Microstructure of the coatings
Coatings sprayed by APS exhibited characteristic wavy, layered structure without showing cracks inside the layers. The average microhardness value of the iron boride layers is about 760 HV which is much lower than borided layers of AISI 1020 steel at 900°C for 4 h. Typical cross-section microstructure of the coatings sprayed with these parameters is shown in Fig. 2, where the sprayed coating exhibits a wavy layered structure and consists of some unmelted particles and pores, with an average thickness of approximately 100 µm. X-ray diffraction patterns of as-sprayed and heat-treated samples are shown in Fig. 3. As seen at higher magnification (Fig. 4(b)), some microcracks are observable inside the layers. In order to eliminate these microcracks and, preheating substrate before spraying, stopping compressed air cooling used for substrate cooling from the backside as well as extending spray distance from 150 to 170 mm were also tried. However, formation of microcracks was generally not avoided. This is attributed to the cooling rate which leads to generation of residual stresses, which might result in cracks in the coating. PTA boriding study conducted by Bourithis et al.16 showed that microcracks were observed inside the borided layers with increasing content of boron, which lead to the local formation of FeB and, eventually, intergranular cracking. Galvenetto et al.2 sprayed Fe2B (%100) and FeB+α-Fe graded powder composition by VPS on AISI 1040 steel, it was concluded that no cracks were observed inside the iron boride layers. Both studies were focused on fast formation of boride layers on low carbon steel which used commonly for pack boriding process.
Figure 4. SEM images of cross-section microstructures of the coatings sprayed by APS (a) as-sprayed and post-treated at (b) 450°C for 1 h, (c) 500°C for 2 h, and (d) 550°C for 2 h.
Download figure to PowerPoint
While, in this study, aluminum surface was modified by depositing boride layers by APS, which its surface is not possible boriding in conventional boronizing. The observed cracks inside the coating produced by APS might be attributed to fast cooling during spray process due to the higher thermal conductivity of Al compared to low carbon steel, such cracks might be eliminated by spraying graded composition of iron boride powders. Postheat treatment process was applied to treat the sprayed boride layers, enabling the boron diffusion which decomposed and formed solid solution during spray process. In order to densify and improve the mechanical properties, the coatings were post-treated in a furnace at temperatures of 450–550°C for 60 to 120 min in an atmosphere of argon. Modification of initial coating microstructure is shown in Fig. 4. Post-treated microstructure became smoother and improved exhibiting decrease in pore size, that is, coating densification, better adhesion of splat layers as well. As a matter of fact that the hardness of predeposited coating increased significantly at the rate of 40%. Depending on composition of borided layers, measured hardness value changes from 400 to 1600 HV. The measured hardness values of as-sprayed and post-treated coatings are much lower compared with data of borided AISI 1020 steel at 900°C for 4 h, yielding maximum hardness value, 1600 HV. On the other hand, several studies showed that iron-based powder sprayed coatings exhibited good antiwear performance.17,18 The improvement at wear performance and reduction in friction coefficient not only depend on coating hardness but also on the formation of protective oxide films which strongly affect the wear mechanisms of the layers. Therefore, coating sprayed with ferroboron powder containing FexB intermetallic phase, boride oxide, and Fe-based oxide expected to experience with low friction coefficient and improved wear resistance in comparison to the steel surface which is conventionally boronized. Moreover, contrary to the classical boronizing, thermal spraying is a cost-effective promising technique to solve problems such as wear, corrosion, and thermal stability by producing rapidly solidified thick materials on any substrate with a relatively short cycle time.
Mechanical properties of coatings
Dynamic Ultra-microhardness test is applied to cross-section part of samples for determination of hardness and Young's modulus variations of FexB under the loads of 640, 320, and 160 mN applied loads. Dynamic Hardness results for without heat treatment, 450°C for 1 h, 500°C for 1 h, and 550°C for 2 h under 640, 320, and 160 mN applied peak loads are shown in Fig. 5. The figures were constructed experimentally using the data taken from the loading part of depth setting Dynamic Hardness Values (DHV) measurements at various loads. It is seen that DHV numbers decrease with increasing applied peak load and indentation depth. Hardness measurements of FexB shows that when the heat treatment process temperature increase from non-heat treated to 550°C, DHV of FexB layer decreases from 1518 DHV to 657 DHV, from 1042 DHV to 475 DHV, from 897 DHV to 653 DHV, and finally from 953 DHV to 245 DHV with increasing applied peak loads from 160 to 640 mN. As hardness is accepted as an inherent material property, it should not vary with indentation load and size. However, investigations have confirmed that DHV number of materials were indentation size dependent especially at lower peak loads. Increase in hardness with decreasing applied peak load cause from differences in indentation depth, therefore this effect is called indentation size effect. When the curves are examined, two different parts of hardness variation can be clearly seen. The first part of the curve represents the contact region between the indenter and coatings. So, at this point, indentation depth influences the dynamic hardness value of coatings at near the surface region. When the applied load and depth increase, depth effect systematically decreases. At the second part of the curve, dynamic hardness value of coatings is approximately constant depending on indentation depth. The first stage of curves shows the load and size dependency of hardness under applied peak loads. Figure 5 exhibits this kind of behavior.
