**Influence of Plasma Electrolytic Oxidation on Fatigue Behaviour of ZK60A-T5 Magnesium Alloy**

#### **Alessandro Morri \* , Lorella Ceschini , Carla Martini and Alessandro Bernardi**

Department of Industrial Engineering (DIN), Alma Mater Studiorum, University of Bologna, Viale Risorgimento 4, 40136 Bologna, Italy; lorella.ceschini@unibo.it (L.C.); carla.martini@unibo.it (C.M.); alessandr.bernardi10@studio.unibo.it (A.B.)

**\*** Correspondence: alessandro.morri4@unibo.it

Received: 4 November 2020; Accepted: 1 December 2020; Published: 2 December 2020

**Abstract:** Magnesium alloys are used in the motorsport and aerospace fields because of their high specific strength. However, due to their low corrosion resistance, protective surface treatments, such as conversion coating or electroless plating, are necessary when they are used in humid or corrosive environments. The present study aimed at evaluating the effect of plasma electrolytic oxidation (PEO), followed by the deposition of a polymeric layer by powder coating, on the rotating bending fatigue behaviour of the wrought magnesium alloy ZK60A-T5. The specimens were extracted from forged wheels of racing motorbikes and were PEO treated and powder coated. Microstructural characterization was carried out by optical (OM) and scanning electron microscopy (SEM) to analyse both the bulk material and the multilayer, consisting of the anodic oxide interlayer with the powder coating top layer (about 40 µm total thickness). Rotating bending fatigue tests were carried out to obtain the S–N curve of PEO-treated specimens. The results of the rotating bending tests evidenced fatigue strength equal to 104 MPa at 10<sup>6</sup> cycles and 90 MPa at 10<sup>7</sup> cycles. The results of the investigation pointed out that PEO led to a reduction in fatigue strength between 14% and 17% in comparison to the untreated alloy. Fracture surface analyses of the fatigue specimens, carried out by SEM and by 3D digital microscopy, highlighted multiple crack initiation sites at the interface between the PEO layer and substrate, induced by the concurrent effects of coating defects, local tensile stresses in the substrate, and increased roughness at the substrate–coating interface.

**Keywords:** magnesium alloy; forging; fatigue; microstructure; plasma electrolytic oxidation (PEO); micro arc oxidation (MAO)

#### **1. Introduction**

Magnesium alloys, because of their low density, high specific strength, high damping capacity, and good castability [1–3], are attractive for lightweight applications in the automotive and aerospace industries, such as transmission housing, engine blocks, steering components, and wheels. Since many structural components are subjected to in-service cyclic stresses, the study of fatigue behaviour of both cast and wrought magnesium alloys, such as magnesium–zinc–zirconium (ZK) alloys, is gaining increasing interest [1,4–9].

Compared to other Mg alloys, ZK series show high strength and formability, mainly due to the presence of Zr, which acts as a grain refiner and leads to the development of a homogeneous equiaxed grain structure, not only in extruded but also in forged and cast-forged components [8]. In fact, even if Zn is added to produce age hardening by precipitation of intermetallic compounds, because of the moderate age hardening response of ZK alloys, the contribution of grain size strengthening is predominant [5]. For this reason, in recent years, some papers focused on the effects of the forging process on the ZK alloys' grain structure, widely used for the production of complex components. Vasilev et al. [7] studied the effects of multiaxial isothermal forging (MIF) on the microstructure and fatigue behaviour of the as-cast ZK60 and demonstrated that MIF is able to refine coarse grains with a consequent improvement of the alloy fatigue response. Karparvarfard et al. compared the tensile and compressive behaviour [4] and the fatigue behaviour [8] of the as-cast and the cast-forged ZK60 alloy, concluding that the superior fatigue strength of the forged alloy compared to the as-cast one is due to grain refinement, the lower amount of porosities, and lower size of the intermetallics induced by plastic deformation.

Due to poor corrosion and wear resistance, when Mg components must be used in a humid or corrosive environment and/or an improvement of wear resistance is required, protective surface treatments are increasingly used [10–14]. For this reason, the fatigue strength of surface-treated Mg alloys has been attracting significant research interest in recent years.

Among surface modification techniques for magnesium alloys, Plasma Electrolytic Oxidation (PEO), also named Micro-Arc Oxidation (MAO), is a well-established solution for the improvement of wear and corrosion behaviour. PEO/MAO is an electrochemical conversion treatment applied not only to Mg alloys but also to Al and Ti alloys and other so-called valve metals [13,15]. This process is based on the modification of the growing anodic film by spark micro-discharges, initiated at potentials above the breakdown voltage of the oxide film [15]. Even though the PEO treatment generally improves wear and corrosion resistance, the production of a hard and rough PEO surface usually leads to a reduction in fatigue strength. This is mainly due to the generation of tensile residual stress during the oxidation treatment or to the presence of defects in the anodic layer, so that the conversion layer may be readily cracked when deformed [10,11,14].

