature [27], which decreases the vaporization of aluminum, and facilitates the formation of IQC rather than any crystalline phase (α, θ or ε). Furthermore, the feedstock particles accelerated to a much higher speed [27] by HVOF flame in the nozzle inside, resides a much briefer time in the nozzle outside to expose in the oxidizing atmosphere, so the oxidation of particles decreases. Consequently, the HVOF-sprayed coatings contained more IQC and fewer crystalline phases (α, θ and ε), compared with the ones deposited by LPAPS (except PS1), as shown in Table 2 and Fig. 2. Even though the same thermal spray technique (HVOF or LPAPS) was used, the energy increase of the heat source caused a rapid growth of heat transfer to in-flight particles, promoted the vaporization of aluminum and -6-

ACCEPTED MANUSCRIPT oxidation of particles (Fig. 3), reduced the formation of IQC, and made cooling rate of droplets or splats so “slow” that some crystalline phases were prone to form. So the higher the input energy was, the fewer IQC and the more crystalline phase (α, θ or ε) the coating

T

contained.

RI P

3.2 Microstructure

The coating cross-sectional SEM micrographs are shown in Fig. 4. PS3 and PS4 comprised

SC

layers of deformed particles, in which porosities, cracks and remains of semi-molten particles

NU

were observed and regarded as typical features of plasma sprayed coatings, shown in Fig. 4(a) and (b). The microstructure characteristics of the coating deposited by LPAPS and its changes

MA

with increasing heat source energy have been discussed in the previous paper [36]. As shown in Fig. 4(c) and (d), porosities, microcracks and partially-melted particles were also detected

ED

in the HVOF-sprayed coatings, but these microstructures were much smaller-sized than those of the coatings deposited by LPAPS [36].

PT

Generally, the density of thermal spray coating depends not only on the state parameters of

AC CE

the feedstock particles, including structure, size, velocity and temperature before impinging on a substrate, but also on the subsequent flattening and solidification of liquid or partially liquefied droplets after collision onto a substrate; and ultimately affects the coating microstructures and properties [50,51]. In the present experiments, because the feedstock QC powders for HVOF and LPAPS were the same, the structure and size of the feedstock particles could be regarded as identical, and their influences could be neglected. The microstructural difference of the as-sprayed coatings should be determined by the velocity and temperature of in-flight particles, as well as the flattening and solidification of liquid or partially liquefied droplets after impact onto the substrate. Compared with the Al65Cu20Cr15 QC particles in LPAPS process [36], those in HVOF spraying had a much higher average velocity and a much lower average temperature [27], leading to insufficient melting of the particles, as confirmed by the SEM micrographs of -7-

ACCEPTED MANUSCRIPT the coating surface morphologies shown in Fig. 5. The molten fraction of spray material plays an important role on producing a high density coating [52], so the low melting degree of HVOF-sprayed Al65Cu20Cr15 QC particles facilitates a low coating density. However, a higher

T

coating density than that by LPAPS could be obtained using the same QC powders (Fig. 4),

RI P

because the in-flight velocity of Al65Cu20Cr15 QC particle reached to 400-700 m/s by the DJ2700 HVOF system, in comparison of 200-400 m/s by LPAPS [36], which were estimated

SC

from [27, 52]. In the present cases, pores in the coatings were originated not only from

NU

insufficient filling of droplet materials into the cavities of previously deposited coating surface, but also from insufficient densification through deformation of unmelted particle

MA

fractions [52]. High particle velocities of HVOF benefited the decrease of apparent pores of the resultant coating [53], overwhelmed the disadvantage of low melting degree and induced a

ED

high coating density.

