Anti-Corrosion Coatings Fabricated by Cold Spray Technique: Optimization of Spray Condition and Relationship Between Microstructure and Performance

Anti-corrosion coatings fabricated by cold spray technique : Optimization of spray condition and relationship between microstructure and performance Elizaveta Lapushkina

To cite this version:

Elizaveta Lapushkina. Anti-corrosion coatings fabricated by cold spray technique : Optimization of spray condition and relationship between microstructure and performance. Materials. Université de Lyon; Tohoku gakuin university (Sendai, Japon), 2020. English. NNT : 2020LYSEI054. tel- 03127323

HAL Id: tel-03127323 https://tel.archives-ouvertes.fr/tel-03127323 Submitted on 1 Feb 2021

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés.

N°d’ordre NNT : 2020LYSEI054

THESE de DOCTORAT DE L’UNIVERSITE DE LYON

opérée au sein de L’INSTITUT NATIONAL DES SCIENCES APPLIQUEES DE LYON

En cotutelle internationale avec UNIVERSITE DE TOHOKU

Ecole Doctorale N° ED34 École doctorale materiaux de Lyon

Spécialité/ discipline de doctorat : Matériaux

Soutenue publiquement le 16/07/2020, par : Elizaveta Lapushkina

Anti-corrosion coatings fabricated by cold spray technique: Optimization of spray condition and relationship between microstructure and performance

Devant le jury composé de :

Hanlin, Liao, Professeur des université, Université de Bourgogne Franche-Comté, Rapporteur Bolot Rodolphe, Professeur des université, Université de Bourgogne-IUT Creusot, Rapporteur Philippe Bertrand, Professeur des université, ENISE, Université Jean Monnet-Saint- Étienne, Examinateur Marie-Pierre Planche, Maitre de conférences HDR, Université de Bourgogne Franche- Comté, Examinatrice Bernard Normand, Professeur des université, INSA de Lyon, Directeur de thèse Kazuhiro Ogawa, Professeur des université, Université de Tohoku, Co-directeur de thèse Sheng Yuan, Professeur agrégé, INSA de Lyon, Invité Nicolas Mary, Professeur agrégé, ElyTMax, Invité

Acknowledgements

First of all, I am extremely grateful to my main supervisors Prof. Bernard Normand and Prof. Kazuhiro Ogawa for their guidance throughout the research work from title’s selection to finding the results. Their deep insights helped me at various stages of my research and gave me more power and spirit to excel in the research writing.

I would like to express my deep gratitude to the examiners of my thesis, Prof. Hanlin Liao, Prof. Rodolphe Bolot, Assoc. Prof. Philippe Bertrand and Maitre de conférences HDR Marie-Pierre Planche for their suggestions, comments and tolerance. I hope that the present outcome will fulfil some of their expectations.

My sincere gratitude is reserved for Dr. Sheng Yuan and Dr. Nicolas Mary for sharing insightful suggestions and giving the encouragement. They have played a major role in improving my research skills.

I would also like to express the gratitude to ElyTMax, MATEIS/CorriS and Ogawa-lab who made my access simpler to the research facilities and laboratory and gave an opportunity to become part of their teams. It wouldn’t have been possible to conduct this research without their precious support.

In the end, I am grateful to my parents, siblings, friends and acquaintances who remembered me in their prayers for the ultimate success. They gave me enough moral support, encouragement and motivation to accomplish my objective without any obstacle on the way.

Département FEDORA – INSA Lyon - Ecoles Doctorales – Quinquennal 2016-2020

SIGLE ECOLEDOCTORALE NOM ET COORDONNEES DU RESPONSABLE

CHIMIE CHIMIE DE LYON M. Stéphane DANIELE

Institut de recherches sur la catalyse et l’environnement de Lyon

http://www.edchimie-lyon.fr IRCELYON-UMR 5256

Sec. : Renée EL MELHEM Équipe CDFA

Bât. Blaise PASCAL, 3e étage 2 Avenue Albert EINSTEIN

[email protected] 69 626 Villeurbanne CEDEX

INSA : R. GOURDON

[email protected]

E.E.A. ÉLECTRONIQUE, M. Gérard SCORLETTI

ÉLECTROTECHNIQUE, École Centrale de Lyon

AUTOMATIQUE 36 Avenue Guy DE COLLONGUE

69 134 Écully

http://edeea.ec-lyon.fr Tél : 04.72.18.60.97 Fax 04.78.43.37.17

Sec. : M.C. HAVGOUDOUKIAN [email protected]

[email protected]

E2M2 ÉVOLUTION, ÉCOSYSTÈME, M. Philippe NORMAND

MICROBIOLOGIE, MODÉLISATION UMR 5557 Lab. d’Ecologie Microbienne

Université Claude Bernard Lyon 1

http://e2m2.universite-lyon.fr Bâtiment Mendel

Sec. : Sylvie ROBERJOT 43, boulevard du 11 Novembre 1918

Bât. Atrium, UCB Lyon 1 69 622 Villeurbanne CEDEX

Tél : 04.72.44.83.62 [email protected]

INSA : H. CHARLES

[email protected]

EDISS INTERDISCIPLINAIRE Mme Sylvie RICARD-BLUM

SCIENCES-SANTÉ Institut de Chimie et Biochimie Moléculaires et Supramoléculaires

(ICBMS) - UMR 5246 CNRS - Université Lyon 1

http://www.ediss-lyon.fr Bâtiment Curien - 3ème étage Nord

Sec. : Sylvie ROBERJOT 43 Boulevard du 11 novembre 1918

Bât. Atrium, UCB Lyon 1 69622 Villeurbanne Cedex

Tél : 04.72.44.83.62 Tel : +33(0)4 72 44 82 32

INSA : M. LAGARDE

[email protected]

INFOMATHS INFORMATIQUE ET M. Hamamache KHEDDOUCI

MATHÉMATIQUES Bât. Nautibus

43, Boulevard du 11 novembre 1918

http://edinfomaths.universite-lyon.fr 69 622 Villeurbanne Cedex France

Sec. : Renée EL MELHEM Tel : 04.72.44.83.69

Bât. Blaise PASCAL, 3e étage [email protected]

Tél : 04.72.43.80.46

[email protected]

MATÉRIAUX DE LYON M. Jean-Yves BUFFIÈRE

Matériaux INSA de Lyon

http://ed34.universite-lyon.fr MATEIS - Bât. Saint-Exupéry

Sec. : Stéphanie CAUVIN 7 Avenue Jean CAPELLE

Tél : 04.72.43.71.70 69 621 Villeurbanne CEDEX

Bât. Direction Tél : 04.72.43.71.70 Fax : 04.72.43.85.28

[email protected] [email protected]

MEGA MÉCANIQUE, ÉNERGÉTIQUE, M. Jocelyn BONJOUR

GÉNIE CIVIL, ACOUSTIQUE INSA de Lyon

Laboratoire CETHIL

http://edmega.universite-lyon.fr Bâtiment Sadi-Carnot

Sec. : Stéphanie CAUVIN 9, rue de la Physique

Tél : 04.72.43.71.70 69 621 Villeurbanne CEDEX

Bât. Direction

[email protected]

ScSo ScSo* M. Christian MONTES

Université Lyon 2

http://ed483.univ-lyon2.fr 86 Rue Pasteur

Sec. : Véronique GUICHARD 69 365 Lyon CEDEX 07

INSA : J.Y. TOUSSAINT [email protected]

Tél : 04.78.69.72.76

[email protected]

*ScSo : Histoire, Géographie, Aménagement, Urbanisme, Archéologie, Science politique, Sociologie, Anthropologie

Table of contents

Acknowledgements ...... 1 Abstract ...... 6 Résumé ...... 7 Introduction ...... 8 Chapter I: Bibliography ...... 10 1.1 Cold Spray Technique ...... 10 1.1.1 Cold Spray Process ...... 10 1.1.2 Coating formation and microstructure (ASI) ...... 11 1.1.3 Effect of spraying parameters ...... 19 1.2 Cold Spray for corrosion protection ...... 22 1.2.1 Noble and sacrificial coatings. Corrosion mechanisms ...... 22 1.2.2 Influence of defects in coatings for corrosion protection ...... 26 1.3 Conclusion ...... 29 Chapter II. Experimental methods and equipments ...... 31 2.1 HPCS and LPCS equipment and parameters ...... 31 2.1.1High-pressure cold spray process...... 31 2.1.2 Low-pressure cold spray process ...... 34 2.2 Experimental methods ...... 35 2.3 Conventional macro-scale measurements, electrochemical 3-electrode cell ...... 38 2.3.1 Electrochemical 3-electrode micro-cell ...... 39 Chapter III ...... 45 3.1 Introduction ...... 45 3.2 Materials ...... 46 3.3 Preliminary tests. Parameters boundary conditions ...... 48 3.4 Analysis ...... 53 3.4.1 Coating roughness/waviness and surfaces ...... 53 3.4.2 Adhesion properties of coating ...... 56 3.5 Doehlert uniform shell design for Zn coatings on carbon steel ...... 59 3.5.1 Cross section Microstructures ...... 62 3.5.2 Coating porosities ...... 65 3.6 Corrosion performance of Zn cold spray coatings ...... 67 3.7 Discussions ...... 74 3.8 Conclusions ...... 77 Chapter IV: Relationship between the microstructure of Zn coating and corrosion behavior ...... 78

4.1 The strategy of research and coating parameters choosing ...... 78 4.2 Experimental part ...... 79 4.3 Analysis ...... 81 4.4 Conclusion ...... 96

Chapter V: Low-pressure cold sprayed Al and Al/Al2O3 coatings on carbon steels ...... 97 5.1 Introduction ...... 97 5.2 Used materials ...... 106 5.3 Experimental procedures ...... 109 5.3.1 Parameters optimization ...... 109 5.3.2 Laser treatment ...... 113 5.4 Characterization ...... 114 5.4.1 Microstructure characterization ...... 114 5.4.2 Porosity measurement ...... 119 5.4.3 Roughness ...... 120 5.4.4 Microhardness ...... 122 5.5 Corrosion performance ...... 124 General Conclusions ...... 130 Future work and perspectives ...... 132 Appendix ...... 133 References ...... 142

Abstract Anticorrosion coatings of Zinc and Aluminium were developed by high pressure and low-pressure Cold Spray techniques, respectively. For Zinc coatings, the dependence of spraying temperature on thickness as the important factor for corrosion resistance has been analyzed. The critical temperature of gas while deposition was found at 230 oC. For lower temperatures, the coating was considerably thinner. Dependence of thickness on pressure variation 2 MPa, 2,5 MPa and 3 MPa at constant working gas temperature 290 oC has shown the highest thickness value at 2 MPa. For all the experiments the substrate was kept at room temperature. It was confirmed that the coating thickness tends to decrease with the pressure rise. The powder feed rate as well as the spraying distance were also found to influence the thickness. The optimal conditions were found for 3 RPS and 30 mm, respectively. Finally, the gas temperature and pressure were optimized by a Doehlert uniform shell design. Their influences on the zinc coating quality were discussed in terms of microstructure, porosity, thickness, and corrosion resistance. A maximum porosity of 4.2% was reached with the highest pressure and with a moderate gas temperature (260 °C< T < 300 °C). These conditions promoted erosion of the substrate and a lower accommodation of particles at the impact. Thicker coatings were obtained at higher temperatures because of better particle straining. Two optimal conditions were then identified: 320 °C - 2.5 MPa and 260 °C–2.5 MPa. Macroscopic and local electrochemical experiments were performed. Higher corrosion resistance was detected for the condition 320 °C - 2.5 MPa. Coatings were thick enough to protect the substrate and the corrosion mechanism was driven by the classical Zn hydroxide and oxide layers. Note that the coating roughness may be optimized later to reduce the corrosion initiation. For aluminum coatings deposited by a low-pressure cold spray method, the optimal spraying parameters according to deposition efficiency were found at 400 °C /0.65 MPa. Ceramic particles were added to densify the coating and allowed to reduce porosity from 8% to 6.4%. Instead of ceramic particle addition, laser surface treatment was performed after coating design. Laser power was not high enough to reach the surface melting. However, the coating microhardness was modified. Results showed a microhardness increase of coatings of 5% with the addition of hard particles whereas the microhardness decreased after the post-heat treatment (pure aluminum coating reduction of 39% and for composite coating 35%). The hardness reduction during the laser treatment was attributed to surface annealing and the release of internal stresses and possible recrystallization with the subsequent grain growth. Finally, the results of the electrochemical investigations showed higher corrosion resistance of ceramic composite coatings in comparison with pure aluminum and laser- treated coatings.

Keywords: Cold Spray, Corrosion Protection, Electrochemical Characterization, Galvanic Corrosion, Microstructure, Porosity, Surface roughness, composite coating, laser treatment.

Résumé Des revêtements anticorrosion de zinc et d'aluminium ont été développés respectivement par des techniques de pulvérisation à froid à haute pression et à basse pression. Pour les revêtements de zinc, la dépendance de la température de pulvérisation sur l'épaisseur a été analysée et la température critique de dépôt a été trouvée à 230°C. Pour des températures plus basses, le revêtement était considérablement plus mince. Des variations de pression de 2 MPa, 2,5 MPa et 3 MPa à température constante 290 °C ont montré la valeur d'épaisseur de couche plus élevée à 2 MPa. Il a été également confirmé que l'épaisseur du revêtement à tendance à diminuer avec la pression. Le taux d'alimentation en poudre ainsi que la distance de pulvérisation ont également été considérés comme des paramètres influençant l'épaisseur. Les conditions optimales de projection ont été trouvées pour 3 rps et 30 mm, respectivement. Enfin, la température et la pression du gaz ont été optimisées par le plan d’experience dit de Doehlert. Leurs influences sur la qualité du revêtement de zinc ont été discutées en termes de microstructure, de porosité, d'épaisseur et de résistance à la corrosion. Une porosité maximale de 4,2% a été atteinte avec la pression la plus élevée et avec une température modérée (260 ° C

Mots-clés: Cold Spray, Protection contre la corrosion, Caractérisation électrochimique, Galvanic Corrosion, Microstructure, Porosité, la rugosité de surface, Les revêtements composite, Le traitement au laser

Introduction

Cold spray technology was developed in the mid-1980s in the Institute of Theoretical and Applied Mechanics in Novosibirsk, Russia [1-3]. Nevertheless, wide application has started at the end of 1990. At providing supersonic wind tunnel tests researches have noticed that at some critical velocity the abrasion of a substrate by particles decreases and deposition increases [1]. In the current decades the interest of cold spray technology increased exponentially. Hundreds of papers, scientific works and patents concerning cold spray technology, materials, physical phenomena and equipments were released [4].

A variety of pure materials on different substrates have been successfully deposited including metal alloys, polymers and composites. Note that .U.S. patent was issued in 1994 [2]. Recently, the deposition of special powder has attracted attention, cold spray coatings on the basis of nanocrystalline composites [5,6], nanocrystalline alloys [7] or bulk metallic glasses [8] could be performed. According to used materials different purpose coatings could be performed: abrasive coatings, conductive, multi-material, anticorrosion coatings, and others.

The application of cold spray coatings as corrosion protection has attracted attention last two decades, Fig. 1. Such metals as titanium, nickel, aluminum, zinc, stainless steel have been successfully deposited for corrosion protection [9-12]. Commonly, the density of the coatings is found to be a crucial parameter of cold spray deposits for corrosion protection [13].

Figure 1, Growth of publications number of cold spray technology during the last two decades

The coating of large objectives with sacrificial or noble material or restoration of damaged parts could be successfully performed with a cold spray method. It is achieved by robotic control or in some cases a portable cold spray system [14].

The present work is dedicated to the optimization of cold spray parameters for anticorrosion coating formation by an example of zinc material. Investigation of the influence of coating microstructure on electro-chemical behavior in alkaline solution was analyzed. And methods of improvement anti-corrosion properties of coatings were proposed by the addition of hard particles or laser treatment in case of aluminum cold spray deposites.

To develop a theme next chapters are proposed:

Chapter I includes an introduction to the topic of cold spray technology and other methods of anticorrosion coating application;

Chapter II describes methods of investigation and characterization of deposited coatings

Chapter III follows up the mechanism of zinc high pressure cold spray coating formation, selection of spraying parameters. Besides, it introduces the Doehlert uniform shell design to reveal a dependence of spraying parameters, gas pressure and temperature on coating properties: interfacial porosity, thickness and corrosion resistance;

Chapter IV realizes local electrochemical investigation of zinc cold spray coatings with the use of microcapillaries. The influence of coatings parameters and microstructure on electrochemical response will be discussed in the chapter;

Chapter V describes the methods of aluminium low pressure cold spray coatings densification. Addition of hard alumina particles and laser treatment will be considered for improvement of anticorrosion properties of aluminium coatings.

Appendix includes the analysis of critical and abrasion velocities of particles with diameters 26 µm, 37 µm and 54 µm based on Equation proposed by Schmidt et. al. [15]. Impact velocities and impact temperatures were found by modeling based on fluid dynamics and heat transfer calculations performed with the help of Xinliang Xie from the University of Technology of Belfort-Montbéliard

Chapter I: Bibliography

In this chapter the review of anticorrosion coating deposition is presented. It includes different coating methods, mechanisms of coating formation and microstructure. It gives also a description of the influence of spraying parameters on microstructure.

1.1 Cold Spray Technique

1.1.1 Cold Spray Process

Flame spray, wire arc spray, HVOF, thermal and cold spray are widely used technologies for additive mono- and multi-material coating formation for corrosion protection use [16-21]. The youngest deposition method is Cold Spray based on a solid-state process. Unlike conventional thermal spray methods and galvanizing cold spray allowed the formation of coatings with low oxidation, minimal internal stress, and phase transformation like recrystallization or grain growth. The cold spray system is flexible and can provide a local restoration or on-site refurbishment of damaged areas, to form thin and thick, durable, composite or anticorrosion coatings [22]. During spraying, a powder could

be applied with working gas (N2 or He) using the De Laval type nozzle. At expansion part, the gas jet inducing particles achieves close to sound velocity (500-1200 m/s), which gives high deformation rates of particles while striking the substrate [23,24]. The wide range of materials, such as metals, ceramic, nanostructure materials or polymers could be applied with cold spray [9,10]. This technique is often compared to explosive welding [25,26]. From one side in both technique coating adhesion depends not only on thermal but also on the kinetic energy of particles. Also, deposition of material is kept in a solid-state, only local heating takes place during particle-particle collision [26,27].

The process can be repeated layer by layer to grow thick coating retaining the composition of the feedstock. Working gas pressure and temperature are the most influencing parameters on the quality of coatings. The gas can reach a pressure of 5 MPa and temperature of 1000 oC. The working gas temperature is lower than the raw material melting point, the whole process is conducted in solid-state. The coating is formed by particles adhesion to the substrate and onto the previously deposited layer due to high deformation rates [9,28,29]. Adhesion mechanisms include mechanical material interlocking, intimate contact of fresh metal activated by oxygen films breaking on particles surfaces and localized melting [30,31].

Aluminum and zinc are two of the most popular materials used for anticorrosion coatings [32]. Zinc is very active due to its low electron potential and it serves well as a sacrificial coating in the marine area. And aluminum has additional protection due to oxide film, it can serve as a noble or sacrificial coating according to the chosen substrate [33]. Due

1.1.2 Coating formation and microstructure (ASI)

During the impact, the kinetic energy of a particle dissipates into heat at a larger fraction for higher strain rate [34]. About 90% of the kinetic energy dissipates into heat and the rest of the energy is stored in the coating as residual stresses or defects, the time of kinetic energy dissipation is about 100 ns [26, 35].

During strain process microstructural changes take place and geometrically necessary dislocations (GNDs) are formed, dynamic recrystallization and as a result a grain-refinement [36-38]. Dislocations can form stable nano-sized loops and influence the mechanical and electrical properties of the coating [39].

Adiabatic shear instability (ASI) appears at high velocities and temperature while spraying, and leads to extra high rate of plastic deformation of the particles due to absence of time for common thermo-dynamic energy exchange. The condition of adiabatic shear instability (ASI) could be indicated by sharp rise of the plastic strain near the edge of the contact zone [26]. Bonding occurs above a certain impact velocity as a result of ASI with plastic strain of about 1000% [40,41]. Thermal softening relates to thermally activated dislocation movement and not to changes of microstructure according to Johnson and Cook [42].

The hardening effect takes place due to high strains and strain rates, but softening effect caused by adiabatic heating can dominate and leads to an adiabatic shear instability [29]. ASI initially forms at the outer rim of the contact zone above a certain velocity of impact or critical velocity (Vcrit).

ASI and metal jetting extend toward the center of joint with increasing impact velocity, Fig. 2. At the center region of the contact zone there is no metallurgical bonding. Deformation and grain refinement has the highest rate at the contact zone and reduces toward the center of joint [35,43-45].

Figure 2, SEM photos of particle impacts, a) jetting of the outer rim of the copper particle on copper, [26], b) and c) detached particle of Ti-6AL-4V by ultrasonic cavitation and its imprint on titanium substrate [46]

Inter-splat (flat deformed particle) bonding of neighbor adhered particles determines the quality and performance of cold spray coatings. In-flight particles have not a high oxidation rate during spraying with N2 or He working gas. The amount of oxide while using air is higher, but following the breaking of oxide films while impact takes place [12]. The near-zero oxide contents in the coatings provides a high electrical and thermal conductivity similar to bulk material [28]. In the work «Evaluation of Parameters for Assessment of Inter- Splat Bond Strength in Cold-Sprayed Coatings» by G. Sundararajan et al., inter-splat bonding was quantified with the use of elastic modulus, electrical conductivity and critical load of inter-splat cracking. Theoretical and phenomenological models proposed for air plasma spray were extended to cold spray coatings. Experiments have shown, that at constant gas temperature the ratio of elastic modulus of the coating to the bulk is (Ec/Eb) is proportional to the parameter (1- p)/(1 + ap) where a is a numerical constant approximately equal to 32 and p is the cumulative volume fraction of defects (porosity, cracks, unbounded regions).

The ratio of electrical conductivity of the coating to that of the bulk material (σc/σb) is proportional to p itself [47,48]. For cold spray p is meanly presented by inter-splat cracks and measured by critical load Lc for tensile test. Investigation of cold spray inter-splat bonding showed, that elastic modulus (Ec/Eb), normalized electrical conductivity (σc/σb) and critical load (Lc) are inter-related to each other and correlate very well with the extent of inter-splat bonding. The electrical conductivity (σc/σb) is in a linear relationship with critical load (Lc) and amount of inter-spat cracks ap, they both are more sensitive to inter-splat bonding than

elastic modulus (Ec/Eb) especially for denser structure [28]. For example, coating/substrate bond strength of LPCS Zn coating on steel and copper substrates showed high values of 33 and 38 MPa respectively. Unlike an adhesion at HPCS indicated 13 MPa with fracture planes inside the coating [29,49]. This investigation has shown the influence of inter-splat bonding and heterogeneities in the coating on its properties. High bonding and fewer heterogeneities could be achieved by increasing of particle deformation rate, commonly due to propellant gas temperature and pressure increase.

Gradient of residual stresses takes place in cold spray coatings. They could be compressive or tensile. At coating top side and bottom of substrate, tensile stresses were revealed [9,50]. Compressive stresses are naturally formed at the coating/substrate interface by peening and deformation of particles [51]. Dynamic recovery and recrystallization while deposition leads to relaxation of residual stresses [50]. While coating thickness growth; the adhesion-strength of the coatings decreases. However, thick Inconel coating of more than 1 mm thickness have been successfully deposited without cracks and peeling-off [51]. In comparison to thermal spray methods, the cold spray provides a low level of tensile stress, reducing the melting of particles.

Common microstructure of cold sprayed coatings consists of deformed particles or splats. Particles usually have oxidized surfaces either coming from the feedstock or appearing during spraying, Fig 3. These oxide layers form a network of inter-splat boundary, it could be visible by etching with 100g/L of NaOH or another solution. At low deformation rate, some amount of porosity between particles could be formed [22,29]. Most of heterogeneities are formed at the deposited particles boundaries: pores, oxidation films, impurities, etc. In the case of Al cold spray coatings, micro-sized pores were reviled at particle/particle boundaries [33]. Cold spray process avoids the creation of shrinkage-driven vertical cracks due to absence of particle melting unlike galvanization or other thermal spray technologies [28,29].

Figure 3, SEM microstructure of a) cold-sprayed Zn coatings etched with 100g/L of NaOH in the as-coated condition, 2Mpar, 350 oC, [28], b), after heat treatment at 150 oC, [22]

Grain refinement due to dynamic recrystallization leads to the formation of the nanostructure, it was observed for zinc [31,52], nickel [53], aluminum [54], titanium [55]. Formation of amorphous phase as a layer about tens of nanometers was observed in several works [56-59]. The change in interfacial microstructure will influence on particle cohesion and hence properties of the coating. There are not many works discussing the evolution of nanostructure during the cold spray process. In the work “Formation of metastable phases in cold-sprayed soft metallic deposit” by Chang-Jiu Li et al., the nanostructure of cold spay Zn coating was investigated. It was found that at the core of deposited particles their structure and grain size remain similar to the feedstock. At the same time near the interfaces between deposited particles the nanograins are formed. At direct particle impact, the width of the fine grains, the interface is narrow, when near the periphery of deposited particles it is up to several hundreds of nanometers. The oriented lamellar microstructure and amorphous phase was observed at the interface region. Figure 4 and 5 represents TEM microstructure of

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University the particle interface and the schematic mechanism of the particle interface microstructure evolution [31,52]. The mechanism of tamping of melting-induced particulates to lamellae takes place in case of local melting of particle boundaries while impact and temping of melted drops by following particle impacts. Recrystallization refinement mechanism is a result of dynamic grain refinement by temping particles. We can suppose the presence of both mechanisms at the same time taking into account different particle size, velocity, impact force, etc.

b Figure 4, ATEM microstructure of the interface of Zn coating deposited at the gas temperature of 410 oC (a) and high magnification of area marked C indicating an oriented lamellar structure (b) [52].

Figure 5, Schematic diagram of the evolution of the particle interface microstructure in cold sprayed coating by recrystallization refinement mechanism (a–c) and tamping of melting- induced particulates to lamellae (d–f). Panels a–c show formation of elongated grains resulting from viscoplastic deformation (a), further elongation of grains by the tamping of following particles (b), and recrystallization (c). Panels d–f illustrate the formation of melt jetting with splashed particulates (d), tamping by the following particles (e), and successive deformation of the fine particulates by tamping effect to lamellae (f)[52].

Another work of the same group of authors concerning inter-splat microstructure at cold spray coatings has observed small spherical particulates adhered to the periphery of splats, which is the evidence of melt-jetting at the splat boundaries. The melting rate of the interfacial area depends on the high velocity and temperature of the spraying jet. For Zn particles intensive plastic flow was highly localized at the interfacial splat area. In the Fig. 5 a. the trace of melted particles interfacial area left after particle deadhesion could be seen. Instead of elongated grains, the nanocrystalline structure was observed due to the recrystallization process Fig. 6 b. Small spherical particulates formed by melting and splashing out during impact were deformed by following deposition particles and formed lamellae structure along interface directly with the thickness of several tens nanometers. Interfacial melting brings the great contribution to adhesion strength. Metallurgical bonding between particles is especially important [31].

Figure 6, a) Surface morphology of cold-sprayed Zn coating at N2 temperature of 410 °C, b)

TEM microstructure of an interface in Zn coating deposited at the N2 temperature of 410 °C illustrating the recrystallized nanograins [31]

Cold spray coatings exhibit a microstructure close to highly cold-worked bulk material, high amount of lattice defects can reduce coating electro conductivity due to limitation of electron mobility [35].

Some additional treatments or methods of improving the quality of cold spray coatings were proposed. For example, heat treatment of Zn cold spray coatings in a vacuum oven at a temperature of 150 oC for 1 h closes porosity and reduces inter-splat defects from 0.47 to 0.25 % in the work of N. Chavan et al. In the case of 350 oC heat treatment closes the porosity, but do not reduce the extent of inter-splat defects. From another side, pre-heating

1.1.3 Effect of spraying parameters

Heterogeneities in the coating (pores, oxidation films, impurities etc.) can be controlled with input parameters such as working gas, feedstock powder shape, temperature, pressure, gun traverse speed, number of coating layers, the distance between spraying passes and stand-off distance.

Working gases helium and nitrogen are the most popular gases used for cold spray process. With helium, it is much easier to reach high velocity jet and it extends the region of high velocities outside the nozzle due to its high specific gas constant and isentropic expansion ratio. He molar mass is 2 g/mol, which is much lower than a molar mass of N2, it is 28 g/mol [23], [395]. It is recommended to use He for materials, which require to reach high critical velocity to obtain high adhesion and coating density. In opposite way, N2 and air are used to reduce costs of manufacturing. However, N2 provides less oxidation than air [60].

The effect of particle shape is largely discussed in the literature. Some authors observed that the rate of porosity is higher with spherical morphology of feedstock than with irregular one [61,62]. Irregular particles reach higher velocity and deformability. A pretreatment of the feedstock powder can enhance the coating performances. For example, if particles exhibit dendritic morphology, their velocities are different due to the fragmentation of the irregular particles [63].

Gun transverse speed is the speed of gun movement along the substrate, it controls the thickness of the coating at one pass, porosity, and microhardness [64]. Relatively slow transverse speed allows to adhere more particles from spraying jet onto the surface of the sample per unit of time. Long time jet exposure influence on deposition efficiency (DE). Thus, DE of Stellite 21 powders sharply increases at transverse speed from 20 to 100 mm·s−1 and then continuously decreases [64]. In the case of too fast transverse speed of the gun, particles adhere far from each other and it leads to thin coating and poor covering of the substrate. With increasing of transverse speed, the amount of particles impinging the surface decreases. It can reduce deposition efficiency because the incubation time necessary for activating of spraying surface by some particles could be too short for high adhesion [64].

Scan number is the number of layers of the coating. Once all the necessary surface of the substrate is fully covered the process of coating could be repeated. It is well used when the high thickness of the coating is needed or for the restoration of details by multilayer growth of coating at a damaged area [65]. However, increasing coating thickness leads to a bonding strength decrease and accumulating of residual stresses [9].

Distance between gun passes is used to describe overlapping of spray spot, while creating full covering coating. Too high distance will lead to strips of gaps at the bare substrate. This parameter also influences the thickness of projecting coating [66].

Stand-off distance from nozzle exit to the substrate influences the size of the spraying spot or diameter of the jet while striking the substrate. This parameter regulates the thickness of the coating by adhesion of particles allowing them to reach the critical velocity. Moreover the strength of shock-wave also depends on stand-off distance. Shock- wave takes place while spraying, it consists of rebounding particles which do not have enough energy nether to adhere no to leave the jet. From one hand low energy particles rebound and from another hand they are exposed to high pressure of shock spraying jet and redirected toward the substrate again.

Too strong shock wave can give an abrasive effect to the already formed coating [56]

Shock-waves reduce the velocity of the spraying jet (gas with particles), its’ strength is higher at short stand-off distance, which reduces coating performance and impede deposition efficiency. The intensity of shock wave reduces at a relatively large stand-off distance [67].

Choosing of stand-off distance depends on particle velocity, the most appropriate location of the substrate is, when maximum particle velocity achieved. For example at 60 mm distance from the nozzle exit, Ti powder with helium working gas has a maximum velocity of 900 ms-1. A similar trend was observed by Pattison et al. for Ti particles with a density 4.5 kg/m3 [23,67].

In the work of N Maledi et al. with increasing of stand-off distance, coating thickness and micro-hardness decreased together with deposition efficiency and more residual stresses were formed [9]. Morsi et al. explain this by a decrease of the amount of spraying particles and oxygen incorporation while using air as working gas, which also creates a kind of stress points [68].

According to computational fluid dynamics models, a shock waves or bow shock can forms at some standoff distances [67]. It is presented as a layer of compressed gas with trapped particles, which rebounds from the substrate and undergoes the pressure of following the working jet. With the decrease of standoff distance, the density and thickness of the bow shock compressed layer grows.

Due to the inertness of particles, their velocity in free jet out of nozzle continues to grow. Depending on the material and size of feedstock powder, some particles start to loose their velocity close to the substrate. Furthermore, particle velocity may be reduced by the obstacles such as the shock wave, ambient mixing or others [69-71]. Thereby, some light or small particles already reached critical velocity could loose it because of the bow shock. Dykhuizen et al. [72] and Gilmore et al [73] have reported that the velocity of small particles with the size of 5-15 µm can be reduced and they could rebound away from the substrate.

Increasing of compressed layer thickness decreases the ability of the powder to pass through the wave obstacle [67,74].

Due to the formation of stagnation bubble, spraying particles are exposed to changes in drag force, deceleration and decreasing in deposition efficiency. In this way, the shock wave acts as a filter not only for small particles, but for all the particles with velocity only slightly above the critical velocity necessary for successful deposition [67].

In the work of J. Pattison et al. “Standoff distance and bow shock phenomena in the Cold Spray process” the confirmation of the discussion above was done on the calculation of deposition efficiency by spraying powders of different density aluminum, copper and titanium. Figure 7 from the paper shows the scheme of bow shock wave formation. Very low stand-off distance (5 mm) gives a high porosity level. At stand-off distance higher than 25 mm, a slight increase in porosity was found [75], [9].

Figure 7, Schematic diagram of the supersonic impingement zone at the substrate and presence of bow shock wave [67].

Temperature and pressure of the gas are the most important parameters for cold spray process, the quality of a coating depends mostly on them. The increase of temperature contributes to the rise of the particle velocity along the nozzle axis at the expense of in-flight particles at the nozzle exit. At the same time, pressure rise gives a more uniform increase of the particles velocity within the flow [23].

The work of K. Binder et al. shows rise of Tubular Coating Tensile (TCT) strength from

120 to 300 MPa for Ti coatings with use of N2 as working gas. From one hand increasing the kinetic energy of particles leads to more efficient acceleration and consequently higher impact velocities. On the other hand, at higher temperatures the ductility of the metal grows and coating strength increases. With high working temperatures, it is easier for particles to reach their critical velocities and obtain adiabatic shear instability (ASI) [24].

Phani et al. investigated the influence of gas temperature on the porosity of Cu coatings based on the Taguchi method. A porosity level lower than 1% was obtained for Cu coatings, porosity decreases with the increase of gas temperature. Additionally, while treatment in vacuum, the lowest porosity values were achieved. The powder feeding rate showed a minimum influence on porosity. Vacuum heat treatment for 1 h showed a similar trend [75].

For Zn coatings, gas temperature rise showed improvement of hardness up to 500 oC and a slight decrease of hardness at 525 oC. The hardness improvement could be a result of strain hardening due to shot penning effect providing microstructure closed to cold-worked bulk material. Striking particles have high energy to the polarization of dislocations and following load resistance of the material [76]. But at high-temperature strain relaxation starts as self- annealing effect and hardness decreases. At the same time, an increase in pressure did not reveal a noticeable effect on hardness. It might be due to the reduced temperature of the gas and lower annealing effect. Temperature and pressure increase create less residual stresses in Zn coatings [9]. We can suppose that, a local temperature increase due to a stronger impact force leads to a self-annealing effect and relies on residual stresses.

Too low temperature as well as low pressure leads to a low velocity of the gas and weak adhesion of particles on the surface. From another side, higher temperature helps to increase deformation rate. Hence coating density increases by particle softening from one side and through an increase of particle velocity and impact force from another side.

1.2 Cold Spray for corrosion protection

1.2.1 Noble and sacrificial coatings. Corrosion mechanisms

Noble coatings provide barrier protection, as they potential is higher than the substrate. Even small defects in the noble coating can lead to rapid localized corrosion. The potential of noble coating is higher than the substrate. Opened areas connected to the electrolyte of the substrate will be directly exposed to the corrosion process. That is why it is necessary to minimize the number of defects and porosity for noble coating protection [33,77].

Sacrificial coatings serve as anodic protection of the substrate, which is under cathodic protection in a galvanic couple, because coating potential is lower than that of the substrate. While exposure to electrolyte, most active material of the coating dissolves and released electrons flow to the substrate preventing its oxidation [33].

Continuous zinc coating provides barrier protection of steel by separating from the aggressive environment [22]. At scratches, damaged areas or cut edges Zn coating provides cathodic protection to iron and steels, with the formation of a protective oxide layer, which

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University increases corrosion resistance. Zinc has less noble potential than steel and serves as a sacrificial coating. A porous passive zinc oxide layer is formed on the surface by dissolution or precipitation and leads to the preferential corrosion pathways across the high porosity areas. After zinc dissolution, the second barrier layer can be formed by precipitation of zinc hydroxide [78,79]. In the case of galfan corrosion, when steel is covered with an alloy of 95% Zinc and 5% Al, two-step protection mechanism is working: at first the temporary aluminum oxide film is forming and then zinc sulfate layer covering for galvanic protection. Zinc is exposed to corrosion preferentially in the proeutectic and eutectic lamellar microstructure [80]. It shows that microstructural heterogeneities are the activating points for corrosion initiation and concentrators for propagation of solution.

As reported by N. Chavan et al. heat treatment of cold spray Zn coating at 150 oC for

1 h increased the corrosion resistance by reducing Icorr by a factor from 7 to 10 and shifting

Ecorr values to less negative region in 3.5% NaCl solution. A two-step corrosion mechanism was described. The first step is governed by charge transfer resistance in parallel with double-layer capacitance at coating/electrolyte interface. Additional second corrosion process can take place at electrolyte-pore interface in the coating with the solution penetration [22].

The corrosion of Zn in the marine atmospheric environment was investigated with the use of wet-dry cycle test exposure, immersion in 0.05 M NaCl solution for 1 h and 7 h of drying at 298 K for 10 days. For bulk Zn after 9th cycle exposure, corrosion products formed a two-layer structure, porous out layer of oxide and chloride rich rust mixture, and an inner layer of mainly oxide rich rust, Fig.89. At the beginning, the impedance response is similar for bulk zinc and galvanized coating, until the dissolution of Fe-Zn alloy layer formed during the galvanizing process. But after several days, the impedance sharply increases due to the commencement of iron corrosion, and provides a porous corrosion layer, which decreases the mass transfer of oxidant, Fig.8-9. Depending on the thickness of the oxide layer, the two- time constant can be obtained in the initial stage. With deposition of rust the oxygen diffusion rate decreases by barrier effect and leads to high charge transfer resistance. Thus, the time constant of the charge transfer reaction increased and the second semicircle loop disappeared, Fig.10 [81].

Figure 8, SEM photographs of the cross-section of bulk zinc after 9th cycle exposure in wet- dry cycles [81].