Despite some drawbacks, these surface modification processes are widely used for the production of high-performance components. Up to now, however, only few data on the fatigue behaviour of anodized or PEO-treated magnesium alloys are available and to the best of our knowledge, no data about the effect of PEO treatment on the fatigue performance of ZK alloys are reported, even if these alloys find several applications in high strength aeronautic, military, and racing parts.

The available data on the effect of PEO on fatigue behaviour of Mg alloys, instead, concern the magnesium–gadolinium–rare earths (EV) and magnesium–aluminium–zinc (AZ) alloys [10,12,14]. Ceschini et al. [14] studied the influence of PEO on the fatigue behaviour of the sand cast EV31A Mg alloy. The results highlighted only a 15% reduction in fatigue strength of the PEO-treated alloy compared with the untreated, thanks to the low amount of defects and the good adhesion of the conversion layer. Similarly, Yerokhin et al. [10] demonstrated that PEO coating on the wrought AZ21 Mg alloy may cause no more than a 10% reduction in fatigue strength, which is lower than the effect of traditional anodizing. On the other hand, Nemcova et al. [12] showed a 35–40% reduction in fatigue properties of PEO-treated AZ61 wrought magnesium alloy with respect to the untreated one.

Based on the above, the present study aimed at evaluating the effect of PEO treatment on the rotating bending fatigue behaviour of a powder-coated ZK60A-T5 alloy. The samples were directly extracted from cast-forged wheels in order to test specimens with the same microstructure of the real component.

#### **2. Materials and Methods**

Samples used in the present investigation were extracted from forged wheels made of the ZK60A-T5 alloy (Figure 1), provided by Ducati Motor Holding SpA, Bologna (BO), Italy.

The nominal chemical composition of the investigated ZK60A alloy, reported in Table 1, is characterized by the presence of Zn to induce age hardening during heat treatment [5,16–18] and Zr as grain refiner [5,19,20].

**Figure 1.** Wheel shape and dimensions (width 152 mm, diameter 431 mm) (**a**) and scheme of the fatigue samples extraction locations (**b**).

**Table 1.** Nominal chemical composition (wt.%) of the ZK60A magnesium alloy [21].


The alloy was cast, forged, and T5 heat-treated according to the ASTM B 661 standard [22] to obtain the final geometry and increase the mechanical properties of the final wheel. The hot-working process was carried out at a temperature in the range 350−<sup>400</sup> ◦C; immediately after forging, the wheels were water quenched and then, artificially aged at 150 ◦C for 24 h with final air cooling. Fatigue test samples were extracted from the wheels, as shown in Figure 1, and machined to the final geometry reported in Figure 2. Subsequently, they were PEO-treated and powder-coated. After PEO, the average surface roughness (*R*a) was about 1.6 µm, but the deposition of the powder coating top layer decreased *R*<sup>a</sup> down between 0.5 and 0.8 µm. The roughness was measured before and after deposition of the powder coating top layer by means of a stylus profilometer Hommelwerke T2000 (Hommelwerke, Schwenningen, Germany).

The PEO treatment was carried out in an industrial facility, using a dilute alkaline solution containing P-based as well as Zr-based compounds, above the dielectric breakdown potential of the anodic oxide in order to induce micro-arc discharges, which facilitates the growth of a thick layer and the incorporation of electrolyte compounds [13]. Further details on the forging process and the PEO treatment cannot be disclosed due to confidentiality reasons. A polymeric top layer was deposited above the PEO layer by powder coating (i.e., electrostatic painting) to improve the corrosion resistance of the component, as well as for aesthetical reasons. The polymeric top layer consisted of a carboxyl polyester resin.

Microstructural characterization was carried out using an Axio Imager optical microscope (OM; Zeiss, Oberkochen, Germany) and a scanning electron microscope (SEM) Tescan Mira-3 (Tescan, Brno, Czech Republic) equipped with an energy dispersive X-ray spectroscopy microprobe (EDS; Oxford Instruments, Abingdon, UK). The samples were prepared by standard metallographic techniques (grinding with SiC emery papers 800, 1200, and 2000 grit; polishing with diamond 9, 3, and 1 µM) and chemically etched with Kroll's reagent (HNO<sup>3</sup> 4% HF 2% vol. with H2O). Image analyses were carried out by the ®Image Pro-Plus software (4.5) to evaluate the recrystallized fraction (induced by forging) and the average grain size, evaluated as the square root of the average grain area.

Brinell hardness tests were performed according to the ISO 6506 standard [23] using a 62.5 kg load and a 2.5 mm steel ball indenter (HB10).