In addition, when HVOF-sprayed Al65Cu20Cr15 QC particles impacted onto the substrate or

PT

previously deposited coating surface, the unmelted particle part broke up due to the inherent

AC CE

brittleness of quasicrystals; and the previously deposited portion was deformed, densified and even cracked by high-velocity-particle impingements. The particle breakages and deformations of particles and the previously deposited coating upon high-velocity-particle impacts, as shown in Fig. 5(a), minified splats, facilitated sufficient filling of droplet materials into the surface cavities and densified the coating. 3.3 Particle impact breakage behavior The impact breakage behavior of HVOF-sprayed Al65Cu20Cr15 QC particles significantly affects the microstructure and properties of the resultant coating. It is valuable and essential to clarify the occurrence conditions for this behavior. Hence, the contrast experiments were performed by LPAPS [36] with the same feedstock, and by HVOF spraying Cu180 powders. Three typical coatings—HF2, HF3 (shown in Table 1) and PS4 (shown in Table 2 and [36]) -8-

ACCEPTED MANUSCRIPT were chosen for comparison. Their SEM surface morphologies and cross-sectional light microscopy images are shown in Fig. 5 and Fig. 6, respectively, and their cross sections were etched with 10wt.% NaOH-water solution after polishing treatment, to make the splats and

T

microstructure in them more visible.

RI P

Under otherwise identical conditions, the feedstock for HF3 has the smaller-sized powders, much higher thermal conductivity and toughness than that for HF2, so the microstructure of

SC

HF3 conventionally may be finer and denser than that of HF2. On the contrary, HF2 presents

NU

a much finer microstructure with much smaller splats compared with HF3, shown in Fig. 6(a) and (b). Why? It positively attributes to the particle breakages (“A” in Fig. 5(a)) and

MA

deformations of particles (“B” in Fig. 5(a)) and the previously-deposited coating portion (“C” in Fig. 5(a)) upon high-velocity-particle impacts. This particle breakage phenomenon can be

ED

clearly confirmed in Fig. 5(a), but nearly imperceptible in Fig. 5(b). Based on the results mentioned above, the high brittleness of feedstock powders (e.g. quasicrystal) should be the

PT

necessary condition for occurrence of the particle impact breakage in thermal spray process.

AC CE

In other words, no high-brittleness feedstock, no visible particle breakage in thermal spraying. HF2 also gives a much finer microstructure than that of PS4, as shown in Fig. 4(b), 4(d), 7(a) and 7(c), although they were fabricated with the same feedstock. The finer microstructure of HF2 directly relates to the characteristics of HVOF-sprayed particles. As mentioned above, HVOF-sprayed QC particles had a much higher average velocity and melting degree than those of LPAPS. Moreover, Al65Cu20Cr15 QC powders have such coarse particles with a very low thermal conductivity (≤ 1/5 that of AISI 304 stainless steel) [39], which can aggravate insufficient melting. Therefore, most in-flight QC particles (>80%) [54] in HVOF were the solid-liquid two-phase particles similar to those reported by Li et al. [51,52,55,56], where unmelted cores and completely melted surface layers were at solid and liquid state, respectively. However, most QC particles of LPAPS [36] melted more sufficiently than those in HVOF, and some small-sized particles even completely melted into single-phase liquid -9-

ACCEPTED MANUSCRIPT droplets. The great difference of particle melting degree can be confirmed in Fig. 5(a) and (c), but the particle-breakage phenomenon mentiond above cannot be perceptible in Fig. 5(c), which indicates the high melting degree and low velocity of in-flight particles of LPAPS

T

significantly suppressed the particle breakage. Hence, the sufficient condition for occurrence

RI P

of thermal spray particle impact breakage may be the high velocity and low melting degree of

SC

spray particles (e.g. HVOF-sprayed low thermal conductivity powders). 3.4 Porosity and microhardness

NU

The porosity and Vickers microhardness (HV0.1) of as-sprayed QC coatings are shown in Table 3. In order to decrease the negative effects of porosity on the accuracy of hardness

MA

examination, the microhardness value in Table 3 is an average of 10 different locations on each sample. Gas porosity and shrinkage porosity, 2 main types of porosity in thermal sprayed

