Corrosion characteristics of plasma spray, arc spray, high velocity oxygen fuel, and diamond jet coated 30MnB5 boron alloyed steel in 3.5 wt.% NaCl solution

2.1 Substrate material

Structurally, the substrate material used was low-carbon special-alloyed boron steel (30MnB5), whose elemental composition is given in Table 1. Samples prepared in dimensions of 15 × 25 × 10 mm were degreased in acetone before coating. To ensure the mechanical bonding of the coating, the sample surfaces were roughened to an average Ra of 8–9 μm by spraying with Al₂O₃ sand at 9 bar pressures.

2.2 Coating method and coating materials

Sample surfaces were coated at Senkron Surface Technologies (San. ve Dış Tic. Ltd., Kocaeli, Turkey). The samples were treated with four different coating materials as listed in Table 2. Although the elemental composition of the coating powder used for HVOF coating and Diamond Jet HVOF coating was the same, different particle sizes were chosen. The powder particle sizes used in HVOF coating were coarser and a conventional-type spray gun was selected. The powder particle sizes used in Diamond Jet HVOF coating were finer and the Diamond Jet spray gun was used. The parameters used in the plasma spray, arc spray, HVOF, and Diamond Jet HVOF coating techniques are listed in Table 3.

2.3 Electrochemical corrosion measurements

Electrochemical experiments were performed in a Gamry instrument, Reference 600 under static and atmospheric conditions. The setup consisted of the coated 30MnB5 boron alloyed steel sample as the working electrode, a graphite rod as the counter or auxiliary electrode, and a Ag_(s)|AgCl_(s)|sat. Cl(aq.)− reference electrode. The coated 30MnB5 boron electrochemical samples were not subjected to mechanical abrasion. Prior to electrochemical impedance spectroscopy (EIS) experiments, the 30MnB5 boron alloyed steel electrode was allowed to attain steady open circuit potential (E_ocp) during 11,000 s free corrosion in 3.5 wt.% NaCl test solution. Thereafter, a sinusoidal excitation potential of ±10 mV was applied over a frequency range of 100,000 and 0.01 Hz to perform the impedance measurements. All the electrochemical experiments were performed in triplicate as such the average value is presented.

2.4 Surface characterization

The surface morphology of the coated 30MnB5 boron alloyed steel samples before and after corrosion in 3.5 wt.% NaCl solution for 6 h was examined using scanning electron microscope (SEM) Quanta FEG 250 model conjoined with an energy dispersive spectroscopy (EDAX) probe (accelerator voltage 10 keV) for the acquisition of elemental composition. The elemental distribution on the steel surface before as after corrosion was also mapped using the EDAX.

3.1 Microstructural of uncoated and coated 30MnB5 samples

The microstructure of the crude 30MnB5 steel sample is given in Figure 1a. The structure of the crude steel consists of ferrite and pearlite matrices and iron carbide structures. Porosity and residues are also seen on the surface. In Table 4 the quantitative analysis (atomic%) of the elements Fe, C, Al, Si, P, Ti, Mn, and B for the full surface area in Figure 1a as determined by the EDAX tests is given. The analysis results show compatibility with the chemical composition of 30MnB5 substrate (Table 1).

Figure 1:

SEM microstructural image of (a) crude 30MnB5 steel at ×220 magnification, (b) plasma spray coated 30MnB5 steel cross-section at ×300 magnification, (c) arc spray coated 30MnB5 steel cross-section at ×500 magnification, (d) HVOF coated 30MnB5 steel cross-section at ×300 magnification, and (e) DJ-HVOF coated 30MnB5 steel cross-section at ×500 magnification.