Figure 9, Digital photographs of galvanized steel after 3rd cycle exposure in a wet-dry cyclic test in 0.05 M NaCl. A schematic model showing the dependence of oxygen reduction rate on the rust layer and thickness is also shown together [81].

Figure 10, Comparison of measured and simulated impedance diagram obtained using the equivalent circuit in (a) pure zinc and (b) galvanized steel. A schematic diagram of the current-potential plot is also shown for zinc (c) [81].

The corrosion mechanisms of zinc in NaCl, NaOH and Na2SO4 solutions have been well studied in the literature. Whatever the media, porous corrosion layers form on the Zn surface, while composition and structure depend on the electrolyte. For example, M. Mouanga et al. have reported, in chloride solution, that corrosion layer consists mainly of zinc hydroxide chloride and zinc hydroxide carbonate. In basic NaOH, the dominant products are zinc hydroxide, zinc oxide and zinc hydroxide carbonate [82]. In a neutral environment as

Na2SO4 solution, the structure of precipitation layer depends on the concentration of the solution. It becomes more porous with an increase of concentration. Corrosion layer contains Zn, ZnO and embedded Zn4SO4(OH)6 [83]. During exposure in NaCl solution the dominant corrosion products are zinc hydroxide chloride and zinc oxide. However Kalinauskas have noticed hydroxide carbonate or hydrozincite, which might be formed with

CO2 participation [84].. Thick and dense cold spray Zn coating can provide sacrificial corrosion protection of mild steel for 72 h. Formation of the oxide layer on coating surface and in the pores provides resistance to penetration of the solution [22].

Aluminum can serve as a noble or sacrificial coating depending on the material of the substrate. In the case of stainless steel substrate, aluminum coating is sacrificial, because its potential is lower and it will be exposed to corrosion process first. However, in couple with magnesium substrate, aluminum provides cathodic protection due to its higher potential value [33].

D. Blusher et al. [85] have performed atmospheric corrosion of aluminum in NaCl and

Na2SO4 environments. The average corrosion rate of aluminum in chloride and sulfate corrosion environments is similar to the presence of CO2. However, without participation of

CO2 , the sodium sulfate is less aggressive than sodium chloride. Due to lower solubility of aluminum hydroxy sulfates in comparison to the chlorides. The difference of electrolytes influences the on corrosion behavior of Al, which is mostly in the creation of pittings in acid or neutral solutions.

Higher pH regions are supposed to be formed due to oxygen cathodic reduction in the surface electrolyte:

- - 1/2O2(g)+H2O+2e → 2OH (aq)

In these areas high pH is supported by sodium ions migrating to the cathodic regions. - Without CO2, circular crusts corrosion products contains unreacted Na2SO4, α Al(OH)3 (bayerite) and NaAl(SO4)2 and 11H2O (mendozite). In the case of CO2 presence, only unreacted Na2SO4 have been detected on the surface after the corrosion test.

In the case of chlorides in CO2 free media, the corrosion process is slowly at high pH, as negatively charged alumina surface makes unfavorable adsorption of chlorides. At lower pH with participation of CO2 chlorides adsorb on positively charged alumina film in the surface electrolyte and they cause local depassivation creating corrosion pits.

A Becceria and B Poggi [86] have investigated the corrosion of aluminum in alkaline -2 -1 solution pH=8.2 with variation of SO4 concentration from 10 to 10 M. After 10 h exposure corrosion products are formed in different concentrations Na2SO4: Al10O15 ∙ H2O, Al(OH)3, -2 5Al2O3 ∙ H2O; in 5∙10 M Na2SO4: αAl2O3.

At corrosion process the following reactions are possible:

+++ - Al + 3OH → AI(OH)3 (1)

+ + - AI(OH) + 2OH → AI(OH)3 (2)

+++ - - AI(OH) SO4 → AI(OH)SO4 (3)

+++ - - + Al + SO4 → AI(SO4) (4)

+ - Al(SO4) + OH → AIOH(SO4) (5)

- - With the increase of SO4 concentration, the formation of aluminum sulfates is competitive to hydroxides.

The impedance test suggested the formation of Me(OH) ads at low concentrations of – - - SO4 and as a next step formation of Me(SO4) ads at higher concentrations. As SO4 ions can be replaced by OH - ions or water molecules, oxysulfates can be easily turned into oxides or hydroxides. Thus, they are hard to be detected as corrosion products.

1.2.2 Influence of defects in coatings for corrosion protection

Hot-dip galvanizing is the most widely used method for anticorrosion coating application. It is a concurrent to cold spray for Zn or Al coating process. Some methods of

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University improving corrosion protection at galvanizing were proposed, like dopping or change of microstructure.

Multi-material or alloyed coatings are applied with the use of hot-dip, electrodeposition, thermal, additive and other methods [87,88]. At galvanic approach, to increase anti-corrosion resistance of Zn doping, elements need to have good film-building properties and ability to enter into solid solution at a certain amount. Dissolution rate and a potential difference between Zn and steel can be decreased with addition elements doping such as Ni, Fe, Co, Sn or Al, which improves the corrosion resistance of the coating [13,14,89- 93].

To increase the corrosion resistance of Zn coatings refining methods of grains and dendritic structure are used. For example by addition of alloys Mg, Al or Sr. In the work of S. Vagge and V. Raja, it was concluded that Zn coating containing 0.02 wt. % strontium refined dendritic grain size by 86%, increased adhesion and corrosion resistance by about 33% and 30 % respectively [90].

At the hot-dip Zn-steel system the presence of localized corrosion on the galvanized layer can be observed due to active phases formation. It can lead to de-adhesion in case of multilayered coating [94,95]. In general, chemical inhomogeneity reduces corrosion resistance due to the presence of dendrites, textures, eutectic, nucleation sites, impurities and inclusions [96]. Aluminum hot-dip coatings also contain these kinds of defects including pores and cracks in some cases [97-99]. Nevertheless, most of the corrosion – resistive metals can be hardly dissolved in matrix, which makes hot-dip galvanizing not efficient [100].

As Cold spray avoids melting of powder to improve service life and corrosion resistance, a coating made with feed-stock of mixture with inert materials could be performed. Composition of the coating is presented with metallic powders like Al, Cu, Ni, Zn or alloys with the addition of ceramic or polymer particles from the mixture or supplied directly from different blenders.

The addition of 10 wt% Ni powder to the initial blend of Zn powder improves sacrificial anodic protection of Zn and physical shielding caused by corrosion products. The corrosion behavior of composite obtained from powders mixture with LPCS is similar to zinc- nickel alloy coating. The attempts to form Zn-Ni coating with a thermal spray method faced some difficulties like complex milling process and oxidation during spraying. With the use of electroplating technology, the plating solution of Zn-Ni is not homogenous and corrosion resistance is not well controlled. From another hand, obtained coating is of several microns and the method brings high rate of pollution. For these reasons the cold spray method is a good solution for composite coating formation, and performance in an aggressive environment.

Corrosion mechanism includes Zn cathodic protection and dissolution with formation

Zn(OH)2 as a first step and as the second step formation of another corrosion layer with

Zn(OH)2, ZnCl2-4Zn(OH)2-H2O, Zn4CO3(OH)6-H2O and zinc oxide ZnO. This layer works as a physical barrier to further propagation of electrolyte [101]. The formation of this special oxides chemistry is possible due to the presence of Ni in the coating. Moreover, another role of Ni is to reduce the potential difference between Zn and Fe, and the galvanic cell electromotive force.

Composite coatings are used for the improvement of corrosion resistance. In the work of K. Brindha et al., the graphene prepared by electrochemical method was applied on Zn using so-called a doctor blade technique. The process of composite coating creation by application of composite suspension in gel state and control of thickness by the blade [102- 105]. The composite performed better corrosion resistance than Zn in a chloride environment [106]. K. Kamburova has demonstrated that incorporation of core- nanocontainers (NCs) with corrosion inhibitor BTA into the Zn matrix and its’ application on steel substrate by electrodeposition from acidic zinc sulfate solutions for anti-corrosion application. The composite demonstrated improvement of steel corrosion protection compared to Zn in 5% NaCl solution due to the release of BTA inhibitor after 15 days of immersion and formation of a passive layer [107].

Aluminum reinforced composites are also presented in the literature like nanotube reinforced aluminum composite powder (CNT-Al) [108], hard FeB particles with Ni, Fe and Al binder [109], nanodiamond-reinforced aluminum metal matrix composites (ND–Al MMC) [110].

The top surface of cold spray coatings has high roughness; it promotes local acidification and crevice corrosion mechanism. Even for chemically uniform metal, the protection of passive film is being reduced by physical unevenness, as it tends to form on homophase material [100]. The evolution of the corrosion mechanism promotes at the boundaries of the particles as autocatalytic dissolution. Heterogeneities are the weakest places in the coating due to a lower Gibbs’ free energy (ΔG). In general, propagation of solution through the particle boundaries enlarge pores and defects and forms a continuous pathway from the surface to the substrate. While reaching the substrate by electrolyte galvanic corrosion starts with creating electrical contact of two metals and solution. Less noble metal of coating corrodes preferentially protecting more noble substrate [33,111].

Comparing cold spray process, Zn coatings show high density, unlike aluminum coatings. Deformation rate of aluminum particles is lower and they often show more porous microstructure (about 1.5% by area) compared to Zn (0.15-0.8 % by area) [29,33,112].

However, in literature, there are works presenting composition Zn with ceramic particles cold spray coatings. Addition of hard particles also gives densification effect and improvement of particle bonding strength and hardness of Zn coatings by strain hardening [29, 113,114].

The deposition of metal-ceramic mixture with cold spray method shows an improvement of anti-corrosion properties in comparison with metallic powder deposition only. Due to the high hardness of ceramic particles, they provide additional pinning effect densifying the already deposited layer. Thus, closing porosity, they prevent fast propagations of the electrolyte through inter-splat cracks of the coating. From another hand reduced active area retards corrosion speed and increases the lifetime of the coating [111]. A wide experience of deposition of bilayer composite coatings is presented in the literature. For example, Al covered with Al-Al2O3 composite coating [111]. Addition of hard particles not only improves the corrosion resistance of the cold spray coatings, but also increases its hardness and durability [115-116].

From another hand depending on ceramic particles, fracture tiny powder can fill the inter-splat boundaries, closing microporosity. Thus, feedstock mixture with a wide size distribution of hard particles would be more effective for coating densification. During the process, ceramic particles activate the metal surfaces, densify the structure, improve deposition efficiency (DE) and bond strength, and also clean the nozzle of the gun [29, 117,118].

In the case of Al5056, porosity is about 2%. With the addition of SiC particles from 15 to 30%, porosity decreases from 0.93 to 0.61%. It is in the same porosity level as other Al- matrix cold spray composite coatings like TiNp/Al 2319 or Al2O3/Al, below 1% [33,119-122].

As ceramic particles are inert, their presence in the coating does not influence greatly the corrosion resistance [33,121]. From another hand, due to coating densification, the addition of ceramic particles reduces the penetration rate of corrosive electrolyte and galvanic current. For example, according to Silva et al. [111] with the addition of Al2O3 particles to Al matrix, corrosion potential was increased from -847 to -826 mV/Ag|AgCl|KCl3mol/1 and corrosion current was reduced from 13.5 to 9.5 µA/cm-2.

1.3 Conclusion

To conclude, cold spray coatings are competitive to hot-dip and other processes for anticorrosion coating application.

Heterogeneities in the cold spray coatings are a key parameter in the case of anticorrosion application. Usually, they are presented as microstructural defects or stress concentrators and serve as corrosion activation points. However, at galvanizing method alloying microstructure contains structural heterogeneities including doping elements, which improves corrosion resistance. Anyway, the cold spray process avoids melting and consequently inhomogeneous distribution of doping elements in the structure.

Another key parameter of Zn anticorrosion coating is a durability of adhesive substrate-coating and particle/particle joints under corrosive environment due to the loss of

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University mechanical integrity [90]. Adhesion strength is based on intermolecular contact between atoms, covalent and van der Waals interaction and can be described by adsorption theory. For interface joint bonding heterogeneity like surface oxide, hydroxyl content or functional groups have to be taken into account due to their high energy [90,123].

Depending on the technology of anticorrosion coating formation, different mechanisms could be used to improve the performance of the anti-corrosion coating and extend its’ lifetime. In the case of the cold spray process, not only the chemical composition of feedstock is important, but also heterogeneities in microstructure, which are always present due to the nature of the process. To create a coating for anti-corrosion application with the use of cold spray method, a dense microstructure with a minimum amount of porosity, internal stresses, oxide layers and other heterogeneity is preferential.

Densification of cold spray coatings could be reached by adjustment of cold spray parameters, addition of ceramic particles or by laser post-treatment.

Chapter II. Experimental methods and equipments

In the present chapter experimental methods and equipments for investigation and characterization of cold spray coatings and feed-stock materials are introduced. Taking into account the purpose of the work, the accent of choosing methods was done on corrosion protection properties.

2.1 HPCS and LPCS equipment and parameters

2.1.1High-pressure cold spray process

The scheme of commonly used cold spray equipment is shown in Fig. 1 while complete descriptions can be found in [124-126]. A gas flow, usually He or N2 at pressure up to 5 MPa is fed to a gas heater, which enables preheating in the range of 100-600 oC. The preheated gas flow is followed to a converging-diverging de Laval-type nozzle to form a working jet. At the same time a second gas vector is fed to a powder feeding system to bring the powder from one or two blenders to the gas jet. The gas flow with unmelted powder is moved to the nozzle, where it joins preheated gas and forms a working jet. The jet of preheated gas with powder is accelerated by the special design of the nozzle and can reach supersonic velocity up to 1200 m/s. The output temperature stream drops to about room temperature [1,70,127].

Figure 1, The scheme of commonly used cold spraying equipment

The de Laval-type nozzle has a special design to create the gas jet with entrained particles, which can achieve critical velocities for successful deposition. Critical velocity depends on powder material density as well as it’s size and shape [128, 129]. Optimization of nozzle design takes into account particle impact parameters and powder feeding rate to

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University achieve maximum deposition efficiency and process productivity [129]. As shown in Fig. 2 the main characteristics of the nozzle includes the throat diameter of the throat (the narrow part), the ratio between the diameter of the throat and the nozzle exit (expansion ratio) and the length of the divergent part. Accelerating through the narrow part of the nozzle the gas loses its temperature and pressure. The process gas flow and particle behavior within the nozzle have been modelized by K. Sakaki in his work “The influence of nozzle design in the cold spray process” [396]. The gas condition could be described as a function of Mach number, which is presented as a ratio of the gas velocity to the local sound speed. To reach a supersonic condition the Mach number has to exceed the value of 1. For the nozzles with Mach numbers ~2.5–3.5 the values of stagnation pressure of working jet are 2.5–5.0MPa and temperature up to 1000 oC. Usually the diameter of the nozzle throat is 2-3 mm. It allows working gas flow rate about 120 m3/h and powder feeding rate of 20 kg/h [1,129].

To increase the efficiency of cold spray deposition it is necessary to increase particle velocities in spraying jet. Nozzle design is a first step and the base of the cold spray process. Rising of working gas temperature could be another solution for flow velocity increase as well as the rise of powder impact temperature and adhesion efficiency. The powder is commonly fed in the prechamber or subsonic part of the nozzle, where the temperature is high, and it is exposed to intensive heat exchange [1,127,129].

Gas velocity also depends on the type of working gas. Gases with a lower molecular weight such as helium are preferential for cold spray, when high velocities are necessary. Increasing of stagnation gas pressure increases the density of working gas, its velocity and drag force. The size of the particles and their density also play a role in impact velocity. Smaller particles achieve critical velocity easily in the spraying jet.

The distribution of particles in the gas flow has a quasi-Gaussian shape, it forms a convex coating, which cross-section has a triangular shape [130]. To control the spraying spot detentions different nozzle design could be used [129,131]. A rectangular type nozzle could be used to obtain uniform particle distribution and hence homogenous thickness of the coating [132]. The track formed with the application of a rectangular nozzle gives a uniform thickness of the coating and a spot size up to 30 mm in diameter, Fig. 2. On the other hand, special shape causes the deposition on the inside walls of the nozzle and changes characteristics of the working jet. Thus, the application of a rectangular nozzle is reasonable in the special case of deposition of light materials like aluminum or materials with high density such as zinc or copper [129]. Klinkov et al have proposed the application of swirled gas flow in a prechamber with oblique side cuts. This nozzle produces a uniform coating and can be used for internal surfaces of tubes or interior angles of complex details [133]. To reach a spatial resolution of cold spray it is necessary to reduce the dimensions of spray spot. In this case, a solution is the application of micro-nozzle with the exit diameter less than 1 mm. By decreasing the nozzle size, not only in diameter but also in length a small nozzle can be

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University created. This kind of nozzle can produce coatings for powder with small granulometry and narrow size distribution only. Large particles cannot reach their critical velocity for a small length of the nozzle, their acceleration becomes not sufficient [134], [135], [129].

Figure 2, The profiles of deposition tracks made after one pass and low velocity of the gun with axisymmetric and rectangular nozzles [129]

Different parameters play roles in coating deposition. There are practical limits for all the parameter variables depending on the used materials and equipments. It is also necessary to take into account possible occurrence of a bow shock wave close to the substrate at large Mach numbers [1,136].

Zn coatings were prepared with a cold spray system developed by Impact Spray System ISS 5/8-N-2 PF (Germany), Fig. 3. A converging-diverging de Laval nozzle has a length of 160 mm and expansion ratio or ratio between narrow part and nozzle exit diameters of 5.6. the nozzle was mounted on the arm-type robot for remote control of deposition. The nozzle is equipped with a cooling system that allows us to avoid overheating the feedstock and consequent clogging of the nozzle. Two separate powder feeding blenders allow to deposit multilateral coatings directly from feeding from different blenders and changing of feeding rate, Fig. 3b.

Figure 3, a) Impact Spray System ISS 5/8-N-2 PF (Germany), b) powder feeding system

Cold sprayed coatings were performed on carbon steel (0.08-0.13 % C, 0.15-0.35% Si,

0.30-0.60% Mn) substrates with dimensions 3 cm x 7 cm x 1 mm sandblasted by Al2O3 (P100) particles and. The following cleaning in the ultrasonic bath was performed with ethanol for 5 minutes to remove impurities and oil.

2.1.2 Low-pressure cold spray process

Low-pressure cold spray system Dymet 403j was used in present work for anticorrosion Al coating formation, Fig. 4. De Laval-type nozzle, with a removable diverging part made of stainless steel was applied. The nozzle dimensions with narrow part d=2.21 mm. Compressed air has been applied as propeller gas.

The cold spray system includes an air compressor, gas heater, two powder feeders and a cooling system. The cold spray system is equipped with an arm-type robot and remote control for accurate position and movement control of De Laval-type nozzle while spraying.

Figure 4, Dymet 403j low pressure cold spray system with two powder feeders and converging-diverging nozzle

2.2 Experimental methods

Contact profilometer

The condition of the top surface of the coating is necessary to describe for anticorrosion applications. Surface irregularities lead to energy concentration and further activation of the corrosion process.

Roughness of as-sprayed coatings was investigated by measuring and calculating of arthimetic mean deviation of the surface Ra and the mean deviation between the five highest peaks and the five deepest holes.

Contact profilometer Mitutoyo, Japan was used for Zn coatings investigation. The length of measurement and scanning rate was 2 cm and 0,5 m/s, respectively and filter ʎc 0.8 x5.

In the case of Al, coating measurements were performed with laser profilometer Surr-corder SE300. The analyses of roughness were performed on 250X250 µm area with a scanning rate of 0.2 mm/s.

Optical microscope

Optical microscope HIROX HK7700, Japan allows 3D measuring of top surface profile of as-sprayed coatings by a non-contact method. Due to the high and not constant homogeneity of the surface, it was characterized by measuring the amplitude between the highest and lowest points.

Commonly optical microscope was used for microstructural investigation of metallurgical sample cross-section. It also allows measuring the coating thickness and porosity by image analysis.

The preparation of cross-section was performed with standard methods. First, all the samples were cut slowly to avoid delamination of the coating and then mounted with cold epoxy resin. The samples were first ground with SiC papers of 600 -1000 grit and then polished with ethanol-based diamond pastes from 9 to 0.05 µm.

Scanning Electron Microscope and Energy-dispersive X-ray spectroscopy

Higher quality image and resolution of coatings and feed-stock can be obtained with electron microscopy. At present work for investigation of Zn cold spray coatings, Scanning Electronic Microscope (SEM Supra55, Zeiss, Germany) coupled with an 80 mm2 Oxford EDS probe for elemental composition analysis was used. In the case of Al coating analysis, Hitachi SU-70 Analytical Scanning Electron Microscope was used and EDS analyses were performed with EDAX Genesis.

SE2 detector of the microscope was used to obtain high-resolution and contrast secondary electrons image. Angle selective Backscattered detector (AsB) allows the analysis

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University of high energy backscattered electrons image showing topography and channeling contrast by crystal orientation.

Energy Dispersive Spectroscopy (EDS) detector allowed us to separate the characteristic x-rays of different elements into an energy spectrum. Accompanying software helps to analyze the energy spectrum of the material spot of several microns, line or an area to build the map of chemical composition. The activation energy of the elements in the material by an electron beam of 15-20kV with a working distance of 8.5 mm. EDS analysis was used in the work to track changes in the chemical composition of powder before spray and in the coating. As well as, the composition of coatings before corrosion tests and after that.

Coating thickness was measured by cross-sectional SEM observations of coatings at 10 separate places of one sample and average values were calculated.

To reveal heterogeneities in the coating and interface, chemical etching with in sulphuric acid and water at proportion 1:4 for 1 minute was used.

Ion Beam polishing

Ion beam (IB) polishing was used instead of common abrasive polishing to avoid flattering of porous while polish since high ductility of aluminum. For polishing with Cross- section polisher IB-09020CP, all coated samples were cut for pieces with dimensions 11x6 mm and about 2 mm thickness.

X-ray diffraction (XRD)

Phase identifications of powder and coatings were based on the correlation of XRD peaks with ICDD database. X-ray diffraction (XRD) diagrams were recorded with a Phillips PW

1830/40 diffractometer at the lab condition (CuKα radiation source at 20 kV and 10 mA). Diffractograms were collected in 2θ/θ from 30 ° to 109.995° with a step size of 0.100 ° and step time 184 s.

X-ray tomography

The volume percentage of porosity in the coating/substrate interface of cold spray coatings was measured with powerful non-destructive technique X-ray tomography with EasyTom tomograph (RX Solutions, France). High-resolution images were acquired on small coated samples (2 × 2 × 30 mm) with a voxel size of 1 µm3. This resolution was ensured by a LaB6 emission tip for the X-ray source for which the resolution is not modified by geometric blur. The Hamamatsu X-Ray source was operated at a voltage of 100 kV with a 0,1 mm thick copper filter. The detector was a Hamamatsu flat panel with a pixel size of 50 μm. Each scan consisted of 1184 projections on a 360 degrees rotation and an averaging of 3 images at each step angle. The exposure time for one radiograph was 3 s and the total time for one analysis is about 4 hours. All the data were reconstructed by a filtered back-projection Feldkamp-algorithm.

ImageJ/Fiji software

Volume porosity and 2D measurements obtained from tomography and SEM captures respectively were calculated with ImageJ/Fiji public domain software. Analyzes were performed by setting grayscale threshold cutoff points. The resolution of x-ray tomography allows us to calculate interface coating/substrate porosity and for more precise measurement of naturally created pores between particles in the coating 2D analysis based on high-resolution microphotograph were performed.

Bending tests machine

Three-point bending tests Zwick/Roell machine and TestXpert II software are helpful to perform bending tests of coated samples to observe damage and delamination resistance of the coatings. Bending force was applied to the sample with a speed of 0.5 mm/min, the diameter of the cylinder tip was 8 mm. The length between the loading tip and supporting points was set as 30 mm (L) symmetrically for both sides, Fig. 5. The coating was placed on the opposite side of the loading tip.

a b

Figure 5, Three-points bending test a) Schematic illustration and b) photo of the process

Microhardness test

Vicker’s micro hardness test was performed with HMV-2, Shimadzu equipment on the aluminum cold spay coatings in as-deposited and heat-treated conditions. The applying force of the indenter was 98.07mN (HV0.01) for 10 sec. Each experiment was performed 10 times and average values were calculated.

Electro-chemical behavior investigation

Standard 3-electrodes electrochemical cell was used for corrosion behavior characterization of both zinc and aluminum cold spray coatings. Zn coatings response was

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University recorded with GAMRY Ref600 (USA) potentiostat and with potentiostat SP-300 in case of aluminum coatings.

Local microcapillary tests were carried out for precise investigation of heterogeneities in Zn coatings and their influence on electrochemical response data. For this purpose standard 3-electrode electrochemical micro-cell attached to optical microscope “OLYMPUS BX51M. Japan” was applied. Electrochemical data was collected with Gamry Instruments reference 600+ Potentiostat/Galvanostat/ZRA.

An open circuit potential (OCP) test is a method of zero-current application. It was performed to show the stabilized potential of a system. It is necessary to provide it before applying additional potential and measuring the system response.

Linear Sweep Voltammetry (LSV) is a voltammetric method for measuring current at a working electrode or the sample while application of extra potential to the system. The potential between the working electrode and a reference electrode is swept linearly in time.

Electrochemical Impedance Spectroscopy test (EIS) was performed for respond of Al coatings system resistance by applying a sinusoidal perturbation.

2.3 Conventional macro-scale measurements, electrochemical 3-electrode cell

Large scale electrochemical cell contains 3 electrodes, according to the specific roles, they are labeled as the working electrode (WE), the reference electrode (RE) and the counter electrode (CE), which is also called an auxiliary electrode. Typically, working electrode is an investigating sample, which presents a cathode. And the reference electrode serves as an anode. It contains a system of phases, which retain a constant composition. These phases present reversible redox reactions at high speed and could rapidly adjust the ionic activity of the solution. At the same time the electrode stays sensitive to larger current changes. This is the mechanism to provide a stable potential for analysis of the working electrode. Auxiliary electrode serves as a removal of excess current from the reference.

While measuring the potential between the working electrode and reference electrode a voltage drop could be observed. According to Ohm’s low:

푉 = 퐼 ∙ 푅 (1)

, where I [A] is the current, which passes through the cell and R [Ω] is the resistance of solution. A major part of solution resistance can be eliminated by the position of working and reference electrodes as close as possible.

The material selection for the electrochemical cell is provided according to easy fabrication, inertness to the electrochemical reactions and cost-efficiency. The most popular

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University materials for electrochemical cell fabrication include Teflon, Nylon, Kel-F, and glass (Pyrex and glass) [163].

2.3.1 Electrochemical 3-electrode micro-cell

Commonly large-scale electrochemical methods are used to study the corrosion behavior of materials. They allow investigating exposed surfaces area in the range of cm2- mm2. With the use of these techniques the information of the electrochemical activity of the whole system could be recorded, but no impute of reactions in local places could be obtained [164]. For electrochemical analysis of solid surfaces, researches always tend to miniaturize the working electrode to be able to investigate local areas with common electrochemical techniques [165]. Micro-area analysis allows us to record data from different phases, inclusions, single grain or cracks. Nowadays, different local electrochemical techniques are proposed for micro- and submicrometer size electrochemical characterization of metals and the alloys, such as scanning vibrating electrode technique (SVET) [166-168], probe microscopy (SPM) [169-173] or electrochemical micro-cell system [174-176].

There are two main groups of local measuring techniques: scanning micro-reference electrode technique and small area techniques [177]. At scanning techniques, the specimen area is mm2-cm2 size, corrosion potential and current is recorded by scanning of micro- or ultramicro- reference electrode over the surface. With the use of different scanning setups, a lateral resolution in the nanometer range could be obtained. Nevertheless, this technique does not give information on local electrochemical behavior as the current flow from the whole wetted surface is normally recorded [166-171,173].

The area of the working electrode is defined by a solution-wetted spot on the specimen. The electrochemical measurement could be localized by decreasing the immersed area. The miniaturization of the area could be done either by reducing testing material surface (masking, cutting) or by reducing contact with electrolyte through the application of microcapillary. Achievement of micro- and even submicrometer size of capillary tip gives access to local measurement of heterogeneities on the specimen surface [164,170,175, 176], [178]. Data of corrosion behavior in local place can be obtained directly from the surface and recorded with high-resolution potentiostat [179].

In general, all the macroscale known electrochemical reactions can occur at the microscale, for example, nucleation at the surface and charge transfer reactions [180]. However, the density of limiting current diffusion layer is high and leads to high convergent diffusions in microsystems according to the small area of exposed surface [181]. Thus, in microsystems not only the area of the working electrode, but also the mass transfer process are two main factors controlling microelectrochemical reactions kinetic. The use of 3- electrode electrochemical micro-cell showed high –resolution electrochemical response, unlike conventional methods [165]. Higher difference between conventional and micro-

In the mid 1990s, the electrochemical micro-cell was proposed by Bohni, Suter and in parallel by Lohrengel to provide the local investigation of corrosion response on single microstructure heterogeneities of the surfaces [175-176,186]. Electrochemical micro-cell system setup could be also named by authors as “chemical micro electro system” [175], “micro-capillary cell” [177] or “scanning droplet cell” [186,187]. Local heterogeneities are crucial for passive materials like aluminum or stainless steel. The heterogeneities like precipitates, inclusion or impurities present themselves as potential initiation sites for corrosion initiation [164]. To reduce the wetted area of working electrode glass microcapillary is used.

Commonly the micro-cell experimental setup is based on the model proposed by Böhni et al. [175] in 1995. The setup represents the 3-electrode microcell equipped by microcapillary and filled with electrolyte, Fig. 6. The diameter of capillary could be from 1000 µm to nano-size depending on the necessary area of investigation. The capillary contacts the surface of the sample and forms the working electrode. The counter electrode is often located not far from the microcapillary tip, it is embedded in the acrylic cell and often made of platinum or gold. The short distance between the tip and counter electrode reduces the resistance of the micro-cell. The type and size of the reference electrode can be different according to the micro-cell design and properties of exposed material. Electrochemical micro-cell is mounted on an optical microscope and it also could be equipped with a camera for precise positioning of the capillary tip on the examined surface [165,188]. The reference electrode is usually located outside the electrochemical cell, which makes it possible to reduce the dimensions of the microcell and to maintain it to the holder of the optical microscope. In this case, the potentiostat will include uncompensated ohmic drop between reference and working electrode to the measurements [13]. This kind of setup has been used by Paussa L et al, Schneider M et al., Krawiec H et al. and others in their works [183,189- 195].

Figure 6, Scheme of electrochemical 3-electrode micro-cell, [2]

A capillary is a less than 1000 µm tube made of glass, metal or plastic. Commonly, capillaries with diameters in the range of 10-500 µm are used [195]. To prepare a glass microcapillary with determined tip diameter a special puller machine could be used. This machine allows controlling of precise inner tip diameter while producing up to nanometers size [196]. With this method has high cost and even some Lab-made alternative methods could be found [183]. For microcapillary fabrication, the use of heating-and-pulling machines is the most approachable method. Its’ main idea is to heat the thin tubes to the melting point of the glass and at the same time to stretch the tube holders in opposite directions until the tear of the tube edges. This way from one glass tube two microcapillary could be formed. They need to be open by grinding with SIC paper. Opened tip can have a size of micrometers [165].

To prevent possible leakage of electrolyte solution at the place of microcapillary and surface connection Suter et al [175,176] have proposed the flexible silicon gasket application. This improvement has sharply increased the reliability, lateral resolution and versatility of the technology and it is widespread nowadays. The hydrophobic properties of the silicon prevents solution penetration through the gasket and its’ flexibility allows precise position and measuring on rough or nonhomogeneous surfaces [176,197]. The tip of microcapillary is dipping into liquid silicon glue several times to form a silicon gasket on the mouth of the tip. For a single measurement the microcapillary with the gasket can be directly attached to the surface of the working electrode without drying. From another hand, after drying, the microcapillary with silicon gasket can be used for multiple measurements [183,195,196]. This method is more useful for investigation of metal surfaces in different places with the same capillary tip diameter.

During performing of electrochemical micro-cell measurements, some difficulties can occur such as crevice corrosion, solution leakage or pH change by oxygen penetration, Fig. 7 a. If the capillary tip is not flat or the silicon gasket has variation in thickness, some crevice can be formed between the microcapillary tip and the working surface, Fig. 7 b. Formation of such places leads to rapid oxygen consumption, cathodic reaction suppression and localization of aggressive species like Cl- or OH- to the crevice from the main solution mass of the cell. All these factors lead to crevice corrosion especially for passive materials [164,165]. Krawiec H et al. [198] have reported the effect of mass transportation into the microcapillary due to crevices.

If the silicon gasket was applied not properly the electrolyte leakage can happen and misrepresent collected data. Furthermore, some oxygen can penetrate through the gap into microcapillary and influence on electrochemical reactions in the cell. The pH of the solution can be changed and affect the cathodic part of the polarization curve [164,198].

The shape of microcapillary can also influence on collecting data and shape of obtained polarization curve. According to Guseva O et al. [199], Kraviec H et al. [198] and Abodi et al. [200] the shape of microcapillary glass, the angle of microcapillary tip, and non homogenous thickness of silicon gasket influence the concentration of components participating in electrochemical reactions and current density measurements. It would be preferential to perform all the set of experiments with the same microcapillary.

a b

Figure 7, Electrochemical micro-cell, a) leakage and oxygen diffusion, b) formation of a crevice, [2]

At electrochemical micro-cell setup with the use of classical glass microcapillary the electrolyte could be refreshed only after each measurement. Due to chemistry change,

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University corrosion products can accumulate at the microcapillary mouth and reduce the access to electrolytes or completely block the capillary tip. From another hand, oxygen or hydrogen bubbles could be formed as a result of cathodic or anodic reactions and also misrepresent or block the response of experiments. Such undesirable effects are often observed at the beginning of the experiment while using microcapillaries with extremely small size of the tip. To avoid the clogging of the capillary tip it is necessary to stop the experiment directly after reaching the passivity breakdown of material [164,165,198,201,202].

Despite the fact of presenting some difficulties during microcapillary preparation and experimental process, the electrochemical micro-cell technique is a powerful method for local investigation of heterogeneities on the specimen surface. In combination with conventional electrochemical cell measurements, it gives the full description of microstructure behavior and comparison of information from large and local spots of microstructure [164]. High current resolution and high limiting current allow collecting data from local places unlike conventional electrochemical setup, but a high-resolution potentiostat is required (in the range of fA). Electrochemical investigation can be performed on a complex shape or rough surface without special preparation. Due to the small sizes of 3-electrode micro cell setup and capillaries, several measurements can be performed on one small sample with small volumes of electrolyte [165]. Examination can be provided with various electrochemical techniques, additional equipment like UV-visible spectrometers [203] or laser beam microscopes [196] and in combination with scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDXS), X-ray photoelectron spectroscopy (XPS), atomic force microscopy (AFM), time of flight-secondary ion mass spectroscopy (TOF-SIMS) and glow discharge optical emission spectroscopy (GDOES) [190,202,204,205]. Suter T et al. have demonstrated local pH measurement by embedded in microcell tungsten wire isolated with Polyimid, the obtained sensitivity of measurement is better than 0.5 pH [197].

The main objective of the work is to define the influence of cold spray coating microstructure on electrochemical behavior and adjust spraying parameters according to higher anticorrosion properties. In the present work, the investigation of corrosion behavior in local areas less than 60 µm size will be studied with the use of a 3-electrode electrochemical micro-cell.

A thin tip of microcapillary allows investigation in different phases of material and even in grain boundaries of metallic grains. In Fig. 8 different possible position of microcapillary is illustrated [165].

Figure 8, Scheme of different microcapillary position, α-metal grain, β-metal grain and grain boundary [165]

The work of this chapter is proposed for a peer review paper submitted in Surface and Coatings Technology Journal with title “Experimental design to optimize corrosion-resistance of Zn cold-sprayed coatins on steel substrates”

Chapter III

3.1 Introduction

Zinc is a versatile material that can be used in several applications including structural and protection, like roofs, facades, ship decks, ballast tank protection, high abrasion areas on bows, etc. [81,83,137,138,139,140]. It can be manufactured as bulk material by forgings, extrusions, drawn wire, rolled sheets, castings and brasses, etc. or as a coating. Zinc coated steel is obtained industrially by several processes such as plating, hot-dip galvanizing, electro-galvanizing, painting with zinc-bearing paints [104,138]. Zinc is very effective for corrosion protection, in most cases. Zn layer behave as sacrificial coating regarding to its low electrical potential [22]. Alternative solutions allowing local coating application and restoration of damaged areas arise from spraying techniques as thermal or cold spray methods [28,52,141]. The main difference between these two techniques concerns the physical state of Zinc in the propellant gas. During thermal spraying, the gas temperature is higher than the melting temperature whereas the temperature stays below the melting point during cold spraying (solid deposition process). Last decade, since it is a flexible technique, cold spray gained popularity. Moreover, it can provide an on-site local restoration of damaged areas. Thick or thin coatings can be performed, with attractive anticorrosion properties [22].

Since thermo-mechanical histories of the particles are different between thermal and cold spraying, adhesion and manufacturing mechanisms are different. With cold spray, the material build-up is based on the impact-induced bonding under high impact velocity. Several mechanisms have been proposed in the literature to explain this phenomenon, such as the adiabatic shear instability, oxide layer break up, mechanical interlocking, etc. [35,142]. If adiabatic shear instability prevails, Hassani-Gangaraj et al. argue recently that hydrodynamic jetting mechanism may occur instead of adiabatic processes [143]. Whatever the mechanism, cold spray coatings exhibit a microstructure close to highly cold-worked bulk material where defects (porosities, oxides, impurities, etc.) are located at the boundaries of the particles [33].

Metallurgical defects in cold spray coating could reduce the corrosion resistance and promote the way for solution propagation up to the substrate [96], [33]. The addition of ceramic particles is proposed to densify the coating by reducing the density of the pores. Consequently, it increases the corrosion resistance [33], [121].

To elaborate coatings, several input parameters on cold spray technique have to be optimized for metallic coatings such as the nature of working gas, its temperature and pressure, the feedstock powder morphology, the gun traverse speed and the number of passes, the spraying stand-off distance. Since some of parameters are interdependent, a strategy is required to find the optimal deposition parameters. When corrosion resistance is the target of the coating, the main descriptors should be the coating morphology (roughness, thickness), it’s chemistry (mono particle, ceramic addition particles, etc.), density (porosity, homogeneity) and polarization resistance (i.e., natural oxide film resistance on the top surface of the coating). To contribute to the optimization and to evaluate the weight of each parameter, the uniform design of experiments had been developed in [144]. In another work, Phani et al. [75] found, from Taguchi method, that a higher gas temperature has a positive effect on the porosity decrease of a Cu coating. To the opposite, the increase of the stand-off distance from 5 mm to 25 mm slightly reduces the density of the coating.

