Impact of TiC/TiB2 Inoculation on the Electrochemical Performance of an Arc-Directed Energy-Deposited PH 13-8Mo Martensitic Stainless Steel

Abstract

This study investigates the impact of incorporating TiC and TiB₂ inoculants on the microstructure and corrosion performance of an arc-directed energy-deposited PH 13-8Mo martensitic stainless steel. The microstructural characterizations revealed partial dissolution of the incorporated ceramic-based nanoparticles, resulting in the formation of in situ TiC phase in the TiC-inoculated sample, while TiC and chromium-enriched M₃B₂ phases were formed in the TiB₂-inoculated sample. Further investigations into the electrochemical response of the fabricated samples confirmed that the applied inoculation strategy slightly enhanced the corrosion resistance of the alloy, offering a valuable advantage for in-service performance for applications in harsher environments. The slight improvement in the corrosion resistance of the inoculated samples was found to be attributed to the formation of a higher fraction of low-angle grain boundaries and enhanced retained austenite content in the microstructure. However, it is essential to note that the formation of chromium-enriched M₃B₂ phases in the TiB₂-inoculated sample led to a slight deterioration in its corrosion resistance compared to the TiC-inoculated counterpart.

1. Introduction

Arc-directed energy deposition, also known as wire arc additive manufacturing (WAAM) is an up-and-coming technique for 3D printing of large metallic components with a medium complex geometry at a relatively high production rate compared to powder-based additive manufacturing methods [1]. However, the strong unidirectional heat sink towards the base plate during WAAM commonly results in the formation of a textured microstructure in as-printed conditions, which could be detrimental to both corrosion and mechanical properties [2]. To overcome these challenges, the in situ addition of refractory nanoparticles as inoculants during the material deposition process is an innovative approach capable of potentially modifying the microstructure and in-service performance of the final product [3].

In one of the authors’ previous studies [3], the effect of in situ inoculation with TiB₂ and TiC nanoparticles was found to improve the mechanical properties of wire arc additive-manufactured PH 13-8Mo martensitic stainless steel (MSS) through modifying the textured as-printed grain structure. However, the impacts of these particles on the corrosion behaviour of the fabricated alloy have not been investigated yet. This study aims to experimentally investigate the effect of TiC and TiB₂ inoculation on the corrosion behaviour of PH 13-8Mo alloys produced by WAAM. The results provide valuable insight into the role of TiC and TiB₂ inoculation in enhancing the corrosion resistance of WAAM-fabricated PH 13-8Mo alloy.

2. Experimental Procedure

2.1. Fabrication of TiC/TiB₂-Inoculated Components Using WAAM

In this study, a commercial grade PH 13-8Mo martensitic stainless steel (MSS) solid wire with a diameter of 1.143 mm and a spool weight of 15 kg was used as the feedstock material. The wire’s chemical composition included 12.25–13.25 wt.% Cr, 7.5–8.5 wt.% Ni, 2–2.5 wt.% Mo, 0.9–1.35 wt.% Al, 0.1 wt.% Mn, 0.1 wt.% Si, 0.05 wt.% C, and the balance was Fe. The WAAM setup utilized a Lincoln Electric S-350 Power Wave and a 6-axis Fanuc robotic arm to build wall-shaped components on an AISI 420 MSS substrate.

Interlayer inoculation involved TiC (99+% purity, with 40 nm average particle size (APS) and TiB₂ (97+% purity, with 70 nm APS) nanopowders, which were mixed with 3 wt.% polyvinyl alcohol (PVA) for 15 min to form a paste. This paste was applied on each layer prior to the deposition of the subsequent track. To ensure consistency, both inoculated and non-inoculated samples were deposited using an optimized set of WAAM processing parameters developed in one of the authors’ previous studies [3]: travel speed of 4 mm/s, wire feed speed of 67 mm/s, arc voltage of 28 V, and arc current of 135 A. To prevent oxidation and other atmospheric contaminations of the melt pools, a shielding gas mixture consisting of 90% helium, 7.5% argon, and 2.5% carbon dioxide (marketed as Blueshield 9) was used at a flow rate of 20 L/min. Furthermore, a bidirectional scanning technique, aligned with the deposition path, was implemented to enhance layer uniformity, resulting in the formation of a straight wall.

2.2. Microstructural Characterization

Metallographic samples extracted from central regions of the printed components were prepared using standard sanding and polishing techniques, followed by etching with Fry’s reagent for 5 s. A range of characterization techniques were employed, including a confocal scanning laser microscope (CSLM, VK-X1000) and a field emission scanning electron microscope (FESEM, FEI MLA 650F) equipped with both electron backscattered diffraction (EBSD) and energy-dispersive X-ray spectroscopy (EDS) detectors. Furthermore, sub-micron and nano-scale features were thoroughly examined using a scanning transmission electron microscope (STEM, Talos 200X) featuring an extreme-field emission gun (X-FEG) source operating at 200 kV.