ED

coatings, as well as their formation mechanism, had been discussed before [36, 57, 58], and

PT

should not be repeated here. Significant phase assemblage and atomic volume differences between QC and crystalline phases facilitate the formation of porosity [59, 60]. Therefore, the

AC CE

QC coatings in the present study may have higher porosities, than the Al-Cu-Cr crystalline coatings thermally sprayed by the same technique, despite having similar elemental compositions. The HVOF-sprayed QC coatings had a much lower porosity compared with those by LPAPS[36], which related to the finer and denser coating microstructures resulted from the particle impact breakage mentioned above. The rise in HVOF energy promoted the oxidation of particles and vaporization of aluminum, as illustrated in Fig. 3, thus inducing the pores. However, the growth of particle velocity facilitated the particle impact breakage, minified splats, increased splat flattening ratio and adhesion among splats to suppress interlamellar shrinkage porosity formation, and consequently, diminished the coating porosity. Under the present experimental conditions, the latter effect may overwhelm the former, thereby causing the coating porosity decrease with increasing the HVOF energy, so HF2 had a lower porosity than that of HF1. The coating microhardness depended on the coating phase - 10 -

ACCEPTED MANUSCRIPT composition and microstructure, and increased with the coating porosity decrease in present study, which agreed well with the experimental results of E. Fleury et al. [61]. As a result, HF2 had a higher microhardness than that of HF1, though HF1 contained more IQC that

RI P

T

contributes to hardness (Fig. 3). 4. Conclusions

SC

Al-Cu-Cr QC coatings were prepared by a HVOF DJ2700 system. The influences of heat source energy and particle impact breakage on the phase constituent, microstructure and

NU

hardness properties of the coatings were investigated. The results are summarized as follow: 1. Compared with the one deposited by LPAPS using the same Al65Cu20Cr15 QC feedstock,

MA

(1) the HVOF-sprayed coating contained more IQC, due to the less oxidation and vaporization of the feedstock; and (2) presented a lower porosity and higher hardness with

ED

a much denser microstructure, which resulted from the particle impact breakage behavior.

PT

2. With increasing the HVOF energy, (1) the volume fraction of such crystalline phases as ε, α and θ in the coating increased, but that of the IQC decreased, because the higher energy

AC CE

reduced the cooling rate and increased the particle oxidation and vaporization of aluminum; (2) the coating porosity decreased whilst its microhardness increased due to the increase of the splat flattening ratio and the adhesion among minified splats. 3. The particle impact breakage behavior can minify splats, deform and densify the coating. The necessary and sufficient condition for its occurrence in the thermal spray process are: (1) the high brittleness of feedstock and (2) the high velocity and low melting degree of spray particles, respectively. Acknowledgement The authors are grateful to the financial support by the National Natural Science Foundation of China (50805012, 51172033 and 21276036) and the Fundamental Research Funds for the Central Universities (3132013056 and 3132013311). - 11 -

ACCEPTED MANUSCRIPT References [1] D. Shechtman, I. Blech, D. Gratias, J.W. Cahn, Phys. Rev. Lett., 53 (1984) 1951. [2] L. Swartzendruber, D. Shechtman, L. Bendersky, J. Cahn, Phys. Rev. B, 32 (1985) 1383. [3] T. Klein, A. Gozlan, C. Berger, F. Cyrot-Lackmann, Y. Calvayrac, A. Quivy, Europhys.

T

Lett., 13 (1990) 129.

RI P

[4] M. Sales, A. Merstallinger, A. Ustinov, S. Polishchuk, T. Melnichenko, Surf. Coat. Technol., 201 (2007) 6206.

SC

[5] W. Yuan, T. Shao, E. Fleury, D. Chen, Surf. Coat. Technol., 185 (2004) 99. [6] J.-M. Dubois, Useful quasicrystals, in, World Scientific, 2005, pp. 305.