**Figure 2.** Three-point rotating bending machine (**a**); geometry and dimensions (mm) of fatigue specimens according to ISO 1143 [24] (**b**).

Fatigue tests were carried out by a three-point rotating bending machine (TP Engineering, Parma, Italy), shown in Figure 2, at a frequency of 47 Hz, at stress ratio R <sup>=</sup> <sup>−</sup>1, testing at least 4 samples at each stress level, in order to obtain the S–N curves according to ISO 1143 [24] and ISO 12107 [25] standards. The influence of the PEO treatment on the fatigue strength was investigated by comparing the experimental S–N curve to the literature data for the untreated alloy. The fracture surfaces, after fatigue failure, were analysed by 3D digital microscope Hirox KH-7700 (Hirox, Tokio, Japan) and by SEM–EDS.

#### **3. Results and Discussion**

#### *3.1. Microstructure*

Representative optical micrographs of the ZK60A-T5 forged alloy are reported in Figure 3, showing the presence of both large un-recrystallized dendrites oriented along the plastic flow (Figure 3a) and zones with fine and equiaxed recrystallized grains (Figure 3b) [8,26–29].

Grain size distribution (Figure 4), evaluated by image analysis, confirmed a bimodal grain structure, with an average grain size of about 4.5 µm for recrystallized grains and 24 µm for un-recrystallized dendrites, even if the length of some dendrites reached 150 µm (Figure 3). The area fraction of recrystallized grains is about 60%. This grain structure is probably ascribable to the synergic effect of localized plastic flow during forging [26] and to elements segregation in the dendrites [28,29].

**Figure 3.** Optical microscope (OM) images of the bimodal grain structure of the ZK60A-T5 alloy etched with Kroll's reagent. Large un-recrystallized dendrites oriented along the plastic flow direction (**a**); fine recrystallized grains (**b**). The specimen was extracted from an untested fatigue sample.

**Figure 4.** Grain size distribution of ZK60A-T5 microstructure, showing a bimodal distribution due to the presence of recrystallized grains, partially recrystallized, and un-recrystallized dendrites.

Moreover, SEM–EDS analyses pointed out the presence of micrometric Zn–Zr intermetallics at the dendrite boundaries (white in Figure 5a) and different amounts of Zn and Zr in the recrystallized equiaxed grains and un-recrystallized dendrites. In Figure 5, the light grey zones, with about 6.4 wt.% of Zn and 1.5 wt.% of Zr, correspond to un-recrystallized dendrites, while the dark grey ones, with about 4.5 wt.% of Zn and 0.5 wt.% of Zr, correspond to recrystallized grains. The higher amount of Zn and Zr in the dendritic zones is probably due to the presence of sub-micrometric and nanometric Zn–Zr based precipitates. These findings are in agreement with other authors [17,27–29], showing that during the recrystallization of Zn–Zr magnesium alloys, fine grains nucleate near the previous casting grain boundaries, while in the grain core, finely dispersed Zn–Zr submicrometric and nanometric 19

precipitates can pin dislocations and prevent the nucleation and growth of the new recrystallized grains. This process results in a "necklace" structure, caused by an incomplete dynamic recrystallization. Dynamic recrystallization (DRX) can occur during the hot deformation of metals and leads to the nucleation and growth of the new grains. Because during hot forging of the wheel, the alloy undergoes inhomogeneous strain rate and temperature fields, there is the possibility that in some regions, temperature or strain rate do not exceed the critical values of these parameters needed for DRX. In these regions, the recrystallization does not take place and therefore, the DRX is incomplete.


**Figure 5.** SEM image highlighting un-recrystallized dendrites (light grey), recrystallized grains (dark grey), and micrometric intermetallics (white) and the EDS analysis locations (spots—Sp.) (**a**); table with the results of EDS analyses (**b**).

Representative SEM micrographs of the polished cross-section of PEO and painting layers are reported in Figure 6. The average thicknesses of the PEO base layer and the powder coating top layer were 11 ± 3 and 26 ± 3 µm, respectively. The cross-sections revealed the typical micro-defective structure of PEO layers, due to discharge events which, on the one hand, favour coating growth and on the other, induce the formation of pores, microcracks, and microchannels, which can negatively affect fatigue strength [30,31]. Micro-arc discharge events also account for the typical micro-undulation of the interface with the substrate, due to localized inward coating growth [32].

**Figure 6.** SEM images at low (**a**) and high (**b**) magnifications of PEO and painting layers cross-section. The cross-sections revealed the micro-defective structure of PEO layers: pores, microcracks, and microchannels. The SEM analyses were carried out on an untested fatigue sample.