2.3. Corrosion Testing

To assess the corrosion performance of the as-printed samples, cyclic potentiodynamic polarization (CPP) and electrochemical impedance spectroscopy (EIS) analysis in a 3.5 wt% NaCl solution at 21 ± 1 °C were carried out. The setup included a saturated calomel electrode (SCE) reference electrode, a graphite counter electrode, and the working electrode (sample). The open circuit potential (OCP) was allowed to stabilize over 7 days. Subsequently, the CPP experiment was initiated at −0.3 V relative to the OCP. The scan direction was reversed once a current density of 1 mA/cm² was achieved, and the potential scan continued through the reverse anodic branch at a rate of 1 mV/s. Also, EIS measurements spanned frequencies from 100 kHz to 10 mHz with a 10 mV perturbation voltage after 7 days of immersion. Each corrosion experiment was meticulously replicated at least five times to guarantee the reliability of the obtained outcomes. Following the completion of the CPP testing, a confocal laser scanning microscope (CLSM, VK-X1000) was employed to measure the depth of pitting or localized corrosion present on the specimens, thereby providing insights into the severity of the corrosion-induced damage.

3. Results and Discussion

3.1. Microstructural Characterization

Figure 1 illustrates the microstructural characteristics of the non-inoculated and TiC/TiB₂-inoculated samples. Figure 1a–f represent the micrograph and EDS maps of the non-inoculated sample, revealing Cr-rich δ-ferrite in lathy and vermicular forms. This retention at room temperature is due to the high solidification rate of the WAAM process and the high Cr content (~13 wt.%) in the alloy, which acts as a ferrite stabilizer [4]. The TEM image shown in Figure 1b depicts a lath martensitic matrix with retained austenite, resulting from the presence of approximately 8 wt.% Ni, which acts as an austenite stabilizer and inhibits the complete transformation from austenite to martensite during cooling. Additionally, the EDS map in Figure 1e identifies Al-rich oxides, likely originating from the shielding gas, ambient air, or residual moisture. For more comprehensive information on the microstructure of WAAM-fabricated PH 13-8Mo parts, refer to the authors’ recent publications [3,5].

Figure 1g–l present SEM micrographs and EDS maps of the TiC-reinforced sample, revealing islands of residual δ-ferrite with a different morphology compared to the lathy and vermicular forms observed in the non-inoculated sample. This difference is likely due to the presence of TiC particles acting as nucleation sites, resulting in the formation of equiaxed primary δ-ferrite grains. The EDS maps, shown in Figure 1i–l, confirm the presence of TiC phases that are evenly distributed throughout the TiC-inoculated sample. These TiC phases can be identified as either intact preplaced TiC nano-powders or larger in situ TiC particles formed on Al₂O₃ inclusions during the solidification process. The formation of larger in situ TiC particles can be justified by partial dissolution of the preplaced powders during the deposition of each layer. Zhai et al. [6] have also observed the partial dissolution of micron-sized TiC particles added to 316L powders during selective laser melting, leading to the formation of in situ TiC particles in the final printed component. Bahramizadeh et al. [7] suggested that the lower Gibbs free energy of formation for Al₂O₃ particles promotes their early formation during solidification, enabling them to act as favorable sites for the heterogeneous nucleation of cubic TiC particles. Similarly, Sharifitabar et al. [8] reported the heterogeneous nucleation of TiC on Al₂O₃ particles during gas tungsten arc cladding of 1045 steel, using a mixture of TiO₂, Al, C, and Fe powders.

According to Figure 1m–r, intergranular residual δ-ferrite and Ti-rich particles were also formed in the TiB₂-inoculated sample, following a mechanism similar to that occurring in the TiC-inoculated sample. Additionally, M₃B₂ phases were found in a skeleton-like morphology. During the deposition process, TiB₂ nano-powders partially dissolved, resulting in the formation of in situ TiC phases due to a rapid Ti-C reaction. The remaining boron segregated into the liquid phase, forming Cr-rich M₃B₂-type borides with a skeleton morphology. Sigolo et al. [9] similarly observed the formation of skeleton-shaped M₃B₂-type borides during plasma transferred arc (PTA) cladding utilizing boron-modified stainless steels.

