Fatigue crack initiation in cold spray coated AZ31B-H24 with AA7075 powder

Highlights

• Cracking mechanism of cold spray coated AA7075 on AZ31B substrate is studied, it was shown that cracking mechanism is a fucnction of coating parameters.

• Residual stresses induced by cold spray were customized for desirable distributions by changing coating parameters.

• In the sample with coating parameters resulting in compressive residual stress at the interface, cracked initiated at the surface, propagated in an inter-splat manner into coating and then substrate.

• Conversely, in the sample with coating parameters resulting in tensile residual stress at the interface, crack/delamination initiated at the interface, propagated into substrate, and in an intra-splat manner into the coating.

• The differences between the state of stress, nano-meter interface grain structure, and local substrate grain refinement at the coat/substrate interface were amongst the reason for different cracking behavior.

Abstract

The effect of cold spray coating parameters on the fatigue life and cracking mechanism of AZ31B-H24 coated with AA7075 powder is investigated. Two sets of coated samples are fabricated based on the selection of different coating parameters. An in-situ control of heat transfer is performed to obtain different residual stress states and microstructure at the aluminum/magnesium interface. Subsequently, the samples are tested under load-controlled fatigue tests at different load amplitudes. Fatigue lives are obtained and the cracking behavior of the two samples is studied and compared with that of uncoated baseline samples. It is revealed that the samples with compressive residual stress at the coating/substrate interface have significantly longer lives (approximately 85% improvement at the same stress) compared with that of the baseline samples. In contrast, samples with tensile residual stress at the interface have similar or slightly improved lives (approximately 24%) compared with that of the baseline samples. The cracking mechanisms of these two samples are considerably different. In the case of compressive samples, cracks initiate at the coating surface and propagate through the splat boundaries of the cold spray coating to the substrate. Conversely, in the case of tensile samples, delamination and cracking initiate at the interface and subsequently propagate to the substrate and through the splats in the coating. The different lives and cracking mechanisms obtained are attributed to the differences in the initial state of stress, details of the microstructure of the nano-size interface layer, and the morphology of the substrate grains adjacent to the interface.

Introduction

Phenomena, such as climate change, global warming, and depleting energy resources, have led to significant research efforts to reduce the carbon footprint. A substantial portion of greenhouse gas emissions is directly emitted due to the transportation sector. Therefore, decreasing the weight of the mobile structures to reduce the fuel consumption has become a significant priority in automotive and aerospace industries. Recently, magnesium (Mg) and its alloys have attracted significant attention as the lightest structural materials with the highest strength-to-weight ratio among all commercially available metals [1]. However, the application of Mg alloys is restricted to non-load-bearing components due to shortcomings, such as low fatigue strength, poor corrosion characteristics, and wear resistance [2]. Coating the surface of manufactured parts with a thin layer of material with higher corrosion-and fatigue-resistance is considered as a practical surface-modification approach to enhance the surface properties and durability of Mg alloys [3].

In an ideal condition, coating a defect-free layer of a tough and hard material with an acceptable cyclic work hardenability and good adherence can enhance the fatigue life of the substrate, regardless of the deposition technique. The fatigue life is improved owing to delay in fatigue crack initiation and propagation [4], [5], [6]. However, the coating surface is prone to defects. The presence of coating defects, such as voids, cracks, undeformed particles, surface imperfections, and tensile residual stress, in the coating or at the interface can have a detrimental effect on the durability of the coated samples [7], [8], [9], [10], [11], [12], [13]. It has been demonstrated that the fatigue crack initiation at the interface, which leads to the delamination of the coating and failure of the sample, is a critical problem that severely affects the fatigue life of coated samples [14]. Defects at the interface, poor splat toughness, and poor bonding between the substrate and coating along with the presence of tensile residual stress at the interface region can result in premature cracking in this particular area [4], [7], [13]. Moreover, it has been reported that the coated samples are susceptible to interfacial cracking at high-stress levels [12], [13], [14]. To address this issue, compressive residual stress is induced in the coating surface using post-processing techniques, such as shot peening and/or grit blasting. This induction can transfer the crack initiation from the interface to the coating surface, which can considerably delay the crack initiation [5], [6], [15], [16], [17]. Therefore, altering the residual stress at the interface during the coating process can be an effective method to transfer the fatigue crack initiation from the interface to the surface of the coated layer and enhance the fatigue life of the coated samples.