NU

[7] D.J. Sordelet, J. Dubois, MRS Bull., 22 (1997) 34.

[8] J. Kong, C. Zhou, S. Gong, H. Xu, Surf. Coat. Technol., 165 (2003) 281. [9] C. Zhou, R. Cai, S. Gong, H. Xu, Surf. Coat. Technol., 201 (2006) 1718.

MA

[10] E. Huttunen-Saarivirta, J. Alloys Compd., 363 (2004) 150. [11] S. Olsson, F. Eriksson, J. Birch, L. Hultman, Thin Solid Films, 526 (2012) 74. [12] R. Anton, P. Kreutzer, J. Alloys Compd., 342 (2002) 464.

ED

[13] R. Teghil, L. d’Alessio, A. Santagata, M. Zaccagnino, D. Ferro, D. Sordelet, Appl. Surf. Sci., 210 (2003) 307.

PT

[14] A. Traverse, E. Belin-Ferré, Z. Dankházi, L. Mendoza-Zélis, O. Laborde, R. Portier, J. Phys.: Condens. Matter, 8 (1996) 3843.

AC CE

[15] L. Chen, L. Wang, Appl. Phys. Lett., 70 (1997) 1825. [16] M. Besser, T. Eisenhammer, MRS Bull., 22 (1997) 59. [17] E. Fleury, Y.-C. Kim, J.-S. Kim, H.-S. Ahn, S.-M. Lee, W.-T. Kim, D.-H. Kim, J. Mater. Res., 17 (2002) 492.

[18] D. Sordelet, M. Besser, J. Logsdon, Mater. Sci. Eng. A, 255 (1998) 54. [19] E. Fleury, Y.-C. Kim, J.-S. Kim, D.-H. Kim, W. Kim, H.-S. Ahn, S.-M. Lee, J. Alloys Compd., 342 (2002) 321. [20] D. Sordelet, S. Widener, Y. Tang, M. Besser, Mater. Sci. Eng. A, 294 (2000) 834. [21] S. De Palo, S. Usmani, S. Sampath, D. Sordelet, M. Besser, Proceedings of the 1st United Thermal Spray Conference, ASM International, Indianapolis, USA, 1997, pp. 135. [22] C. Zhang, Y. Wu, L. Liu, Appl. Phys. Lett., 101 (2012) 121603. [23] P.L. Fauchais, J.V. Heberlein, M.I. Boulos, 1.4 Description of Different Thermal Spray Coating Processes, in: Thermal Spray Fundamentals, Springer, 2014, pp. 4. [24] V.V.V. Sobolev, J. Nutting, J. Guilemany, High velocity oxy-fuel spraying, in, Maney Pub., 2004, pp. 30. - 12 -

ACCEPTED MANUSCRIPT [25] J.R. Davis, Handbook of thermal spray technology, in, ASM international, 2004, pp. 43. [26] L. Pawlowski, The science and engineering of thermal spray coatings, in, John Wiley & Sons, 2008, pp. 89.

[28] S. Kamnis, S. Gu, Chem. Eng. Process., 45 (2006) 246.

T

[27] X. Liu, P.K. Chu, C. Ding, Mater. Sci. Eng. R, 47 (2004) 49.

RI P

[29] M. Verdian, K. Raeissi, M. Salehi, Appl. Surf. Sci., 273 (2013) 426. [30] B. Irving, R. Knight, R.W. Smith, Weld. J., 72 (1993) 25.

[31] M. Li, P.D. Christofides, Chem. Eng. Sci., 61 (2006) 6540.

SC

[32] C.-J. Li, W.-Y. Li, Surf. Coat. Technol., 162 (2003) 31.

[33] M. Morks, Y. Gao, Y. Fu, Surf. Coat. Technol., 199 (2005) 66.