3.2. Crystallographic Orientation Characterization

To comprehensively characterize the crystallographic orientation of the samples, the inverse pole figure (IPF) map, phase map, and grain boundaries misorientation maps of each sample at high magnifications are shown in Figure 2. In the non-inoculated sample, the microstructure predominantly exhibited a lath martensitic structure with a dominant body-centered tetragonal (BCT) structure, with only ~2% of retained austenite (FCC), which increased to ~12% and ~7% in the TiC-inoculated and TiB₂-inoculated samples. The higher fraction of retained austenite in the TiC-inoculated sample can be attributed to the introduction of carbon as a strong austenite stabilizer into the microstructure due to the partial dissolution of preplaced TiC nanoparticles. Chen et al. [10] also reported that the incorporation of TiC particles into stainless steels during additive manufacturing processes increases the fraction of retained austenite. On the other hand, the segregation of additional boron in the TiB₂-inoculated sample at the grain boundaries of primary austenite could reduce its grain size and increase the phase stability of austenite at room temperature [11].

According to the grain boundary maps, the inoculated samples demonstrated a significantly high content of low-angle grain boundaries (LAGBs) (Figure 2h,i), surpassing that observed in the non-inoculated counterpart (Figure 2g). This can be attributed to the effective role of TiC/TiB₂ nanoparticles serving as heterogeneous nucleation sites for grain formation. Upon introduction into the molten metal, these particles facilitate grain nucleation at multiple locations simultaneously. This controlled nucleation process promotes a more uniform distribution of grains, which may favour the formation of low-angle grain boundaries. The specific interactions between the growing grains and the inoculants could encourage alignment along specific axes, thereby creating conditions conducive to the formation of low-angle boundaries. Also, the presence of inoculants can influence the solidification process, affecting the cooling rate and thermal gradients established during solidification, which, in turn, can impact how grains nucleate and grow relative to inoculants, favoring the formation of low-angle boundaries over high-angle ones.

3.3. Corrosion Properties

Electrochemical impedance spectroscopy (EIS) results, illustrated through Nyquist and Bode plots (Figure 3a,b), provide a detailed insight into the electrochemical performance and stability of the passive films on the non-inoculated and inoculated samples. The EIS data highlight the direct influence of microstructure, particularly the distribution of grain boundaries on corrosion resistance. According to Figure 3a, the Nyquist plots of the inoculated samples showed a relatively larger capacitive loop radius compared to the non-inoculated sample, indicating the presence of a more robust and stable passive film, which can be attributed to the high-volume fraction of LAGBs in their microstructure. LAGBs are known for their low energy levels, which makes them less susceptible to corrosion attacks [12]. Certain microstructural features, such as retained austenite content, can also enhance the electrochemical stability of stainless steels due to the higher solubility of substitutional alloying elements, such as Ni and Mo in the austenite phase, with an FCC crystal structure compared to BCC-Fe [13]. As observed in the Bode plots, the inoculated samples exhibited higher impedance at low frequencies, indicative of better passive film barrier properties than the non-inoculated sample. The EIS data were fitted using an equivalent circuit (EC) commonly applied to stainless steels to quantify the EIS measurements [14]. This selected EC, R_s(CPE_P[R_P(R_ct CPE_dl)]) [1], represents an electrode coated by a porous layer, which aligns with the microporous passive films on the surface of PH 13-8Mo MSS. Based on the fitted data, the overall corrosion resistance of the passive layers (R_Total) was measured to be 5.99 × 10⁶ Ω·cm², 5.02 × 10⁶ Ω·cm², and 3.03 × 10⁶ Ω·cm² for the non-inoculated, TiC-inoculated, and TiB₂-inoculated samples, confirming the slightly higher corrosion resistance of the inoculated samples.

The CPP results (Figure 3c) show that all samples maintained a robust passive behaviour, with a wide passivation range and consistent pitting potentials (E_pit), indicating minimal degradation of the passive layer. It is evident that the corrosion potentials of inoculated samples were higher compared to the non-inoculated sample. The corrosion resistance did not decrease due to the unchanged chemical composition, especially Cr content, while the observed improvement in electrochemical stability could be attributed to two main microstructural alterations induced by the inoculation process, including (i) higher fraction of LAGBs and (ii) higher content of retained austenite in the inoculated samples. In corrosion science, the intricate relationship between microstructure and passivity breakdown, particularly pitting behaviour, is a complex phenomenon. Several microstructural factors, including grain boundary misorientation angles, the presence of precipitates, phase structures, and residual stresses, significantly influence the passivity behaviour of stainless steels.