This work aims to investigate the correlation between input parameters of the cold spray system and the coating performances. Zinc was chosen as a sacrificial element in cathodic coupling with iron [145]. Few works could be found to deal with this system prepared by cold spray [30,31,52,146,147]. Dense and homogeneous zinc coating provides barrier protection of steel against an aggressive environment. It provides the formation of oxide or hydroxide layers on coating surface and in emergent pores, promoting healing phenomena [22,78,79,148]. Under scratches, damaged areas or cut edges Zn coating provides cathodic protection. It serves as a sacrificial coating ensuring cathodic protection.

Instead of using a systematic method where one parameter evolves and the others are fixed, an experimental design was applied in this work. It could be described as rational planning of experiments with high efficiency and less number of experiments [149]. After the preliminary tests, two input parameters were chosen (propellant gas temperature and pressure). Their influence on the porosity, and the thickness of the coating have been studied. The trends obtained from experimental design were compared with corrosion performances established from electrochemical measurements. Adhesion and roughness have been previously characterized by bending tests and profilometry respectively.

3.2 Materials

Carbon steel plates with a chemical composition of 0.08-0.13 % C, 0.15-0.35% Si, 0.30-0.60% Mn, were used as substrates. Samples with dimensions of 30 x 70 x 1 mm were sandblasted with Al2O3 (P100). This surface treatment aimed to remove mechanically the native oxide layer from the surface. It increased the sample roughness to reach higher mechanical adhesion of Zn particles.

Commercially gas atomized Zn powder was provided by Grillo (Germany) and presents a purity of 99,9%. Figure 1a shows the morphology of the particles. They are either spherical or elongated. Cross-section in figure 1b shows dense and homogeneous Zn particles. The particle size distribution was measured by laser diffractometry (Mastersizer 2000, Malvern Instruments Ltd., UK). The mean diameter was measured to 37.3µm (+20 - 90µm) (Figure 1c). Chemical etching in sulphuric acid and water solution 1:4 (1 minute) [150] was performed to reveal the microstructure of cross-section particles. Figure 1d shows the crystallized structure, and some gas pores as usually observed in atomized powders.

d Figure 1 a, SEM images a) of Zn powder and EDS analyses (insert), b) of Zn powder on the cross-section, c) the size distribution measured with laser granulometry d) Zn powder etched in sulphuric acid and water solution 1:4.SEM

3.3 Preliminary tests. Parameters boundary conditions

A few amount of information about Zn powder deposition with cold spray method could be found in the literature nowadays. They are most concerning coating deposition parameters, surface morphology, microstructure, and mechanical properties [28,31,52,133,147] and rarely the corrosion behavior [16,101]. In this work it has been decided to perform preliminary tests of Zn powder deposition to find out the limitation of the

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University conditions of the feasibility of the material deposition. To find out the influence of one parameter on the deposition process and coating properties, this parameter was manipulated and other parameters were fixed. Table 1 summarizes the fixed parameters established for coating elaboration. They have been selected in regard of their relevance to the process identified in the literature and will not be not changed during all the experiments in present work for experiment design. Stand-off distance between nozzle exit and the substrate was fixed at 30 mm, a traverse speed of 200 mm/s and a distance between gun passes of 1 mm. The nozzle/substrate angle was fixed to 90°. N2 was used as a gas carrier and spray gas and its debit is equal to 50 m3/h. Coating was elaborated from only one scan of the gun.

Table 1. Fixed parameters of the Cold Spray system in the frame of present work.

Traverse Coating Distance Stand-off speed, layer between gun distance, (mm/s) passes, (mm) (mm) 200 1 1 30

The temperature of working gas and its pressure are crucial parameters in the cold spray process as they both show the greatest influence on gas velocity and therefore the critical velocity of the particles at impact. [22,28,31,52,147]. They have been chosen as two variables to be adjusted according to the Doehlert uniform shell design. Only a change in gas temperature during spraying will be applied in the present work keeping the substrate at initial room temperature for all the tests.

In the frame of these experiments, the working temperature of the gas was changed from 100 oC to 320 oC at constant pressure 2 MPa . The maximum temperature was chosen according to the melting temperature of Zn, which is about 420 oC. As the cold spray process avoids melting of feedstock materials, it is not necessary to increase the temperature and overheat the feedstock. Anyway, Zn has high ductility on one side, which gives high bonding strength, and it is easy to oxidize at increased temperatures on the other side. Fresh Zn surface is covering with ZnO with a constant rate at room temperature and at saturated thickness it is nearly stopped. With temperature increase, the oxide layer thickness increases. At point 90 oC oxidation rate suddenly growth showing a second-stage of the o process [151]. The lowest temperature boundary 100 C was chosen according to low deposition efficiency. At lower temperatures the loss of powder is too high and there is no economically profitable interest for this coating deposition.

Thickness measurements of these coatings allowed us to find out the gas critical temperature for Zn deposition, which is around 230 oC see Fig 2. In a general way, the coating thickness represents indirectly the deposition efficiency. When the temperature is lower than this value, the coating is considerably thinner than that obtained above 230°C.

This phenomenon is generally explained by the elastic impact of the particles with a relatively low velocity that causes the loss of the feedstock powder. The thickness was estimated from 60 µm to 115 µm ± 5µm for one nozzle pass and stay stable for higher temperatures.

Figure 2, Determination of the critical temperature of deposition, 2MPa

Based on the literature review for initial tests, the pressure was applied at respectively 2 MPa, 2,5 MPa and 3 MPa which is commonly used for Zn coating [22,28,31,52,147].

Unlike helium, nitrogen has higher molecular weight and needs higher pressure to reach the critical velocitiy. For the coating prepared at different gas pressure and constant temperature 290 oC, the highest thickness value was obtained at 2 MPa, then the coating thickness tends to decrease with the pressure rise, Fig 3a. In contrast, the coating density calculated from the increasing weight and coating volume shows an incise with increase in pressure. It could be evidence of coating densification with pressure rise, Fig. 3b. This phenomenon can be explained by the fact that, the increase of pressure leads to a higher velocity of the particles in the working jet, and their plastic deformation rate/level during the impact gets also higher. Densification of coatings with pressure rise could be a result of pores closing between neighbor deposited particles. On the other hand, the abrasion effect at high particle velocities can take place and reduce the coating thickness.

Figure 3, a) Dependence of coating thickness on spraying pressure, 290 oC, b) Dependence of coatings density on spraying pressure

Another parameter influencing coating morphology is the powder feeding rate, which is determined by the number of blander disk rotations per second and gas pressure. For the setup used, it is expressed as 1/3 the value 1 is the number of blander disk rotations per second and 3 is powder feeding gas pressure in MPa. A sequence of the samples with different powder feeding parameters were prepared at constant gas temperature 290oC and pressure 2 MPa. These chosen parameters are 1/3, 3/3, 3/5 and 5/5.

The thickness of the coating with the same power consumption is evidence of the high indirect efficiency of spraying. In the plot of thickness dependence on the powder feeding rate (Fig. 4) one could see that the highest efficiency was obtained for parameters with equal values 3/3 and 5/5. The feeder rotation increase promotes deposition at a fixed pressure, however, further increasing of pressure reduces the efficiency of spraying. It could be suggested that the higher pressure of transport gas impede blender disk rotation and

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University continues flow of powder feeding. Thus, it is necessary to increase the number of rotation to form continuous suppling of powder to the jet.

Figure 4, Dependence of the coatings thickness on powder feeder modes, at 290oC and 2 MPa

Another important parameter at the cold spray process is the distance between the exit of the spraying nozzle and the substrate, the so-called standoff distance.

In present work, the influence of three values of standoff distance 30, 40 and 50 mm on obtained coating thickness was studied. The curve of thickness on the plot in Fig. 5 shows a decrease when the standoff distance goes from 30 to 50 mm. In general, three regions of standoff distance could be identified according to deposition efficiency: 1) the small standoff region with the appearance of the shock wave, which in turn reduces the deposition efficiency; 2) the medium standoff region, where there is no shock wave and particles keeps they critical velocities, this region shows the highest deposition efficiency; and 3) the large standoff region, where the gas velocity is not sufficient, it leads to particle deceleration and a decrease of the deposition efficiency [67].

Figure 5, Dependence of coatings thickness on spraying distance, at 290oC and 2 MPa

At increase of standoff distance keeping same spraying conditions, a decrease of thickness or fewer amounts of deposited particles was detected. For this reason the standoff distance will be kept constant at 30 mm for all future experiments in the present study.

3.4 Analysis

3.4.1 Coating roughness/waviness and surfaces

Different coatings were elaborated with gas temperature from 100 oC to 320 oC at constant pressure 2 Mpa, the samples were metallurgically characterized. 3D images of surfaces were collected by a confocal microscope. Since all coating wavinesses are comparable, only the observation performed on the coating layer built at 320 oC and 2 MPa is given in Figure 6 a. The surface consists in a valley and hollow with an amplitude of 91 µm. A modification of the waviness amplitude could be observed depending on temperature variation from 100 °C to 320 °C at constant pressure 2 MPa obtained at a preliminary test, Figure 6b . The maximum variation was found at 260°C (119 µm). SEM images reported in Figure 6c and 6d show the top surface and cross-section of the coatings.

The measurements of arithmetic roughness values Ra and Rz with contact profilometer showed the tendency of roughness decrease with temperature rise, Fig.6 e. It could be attributed to the consequence of a higher deformation rate of particles on the surface due to thermal softening.

The top layer of adhered particles exhibits a weak deformation rate compared to dense inner layers. It is due to the absence of a peening effect performed by following particles from spraying jet (Figure 6 b). Topography is controlled by the different size particles of feedstock. The highest «hills» correspond to slightly deformed particles or particles of a bigger size. At the same time, hollows correspond to single or clusters of low particle energy adhesion removed by spraying jet (Figure 6 c).

Figure 6, Surface analysis of roughness with a) optical microscope, b) Dependence of irregularities amplitude dZ from spraying temperature at 2 MPa, c) SEM photo of the top surface, 2MPa and 320 oC,, d) SEM cross-section of coating profile of zinc coating elaborated o with 2MPa pressure and 320 C, e) Plot of dependence Ra and Rz on T for the difference (±0.01 mm).

3.4.2 Adhesion properties of coating

To increase the adhesion of particles all the substrates were sand-blasted and cleaned in an ultrasonic bath for 5 minutes prior to cold spray deposition. High roughness is favorable since it helps to form stronger bending through mechanical interlocking. Christou et al. [159] and Danlos et al. [160] applied laser pre-treatment on the substrate surface to form special relief, remove the native oxide layer and also pre-heat the substrate. Deposition of aluminum particle showed higher adhesion result on laser treated surface than sandblasting surface. Anyway, the polished substrate gave a lower adhesion than the sandblasted one.

The particles with not sufficient velocity for adhesion strike the substrate and abrade the surface creating local morphological arrangement. From another hand these particles break oxide films on impacting surfaces, as the process happens at ultrasonic speeds there is no time for new oxide films to be formed between impacts [127,129].

However, all discussed mechanisms of surface preparation improve adhesion between the substrate and the first layer of incident particles. At the same time, the

Figure 7 presents the cross-section of Zn coating on carbon steel prepared at 230 oC and 2 MPa. It is seen that the size and shape of substrate grains changes close to the interface. Big axisymmetric grains show the nature microstructure of the substrate. At the same time during sand-blasting, the grain size has reduced and they undergo plastic deformation. During microstructure deformation, work hardening effect can take place and influence on bonding strength and adhesion in the interface.

Figure 7, Cross-section of ion beam polished sample of Zn coating on carbon steel prepared at 230 oC and 2 MPa, AsBin sulphuric acid with water at proportion 1:4

Bending test was chosen to evaluate the adhesion properties location of defects placed mostly at the interface. These defects could serve as stress concentration and lead to coating deadhesion. In Figure 8 a, a plot of loading under straining is shown. As all the coatings have similar curves, only one spraying condition 290 oC and 2 MPa is plotted. The bending test was performed at both sides of the sample. Both curves indicate constant elastic deformation with continuous loading from 0 to 350 N. Above 400 N, elastic deformation turns to plastic deformation. That indicates the beginning of bending. No damage of coatings was noticed during the test, SEM microscopy of the cross-section of the

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University coatings did not reveal any cracks or delamination at the bent area, adhesion of coating can be considered as very high, Figure 8 b.

Figure 8, a) Curves of displacement vs loading force and b) observations of bended sample, 290 oC, 2 MPa.

3.5 Doehlert uniform shell design for Zn coatings on carbon steel

Preliminary tests have shown the behavior of the material at different spraying parameters. Anyway, the cold spray process involves a wide variety of parameters and the process might be material and time consuming to investigate the influence of each parameter on resulting coating properties. In literature, examples of parameters application according to the special plan could be found. Different mathematical systems were created to simplify the procedure of parameter investigation. They allowed to avoid preparation of many samples and reduce the number of experiments. However, these Experimental design can predict the properties of coatings based on samples with neighbor conditions.

In the literature, several experimental designs exist such as Taguchi designs[152-154], Plackett-Burman designs [152,155,156] Box-Behnken designs [153,155,157] factorial designs [48,158] simplex design [152,158], star design [152,156], Fang and Wang statistical experiment method [144] and central composite designs [152,155,156,158]. Due to the visibility of results, ease of interpretation, the Doehlert uniform shell designs was chosen among them [155], [153]. The name “uniform shell designs” means uniform distribution of the points or chosen parameters on the spherical shell surface, Fig. 9a. By Doehlert design chosen experimental points are always located on the surface of hyper-sphere for k˃3 variables. Experimental points for two and three variables enclosed into a sphere with radius 1. The Figure 9b depicts projecting cross-section of the sphere in two dimensions giving the combination of the different variables at every experimental point. When ƞ is the total number of experimental combinations, and k is the number of variables, ƞ could be described as:

ƞ=k2+k+1 (1)

For two variables 7 experiments have to be chosen, their spatial distributions will form a hexagon with a point at the central point shown in Figure 9.

Figure 9, Uniform shell of experimental points in a Doehlert design, a) for 2 (points 1–7) and 3 (points 1–13) factors; b) hexagon, cross-section projection of the initial sphere [149].

To investigate a correlation between input parameters of system and coating properties the standard matrix for Doehlert’s uniform shell designs with two variables X1, X2 and 7 experiments was applied, (Table 2). In present work variables X1, X2 will correspond to spraying gas preheating temperature and pressure respectively. They are considered as the most influential parameters on coating properties and performances (porosity, thickness and corrosion resistance).

Table 2. Matrix for Experimental plan

№ X1 X2 1 +0.000 +0.000 2 +1.000 +0.000 3 +0.500 +0.866 4 -1.000 +0.000 5 -0.500 -0.866 6 +0.500 -0.866 7 -0.500 +0.866

Preliminary tests have shown, that temperature and pressure of working gas have the highest influence on properties of final coatings. Thus, they were chosen as two variables to be adjusted according to the Doehlert uniform shell design. Boundary conditions of temperature are set between 200°C and 320°C and the gas pressure 2 MPa, 2.5 MPa and 3 MPa. Table 3 represents the standard matrix of Doehlert uniform shell design with replaced standard X1 and X2 values by selected temperature and pressure and the final rebuilt matrix is presented in Table 4. Therefore, 7 coatings corresponding to the matrix values were elaborated for their response features characterization: porosity, thickness and corrosion resistance according to Doehlert uniform shell design.

Table 3. Replacement of matrix values X1 and X2 with working temperatures and pressure respectively

X1 T, oC -1 200 -0.5 230 0 260 0.5 290 1 320 X2 P, MPa -0.86 2 0 2.5 0.86 3

Table 4. Rebuilt matrix with parameters taking into account the temperature and gas boundaries conditions

№ T, oC P, MPa 1 260 2.5 2 320 2.5 3 290 3 4 200 2.5 5 230 2 6 290 2 7 230 3

3.5.1 Cross section Microstructures

Cross-sections of Zn coatings performed with different temperatures 230 - 320 oC and pressures 2 – 3 MPa are shown in Fig 10. a-d. Highly dense coatings are observed. However, some porosities remain. That is confirmed by X-ray tomography analysis and ImageJ software (shown in the next sub-section). Zn has the melting point close to spraying temperature, it leads to higher deformation and adhesion rates of powder during cold spray [30].

The surface deformation of the substrate influences the coating adhesion of the particles, creating porosity. When waviness of the substrate is not consistent with the particles size and deformation rate, weak interlocking and entrapped gas can be observed, see Fig. 10 b, c.

However, coating delamination or cracks are not detected which means that the adhesion between Zn and steel is good. After the chemical etching of the coating (Figure 11a), particle stacking is clearly observed in Figure 11b. One pass of spraying torch consists of about 7 layers of particles. No new phase or oxide in the coatings was detected by X-ray diffraction, Fig. 11c.

Table 5 shows the thicknesses of the coatings. A Maximum of 115 µm is obtained at 260 oC and 2.5 MPa. The minimum is obtained at 230 oC and 3 MPa; it reaches 60 µm.

Figure 10, SEM observations of coating cross section, elaborated with a) 260 oC and 2.5 MPa, b) 320 oC and 2.5 MPa, c) 290 oC and 3 MPa, d) 230 oC and 2 MPa

Table 5. Thickness of Zn coatings of the matrix

№ 1 2 3 4 5 6 7 Temperature, oC 260 320 290 200 230 290 230 Pressure, MPa 2.5 2.5 3 2.5 2 2 3 Thickness, ±5 µm 115 108 64 70 88 101 60

Figure 11, The SEM observations a) of the cross-section along the coating 320 oC, 2.5 MPa, and b) etching in sulphuric acid with water at proportion 1:4, c) X-ray diffractograms of Zn powder and as-sprayed Zn coatings.

3.5.2 Coating porosities

SEM cross-section micrograph show porosities in volume and at the coating/metal interface. Since surface preparation can affect their quantifications, X-ray tomography tests were performed on the samples. The results are presented in Figure 12. As observed on SEM cross-section, most of the heterogeneities are located at the coating/substrate interface highlighting the role of the roughness of the substrate on the interlocking process of Zn particles. Heterogeneity could be a result of internal stresses created by impact related to the different nature of the powder and substrate materials. Quantification was done by image analysis in volume and the percentage of the volume of porosity was extracted for

Table 6 shows low interfacial porosity percentages with a maximum of 4% for a gas temperature and pressure of 290°C and 3 MPa, respectively.

Table 6. Porosity values at the coating/substrate interface in volume

№ 1 2 3 4 5 6 7 Temperature, oC 260 320 290 200 230 290 230 Pressure, MPa 2.5 2.5 3 2.5 2 2 3 Porosity V, % 1.1 0.8 4.2 2.2 1.1 3.2 0.8

Different values of the interfacial porosity shows, that different spraying parameters temperature and pressure governs the density of coverage of the surface on the substrate by Zn particles. That can be associated with the low particle velocity which leads to low deformation rate and low adhesion of the particles, creating porosity. However, too strong particle impact can increase internal stresses at the interface or abrasive wear of as-deposited coating layer [15]. However, the rebound effect of the particles also take place if the particle energy is too low to ensure adhesion [67].

a b

c d

e f

g h

Figure 12, presents 3D images of Zn coatings with different parameters: a) overview of coated system scheme, b) 260 oC and 2.5 MPa, c) 320 oC and 2.5 MPa, d) 290 oC and 3 MPa e) 200 oC and 2.5 MPa, f) 230 oC and 2 MPa, g) 290 oC and 2 MPa , h) 230 oC and 3 MPa

3.6 Corrosion performance of Zn cold spray coatings

Figure 13a depicts the OCP evolution of the zinc coatings, composed of carbon steel and the bulk zinc sample. In an aerated and unstirred 0.1-M Na2SO4 solution, all the potentials attained a steady state in the first seconds of immersion. As expected from the thermodynamic tables, Zn materials attained cathodic potential (–1.5 V/MSE) before carbon

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University steel (–0.4 V/MSE). This confirmed its sacrificial role. At this stage, the effect of differences in porosity at the interface of the metal and coating could not be detected. Thus, the waviness and particle deformation modifications with the spray condition were considered. Also, the variation in potential among the seven samples was not significant to indicate any trend.

Figure 13b shows the linear sweep voltammograms recorded after OCP and the representative curves. The electrochemical behavior of the samples exhibited cathodic and anodic domains. The latter indicated charge transfer which allowed the use of Tafel plot extrapolation to quantify Icorr at the corrosion potential (Ecorr). At overpotentials (+0.3 V/Ec), limiting currents were detected at 10 mA/cm² due to limitations by the mass transport of dissolved species (Zn 2+). In the cathodic domain, the reduction of the media and the native oxide-hydroxide layers on Zn were expected. Zn oxide and hydroxide ratios slightly differed because of the local variation in pH due to the waviness of the surfaces. The current densities recorded for the bulk Zn and coating (1–10 mA/cm²) were higher than that of the steel (10 µA/cm²). This confirmed the high reactivity of Zn.

Figure 13, Results of corrosion experiments for Zn coatings on carbon steel a) Open circuit potentials of Zn coatings in comparison with bulk Zn and steel substrate, b) Potentiodynamic polarization curves of Zn coatings, bulk Zn and steel substrate.

The corrosion current density and corrosion potential were calculated based on the Tafel plot approximation. The authors considered these values as a trend, taking into account that the solution was aerated and the kinetics was partially controlled by the diffusion of dissolved oxygen. For this study, the obtained values can indicate a ranking used for the discussion. The corrosion resistance for each sample Rp was estimated using EC-lab software in a potential range of ±20 mV around Ecorr for each sample. Table 7 lists the corrosion potential (Ecorr), polarization resistance (Rp), and the current of corrosion (Icorr) for the 7 samples.

Table 7. Corrosion values of samples according to Doehlert design

-2 -² Sample name Ecorr (mV) Rp (Ohm.cm ) Icorr (µA.cm ) 1 260 oC 2.5 MPa -1595.4 1183 13.2 2 320 oC 2.5 MPa -1574.3 1346.8 11.9 3 290 oC 3 MPa -1590,7 1524.6 10 4 200 oC 2.5 MPa -1581.9 996.5 21.2 5 230 oC 2 MPa -1545.7 1234.8 11.6 6 290 oC 2 MPa -1568.8 1339 19 7 230 oC 3 MPa -1601.2 1180.8 10.8 Zn -1501.3 3511.5 7.1 Substrate -1049.6 983.4 4.6

Regarding the density of the coating, corrosion was primarily due to the waviness rather than porosity. Figure 7b shows a trend of decreasing the roughness at higher spraying temperature, except too high temperature 290 and 320 oC, which can be contributed to the abrasion effect. According to Table 7, higher temperature and pressure shows lower corrosion current density. Analyzing these results we can conclude, that higher roughness gives higher current density. This effect explains that high roughness of the coatings promoted occluded cells, which initiated localized corrosion. The corrosive solution propagated into the coating through the pathway of the heterogeneities. As the porosity level in the coating body was considerably low, the corrosive media propagated primarily through the interlamellar boundaries [33].

Cross-sections of the coating were observed using SEM after the electrochemical tests (6 h at OCP and LSV from OCP to 1.5 V/MSE), see Figure. 14. As the potentials of all the coatings indicated similar values, only the cross-section of one coating with spraying parameters of 290 oC and 2 MPa after the LSV test was investigated.

The dissolution of the coating was observed from the top layer. The thickness decreased twice from the initial coating, which was approximately 60 µm (Fig. 14 a,b). Corrosion was observed through the interlamellar pathway to the substrate. Unlike noble coatings for sacrificial Zn-coated steel systems, corrosion occurred on the surface and between particles of the coating. Some parts of the coating dissolved and the generation of defects at the metal-coating interface was observed. Some non-dissolved particles were still present, indicating that the weakest sites of the coatings were the boundaries between the deformed particles and the substrate-coating interface.

Corrosion was initiated on the surface of the particles due to the roughness, caused by local acidification. The particle boundaries exhibiting higher free energy (ΔG) were the preferred sites for corrosion propagation. Some of the particles may have separated from the coating before their total dissolution. EDS analysis revealed that oxidation occurred through the coating, mostly at the surface of the coating and coating-substrate interface, and a small amount in the coating; it corresponded to the solution penetration (Figure 14c).

Figure 14, Zn coating on carbon steel 290 oC and 2 MPa, a) SEM photo of the coating after LSV, b) not corroded coating, c) EDS analysis of Zn coating after LSV.

The photo of Zn cold spray coating after immersion in 0.1-M Na2SO4 solution for 1 year and EDS analysis of its’ cross-section are presented in Figure 15. The plot of atomic percentage distribution of chemical elements illustrates the presence of Fe at the substrate and Zn at the coating body. On the surface, oxygen was detected, while intensity of Zn reduced, it can be attributed to the formation of ZnO corrosion products. A small amount of sulfur indicates the nature of the electrolytic solution, Fig. 15b.

Figure 15, 290 oC and 2 MPa Zn coating on carbon steel a) The top side photo exposed to

0.1M Na2SO4 solution for 1 year, b) SEM image of cross-section and EDS analysis

SEM microphotographs of Zn coating cross-section in Figure 16 a and b show the propagation of corrosion through the heterogeneity of the coating. The top layer illustrates dissolution and active corrosion on the particle boundaries. At the same time inner layers remain dense and keep the same microstructure as non corroded samples, Fig 16.

During 1 year of immersion, the solution did not reach the substrate, no interface corrosion occurred. In the case of connected porosity or heterogeneity, an aggressive environment can pass through the coating and reach the substrate nullifying the protection properties of the anticorrosion coating. Thus, the densification of the coatings turns to a key parameter for cold spray anticorrosion coating formation. From another hand, small pores (0.1-0.5%) are not connected and cannot directly influence the electrochemical behavior [33]. Moreover small porosities could be closed by the formation of corrosion products and prevent penetration of electrolyte directly to the substrate.

Figure 16. a-b) SEM microphotographs of Zn coating cross-section after 1 year of immersion

o in 0.1M Na2SO4 solution, 290 C and 2 MPa

3.7 Discussions

This section discusses the relationship between spraying parameters and coating anti-corrosion performance. The analysis was based on hexagonal diagrams built from the Doehlert uniform shell design matrix; the correlation between spraying parameters, interface porosity, and coating thickness was observed. This correlation will be the first part of this discussion. In the second part, the observed trends will be compared with corrosion performances to validate the selection of parameters. The experimental points are placed at the corners and center of each hexagon (Figure 16-17).

Figure 17a shows the evolution of the interfacial porosity with the gas temperature and pressure. The wide part of the diagram indicates a low rate of porosity. The higher rate (4.2%) corresponds to the sampe № 3 with higher pressure 3 MPa and moderate temperature 290°C which promoted erosion of the substrate, as mentioned before, and lower accommodation of particles at the impact. The critical temperature which ensured the adhesion of particles and promoted the coating accumulation was 230 °C. Temperatures under 230 °C cannot be considered for industrial applications. The cold spraying parameters would not be efficient, which was confirmed by the analysis of the thickness diagram. As Figure 17b shows, the thickness of the coating was essentially thinner. The spray conditions induced erosion or rebounding of particles on the substrate. Thicker coatings were obtained at higher temperatures by promoting particle straining at the impact with moderate pressure to limit erosion phenomenon. Thus, temperatures must range from 230 to 320 °C and pressure from 2.0 to 2.5 MPa to obtain thicker coating. For material with a low melting point, the pressure is seemingly the first-order parameter. In summary of this first analysis, two sets of parameters should be selected: 320 °C–2.5 MPa (№2) and 260 °C–2.5 MPa (№1) , which are medium pressure and medium and higher temperature.

Figure 17, Hexagons based on Doehlert uniform shell design, correlation of spraying parameters T, P and a) interfacial porosity volume, b) thickness.

Corrosion performances were evaluated from electrochemical measurements to validate the trends observed. The analysis was based on polarisation resistance, and the results are shown in Figure 18. Higher values of polarisation resistance are shown in the upper part of the diagram. This trend can be discussed in two dimensions. Better efficiency was observed for 3.0 MPa and 290 °C (№3) . This field corresponded to a thinner coating with heterogeneities. This was an interesting observation because the results indicated the efficiency of the coating in terms of sacrificial coating. As mentioned in the previous part, a defective coating essentially promotes cathodic protection, which promotes Zn dissolution. As observed by Souto et al., corrosion products exhibit self-healing phenomena which ensure inhibitive protection [148]. A higher corrosion resistance was also obtained at a higher temperature (290 °C

As Figures 17a and b indicate, because the porosity rate was lower and thickness higher for 260 °C and 2.5 MPa (№1) , the coating properties promoted a higher corrosion resistance. However, the polarisation resistance was not optimum. This result can be explained by remembering that roughness promotes corrosion initiation; thus, a higher roughness increases corrosion initiation. As Figure 5d shows, this was the scenario at 260 °C and 2.5 MPa №1), and explains the lower corrosion resistance of this set of parameters in comparison with 320°C and 2.5 MPa (№2).

Figure 18, Hexagons based on Doehlert uniform shell design, correlation of spraying parameters T, P and Polarization resistance.

The best compromise was observed at 320 °C and 2.5 MPa (№2) . A high quality of Zn coating could be obtained that exhibited low porosity, larger thickness, and high corrosion resistance. This coating had the lowest interfacial porosity (0.8%) and a high coating thickness of 108 µm whereas the corrosion resistance was also high at 1347 Ω/cm2.

3.8 Conclusions

 The Doehlert uniform shell design is a useful method to lower the number of experiments. This paper describes the effect of gas temperature and pressure on the coating microstructure and predicates further changes in the microstructure.

 Different sets of parameters enable two protection modes: the sacrificial mode for heterogeneities in the coating (290 °C and 3.0 MPa), and corrosion product protection for homogeneous and dense coatings (320 °C and 2.5 MPa).

 The mechanism of corrosion is described as conditioned by corrosion initiated by roughness and corrosion propagation through interlamellar microstructure controlled by an adiabatic shearing.

 A temperature of 320 oC and pressure 2.5 MPa represent the set of parameters that provide the highest corrosion performances of Zn cold spray coatings.

Chapter IV: Relationship between the microstructure of Zn coating and corrosion behavior

The corrosion behavior of the material is influenced by several factors. Electrochemical nature and composition have the most significant effect on corrosion mechanisms. At the same time, morphology and microstructure of material also play an important role. In this chapter, the influence of microstructure on the corrosion behavior of cold spray Zn coatings will be analyzed. Open circuit potential (OCP) and Linear sweep voltammetry (LSV) will be performed at macro- and micro- scale measurements.

4.1 The strategy of research and coating parameters choosing

Richardson and Wood have proposed the theory of the role of defects like inclusions, dislocations and grain boundaries as privileged sits of corrosion pit initiation [206]. Other authors also have reported a higher reactivity and localized corrosion of grain boundaries and different phases [90,94]. Creation of local corrosion cells between phases of the coatings or high porosity also lead to poor corrosion resistance [207]. Local corrosion attack could be formed as a result of the potential difference between grain boundaries, impurities, and matrix of material [208]. K. Yasakau et al. have demonstrated that in first seconds of galvanized Zn coatings immersion in 5 M NaCl solution, the starting point of corrosion takes place on grain boundaries and then expends to the surrounding matrix at the longer time of immersion [90]. Thereby, particle boundaries formed by cold spray show weak place in the coating with a higher rate of corrosion exposing analogically to grain boundaries and other defects.

The influence of crystallographic orientation on corrosion resistance of Zn was reported in literature [207,209]. Low index grains are exposed to thicker crystalline oxide formation. While electropolishing, a gray coating with similar to passive film characteristics has covered Zn surface. The growth of oxides preferentially takes place at a metal/film interface with a strong dependence of the crystallographic surface structure. Oxide film shows a hexagonal surface structure with a cell size about 3000 Ao [210,211]. Different grains show different corrosion rates depending on the binding energy of atoms between the crystallographic plans [212,213]. J. Scully has reported that the closed packed or index-low planes are more corrosion resistant due to the higher binding energy of the surface atoms. The higher number of nearest neighbor atoms forms stronger bonds and needs more energy for they break [212]. The character of grain boundary distribution or particle boundaries in case of cold spray coating also influence the corrosion resistance [214]. In the case of hot-dip zinc coatings, a better corrosion resistance is attributed to the basal (0001) texture than (1122) according to Takechi et al [215]. Raichevski and Vitkova have investigated hcp metals for corrosion resistance and have concluded that the corrosion resistance of the texture decreases as (0001) ˃ (1011) ˃( 1120) ˃ (1010) [216].

In present work 3 different surface conditions will be considered for the electrochemical test: rough surface, polished surface and etched after polishing. Each surface condition presents different heterogeneities exposed to the capillary area.

According to Doehlert uniform shell design (Fig. 1) from the previous chapter it was concluded, that the highest corrosion resistance is achieved with the use of 290 oC and 3 MPa. High temperature and pressure leads to particle velocity increase and stronger impact of particles with the substrate and deposited layer. Higher rates of plastic deformation densify coatings and may change nanostructure and internal stresses balance. All these factors can lead to higher corrosion resistance of the coating.

However, interfacial porosity at this point is also the highest due to the increase of the erosion rate of the substrate at high velocities of particles (see Appendix). Thus, the highest protection rate of the sample 290 oC /3 MPa keeps until the solution reaches the interface. After that, the corrosion rate will increase faster due to solution propagation into pores and the creation of local corrosion cells.

From another hand, internal porosity of all coatings is very low, about 0.1-0.5 % at the same time at 260 oC and 2,5 MPa the coating thickness is the highest.

All features described above gives the interest of more investigation of samples 260 oC /2,5 MPa as the thick dense coating with good anticorrosion properties, 290 oC /3 MPa. with the highest anticorrosion properties and 200 oC /2,5 MPa the coating with the lowest response parameters

a b c Figure 1, Hexagons based on Doehlert uniform shell design, correlation of spraying parameters T, P and a) interfacial porosity volume, b) thickness) and c) corrosion resistance

4.2 Experimental part

At present work a standard 3-electrode electrochemical micro-cell attached to optical microscope “OLYMPUS BX51M .Japan” was used, Fig. 2. Aerated and stagnate 0,1 M Na2SO4 electrolyte was provided to the surface of zinc coatings (working electrode) by glass microcapillaries with tip sizes 40-60 µm. The platinum counter electrode was embedded into the acrylic microcell close to the capillary, reference electrode Ag/AgCl, KCl (saturated)

(EAg/AgCl = 0.199 V/NHE) was placed outside the microcell. To cut electromagnetic noise the experiments were conducted inside a copper faraday cage.

Microcapillaries were cut with pull-off machine “FLEMING/BROWN Micropipette Puller P-1000, USA ” by heating of small glass tubes (with internal diameter 0,84 mm) in the middle up to their melting point and pulling out in opposite directions. Then, to open the mouth of microcapillaries, they were polished with grinding paper 2000 and distilled water. To form a silicon gasket each microcapillary was dipped into “DOW CORNING 732 MULTI- PURPOSE SEALANT MIL-A-46106” silicon drop, Fig 3. A camera “DIGITAL.B006 MICROSCOPE ” with “AMCAP” software was used for accurate dipping into the silicon and precise position of microcapillary on top of the sample. The opening of the tip mouth after dipping into the liquid silicon was performed by pressing a stream of ethanol with a syringe before solidification of the silicone. Then, the microcapillary with the silicon gasket was dried in the air for 10 minutes and the process was repeated 5 times. Obtained microcapillary with silicon gasket was suitable for multiple measurements. Each set of experiments was provided with the same microcapillary. Electrochemical measurements were performed with “Gamry Instruments reference 600+ Potentiostat/Galvanostat/ZRA”.

All experienced samples were polished and etched for 1 minute with sulphuric acid with water at proportion 1:4. Open Circuit Potential (OCP) and Linear Sweep Voltammetry (LSV) were measured for all three samples with the use of GAMRY Ref600 (USA) potentiostat.

b Figure 2, a) Electrochemical 3-electrode micro-cell setup, b) glass microcapillary on sample surface

Figure 3, Camera observation of glass microcapillary with silicon gasket on sample surface

4.3 Analysis

To characterize electrochemical behavior of the seven samples of matrix from previous Chapter 3 macro- experiments were performed with the use of conventional 3- 2 electrode cell, at the apparent surface of 0.28 cm . A mercury saturated electrode (EMSE = 0.658 V/NHE) and carbon rods of 6.7 cm2 were used as reference and counter electrodes respectively. Open circuit potential was recorded for 6 h and Linear Sweep Voltammetry (LSV) were carried out from -0.1 V to 2 V according to OCP with a scan rate of 1 mV/s.

Electrolyte was an unstirred 0.1-M Na2SO4 solution at pH 6.5 and 20°C. OCP and LSV curves were also plotted on Zn plates (99,9% purity, Goodfellow) to compare the electrochemical behavior of bulk material with cold sprayed Zn coatings.

After immersion in electrolyte the potentials of all coatings were stabilized during several seconds. Zinc coatings confirms theis sacrificial behavior by presenting a higher cathodic potential (-1.5 V/MSE) than the carbon steel (-0.4 V/MSE), Fig 4.

Figure 4, OCP at macro-scale measurements of 7 Zn coatings and bulk Zn and build in Doehlert plot of corrosion resistance depending on spraying parameters

After OCP Linear sweep voltammograms were recorded, corresponding curves of 7 samples are presented in Fig. 10. The plot indicates the anodic domain of corrosion behavior of the surface after the OCP test. Close to the corrosion potential electrochemical reactions could be limited by charge transfer due to Tafel behaviors [217]. At overpotentials (+0.3

V/Ec), limiting currents were detected at 10 mA/cm² due to limitations by the mass transport of dissolved species (Zn2+). The difference in the current is not very high, but still it represents some change in corrosion resistance Fig. 4-5.