The cross-section SEM–EDS images in Figure 7 highlight the rough, micro-undulated morphology of the PEO layer/alloy interface, as already pointed out by Figure 6. Moreover, EDS X-ray maps recorded on the polished cross-section (Figure 7) revealed a homogeneous distribution of phosphorus, zirconium, and oxygen in the PEO layer, due to incorporation of P- and Zr-based compounds from the electrolyte. Moreover, EDS maps also showed the presence of C (main constituent of the polymeric powder coating) as well as of Al- and Si-based compounds, typical inorganic fillers added to the powder coating top layer. Mechanical and fatigue properties of the polymeric top layer are not available. However, the effect of this type of layer on fatigue strength is usually considered negligible [11].

**Figure 7.** SEM–EDS and X-ray EDS elemental maps of a polished cross section of the ZK60A coated alloy. SEM image of PEO and painting layers cross-section (**a**); X-ray EDS elemental maps of the PEO base layer and powder coating top layer (**b**–**h**). The SEM–EDS analyses were carried out on an untested fatigue sample.

#### *3.2. Mechanical Behaviour*

The results of the hardness tests, carried out on the polished cross-section of metallographic samples of the ZK60A-T5 magnesium alloy extracted from the tested fatigue samples, gave hardness values between 70 and 71 HB10, which correspond to about 73−74 HV according to ASTM E 140-07 [33].

The results of the rotating bending fatigue tests, carried out by a three-point rotating bending testing machine according to ISO 12107 [25], are reported in the S–N curves of Figure 8, corresponding to a failure probability of 10% and 50%. In fact, since fatigue tests are typically characterized by large scatter, due to several factors like material inhomogeneity or incorrect specimen alignment,

(104 vs. 90 MPa). When comparing the fatigue behaviour of the PEO-treated alloy with those reported for the untreated substrate in [34], the reduction in fatigue strength was 8% at 10<sup>6</sup> cycles (113 vs. 104 MPa) and 17% at 10<sup>7</sup> cycles (108 vs. 90 MPa).

**Figure 8.** Stress amplitude versus number of cycles to failure (S–N curves) of the PEO-treated and powder coated ZK60A-T5 alloy (ZK60A-T5 + PEO), showing the typical linear fatigue response: the two straight lines correspond to 50% and 10% fatigue failure probability.

Similar results were also reported by Ceschini et al. [14] for the PEO-treated cast EV31A magnesium alloy, showing that the conversion treatment induced a 15% reduction in fatigue strength. Yerokhin et al. [10] also reported a fatigue strength reduction in the range 3–10% induced by PEO treatment on the extruded AZ21 alloy. Nemcova et al. [12], instead, showed a 35–40% reduction in fatigue properties of PEO-treated extruded AZ61 alloy compared to the untreated one. The detrimental influence of the PEO layer on fatigue behaviour of Mg alloys can be ascribed to a combination of factors

leading to enhanced surface stresses: high defect density and microcracks in the coating as well as roughness of the alloy/coating interface (Figures 6 and 7), which can produce multiple crack initiation sites [12].

According to [10,12–14], the reduction in fatigue strength, in fact, is mainly related to the intrinsic defectiveness of the ceramic conversion layer. The micro-discharges cause local melting of the growing oxide layer, inducing the formation of pores, microchannels, and microcracks, as well as the typical volcano-like surface features responsible for surface roughness increase. The subsequent very rapid solidification also generates tensile residual stresses that notoriously negatively affect fatigue strength.

In order to further confirm the previous assessments, an evaluation of the effect of PEO treatment on fatigue behaviour of the ZK60A-T5 alloy has been also carried out by comparing the results of fatigue tests with the fatigue strength of the untreated alloy evaluated from hardness data. According to [6], indeed, a good linear relationship can be established between rotating fatigue strength at 10<sup>7</sup> cycles and the hardness of different cast and wrought heat-treated magnesium alloys. The linear relationship between fatigue strength (*S*<sup>f</sup> ) and Vickers hardness of magnesium alloys is given by:

$$\mathbf{S}\_{\mathbf{f}} \approx \boldsymbol{n} \cdot \mathbf{H} \mathbf{V} \tag{1}$$

where *HV* is the Vickers hardness, and *n* is a coefficient generally equal to 1.32 and 1.66 for artificially and naturally aged alloys, respectively. On the basis of the fatigue strength and hardness values, the *n* coefficient for the PEO-treated ZK60A-T5 is approximately 1.21, which is about 10% lower than the value reported in [6] for artificially aged magnesium alloys. Therefore, the fatigue strength of the PEO-treated alloy is about 10% lower than the fatigue strength expected for an untreated alloy according to (1), that is equal to 98 MPa (Figure 9). This is in agreement with the results of the previous comparison of the experimental and literature fatigue data, respectively, for PEO-treated and untreated alloys.