Thermal spray coatings generally produce porous structures. The image analysis software ImageJ was used to assess whether the coatings met the required specification (Singh and Kaur 2020b). The porosity rate in the microstructure of the coatings was determined via SEM images as 2–3%. Figure 1b shows the SEM micrograph taken for the section coated using the plasma spray technique. It can be seen that the 60(Ni-15Cr-4.4Si-3.5Fe-3.2B 0.7C)-40(WC 12Co) coating applied using the NiCrAlY intermediate binder bonded well to the substrate mechanically and metallurgically. The EDAX analysis given in Table 4 shows the presence of the powder elements on the surfaces. Moreso, high percentage of Fe (90.04%) was detected in Section 1 of Figure 1b. The binder, Ni-22Cr-10Al-1Y, may have contained some Fe or small Fe debris carried into the pores of the coating during grinding/polishing. Chen and Pfender (1996) had also reported the possibility of metallurgical diffusion and alloying during the coating process. It can be observed from the micrograph that the fully melted or semi-melted particles in the coating increased the mechanical interaction. Pores also observed in the micrograph are a characteristic feature of plasma coatings and are caused by the high pressure and temperature to which the coating is subjected to and are believed to result from the breakdown of the lamellae (Yilmaz et al. 2011).

As seen in Figure 1c, the SEM examination of the 28Cr 5C1Mn-based cored wire coatings produced using optimized coatings parameters via the arc spray technique shows that the lamellae typically produced by thermal spray coatings were formed in the structure (Stoica et al. 2004). It can be seen that the coatings made using the NiAl intermediate binder bonded well both mechanically and metallurgically to the substrate. Porosities and oxide zones are seen in the coatings’ structure. In addition, microcracks and particles that solidified afterwards are present in the coatings’ structure. Depending on the chemical content of the coatings’ powder, complex carbides were found in the structure, and carbide phases were deposited in the coating structure, especially along the grain boundaries. This can be explained using the results of the EDAX analysis given in Table 4. The EDAX results reveal that in the regions with darker color contrast than the main phase, Cr is atomically higher.

Figure 1d and 1e, respectively, show the SEM micrographs of cross-sections of coarse-grained WC-10Co-4Cr coatings applied using the kerosene-fueled HVOF technique and of fine-grained WC-10Co-4Cr coatings applied using the propane gas-fired diamon jet-HVOF technique. The general appearance on the micrographs reveals that the coatings are mechanically and metallurgically bonded to the substrate. It can be observed from the micrographs that the fully melted or semi-melted particles in the coatings increased the mechanical interaction. Quantitative analyses (atomic%) of the areas 1, 2, 3 and 4 in Figure 1d and 1e are also given in Table 4. The EDAX spectrum of the area designated 1 was taken from the substrate regions of the migrorographs in Figure 4d and 4e. These results describe the typical parent material. The EDAX analyses from areas 2 and 3 near the melding border provide a composition that almost matches the standard composition of the WC-10Co-4Cr powder. The presence of the element Fe in these coatings regions, respectively, is accepted as its inclusion in the structure by diffusion (Singh and Kaur 2020b). This is an indication that metallurgical bonding has occurred and strengthened the coatings (Singh and Kaur 2020b). It reveals that the light-gray particles in the same areas primarily contain W and C, indicating the presence of the tungsten carbide phase. The dark-gray matrix in area 4 shows the presence of significant C and Co as well as small amounts of other elements such as W, Cr, O, and so on. The surface cross-section of the WC-10Co-4Cr coatings indicates that under the effect of the high temperature of the coatings’ process,WC, W2C Cr, and W parent matrix carbide structures were formed. That is, the carbide matrices are represented by the dark gray region containing the phases in the morphology of the coatings. This matrix shows a low W content, representing low temperature and low WC solubility, causing the carbides to cluster. The porous WC particle grains (Figure 1d and 1e) are well bonded due to the dark-colored CoCr matrix material. Moreover, regions with a bright contrast have carbides with rounded morphology in their microstructure. Generally, the disadvantage of HVOF spray coatings is that more decarburization occurs (Singh and Kaur 2020b).