At macro-size experiments the investigating area is high and takes 6 mm in diameter. The exposed surface contains many corrosion activation points, which serve as a cathode in the electrochemical cell-like surface cavities, grain boundaries, inclusions and others. Only general information can be obtained at the macro –size experiments, as the local behavior is masked by activating points’ reactivity. To receive more information about the role of the

Figure 5, Conventional large scale LSV measurements of 7 samples of matrix

Ectrochemical micro-cell OCP measurements of samples at 260 oC/2.5 MPa with rough, polished and etched polished top surfaces are presented in Figure 6. Stabilized potentials of samples are close, in the range from -0.96V to -1.02V. It differs from general OCP test for about 0.5 V, Fig.9.

a) b) c)

Figure 6, OCP micro-scale measurements of Zn coatings with 260 oC/2.5 MPa top surfaces, a) on rough surface, b) on polished surface, c) etched after polishing surface

A shift of LSV potential between micro and general measurments can be seen in Figure 7. Unlike local measurements of a precise area at a conventional cell, all the exposed area gives electrochemical response including overlapping of current from heterogeneities [197]. All particles, grains and heterogeneity exposed at macroscale measurement

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University contribute to the sum current by their individual degree of coverage and electrochemical response [210]. Farzin Arjmand and Annemie Adriaens have studied potentiodynamic polarization curves of 304-L stainless steel in a 1-M sodium chloride solution, by comparison in macro- and micro-scale. They have reported smaller current density at the macro-scale and the shift of corrosion potential to nobler at micro-scale. This change was attributed to a higher rate of the oxygen reduction [165,183]. The ratio of heterogeneity on the surface/matrix means the ratio of cathode/anode in the electrochemical system. It is higher at the macro-scale, which means more cathodic (defects) area, more oxidation and higher corrosion rate.

From another side, miniaturizing of electrochemical cell gives a higher influence of oxygen penetration and Ohmic drop between reference and working electrodes, especially for cathodic reactions [218]. These factors can shift corrosion potential at a comparison of electrochemical response at macro- and micro- scale.

Figure 7, Comparaison of LSV curves obtained with conventional and micro-cell

Small micrometer-scale sizes of microcapillaries and application of silicon gasket allow measurements on rough surfaces without special preparation. Cold spray coating applied at 260 oC and 2.5 MPa shows the highest thickness, low interfacial porosity and high density. The electrochemical behavior of this sample will be investigated with 3-electrode micro-cell at 3 surface state: as-received after cold spray process (rough), polished and etched after polishing.

At rough state surface measurements, several curves from different areas were received with the same microcapillary after 9 minutes of OCP, Fig. 8. Corrosion potential was stabilized around -1,4 V/MSE for the different curves.

Figure 8, LSV curves of rough coating in different places with 260 oC and 2.5 MPa

The reaction of Zn anodic dissolution was presented by Dirkse [219]:

- Zn + 4OH = Zn(OH)4 + 2e

After this reaction formation of oxides ZnO , Zn(OH)2 or both is possible by following reactions:

Zn + 2OH = Zn(OH)2 + 2e

- Zn(OH)2 + 2OH = Zn(OH)4

- Zn+2(OH) = ZnO + H2O + 2e

- ZnO+2OH + H2O = Zn(OH)4

Cathodic overpotential initiate the partial reduction of the oxide-hydroxide layers on Zn whereas the metallic surface composed by oxi-hydroxide directly dissolved:

- O2+2H2O+4e = 4OH

Due to the release of hydrogen, the effect of capillary blockage often takes place at the cathodic branch, which obstructs measurements. Anodic part is of the most interest in this work, as it represents coating dissolution and corrosion behavior of Zn.

Most of the curves of the anodic part are very close, which shows that microcapillary could be well-positioned on the surface and the cavity of roughness could be compensated with a flexible silicone gasket. On another hand, some places on the surface show an increase in current density response. This weak places should correspond to surface defects, like particle/particle boundary, which demonstrate high-stress concentration, change in nanostructure and faster corrosion activation and propagation, [206-208].

Polishing is often performed before electrochemical investigations. Analysis of smooth surface allows excluding the effect of surface morphology heterogeneity on corrosion initiation.

For polished surfaces, micro dimensions of experiments bring some difficulties as the accurate location of the capillary on the top surface of the coating, but not the bare substrate. While polishing at thin coating areas some islands of bare substrate could be open, Fig. 9, 10. EDS analyses of the corroded area have proved the Zn nature of coating contained some oxygen on top and presente of Fe in the islands of bare substrate. The corrosion test was successfully done on the Zn coating surface for all the samples, Fig 10-12. The size of capillary trace after LSV experiment is about 50 µm for the coating with 260 oC / 2.5 MPa, 85 µm for 290 oC /3 MPa and 60 µm for200 oC/ 2,5 MPa.

Figure 9, Cross-section of Zn coating on carbon steel and possible polishing plane

Figure 10, SEM photo of the top surface with capillary trace, 260 oC /2.5 MPa, b) EDS analysis of capillary trace

Figure 11, SEM photo of the top surface with capillary trace, 290 oC /3 MPa, b) EDS analysis of capillary trace

Figure 12, a) SEM photo of the top surface with capillary trace, 200 oC / 2,5 MPa, b) EDS analysis of capillary trace and islands of bare substrate

Potential curves of polished surfaces are very close and show a similar tendency, Fig. 13. Polished surfaces do not have high cavities like as-resaved rough samples, but still can demonstrate some porosities, biggest particle boundaries and have internal stresses, Fig. 14a.

Figure 13, LSV curves of polished surface 260 oC/2.5 MPa

Figure 14, a) polished surface of sample 260 oC /2.5 MPa, b) trace of LSV and etching droplet on the polished surface, 260 oC /2.5 MPa

In Fig. 14 b polished top surface of the sample 260 oC /2.5 MPa after the LSV test is presented. The surface after polishing presents only the biggest defects of particle boundaries. At the same time a drop of etching solution of sulphuric acid and water in proportion 1:4 has revealed fine heterogeneities in the coating, Fig. 14 b. These heterogeneities have cold sprayed nature, they are formed by collision, deformation and adhesion while spraying. The location of the capillary for the LSV test could be seen, the diameter of corrosion spot is about 54 µm, it corresponds to a capillary diameter of 50 µm. One can admit the propagation of corrosion solution through the particle boundaries into the coating and thus the creation of localized corrosion in places with more heterogeneities.

LSV curves obtained after surface polishing and etching preparation are shown in Fig.15. Corrosion potentials have stabilized at values from -1.37 to -1.5 V/MSE. The curves represent the same trend as all Zn coatings, but also show weaker place with higher current density. In Fig. 16 polished and etched coating 260 oC /2.5 MPa with an approximate area of capillary of 60 µm is shown. The density of particle boundaries is not constant in the chosen area. From one side the feedstock particles differ by size, some particles are bigger than others and their exposed area has fewer heterogeneities. On another hand, particle velocity, temperature during impact, angle of collision and other factors are different. Consequently, deformation rates, which form variation in boundary density change, a higher deformation rate gives a higher covering area. Different places of coating surface could be exposed to microcapillary experiments. The same exposed area may contain clusters of small particles with more boundaries and show higher heterogeneity density or cover only one particle with a low amount of defects, Fig. 16 areas I and II.

Figure 15, LSV curves of top polished and etched coating with 260 oC / 2.5 MPa

Figure 16, 260 oC / 2.5 MPa top polished, etched

A comparison of polished and etched plots of 260 oC /2.5 MPa sample shows very close curves and similar tendency, Fig. 17. A small shift in potential and current density corresponds to structural heterogeneities in the coating.

Figure 17, LSV curves 260 oC / 2.5 MPa, comparison of top polished and polished etched surfaces electrochemical response

In Fig. 18. the plot of LSV curves obtained from the polished and polished etched sample condition with 200 oC / 2.5 MPa is presented. In general, obtained LSV curves are similar, anyway some differences in current density could be registered.

The density of heterogeneities on the chosen area is different, thus the amount and density of forming oxides could vary. A similar effect was found at the investigation of the pit initiation potential of stainless steel. When more weak points were exposed to measurements corrosion potential was moving to more negative values [165]. Suter and Böhni also reported, that higher amount of weak points for corrosion initiation reports less noble potential [197]. From another hand, presence of defects, internal stresses different grain orientation and the deformation rate also can give influence on obtained current density [90,94]. During cold spray some load from the pressure of spraying jet and peening effect take place, it leads to high deformation of particles. At cold spray particles are exposed to a different rate of shear stress and shear deformation. Moreover, first layers of deposited particles or sublayers undergo higher deformation rate and densification, than the top layer of particles. Extremely fine grains were detected at the boarders of particles, they appeared due to fracturing of big grains while impact according to Chang-Jiu Li et al. [52]. All of these factors lead to the creation of internal stresses. Accumulation of residual stresses in the coating increase dislocation density, which influences the energy of charged defects formation and donor-acceptor densities at oxide film formation [220,221]. With residual stress increase the anodic current of copper increases in the passive region [222].

It has to be noticed, that even with the application of silicone gasket on the tip of microcapillary it is not preventing the penetration of some oxygen into the micro-cell, which could change pH of the solution and give some differences in obtained values. The shift of corrosion potential could be attributed to a higher rate of oxygen reduction, which increases cathodic current density [165,183]. For copper in NaCl solution with alkaline pH values, the pit initiation occurs at a more negative potential. From another hand, the passive current is higher stabilized at low acidic pH values [184].

Figure 18, Anodic part of LSV curves of polished and polished etched after polishing of the top surface of the coating (200 oC / 2.5 MPa)

In some cases, obtained LSV curves can show unusual corrosion behavior, for example slowly increase in current density before reaching limit current at 290 oC/ 3 MPa sample, Fig. 19. Due to the irregular area of heterogeneities, their role in electrochemical reactions also differs. We can suggest, that thick or dense oxide layer at one point reduces the activity of covered surface compared to another region of the same sample [90].

Figure 19, LSV curves of polished and etched after polishing top surface of the coating with 290 oC and 3 MPa

In previous Chapter 3 the influence of spraying parameters on corrosion resistance of cold spray coating where studied. Сorrosion resistance Rp of three chosen samples of the DOE matrix from previous Chapter III increase respectively: 200 oC/ 2.5 MPa – 996.5 Ohm.cm-2 ; 260 oC / 2.5 MPa – 1183 Ohm.cm-2 ; 290 oC/3 MPa – 1524.6 Ohm.cm-2

Comparison of current densities for these three samples obtained with electrochemical micro-cell showed a similar trend. The highest current density is measured for the sample 200 oC/ 2.5 MPa, and the lowest for 290 oC/ 3 MPa corresponding to the highest corrosion resistance, Fig. 20. We can conclude that the microstructure heterogeneities may influence the coating corrosion protection. Microcapillary test proves the tendency of corrosion resistance obtained from general electrochemical measurements at previous Chapter III, Fig. 4 c.

Figure 20, Comparaison of LSV curves of Zn coatings with 200 oC/ 2.5 MPa, 260 oC/ 2.5 MPa and 290 oC/ 3 MPa and build in Doelert plot of corrosion resistance depending on spraying parameters

4.4 Conclusion

 Different area of heterogeneities on the exposed surface can shift the corrosion potential and current density;

 In general micro-scale experiments have proved the trend of corrosion resistance of Zn coatings sprayed at different parameters obtained at the macro-scale. The highest current density was obtained for the coating with 200 oC/ 2.5 MPa and the lowest - with 290 oC/ 3 MPa, which is in good correlation with corrosion resistance Rp;

 At precise investigations we can conclude, that the change of cold spray parameters indicates some difference in corrosion resistance of Zn coatings. However, the variations are very small and taking into account the possible oxygen penetration with following a change in pH value, the influence of spraying parameters on corrosion resistance is not crucial. It means, at macro-scale all Zn cold spray coatings present a high corrosion resistance and spraying parameters could be adjusted according to mechanical properties of the coating keeping high corrosion resistance.

Chapter V: Low-pressure cold sprayed Al and Al/Al2O3 coatings on carbon steels

In this chapter, the work for aluminum coating deposition for corrosion protection of carbon steel with a low-pressure cold spray method (LPCS) is presented. Optimization of parameters is described and coating densification methods: addition of Al2O3 particles and laser post-treatments are proposed.

5.1 Introduction

Aluminum, its alloys, and composites are widely used for corrosion protection, in aviation, aerospace, automotive, military, electronic packaging industries on bridges and water towers [223,224]. Aluminum has been recognized to be the most effective for steel protection in offshore structures. It can serve in the most aggressive corrosion environment and give protection for 15-50 years [223]. Moreover, aluminum can be applied for repair and restoration of details. Nickel 5 wt.% Aluminum powder is an example of a popular restoration material [223]. Aluminum well serves as anodic coatings for sacrificial corrosion protection of details, which represents a cathode [223].

Different methods of Al application: anodization is one of widely spread method of aluminum coating application [225], chemical vapor deposition [226], electrodeposition [227], micro-arc oxidation [228], conversion method [229] and recently additive manufacturing of aluminum and its alloys have attracted broad attention from industrials customers [230].

The thermal spray process is widely used for surfacing and resurfacing of engineered components [231,232]. Depending on energy source thermal spray could be divided on combustion (Flame spray, Detonation deposition, High-Velocity Oxygen Fuel (HVOF)) and electric energy (Wire Arc and Plasma) [223,233]. This technology allows the deposition of pure materials, alloys and powder mixtures to protect or improve the performance of the substrate [223].

In thermal spray methods, the coating is formed by the heating of the feedstock material, melting and then quenching the molten droplets by working gas or plasma at temperatures higher than 15000oC [223]. Resulting coatings often have some inclusions, oxides, diffusion to the substrate and varying degree of porosity. These kinds of coatings can undergo chemical changes or phase transportation and usually present low corrosion resistance characteristics [234]. Resulting coating properties depend on both thermal and kinetic energy [223]. For materials sensitive to heat and oxidation, such as aluminum the use of cold spray technology would be more suitable. The cold spray processing was introduced in the surface

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University engineering as a relatively new technology of coating application with high performance and low cost. This technology relies mostly on kinetic energy and less on heating, it can provide a high thickness of deposits in the range of 200-250 micron [111,223]. Unlike thermal spray, cold spray coatings show less porosity, oxidation, change of structure and it avoids chemical or phase transformation [233,235].

Various materials were successfully deposited with the cold spray method: Cu [236],

Zn [237], Ni [238], Ti [239], cermet WC–Co [240] and ceramics like WO3 and Y2O3 [241]. Al and Zn cold sprayed coatings are widely used for corrosion protection in oil and auto industries and others [234,242].

Cold sprayed Al coatings are a relatively new technology for corrosion protection among other methods. Anyway, the technology has shown promising results in the formation of aluminum anti-corrosion coatings due to the low processing temperature of the process [243-248].

Cold spray system could be designed as a manual (portable) or robotic (fixed) system. The method can be divided into the high-pressure cold spray (HPCS) and low-pressure cold spray (LPCS). Both systems usually include a source of compressed gas (He, N2, a mixture of

He and N2 or dry air about 79% N2 and 21 % O2), gas heater, supersonic de Laval-type nozzle, powder feeder (for powders in the range of 1-50 µm size), monitoring system to control spraying parameters and spraying chamber with motion system [234,249].

In the case of HPCS, nitrogen or helium as the working gas is applied at a pressure of about 1.5-4 MPa [234] and could be preheated up to 1000 oC to optimize its aerodynamic properties. Passing through the converging-diverging nozzle the working gas expands and converts the pressure into kinetic energy accelerating to supersonic mode up to 1200 m/s and reducing the temperature [249]. The powder is axially fed in a solid-state into the system in the pre-chamber of the nozzle. At diverging part of the nozzle, accelerated particles in solid-state reach velocity in the range of 600-1500 m/s. Mixed working jet strikes the substrate with kinetic energy sufficient for high adhesion and creation of mechanical/metallurgical bonding. Due to the short time of contact the particle and spraying gas, its temperature remains below initially preheated working gas temperature [250]. High- pressure cold spray deposition efficiency is very high and can reach 90% depending on materials.

The low-pressure cold spray (LPCS) uses air or nitrogen as accelerating gas at a pressure in the range of 0.5 – 1 MPa. The working gas could be preheated up to 550 oC and reach a velocity of about 300 – 600 m/s at the exit of the nozzle [249]. For LPCS, the feedstock powder in a solid-state is radially fed through the throat section of the supersonic nozzle and injected in the gas where it is accelerated toward the substrate. The powder is

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University drown in from the powder supplier by the effect of keeping the static pressure in the nozzle lower than the atmospheric pressure, which is called the Venturi effect. It can be achieved, when the ratio of the nozzle cross-section area at the point of powder entry (Ai, m2) and throat diameter (A*, m2) [250] satisfies the equation 1:

Ai/ A* ≥ 1.3Po + 0.8 (1) where Po is the pressure at the nozzle inlet (MPa), Low-pressure cold spray methods have been used for deposition of materials such as Al, Cu, Zn, Co and Ni [18]. However, deposition efficiency of low-pressure cold spray coatings typically does not exceed 50%, unlike HPCS, where 90% efficiency can be reached. However, spraying cost is significantly lower than that of the HPCS system and niche applications can support this technology as the low-operation costs can overcome the higher power consumption [18]. The LPCS system is improved in its operational safety, it is more portable, flexible and automated. Moreover, the service life of the nozzle is longer at LPCS due to the powder feeding supplier into the diverging part of the nozzle avoiding passing of powder through the throat and keeping it out of wear [249]. Aluminum is well suitable for cold spraying due to its high ductility and low density [111]. Commercially Al powders are available in a variety of composition. Researches have performed cold spray deposition of Al on different substrates [244-248,251-254] as pure aluminum deposition [129,247,255,256] and al-based coating formation [122,257,258].

Particle critical velocity is an important parameter for successful coating formation, but deposition and adhesion at high rate depend on morphology and state of the surface. For high-quality adhesion, a special preparation is needed, including desorption of adsorbed pollutants, oil, and dust by sandblasting and following chemical degreasing [26,259-261]. In many cases, surface modification is necessary to reach sufficient adhesion between the substrate and depositing particles. Ghristoulis et al. [262] have studded aluminum - aluminum alloy system and shown that non-treated substrates remain its oxide layer. The fracture test has shown primary separation at entire particles and particle boundaries. In this case, the bonding mechanism was attributed substantially to mechanical interlocking and less to metallurgical bonding [263-265]. Dense coating with a high rate of adhesion could be formed only by metallurgical bonding while braking of initial oxide films on working surfaces and creating the contact of fresh material. To achieve metallurgical bonding, spraying velocity have to be high enough to brake natural oxide films on particles and substrate [26].

The thickness of the oxide layer on aluminum could be in the range of 20-200 nm [266]. For aluminum surface preparation grit blasting, sandblasting or laser process could be used. Blasting operations not only removes the oxides, but also modifies the surface morphology in one hand, and brings impurities on another hand. Thus, the following rinsing

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University in ethanol for several minutes is necessary [267]. Increasing the interfacial area of contact by surface variation promotes mechanical anchoring and leads to adhesion rise [160,268,269]. The adhesion of aluminum coatings was reported about 25 - 30 MPa [111]. Higher efficiency of Al deposition was noticed while spraying Al particles on grit blasted steel, then on ground steel substrate [111]. As well as higher deposition efficiency of Al powder on grit blasted aluminum, than on polished aluminum plates [270].

During impact the temperature of particles and substrate in contact area increases, it causes softening of material by adiabatic shear instability (ASI) [256]. ASI is a dominant mechanism of nanostructure formation by recrystallization of grains [271,272]. Resulting in severe deformation of particles while deposition forms the different regions of the coating structure. It could be divided into three parts: 1 – interior particle, it is slightly deformed particles keeping initial microstructure of feedstock, poorly organized network of dislocation and a low angle of grain boundaries; 2 – pancake grains morphology with higher deformation rate, more dislocation density, and high angle grain boundaries; and 3 – the region with nano-scale grain refinements at severe plastic deformation, when the shape of feedstock particle cannot be seen. These three regions are well studied in the literature on 7075 Al cold spray deposits [273-275].

During cold spray process, high rates of plastic deformation at cold spray gives high strength to deposited coatings due to cold working effect. From another hand, this effect can lead to ductility decrease [15,258,274,276].

The main challenge of cold spray research is the optimization of parameters. Since the final coating microstructure is very sensitive to processing conditions (gas speed, temperature, number of layers etc.) different rates of severe plastic deformation could be formed [277]. By tuning the effective spraying parameters of the carrier gas, the powder velocity and thus coating best quality without interfacial delamination is achieved [278- 280]. Moreover, a different area of grain boundaries, dislocation density, pores end defects lead to changes in the coating strength and corrosion behavior [271].

To improve the quality of cold spray coatings a row of spraying parameters was studied in the literature to optimize coating properties and to reduce porosity [258,274,275,281].

Different researches have studied ductility increase during temperature rise in aluminum alloys. Wang et al. [282] have shown temperature influence on Al alloy AA2024 formability, the maximum ductility was obtained at gas preheating 450 oC. Gao et al. [283] have reported similar data for AA2060. Mohamed et al. [284] have measured the uniaxial ductility of AA6082 at three high temperatures preliminarily. He has reported the ductility increase with forming temperature rise and decrease of flow stress [285].

100

Marzbanrad B et al. [286] have demonstrated that high temperature is the most important parameter for residual stress formation. Moreover, increased deposition temperature together with the peening effect leads to dynamic recrystallization of the substrate surface and formation of nano-structure on the boundaries [286]. The residual stress distribution influences the coating and substrate properties as well [287-288]. In the case of Al coating deposition by cold spray, residual stress accumulation can play a role in the formation of a dense coating, since Al has a low density and provides low impact pressure [289]. In literature, some findings on the influence of propellant gas temperature/pressure on final properties of the coating, residual stress induced and fatigue performance are presented [290-292].

In the work of Marzbanrad B et al. [286] while studying of cold spray deposition of aluminum 7075 powder onto AZ31B-H24 magnesium substrate, it was concluded that during spraying process the substrate can reach the temperature of 330 oC and release obtained compressive strain. The size of substrate grains is a function of absorbed thermal energy and grain size is larger with a high amount of transferred thermal energy. The energy generated by the peening effect of striking aluminum 7075 particles led to substrate surface grain refinement [286]. Usually it is forming compressive residual stresses in the coating and at coating/substrate interface [293]. The rate of residual stress in Al and Cu coatings depends on a degree of plastic or, strength level of the coating and substrate and their elastic modulus [292]. In the case of different materials of powder and substrate, the difference in the thermal expansion coefficients also influences the formation and magnitude of residual stresses near the interface [294].

Micro-pores are one of the most crucial defects in aluminum coatings. It reduces strength, fatigue, the toughness of Al and its alloys [295]. Depending on the coating application, a decrease in mechanical properties can act as a key weakness. Talbot et al. [296] have reported a crucial weakening of aluminum alloy mechanical properties at porosity volumes of 0.5% - 1%. In the case of anti-corrosion application, porosity is a key factor of coating quality.

On one hand, pores serve as the path for corrosion solution and on the other hand as the accumulation of stress concentration yielding the load-bearing capacity by microspores and leading to a fracture [297]. According to Mayer et al. pores in size larger than 50 µm are generally crack initiators [298].

By optimization of spraying parameters, some metallic materials like steel [299], titanium [300] or Inconel 718 [301] can reach a high density of about 99,5%. However, the aluminum alloy system is highly undergoing porosity effect at additive technologies [302].

101

It was reported by J. Gu et al., that volume, size, number and roundness of porous aluminum alloys decreasing with increasing of external load [230]. High-pressure cold spray system allows easier controlling of the deposits quality by reducing porosity and improving the mechanical properties due to higher gas velocity and particle impact [4,15,249,258,274,275,281]. It shows higher deformation rates of particles and presents ultrafine grains (UFG) structures in the area of particle boundaries. This structure can improve mechanical properties of the coatings, e.g. hardness, strength and ductility according to feed-stock [11,258,281,303-309].

However, while the application of a more cost-effective method low-pressure cold spray to achieve high loading by peening effect is a more difficult task due to less velocity of particles in the working jet. Anyway, good quality of Al coatings formed with low-pressure methods is mostly formed with the use of He as propellant gas. It can allow to reach high velocity at a lower temperature due to its lower density. Nevertheless, helium is quite expensive and dangerous in the use of the gas. Optimization of cost-effective, portable and easy-to-use low-pressure cold spray method with the application of compressed air as working gas represents interest in the present work.

In the last decades, researches have reported improvement of mechanical properties of cold spray coatings reinforced by ceramic particle inclusion [86,311,312,313]. Aluminum and its alloys serve well as matrix materials due to high strength to weight ratio, low density, good corrosion properties and low cost [314]. To improve ductility during coating formation, mechanical strength, hardness, stiffness ceramic particles like WC [315], TiB2 [316], TiN [317], could be embedded in the matrix. The most popular among ceramic fillers for the aluminum matrix is SiC. The second commonly used ceramic reinforcement powder is Al2O3, it is more inert and stable than SiC and shows better corrosion resistance and high-temperature resistance [318- 319].

To improve coating bonding strengths, reduce porosity [315] and increase deposition efficiency Al matrix powder can be blended with reinforcing particles in different proportions [235,320]. According to Irissou et al. [244] among studied coatings obtained from pure Al and mixture of Al with 7, 10, 20, 30, 50 and 75% Al2O3, the coating formed from 70%Al-30%Al2O3 feedstock mixture showed the best tribological and anticorrosion properties.

The same results were obtained by Wang et al. [33], with an increase of SiC content in the Al5056 matrix from 15% to 30%, porosity showed a decrease from 0.93 % to 0.61 %. Nevertheless, 60% of ceramic content showed no more decrease in porosity, but a small increase to 85%. From the other side, addition of reinforcement over 30% did not show an increase in adhesion, as ceramic cannot adhere to ceramic, but it shows a continuous increase with the addition of hard particles from 0% to 30% [121,244,320]. Thus DE is maximum at 30 % and there is no improvement of coating quality over this value.

102

Other authors have also observed improvement of corrosion resistance of Al at the same spraying parameters by the addition of Al2O3 particles due to coating densification [243-246,248]. The addition of ceramic particles has to be considered together with spraying parameters. At constant spraying parameters, the influence of ceramic particle addition has shown improvement, the maximum efficiency was reached at 30%.

Reinforcing by hard particles helps to activate the pre-sprayed surface removing oxide films [29]. It improves adhesion bonding since the residual oxide debris, which naturally exists on aluminum particle/particle and particle/substrate interfaces undermines the coating bonding performance [321]. Hard particles create additional roughness to the substrate and previously deposited layer of particles. It increases the area of contact between surfaces and consequently leads to a higher bonding strength [244]. In the work of Bu H et al [248]. Investigation of Al + Mg17Al12 mixture deposition on the AZ91D substrate showed deformation of the substrate by hard particles and traces of semi-circle indentation on the surface due to peening effect. Thus, interfacial adhesion between substrate and coating was improved.

The improvement of bonding strength was demonstrated by Bu H et al [248], while adhesion test, the failure took place within the Al+Mg17Al12 coating, but not at the coating/substrate interface. It indicates higher adhesive bonding of interface than cohesive bonding in the coating. They also have observed adhesion strength of pure Al coating as 7.2 ±0.4 MPa, when the strength of composite coating with initial blend Al+50% Mg17Al12 was determined to be 32±0.6 MPa.

Another solution for coating densification could be heat post-treatment of the coating with the use of laser. According to the works of Z. Jing et al. [322], V. Cannillo [323], N. Kang et al. [324], T. Marrocco et al. [325], A. Sova et al. [326] after laser post-treatment with melting effect some defects in the coating like porosity, micro cracks, loosing of particles have been eliminated or reduced. The microstructures have shown refinement, the crystal boundaries of feed-stock particles disappeared and bonding strength has been improved, Fig. 1.

While melting, the interfacial porosity was reduced due to bubbles migration to the free surface and following recrystallization of the microstructure [325]. The porosity of laser- treated cold spray coating can be reduced to a value of less than 1% for different materials and process parameters. At the same time the coating/substrate interface could be not influenced by melting [326].

On another hand, the roughness of the as-sprayed coating dramatically decreases after the laser melting process, which is a crucial parameter for corrosion protection [327].

103

As reported by n. Kang et al. [324] the parameter Ra of Al-Si cold sprayed coatings has reduced from 12.2 µm to 5 µm after laser melting, Fig 2.

Figure 1, Backscattered electron (BSE) image of cold spray titanium coating with a laser-melted surface layer [325]

Figure 2, 3D images of the surface of (a) as-sprayed Al-6 vol.% Si deposit and (b, c, d) laser re- melted samples at a laser power of respectively 200W, 250Wand 300 W [324].

Laser heat treatment shows a strengthening of the coating by the change of the bonding mechanism from mechanical interlocking to metallurgical bonding. This microstructural sintering in particles boundaries have made more homogenous coatings and improved its mechanical properties like ductility and strength [263-265,273,328- 330].

With the decreasing of heterogeneities in the coatings, the corrosion resistance improves. The reasons are stronger barrier effect, longer corrosion activation time on the less rough surface and longer solution propagation time through the more homogenous

104

However, during laser melting, a diffusion of oxygen into the matrix was revealed. Traces of oxygen present at molten particle boundaries and the entire surface of the coating. The pores also can represent an evolution of trapped gas. The number of small pores increases with the increase of laser power. From another hand metals with low melting point could be melted and boiled with the creation of bobbles, which forms pores while solidification [323-325]. Anyway, argon shielding atmosphere can prevent the oxidation of Al coating surface during laser treatment [331].

At fast cooling, formation of some cracks is possible [327], further solidification takes place with a new preferential orientation according to the heating direction. The microstructure shows a gradual effect with reducing heating influence deeper to the substrate Fig. 3. Equiaxed grains can be observed in the top of the re-melted surface and columnar in deeper layers [323-325].

Figure 3, Cross-sections of a) as-sprayed Al-6 vol.% Si and b), c), d) the same sample after laser surface re-melting at respectively 200 W, 250 Wand 300W and d), e), f) the corresponding microstructure of several areas (respectively top, middle, bottom) of the re- melted layer of sample c) [324].

105

Conclusion  The addition of hard particles to powder feedstock helps to close porosity in the coating and thus improves corrosion resistance preventing fast propagation of electrolyte. The cases of elimination of small homogenous pores were reported in the literature [230].

 Laser post-treatment is another way of coating densification and increasing corrosion protection properties. It reduces porosity, change microstructure and relax internal stresses.

 However, the proper mode of heat treatment is necessary to optimize as porosity evolution mechanism includes re-precipitation of hydrogen pores, oxidation, formation of vacant voids and reopening of closed pores [230].

In the literature, most experiments have been performed with the use of HPCS, expensive gases He / N2 or expensive high power laser equipment. The aim of the present work is to deposit an aluminum coating on carbon steel for sacrificial corrosion protection with the use of cost-effective equipment of LPCS coupled with laser and to improve the quality of obtained coatings by addition of ceramic particles or heat post-treatment.

5.2 Used materials

Carbon steel plates with composition of 0.08-0.13 % C, 0.15-0.35% Si, 0.30-0.60% Mn were used substrates. Coupons of 3 cm x 5 cm x 1 mm size were cut for cold spray performance. Necessary preliminar treatments were performed by sandblasting with Al2O3 (P100) particles and following cleaning in an ultrasonic bath with ethanol for 5 minutes.

Commercially available gas atomized aluminum powder Mi Fabrica (Japan) with 99,9% purity was used for pure Al coating formation and as a matrix for composition coatings. The SEM observation of Al powder made with Hitachi SU-70 Analytical Scanning Electron Microscope and EDS analyses performed with EDAX Genesis software is presented in Fig. 4. Al particles show irregularly or not perfect round shape.

Al2O3 powder was used as a reinforcing filler to the Al matrix in composition coatings, its SEM microphotograph is presented in Fig. 5. Crash-typ Al2O3 particles have typical sharp edges morphology. For composite coating formation the mixture of 70% Al and

30% Al2O3 were blended with machine Orie Shokai Co., Ltd. during 5 hours.

106

Laser diffractometer’s analyses made with software HORIBA LA-300 for Windows

(TM) Ver. 3.74 showed a quite narrow size distribution of Al and more spread for Al2O3 powder. Size distribution of Al particles has shown an average diameter of particles of 37 µm, as shown in Fig. 5. The mean diameter of alumina powder is 67 µm according to size distribution analyses, as shown in Fig. 6.

107

Figure 4, a, b) SEM images of Al powder c) SEM images of Al powder and EDS analyses (inset)

Figure 5. SEM image of Al2O3 powder

108

Figure 6, a) Size distribution of Al powder, b) Size distribution of Al2O3 powder measured with laser diffractometer

5.3 Experimental procedures

5.3.1 Parameters optimization

Optimizations of coating deposition were conducted by choosing and fixing one spraying parameter (temperature or pressure) and variation of another parameter. So, in the first test spraying of Al powder has been performed with fixed preheating temperature of 400 oC and 3 different pressures: 0.4 MPa, 0.5 MPa and 0.65 MPa. In the second test the pressure was fixed at 0.65 MPa and temperature was varying as 200 oC, 300 oC, 350 oC, 380 oC, 400 oC and 500 oC. It is worth mentioning that at a temperature of 500 oC clogging of nozzle occured, so that it was not possible to perform clear experiments even with the cooling system applied. A similar effect of a particle deposition at the inner surface of the nozzle and clogging at the narrow part was observed by other authors [248,332]. Two layers

109

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University of the coating were performed, the optimization parameters of Al deposition used for this work are gathered in Table 1.

Table 1, Spraying parameters of Al coatings

Deposition efficiency (DE) was used as a key parameter to optimize the coating. DE was counted by the mass difference between feedstock powder before spraying and deposited coating. Experiments showed that the highest deposition efficiency is achieved at pressure 0.65 MPa and temperature 400 oC as shown in Fig. 7-8. The highest pressure possible to achieve with LPCS and air is 0.65 MPa. As was mentioned above at 500 oC the powder stick at nozzle throat and disturb the accuracy of experiments, thus parameters 400 oC and 0.65 MPa are chosen for future experiments as the most efficient ones. The highest deposition efficiency of Al and 70%Al-30%Al2O3 mixture is 66% and 34% respectively, as shown in Fig. 9. In comparison, Shkodkin et al. have obtained 30%DE of Al with the addition of 30% Al2O3 particles to the initial blend.

Figure 7, Dependence of deposition efficiency of Al on gas pressure

110

Figure 8, Dependence of deposition efficiency of Al on gas temperature

Figure 9, Dependence of deposition efficiency of 70%Al-30%Al2O3 on gas temperature and pressure

111

Increasing the working gas pressure reports a higher velocity of particles in the jet. As feedstock contains different size and shape particles, the critical velocity necessary for successful deposition is different. Anyway, at higher gas velocity, more particles would be able to reach their critical velocity and adhere to the substrate and a previously deposited layer of coating [30,333,334]. Thereby, increasing of propellant gas pressure leads to an increase of the deposition efficiency.

Elevated temperature also increases gas and particle velocity of the jet. Moreover, heating energy transmitted to the powder leads to softening of the material [283,284]. High deformation rates while striking the substrate increase the deposition efficiency.

Feedstock mixture containing hard particles shows a lower deposition efficiency than pure aluminum powder, due to its high ductility. While deposition of hard particles on a soft substrate or deposited aluminum particle, they can be easily embedded in the surface. On another hand, the collision of two hard ceramic particles leads to bouncing back. Moreover, at high particle velocities, the reinforcement particles may lead to erosion and abrasion of the previously deposited layer [33,235,246,335,336].

The thickness of coatings also increases with increasing pressure and temperature of working gas, is about 400 µm for two layers coating.

Before coating deposition, the carbon steel substrate has been sandblasted with

Al2O3 (P100) particles. Substrate preparation is a crucial operation for stronger adhesion of the coating, higher surface irregularity increases the contact area of the substrate and coating, providing a stronger bending. The addition of ceramic particles creates asperities. They leave craters or protrude from the surface if deposited. In this case following soft aluminum particle strikes the hard surface of the alumina particle and it is exposed to a higher deformation rate. Thereby, the critical velocity of Al particle decreases while collision with the hard surface of alumina and allows deposition at lower temperature and pressure [235], [244].

Particle size and shape also play an important role in the deposition efficiency. One aspect is that, the particle needs to obtain enough energy to pass the shock-wave and adhere to the surface. Thus, too small particles may have not enough inertness and big particles could be too slow. On another hand, there is a critical size of the particle, below which it is not possible to obtain adiabatic shear instability and as a consequence high adhesion and deformation rate due to high thermal diffusion. This critical particle size is different for different materials and composition of the initial blend [15,33,320].

Most of the particles have obtained impact velocity enough to adhere, but not enough energy to densify the coating. However, impact energy shows a low abrasion effect

112

Conclusion:  High deposition efficiency obtained in this work could be a result of a combination of moderate particle size of aluminum (37.4 µm), which has high adhesion and low abrasion; the big size of alumina promotes higher densification (67.4 µm), which can leave deep craters and more additional roughness for high adhesion;  High impact velocities of particles give dense coatings according to the literature, but deposition efficiency is reduced. The reason for this effect might be the abrasion of the deposited coating by ceramic particles or big size aluminum particles. Using air as a working gas, which reduces jet velocity and allows us to keep high adhesion, but low deformation rate of particles.

5.3.2 Laser treatment

Samples with the best deposition efficiency obtained while spraying at 400 oC and 0.65MPa were laser treated with CW fiber laser, with spot diameter 1.6 mm. Maximum laser power 81 W and scanning speed 0.05 mm/s wavelength 940 nm±20 nm. The laser beam 60Hz pulse repetition with Gaussian energy distribution. The head of the laser is attached to a moving system with remote control, which allows laser scanning with defined parameters. The process is being performed in a transparent cabin, in Fig. 10.

Figure 10, Dymet system equipped with a laser movement system with remote control and transparent cabine.

Laser post-treatment with 30 A (40W) 50 A (81W) currents was done as an alternative way for closing porosity in the coating by the melting process. However, the

113

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University melting did not occur due to high heat conductivity of Al and low laser power 81 W, on the other hand, defects like pores and cracks can complicate the spreading of heat into the coating [337]. But the coating underwent heat treatment and annealing effect. A similar experiment of laser annealing of Al-Ni coating on Al substrate has been performed by Podrabinnik et Shishkovsky, [337]. With the use of laser power in the range of 10-100 W they did not reach melting, but heat treatment intensified intermetallic phase formation and reduced inhomogeneity and defects in the coating.

Parameters of cold spray deposition of Al powder, 70%Al-30%Al2O3 mixture and laser post-treatment are gathered in Table 2.

Table 2, Parameters of pure Al deposition, composite coating, and laser post-treatment

5.4 Characterization

5.4.1 Microstructure characterization

SEM photos of the top surface were taken for pure Al coating, 70%Al-30%Al2O3 coating and laser-treated aluminum coating, as shown in Figs. 11-13. All of the coatings show non-homogenous surface with cavities and peaks, which is typical for cold spray coatings. One can notice, that particles on the surface are not highly deformed and their initial shape could be recognized. This effect happened due to the absence of the following particles, which normally promote deformation by peening effect. In composite coating Al2O3 particles are visible, as shown in Fig. 12. As ceramic particles have very low ductility, they adhere by embedding into a soft aluminum layer of the previously deposited coating. From another side, due to this feature, hard particles generate a higher deformation rate for already deposited aluminum particles, than bonding Al particles [239,246]. The top surface of aluminum coating after laser heat treatment does not show melting smooth layer, coating surface same heterogamous as no- treated coatings, as shown in Fig. 13.