NU

[34] L. An, Y. Gao, T. Zhang, J. Therm. Spray Technol., 16 (2007) 967. [35] Y. Gao, X. Xu, Z. Yan, G. Xin, Surf. Coat. Technol., 154 (2002) 189.

MA

[36] Y. Fu, L. An, F. Zhou, Y. Zhao, D. Yang, Y. Gao, Surf. Coat. Technol., 202 (2008) 4964. [37] Y.Q. Fu, F. Zhou, D.M. Yang, N.X. Li, Y. Gao, Mater. Sci. Forum, 610-613 (2009) 641. [38] Y. Qi, Z. Zhang, Z. Hei, C. Dong, J. Alloys Compd., 285 (1999) 221.

ED

[39] C. Dong, Quasicrystal Materials, in, China National Defence Industry Press, 1998, pp. 52. [40] A.-P. Tsai, A. Inoue, T. Masumoto, J. Mater. Sci. Lett., 7 (1988) 322.

PT

[41] J. Wu, Philos. Mag. Lett., 73 (1996) 163. [42] T. Okabe, J.-I. Furihata, K. Morishita, H. Fujimori, Philos. Mag. Lett., 66 (1992) 259.

AC CE

[43] J.-I. Furihata, T. Okabe, J. Electron Microsc., 48 (1999) 761. [44] A. Shevchukov, T. Sviridova, S. Kaloshkin, V. Tcherdyntsev, M. Gorshenkov, M. Churyukanova, D. Zhang, Z. Li, J. Alloys Compd., 586 (2014) S391. [45] T. Sviridova, A. Shevchukov, E. Shelekhov, D. Diakonov, V. Tcherdyntsev, S. Kaloshkin, J. Alloys Compd., 509 (2011) S299. [46] J. Friedel, F. Denoyer, Comptes rendus de l'Académie des sciences. Série 2, Mécanique, Physique, Chimie, Sciences de l'univers, Sciences de la Terre, 305 (1987) 171. [47] C. Dong, Q. Zhang, D. Wang, Y. Wang, Eur. Phys. J. B, 6 (1998) 25. [48] D. Sordelet, M. Kramer, O. Unal, J. Therm. Spray Technol., 4 (1995) 235. [49] D. Sordelet, M. Besser, I. Anderson, J. Therm. Spray Technol., 5 (1996) 161. [50] R. McPherson, Surf. Coat. Technol., 39 (1989) 173. [51] C.-J. Li, A. Ohmori, Y. Harada, J. Mater. Sci., 31 (1996) 785. [52] C.-J. Li, G.-J. Yang, Int. J. Refract. Met. Hard Mater, 39 (2013) 2. [53] H. de Villiers Lovelock, J. Therm. Spray Technol., 7 (1998) 357. [54] Y.Q. Fu, PhD Thesis, Dalian Maritime University, China (2010). - 13 -

ACCEPTED MANUSCRIPT [55] C.-J. Li, Y.-Y. Wang, G.-J. Yang, A. Ohmori, K. Khor, Mater. Sci. Technol., 20 (2004) 1087. [56] C.-J. Li, Y.-Y. Wang, T. Wu, G.-C. Ji, A. Ohmori, Surf. Coat. Technol., 145 (2001) 113. [57] V. Sobolev, J. Guilemany, J. Mater. Process. Technol., 58 (1996) 227.

T

[58] C.C. Berndt, W. Brindley, A. Goland, H. Herman, D. Houck, K. Jones, R. Miller, R.

RI P

Neiser, W. Riggs, S. Sampath, J. Therm. Spray Technol., 1 (1992) 341.

[59] V. Tcherdyntsev, S. Kaloshkin, A. Salimon, E. Leonova, I. Tomilin, J. Eckert, F. Schurack, V. Rogozin, S. Pisarev, Y.P. Trykov, Mater. Manuf. Processes, 17 (2002) 825.

SC

[60] M. Suarez, I. Figueroa, G. Gonzalez, G. Lara-Rodriguez, O. Novelo-Peralta, I. Alfonso, I. Calvo, J. Alloys Compd., 585 (2014) 318.