3.2 E _ocp studies

The variation of open circuit potential with time for the coated 30MnB5 boron alloyed steel in 3.5 wt.% NaCl solution at 25 °C is shown in Figure 2. For all samples, a steady-state E_ocp was achieved at about 3000 s fulfilling one of the cardinal requirements for impedance measurement. The E_ocp, as a thermodynamic property, can be used to examine the extent of metal susceptibility to corrosion in a corrosive medium (Solomon et al. 2021). The nobler the E_ocp, the lesser the anodic dissolution reactions and vice versa (Aydın and Yazıcı 2019). In Figure 2, it is observed that the E_ocp of all the tested samples decreased upon immersion in the corrosive medium indicating the corrosion of the coated 30MnB5 boron alloyed steel samples in 3.5 wt.% NaCl solution. However, the susceptibility of the samples to corrosion in the corrosive solution varies. For instance, the E_ocp value at 11,000 s for plasma spray, arc spray, HVOF, and diamond jet coated samples is −603.3, −642.3, −461.1, and −588.2 mV, respectively. Based on these results (Figure 2), the corrosion resistance of the samples in 3.5 wt.% NaCl solution is in the order: HVOF coated ≫ diamond jet coated > plasma spray coated > arc spray coated.

Figure 2:

Variation of open circuit potential with time for coated 30MnB5 boron alloyed steel in 3.5 wt.% NaCl solution at 25 °C.

3.3 EIS studies

The impedance characteristics of 30MnB5 boron alloyed steel coated via plasma spray, arc spray, HVOF, and diamond jet techniques in 3.5 wt.% NaCl solution at 25 °C in (a) Nyquist and (b) Bode phase formats are shown in Figure 3 In the Nyquist graphs (Figure 3a), a single and imperfect capacitive loop is observed at high frequencies for each of the studied samples. The imperfectness is a common characteristic of a solid electrode and is caused by micro roughness and other heterogeneity on the solid working electrode (Masroor et al. 2017). The singleness indicates a charge transfer phenomenon (Masroor et al. 2017). The capacitive loop, however, is seen to be an incomplete semi-circle suggesting that diffusion equally plays a part in the corrosion process. In the Bode phase representation (Figure 3b), the phase angle for the arc spray, plasma spray, diamond jet, and HVOF coated samples is 31°, 41°, 50°, and 57°, respectively, and confirms contribution from diffusion in the corrosion process. For a completely diffusion controlled corrosion system, the phase angle is expected to be 45° (Gerengi et al. 2016). These observations informed the selection of the equivalent circuits given in Figure 4a and 4b in the modeling of the corrosion of coated 30MnB5 boron alloyed steel in 3.5 wt.% NaCl solution. The selected equivalent circuits gave a good fit as could be seen in the representative fitted graphs in Figure 4c and 4d and the low chi-square values given in Table 5. The equivalent circuits consist of a solution resistance (R_s) that accounts for the resistance between the working electrode surface and the tip of the Luggin capillary holding the reference electrode, a constant phase element (∅) that describes the charge capacitance at the working electrode–solution interface, a charge transfer resistance (R_ct) that indicates the resistance to charge transfer across the interface, and a Warburg impedance (W) that accounts for the diffusion process. Figure 4b has additional resistor (R_f) that describes oxide film resistance. All the electrochemical parameters derived from the fitting process are listed in Table 5.

Figure 3:

Electrochemical impedance spectra for coated 30MnB5 boron alloyed steel in 3.5 wt.% NaCl solution at 25 °C in (a) Nyquist and (b) Bode phase formats.

Figure 4:

Equivalent circuit model used in modeling the corrosion of (a) plasma spray coated, arc spray coated, HVOF coated, and (b) diamond jet coated 30MnB5 boron alloyed steel samples in 3.5 wt.% NaCl solution at 25 °C; (c) and (d) show the representative fitted graphs.