114

o Figure 11, SEM microphotograph of the top surface of Al coating, 400 C /0.65 MPa

Figure 12, SEM microphotograph of the top surface of 70%Al-30%Al2O3 coating,

400 oC /0.65 MPa

115

o Figure 13, SEM microphotograph of the top surface of Al coating, 400 C /0.65 MPa and laser treatment at 50 A (81W) current

Cross-sections of coatings were observed with Optical Microscope, Scanning Electron Microscope (SEM SU-70) and EDAX Genesis software. Ion Beam polishing was used instead of common abrasive polishing to avoid closing of porous while polishing due to high ductility of aluminum. For polish with Cross-section polishe machine IB-09020CP all coated samples were cut for pieces with dimensions 11x6 mm and about 2 cm thickness. On the figures 14 and 15 SEM microphotographs of Al coating cross-section with the most effective parameters 400 oC and pressure 0.65 MPa are presented. The structure of coatings is porous, the rate of powder deformation while spraying is not very high so that we can observe a powder nature of the coating, some grains keeping shape close to feedstock powder, as shown in Fig. 15.

As the deformation force of aluminum particles was insufficient to close porosity, hard reinforcement 30%Al2O3 was added to the Al matrix. The cross-section of composite coating shows the denser structure and higher rate of powder deformation, as shown in Fig. 16. Anyway, the coating remains a high porosity rate due to not enough velocity of the particles in the working gas.

116

Figure 14, SEM microphotograph of Al coating cross-section, 400 oC /0,65 MPa

Figure 15, SEM microphotograph of Al coating cross-section, 400 oC /0,65 MPa, grains shape

117

o Figure 16, SEM microphotograph of 70%Al-30%Al2O3 coating cross-section, 400 C /0.65 MPa

Aluminum coatings deposited at 400 oC and 0.65 MPa and subject to laser post- treatment were cut and polished with ion-beam similarly. Cross-sectional photos of samples treated with a laser at 30 A and 50 A are shown in Figs. 14 and 15, respectively. No trace of melting on the coating surface could be visible at both 30A and 50A currents.

The image analyses show high adhesion of aluminum coatings on carbon steel substrates, as shown in Fig. 17, 18. The substrate has irregularity due to the previous sandblasting, no delamination or cracks could be observed.

Figure 17, SEM microphotograph of Al coating cross-section, 400 oC /0.65 MPa with laser treatment at 30A (40W)

118

Figure 18, SEM microphotograph of Al coating cross-section, 400 oC /0.65 MPa with laser treatment at 50A

In Fig. 17 it is visible, that the top layer shows less compaction in comparison to coating substrate interface layers. This phenomenon occurs due to the continuous loading of depositing particles on the already formed coating layer. The same results have been obtained in the literature [338,339].

5.4.2 Porosity measurement

The amount of porosity was calculated by the ratio of the area of porosity and the remaining area of the material with the use of public domain software ImageJ. The measurements were performed in 5 pictures and the average area percentage is presented in Table 3. It is necessary to avoid a highly porous top layer while counting average porosity, as a part of that pores would be closed by the peening effect of following particles in case of second layer deposition [111]. The average level of porosity for Al coatings is 8%, and 6.4% for a mixture of Al and 30% Al2O3 composite coatings. Different results of porosity were reported in the literature. In some cases, it was lower than 1%. In another more than 6% depending on used parameters [248,340]. Anyway, less porosity was observed at higher gas speed with the help of high temperature/pressure or with the use of He as propellant gas. Addition of hard particles always leads to considerable porosity decrease. In some cases to 0.5% [246,248,341]. A. Sova et al. [326] have demonstrated, that porosity of 316 L coatings takes place mostly at the particle boundaries and depends on feed-stock particle size distribution. For powder about 18 µm size the porosity value was 3% and for the powder of 36 µm, it took 8%. At present work, the mean Al particle size is 37 µm and the obtained porosity level is also high.

119

High values of measured porosity at present experiments can be obtained due to not only low working temperature but also due to high accuracy obtained with gentle ion beam polishing.

Another aspect is a possible lack of energy due to high heat conductivity, which turns the deformation process to not adiabatic process. This effect sharply reduces the deformation rate of particles. It could take place because of relatively small particle size or strong shock-wave, at which particles can lose a part of kinetic energy. The spraying distance is mostly affected by the density of shock-wave. However, the presence of non deposited particles, which have no energy to leave the shock-wave also can disturb the deposition of powder and reduce particles energy. Thus, particles show not high deformation rates and subsequently high porosity level. [15, 4.6244,320].

Besides, the low velocity of air used as the working and powder feeding gas is a reason for high porosity in the coating. Application of helium or nitrogen can improve the quality of the coatings while using the same feed-stock, spraying conditions and equipment.

Table 3, Porosity of aluminum coatings

5.4.3 Roughness

All the coatings have a high rate of roughness/waviness made by powder spraying, high roughness leads to pitting corrosion due to high-stress concentrations at peaks.

A laser profilometer Surr-order SE300 was used to investigate the as-received surface of coatings and modification induced by laser treatment. The analyses of roughness were performed on 250X250 µm area with a scanning rate of 0.2 mm/s. Parameters: arithmetic mean deviation of the surface Ra and the mean deviation between the five highest peaks and the five deepest holes Rz were chosen for surface characterization.

The roughness of all coatings is very high, it is created by different diameters of non deformed top layer particles [294]. It could be also called waviness due to high amplitude, measured profiles of Al, composite and laser-treated coatings are presented in Fig. 19. The obtained values of Ra and Rz of all the coatings are presented in Table 4. All the coatings show similar roughness, no smooth melted surface was observed for laser treatment at 50A.

120

Figure 19, Roughness profile: a) aluminum coating, b) composite coating, c) aluminum coating with laser treatment at 50 A

121

Table 4, Roughness of aluminum coatings

The roughness of the metallic coatings greatly affects the corrosion potential. It is making easier formation of corrosion pits and further corrosion paths for solution propagation [342]. Investigation has shown, that smoother surface demonstrates higher corrosion potential than a rough surface. And increasing roughness increase pitting corrosion [343-349]. Pitting initiation includes two processes. The first is the formation of pits preferentially on defects, which are more numerous on rough surfaces. And the propagation of corrosion takes place along grain boundaries.

When aggressive species from the solution penetrate to the bottom of deep grooves of the surface, they are trapped in the cavities and cannot diffuse out to the bulk solution. So, pits continue to growth and solution penetrates deeper. From another side, ions necessary for metal passivation are hard to diffuse to the grooves to stop corrosion pits growth [346, 350,351].

At lower roughness level, a diffusion of corrosion products is less easier to occur and no accumulation of aggressive species happens. Thus, the nucleation of pits and their growth is stopped by re-passivation. A stable oxide film grows and reduces corrosion due to its higher corrosion potential than active surface [346,350-352].

5.4.4 Microhardness

Vicker’s microhardness test was performed with HMV-2, Shimadzu equipment on the cold sprayed aluminum coatings in as-deposited and heat-treated conditions. The Applied force of the indenter was 98.07mN (HV0.1) for 10 sec. Each experiment was performed 10 times and average values are presented in Fig. 20. Results showed an increase in microhardness of coatings of 5% whith the addition of hard particles. From another side, degreasing of microhardness after post heat treatment takes place by 39% for pure aluminum coating.

Increase in microhardness with the addition of reinforcing particles to cold spray aluminum coatings was also observed by other authors, for example, Bu H et al. [248] have demonstrated an increase from 47±5HV0.1 to 58±3HV0.1 with the addition of Mg17Al12 and

[246] obtained an increase from 51±3HV0.1 to 65±5HV0.1 by adding Al2O3 particles.

122

In contrast, a decrease in hardness usually happens at the annealing process due to releasing of internal stresses and possible recrystallization with the subsequent grain growth [353]. Ductility of cold spray coatings highly increase with loosing in strength at high temperature heat treatment like annealing and solutionizing [263-265,328-330,353]. Usually, the process is accompanied by grain growth, but it could be stabilized by ultra-fine grain structures at grain boundaries [305,330,354]. At high particle impact velocities more dislocations are formed, which leads to nucleation of new grains at recrystallization. Thus, at higher temperature fine grain microstructure forms. Inter-particle sintering takes place during annealing process indicated by the growth of ductility. Dislocations and other defects migrate from the particle-particle interface and adhesion mechanism changes from mechanical interlocking to metallurgical bonding [273].

Thus, a decrease of microhardness is evidence for laser annealing due to internal stress release and annihilation of dislocations [245,273]. Annealing can homogenize the coating structure and reduce the number of defects, crack nucleation sits and internal stresses [273].

On another hand, a small increase in hardness of composite coating can be higher due to internal stress concentration brought by higher peening load of hard particles and subsequently higher deformation rate of Al particles. On another hand, while striking of hard particles the cold working effect increasing dislocation density takes place in composite coatings [121,355].

Figure 20, Diagram of microhardness of aluminum coatings

123

Conclusion: It is difficult to design an optimal heat treatment for all similar coatings due to nonhomogeneous microstructure of coatings, which depends on deformation rates and many spraying parameters [15,258,274]. Anyway, adjusted heat treatment can help to homogenize the microstructure of the cold sprayed coating and generate appropriate microstructural features [273]. According to Rokni M et al. annealing leads to fusion of grain boundaries and a metallurgical bonding increase [273]. The same effect was noticed for other cold spray materials [263-265,328-330]. They have also reported, that additional sintering while annealing reduces crack nucleation sites [349].

5.5 Corrosion performance

Aluminum (Al) is a widely used material for corrosion protection due to its high anticorrosion properties. It serves as a sacrificial coating for the more noble substrate metal. The high anti-corrosion properties are mainly attributed to the ability of forming a thin aluminum oxide layer while exposed to air, which prevents further oxidation [127,356].

Common steel usually becomes active in an aggressive environment: acid, alkali, oxygen, water and on the surface of steel native iron oxides growth due to thermodynamic reasons [111].

There are two main types of localized corrosion in aluminum alloys, the first is pitting and the second is intergranular corrosion. Pitting corrosion happens as a result of passive film breaking around inclusions, usually in a chloride-rich environment [357], [358]. Also in composite coatings pits starts around Al2O3 particles, corrosion happens by the formation of local cells between pits and more active Al matrix [111]. To improve the corrosion resistance of aluminum, a heat treatment or addition of alloying elements is used [359,360].

Investigations of different researches have shown that there is chemical heterogeneity along grain boundaries, that create micro coupling and lead to intergranular corrosion [359-365]. Last researches have shown that intergranular attack can happen without chemical heterogeneity [366]. It can be a piece of evidence, that instead of chemical heterogeneity, structural heterogeneity influences the intergranular corrosion [145,368-372]. Intergranular corrosion is influenced by energy stored in the grains as a result of dislocation accumulation [366,367,370,373-375]. Immersion experiments into Na-Cl solution have shown that intergranular corrosion primarily occurs in the periphery of grains with higher stored energy and then on the boundaries of grains with lower energy. Both chemical and structural heterogeneities of grain boundaries can influence the intergranular corrosion in aluminum alloys [376]. The distribution of dislocations is non uniform, it can form galvanic coupling between grains with different stored energy which promotes the development of boundary corrosion [376].

124

On another hand certain morphology of plastically deformed materials represents structural heterogeneities. Experiments showed that corrosion is localized along with the shear directions. The dissolution is restricted by the shear band area and several pits have appeared along with the bands not far from each other. It shows that the most susceptible places for pit initiation are structural heterogeneities like twins, texture. Small pits may coalesce to larger pits and easily propagates.

Laser treatment is expected to improve the corrosion behavior of aluminum coatings. Rodopoulos et al. have reported an anodic shift of the pitting potential of aluminum alloys after laser treatment [132].

Corrosion performance of aluminum coatings was investigated with standard 3 electrodes electrochemical cell connected to a SP-300 potentiostat (bio-logic company). Ag/AgCl, KCl (sat-d) electrode was used as a reference, graphite rode served as a counter electrode and the coating surface was the working electrode. All the electrochemical tests were performed in an aerated electrolyte 0,1M Na2SO4 pH 6.5.

Open circuit potential of pure Al coating, composite coating and Al coating after laser treatment at 50 A (81W) were recorded for 1600 seconds. All the curves show a continued drop of potential which later stabilizes at about -1.25 V at about 900 seconds, Fig. 21.

Figure 21, – Open circuit potential (OCP) measurements in 0.1 M Na2SO4 for 3 hours of cold

sprayed coatings from pure Al, Al and 30%Al2O3 and Al with laser post-treatment

Linear Sweep Voltammetry tests were performed from -1.5 V to 0.23 V with a scan rate of 1 mV/s in 0.1M Na2SO4 solution with Ag/AgCl, KCl (sat-d) reference electrode and graphite counter electrode.

125

Polarization curves of aluminum samples show similar corrosion potentials at -1.2 V. Cathodic behavior is not influenced by the surface whereas anodic branches show different trends. For reference aluminum coating, the electric current increases after -1.0 V. Above this potential, it is assumed that localized corrosion takes place as intergranular impurities or heterogeneities, which serve as an anodic sits and preferentially dissolve. H. Amar et al. [378] have performed large scale and local electrochemical tests of aluminum coating. They have found that at large scale, test pitting potential was fixed by the corrosion behavior of the most active particles and at a local test of single-particle the pitting potential was taken from the weakest place of the particle. It has been reported by authors [379,380] that the number of stages of the polarization curve depends on the detected surface area [378]. For treated aluminum coatings, the current densities remain lower than for the reference pure Al coating in the anodic polarization domain.

1 ) 2

0.1

0.01

0.001

1E-4 70Al30Al2O3 1E-5 Al+laser 50A

Current density log(I), (mA/cm log(I), density Current Al coating 1E-6

1E-7 -1.6 -1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 Ewe vs Ag/AgCl

Figure 22, Linear sweep voltammetry of aluminum coatings

Cold sprayed Aluminum coating after laser treatment shows the same stages of localized corrosion and intergranular corrosion. But the corrosion current is lower, as shown in Fig. 22. The reason has to be less internal energy of dislocation clusters as laser post- treatment influences the relaxation of internal stress in the coating by annealing effect [378]. Heating leads to easier dislocation movement, they annihilation and recrystallization of grains with releasing of energy. Less internal stresses have lower potential for pitting

126

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University initiation and corrosion propagation. On another hand, sintering between particles can take place, which reduces the area of heterogeneities. Thus, laser annealing has a beneficial effect on the electrochemical behavior and corrosion performance of an aluminum coating [378]. Composite coating has shown the same tendency and the curve is very similar to laser-treated coating. The effect of improvement in corrosion resistance of the composite coating is achieved by coating densification, which makes slower propagation of the solution.

Electrochemical Impedance Spectroscopy tests (EIS) were performed at the potentiostat mode at the OCP with a sinused amplitude of 10 mV and frequency range between 1000 kHz and 10 MHz.

Figure 23 shows the electrochemical impedance measurements plotted in Nyquist diagrams for the reference Al coating. The reinforced coating (70Al30Al2O3) and the laser- treated surface (Al+laser 50A). These data were recorded during 1600 seconds at the OCP. The analyses were performed on two different samples and the results are consistent between them. This also demonstrates a good reproducibility on the coating/processes. Electrical equivalent circuits EEC are usually applied to EIS data to quantify the electrochemical behavior of the material. Before that, the size of the impedance loop is the first parameter to consider. In Figure 23, reinforced coating tends to form a higher semicircle than pure aluminum and laser-treated coatings, and shows higher corrosion resistance [384]. Composite particles replace some of the aluminum and decrease the active area of corrosion [111]. Accumulation of corrosion products around Al2O3 particles can slow the corrosion process.

Typical approximately straight line of plots in the low-frequency range shows a presence of mass-transport step of corrosion process [381]. This step usually involves diffusion of reactants, like oxygen from the solution and defects with narrow and deep geometry [382]. Furthermore, the limited mass-transport of oxygen could also be influenced by the porous layer of deposited corrosion products [382,383].

Two times constants can be observed for each sample, one at high frequency and another at low-middle frequency. Therefore a classical EEC model with two-time constants was selected: R1+C2/R2+C3/R3 corresponds to the electrolyte resistance and does not count in the total polarization resistance. Graphically, the time constant at high frequency has a slope at approximatively 45°-55° whatever the sample. This angle usually means that the reactivity of porosities or surface defects becomes predominant [385-387]. In a simple approach, this can be modelized by a phase constant element in parallel to a resistance (CPE//R, i.e. Q1/R2) [387]. It has to be noticed that active area might be higher than the apparent surface in one hand and on another hand, diffusion of reactants into pores or grain boundaries was microscopically proved [388]. To go further in the analysis, a detailed description of the surface morphology including the porosity geometry, number, etc. should be carried out to extract information such as the ionic charge transfer resistance on the

127

Cette thèse est accessible à l'adresse : http://theses.insa-lyon.fr/publication/2020LYSEI054/these.pdf © [E. Lapushkina], [2020], INSA Lyon, tous droits réservés Elizaveta Lapushkina/ Engineering materials / INSA de LYON&Tohoky University pores wall, the pore wall capacitance, etc [389,390]. The second time constant can be modelized by a CPE // R (i.e., Q3/R3) which informs about the reactivity of the surface of the material. [391,392]. Table 5 reports the EEC values. Even if the results of the total resistance

(R3 + R2) are in the same order of magnitude, Table 5 shows that Al2O3 incorporation slightly decrease the corrosion resistance of the coating compared to the reference (i.e., Al). Even the laser treatment was not able to melt the surface to close porosities, the thermo- mechanical shock improves corrosion resistance of the surface.

Table 5, A classical EEC model data Sample R1 Q1 N1 R2 Q3 N3 R3 Ohm.cm² F.s^(a- Ohm.cm² F.s^(a- Ohm.cm² 1)/cm² 1)/cm² Al coating 105 0.48 e-3 0.599 1 366 0.22 e-3 0.95 33 664

Al+Al2O3 47 1.16 e-3 0.612 2 520 1.16 e-3 0.96 14 256 Laser 125 0.48 e-3 0.623 952 0.33 e-3 0.12 54 464

a b

Figure 23, a) EIS plots of aluminum coatings and b) application of EEC on EIS data.

128

Summary

 In this chapter, the formation of aluminum coatings on carbon steel substrate with the use of the Low-Pressure Cold Spray method was considered. Optimization of parameters was performed according to deposition efficiency and spraying parameters 400 oC /0.65 MPa were found to provide the best result  Investigation of microstructure have shown high porosity ratio and two densification

methods were applied: addition of ceramic particles 30% Al2O3 to the Al feedstock and laser post-treatment. Ceramic particles have densified the coating and allowed to reduce porosity from 8% to 6.4%. Laser treatment did not perform coating melting, however, it released coating microhardness due to relaxation of residual stresses and possible change in microstructure;  Corrosion performance of pure aluminum, composite and laser-treated coatings was investigated. Laser treated coating showed higher corrosion resistance, in comparison with pure aluminum coating due to the relaxation of stresses. From another hand, composite coatings have shown a higher corrosion resistance than both pure aluminum and laser-treated coatings. Closing of porosity reduces the rate of solution propagation from one hand and reduces the coating area exposed to corrosion from another.

It could be concluded, that the low-pressure cold spray method allowed to form the aluminum coating. In the case of corrosion application, the coating has to be densified as porosity is a key factor for corrosion resistance. The addition of ceramic particles reduces coating porosity and improves corrosion resistance. Laser heat treatment also can improve corrosion resistance by releasing residual stresses in the coating.

There is a potential of LPCS technique combined with laser for cost-effective reparation, but it needs further improvement in coating densification. In the future work for laser melting, more powerful laser is needed. From another hand shielded atmosphere or vacuum can help to improve the coating microstructure.

Increasing laser power would lead to the melting of the surface layer reducing roughness and closing porosity in the coating body. Laser post-treatment strong enough for melting can increase corrosion performance of LPCS aluminum and composite coatings.

129

General Conclusions

This thesis, of the topic: cold spray coatings, were manufactured in collaboration between two academic laboratories that one is MATEIS in France and the other one is Ogawa-lab (FRI) in Japan. As presented in the introduction, many efforts have been devoted to develop this additive manufacturing and have harvested great promising results. For anticorrosion, cold spray allows us to deposit both noble materials or sacrificial metals according to different protection strategies replying to different industrial requirements. Within the context of this study, we focused on the sacrificial strategy based on:

- the pure Zn coating deposited by a high-pressure cold spray process in MATEIS; - the alumina filled aluminum coating prepared by the low-pressure process in Ogawa- lab.

At first, it was an opportunity to participate in the installation of such industry-level equipment in MATEIS. We attempted to understand the roles of different process parameters and achieved the first successful deposition of Zn coating on mild steel very quickly. It has been found that the deposition efficiency of Zn is significantly improved for a projection gas temperature getting as high as 200 °C under a pressure of 2 MPa. It means that under these conditions, the particle velocity upon impact becomes higher than the critical value, Vcrit, leading a high deposition efficiency. The further depositions were carried on, keeping the parameters higher than these onset values.

A design of experiments with Doehlert uniform shell was applied to evaluate the influence of the gas temperature and pressure on the coating microstructure as well as the anti-corrosion performance. For this purpose, seven temperature-pressure coupling values were chosen that provides the average particle velocity recovering the whole parameter window for cold spray deposition (from Vcrit to Verr). The as-prepared coatings presented satisfactory adhesion since no significant cracks and delamination can be observed from the cross-section. The three-point bending tests also proved their good coating/ substrate adhesion.

Different temperature-pressure couplings derived coatings displayed two protection modes: the sacrificial mode for less dense coating sprayed under 290 °C and 3.0 MPa, and corrosion product protection for homogeneous and dense coatings sprayed under 320 °C and 2.5 MPa. In the first case, the presence of hetérogéities, including pores, grain boundaries, phase transition and residual stresses, are considered as the active corrosion points. The electrolyte penetration took place so that the electrolyte was in contact simultaneously with the coating and substrate. The protection is attributed to a cathodic protection mechanism. On the other hand, for the dense coating, the direct contact of the underlying substrate with the surrounding electrolyte is prevented by the coating.

In a microscopy scale, the surface morphology is the most critical factor influencing the coating electrochemical behavior. It was intended to evaluate the effect of the interparticle interphase using the 3-electrode micro-cell. With the microcapillary, an analysis area with a diameter of 60 µm could be selected. The LSV results were recorded with the polished and etched coating surface of the sample prepared under 260 oC /2.5 MPa. The

130

On the other hand, the formation of aluminum coatings on carbon steel substrate with the low-Pressure Cold Spray method was considered. Optimization of parameters was performed according to deposition efficiency and spraying parameters 400 °C /0.65 MPa have shown the most best results.

Without any improvement or post-treatment, the aluminum coating has been found porous. Therefore, to enhance its anti-corrosion performance, two densification methods were applied. The first one consists in the incorporation of 30 wt% Al2O3 ceramic particles to the Al feedstock. Ceramic particles have a densification effect in the coating and allowed to reduce the porosity from 8% to 6.4%. The second way consists in a post-treatment with a low energy laser to reclose the open pores of the coating surface. Indeed, the Laser treatment did not achieve coating melting, but released the coating microhardness due to the relaxation of residual stresses and possible change in microstructure. As the results show, the laser-treated coating showed higher corrosion resistance than pure aluminum coating due to the relaxation of stresses. From another hand, composite coatings have shown higher corrosion resistance than both pure aluminum and laser-treated coatings.

131

Future work and perspectives

Concerning the pure Zn coating, it is interesting to optimize further the coating system in terms of the different protection modes. It requires a long term corrosion analysis to evaluate the performance of high-density and low-density coatings.

The 3-electrode micro-cell is a powerful method to analyze the electrochemical behavior in a very local area with different microstructures. Deposition of thick coatings would allow cross-section analysis. Therefore, the electrochemical test in the deepness of the coating allows us to reveal the gradually evaluated properties as a function of residual stresses. Meanwhile, the investigation on the influence of different heterogeneity provides a complementary argument for the coating optimization.

There is a potential of LPCS technique combined with laser for cost-effective reparation, but it needs further improvement in coating densification. A more powerful laser could be used to achieve laser melting. From another hand shielded atmosphere or vacuum can help to improve the coating microstructure.

Increasing laser power would lead to the melting of the surface layer reducing roughness and closing porosity in the coating body. It should increase the anticorrosion performance of LPCS aluminum coating and composite coatings.

132

Appendix Critical velocity and window of spraying parameters

At cold spray process for successful bonding impact particle have to obtain localized deformation and adiabatic shear instabilities (ASI), which appears at specific critical velocity of the powder particles [26,126]. Not all the particles from the working flow necessary achieve a sufficient velocity for bonding. Only particles, whose impact velocities exceed the critical velocity.

Three velocities can be distinguished for Zn: the first for particle/substrate impact, it was reported 375-390 m/s [15,30]; the second one is related to the particle/particle cohesion, it gives the ability of thick coating formation; and the third can be attributed to erosion velocity of already formed coating. The rise of impact velocity increase the deposition efficiency and inter-splat bonding strength, as well as process temperature and pressure rise [30,333,334,393].

Material density, crystallographic structure (hexagonal close-packed in case of Zn), ductility play a role in particle adhesion and strength of bonding. Finally, high ductility of Zn allows formation of dense coating with use of lower gas temperatures [15,126,127]. It is also known from convention aerodynamics that the drag coefficient of irregular powder is higher than that of a spherical sphere powder, so they should have higher Vimpact [67].

Particles with low velocity activate the surface by removing the oxide layers, it helps to prepare the surface for following particles deposition and reach high bonding strength. The velocity of the particle could be described as a function of its size, thus the number of deposited particles is correlated to a particle size distribution of the feedstock for the chosen material. The critical velocity of the largest particle could be taken as a critical velocity of the powder. In general small particles could be accelerated more efficiently than large particles. But the particles of very small size could be decelerated by shock wave close to the substrate [15,126,127,394].

Shear instabilities could be indicated by numerical analyses and critical velocity could be calculated according to materials properties and spraying parameters. For example, Assadi et al. [26] have investigated the effect of material properties on the critical velocity at cold spray with the use of numerical simulation. To describe this effect they have proposed the following expression:

Equation 1:

−7 푣푐푟𝑖푡 = 667 − 0.014𝜌 + 0.008(푇푚 − 푇푅) + 10 𝜎푈 (1)

, the used parameters are defined in Table 1, copper served as a reference material.

133

Table 1 - Nomenclature of used parameters

Parameters Description cp Specific heat d, x Diameter, distance

F1, F2 Empirical factors

Ti Impact temperature

Tm Melting temperature

TR Reference temperature (293K) t Time v Velocity ρ Density

σU Yield stress

σTS Tensile strength ʎ Thermal conductivity

However, only small changes in parameters can be applied to this expression. As copper has been used as a reference, only materials with similar properties can be described with Equation 1. Besides, they did not consider the influence of particle size.

Schmidt et. al. [15] have studied the influence of particle properties on bonding conditions based on ASI and critical velocity. They have performed a numerical simulation and demonstrated a strong influence of heat conduction on bonding at cold spray, and described this effect analytically. Based on medialization and the row of experiments they also found that with increasing particle size the critical velocity decrease. A simple expression was proposed based on available material data. It can predict a critical velocities for different metallic materials. Impact dynamic effects were described with material strength and dynamic load, see Equation 2 and nomenclature of used parameters in Table 1. The temperature-dependent strength of material is presented by the Johnson-Cook equation for thermal softening and the tensile strength, on the left side of Equation 2 (red color). And the right side represents characterizes the dynamic load, by a ballistic expression for the crater ground pressure in hydrodynamic penetration.

134

푇푖−푇푅 1 2 퐹1 ∙ 𝜎푇푆 ∙ (1 − ) = ∙ ρ ∙ 푣푐푟𝑖푡 (2) 푇푚−푇푅 8

Introducing calibration factor F1 Equation 3 was obtained for description of mechanical balance :

푇푖−푇푅 퐹1∙8∙휎푇푆∙(1− ) 푣푚푒푐ℎ = √ 푇푚−푇푅 (3) ρ

The energy balance between thermal dissipation and provided kinetic energy can be described as Equation (4), with a correlation factor F2.

1 퐹 ∙ 𝜎 ∙ (푇 − 푇 ) = ∙ 푣2 (4) 2 푝 푚 𝑖 2 푐푟𝑖푡 The combination of Eq. 3 and Eq. 4 leads to Eq. 5 with applying of weight factor of 0.5 for both:

푇 −푇 퐹 ∙4∙휎 ∙(1− 푖 푅 ) 1 푇푆 푇 −푇 푣푡ℎ,푚푒푐ℎ = √ 푚 푅 + 퐹 ∙ 푐 ∙ (푇 − 푇 ) , [15] (5) 푐푟𝑖푡 ρ 2 푝 푚 𝑖

This equation provides a mechanical balance of an impact, a balance between thermal dissipation and provided kinetic energy. The powder purity, high-strain-rate and size effects are not included in these equations.

Based on the correlation of calculated and experimentally determined critical velocities the calibration factors F1 = 1.2 and F2 = 0.3 where proposed for critical velocity and F1 = 4.8 and F2 = 1.2. for velocity of erosion, which kinetic energy is too much to create strong bonding force exceeding rebound effect. The properties of bulk materials are used in the equation and they can be different for powders taking into account the microstructure. This experiment leads to a proposal of a parameter window for cold spray deposition, which has to cause the impact velocities between Vcrit and Verr. It allows predicting maximum coating qualities according to particle size and impact conditions

In present work with the help of our colleague Xinliang Xie from University of Technology of Belfort-Montbéliard the medialization with by use of fluid dynamics and heat transfer calculations were performed to predict particle impact velocities and temperatures for particle sizes 26 µm, 37 µm and 54 µm (Fig.1) for all 7 conditions of the matrix according to Doehlert uniform shell design. Next parameters were considered for modeling and the following assumptions have been made : melting temperature of Zn=420 oC, powder and

135

o substrate of room temperature 20 C, nitrogen as the working and powder feeding gas, Zn density 7.14 kg/m3, the spherical shape of particles.

The calculations also considered particle deceleration in the bow-shock at the substrate.

Figure 1, Particle size distribution of used Zn powder

Based on obtained data for Vimpact and Timpact and Equation 5 analyzation of critical and erosion velocities were performed and the spraying window between critical and erosion velocities was found.

From Figure 2 a,b one can notice that particles show higher impact velocities at lower powder size, which is caused by the lightweight. On another hand, with increasing particle size, its impact temperature increases. As bigger particles are exposed to lower cooling rate, they can keep heating energy longer, so, they have more time for diffusion and bonding at the same impact velocity [15,128]. Moreover, initial parameters of spraying (temperature and pressure) influence on impact velocities and temperature, but not depend on powder size. Thus, the highest impact velocities were detected at 290 oC /3 MPa for all three powder o sizes. And the lowest Vimpact at 230 C and 2 MPa in all three cases. For impact temperature the highest values are at 320 oC /2 MPa and the lowest at 200 oC/ 2.5 MPa for all the powders, which corresponds to two heating energy. In Figure 2b it is clear, that particle impact temperatures grow for all the particles with an increase of size by similar behavior.

136

180

160

140 o 200 C 2.5 MPa 230 oC 2 MPa

C) o o

( 230 C 3 MPa 120 260 oC 2.5 MPa o

impact impact 290 C 2 MPa

T 290 oC 3 MPa 100 320 oC 2.5 MPa

60 25 30 35 40 45 50 55 60 Particle size (µm)

b Figure 2, a) dependence of impact velocity and temperature on particle size, b) behavior of impact temperature change with particle size increase

137

Applying Equation 5 critical and erosion velocities were found for all three particle sizes. From Figure 3a it is obvious, that all the powder impact velocities have exceeded their critical velocities, which are also different for each spraying condition and particle size. It shows enough energy for ASI formation and high adhesion of the powder. In case of 54 µm particles, they erosion velocities always exceed Vimpact. It means, that there is no erosion of the coating by large particles at applied conditions. To analyses the possibility of the smaller particles to erode the coating, the difference

Vimpact – Vcrit/Ver were found. As can be seen in Figure 3b, all the values of Vimpact – Vcrit are positive, which shows adhesion of all the particles at all spraying parameters. It is interesting to note, that smaller particles need higher velocity for adhesion (Vcrit). This effect could be explained by the ease of cold working effect impeding local deformation and by different particle surface to volume ratio including more defects, like oxides and adsorbents [15].

In opposite, negative values of difference Vimpact – Ver will show the absence of erosion. However, in our case impact velocities of several parameters 320 oC/2.5 MPa and o 290 C/3 MPa for particles with 26 µm and 37 µm have exceed Ver, which should cause abrasion effect. That is in agreement with previously analyzed Doehlert uniform shell design, Fig. 4. In general there is a tendency of smaller particles to erode the coating at higher spraying temperatures due to the high impact velocities. As our powder is not sieved and all the particles of size distribution shown in Figure1 are mixed in spraying jet, we can conclude that at higher temperatures small particles faster achieve the substrate and give adhesion and some erosion at the same time. Moreover, with decreasing of spraying temperature the rate of erosion decreasing and at 260 oC and 2.5 MPa we obtained the thickest coating. However, at 320 oC and 2.5 MPa conditions particle impact temperature is high 153-174 oC , moreover particle should be already soft due to the close melting point of Zn (420 oC) it should lead to particle softening with decreasing erosion rate.Thus, we have a high thickness of the coating at these spraying parameters. On the other hand, for the high quality of the coating not only thickness, but also density should be considered. Figures 3 and 4b, show a good correlation of interfacial porosity between coatings and substrate and particle velocities. Thus, at 290 oC/3 MPa the highest erosion is indicated, which leads to low thickness and high porosity at the coating/substrate interface. From another side, parameters 260 oC/2.5 MPa and 320 oC/2.5 MPa have a good correlation between critical/erosion velocities, coatings thickness and interfacial porosity caused by abrasion of the surface with the not deposited particles. It should be taken into account, that 320 oC leads to increasing of particles softening due to material properties from one side and from another side the smallest particles could have a high gradient of temperature leading to the quenching by high cooling rate and following strain-rate hardening [15]. Consequently, these particles need a higher temperature for softening and successful adhesion, which is easier to perform at 320 oC.

138

b Figure 3, dependence of a) impact, critical and erosion velocities on powder size, b) the

difference of Vimpact – Vcrit (Ver) on powder size

139

Figure 4, Doehlert uniform shell design analyzation of correlation a) temperature, pressure and thickness, b) temperature, pressure and volume porosity

140

Summary:

 Based on Equation 5 from the work of Schmidt et. al., fluid dynamic and heat transfer modeling critical, erosion and impact velocities of powder can be found, as well as impact temperature and they adhesion properties could be analyzed;

 Technically the deposition window is presented by the gap of Vimpact between Vcrit and Ver, it gives more adhered particles and higher thickness of the coating. However, the microstructure has to be taken into account while choosing spraying parameters;

 Medialization of particle velocities correlates with previously Doehlert uniform shell design analyzation.

141

References

[1] R. C. Dykhuizen and M. F. Smith, ‘Gas dynamic principles of cold spray’, J Therm Spray Tech, vol. 7, no. 2, pp. 205–212, Jun. 1998, doi: 10.1361/105996398770350945.

[2] A. P. Alkhimov, A. N. Papyrin, V. F. Kosarev, N. I. Nesterovich, and M. M. Shushpanov, ‘Gas- dynamic spraying method for applying a coating’, 25 Feb. 1997, US5302414B1

[3] A. O. Tokarev, ‘Structure of aluminum powder coatings prepared by cold gasdynamic spraying’, Met Sci Heat Treat, vol. 38, no. 3, pp. 136–139, Mar. 1996, doi: 10.1007/BF01401446.

[4] E. Irissou, J.-G. Legoux, A. N. Ryabinin, B. Jodoin, and C. Moreau, ‘Review on Cold Spray Process and Technology: Part I—Intellectual Property’, J Therm Spray Tech, vol. 17, no. 4, pp. 495–516, Dec. 2008, doi: 10.1007/s11666-008-9203-3.

[5] R. Lima, J. Karthikeyan, C. M. Kay, J. Lindemann, and C. Berndt, ‘Microstructural characteristics of cold-sprayed nanostructured WC-Co coatings’, Thin Solid Films, vol. 416, pp. 129–135, Jan. 2002, doi: 10.1016/S0040-6090(02)00631-4.

[6] H.-J. Kim, C.-H. Lee, and S.-Y. Hwang, ‘Superhard nano WC–12%Co coating by cold spray deposition’, Materials Science and Engineering: A, vol. 391, pp. 243–248, Jan. 2005, doi: 10.1016/j.msea.2004.08.082.

[7] L. Ajdelsztajn, J. M. Schoenung, B. Jodoin, and G. E. Kim, ‘Cold spray deposition of nanocrystalline aluminum alloys’, Metall and Mat Trans A, vol. 36, no. 3, pp. 657–666, Mar. 2005, doi: 10.1007/s11661-005-0182-4.

[8] S. Yoon, H. Kim, and C. Lee, ‘Deposition behavior of bulk amorphous NiTiZrSiSn according to the kinetic and thermal energy levels in the kinetic spraying process’, Surface and Coatings Technology, vol. 200, pp. 6022–6029, May 2006, doi: 10.1016/j.surfcoat.2005.09.022.

[9] N. B. Maledi, O. P. Oladijo, I. Botef, T. P. Ntsoane, A. Madiseng, and L. Moloisane, ‘Influence of cold spray parameters on the microstructures and residual stress of Zn coatings sprayed on mild steel’, Surface and Coatings Technology, vol. 318, pp. 106–113, May 2017, doi: 10.1016/j.surfcoat.2017.03.062.

[10] L. Ajdelsztajn, B. Jodoin, and J. M. Schoenung, ‘Synthesis and mechanical properties of nanocrystalline Ni coatings produced by cold gas dynamic spraying’, Surface and Coatings Technology, vol. 201, no. 3, pp. 1166–1172, Oct. 2006, doi: 10.1016/j.surfcoat.2006.01.037.

[11] S. H. Zahiri, D. Fraser, and M. Jahedi, ‘Recrystallization of Cold Spray-Fabricated CP Titanium Structures’, J Therm Spray Tech, vol. 18, no. 1, pp. 16–22, Mar. 2009, doi: 10.1007/s11666-008- 9212-2.