In the present study, the relative fractions of low-angle and high-angle grain boundaries could be one of the contributing factors to the corrosion performance of the non-inoculated and inoculated samples. LAGBs, characterized by dislocations that form regular arrays across the boundary plane, have lower energy levels compared to high-angle grain boundaries (HAGBs). This difference is primarily due to the fewer atomic misorientations involved in LAGBs [12]. As a result, LAGBs tend to be more stable and less prone to corrosion-induced degradation. On the other hand, the point defect model (PDM) suggests that passivity breakdown is primarily driven by cation vacancies’ condensation at the metal–barrier layer interface [13]. When the flux of cation vacancies surpasses their annihilation rate, accumulation of vacancies leads to the detachment of the barrier layer, ultimately triggering passivity breakdown [15]. The elevated concentration of Mo in the reversed austenite phase contributes to improved pitting resistance, as Mo reduces cation vacancy flux, thereby increasing the breakdown potential and delaying the onset of pitting [16]. Furthermore, the high Ni content in reversed austenite also enhances pitting resistance by promoting the crystallization of the passive film and strengthening the barrier layer’s resistance against breakdown [17]. It has also been reported that increased Ni content in the Cr₂O₃ passive layer fortifies the barrier layer, thereby improving the resistance of the material to pitting corrosion [18]. These factors collectively explain the observed improvement in pitting resistance in the inoculated samples.

However, it is notable that the E_pit of the TiB₂-inoculated sample did not improve as significantly as that of the TiC-inoculated sample. This indicates a weaker protective passive film and less stable electrochemical behaviour for the TiB₂-inoculated sample. This observation aligns with the microstructural heterogeneities posed by the presence of high-chromium M₃B₂ phases in the TiB₂-inoculated sample, which enhances micro-galvanic coupling effects between the chromium-enriched area (M₃B₂) and the martensitic matrix. This interaction leads to localized corrosion sites, thereby deteriorating the overall corrosion resistance.

These findings emphasize the critical role of microstructural characteristics, particularly the type and distribution of grain boundaries and the presence of different phases, in influencing the corrosion resistance of stainless steels. The superior performance of the TiC-inoculated samples underscores the benefits of using inoculation techniques that promote a higher fraction of LAGBs and higher content of retained austenite. The challenges observed with the TiB₂-inoculated samples highlight the detrimental effects of unfavorable chromium-enriched borides (M₃B₂), which result in micro-galvanic interactions within a heterogeneous microstructure.

Confocal microscopy (Figure 3d) was employed to further analyze the corrosion effects on the samples following cyclic potentiodynamic polarization tests. The TiC-inoculated samples exhibited minimal pitting, with pit depths significantly shallower than those found in other samples, indicating superior resistance to localized corrosion. This finding aligns with their enhanced electrochemical stability demonstrated in previous tests.

The presence of large pits observed in Figure 3d is attributed to the unfavorable area ratio that develops between the corroding pit (small anode) and the passivated surface (large cathode). The 7-day passivation process results in the formation of a stable passive film on the surface. Consequently, when pitting initiates, the establishment of an unfavorable area ratio between the corroding pit and the surrounding surface becomes inevitable. This phenomenon is comparable to the large pits observed on the surface of galvanostatically passivated 316 stainless steel [19]. Furthermore, the absence of fluctuations in the anodic branch of the CPP curves (Figure 3c), which would indicate metastable pitting, further supports this observation.

There is a well-founded concern regarding the potential impact of strengthening mechanisms on the corrosion resistance of materials. However, the corrosion analysis conducted in this study has demonstrated that applying nano-inoculation strengthening not only did not compromise the corrosion resistance but also resulted in a slight improvement. This enhancement in corrosion resistance is a significant advantage for the in-service performance of materials in industrial applications. This finding underscores the potential of nano-inoculation strengthening as a valuable technique for enhancing both the mechanical and corrosion properties of materials used in challenging industrial applications.

4. Conclusions

This study confirms that the TiC/TiB₂ inoculation strengthening mechanism in WAAM-fabricated PH 13-8Mo martensitic stainless steel not only maintains the material’s corrosion resistance but also slightly enhances its electrochemical stability and corrosion performance in the as-printed condition. This improvement is attributed to the promotion of a higher fraction of LAGBs and increased retained austenite content in the TiC/TiB₂-inoculated samples. However, it is essential to note that the formation of in situ chromium-enriched M₃B₂ phases in the TiB₂-inoculated sample can lead to a more heterogeneous microstructure, resulting in micro-galvanic coupling and potentially facilitating localized corrosion. These findings underscore the necessity of selecting appropriate inoculants and controlling the resultant microstructural features to optimize the corrosion resistance of PH 13-8Mo alloys in demanding environments. By carefully managing these factors, the benefits of nano-inoculation strengthening can be fully realized, enhancing both the mechanical and corrosion performance of materials used in industrial applications.