**Figure 9.** Comparison of the stress amplitude versus number of cycles to failure (S–N curves) at 50% failure probability of the PEO-treated ZK60A-T5 alloy (ZK60A-T5 + PEO), the uniaxial forged untreated ZK60A-T6 alloy [34], and the multiaxial isothermal forged untreated ZK60A-T6 alloy [7]. The fatigue strength for the ZK60A-T5 at 10<sup>7</sup> cycles, evaluated with Equation (1), is also reported.

#### *3.3. Fracture Surfaces Analysis*

Fractographic analyses were carried out firstly by 3D digital microscopy and then, by SEM. A representative low magnification micrograph of the fatigue fracture surface is reported in Figure 10. The surface is characterized by the presence of multiple and superficial nucleation sites, radial ratchet marks in the crack propagation region, and a rough feature in the final fracture zone due to overloading. The presence of the ratchet marks is a consequence of the multiple nucleation sites, since they develop when fatigue cracks, initiated at different positions and propagated in different planes, join together, creating steps on the fracture surface [35,36].

High magnification analyses of the nucleation sites (Figure 10b) pointed out the presence of zones with cracked, fragmented, and even detached PEO coating, which led to an increased roughness of the interface between substrate and PEO layer in comparison to the untested material. Therefore, the multiple nucleation sites, slightly below the specimen surface, could be the result of the synergic effect of the coating defects, tensile stresses in the substrate at the interface with the coating [10,12], and the increased roughness of the interface induced by fragmentation and debonding of the coating during cyclic loading. According to [11,12], in fact, not only tensile stresses in the substrate and coating defects, but also roughness at the interface, acting as a stress riser, can facilitate fatigue crack nucleation during cyclic loading.

Furthermore, the absence, next to the initiation sites, of typical forging defects (e.g., oxides), large intermetallics, or slip bands, that usually induce local stress concentration and the formation of cracks [36] confirms that the observed fatigue strength reduction is mainly due to the detrimental effect of the PEO layer whose typical microstructure facilitates crack nucleation.

A representative SEM image of the crack growth region is shown in Figure 11. The propagation zone (Figure 11a) shows the presence of classic fatigue striations, bright micro-cliffs, respectively, perpendicular and parallel to the direction of crack propagation [8,35], as well as secondary cracks probably due to the local strengthening of the matrix [8] or the activation of twinning during cycling loading [9]. Both for primary and secondary cracks, the crack path is mainly transgranular [12], as highlighted in the OM cross-sections in Figure 12.

SEM image of the overloading region in Figure 11b, instead, display a mixed morphology with quasi-cleavage features and zones with dimples and tear ridges, the latter typical of a ductile fracture [8].

**Figure 10.** Representative 3D digital microscopy of the fracture surface of a PEO-treated ZK60A-T5 fatigue sample tested at 100 MPa and failed after 1.5 <sup>×</sup> <sup>10</sup><sup>6</sup> cycles (**a**); SEM image of the morphology of the PEO layer next to a crack nucleation site (**b**).

**Figure 11.** Representative SEM images at high magnification of the crack growth region with fatigue striations and secondary cracks (**a**), and of the overloading region with dimples and cleavage planes (**b**). Fatigue samples tested at 100 MPa and failed after 1.5 <sup>×</sup> <sup>10</sup><sup>6</sup> cycles.

**Figure 12.** OM image of a fracture surface cross section, highlighting the presence of transgranular secondary cracks. Fatigue samples tested at 95 MPa and failed after 3.5 <sup>×</sup> <sup>10</sup><sup>6</sup> cycles.

#### **4. Conclusions**

The present study investigated the effect of PEO-treatment on fatigue behaviour of the ZK60A-T5 alloy by testing samples directly extracted from forged wheels under rotating bending conditions. The following conclusions can be drawn:


**Author Contributions:** Conceptualization, A.M., L.C., and C.M.; data curation, A.M. and A.B.; formal analysis, A.M., L.C., and C.M.; investigation, A.M. and A.B.; methodology, A.M., L.C., and C.M.; validation, A.M., C.M., and A.B.; writing—original draft, A.M. and A.B.; writing—review and editing, L.C. and C.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. We wish to thank Iuri Boromei at the Dept. of Industrial Engineering (University of Bologna) for SEM observations and EDS analyses of PEO-treated samples, as well as Andrea Morri for his support of experimental work. We also wish to thank Simone Messieri for providing fatigue samples.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **The Effect of Co-Deposition of SiC Sub-Micron Particles and Heat Treatment on Wear Behaviour of Ni–P Coatings**

**Donya Ahmadkhaniha 1,\*, Lucia Lattanzi 2,\* , Fabio Bonora <sup>2</sup> , Annalisa Fortini <sup>2</sup> , Mattia Merlin <sup>2</sup> and Caterina Zanella <sup>1</sup>**


fabio01.bonora@edu.unife.it (F.B.); annalisa.fortini@unife.it (A.F.); mattia.merlin@unife.it (M.M.)