Figure 5. Hardness-penetration depth curves of the coatings sprayed by APS as-sprayed and post-treated at 450°C for 1 h, 500°C for 1 h, 550°C for 2 h under (a) 160 mN, (b) 320 mN, and (c) 640 mN applied peak loads.
Download figure to PowerPoint
In order to determine the Young's modulus, maximum applied load should be sufficient to produce a permanent deformation on the coating surface. So that the values of Young's modulus, determined from indentations, do not depend on the value of h and, therefore, on the value of the maximum load, the indentation depth should not exceed 10–20% of the layer thickness, otherwise the results will be affected by the properties of the substrate. In this study, thicknesses of FexB layers are between 442 and 298 µm depending on heat-treatment conditions and maximum indentation depth changes among 3.19 and 0.627 µm.
The load–unload (load–displacement) curves shown in Fig. 6 represent the 640, 320, and 160 mN applied load as a function of the displacement of the indenter with respect to the initial position of the cross-section part of layers. Using the experimentally determined Sand hc, the reduced elastic modulus by micro-indentation was calculated and the results are shown in Fig. 7 for heat-treated and non-heat-treated samples. It is clearly seen from the figures that the extracted reduced elastic modulus also exhibits a strong peak load dependency as shown in Table 1. According to the result, Young's modulus values increase with decreasing applied peak loads. The Young's modulus analysis of FexB layer shows that when the heat-treatment temperature increases from non-heat treated to 550°C, Young's modulus of FexB layer decreases from 678 to 202 GPa, from 300 to 110 GPa, and finally from 235 to 52 GPa with increasing applied peak loads from 160 to 640 mN.
Figure 6. A series of load-displacement plots of the coatings sprayed by APS (a) as-sprayed and post-treated at (b) 450°C for 1 h, (c) 500°C for 2 h, and (d) 550°C for 2 h under 160, 320, and 640 mN applied peak loads.
Download figure to PowerPoint
Figure 7. Young's modulus variations of as-sprayed and post-treated at 450°C for 1 h, 500°C for 2 h, and 550°C for 2 h under 160, 320, and 640 mN applied peak loads.
Download figure to PowerPoint
Table 1. Dynamic hardness results of samples
|Material type||Load (mN)||Maximum depth (µm)|| h c (µm)||Dynamic hardness value (DHV)||Young's modulus (GPa)|
|Without heat treatment||160||0.640||0.432||1518||678.34|
Furthermore, elasto-plastic properties of coatings, such as C constant, yield strength, indentation hardness, and strain hardening exponent variation of FexB coatings depending on applied loads and with–without heat treatment was calculated by indentation algorithms, as represented Fig. 8. According to the results, C constants, yield strengths, indentation hardness's, and strain hardening exponents of FeB coatings were decreased from 325 to 76 GPa, from 4.62 to 1.15 GPa, from 27.95 to 4.41, and finally from 0.2621 to 0.2600 by increasing applied load from 160 to 640 mN and heat-treatment temperature from 450 to 550°C, respectively. Indentation size and heat-treatment effect on elasto-plastic properties of FexB coatings were clearly seen in Fig. 7. According to the several studies,19,20 load depended elastic modulus (125–624 GPa) and hardness (17–33 GPa) were obtained at 80, 160, 320, and 640 mN applied peak loads depending on boriding process time. In addition, finite element modeling was applied to simulate the mechanical response of FexB layer on low alloy steel substrate. When the results of mechanical properties were compared from current studies, it can be considered that indentation hardness, Young's modulus, and elasto-plastic properties of coatings show similarities. When the diffusion-controlled production regime and atmospheric plasma spraying compared each other, FexB layer on different substrate shows same indentation properties.20,21
As known, indentation size effect played an important role while increasing indentation loads and depths. In addition to this effect, heat treatment at 450–550°C at inert atmosphere influenced the indentation algorithm results. Concurrently, comprehensive theoretical and computational studies have emerged to elucidate the deformation mechanism in order to systematically extract material properties from P versus h curves obtained from instrumented indentation. Analysis algorithms for determination of elasto-plastic properties were established based on the identified dimensionless functions.21 These algorithms allow for the calculation of the indentation response for a given set of properties, and also for extraction of yield strengths and strain hardening exponent from a given set of indentation data. The main aim of heat treatment was to obtain the improvement of mechanical properties of FexB coatings on Al substrate by APS. Young's modulus, yield strength, indentation hardness, and strain hardening exponents of coatings systematically decreased and mechanically diverged from brittle behavior. In addition, residual stresses and mechanical mismatch effect of Al substrate-FeB coating was also decreased due to the heat treatments.