[12] A. Malachowska, M. Winnicki, L. Pawlowski, B. Pateyron, A. Ambroziak, and P. Sokołowski, ‘Oxidation of Titanium Particles during Cold Gas Dynamic Spraying’, COMSOL Conference in Rotterdam, 2013.

[13] K. H. W. Seah, R. Thampuran, and S. H. Teoh, ‘The influence of pore morphology on corrosion’, Corrosion Science, vol. 40, no. 4–5, pp. 547–556, Apr. 1998, doi: 10.1016/S0010-938X(97)00152- 2.

142

[14] R. Maev, V. Leshchynsky, and E. Strumban, ‘Portable, low pressure cold spray systems for industrial applications’, in The Cold Spray Materials Deposition Process, V. K. Champagne, Ed. Woodhead Publishing, 2007, pp. 217–231, DOI: 10.1533/9781845693787.2.217

[15] T. Schmidt, F. Gärtner, H. Assadi, and H. Kreye, ‘Development of a generalized parameter window for cold spray deposition’, Acta Materialia, vol. 54, no. 3, pp. 729–742, Feb. 2006, doi: 10.1016/j.actamat.2005.10.005.

[16] N. M. Chavan, B. Kiran, A. Jyothirmayi, P. S. Phani, and G. Sundararajan, ‘The Corrosion Behavior of Cold Sprayed Zinc Coatings on Mild Steel Substrate’, J Therm Spray Tech, vol. 22, no. 4, pp. 463–470, Apr. 2013, doi: 10.1007/s11666-013-9893-z.

[17] L. L. D. Chaliampalias, G. Vourlias, N. Pistofidis, G. Stergioudis, and E.K. Polychroniadis, A Morphological and Microstructural Study of Flame-Sprayed Zinc Coatings on Low-Alloyed Steels as a Contribution to Explaining Their Corrosion Resistance, Phys. Status Solidi. (a), 205. 2008.

[18] L. L. T. Kobayashi, T. Maruyama, and M. Kano, ‘Characterization of Pure Aluminum and Zinc Sprayed Coatings produced by Flame Spraying’, Materials transactions 44(12), pp. 2711-2717, December 2003, DOI: 10.2320/matertrans.44.2711

[19] L. L. W. Zhao, Y. Wang, C. Liu, L. Dong, H. Yu, and H. Ai, ‘Erosion- Corrosion of Thermally Sprayed Coatings in Simulated Splash Zone’, Surface and Coatings Technology 205(7): pp. 2267-2272, December 2010, DOI: 10.1016/j.surfcoat.2010.09.011

[20] L. L. E. Almeida, D. Pereira, and O. Figueiredo, ‘The Degradation of Zinc Coatings in Salty Atmospheres’, Prog. Org. Coat., 17. 1989, https://doi.org/10.1016/0033-0655(89)80022-6

[21] L. L. C.J. Li, W.-Y. Li, and H. Fukanuma, ‘Impact Fusion Phenomenon During Cold Spraying of Zinc’, Proceedings of the International Thermal Spray Conference, 2004, https://www.researchgate.net/publication/279896055_Impact_fusion_phenomenon_during_col d_spraying_of_zinc

[22] N. M. Chavan, B. Kiran, A. Jyothirmayi, P. S. Phani, and G. Sundararajan, ‘The Corrosion Behavior of Cold Sprayed Zinc Coatings on Mild Steel Substrate’, Journal of Thermal Spray Technology, vol. 22, no. 4, pp. 463–470, Apr. 2013, doi: 10.1007/s11666-013-9893-z.

[23] S. H. Zahiri, W. Yang, and M. Jahedi, ‘Characterization of Cold Spray Titanium Supersonic Jet’, Journal of Thermal Spray Technology, vol. 18, no. 1, pp. 110–117, Mar. 2009, doi: 10.1007/s11666-008-9278-x.

[24] K. Binder, J. Gottschalk, M. Kollenda, F. Gärtner, and T. Klassen, ‘Influence of Impact Angle and Gas Temperature on Mechanical Properties of Titanium Cold Spray Deposits’, Journal of Thermal Spray Technology, vol. 20, no. 1–2, pp. 234–242, Jan. 2011, doi: 10.1007/s11666-010-9557-1.

[25] C.-J. Li, H.-T. Wang, Q. Zhang, G.-J. Yang, W.-Y. Li, and H. L. Liao, ‘Influence of Spray Materials and Their Surface Oxidation on the Critical Velocity in Cold Spraying’, J Therm Spray Tech, vol. 19, no. 1, pp. 95–101, Jan. 2010, doi: 10.1007/s11666-009-9427-x.

[26] H. Assadi, F. Gärtner, T. Stoltenhoff, and H. Kreye, ‘Bonding mechanism in cold gas spraying’, Acta Materialia, vol. 51, no. 15, pp. 4379–4394, Sep. 2003, doi: 10.1016/S1359-6454(03)00274- X.

143

[27] A. Loureiro, R. Mendes, J. B. Ribeiro, R. M. Leal, and I. Galvão, ‘Effect of explosive mixture on quality of explosive welds of copper to aluminium’, Materials & Design, vol. 95, pp. 256–267, Apr. 2016, doi: 10.1016/j.matdes.2016.01.116.

[28] G. Sundararajan, N. M. Chavan, G. Sivakumar, and P. Sudharshan Phani, ‘Evaluation of Parameters for Assessment of Inter-Splat Bond Strength in Cold-Sprayed Coatings’, Journal of Thermal Spray Technology, vol. 19, no. 6, pp. 1255–1266, Dec. 2010, doi: 10.1007/s11666-010- 9527-7.

[29] H. Koivuluoto, J. Lagerbom, M. Kylmälahti, and P. Vuoristo, ‘Microstructure and Mechanical Properties of Low-Pressure Cold-Sprayed (LPCS) Coatings’, Journal of Thermal Spray Technology, vol. 17, no. 5–6, pp. 721–727, Dec. 2008, doi: 10.1007/s11666-008-9245-6.

[30] J. G. Legoux, E. Irissou, and C. Moreau, ‘Effect of Substrate Temperature on the Formation Mechanism of Cold-Sprayed Aluminum, Zinc and Tin Coatings’, Journal of Thermal Spray Technology, vol. 16, no. 5–6, pp. 619–626, Dec. 2007, doi: 10.1007/s11666-007-9091-y.

[31] W.-Y. Li, C.-J. Li, and G.-J. Yang, ‘Effect of impact-induced melting on interface microstructure and bonding of cold-sprayed zinc coating’, Applied Surface Science, vol. 257, no. 5, pp. 1516–1523, Dec. 2010, doi: 10.1016/j.apsusc.2010.08.089.

[32] Y. Qian, Y. Li, S. Jungwirth, N. Seely, Y. Fang, and X. Shi, ‘The Application of Anti-Corrosion Coating for Preserving the Value of Equipment Asset in Chloride-Laden Environments: A Review’, International journal of electrochemical science, vol. 10, pp. 10756–10780, Nov. 2015.

[33] W. Yingying, ‘Electrochemical behavior of cold sprayed coatings dedicated to corrosion protection applications -Role of microstructure-’, p. 196, 2015, http://theses.insa- lyon.fr/publication/2015ISAL0026/these.pdf

[34] L. Stainier and M. Ortiz, ‘Study and validation of a variational theory of thermo-mechanical coupling in finite visco-plasticity’, International Journal of Solids and Structures, vol. 47, no. 5, pp. 705–715, Mar. 2010, doi: 10.1016/j.ijsolstr.2009.11.012.

[35] H. Assadi, H. Kreye, F. Gärtner, and T. Klassen, ‘Cold spraying – A materials perspective’, Acta Materialia, vol. 116, pp. 382–407, Sep. 2016, doi: 10.1016/j.actamat.2016.06.034.

[36] N. A. Fleck, G. M. Muller, M. F. Ashby, and J. W. Hutchinson, ‘Strain gradient plasticity: Theory and experiment’, Acta Metallurgica et Materialia, vol. 42, no. 2, pp. 475–487, Feb. 1994, doi: 10.1016/0956-7151(94)90502-9.

[37] H. Hallberg, B. Svendsen, T. Kayser, and M. Ristinmaa, ‘Microstructure evolution during dynamic discontinuous recrystallization in particle-containing Cu’, Computational Materials Science, vol. 84, pp. 327–338, Mar. 2014, doi: 10.1016/j.commatsci.2013.12.021.

[38] H. Kreye, ‘Über den Bindemechanismus beim Explosivschweißen’, Schweissen Schneiden, vol. 37, pp. 297–302, 1985, https://www.semanticscholar.org/paper/Modelling-Pulse-Magnetic- Welding-Processes-%E2%80%93-An-Uhlmann- Ziefle/0baec2898ca0c030e798392450ab37ca46017667

[39] C. Borchers, F. Gärtner, T. Stoltenhoff, and H. Kreye, ‘Formation of persistent dislocation loops by ultra-high strain-rate deformation during cold spraying’, Acta Materialia, vol. 53, no. 10, pp. 2991–3000, Jun. 2005, doi: 10.1016/j.actamat.2005.02.048.

144

[40] H. Kreye, M. Hammerschmidt, and C.-P. Woidneck, ‘Discussion on the fusion characteristics during the process of explosion welding.’, vol. 37, pp. e98–e100, Jul. 1985.

[41] A. Berlin, T. C. Nguyen, M. J. Worswick, and Y. Zhou, ‘Metallurgical analysis of magnetic pulse welds of AZ31 magnesium alloy’, Science and Technology of Welding and Joining, vol. 16, no. 8, pp. 728–734, Nov. 2011, doi: 10.1179/1362171811Y.0000000070.

[42] R. Johnson and W. K. Cook, ‘A constitutive model and data for metals subjected to large strains high strain rates and high temperatures. The 7th International Symposium on Ballistics’, The Hague, pp. 541–547, Jan. 1983, https://pdfs.semanticscholar.org/40b2/4ee656f8f82a1927338117cd1999c96517a7.pdf

[43] T. Schmidt, ‘Kaltgasspritzen : eine Analyse des Materialverhaltens beim Partikelaufprall und die daraus abgeleitete Prozessoptimierung’, Hamburg, Helmut-Schmidt-Univ., Diss., 2007.

[44] M. V. Vidaller, A. List, F. Gaertner, T. Klassen, S. Dosta, and J. M. Guilemany, ‘Single Impact Bonding of Cold Sprayed Ti-6Al-4V Powders on Different Substrates’, J Therm Spray Tech, vol. 24, no. 4, pp. 644–658, Apr. 2015, doi: 10.1007/s11666-014-0200-4.

[45] K. Kim, M. Watanabe, K. Mitsuishi, K. Iakoubovskii, and S. Kuroda, ‘Impact bonding and rebounding between kinetically sprayed titanium particle and steel substrate revealed by high- resolution electron microscopy’, J. Phys. D: Appl. Phys., vol. 42, no. 6, p. 065304, Mar. 2009, doi: 10.1088/0022-3727/42/6/065304.

[46] H. ASSADI, ‘Application of Analytical Methods for Understanding and Optimization of the Cold Spray Process’, Proc. of 6th HVOF Colloquium, 2003, 2003.

[47] L. L. F. Azarmi, T. Coyle, and J. Mostaghimi, ‘Young’s Modulus Measurement and Study of the Relationship Between Mechanical Properties and Microstructure of Air Plasma Sprayed Alloy’ 625, Surf. Coat. Technol, 23. 2009, DOI: 10.1016/j.surfcoat.2008.09.035

[48] R. McPherson, ‘A model for the thermal conductivity of plasma-sprayed ceramic coatings’, Thin Solid Films, vol. 112, no. 1, pp. 89–95, Feb. 1984, doi: 10.1016/0040-6090(84)90506-6.

[49] E. by B. R. M. Montavon Margaret M. Hyland, Yuk-Chiu Lau, Chang-Jiu Li, Rogerio S. Lima, Ghislain, ‘Mechanical Properties and Corrosion Resistance of Cold Sprayed Coatings’, Building on 100 Years Success: Proceedings of the 2006 International Thermal Spray Conference. Seattle, Washington, USA: ASM International, 2006.

[50] D. Boruah et al., ‘Evaluation of residual stresses induced by cold spraying of Ti-6Al-4V on Ti-6Al- 4V substrates’, Surface and Coatings Technology, vol. 374, pp. 591–602, Sep. 2019, doi: 10.1016/j.surfcoat.2019.06.028.

[51] R. Singh, S. Schruefer, S. Wilson, J. Gibmeier, and R. Vassen, ‘Influence of coating thickness on residual stress and adhesion-strength of cold-sprayed Inconel 718 coatings’, Surface and Coatings Technology, vol. 350, pp. 64–73, Sep. 2018, doi: 10.1016/j.surfcoat.2018.06.080.

[52] C.-J. Li, W.-Y. Li, and Y.-Y. Wang, ‘Formation of metastable phases in cold-sprayed soft metallic deposit’, Surface and Coatings Technology, vol. 198, no. 1–3, pp. 469–473, Aug. 2005, doi: 10.1016/j.surfcoat.2004.10.063.

145

[53] G. Bae, K. Kang, and C. Lee, ‘Nanoscale deformation twinning at ultrahigh strain rates during kinetic spraying of nickel’, Materials Letters, vol. 89, pp. 320–323, Dec. 2012, doi: 10.1016/j.matlet.2012.08.133.

[54] K. Kang, J. Won, G. Bae, S. Ha, and C. Lee, ‘Interfacial bonding and microstructural evolution of Al in kinetic spraying’, J Mater Sci, vol. 47, no. 11, pp. 4649–4659, Jun. 2012, doi: 10.1007/s10853- 012-6332-3.

[55] G. Bae, K. Kang, J.-J. Kim, and C. Lee, ‘Nanostructure formation and its effects on the mechanical properties of kinetic sprayed titanium coating’, Materials Science and Engineering: A, vol. 527, no. 23, pp. 6313–6319, Sep. 2010, doi: 10.1016/j.msea.2010.06.037.

[56] Y. Xiong, K. Kang, G. Bae, S. Yoon, and C. Lee, ‘Dynamic amorphization and recrystallization of metals in kinetic spray process’, Appl. Phys. Lett., vol. 92, no. 19, p. 194101, May 2008, doi: 10.1063/1.2928218.

[57] S. Guetta et al., ‘Influence of Particle Velocity on Adhesion of Cold-Sprayed Splats’, J Therm Spray Tech, vol. 18, no. 3, pp. 331–342, Sep. 2009, doi: 10.1007/s11666-009-9327-0.

[58] K. H. Ko, J. O. Choi, and H. Lee, ‘Pretreatment effect of Cu feedstock on cold-sprayed coatings’, Journal of Materials Processing Technology, vol. 214, no. 8, pp. 1530–1535, Aug. 2014, doi: 10.1016/j.jmatprotec.2014.02.020.

[59] Y. Xiong, X. Xiong, S. Yoon, G. Bae, and C. Lee, ‘Dependence of Bonding Mechanisms of Cold Sprayed Coatings on Strain-Rate-Induced Non-Equilibrium Phase Transformation’, J Therm Spray Tech, vol. 20, no. 4, pp. 860–865, Jun. 2011, doi: 10.1007/s11666-011-9634-0.

[60] R. N. Raoelison et al., ‘Cold gas dynamic spray technology: A comprehensive review of processing conditions for various technological developments till to date’, Additive Manufacturing, vol. 19, pp. 134–159, Jan. 2018, doi: 10.1016/j.addma.2017.07.001.

[61] V. N. V. Munagala, V. Akinyi, P. Vo, and R. R. Chromik, ‘Influence of Powder Morphology and Microstructure on the Cold Spray and Mechanical Properties of Ti6Al4V Coatings’, Journal of Thermal Spray Technology, vol. 27, no. 5, pp. 827–842, Jun. 2018, doi: 10.1007/s11666-018- 0729-8.

[62] Y.-J. Li, X.-T. Luo, and C.-J. Li, ‘Dependency of deposition behavior, microstructure and properties of cold sprayed Cu on morphology and porosity of the powder’, Surface and Coatings Technology, vol. 328, pp. 304–312, Nov. 2017, doi: 10.1016/j.surfcoat.2017.08.070.

[63] K. H. Ko, J. O. Choi, and H. Lee, ‘Characteristics of cold sprayed dendritic Cu coatings’, Surface Engineering, vol. 32, no. 9, pp. 650–654, Feb. 2016, doi: 10.1179/1743294415y.0000000044.

[64] R. A. Seraj, A. Abdollah-zadeh, S. Dosta, H. Canales, H. Assadi, and I. G. Cano, ‘The effect of traverse speed on deposition efficiency of cold sprayed Stellite 21’, Surface and Coatings Technology, vol. 366, pp. 24–34, May 2019, doi: 10.1016/j.surfcoat.2019.03.012.

[65] A. W.-Y. Tan et al., ‘Effect of coating thickness on microstructure, mechanical properties and fracture behaviour of cold sprayed Ti6Al4V coatings on Ti6Al4V substrates’, Surface and Coatings Technology, vol. 349, pp. 303–317, Sep. 2018, doi: 10.1016/j.surfcoat.2018.05.060.

[66] C. Chen et al., ‘A novel spiral trajectory for damage component recovery with cold spray’, Surface and Coatings Technology, vol. 309, pp. 719–728, Jan. 2017, doi: 10.1016/j.surfcoat.2016.10.096.

146

[67] J. Pattison, S. Celotto, A. Khan, and W. O’Neill, ‘Standoff distance and bow shock phenomena in the Cold Spray process’, Surface and Coatings Technology, vol. 202, no. 8, pp. 1443–1454, Jan. 2008, doi: 10.1016/j.surfcoat.2007.06.065.

[68] M. S. Morsi, ‘Effect of Air Plasma Sprays Parameters on Coating Performance in Zirconia–Based Thermal Barrier Coatings’, Int. J. Electrochem. Sci., vol. 7, p. 21, 2012, http://www.electrochemsci.org/papers/vol7/7042811.pdf

[69] J.D. Anderson, ‘Modern Compressible Flow: with historical perspective’, 3rd ed., McGraw-Hill, New York, vol. 15, pp. 223–232, 2003, doi: 10.1361/105996306X108110.

[70] A.N. Papyrin, et al, ‘ITSC’, ASM International, 2001, doi: 10.1361/105996306X108110.

[71] V. Shukla, G. S. Elliott, and B. H. Kear, ‘Nanopowder deposition by supersonic rectangular jet impingement’, J Therm Spray Tech, vol. 9, no. 3, pp. 394–398, Sep. 2000, doi: 10.1361/105996300770349854.

[72] ‘R.C. Dykhuizen, R.A. Neiser, ITSC, ASM International, 2003’, doi: 10.1361/105996306X108110.

[73] ‘D.L. Gilmore, et al., J. Therm. Spray Technol. 8 (4) (1999) 576’, doi: 10.1361/105996306X108110.

[74] A. P. Alkhimov, V. F. Kosarev, and S. V. Klinkov, ‘The features of cold spray nozzle design’, J Therm Spray Tech, vol. 10, no. 2, pp. 375–381, Jun. 2001, doi: 10.1361/105996301770349466.

[75] P. Sudharshan Phani, D. Srinivasa Rao, S. V. Joshi, and G. Sundararajan, ‘Effect of Process Parameters and Heat Treatments on Properties of Cold Sprayed Copper Coatings’, Journal of Thermal Spray Technology, vol. 16, no. 3, pp. 425–434, Aug. 2007, doi: 10.1007/s11666-007- 9048-1.

[76] G. Purcek, H. Yanar, D. Shangina, M. Demirtaş, N. R. Bochvar, and S. Dobatkin, ‘Influence of high pressure torsion-induced grain refinement and subsequent aging on tribological properties of Cu-Cr-Zr alloy’, Journal of Alloys and Compounds, vol. 742, pp. 325–333, Apr. 2018, doi: 10.1016/j.jallcom.2018.01.303.

[77] H. Koivuluoto, G. Bolelli, L. Lusvarghi, F. Casadei, and P. Vuoristo, ‘Corrosion resistance of cold- sprayed Ta coatings in very aggressive conditions’, Surface and Coatings Technology, vol. 205, no. 4, pp. 1103–1107, Nov. 2010, doi: 10.1016/j.surfcoat.2010.02.052.

[78] A. R. Marder, ‘The metallurgy of zinc-coated steel’, Progress in Materials Science, vol. 45, no. 3, pp. 191–271, Jun. 2000, doi: 10.1016/S0079-6425(98)00006-1.

[79] G. E. Dieter, Ed., ‘Effects of Surface Treatments on Materials Performance’, in Materials Selection and Design, ASM International, 1997, pp. 470–490, DOI: 10.11868/j.issn.1001-4381.2016.000590

[80] E. Palma, J. M. Puente, and M. Morcillo, ‘The atmospheric corrosion mechanism of 55%Al-Zn coating††Made in Spain by CSI Pianos under the trademark Algafort, which has been licensed by BIEC International, Inc. on steel’, Corrosion Science, vol. 40, no. 1, pp. 61–68, Jan. 1998, doi: 10.1016/S0010-938X(97)00112-1.

[81] A. P. Yadav, ‘Electrochemical Impedance Response of Zn and Galvanized Steel Corroding under Marine Atmospheric Environments’, Journal of Nepal Chemical Society, vol. 23, pp. 33–42, Jan. 1970, doi: 10.3126/jncs.v23i0.2094.

147

[82] M. Mouanga, P. Berçot, and J. Y. Rauch, ‘Comparison of corrosion behaviour of zinc in NaCl and in NaOH solutions. Part I: Corrosion layer characterization’, Corrosion Science, vol. 52, no. 12, pp. 3984–3992, Dec. 2010, doi: 10.1016/j.corsci.2010.08.003.

[83] S. Bonk, M. Wicinski, A. W. Hassel, and M. Stratmann, ‘Electrochemical characterizations of precipitates formed on zinc in alkaline sulphate solution with increasing pH values’, Electrochemistry Communications, vol. 6, no. 8, pp. 800–804, Aug. 2004, doi: 10.1016/j.elecom.2004.05.012.

[84] P. Kalinauskas, I. Valsiūnas, M. Samulevičien≐, and E. Juzeliūnas, ‘Zinc photo-corrosion in neutral solutions’, Corrosion Science, vol. 43, no. 11, pp. 2083–2092, Nov. 2001, doi: 10.1016/S0010- 938X(01)00013-0.

[85] D. B. Blücher, J.-E. Svensson, and L.-G. Johansson, ‘The influence of CO2, AlCl3·6H2O, MgCl2·6H2O, Na2SO4 and NaCl on the atmospheric corrosion of aluminum’, Corrosion Science, vol. 48, no. 7, pp. 1848–1866, Jul. 2006, doi: 10.1016/j.corsci.2005.05.027.

[86] A. M. Beccaria and G. Poggi, ‘Aluminum Corrosion in Slightly Alkaline Sodium Sulfate Solutions at Different Hydrostatic Pressures’, CORROSION, vol. 43, no. 3, pp. 153–158, Mar. 1987, doi: 10.5006/1.3583128.

[87] M. Tafreshi, S. R. Allahkaram, and H. Farhangi, ‘Comparative study on structure, corrosion properties and tribological behavior of pure Zn and different Zn-Ni alloy coatings’, Materials Chemistry and Physics, vol. 183, pp. 263–272, Nov. 2016, doi: 10.1016/j.matchemphys.2016.08.026.

[88] L. L. Shreir, Corrosion, Volume 1 Metal/Environment reactions; Volume 2 Corrosion control. 1976, ISBN-10 : 0408001097

[89] C.-C. Lin and C.-M. Huang, ‘Zinc-Nickel alloy coatings electrodeposited by pulse current and their corrosion behavior’, J Coat. Technol. Res., vol. 3, no. 2, pp. 99–104, Apr. 2006, doi: 10.1007/s11998-006-0011-8.

[90] K. A. Yasakau et al., ‘Influence of stripping and cooling atmospheres on surface properties and corrosion of zinc galvanizing coatings’, Applied Surface Science, vol. 389, pp. 144–156, Dec. 2016, doi: 10.1016/j.apsusc.2016.07.093.

[91] L. L. P. Ozga, E. Biela’nska,’ Determination of the corrosion rate of Zn and ZneNi layers by the EDS technique’, Mater. Chem. Phys. 81. 2003.

[92] L. L. S. Dubent, M.L.A.D. Mertens, M. Saurat, ‘Electrodeposition, characterization and corrosion behaviour of tine20 wt.% zinc coatings electroplated from a non-cyanide alkaline bath’, Mater. Chem. Phys. 120. 2010, https://hal.archives-ouvertes.fr/hal-02458179/document

[93] L. L. M. Abou-Krisha, ‘Influence of Ni2þ concentration and deposition potential on the characterization of thin electrodeposited ZneNieCo coatings’, Mater. Chem. Phys. 125. 2011.

[94] J. Duchoslav et al., ‘Nanoscopic view on the initial stages of corrosion of hot dip galvanized Zn– Mg–Al coatings’, Corrosion Science, vol. Complete, no. 83, pp. 327–334, 2014, doi: 10.1016/j.corsci.2014.02.027.

148

[95] L. L. A. Nazarov, M.-G. Olivier, D. Thierry, ‘SKP and FT-IR microscopy study of thepaint corrosion de-adhesion from the surface of galvanized steel’, Prog. Org.Coat. 74. 2012, https://doi.org/10.1016/j.porgcoat.2011.10.009

[96] A. Venugopal, R. D. Angal, and V. S. Raja, ‘Effect of Grain-Boundary Corrosion on Impedance Characteristics of an Aluminum-Zinc-Indium Alloy in 3.5% Sodium Chloride Solution’, CORROSION, vol. 52, no. 2, pp. 138–142, Feb. 1996, doi: 10.5006/1.3292104.

[97] J.-C. Hung, Y.-R. Liu, H.-P. Tsui, and Z.-W. Fan, ‘Electrode insulation layer for electrochemical machining fabricated through hot-dip aluminizing and microarc oxidation on a stainless-steel substrate’, Surface and Coatings Technology, vol. 378, p. 124995, Nov. 2019, doi: 10.1016/j.surfcoat.2019.124995.

[98] V. Saarimaa, A. Markkula, J. Juhanoja, and B.-J. Skrifvars, ‘Novel insight to aluminum compounds in the outermost layers of hot dip galvanized steel and how they affect the reactivity of the zinc surface’, Surface and Coatings Technology, vol. 306, pp. 506–511, Nov. 2016, doi: 10.1016/j.surfcoat.2015.11.014.

[99] S.-C. Jeng, ‘Oxidation behavior and microstructural evolution of hot-dipped aluminum coating on Ti-6Al-4V alloy at 800°C’, Surface and Coatings Technology, vol. 235, pp. 867–874, Nov. 2013, doi: 10.1016/j.surfcoat.2013.09.023.

[100] C. W. Borgmann and U. R. Evans, ‘The Corrosion of Zinc in Chloride Solutions’, Trans. Electrochem. Soc., vol. 65, no. 1, pp. 249–274, Jan. 1934, doi: 10.1149/1.3498027.

[101] Y. Bai, Z. H. Wang, X. B. Li, G. S. Huang, C. X. Li, and Y. Li, ‘Corrosion behavior of low pressure cold sprayed Zn-Ni composite coatings’, Journal of Alloys and Compounds, vol. 719, pp. 194–202, Sep. 2017, doi: 10.1016/j.jallcom.2017.05.134.

[102] A. Berni, M. Mennig, and H. Schmidt, ‘Doctor Blade’, in Sol-Gel Technologies for Glass Producers and Users, M. A. Aegerter and M. Mennig, Eds. Boston, MA: Springer US, 2004, pp. 89–92, https://doi.org/10.1007/978-0-387-88953-5_10

[103] S. Xing, Y. Li, J. Wu, and Y. Yan, ‘Electrochemical methods study on corrosion of 5% Al–Zn alloy-coated steel under thin electrolyte layers’, Materials and Corrosion, vol. 61, no. 5, pp. 428– 431, 2010, doi: 10.1002/maco.200905313.

[104] R Brindha, Karthigai Selvii B, and M Selvam, ‘Corrosion Behavior of Zn/Graphene Composite in Aqueous Electrolyte System’, Journal of Nanoscience and Nanotechnology Applications, ISSN: 2577-7920

[105] S. Gayathri, P. Jayabal, M. Kottaisamy, and V. Ramakrishnan, ‘Synthesis of few layer graphene by direct exfoliation of graphite and a Raman spectroscopic study’, AIP Advances, vol. 4, no. 2, p. 027116, Feb. 2014, doi: 10.1063/1.4866595.

[106] R Brindha, Karthigai Selvii B, and M Selvam, ‘Corrosion Behavior of Zn/Graphene Composite in Aqueous Electrolyte System’, Journal of Nanoscience and Nanotechnology Applications, ISSN: 2577-7920

[107] K. Kamburova, N. Boshkova, N. Boshkov, G. Atanassova, and T. Radeva, ‘Hybrid zinc coatings for corrosion protection of steel using polyelectrolyte nanocontainers loaded with benzotriazole’, Colloids and Surfaces A: Physicochemical and Engineering Aspects, vol. 559, pp. 243–250, Dec. 2018, doi: 10.1016/j.colsurfa.2018.09.039.

149

[108] Y. Zhang, Q. Wang, and C. S. Ramachandran, ‘Synthesis of carbon nanotube reinforced aluminum composite powder (CNT-Al) by polymer pyrolysis chemical vapor deposition (PP-CVD) coupled high energy ball milling (HEBM) process’, Diamond and Related Materials, vol. 104, p. 107748, Apr. 2020, doi: 10.1016/j.diamond.2020.107748.

[109] I. Ozdemir, K. Ogawa, and K. Sato, ‘Iron boron based powders sprayed by high velocity spray processes’, Surface and Coatings Technology, vol. 240, pp. 373–379, Feb. 2014, doi: 10.1016/j.surfcoat.2013.12.056.

[110] D. J. Woo et al., ‘Synthesis of nanodiamond-reinforced aluminum metal composite powders and coatings using high-energy ball milling and cold spray’, Carbon, vol. 63, pp. 404–415, Nov. 2013, doi: 10.1016/j.carbon.2013.07.001.

[111] F. S. da Silva et al., ‘Corrosion characteristics of cold gas spray coatings of reinforced aluminum deposited onto carbon steel’, Corrosion Science, vol. 114, pp. 57–71, Jan. 2017, doi: 10.1016/j.corsci.2016.10.019.

[112] Y.-K. Wei, X.-T. Luo, X. Chu, G.-S. Huang, and C.-J. Li, ‘Solid-state additive manufacturing high performance aluminum alloy 6061 enabled by an in-situ micro-forging assisted cold spray’, Materials Science and Engineering: A, vol. 776, p. 139024, Mar. 2020, doi: 10.1016/j.msea.2020.139024.

[113] J. Morimoto, T. Onoda, Y. Sasaki, and N. Abe, ‘Improvement of solid cold sprayed TiO2–Zn coating with direct diode laser’, Vacuum, vol. 73, no. 3–4, pp. 527–532, Apr. 2004, doi: 10.1016/j.vacuum.2003.12.157.

[114] J. W. Murray, M. V. Zuccoli, and T. Hussain, ‘Heat Treatment of Cold-Sprayed C355 Al for Repair: Microstructure and Mechanical Properties’, J Therm Spray Tech, vol. 27, no. 1, pp. 159– 168, Jan. 2018, doi: 10.1007/s11666-017-0665-z.

[115] S. Yin et al., ‘Advanced diamond-reinforced metal matrix composites via cold spray: Properties and deposition mechanism’, Composites Part B: Engineering, vol. 113, pp. 44–54, Mar. 2017, doi: 10.1016/j.compositesb.2017.01.009.

[116] M. Yu and W. Li, ‘Metal Matrix Composite Coatings by Cold Spray’, in Cold-Spray Coatings, P. Cavaliere, Ed. Cham: Springer International Publishing, 2018, pp. 297–318, DOI: 10.1007/978-3- 319-67183-3_10

[117] B.B. Djordjevic, R.G. Maev, ‘Application for Aerospace Corrosion Protection and Structural Repair, Building on 100 Years Success Proceedings of the 2006 International Thermal Spray Conference’, ResearchGate, 15-May-2006.

[118] A. P. I. Popoola, V. S. Aigbodion, and O. S. I. Fayomi, ‘Surface characterization, mechanical properties and corrosion behaviour of ternary based Zn–ZnO–SiO2composite coating of mild steel’, Journal of Alloys and Compounds, vol. 654, pp. 561–566, Jan. 2016, doi: 10.1016/j.jallcom.2015.09.090.

[119] S. Rech, A. Trentin, S. Vezzù, J.-G. Legoux, E. Irissou, and M. Guagliano, ‘Influence of Pre- Heated Al 6061 Substrate Temperature on the Residual Stresses of Multipass Al Coatings Deposited by Cold Spray’, Journal of Thermal Spray Technology, vol. 20, no. 1–2, pp. 243–251, Jan. 2011, doi: 10.1007/s11666-010-9596-7.

150

[120] W.-Y. Li, G. Zhang, H. L. Liao, and C. Coddet, ‘Characterizations of cold sprayed TiN particle reinforced Al2319 composite coating’, Journal of Materials Processing Technology, vol. 202, no. 1, pp. 508–513, Jun. 2008, doi: 10.1016/j.jmatprotec.2007.09.045.

[121] K. Spencer, D. M. Fabijanic, and M.-X. Zhang, ‘The use of Al–Al2O3 cold spray coatings to improve the surface properties of magnesium alloys’, Surface and Coatings Technology, vol. 204, no. 3, pp. 336–344, Oct. 2009, doi: 10.1016/j.surfcoat.2009.07.032.

[122] B. Jodoin, L. Ajdelsztajn, E. Sansoucy, A. Zúñiga, P. Richer, and E. J. Lavernia, ‘Effect of particle size, morphology, and hardness on cold gas dynamic sprayed aluminum alloy coatings’, Surface and Coatings Technology, vol. 201, no. 6, pp. 3422–3429, Dec. 2006, doi: 10.1016/j.surfcoat.2006.07.232.

[123] S. R. Hartshorn, Ed., Structural Adhesives: Chemistry and Technology. Springer US, 1986, https://www.springer.com/gp/book/9781468477832

[124] R. C. McCune, A. N. Papyrin, J. N. Hall, W. L. I. Riggs, and P. H. Zajchowski, ‘An exploration of the cold gas-dynamic spray method for several materials systems’, Dec. 1995, ISBN 0-87170-541- 9

[125] A. P. Alkhimov, S. V. Klinkov, V. F. Kosarev, and A. N. Papyrin, ‘Gas-dynamic spraying study of a plane supersonic two-phase jet’, J Appl Mech Tech Phys, vol. 38, no. 2, pp. 324–330, Mar. 1997, doi: 10.1007/BF02467920.

[126] T. Stoltenhoff, H. Kreye, and H. J. Richter, ‘An analysis of the cold spray process and its coatings’, J Therm Spray Tech, vol. 11, no. 4, pp. 542–550, Dec. 2002, doi: 10.1361/105996302770348682.

[127] F. Gärtner, T. Stoltenhoff, T. Schmidt, and H. Kreye, ‘The cold spray process and its potential for industrial applications’, J Therm Spray Tech, vol. 15, no. 2, pp. 223–232, Jun. 2006, doi: 10.1361/105996306X108110.

[128] T. Schmidt et al., ‘From Particle Acceleration to Impact and Bonding in Cold Spraying’, Journal of Thermal Spray Technology, vol. 18, pp. 794–808, Dec. 2009, doi: 10.1007/s11666-009-9357-7.

[129] S. Grigoriev, A. Okunkova, A. Sova, P. Bertrand, and I. Smurov, ‘Cold spraying: From process fundamentals towards advanced applications’, Surface and Coatings Technology, vol. 268, pp. 77–84, Apr. 2015, doi: 10.1016/j.surfcoat.2014.09.060.

[130] J. Pattison, S. Celotto, R. Morgan, M. Bray, and W. O’Neill., ‘Cold gas dynamic manufacturing: A non-thermal approach to freeform fabrication’, International Journal of Machine Tools and Manufacture, vol. 47, no. 3, pp. 627–634, Mar. 2007, doi: 10.1016/j.ijmachtools.2006.05.001.

[131] E. Irissou, J.-G. Legoux, A. Ryabinin, and B. Jodoin, ‘Review on Cold Spray Process and Technology: Part I - Intellectual Property’, Journal of Thermal Spray Technology, vol. 17, Dec. 2008, doi: 10.1007/s11666-008-9203.

[132] A. Papyrin, V. Kosarev, S. Klinkov, A. Alkhimov, and V. M. Fomin, ‘Cold Spray Technology, 1 edition’. Amsterdam ; London: Elsevier Science, 2006, DOI: 10.1063/1.5065141

[133] S. V. Klinkov, V. F. Kosarev, and V. N. Zaikovskii, ‘Influence of Flow Swirling and Exit Shape of Barrel Nozzle on Cold Spraying’, J Therm Spray Tech, vol. 20, no. 4, pp. 837–844, Jun. 2011, doi: 10.1007/s11666-011-9621-5.

151

[134] R. Lupoi and W. O’Neill, ‘Powder stream characteristics in cold spray nozzles’, Surface and Coatings Technology, vol. 206, no. 6, pp. 1069–1076, Dec. 2011, doi: 10.1016/j.surfcoat.2011.07.061.

[135] A. Sova, S. Grigoriev, A. Okunkova, and I. Smurov, ‘Potential of cold gas dynamic spray as additive manufacturing technology’, Int J Adv Manuf Technol, vol. 69, no. 9, pp. 2269–2278, Dec. 2013, doi: 10.1007/s00170-013-5166-8.

[136] R. A. Granger, ‘Solutions manual for fluid mechanics’. New York: Holt, Rinehart, and Winston, 1985, ISBN 0030706378

[137] L. L. American Welding Society, ‘Corrosion Tests of Flame Sprayed Coated Steel’ 19 Year Report, C2 ed. 1974, ISBN-10 : 0871711117

[138] ‘Metals HandBook VOL 2, Properties and Selection: Nonferrous Alloys and Special-Purpose Materials. Available: https://authorzilla.com/nW2zv/metals-handbook-vol-2-properties-and- selection-nonferrous-alloys-and-special-purpose-materials.html. [Accessed: 30-Mar-2019].

[139] R. V. Dennis, L. T. Viyannalage, T. K. Rout, and S. Banerjee, ‘Graphene nanocomposite coatings for protecting low- alloy steels from corrosion’, American Ceramic Society Bulletin, vol. 92, no. 5, p. 7, https://pdfs.semanticscholar.org/a606/73b19aa039f8e90324bb64a0eb72b910b94f.pdf?_ga=2.1 67354941.266324386.1598351626-816649582.1598351626

[140] S. Majumdar and P. Banerji, ‘Hopping conduction in nitrogen doped ZnO in the temperature range 10–300 K’, Journal of Applied Physics, vol. 107, no. 6, p. 063702, Mar. 2010, doi: 10.1063/1.3353862.