Generally, the size of a capacitive loop is indicative of corrosion resistance by a sample in a corrosive environment (Masroor et al. 2017). The larger the capacitive loop, the higher the corrosion resistance property. From Figure 3, it is obvious that the HVOF coated specimen exhibited the highest corrosion resistance property with a measured total resistance (R_total) of 1233.00 Ω cm² (Table 5). This is followed by the diamond jet and plasma spray coated samples having a R_total of 442.24 and 432.70 Ω cm², respectively. The most susceptible to corrosion is the arc spray coated sample (Al-riched) with a R_total of 185.00 Ω cm². Godlewska et al. (2020) recently reported a detrimental role of Al component in the corrosion of high-entropy alloys, AlCrFe₂Ni₂Mo_x in 3.5 wt.% NaCl solution.

It had been reported that the coating technique could affect the properties of coatings. Huang et al. (2020) noted that high velocity oxygen liquid fuel (HVOLF) sprayed multimodal WC-10Co4Cr coating exhibited good fracture toughness and corrosion resistance, while high velocity oxygen gas fuel (HVOGF) sprayed coating had poor mechanical and electrochemical properties. Similarly, Ding et al. (2018a,b) found the HVOLF WC-10Co4Cr sprayed bimodal coating exhibited lesser WC decarburization, lower porosity, higher hardness, fracture toughness, and erosion resistance compared to HVOGF WC-10Co4Cr sprayed bimodal coating.

3.4 SEM and EDAX studies

Figure 5 displays the SEM images and EDAX data before and after experiments for (a, b) plasma spray coated, (c, d) arc spray coated, (e, f) HVOF coated, and (g, h) diamond jet coated 30MnB5 boron alloyed steel samples in 3.5 wt.% NaCl solution at 25 °C. All the coated surfaces (Figure 5a, 5c, 5e and 5g) exhibit rough morphology but the HVOF technique produced the smoothest morphology (Figure 5e) followed by the diamond jet technique (Figure 5g). Splats and semi-molten powder particles are seen on the coated 30MnB5 boron alloyed steel surfaces particularly the plasma spray and arc spray coated surfaces. This is caused by the pinning effect of molten and semi-molten powder particles during the spraying process (Wang et al. 2020). Expectedly, the plasma spray coated surface (Figure 5a) is enriched with the binder and coated powder components (Table 2): Ni (26.35%), Cr (10.26%), and Al (1.22%). In addition, a high concentration of O (31.15%) is observed and is unarguably from elemental oxidization during the heating process. On the arc spray coated surface (Figure 5c), the content of O, Cr, and Ni detected is 12.24, 24.03, and 0.41%, respectively. On the HVOF coated surface (Figure 5e), the Ni content is 3.41%, Cr is 6.64%, W is 73.95%, and Co is 13.36%. Finally, the content of W, Ni, Co, and Cr detected on the diamond jet surface before corrosion experiments (Figure 5g) is 64.27, 9.22, 9.18, and 1.12%, respectively. Tungsten has an excellent solid solution strengthening effect and should improve the hardness and strength of the coating (Wang et al. 2020). Interestingly, less O content is detected on the HVOF (1.45%) and diamond jet (5.25%) coated surfaces and is suggestive of lesser elemental oxidation during the HVOF coating process relative to other techniques.

Figure 5:

SEM images and EDAX data before and after experiments for (a, b) plasma spray coated, (c, d) arc spray coated, (e, f) HVOF coated, and (g, h) diamond jet coated 30MnB5 boron alloyed steel samples in 3.5 wt.% NaCl solution at 25 °C.

On the surfaces exposed to 3.5 wt.% NaCl solution (Figure 5b, 5d, 5f and 5h), loosely held flake-like products are seen, which is in agreement with the electrochemical results (Figure 2) that the coated samples corroded in 3.5 wt.% NaCl solution. The corresponding EDAX results (Figure 5b, 5d, 5f and 5h), however, provide an insight into the corrosion process. The corrosion of the plasma spray coated sample (Figure 5a and 5b) is basically along the elements: Cr, Al, and Ni phases. For instance, the Cr, Al, and Ni contents in the corroded surface (Figure 5b) decreased from 10.26, 1.22, and 26.35% to 0.75, 0.72, and 10.39%, respectively. This corresponds to 93, 41, and 61% decrease, respectively. A similar decrease in the Cr and Ni contents is observed on the corroded arc spray coated (Figure 5d), HVOF coated (Figure 5f), and diamond jet coated (Figure 5h) surfaces. Additionally, the Co and W elements also contributed to the corrosion of the HVOF and diamond jet coatings. The corroded surfaces are also enriched with chloride ions (Figure 5b, 5d, 5f and 5h).