**\*** Correspondence: donya.ahmadkhaniha@ju.se (D.A.); lucia.lattanzi@unife.it (L.L.)

**Abstract:** The purpose of the study is to assess the influence of SiC particles and heat treatment on the wear behaviour of Ni–P coatings when in contact with a 100Cr6 steel. Addition of reinforcing particles and heat treatment are two common methods to increase Ni–P hardness. Ball-on-disc wear tests coupled with SEM investigations were used to compare as-plated and heat-treated coatings, both pure and composite ones, and to evaluate the wear mechanisms. In the as-plated coatings, the presence of SiC particles determined higher friction coefficient and wear rate than the pure Ni–P coatings, despite the limited increase in hardness, of about 15%. The effect of SiC particles was shown in combination with heat treatment. The maximum hardness in pure Ni–P coating was achieved by heating at 400 ◦C for 1 h while for composite coatings heating for 2 h at 360 ◦C was sufficient to obtain the maximum hardness. The difference between the friction coefficient of composite and pure coatings was disclosed by heating at 300 ◦C for 2 h. In other cases, the coefficient of friction (COF) stabilised at similar values. The wear mechanisms involved were mainly abrasion and tribo-oxidation, with the formation of lubricant Fe oxides produced at the counterpart.

**Keywords:** electroplating; Ni–P coatings; SiC particles; heat treatment; wear

#### **1. Introduction**

Coatings are often applied to industrial components to enhance the durability of materials in abrasive conditions or corrosive environments. Nickel-phosphorous (Ni–P) alloys are one of the most applied alternatives for applications such as aerospace, electronics, machinery, automotive, oil and gas [1–4]. Ni–P coatings have been mainly obtained through electroless plating [2]. However, due to the low overall speed (the deposition rate is only a few micrometres per hour) and continuous maintenance, electroplating could be a valid alternative [3]. The properties of Ni–P coatings depend on their phosphorus (P) content. Based on the P content, Ni–P coatings can be classified into low (2 wt.%–4 wt.%), medium (5 wt.%–9 wt.%) and high (>10 wt.%) P coatings, according to the ASTM B733-15 [5] standard. The increase of P leads to microstructural changes from crystalline to amorphous structures (P > 10 wt.%) [6,7]. Amorphous Ni–P coatings present good corrosion resistance with a hardness around ~600 HV. Heat treatment and addition of reinforcing particles (to produce composite coatings) can be applied to enhance the hardness of these coatings for the demanding situation. The hardness of the coatings depends on the heating time and temperature. Biswas et al. [8] studied the effect of heating temperature for 1 h on the tribological behaviour of Ni–P coatings. They obtained the maximum hardness of ~1085 HV0.1 at 400 ◦C, which slightly decreased to 975 HV0.1 at 600 ◦C. In the case of composite coatings, the nature of the reinforcing particles, their size, their percentage, and their distribution within the matrix can affect the hardness of the coatings [9–12]. Metzger et al. [13]

**Citation:** Ahmadkhaniha, D.; Lattanzi, L.; Bonora, F.; Fortini, A.; Merlin, M.; Zanella, C. The Effect of Co-Deposition of SiC Sub-Micron Particles and Heat Treatment on Wear Behaviour of Ni–P Coatings. *Coatings* **2021**, *11*, 180. https://doi.org/ 10.3390/coatings11020180

Academic Editor: Armando Yáñez-Casal Received: 8 January 2021 Accepted: 29 January 2021 Published: 3 February 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

demonstrated that SiC and Al2O<sup>3</sup> particles resulted in better wear resistance of Ni–P coatings than SiO<sup>2</sup> particles. Sliem et al. [14] co-deposited ZrO<sup>2</sup> nanoparticles in Ni–P coating by means of pulse electrodeposition and showed a gradual improvement in the mechanical properties of the coating. The hardness and modulus of the coating reached to the maximum value of 6.7 and 21.72 GPa, respectively, by adding 1 g/L ZrO<sup>2</sup> to the plating bath. Tamilarasan et al. [15] investigated the effect of surfactant on wear and friction behaviour of Ni–P/TiO<sup>2</sup> coatings. They found that by optimum concentration of the surfactants, a smoother surface with better distribution of TiO<sup>2</sup> particles were achieved which in turn resulted to low frictional coefficient, and high wear resistance.