[141] Y. Xiao, X. Jiang, Y. Xiao, and L. Ma, ‘Research on Zn-Al15 thermal spray metal coating and its organic painting composite system protection performance’, Procedia Engineering, vol. 27, pp. 1644–1653, Jan. 2012, doi: 10.1016/j.proeng.2011.12.632.

[142] M. Grujicic, C. L. Zhao, W. S. DeRosset, and D. Helfritch, ‘Adiabatic shear instability based mechanism for particles/substrate bonding in the cold-gas dynamic-spray process’, Materials & Design, vol. 25, no. 8, pp. 681–688, Dec. 2004, doi: 10.1016/j.matdes.2004.03.008.

[143] M. Hassani-Gangaraj, D. Veysset, V. K. Champagne, K. A. Nelson, and C. A. Schuh, ‘Adiabatic shear instability is not necessary for adhesion in cold spray’, Acta Materialia, vol. 158, pp. 430– 439, Oct. 2018, doi: 10.1016/j.actamat.2018.07.065.

[144] J. F. Li et al., ‘Uniform design method for optimization of process parameters of plasma sprayed TiN coatings’, Surface and Coatings Technology, vol. 176, no. 1, pp. 1–13, Nov. 2003, doi: 10.1016/S0257-8972(03)00019-7.

[145] A. Bai, K.-L. Yang, H.-L. Chen, Y. Hong, and S.-B. Chang, ‘High current density on electroplating smooth alkaline zinc coating’, MATEC Web Conf., vol. 123, p. 00024, 2017, doi: 10.1051/matecconf/201712300024.

[146] H. Kim, B. N. Popov, and K. S. Chen, ‘Comparison of corrosion-resistance and hydrogen permeation properties of Zn–Ni, Zn–Ni–Cd and Cd coatings on low-carbon steel’, Corrosion Science, vol. 45, no. 7, pp. 1505–1521, Jul. 2003, doi: 10.1016/S0010-938X(02)00228-7.

152

[147] Z. B. Zhao, B. A. Gillispie, and J. R. Smith, ‘Coating deposition by the kinetic spray process’, Surface and Coatings Technology, vol. 200, no. 16–17, pp. 4746–4754, Apr. 2006, doi: 10.1016/j.surfcoat.2005.04.033.

[148] R.M. Souto a, B. Normand b,1, H. Takenouti b , M. Keddam, ‘Self-healing processes in coil- coated cladding studied by the scanning’, Electrochimica Acta, 55, pp.4551–4557, 2010, doi:10.1016/j.electacta.2010.03.008

[149] P. Araujo and S. Janagap, ‘Doehlert uniform shell designs and chromatography’, Journal of Chromatography B, vol. 910, pp. 14–21, Dec. 2012, doi: 10.1016/j.jchromb.2012.05.019.

[150] V. V. Lozenko and V. G. Shepelevich, ‘Grain and subgrain structure of rapidly solidified Zn, Zn- Cd, Zn-Sn, and Zn-Sb foils’, Inorg Mater, vol. 43, no. 1, pp. 20–24, Jan. 2007, doi: 10.1134/S0020168507010062.

[151] Suttinart Noothongkaewa, Hideki Nakajima, Anusorn Tong-on, Worawat Meevasana, Prayoon Songsiriritthigul ‘Oxidation of Zn in UHV environment at low temperature’, Applied Surface Science, 258, pp. 1955– 1957, 2012, doi:10.1016/j.apsusc.2011.06.071

[152] Brereton, R.G., Chemometrics, ‘Data Analysis for the Laboratory and Chemical Plant’, Chichester,Available:https://www.researchgate.net/publication/244652148_Brereton_RG_Che mometrics_Data_Analysis_for_the_Laboratory_and_Chemical_Plant_Chichester_Wiley_2003_48 9_pp. [Accessed: 22-Feb-2019].

[153] O. Sailer, ‘Thomas P. Ryan: Modern Experimental Design, Statistical Papers, vol. 49, no. 3, pp. 597–598, Jul. 2008, doi: 10.1007/s00362-007-0090-3.

[154] D. Robinson, ‘Statistical Design: Chemometrics By R. E. Bruns, I. S. Scaramino, and B. De Barros Neto’. Elsevier: Org. Process Res. Dev., vol. 10, no. 5, pp. 1082–1083, Sep. 2006, doi: 10.1021/op060138n.

[155] S. D. Brown, R. Tauler, and B. Walczak, ‘Comprehensive Chemometrics: Chemical and Biochemical Data Analysis’, Illustrated edition. Amsterdam: Elsevier, 2009, DOI: 10.1016/j.chemolab.2010.05.003

[156] S. D. Brown, ‘Chemometrics: A textbook. D. L. Massart. B. G. M. Vandeginste, S. N. Deming, Y. Michotte, and L. Kaufman, Elsevier, Amsterdam, 1988. ISBN 0-444-42660-4. Price Dfl 175.00’, Journal of Chemometrics, vol. 2, no. 4, pp. 298–299, 1988, doi: 10.1002/cem.1180020409.

[157] G. Hanrahan and F. A. Gomez, Eds., ‘Chemometric methods in capillary electrophoresis’. Hoboken, N.J: Wiley, 2010., ISBN-10: 0470393297

[158] M. Feinberg, ‘Experimental design: A chemometric approach, S. N. Deming and S. L. Morgan; from the series: “Data Handling in Science and Technology”, Journal of Chemometrics, vol. 2, no. 2, pp. 169–169, 1988, doi: 10.1002/cem.1180020208.

[159]. D.K. Christoulis et al., ‘J. Therm. Spray Technol’, Journal of Thermal Spray Technology.19, vol. 5, pp. 1062–1073, 2009, doi: 10.1361/105996301770349466.

[160] Y. Danlos, S. Costil, X. Guo, H. Liao, and C. Coddet, ‘Ablation laser and heating laser combined to cold spraying’, Surface and Coatings Technology, vol. 205, no. 4, pp. 1055–1059, Nov. 2010, doi: 10.1016/j.surfcoat.2010.06.018.

153

[161] R. Maev and V. Leshchynsky, ‘Introduction to Low Pressure Gas Dynamic Spray: Physics & Technology’, Introduction to Low Pressure Gas Dynamic Spray: Physics & Technology, pp. 1–234, Mar. 2008, doi: 10.1002/9783527621903.

[162] A. Shkodkin, , A. Kashirin, O. Klyuev, and T. Buzdygar, ‘J. Therm. Spray Technol. 15’, pp. 382– 385, 2006, https://doi.org/10.1361/105996306X124383

[163] C. G. Zoski, Ed., ‘Handbook of electrochemistry’, 1st ed. Amsterdam ; Boston: Elsevier, 2007, ISBN: 9780444519580

[164] F. Andreatta and L. Fedrizzi, ‘The use of the electrochemical micro-cell for the investigation of corrosion phenomena’, Electrochimica Acta, vol. 203, pp. 337–349, Jun. 2016, doi: 10.1016/j.electacta.2016.01.099.

[165] F. Arjmand and A. Adriaens, ‘Microcapillary electrochemical droplet cells: applications in solid-state surface analysis’, J Solid State Electrochem, vol. 18, no. 7, pp. 1779–1788, Jul. 2014, doi: 10.1007/s10008-014-2413-3.

[166] H. S. Isaacs and G. Kissel, ‘Surface Preparation and Pit Propagation in Stainless Steels’, J. Electrochem. Soc., vol. 119, no. 12, pp. 1628–1632, Jan. 1972, doi: 10.1149/1.2404061.

[167] H. S. Isaacs, ‘The Use of the Scanning Vibrating Electrode Technique for Detecting Defects in Ion Vapor-Deposited Aluminum on Steel’, CORROSION, vol. 43, no. 10, pp. 594–596, Oct. 1987, doi: 10.5006/1.3583835.

[168] K. R. Trethewey, D. A. Sargeant, D. J. Marsh, and A. A. Tamimi, ‘Applications of the scanning reference electrode technique to localized corrosion’, Corrosion Science, vol. 35, no. 1, pp. 127– 134, Jan. 1993, doi: 10.1016/0010-938X(93)90142-4.

[169] F.-R. F. Fan and A. J. Bard, ‘In Situ Scanning Tunneling Microscopic Study of the Corrosion of Type 304L Stainless Steel in Aqueous Chloride Media’, J. Electrochem. Soc., vol. 136, no. 1, pp. 166–170, Jan. 1989, doi: 10.1149/1.2096578.

[170] T. P. Moffat, F.-R. F. Fan, and A. J. Bard, ‘Electrochemical and Scanning Tunneling Microscopic Study of Dealloying of Cu3Au’, J. Electrochem. Soc., vol. 138, no. 11, pp. 3224–3235, Jan. 1991, doi: 10.1149/1.2085396.

[171] R. M. Rynders, C.-H. Paik, R. Ke, and R. C. Alkire, ‘Use of In Situ Atomic Force Microscopy to Image Corrosion at Inclusions’, J. Electrochem. Soc., vol. 141, no. 6, pp. 1439–1445, Jan. 1994, doi: 10.1149/1.2054943.

[172] A. A. Gewirth and B. K. Niece, ‘Electrochemical Applications of in Situ Scanning Probe Microscopy’, Chem. Rev., vol. 97, no. 4, pp. 1129–1162, Jun. 1997, doi: 10.1021/cr960067y.

[173] P. Schmutz and G. S. Frankel, ‘Influence of Dichromate Ions on Corrosion of Pure Aluminum and AA2024‐T3 in NaCl Solution Studied by AFM Scratching’, J. Electrochem. Soc., vol. 146, no. 12, pp. 4461–4472, Jan. 1999, doi: 10.1149/1.1392659.

[174] NACE International, Ed., Corrosion control for low-cost reliability: 12th International Corrosion Congress: preceedings [i.e. proceedings]. Houston, TX: NACE International, 1993, DTIC_ADA273665

154

[175] H. Böhni, T. Suter, and A. Schreyer, ‘Micro- and nanotechniques to study localized corrosion’, Electrochimica Acta, vol. 40, no. 10, pp. 1361–1368, Jul. 1995, doi: 10.1016/0013- 4686(95)00072-M.

[176] T. Suter and H. Böhni, ‘A new microelectrochemical method to study pit initiation on stainless steels’, Electrochimica Acta, vol. 42, no. 20–22, pp. 3275–3280, Jan. 1997, doi: 10.1016/S0013- 4686(70)01783-8.

[177] M. Andreatta, ‘Discovering sequence motifs in quantitative and qualitative pepetide data’. Kgs. Lyngby: Technical University of Denmark, 2012, https://backend.orbit.dtu.dk/ws/portalfiles/portal/74972181/PhD_thesis_MA_280912

[178] T. Suter, P. Schmutz, and O. von Trzebiatowski, ‘Electrochemical Characterization of Submicrometer Structures’, ECS Trans., vol. 3, no. 31, pp. 29–37, Sep. 2007, doi: 10.1149/1.2789213.

[179] R. C. Alkire and K. P. Wong, ‘The corrosion of single pits on stainless steel in acidic chloride solution’, Corrosion Science, vol. 28, no. 4, pp. 411–421, Jan. 1988, doi: 10.1016/0010- 938X(88)90060-1.

[180] J. W. Schultze and A. Bressel, ‘Principles of electrochemical micro- and nano-system technologies’, Electrochimica Acta, vol. 47, no. 1, pp. 3–21, Sep. 2001, doi: 10.1016/S0013- 4686(01)00584-9.

[181] J. C. Eklund, A. M. Bond, J. A. Alden, and R. G. Compton, ‘Perspectives in Modern Voltammetry: Basic Concepts and Mechanistic Analysis’, in Advances in Physical Organic Chemistry, vol. 32, D. Bethell, Ed. Academic Press, 1999, pp. 1–120, https://doi.org/10.1016/S0065-3160(08)60006-4

[182] M. Sánchez, N. Aouina, D. Rose, P. Rousseau, H. Takenouti, and V. Vivier, ‘Assessment of the electrochemical microcell geometry by local electrochemical impedance spectroscopy of copper corrosion’, Electrochimica Acta, vol. 62, pp. 276–281, Feb. 2012, doi: 10.1016/j.electacta.2011.12.041.

[183] F. Arjmand and A. Adriaens, ‘Investigation of 304L stainless steel in a NaCl solution using a microcapillary electrochemical droplet cell: Comparison with conventional electrochemical techniques’, Electrochimica Acta, vol. 59, pp. 222–227, Jan. 2012, doi: 10.1016/j.electacta.2011.10.065.

[184] F. Arjmand and A. Adriaens, ‘Influence of pH and Chloride Concentration on the Corrosion Behavior of Unalloyed Copper in NaCl Solution: A Comparative Study Between the Micro and Macro Scales’, Materials, vol. 5, no. 12, pp. 2439–2464, Dec. 2012, doi: 10.3390/ma5122439.

[185] J.-B. Jorcin, H. Krawiec, N. Pébère, and V. Vignal, ‘Comparison of local electrochemical impedance measurements derived from bi-electrode and microcapillary techniques’, Electrochimica Acta, vol. 54, no. 24, pp. 5775–5781, Oct. 2009, doi: 10.1016/j.electacta.2009.05.029.

[186] M. M. Lohrengel, ‘Interface and volume effects in biological cells and electrochemical microcells’, Electrochimica Acta, vol. 42, no. 20, pp. 3265–3271, Jan. 1997, doi: 10.1016/S0013- 4686(97)00177-1.

155

[187] A. W. Hassel and M. M. Lohrengel, ‘The scanning droplet cell and its application to structured nanometer oxide films on aluminium’, Electrochimica Acta, vol. 42, no. 20, pp. 3327–3333, Jan. 1997, doi: 10.1016/S0013-4686(97)00184-9.

[188] H. Krawiec, ‘Application of the Electrochemical Microcell Technique in Solid State Surface Analysis’, SSP, vol. 227, pp. 549–552, Jan. 2015, doi: 10.4028/www.scientific.net/SSP.227.549.

[189] M. Schneider, ‘Buytaert G, Premendra P, de Wit JHW, Katgerman L, Kering B, Brinkman HJ, Terryn H, Surf Coat Technol 201: 4553–4560’. 2007

[190] L. Paussa, F. Andreatta, N. C. Rosero Navarro, A. Durán, and L. Fedrizzi, ‘Study of the effect of cerium nitrate on AA2024-T3 by means of electrochemical micro-cell technique’, Electrochimica Acta, vol. 70, pp. 25–33, May 2012, doi: 10.1016/j.electacta.2012.02.099.

[191] H. Krawiec, V. Vignal, and R. Akid, ‘Numerical modelling of the electrochemical behaviour of 316 stainless steel based upon static and dynamic experimental microcapillary-based techniques: effect of electrolyte flow and capillary size’, Surface and Interface Analysis, vol. 40, no. 3–4, pp. 315–319, 2008, doi: 10.1002/sia.2753.

[192] S. M. Gateman, L. I. Stephens, S. C. Perry, R. Lacasse, R. Schulz, and J. Mauzeroll, ‘The role of titanium in the initiation of localized corrosion of stainless steel 444’, npj Mater Degrad, vol. 2, no. 1, p. 5, Dec. 2018, doi: 10.1038/s41529-018-0026-5.

[193] C. Stromberg, P. Thissen, I. Klueppel, N. Fink, and G. Grundmeier, ‘Synthesis and characterisation of surface gradient thin conversion films on zinc coated steel’, Electrochimica Acta, vol. 52, no. 3, pp. 804–815, Nov. 2006, doi: 10.1016/j.electacta.2006.06.014.

[194] N. Birbilis, B. N. Padgett, and R. G. Buchheit, ‘Limitations in microelectrochemical capillary cell testing and transformation of electrochemical transients for acquisition of microcell impedance data’, Electrochimica Acta, vol. 50, no. 16–17, pp. 3536–3544, May 2005, doi: 10.1016/j.electacta.2005.01.010.

[195] M. M. Lohrengel, S. Heiroth, K. Kluger, M. Pilaski, and B. Walther, ‘Microimpedance— Localized material analysis’, Electrochimica Acta, vol. 51, no. 8–9, pp. 1431–1436, Jan. 2006, doi: 10.1016/j.electacta.2005.02.120.

[196] L. Staemmler, T. Suter, and H. Böhni, ‘Nanolithography by Means of an Electrochemical Scanning Capillary Microscope’, J. Electrochem. Soc., vol. 151, no. 11, pp. G734–G739, Jan. 2004, doi: 10.1149/1.1803834.

[197] T. Suter and H. Böhni, ‘Microelectrodes for corrosion studies in microsystems’, Electrochimica Acta, vol. 47, no. 1–2, pp. 191–199, Sep. 2001, doi: 10.1016/S0013-4686(01)00551-5.

[198] H. Krawiec, V. Vignal, and R. Akid, ‘Numerical modelling of the electrochemical behaviour of 316L stainless steel based upon static and dynamic experimental microcapillary-based techniques’, Electrochimica Acta, vol. 53, no. 16, pp. 5252–5259, Jun. 2008, doi: 10.1016/j.electacta.2008.02.063.

[199] O. Guseva, P. Schmutz, T. Suter, and O. von Trzebiatowski, ‘Modelling of anodic dissolution of pure aluminium in sodium chloride’, Electrochimica Acta, vol. 54, no. 19, pp. 4514–4524, Jul. 2009, doi: 10.1016/j.electacta.2009.03.048.

156

[200] L. C. Abodi, J. A. DeRose, S. Van Damme, A. Demeter, T. Suter, and J. Deconinck, ‘Modeling localized aluminum alloy corrosion in chloride solutions under non-equilibrium conditions: Steps toward understanding pitting initiation’, Electrochimica Acta, vol. 63, pp. 169–178, Feb. 2012, doi: 10.1016/j.electacta.2011.12.074.

[201] M. Andretta ‘Discovering sequence motifes in quantitative and qualitive pepetide data’, Thechnical University of Denmark, p. 105, 2012

[202] T. Suter and R. C. Alkire, ‘Microelectrochemical Studies of Pit Initiation at Single Inclusions in Al 2024-T3’, J. Electrochem. Soc., vol. 148, no. 1, pp. B36–B42, Jan. 2001, doi: 10.1149/1.1344530.

[203] S. O. Klemm, J.-C. Schauer, B. Schuhmacher, and A. W. Hassel, ‘A microelectrochemical scanning flow cell with downstream analytics’, Electrochimica Acta, vol. 56, no. 11, pp. 4315– 4321, Apr. 2011, doi: 10.1016/j.electacta.2011.01.052.

[204] V. Vignal, H. Krawiec, O. Heintz, and R. Oltra, ‘The use of local electrochemical probes and surface analysis methods to study the electrochemical behaviour and pitting corrosion of stainless steels’, Electrochimica Acta, vol. 52, no. 15, pp. 4994–5001, Apr. 2007, doi: 10.1016/j.electacta.2007.01.079.

[205] F. Andreatta et al., ‘Localized corrosion inhibition by cerium species on clad AA2024 aluminium alloy investigated by means of electrochemical micro-cell’, Corrosion Science, vol. 65, pp. 376–386, Dec. 2012, doi: 10.1016/j.corsci.2012.08.053.

[206] J. A. Richardson and G. C. Wood, ‘A study of the pitting corrosion of Al byscanning electron microscopy’, Corrosion Science, vol. 10, no. 5, pp. 313–323, Jan. 1970, doi: 10.1016/S0010- 938X(70)80023-3.

[207] H. Park and J. A. Szpunar, ‘The role of texture and morphology in optimizing the corrosion resistance of zinc-based electrogalvanized coatings’, Corrosion Science, vol. 40, no. 4–5, pp. 525– 545, Apr. 1998, doi: 10.1016/S0010-938X(97)00148-0.

[208] P. Schmutz and G. S. Frankel, ‘Characterization of AA2024‐T3 by Scanning Kelvin Probe Force Microscopy’, J. Electrochem. Soc., vol. 145, no. 7, pp. 2285–2295, Jan. 1998, doi: 10.1149/1.1838633.

[209] R. F. Ashton and M. T. Hepworth, ‘Effect of crystal orientation on the anodic polarization and passivity of zinc’, Corrosion, vol. 24, no. 2, 1968, https://doi.org/10.5006/0010-9312-24.2.50

[210] C. J. Park, M. M. Lohrengel, T. Hamelmann, M. Pilaski, and H. S. Kwon, ‘Grain-dependent passivation of surfaces of polycrystalline zinc’, Electrochimica Acta, vol. 47, no. 21, pp. 3395– 3399, Aug. 2002, doi: 10.1016/S0013-4686(02)00221-9.

[211] H. Fry and M. Whitaker, ‘The Anodic Oxidation of Zinc and a Method of Altering the Characteristics of the Anodic Films’, J. Electrochem. Soc., vol. 106, no. 7, pp. 606–611, Jan. 1959, doi: 10.1149/1.2427446.

[212] J. C. Scully, ‘The fundamentals of corrosion’. Oxford; New York: Pergamon Press, 1990, ISBN 0080378757

[213] J. O. Bockris and S. U. M. Khan, Surface Electrochemistry: A Molecular Level Approach. Springer US, 1993, DOI: 10.1007/978-1-4615-3040-4

157

[214] H. M. Kim and J. A. Szpunar, ‘The Effect of Texture and Grain Boundary Structure on the Intergranular Corrosion of Stainless Steel’, Materials Science Forum, 1994. [Online]. Available: https://www.scientific.net/MSF.157-162.1997. [Accessed: 15-Nov-2019].

[215] L. L., H. Takechi, M. Matsuo, K. Kawasaki, T.H.E. Tamura (74th edn.), Proc. 6th Int’l. Conf. on Textures of Materials, ISIJ, Vol. 2 (1981), p. 210 Tokyo, Japan.

[216] L. L., Raichevski, Vitkova Zashchita Metalloc, 1973, 11, 418

[217] Richard F. Ashton, Malcolm T. Hepworth, ‘Effect of crystal orientation on the anodic polarization and passivity of zinc’, Corrosion, 1968;24(2):50-53, https://doi.org/10.5006/0010- 9312-24.2.50

[218] G. Buytaert et al., ‘Electrochemical investigation of rolled-in subsurface layers in commercially pure aluminium alloys with the micro-capillary cell technique’, Surface and Coatings Technology, vol. 201, no. 8, pp. 4553–4560, Jan. 2007, doi: 10.1016/j.surfcoat.2006.09.096.

[219] T. P. Dirkse, ‘Electrolytic Oxidation of Zinc in Alkaline Solutions’, J. Electrochem. Soc., vol. 102, no. 9, pp. 497–501, Jan. 1955, doi: 10.1149/1.2430136.

[220] Ö. Özkanat, B. Salgin, M. Rohwerder, J. M. C. Mol, J. H. W. de Wit, and H. Terryn, ‘Scanning Kelvin Probe Study of (Oxyhydr)oxide Surface of Aluminum Alloy’, J. Phys. Chem. C, vol. 116, no. 2, pp. 1805–1811, Jan. 2012, doi: 10.1021/jp205585u.

[221] R. Hausbrand, M. Stratmann, and M. Rohwerder, ‘The Physical Meaning of Electrode Potentials at Metal Surfaces and Polymer/Metal Interfaces: Consequences for Delamination’, J. Electrochem. Soc., vol. 155, no. 7, pp. C369–C379, Jan. 2008, doi: 10.1149/1.2926589.

[222] S. Maeda, ‘Surface chemistry of galvanized steel sheets relevant to adhesion performance’, Progress in Organic Coatings, vol. 28, no. 4, pp. 227–238, Aug. 1996, doi: 10.1016/0300- 9440(95)00610-9.

[223] M. Kutz, Ed., ‘Handbook of environmental degradation of materials’. Norwich, NY: William Andrew Pub, 2005, ISBN: 0323524729

[224] S. A. Sajjadi, H. R. Ezatpour, and H. Beygi, ‘Microstructure and mechanical properties of Al– Al2O3 micro and nano composites fabricated by stir casting’, Materials Science and Engineering: A, vol. 528, no. 29, pp. 8765–8771, Nov. 2011, doi: 10.1016/j.msea.2011.08.052.

[225] Z. Shi, G. Song, and A. Atrens, ‘Influence of the β phase on the corrosion performance of anodised coatings on magnesium–aluminium alloys’, Corrosion Science, vol. 47, no. 11, pp. 2760– 2777, Nov. 2005, doi: 10.1016/j.corsci.2004.11.004.

[226] Ch. Christoglou, N. Voudouris, G. N. Angelopoulos, M. Pant, and W. Dahl, ‘Deposition of aluminium on magnesium by a CVD process’, Surface and Coatings Technology, vol. 184, no. 2, pp. 149–155, Jun. 2004, doi: 10.1016/j.surfcoat.2003.10.065.

[227] P. Zhu, W. L. You, Q. G. Ren, C. T. Hua, and M. Zhou, ‘Copper coating electrodeposited directly onto AZ31 magnesium alloy’, Anti-Corrosion Methods and Materials, vol. 60, no. 3, pp. 127–133, Jan. 2013, doi: 10.1108/00035591311315247.

158

[228] F. Chen, H. Zhou, B. Yao, Z. Qin, and Q. Zhang, ‘Corrosion resistance property of the ceramic coating obtained through microarc oxidation on the AZ31 magnesium alloy surfaces’, Surface and Coatings Technology, vol. 201, no. 9, pp. 4905–4908, Feb. 2007, doi: 10.1016/j.surfcoat.2006.07.079.

[229] X. B. Chen, N. Birbilis, and T. B. Abbott, ‘Review of Corrosion-Resistant Conversion Coatings for Magnesium and Its Alloys’, CORROSION, vol. 67, no. 3, pp. 035005-1-035005–16, Mar. 2011, doi: 10.5006/1.3563639.

[230] J. Gu, S. Yang, M. Gao, J. Bai, Y. Zhai, and J. Ding, ‘Micropore evolution in additively manufactured aluminum alloys under heat treatment and inter-layer rolling’, Materials & Design, vol. 186, p. 108288, Jan. 2020, doi: 10.1016/j.matdes.2019.108288.

[231] H. Herman, S. Sampath, and R. McCune, ‘Thermal Spray: Current Status and Future Trends’, MRS Bulletin, vol. 25, no. 7, pp. 17–25, Jul. 2000, doi: 10.1557/mrs2000.119.

[232] M. R. Dorfman, ‘Thermal spray basics’, Advanced Materials and Processes, vol. 160, pp. 47– 50, Jul. 2002, https://www.researchgate.net/publication/280014878_Thermal_spray_processes

[233] M. Grujicic, C. L. Zhao, C. Tong, W. S. DeRosset, and D. Helfritch, ‘Analysis of the impact velocity of powder particles in the cold-gas dynamic-spray process’, Materials Science and Engineering: A, vol. 368, no. 1, pp. 222–230, Mar. 2004, doi: 10.1016/j.msea.2003.10.312.

[234] J. Karthikeyan, ‘Cold spray technology: international status and USA efforts’, p. 14, 2004, https://www.asbindustries.com/documents/int_status_report.pdf

[235] H. Y. Lee, S. H. Jung, S. Y. Lee, Y. H. You, and K. H. Ko, ‘Correlation between Al2O3 particles and interface of Al–Al2O3 coatings by cold spray’, Applied Surface Science, vol. 252, no. 5, pp. 1891–1898, Dec. 2005, doi: 10.1016/j.apsusc.2005.03.148.

[236] D. L. Gilmore, R. C. Dykhuizen, R. A. Neiser, M. F. Smith, and T. J. Roemer, ‘Particle velocity and deposition efficiency in the cold spray process’, J Therm Spray Tech, vol. 8, no. 4, pp. 576– 582, Dec. 1999, doi: 10.1361/105996399770350278.

[237] ‘Thermal spray 2004 : advances in technology and application, 10-12 May 2004, Osaka, Japan : proceedings of the International Thermal Spray Conference - PDF Drive’. [Online]. Available: https://www.pdfdrive.com/thermal-spray-2004-advances-in-technology-and- application-10-12-may-2004-osaka-japan-proceedings-of-the-international-thermal-spray- conference-d166856887.html. [Accessed: 10-Dec-2019].

[238] A. P. Alkhimov, V. F. Kosarev, and A. N. Papyrin, ‘Gas-dynamic spraying. An experimental study of the spraying process’, J Appl Mech Tech Phys, vol. 39, no. 2, pp. 318–323, Mar. 1998, doi: 10.1007/BF02468100.

[239] C.-J. Li and W.-Y. Li, ‘Deposition characteristics of titanium coating in cold spraying’, Surface and Coatings Technology, vol. 167, no. 2–3, pp. 278–283, Apr. 2003, doi: 10.1016/S0257- 8972(02)00919-2.

[240] H.-J. Kim, C.-H. Lee, and S.-Y. Hwang, ‘Fabrication of WC–Co coatings by cold spray deposition’, Surface and Coatings Technology, vol. 191, no. 2, pp. 335–340, Feb. 2005, doi: 10.1016/j.surfcoat.2004.04.058.

159

[241] H. Y. Lee, Y. H. Yu, Y. C. Lee, Y. P. Hong, and K. H. Ko, ‘Interfacial studies between cold- sprayed WO3, Y2O3 films and Si substrate’, Applied Surface Science, vol. 227, no. 1, pp. 244–249, Apr. 2004, doi: 10.1016/j.apsusc.2003.11.073.

[242] T. H. Steenkiste, J. R. Smith, and R. E. Teets, ‘Aluminum coatings via kinetic spray with relatively large powder particles’, Surface and Coatings Technology, vol. 154, pp. 237–252, May 2002, doi: 10.1016/S0257-8972(02)00018-X.

[243] B. S. DeForce, T. J. Eden, and J. K. Potter, ‘Cold Spray Al-5% Mg Coatings for the Corrosion Protection of Magnesium Alloys’, J Therm Spray Tech, vol. 20, no. 6, pp. 1352–1358, Dec. 2011, doi: 10.1007/s11666-011-9675-4.

[244] E. Irissou, J.-G. Legoux, B. Arsenault, and C. Moreau, ‘Investigation of Al-Al2O3 Cold Spray Coating Formation and Properties’, J Therm Spray Tech, vol. 16, no. 5–6, pp. 661–668, Dec. 2007, doi: 10.1007/s11666-007-9086-8.

[245] K. Spencer, D. M. Fabijanic, and M.-X. Zhang, ‘The use of Al–Al2O3 cold spray coatings to improve the surface properties of magnesium alloys’, Surface and Coatings Technology, vol. 204, no. 3, pp. 336–344, Oct. 2009, doi: 10.1016/j.surfcoat.2009.07.032.

[246] Y. Tao, T. Xiong, C. Sun, H. Jin, H. Du, and T. Li, ‘Effect of α-Al2O3 on the properties of cold sprayed Al/α-Al2O3 composite coatings on AZ91D magnesium alloy’, Applied Surface Science, vol. 256, no. 1, pp. 261–266, Oct. 2009, doi: 10.1016/j.apsusc.2009.08.012.

[247] Y. Tao et al., ‘Microstructure and corrosion performance of a cold sprayed aluminium coating on AZ91D magnesium alloy’, Corrosion Science, vol. 52, no. 10, pp. 3191–3197, Oct. 2010, doi: 10.1016/j.corsci.2010.05.023.

[248] H. Bu, M. Yandouzi, C. Lu, D. MacDonald, and B. Jodoin, ‘Cold spray blended Al+Mg17Al12 coating for corrosion protection of AZ91D magnesium alloy’, Surface and Coatings Technology, vol. 207, pp. 155–162, Aug. 2012, doi: 10.1016/j.surfcoat.2012.06.050.

[249] H. Singh, T. S. Sidhu, and S. B. S. Kalsi, ‘Cold spray technology: future of coating deposition processes’, Frattura ed Integrità Strutturale, vol. 6, no. 22, pp. 69–84, Sep. 2012, doi: 10.3221/IGF-ESIS.22.08.

[250] S. E. Tinashe, ‘Conceptual design of a low pressure cold gas dynamic spray (LPCGDS) system’, Thesis, 2010, http://hdl.handle.net/10539/8398

[251] R. Morgan, P. Fox, J. Pattison, C. Sutcliffe, and W. O’Neill, ‘Analysis of cold gas dynamically sprayed aluminium deposits’, Materials Letters, vol. 58, no. 7, pp. 1317–1320, Mar. 2004, doi: 10.1016/j.matlet.2003.09.048.

[252] K. Balani, T. Laha, A. Agarwal, J. Karthikeyan, and N. Munroe, ‘Effect of carrier gases on microstructural and electrochemical behavior of cold-sprayed 1100 aluminum coating’, Surface and Coatings Technology, vol. 195, no. 2, pp. 272–279, May 2005, doi: 10.1016/j.surfcoat.2004.06.028.

[253] W. B. Choi et al., ‘Integrated characterization of cold sprayed aluminum coatings’, Acta Materialia, vol. 55, no. 3, pp. 857–866, Feb. 2007, doi: 10.1016/j.actamat.2006.09.006.

160

[254] M. Hishida, M. Fujita, and K. Sakaki, ‘Fabrication of Aluminum Coating with Dispersed Nanoscale Quasicrystalline Particles by Cold Spray’, Journal of the Japan Institute of Metals and Materials, vol. 73, no. 6, pp. 421–428, 2009, doi: 10.2320/jinstmet.73.421.

[255] M. M. Sharma, T. J. Eden, and B. T. Golesich, ‘Effect of Surface Preparation on the Microstructure, Adhesion, and Tensile Properties of Cold-Sprayed Aluminum Coatings on AA2024 Substrates’, J Therm Spray Tech, vol. 24, no. 3, pp. 410–422, Feb. 2015, doi: 10.1007/s11666-014- 0175-1.

[256] L. Ajdelsztajn, A. Zúñiga, B. Jodoin, and E. J. Lavernia, ‘Cold gas dynamic spraying of a high temperature Al alloy’, Surface and Coatings Technology, vol. 201, no. 6, pp. 2109–2116, Dec. 2006, doi: 10.1016/j.surfcoat.2005.06.001.

[257] T. Suhonen, T. Varis, S. Dosta, M. Torrell, and J. M. Guilemany, ‘Residual stress development in cold sprayed Al, Cu and Ti coatings’, Acta Materialia, vol. 61, no. 17, pp. 6329–6337, Oct. 2013, doi: 10.1016/j.actamat.2013.06.033.

[258] M. R. Rokni, C. A. Widener, and V. R. Champagne, ‘Microstructural Evolution of 6061 Aluminum Gas-Atomized Powder and High-Pressure Cold-Sprayed Deposition’, J Therm Spray Tech, vol. 23, no. 3, pp. 514–524, Feb. 2014, doi: 10.1007/s11666-013-0049-y.

[259] X. Jiang, Y. Wan, H. Herman, and S. Sampath, ‘Role of condensates and adsorbates on substrate surface on fragmentation of impinging molten droplets during thermal spray’, Thin Solid Films, vol. 385, no. 1, pp. 132–141, Apr. 2001, doi: 10.1016/S0040-6090(01)00769-6.

[260] R. S. C. Paredes, S. C. Amico, and A. S. C. M. d’Oliveira, ‘The effect of roughness and preheating of the substrate on the morphology of aluminium coatings deposited by thermal spraying’, Surface and Coatings Technology, vol. 200, no. 9, pp. 3049–3055, Feb. 2006, doi: 10.1016/j.surfcoat.2005.02.200.

[261] C.-J. Li and J.-L. Li, ‘Evaporated-gas-induced splashing model for splat formation during plasma spraying’, Surface and Coatings Technology, vol. 184, no. 1, pp. 13–23, Jun. 2004, doi: 10.1016/j.surfcoat.2003.10.048.

[262] ‘Cold Spraying Combined to Laser Surface Pre-treatment using PROTAL® - Thermal Spray Society’. [Online]. Available: https://www.asminternational.org/web/tss/technical-resources/- /journal_content/56/10192/CP2009ITSC1151/PUBLICATION;jsessionid=20F8299B6DE86AFF1221 F26279FD8144?p_p_id=101&p_p_lifecycle=0&p_p_state=maximized&p_p_mode=view. [Accessed: 09-Dec-2019].

[263] A. C. Hall, D. J. Cook, R. A. Neiser, T. J. Roemer, and D. A. Hirschfeld, ‘The effect of a simple annealing heat treatment on the mechanical properties of cold-sprayed aluminum’, J Therm Spray Tech, vol. 15, no. 2, pp. 233–238, Jun. 2006, doi: 10.1361/105996306X108138.

[264] W.-Y. Li, X. P. Guo, C. Verdy, L. Dembinski, H. L. Liao, and C. Coddet, ‘Improvement of microstructure and property of cold-sprayed Cu–4at.%Cr–2at.%Nb alloy by heat treatment’, Scripta Materialia, vol. 55, no. 4, pp. 327–330, Aug. 2006, doi: 10.1016/j.scriptamat.2006.04.041.

[265] X.-M. Meng, J.-B. Zhang, W. Han, J. Zhao, and Y.-L. Liang, ‘Influence of annealing treatment on the microstructure and mechanical performance of cold sprayed 304 stainless steel coating’, Applied Surface Science, vol. 258, no. 2, pp. 700–704, Nov. 2011, doi: 10.1016/j.apsusc.2011.07.107.

161

[266] L. A. Malik, ‘ASM Metals HandBook Volume 13 Corrosion’, Materials, ISBN: 978-0-87170- 707-9

[267] A. F. Harris and A. Beevers, ‘The effects of grit-blasting on surface properties for adhesion’, International Journal of Adhesion and Adhesives, vol. 19, no. 6, pp. 445–452, Dec. 1999, doi: 10.1016/S0143-7496(98)00061-X.

[268] P. Fauchais, A. Vardelle, M. Vardelle, and M. Fukumoto, ‘Knowledge concerning splat formation: An invited review’, J Therm Spray Tech, vol. 13, no. 3, pp. 337–360, Sep. 2004, doi: 10.1361/10599630419670.

[269] P. Białucki and S. Kozerski, ‘Study of adhesion of different plasma-sprayed coatings to aluminium’, Surface and Coatings Technology, vol. 201, no. 5, pp. 2061–2064, Oct. 2006, doi: 10.1016/j.surfcoat.2006.04.043.

[270] R. Drehmann et al., ‘Splat Formation and Adhesion Mechanisms of Cold Gas-Sprayed Al Coatings on Al2O3 Substrates’, J Therm Spray Tech, vol. 23, no. 1, pp. 68–75, Jan. 2014, doi: 10.1007/s11666-013-9966-z.