3.5 Explanation of the corrosion mechanism of the coated 30MnB5 boron alloyed steel

To gain insight into the corrosion of 30MnB5 boron alloyed steel coated via plasma spray, arc spray, HVOF, and diamond jet techniques in 3.5 wt.% NaCl solution at normal atmospheric conditions, EDAX mapping was performed on the coated surfaces. Sample surface was segmented into two sections: one completely insulated from the corrosive solution and the other exposed to corrosive medium for 6 h (see Figure 6a). After the corrosion of the exposed area, the entire surface (exposed and unexposed) was mapped for elemental composition. The EDAX mapping results are given in Figure 6b–e. In the uncorroded portion of the surfaces (Figure 6), high content of Al, Cr, Mn, Ni, Co, and W is noted, which is expected as these elements constitute the undercoat binder or coating powder applied on the steel surface (Table 2). Additionally, O is found on the surfaces and is unarguably due to the elemental oxidation during heating (Huang et al. 2020). The Fe content is however found to be lesser relative to the aforementioned elements, which is due to the covering of the steel surface by the coating. In the exposed section, Cl content from the corrosive solution is observed in addition to the aforementioned elements. Interestingly, the Al, Cr, Mn, Ni, Co, and W content is seen to decrease substantially on the exposed portion and follow the order: Al > Mn > Cr > Co > Ni > W. The O content seems to increase on the corroded portion and could be due to the formation of more metal oxide film. The observed pattern of decrease in elemental content can be reasonably explained by considering the standard electrode potential of the detected elements (Table 6). As it is known, the higher the standard electrode potential of an element, the higher the tendency to corrode (Wang et al. 2020). It could be seen in Table 6 that the standard electrode potential of the elements mostly affected by the corrosion process is in the order: Al > Mn > Cr > Co > Ni > W. In the corrosive medium (3.5 wt.% NaCl), Cl⁻ is enriched on the coating pores (see Figure 6) and consequently created an unbalance oxygen concentration between the inside and outside coatings. This resulted in the formation of oxygen concentration cells (Huang et al. 2020; Wang et al. 2020). The elements acting as the anode gives up electrons and corrode in order of their standard electrode potential to form the cations (mostly Al³⁺, Mn²⁺, Cr³⁺, Ni²⁺, Co³⁺ and W⁶⁺). The electrons released from the anodic oxidation react with oxygen within the solution to form hydroxyl ions (OH⁻). Meanwhile, Figure 3 clearly shows that mass transfer processes played a role in the corrosion process. These processes encourage the rapid combination of the cations with the OH⁻ to form insoluble hydroxides as the corrosion products, which is evidenced in Figure 3b, 3d, 3f and 3h (the flake-like and loosely adhered products). The corrosion products could hinder further corrosion reactions (Huang et al. 2020; Wang et al. 2020) but Huang et al. (2020) noted that, the corrosion products are rapidly stripped off the coating surface by slurry erosion such that the elements continue to corrode. The fact that HVOF and diamond jet coated samples are devoid of undercoat binders, have no or lower amount of Al and Cr, and W has the least standard electrode potential in comparison to other constituent elements may be the reason for the high corrosion resistance of the HVOF and diamond jet coated samples relative to the plasma spray and arc spray coated samples.

Figure 6:

(a) Sample preparation illustration; EDAX elemental mapping of (b) plasma spray coated, (c) Arc spray coated, (d) HVOF coated, and (e) diamond jet coated 30MnB5 boron alloyed steel samples after immersion in 3.5 wt.% NaCl solution at 25 °C.