Besides, the effect of heat treatment on properties of Ni–P composite coatings has also been investigated. Karthikeyan et al. [16] studied the effect of heat treatment on indentation behaviour of electroless Ni–P/Al2O<sup>3</sup> coating. They reported that incorporation of Al2O<sup>3</sup> nanoparticles induces strengthening. They also found that heat treatment results in the precipitation of Ni3P intermetallic compound, which was increased with the heat treatment temperature up to 400 ◦C. However, the maximum hardness (16.4 GPa) was gained by heat treatment at 300 ◦C. Ram et al. [17] studied the wear behaviour of as deposited and heat-treated Ni–P and Ni–P/Al2O<sup>3</sup> coatings in dry sliding conditions. Oxidation and adhesion in Ni–P coating and a combination of oxidation, adhesion, and abrasion in Ni–P/Al2O<sup>3</sup> coatings were reported as the main wear mechanisms. They demonstrated that heat treatment enhanced the hardness of both Ni–P and Ni–P/Al2O<sup>3</sup> coatings. The incorporation of Al2O<sup>3</sup> nanoparticles reduced the propagation of micro cracks during the wear test of heat-treated coatings. Thus, maximum wear resistance was achieved for Ni–P/Al2O<sup>3</sup> heat treated at 400 ◦C. De Hazan et al. [18] investigated three heat treatment procedures, at 270 ◦C for 10 h, 390 ◦C for 2 h, and 480 ◦C for 1 h. The authors observed a reduction in hardness HV0.3 of heat-treated Ni–P/SiO<sup>2</sup> coatings in all cases, while the wear resistance was improved. Chang et al. [19] annealed the Ni–P and Ni–P/SiC coatings at 200, 400, and 600 ◦C for 1 h. SiC particles led to increased hardness of minimum 30% compared to Ni–P coatings, in all the annealing conditions investigated. The authors also found a decreasing trend in hardness for heat-treated coatings at 600 ◦C, and they attributed it to an excessive number of porosities related to Ni3P precipitation during heat treatment. Besides, heating at a temperature higher than 450 ◦C resulted in decomposition of SiC, which then reacted with Ni to form γ-Ni5Si<sup>2</sup> and β-Ni3Si phases and produced carbon (C) precipitation. Apachitei et al. [20] found that the heat treatments at 500 ◦C changed the deformation mechanisms in electroless Ni–P and reduced the Ni–P coatings' abrasive wear resistance. Besides, Ni3Si formed, and the adhesion between the reinforcement and the matrix was improved. Aslanyan et al. [21] observed different wear mechanisms in Ni–P and composite coatings. Although the hardness of the coatings increased by heat treatment, composite coatings' wear rate was enhanced with respect to Ni–P one [22].

According to the above information, it is necessary to find a proper heat treatment procedure that can enhance the hardness and wear resistance of electrodeposited Ni–P and Ni–P composite coatings. The present study aims to evaluate the effect of SiC particles in combination with different heat treatment conditions on the wear behaviour of Ni–P coating. For this purpose, the effect of SiC particles addition on composition, hardness and wear resistance was evaluated to find out the relation between the composition and wear behaviour of Ni–P coatings. Besides, heat treatment on Ni–P as well as composite coatings was carried out and the wear behaviour of the heat-treated samples was studied to disclose any difference between the wear behaviour of heat-treated Ni–P and Ni–P/SiC coatings. For wear behaviour studies, the 100Cr6 steel was chosen as the antagonist in the wear system to describe non-abrasive conditions since the literature of electrodeposited Ni–P coatings investigated abrasive systems [15–21].

#### **2. Experimental Methods**

Ni–P coatings were deposited by direct plating on low alloyed steel (UNI EN ISO 683-4:2018 [22]) pins. The pins with the geometry shown in Figure 1 were machined by

computer numerical controlled HAAS CNC ST 10 lathe (Haas Automation, Oxnard, CA, USA). Before deposition, the steel substrates were ultrasonically cleaned in an alkaline soap and activated by pickling for 8 min in 2.5 M H2SO4, between each step the samples were rinsed with distilled water.

**Figure 1.** Dimensions of the pin in mm.

The electrodeposition was carried out in 2 L modified Watts bath containing NiSO4·7H2O, NiCl2·6H2O, H3PO3, H3BO3, and two additives (saccharin and sodium dodecyl sulfate). In the case of composite coatings, 20 g/L of SiC particles (β-SiC provided by Get Nano Materials, Saint-Cannat, France) were added to the electrolyte 24 h in advance, and they were stirred by a magnetic stirrer. The particles have an average particle size of 100 nm and irregular morphology. For better dispersion of the particles, the electrolyte was stirred ultrasonically for 30 min before plating. For deposition, the pin (cathode) was immersed in the electrolyte, and two pure bent Ni anode sheets surrounded it. The distance between anode and cathode was 5 cm. Two different cathode configurations were adopted for pure (Figure 2a) and composite (Figure 2b) coatings, the latter was chosen to achieve the maximum SiC particles co-deposition.

**Figure 2.** Configurations of the deposition cells for (**a**) pure coatings and (**b**) composite coatings.

The deposition was carried out at a current density of 4 A/dm<sup>2</sup> , for a total time of 105 min at 70 ◦C with a pH of 2.15. The deposition parameters were kept the same for pure and composite coatings. The deposited layers have an average thickness of about 38 µm, depending on the current efficiency (CE). Different heat treatments, optimised in previous work [23] and listed in Table 1, were applied to some coated samples while the remaining were kept in the as-plated (AP) condition.