ED

MA

NU

[61] E. Fleury, S. Lee, W. Kim, D. Kim, J. Non-Cryst. Solids, 278 (2000) 194.

List of figure captions:

PT

Fig. 1 SEM micrographs of Al65Cu20Cr15 QC (a) and Cu180 (b) feedstock powders

AC CE

Fig. 2 XRD peak intensity ratio of α/I, ε/I and θ/I in the feedstock and as-sprayed coatings Fig. 3 Effect of aluminum and oxygen mean content in as-sprayed coatings on the loss of IQC (based on EDS analyses) Fig. 4 Cross-sectional SEM micrographs of as-sprayed Al-Cu-Cr QC coatings: (a)PS3, (b)PS4, (c)HF1, (d)HF2 Fig. 5 SEM micrographs of coating surface morphologies: (a) HF2, (b) HF3, (c) PS4 Fig. 6 Cross-sectional light microscopy images of coatings: (a) HF2, (b) HF3, (c) PS4

- 14 -

ACCEPTED MANUSCRIPT Table 1 High velocity oxy-fuel spray parameters

Al65Cu20Cr15 Al65Cu20Cr15 Cu180

Feed rate (g/min) 10 10 32

Compressed air flow (L/min) 256 262 262

Propane flow (L/min) 53 58 58

Oxygen flow Fuel-Oxygen (L/min) ratio 208 0.204 243 0.196 243 0.196

Spray distance (mm) 180 180 180

RI P

HF1 (R) HF2 (O) HF3 (O)

Particle size (μm) 61-74 61-74 44-61

T

Coating code Powder

SC

Table 2

XRD peak intensity ratio of α/I, ε/I and θ/I in the Al65Cu20Cr15 QC feedstock and as-sprayed coatings [36] PS2 LPAPS 56.5 28.7 14.8

PS3 LPAPS 61.7 31.0 16.0

PS4 LPAPS 66.8 34.6 16.9

PS5 LPAPS 73.5 38.5 18.3

NU

PS1 LPAPS 49.5 25.1 13.1

HF1 HVOF 47.3 20.5 13.9

HF2 HVOF 48.8 22.6 14.6

Feedstock 46.8 18.7 12.8

MA

Coating code Thermal spray technique θ/I peak intensity ratio (%) α/I peak intensity ratio (%) ε/I peak intensity ratio (%)

ED

Table 3

Cross-sectional porosity and microhardness HV0.1 of as-sprayed Al-Cu-Cr QC coatings [36] PS2 12±2 4.64±0.78

PS3 10±2 4.84±0.50

AC CE

PT

Coating code PS1 Porosity (%) 13±3 Microhardness HV0.1 (GPa) 4.58±0.58

- 15 -

PS4 8±2 4.98±0.70

PS5 15±4 4.75±0.66

HF1 8±3 5.16±0.45

HF2 6±2 5.62±0.36

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 1a

- 16 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 1b

- 17 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 2

- 18 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 3

- 19 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 4a

- 20 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 4b

- 21 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 4c

- 22 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 4d

- 23 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 5a

- 24 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 5b

- 25 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 5c

- 26 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 6a

- 27 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 6b

- 28 -

SC

RI P

T

ACCEPTED MANUSCRIPT

AC CE

PT

ED

MA

NU

Figure 6c

- 29 -

ACCEPTED MANUSCRIPT Highlights We examine particle breakage of Al65Cu20Cr15 quasicrystal coarse powders in HVOF.



Particle breakages after impacts minify splats, deform and densify the coating.



High brittleness of thermal spray particle is necessary for its impact breakage.



High velocity and low melting degree of particle is sufficient for its breakage.



Particle impact breakage contributes to a high-quality quasicrystalline coating.

AC CE

PT

ED

MA

NU

SC

RI P

T



- 30 -