[271] R. Ly, K. T. Hartwig, and H. Castaneda, ‘Effects of strain localization on the corrosion behavior of ultra-fine grained aluminum alloy AA6061’, Corrosion Science, vol. 139, pp. 47–57, Jul. 2018, doi: 10.1016/j.corsci.2018.04.023.

[272] M. A. Meyers, A. Mishra, and D. J. Benson, ‘Mechanical properties of nanocrystalline materials’, Progress in Materials Science, vol. 51, no. 4, pp. 427–556, May 2006, doi: 10.1016/j.pmatsci.2005.08.003.

[273] M. R. Rokni, C. A. Widener, V. K. Champagne, G. A. Crawford, and S. R. Nutt, ‘The effects of heat treatment on 7075 Al cold spray deposits’, Surface and Coatings Technology, vol. 310, pp. 278–285, Jan. 2017, doi: 10.1016/j.surfcoat.2016.10.064.

[274] M. R. Rokni, C. A. Widener, and G. A. Crawford, ‘Microstructural evolution of 7075 Al gas atomized powder and high-pressure cold sprayed deposition’, Surface and Coatings Technology, vol. 251, pp. 254–263, Jul. 2014, doi: 10.1016/j.surfcoat.2014.04.035.

[275] M. R. Rokni, C. A. Widener, G. A. Crawford, and M. K. West, ‘An investigation into microstructure and mechanical properties of cold sprayed 7075 Al deposition’, Materials Science and Engineering: A, vol. 625, pp. 19–27, Feb. 2015, doi: 10.1016/j.msea.2014.11.059.

[276] ‘The Cold Spray Materials Deposition Process | ScienceDirect’. [Online]. Available: https://www.sciencedirect.com/book/9781845691813/the-cold-spray-materials-deposition- process. [Accessed: 03-Dec-2019].

[277] R. Valiev, ‘Nanostructuring of metals by severe plastic deformation for advanced properties’, Nature Mater, vol. 3, no. 8, pp. 511–516, Aug. 2004, doi: 10.1038/nmat1180.

[278] J. Wang, L. Nong, Y. Wei, Y. Tang, J. Wang, and C. Wang, ‘Association of interleukin-8 gene polymorphisms with the risk of hepatocellular carcinoma’, Mol Biol Rep, vol. 41, no. 3, pp. 1483– 1489, Mar. 2014, doi: 10.1007/s11033-013-2993-5.

[279] A. N. Ryabinin, ‘Calculation of the particle velocity in cold spray in the one-dimensional non- isentropic approach’, vol. 10, no. 6, p. 5, 2015,

162

https://www.researchgate.net/publication/281927344_Calculation_of_the_particle_velocity_in _cold_spray_in_the_one-dimensional_non-isentropic_approach

[280] V. K. Champagne, D. J. Helfritch, M. D. Trexler, and B. M. Gabriel, ‘The effect of cold spray impact velocity on deposit hardness’, Modelling Simul. Mater. Sci. Eng., vol. 18, no. 6, p. 065011, Aug. 2010, doi: 10.1088/0965-0393/18/6/065011.

[281] D. Poirier, J.-G. Legoux, R. A. L. Drew, and R. Gauvin, ‘Consolidation of Al2O3/Al Nanocomposite Powder by Cold Spray’, J Therm Spray Tech, vol. 20, no. 1, pp. 275–284, Jan. 2011, doi: 10.1007/s11666-010-9595-8.

[282] L. Wang, M. Strangwood, D. Balint, J. Lin, and T. A. Dean, ‘Formability and failure mechanisms of AA2024 under hot forming conditions’, Materials Science and Engineering: A, vol. 528, no. 6, pp. 2648–2656, Mar. 2011, doi: 10.1016/j.msea.2010.11.084.

[283] H. Gao, T. Weng, J. Liu, C. Li, Z. Li, and L. Wang, ‘Hot stamping of an Al-Li alloy: a feasibility study’, Manufacturing Rev., vol. 3, p. 9, 2016, doi: 10.1051/mfreview/2016006.

[284] M. S. Mohamed, A. D. Foster, J. Lin, D. S. Balint, and T. A. Dean, ‘Investigation of deformation and failure features in hot stamping of AA6082: Experimentation and modelling’, International Journal of Machine Tools and Manufacture, vol. 53, no. 1, pp. 27–38, Feb. 2012, doi: 10.1016/j.ijmachtools.2011.07.005.

[285] K. Zheng, Y. Dong, D. Zheng, J. Lin, and T. A. Dean, ‘An experimental investigation on the deformation and post-formed strength of heat-treatable aluminium alloys using different elevated temperature forming processes’, Journal of Materials Processing Technology, vol. 268, pp. 87–96, Jun. 2019, doi: 10.1016/j.jmatprotec.2018.11.042.

[286] B. Marzbanrad, H. Jahed, and E. Toyserkani, ‘On the evolution of substrate’s residual stress during cold spray process: A parametric study’, Materials & Design, vol. 138, Oct. 2017, doi: 10.1016/j.matdes.2017.10.062.

[287] T. S. Price, P. H. Shipway, and D. G. McCartney, ‘Effect of cold spray deposition of a titanium coating on fatigue behavior of a titanium alloy’, J Therm Spray Tech, vol. 15, no. 4, pp. 507–512, Dec. 2006, doi: 10.1361/105996306X147108.

[288] W. Li, K. Yang, D. Zhang, X. Zhou, and X. Guo, ‘Interface behavior of particles upon impacting during cold spraying of Cu/Ni/Al mixture’, Materials & Design, vol. 95, pp. 237–246, Apr. 2016, doi: 10.1016/j.matdes.2016.01.122.

[289] K. Spencer, V. Luzin, N. Matthews, and M.-X. Zhang, ‘Residual stresses in cold spray Al coatings: The effect of alloying and of process parameters’, Surface and Coatings Technology, vol. 206, no. 19, pp. 4249–4255, May 2012, doi: 10.1016/j.surfcoat.2012.04.034.

[290] R. Ghelichi, D. MacDonald, S. Bagherifard, H. Jahed, M. Guagliano, and B. Jodoin, ‘Microstructure and fatigue behavior of cold spray coated Al5052’, Acta Materialia, vol. 60, no. 19, pp. 6555–6561, Nov. 2012, doi: 10.1016/j.actamat.2012.08.020.

[291] P. Cavaliere and A. Silvello, ‘Processing conditions affecting residual stresses and fatigue properties of cold spray deposits’, Int J Adv Manuf Technol, vol. 81, no. 9, pp. 1857–1862, Dec. 2015, doi: 10.1007/s00170-015-7365-y.

163

[292] V. Luzin, K. Spencer, and M.-X. Zhang, ‘Residual stress and thermo-mechanical properties of cold spray metal coatings’, Acta Materialia, vol. 59, no. 3, pp. 1259–1270, Feb. 2011, doi: 10.1016/j.actamat.2010.10.058.

[293] S. Sampath, X. Y. Jiang, J. Matejicek, L. Prchlik, A. Kulkarni, and A. Vaidya, ‘Role of thermal spray processing method on the microstructure, residual stress and properties of coatings: an integrated study for Ni–5 wt.%Al bond coats’, Materials Science and Engineering: A, vol. 364, no. 1, pp. 216–231, Jan. 2004, doi: 10.1016/j.msea.2003.08.023.

[294] M. Diab, X. Pang, and H. Jahed, ‘The effect of pure aluminum cold spray coating on corrosion and corrosion fatigue of magnesium (3% Al-1% Zn) extrusion’, Surface and Coatings Technology, vol. 309, pp. 423–435, Jan. 2017, doi: 10.1016/j.surfcoat.2016.11.014.

[295] Y. Zhang, J. Xu, and T. Zhai, ‘Distributions of pore size and fatigue weak link strength in an A713 sand cast aluminum alloy’, Materials Science and Engineering: A, vol. 527, no. 16, pp. 3639–3644, Jun. 2010, doi: 10.1016/j.msea.2010.03.104.

[296] D. E. J. Talbot, ‘Effects of Hydrogen in Aluminium, Magnesium, Copper, and Their Alloys’, International Metallurgical Reviews, vol. 20, no. 1, pp. 166–184, Jan. 1975, doi: 10.1179/imtlr.1975.20.1.166.

[297] ASM International, Ed., ASM handbook, 10th editon. Materials Park, Ohio: ASM International, 1990, ISBN: 9780871703774

[298] H. Mayer, M. Papakyriacou, B. Zettl, and S. E. Stanzl-Tschegg, ‘Influence of porosity on the fatigue limit of die cast magnesium and aluminium alloys’, International Journal of Fatigue, vol. 25, no. 3, pp. 245–256, Mar. 2003, doi: 10.1016/S0142-1123(02)00054-3.

[299] T. A. Rodrigues, V. Duarte, J. A. Avila, T. G. Santos, R. M. Miranda, and J. P. Oliveira, ‘Wire and arc additive manufacturing of HSLA steel: Effect of thermal cycles on microstructure and mechanical properties’, Additive Manufacturing, vol. 27, pp. 440–450, May 2019, doi: 10.1016/j.addma.2019.03.029.

[300] H. Attar, M. J. Bermingham, S. Ehtemam-Haghighi, A. Dehghan-Manshadi, D. Kent, and M. S. Dargusch, ‘Evaluation of the mechanical and wear properties of titanium produced by three different additive manufacturing methods for biomedical application’, Materials Science and Engineering: A, vol. 760, pp. 339–345, Jul. 2019, doi: 10.1016/j.msea.2019.06.024.

[301] C. E. Seow, H. E. Coules, G. Wu, R. H. U. Khan, X. Xu, and S. Williams, ‘Wire + Arc Additively Manufactured Inconel 718: Effect of post-deposition heat treatments on microstructure and tensile properties’, Materials & Design, vol. 183, p. 108157, Dec. 2019, doi: 10.1016/j.matdes.2019.108157.

[302] J. P. Oliveira, T. G. Santos, and R. M. Miranda, ‘Revisiting fundamental welding concepts to improve additive manufacturing: From theory to practice’, Progress in Materials Science, vol. 107, p. 100590, Jan. 2020, doi: 10.1016/j.pmatsci.2019.100590.

[303] Y. Zou, D. Goldbaum, J. A. Szpunar, and S. Yue, ‘Microstructure and nanohardness of cold- sprayed coatings: Electron backscattered diffraction and nanoindentation studies’, Scripta Materialia, vol. 62, no. 6, pp. 395–398, Mar. 2010, doi: 10.1016/j.scriptamat.2009.11.034.

[304] Y. Zou, W. Qin, E. Irissou, J.-G. Legoux, S. Yue, and J. A. Szpunar, ‘Dynamic recrystallization in the particle/particle interfacial region of cold-sprayed nickel coating: Electron backscatter

164

diffraction characterization’, Scripta Materialia, vol. 61, no. 9, pp. 899–902, Nov. 2009, doi: 10.1016/j.scriptamat.2009.07.020.

[305] M. R. Rokni, C. A. Widener, and V. R. Champagne, ‘Microstructural stability of ultrafine grained cold sprayed 6061 aluminum alloy’, Applied Surface Science, vol. 290, pp. 482–489, Jan. 2014, doi: 10.1016/j.apsusc.2013.11.127.

[306] R. Z. Valiev, Y. Estrin, Z. Horita, T. G. Langdon, M. J. Zechetbauer, and Y. T. Zhu, ‘Producing bulk ultrafine-grained materials by severe plastic deformation’, JOM, vol. 58, no. 4, pp. 33–39, Apr. 2006, doi: 10.1007/s11837-006-0213-7.

[307] J. Jiang, Y. Ding, F. Zuo, and A. Shan, ‘Mechanical properties and microstructures of ultrafine- grained pure aluminum by asymmetric rolling’, Scripta Materialia, vol. 60, no. 10, pp. 905–908, May 2009, doi: 10.1016/j.scriptamat.2009.02.016.

[308] M. R. Rokni, C. A. Widener, A. T. Nardi, and V. K. Champagne, ‘Nano crystalline high energy milled 5083 Al powder deposited using cold spray’, Applied Surface Science, vol. 305, pp. 797– 804, Jun. 2014, doi: 10.1016/j.apsusc.2014.04.010.

[309] A. V. Sergueeva, V. V. Stolyarov, R. Z. Valiev, and A. K. Mukherjee, ‘Advanced mechanical properties of pure titanium with ultrafine grained structure’, Scripta Materialia, vol. 45, no. 7, pp. 747–752, Oct. 2001, doi: 10.1016/S1359-6462(01)01089-2.

[310] H. Y. Lee, Y. H. Yu, Y. C. Lee, Y. P. Hong, and K. H. Ko, ‘Cold spray of SiC and Al2O3 with soft metal incorporation: A technical contribution’, J Therm Spray Tech, vol. 13, no. 2, pp. 184–189, Jun. 2004, doi: 10.1361/10599630419355.

[311] A. P. Sannino and H. J. Rack, ‘Dry sliding wear of discontinuously reinforced aluminum composites: review and discussion’, Wear, vol. 189, no. 1, pp. 1–19, Oct. 1995, doi: 10.1016/0043-1648(95)06657-8.

[312] T. Miyajima and Y. Iwai, ‘Effects of reinforcements on sliding wear behavior of aluminum matrix composites’, Wear, vol. 255, no. 1, pp. 606–616, Aug. 2003, doi: 10.1016/S0043- 1648(03)00066-8.

[313] Q. Wang, K. Spencer, N. Birbilis, and M.-X. Zhang, ‘The influence of ceramic particles on bond strength of cold spray composite coatings on AZ91 alloy substrate’, Surface and Coatings Technology, vol. 205, no. 1, pp. 50–56, Sep. 2010, doi: 10.1016/j.surfcoat.2010.06.008.

[314] I. A. Ibrahim, F. A. Mohamed, and E. J. Lavernia, ‘Particulate reinforced metal matrix composites — a review’, J Mater Sci, vol. 26, no. 5, pp. 1137–1156, Mar. 1991, doi: 10.1007/BF00544448.

[315] N. Kang et al., ‘Selective laser melting of tungsten carbide reinforced maraging steel composite’, Additive Manufacturing, vol. 22, pp. 104–110, Aug. 2018, doi: 10.1016/j.addma.2018.04.031.

[316] B. AlMangour, D. Grzesiak, and J.-M. Yang, ‘Rapid fabrication of bulk-form TiB2/316L stainless steel nanocomposites with novel reinforcement architecture and improved performance by selective laser melting’, Journal of Alloys and Compounds, vol. 680, pp. 480–493, Sep. 2016, doi: 10.1016/j.jallcom.2016.04.156.

165

[317] H. Proudhon, J. Savkova, S. Basseville, V. Guipont, M. Jeandin, and G. Cailletaud, ‘Experimental and numerical wear studies of porous Reactive Plasma Sprayed Ti–6Al–4V/TiN composite coating’, Wear, vol. 311, no. 1, pp. 159–166, Mar. 2014, doi: 10.1016/j.wear.2014.01.012.

[318] B. Ralph, H. C. Yuen, and W. B. Lee, ‘The processing of metal matrix composites — an overview’, Journal of Materials Processing Technology, vol. 63, no. 1, pp. 339–353, Jan. 1997, doi: 10.1016/S0924-0136(96)02645-3.

[319] Q. Han, Y. Geng, R. Setchi, F. Lacan, D. Gu, and S. L. Evans, ‘Macro and nanoscale wear behaviour of Al-Al2O3 nanocomposites fabricated by selective laser melting’, Composites Part B: Engineering, vol. 127, pp. 26–35, Oct. 2017, doi: 10.1016/j.compositesb.2017.06.026.

[320] A. Shkodkin, A. Kashirin, O. Klyuev, and T. Buzdygar, ‘Metal particle deposition stimulation by surface abrasive treatment in gas dynamic spraying’, J Therm Spray Tech, vol. 15, no. 3, pp. 382– 386, Sep. 2006, doi: 10.1361/105996306X124383.

[321] W.-Y. Li et al., ‘Significant influences of metal reactivity and oxide films at particle surfaces on coating microstructure in cold spraying’, Applied Surface Science, vol. 253, no. 7, pp. 3557–3562, Jan. 2007, doi: 10.1016/j.apsusc.2006.07.063.

[322] J. Zhang and D. Kong, ‘Effect of laser remelting on microstructure and immersion corrosion of cold–sprayed aluminum coating on S355 structural steel’, Optics & Laser Technology, vol. 106, pp. 348–356, Oct. 2018, doi: 10.1016/j.optlastec.2018.04.026.

[323] V. Cannillo, A. Sola, M. Barletta, and A. Gisario, ‘Surface modification of Al–Al2O3 composites by laser treatment’, Optics and Lasers in Engineering, vol. 48, no. 12, pp. 1266–1277, Dec. 2010, doi: 10.1016/j.optlaseng.2010.06.004.

[324] N. Kang et al., ‘Effects of laser remelting process on the microstructure, roughness and microhardness of in-situ cold sprayed hypoeutectic Al-Si coating’, Surface and Coatings Technology, vol. 318, pp. 355–359, May 2017, doi: 10.1016/j.surfcoat.2017.01.057.

[325] T. Marrocco, T. Hussain, D. G. McCartney, and P. H. Shipway, ‘Corrosion Performance of Laser Posttreated Cold Sprayed Titanium Coatings’, J Therm Spray Tech, vol. 20, no. 4, pp. 909–917, Jun. 2011, doi: 10.1007/s11666-011-9637-x.

[326] A. Sova, S. Grigoriev, A. Okunkova, and I. Smurov, ‘Cold spray deposition of 316L stainless steel coatings on aluminium surface with following laser post-treatment’, Surface and Coatings Technology, vol. 235, pp. 283–289, Nov. 2013, doi: 10.1016/j.surfcoat.2013.07.052.

[327] M. Kulmala and P. Vuoristo, ‘Influence of process conditions in laser-assisted low-pressure cold spraying’, Surface and Coatings Technology, vol. 202, no. 18, pp. 4503–4508, Jun. 2008, doi: 10.1016/j.surfcoat.2008.04.034.

[328] R. Huang, M. Sone, W. Ma, and H. Fukanuma, ‘The effects of heat treatment on the mechanical properties of cold-sprayed coatings’, Surface and Coatings Technology, vol. 261, pp. 278–288, Jan. 2015, doi: 10.1016/j.surfcoat.2014.11.017.

[329] W.-Y. Li, C. Zhang, H. Liao, and C. Coddet, ‘Effect of heat treatment on microstructure and mechanical properties of cold sprayed Ti coatings with relatively large powder particles’, J Coat Technol Res, vol. 6, no. 3, pp. 401–406, Sep. 2009, doi: 10.1007/s11998-008-9133-5.

166

[330] M. R. Rokni, C. A. Widener, V. K. Champagne, and G. A. Crawford, ‘Microstructure and mechanical properties of cold sprayed 7075 deposition during non-isothermal annealing’, Surface and Coatings Technology, vol. 276, pp. 305–315, Aug. 2015, doi: 10.1016/j.surfcoat.2015.07.016.

[331] H. Wan, J. Lin, and J. Min, ‘Effect of laser ablation treatment on corrosion resistance of adhesive-bonded Al alloy joints’, Surface and Coatings Technology, vol. 345, pp. 13–21, Jul. 2018, doi: 10.1016/j.surfcoat.2018.03.087.

[332] A. Sova, A. Papyrin, and I. Smurov, ‘Influence of Ceramic Powder Size on Process of Cermet Coating Formation by Cold Spray’, Journal of Thermal Spray Technology, vol. 18, pp. 633–641, Dec. 2009, doi: 10.1007/s11666-009-9359-5.

[333] J. Lee, S. Shin, H. Kim, and C. Lee, ‘Effect of gas temperature on critical velocity and deposition characteristics in kinetic spraying’, Applied Surface Science, vol. 253, no. 7, pp. 3512– 3520, Jan. 2007, doi: 10.1016/j.apsusc.2006.07.061.

[334] J. Wu, H. Fang, S. Yoon, H. Kim, and C. Lee, ‘Measurement of particle velocity and characterization of deposition in aluminum alloy kinetic spraying process’, Applied Surface Science, vol. 252, no. 5, pp. 1368–1377, Dec. 2005, doi: 10.1016/j.apsusc.2005.02.108.

[335] M. Yandouzi et al., ‘Microstructure and mechanical properties of B4C reinforced Al-based matrix composite coatings deposited by CGDS and PGDS processes’, Surface and Coatings Technology, vol. 205, no. 7, pp. 2234–2246, Dec. 2010, doi: 10.1016/j.surfcoat.2010.08.143.

[336] A. Sova, D. Pervushin, and I. Smurov, ‘Development of multimaterial coatings by cold spray and gas detonation spraying’, Surface and Coatings Technology, vol. 205, no. 4, pp. 1108–1114, Nov. 2010, doi: 10.1016/j.surfcoat.2010.07.092.

[337] P. A. Podrabinnik and I. V. Shishkovsky, ‘Laser Post Annealing of Cold-sprayed Al–Ni Composite Coatings for Green Energy Tasks’, Procedia IUTAM, vol. 23, pp. 108–113, 2017, doi: 10.1016/j.piutam.2017.06.010.

[338] R. M. Souto, M. M. Laz, and R. L. Reis, ‘Degradation characteristics of hydroxyapatite coatings on orthopaedic TiAlV in simulated physiological media investigated by electrochemical impedance spectroscopy’, Biomaterials, vol. 24, no. 23, pp. 4213–4221, Oct. 2003, doi: 10.1016/S0142-9612(03)00362-4.

[339] X. Zhou and P. Mohanty, ‘Corrosion behaviour of cold sprayed titanium coatings in simulated body fluid’, Corrosion Engineering, Science and Technology, vol. 47, no. 2, pp. 145–154, Apr. 2012, doi: 10.1179/1743278211Y.0000000037.

[340] H. Bu, M. Yandouzi, C. Lu, and B. Jodoin, ‘Effect of heat treatment on the intermetallic layer of cold sprayed aluminum coatings on magnesium alloy’, Surface and Coatings Technology, vol. 205, no. 19, pp. 4665–4671, Jun. 2011, doi: 10.1016/j.surfcoat.2011.04.018.

[341] E. Sansoucy, P. Marcoux, L. Ajdelsztajn, and B. Jodoin, ‘Properties of SiC-reinforced aluminum alloy coatings produced by the cold gas dynamic spraying process’, Surface & Coatings Technology - SURF COAT TECH, vol. 202, pp. 3988–3996, May 2008, doi: 10.1016/j.surfcoat.2008.02.017.

167

[342] R. B. Alvarez et al., ‘Corrosion relationships as a function of time and surface roughness on a structural AE44 magnesium alloy’, Corrosion Science, vol. 52, no. 5, pp. 1635–1648, May 2010, doi: 10.1016/j.corsci.2010.01.018.

[343] L. R. Hilbert, D. Bagge-Ravn, J. Kold, and L. Gram, ‘Influence of surface roughness of stainless steel on microbial adhesion and corrosion resistance’, International Biodeterioration & Biodegradation, vol. 52, no. 3, pp. 175–185, Oct. 2003, doi: 10.1016/S0964-8305(03)00104-5.

[344] T. Hong and M. Nagumo, ‘Effect of surface roughness on early stages of pitting corrosion of Type 301 stainless steel’, Corrosion Science, vol. 39, no. 9, pp. 1665–1672, Sep. 1997, doi: 10.1016/S0010-938X(97)00072-3.

[345] D. Krouse, ‘Computer simulation of pitting potential measurements’, Corrosion Science, Volume 47, Issue 12, 2005, Pp. 3140-3177, https://doi.org/10.1016/j.corsci.2005.07.003

[346] Y. Zuo, H. Wang, and J. Xiong, ‘The aspect ratio of surface grooves and metastable pitting of stainless steel’, Corrosion Science, vol. 44, no. 1, pp. 25–35, 2002, https://doi.org/10.1016/S0010-938X(01)00039-7

[347] K. Sasaki and G. T. Burstein, ‘The generation of surface roughness during slurry erosion- corrosion and its effect on the pitting potential’, Corrosion Science, vol. 38, no. 12, pp. 2111– 2120, Dec. 1996, doi: 10.1016/S0010-938X(96)00066-2.

[348] H. Lee, D. Kim, J. Jung, Y. Pyoun, and K. Shin, ‘Influence of peening on the corrosion properties of AISI 304 stainless steel’, Corrosion Science, vol. 51, no. 12, pp. 2826–2830, Dec. 2009, doi: 10.1016/j.corsci.2009.08.008.

[349] G. T. Burstein and S. P. Vines, ‘Repetitive Nucleation of Corrosion Pits on Stainless Steel and the Effects of Surface Roughness’, J. Electrochem. Soc., vol. 148, no. 12, pp. B504–B516, Jan. 2001, doi: 10.1149/1.1416503.

[350] P. C. Pistorius and G. T. Burstein, ‘Growth of corrosion pits on stainless steel in chloride solution containing dilute sulphate’, Corrosion Science, vol. 33, no. 12, pp. 1885–1897, Dec. 1992, doi: 10.1016/0010-938X(92)90191-5.

[351] G. T. Burstein and P. C. Pistorius, ‘Surface Roughness and the Metastable Pitting of Stainless Steel in Chloride Solutions’, Corrosion, vol. 51, no. 5, pp. 380–385, May 1995, doi: 10.5006/1.3293603.

[352] T. Suter, Y. Müller, P. Schmutz, and O. von Trzebiatowski, ‘Microelectrochemical Studies of Pit Initiation on High Purity and Ultra High Purity Aluminum’, Advanced Engineering Materials, vol. 7, no. 5, pp. 339–348, 2005, doi: 10.1002/adem.200500067.

[353] E. Calla, D. G. McCartney, and P. H. Shipway, ‘Effect of deposition conditions on the properties and annealing behavior of cold-sprayed copper’, J Therm Spray Tech, vol. 15, no. 2, pp. 255–262, Jun. 2006, doi: 10.1361/105996306X108192.

[354] M. R. Rokni, C. A. Widener, S. P. Ahrenkiel, B. K. Jasthi, and V. R. Champagne, ‘Annealing behaviour of 6061 aluminium deposited by high pressure cold spray’, Surface Engineering, vol. 30, no. 5, pp. 361–368, Jun. 2014, doi: 10.1179/1743294413Y.0000000209.

168

[355] J. K. Kim, H. G. Jeong, S. I. Hong, Y. S. Kim, and W. J. Kim, ‘Effect of aging treatment on heavily deformed microstructure of a 6061 aluminum alloy after equal channel angular pressing’, Scripta Materialia, vol. 45, no. 8, pp. 901–907, Oct. 2001, doi: 10.1016/S1359-6462(01)01109-5.

[356] B. DeForce, T. J. Eden, and H. W. Pickering, ‘Cold-Sprayed aluminum coatings for magnesium aircraft components’, Materials performance 48(2): pp.40-44Feb. 2009, https://www.researchgate.net/publication/279909925_Cold- Sprayed_Aluminum_Coatings_for_Magnesium_Aircraft_Components

[357] G. S. Frankel, ‘Pitting Corrosion of Metals A Review of the Critical Factors’, J. Electrochem. Soc., vol. 145, no. 6, pp. 2186–2198, Jan. 1998, doi: 10.1149/1.1838615.

[358] G. S. Chen, M. Gao, and R. P. Wei, ‘Microconstituent-Induced Pitting Corrosion in Aluminum Alloy 2024-T3’, CORROSION, vol. 52, no. 1, pp. 8–15, Jan. 1996, doi: 10.5006/1.3292099.

[359] G. Svenningsen, M. H. Larsen, J. H. Nordlien, and K. Nisancioglu, ‘Effect of thermomechanical history on intergranular corrosion of extruded AlMgSi(Cu) model alloy’, Corrosion Science, vol. 48, no. 12, pp. 3969–3987, Dec. 2006, doi: 10.1016/j.corsci.2006.03.018.

[360] G. Svenningsen, M. H. Larsen, J. C. Walmsley, J. H. Nordlien, and K. Nisancioglu, ‘Effect of artificial aging on intergranular corrosion of extruded AlMgSi alloy with small Cu content’, Corrosion Science, vol. 48, no. 6, pp. 1528–1543, Jun. 2006, doi: 10.1016/j.corsci.2005.05.045.

[361] G. Svenningsen, M. H. Larsen, J. H. Nordlien, and K. Nisancioglu, ‘Effect of high temperature heat treatment on intergranular corrosion of AlMgSi(Cu) model alloy’, Corrosion Science, vol. 48, no. 1, pp. 258–272, Jan. 2006, doi: 10.1016/j.corsci.2004.12.003.

[362] G. Svenningsen, J. E. Lein, A. Bjørgum, J. H. Nordlien, Y. Yu, and K. Nisancioglu, ‘Effect of low copper content and heat treatment on intergranular corrosion of model AlMgSi alloys’, Corrosion Science, vol. 48, no. 1, pp. 226–242, Jan. 2006, doi: 10.1016/j.corsci.2004.11.025.

[363] L. Lacroix, L. Ressier, C. Blanc, and G. Mankowski, ‘Combination of AFM, SKPFM, and SIMS to Study the Corrosion Behavior of S-phase particles in AA2024-T351’, J. Electrochem. Soc., vol. 155, no. 4, p. C131, 2008, doi: 10.1149/1.2833315.

[364] L. Lacroix, L. Ressier, C. Blanc, and G. Mankowski, ‘Statistical Study of the Corrosion Behavior of Al2CuMg Intermetallics in AA2024-T351 by SKPFM’, J. Electrochem. Soc., vol. 155, no. 1, pp. C8–C15, Jan. 2008, doi: 10.1149/1.2799089.

[365] C. Blanc, A. Freulon, M.-C. Lafont, Y. Kihn, and G. Mankowski, ‘Modelling the corrosion behaviour of Al2CuMg coarse particles in copper-rich aluminium alloys’, Corrosion Science, vol. 48, no. 11, pp. 3838–3851, Nov. 2006, doi: 10.1016/j.corsci.2006.01.012.

[366] C. Luo, X. Zhou, G. E. Thompson, and A. E. Hughes, ‘Observations of intergranular corrosion in AA2024-T351: The influence of grain stored energy’, Corrosion Science, vol. 61, pp. 35–44, Aug. 2012, doi: 10.1016/j.corsci.2012.04.005.

[367] R. Parvizi et al., ‘Probing corrosion initiation at interfacial nanostructures of AA2024-T3’, Corrosion Science, vol. 116, pp. 98–109, Feb. 2017, doi: 10.1016/j.corsci.2016.12.006.

[368] R. Parvizi, A. E. Hughes, A. M. Glenn, P. Cizek, M. Y. Tan, and M. Forsyth, ‘Role of microstructure in corrosion initiation of a highly-deformed AA2024 wire’, Corrosion Science, vol. 144, pp. 184–197, Nov. 2018, doi: 10.1016/j.corsci.2018.08.052.

169

[369] M. Guérin et al., ‘Identification of the metallurgical parameters explaining the corrosion susceptibility in a 2050 aluminium alloy’, Corrosion Science, vol. 102, pp. 291–300, Jan. 2016, doi: 10.1016/j.corsci.2015.10.020.

[370] X. Zhou et al., ‘Grain-stored energy and the propagation of intergranular corrosion in AA2xxx aluminium alloys’, Surface and Interface Analysis, vol. 45, no. 10, pp. 1543–1547, 2013, doi: 10.1002/sia.5218.

[371] Y. Ma et al., ‘Self-assembly, optical and electrical properties of five membered O- or S- heterocyclic annulated perylene diimides’, Dyes and Pigments, vol. 135, pp. 41–48, Dec. 2016, doi: 10.1016/j.dyepig.2016.06.027.

[372] A. M. Glenn, A. E. Hughes, A. Torpy, G. Nolze, and N. Birbilis, ‘Defect density associated with constituent particles in AA2024‐T3 and its role in corrosion’, 2016, doi: 10.1002/sia.5813.

[373] X. Zhang et al., ‘The influence of grain structure on the corrosion behaviour of 2A97-T3 Al-Cu- Li alloy’, Corrosion Science, vol. 116, pp. 14–21, Feb. 2017, doi: 10.1016/j.corsci.2016.12.005.

[374] C. Luo et al., ‘Characterization of Localized Corrosion in an Al-Cu-Li Alloy’, J. of Materi Eng and Perform, vol. 25, no. 5, pp. 1811–1819, May 2016, doi: 10.1007/s11665-016-2010-y.

[375] X. Zhang et al., ‘The propagation of localized corrosion in Al-Cu-Li alloy’, Surface and Interface Analysis, vol. 48, no. 8, pp. 745–749, 2016, doi: 10.1002/sia.5890.

[376] X. Zhang et al., ‘Intergranular corrosion in AA2024-T3 aluminium alloy: The influence of stored energy and prediction’, Corrosion Science, vol. 155, pp. 1–12, Jul. 2019, doi: 10.1016/j.corsci.2019.04.031.

[377] C. A. Rodopoulos, J. S. Romero, S. A. Curtis, E. R. de los Rios, and P. Peyre, ‘Effect of controlled shot peening and laser shock peening on the fatigue performance of 2024-T351 aluminum alloy’, J. of Materi Eng and Perform, vol. 12, no. 4, pp. 414–419, Aug. 2003, doi: 10.1361/105994903770342944.

[378] H. Amar et al., ‘Influence of the microstructure and laser shock processing (LSP) on the corrosion behaviour of the AA2050-T8 aluminium alloy’, Corrosion Science, vol. 53, no. 10, pp. 3215–3221, Oct. 2011, doi: 10.1016/j.corsci.2011.05.066.

[379] C. Augustin, E. Andrieu, C. Blanc, G. Mankowski, and J. Delfosse, ‘Intergranular Corrosion of 2024 Alloy in Chloride Solutions’, J. Electrochem. Soc., vol. 154, no. 11, p. C637, 2007, doi: 10.1149/1.2778224.

[380] V. Guillaumin and G. Mankowski, ‘Localized corrosion of 2024 T351 aluminium alloy in chloride media’, Corrosion Science, vol. 41, no. 3, pp. 421–438, Mar. 1998, doi: 10.1016/S0010- 938X(98)00116-4.

[381] L. Y. Xu and Y. fei Cheng, ‘Effect of fluid hydrodynamics on flow-assisted corrosion of aluminum alloy in ethylene glycol–water solution studied by a microelectrode technique’, 2009, doi: 10.1016/j.corsci.2009.06.005.

[382] D. Cf, F. Aq, L. Xg, and C. Yf, ‘Localized EIS characterization of corrosion of steel at coating defect under cathodic protection’, Electrochim. Acta, vol. 54, no. 2, pp. 628–633, 2008, https://doi.org/10.1016/j.electacta.2008.07.016

170

[383] C. Zhong, X. Tang, and Y. F. Cheng, ‘Corrosion of steel under the defected coating studied by localized electrochemical impedance spectroscopy’, Electrochimica Acta, vol. 53, no. 14, pp. 4740–4747, https://doi.org/10.1016/j.electacta.2008.02.014

[384] J. V. Nardeli et al., ‘Localised corrosion assessement of crambe-oil-based polyurethane coatings applied on the ASTM 1200 aluminum alloy’, Corrosion Science, vol. 111, pp. 422–435, Oct. 2016, doi: 10.1016/j.corsci.2016.05.034.

[385] ‘THE CHARACTERIZATION OF POROUS ELEdefects 10.1016-0013-4686(81)85072-4 1981ElectrochemActaPoreImpedanceCandy.pdf’. .

[386] Samuel J. Cooper, Antonio Bertei, Donal P. Finegan, Nigel P. Brandon, ‘Simulated impedance of diffusion in porous media’, Electrochimica Acta, 251, pp. 681–689, 2017, http://dx.doi.org/10.1016/j.electacta.2017.07.152

[387] Jean-Baptiste Jorcin, Mark. E. Orazem, NadinePébère, Bernard Tribollet, ‘CPE Analysis by Local Electrochemical Impedance Spectroscopy’, Electrochimica Acta 51(8-9), 2006, DOI: 10.1016/j.electacta.2005.02.128

[388] J. R. Macdonald, ‘Impedance Spectroscopy: Emphasizing Solid Materials and Systems’. New York: John Wiley & Sons, 1987, ISBN-10 : 0471831220

[389] ‘Further to the paper “Application of the impedance model of de Levie for the characterization of porous electrodes” by Barcia’, Electrochimica Acta, 51(10) pp. 2096-2097, 2006, DOI: 10.1016/j.electacta.2005.06.035

[390] Morgane Mokhtari, Takeshi Wada, Christophe Le Bourlot, Nicolas Mary, Jannick Duchet- Rumeau, Hidemi Kato, Eric Maire, ‘Low cost high specific surface architectured nanoporous metal with corrosion resistance produced by liquid metal dealloying from commercial nickel superalloy’, Volume 163, 1 April, Pages 5-8, 2019, https://doi.org/10.1016/j.scriptamat.2018.12.023

[391] H. J. W. Lenderink, M. V. D. Linden, and J. H. W. De Wit, ‘Corrosion of aluminium in acidic and neutral solutions’, Electrochimica Acta, vol. 38, no. 14, pp. 1989–1992, Oct. 1993, doi: 10.1016/0013-4686(93)80329-X.

[392] S. E. Frers, M. M. Stefenel, C. Mayer, and T. Chierchie, ‘AC-Impedance measurements on aluminium in chloride containing solutions and below the pitting potential’, J Appl Electrochem, vol. 20, no. 6, pp. 996–999, Nov. 1990, doi: 10.1007/BF01019578.

[393] L. L. T. Schmidt, F. Gaertner, and H. Kreye, ‘New Developments in Cold Spray Based on Higher Gas and Particle Temperatures’, J. Therm. Spray Technol. 2006, https://doi.org/10.1361/105996306X147144

[394] J. Voyer, T. Stoltenhoff, and T. Schmidt, ‘Method and Potential of the Cold Spray Process’, Journal of Thermal Spray Technology, vol. 15, no. 2, pp. 39–47, 2003, doi: 10.1361/105996306X108110.

[395] Adrian Wei-Yee Tan, Jun Yan Lek, Wen Sun, Ayan Bhowmik, Iulian Marinescu, Xu Song, Wei Zhai, Feng Li, Zhili Dong , Chris B. Boothroyd Erjia “LiuInfluence of Particle Velocity When Propelled Using N2 or N2-He Mixed Gas on the Properties of Cold-Sprayed Ti6Al4V Coatings”, Coatings 2018, 8, 327; doi:10.3390/coatings8090327.

171

[396] K. Sakaki, “The influence of nozzle design in the cold spray process”, The Cold Spray Materials Deposition Process, 2007, pp 117-126, DOI: 10.1533/9781845693787.2.117

172