Solid-state cold spray coating can form a dense coating layer without introducing the deleterious effects of alternative high-temperature coating methods, such as oxidation, evaporation, melting, phase transformation, and detrimental residual stresses [18], [19]. In cold spraying, a coating layer is formed due to the high kinetic energy of micron-sized coating particles. The particles are supersonically accelerated by passing through a relatively low temperature and pressurized gas via a de Laval nozzle to impact and adhere to a substrate rather than using thermal energy for binding. Intensive plastic deformation of particles upon impact results in the mechanical and metallurgical bonding of the particles with the substrate [20]. The impingement of particles with high kinetic energy during the process can induce compressive residual stress in coated samples. However, an increase in thermal energy of the system due to the carrier gas, particle deformation upon impact, and thermal mismatch effects can provide destructive conditions and reverse the residual stress from negative to positive stress [21], [22]. This situation can be highlighted for temperature-sensitive materials, such as Mg alloys [23].

Although cold spraying technology can enhance the fatigue resistance of materials, the fatigue behavior of Mg alloy coated samples has not been extensively investigated [24], [25], [26], [27]. Cavaliere et al. [25] investigated the effect of processing parameters on the residual stress and fatigue performance of Al2024/AZ91 coated samples. The effect of carrier gas pressure and temperature, and particle speed in the coating microstructure, porosity, bonding strength, and residual stress formation were investigated. It was demonstrated that an increase in the carrier gas temperature and pressure increased the surface compressive residual stress, and improved the mechanical properties and fatigue limit of the coated samples [25]. In contrast, the temperature of the system was considered as a dominant effect that induced tensile residual stress in cold spray additively manufactured hollow titanium cylinders [28]. It was concluded that the thermal stresses were overcome due to the peening effects at a lower nozzle speed. Therefore, a significant tensile residual stress was developed near the outer and inner surfaces of the cylinder [28]. Dayani et al. [27] studied the fatigue behavior of AZ31B cast Mg alloy samples coated with AA7075 powder. The residual stress profiles of the coated samples revealed that significant compressive residual stress was induced in the coating, and a tensile residual stress of 12 MPa was induced at the substrate interface. The higher hardness and fatigue strength of the coating material on the substrate and the induced compressive residual stress of the coating enhanced 25% the fatigue strength of the coated samples at 10⁷ cycles compared with that of as-received samples. The examination of fracture surfaces exhibited delamination and cracking at the interface. It was observed that primary cracks were initiated and propagated from the substrate due to pores and casting defects on the Mg side of the interface [27]. The pores and casting defects might cause primary crack initiation. However, the presence of tensile residual stress in the substrate due to thermal mismatch and substrate microstructural changes due to high thermal energy can cause an undesirable crack initiation from the substrate [29], [30].

The residual stress of AZ31B-H24 substrate near the interface after the deposition AA7075 was customized [31]. The coating parameters, particularly coating temperature, was altered to induce a significant compressive residual stress in the coating and substrate interface [23], [31]. It was observed that the Mg-coated samples with compressive residual stress exhibited enhanced mechanical properties, such as higher hardness of the coating and interface, lower porosity and surface roughness, and finer grain microstructure, compared with that of samples with tensile residual stress [23], [29].

This research aims to study the effect of cold spray coating parameters on the fatigue crack mechanism and fatigue resistance of AA7075/AZ31B-H24 coated samples. Two different samples were fabricated using different coating setups and parameters. Coated samples with tensile and compressive residual stress at the substrate/coating interface were fabricated. Fatigue tests were performed to determine the fatigue life of the tensile and compressive specimens. The crack initiation locations in the tensile and compressive samples before failure were identified and compared. The cause of crack initiation was discussed in detail.