**Table 1.** Heat treatment conditions.

Surface morphology and composition of the coatings were characterised by a scanning electron microscope (SEM, JSM-7001F, JEOL, Akishima, Japan) equipped with energy dispersive spectroscopy (EDS, Octane Pro, EDAX, Mahwah, NJ, USA). The coatings' microhardness was measured on the cross-section of the pin by a Vickers indenter (NanoTestTM Vantage, 40.36, Micro Materials, Wrexham, UK) with a load of 200 mN and dwell time of 10 s. A total of 15 measurements were carried out on two samples for each investigated condition.

Wear tests were performed at room temperature in a ball-on-disc configuration, with uni-directional sliding, on a 100Cr6 steel (EN ISO 683-17 [24], AISI 52100) disc. The tests were carried out on a pin-on-disc tribometer (Pin-on-disc tester TR-20LE, Ducom instruments PVT LTD, Bangalore, India) with a load of 20 N under dry conditions. The tribometer tests were run at a constant linear sliding speed of 0.15 m/s and a total sliding length of 200 m for the as-plated samples and of 1000 m for the heat-treated samples. The trends of the coefficient of friction (COF) were acquired and registered during the tests. A minimum of three samples for each condition was tested. The wear debris and the worn surfaces on the pins were investigated by SEM (Evo MA15, Carl Zeiss Microscopy, Milan, Italy) coupled with EDS (Oxford Instruments, Abingdon, UK) analysis.

The tip surface of each pin was acquired before (Figure 3a) and after (Figure 3b) the wear test by a non-contact 3D optical profilometer (Talysurf CCI Lite, Taylor-Hobson Limited, Leicester, UK) with an optical resolution of 0.76 µm.

**Figure 3.** Point clouds of the pin tip surface from the profilometer acquisitions: (**a**) unworned; (**b**) worned, at the end of the wear test.

The comparison of the point clouds before and after wear enabled us to evaluate the volume loss with a tailored Matlab ® code, integrating the volume under the point clouds of the surfaces. The wear rate (WR) was defined via Equation (1):

$$\text{WR} = \frac{\Delta V}{F \times L} \left( \frac{\text{mm}^3}{\text{N} \times \text{m}} \right) \tag{1}$$

where ∆*V* (mm<sup>3</sup> ) is the volume loss, *F* (N) is the applied load, and *L* (m) is the total sliding length.

#### **3. Results**

*3.1. Coating Appearance and Morphology*

Figure 4 shows the coated pin with both Ni–P and Ni–P/SiC. The Ni–P coating (Figure 4a) had a shiny appearance while the Ni–P/SiC was a matt coating (Figure 4b). This difference was attributed to the presence of SiC particles, which increased the surface roughness of the coatings and therefore changed the reflection of the light.

**Figure 4.** Coated pins with (**a**) Ni–P and (**b**) Ni–P/SiC coatings.

The morphology of the coatings on the surface of the pin can be seen in Figure 5. Figure 5a demonstrates a nodular morphology for Ni–P coating. In contrast, by adding SiC particles, the morphology was changed to smaller nodules than those in Ni–P (Figure 5b). A cross-sectional image of the composite coating (Figure 5c) exhibited the SiC particles distributed homogeneously along the coating thickness, as highlighted in the EDS map in Figure 5d. According to the SEM cross-section image, it was evident that SiC particles agglomerated during the coating process.

**Figure 5.** SEM image of (**a**) Ni–P surface and (**b**) Ni–P/SiC surface; (**c**) SEM image of the cross-section of Ni–P/SiC coating; and (**d**) EDS map of Si element related to (**c**).

The composition of the surface of the coatings was measured by EDS, and the results are listed in Table 2. Both Ni–P and Ni–P/SiC coating had more than 12 wt.% of P, and they were categorised as Ni high-P coatings. The addition of SiC particles slightly reduced the P content of the coating. This reduction was not observed in the previous study [25].

**Table 2.** Surface composition and current efficiency (CE) of Ni–P and Ni–P/SiC coatings.


Table 2 also lists the CE values of the exposition that was not affected by the addition of SiC particles.

#### *3.2. Coatings Hardness*

Coatings hardness was measured by Vickers indenter on the cross-section of the coatings, and the results are shown in Figure 6. Ni–P coating had 650 ± 30 HV0.02 and the addition of SiC particles increased the hardness values of 15% to 740 ± 10 HV0.02. However, heat treatment had more impact on enhancing the hardness values of these coatings. The maximum hardness in Ni–P coating was 1130 ± 30 HV0.02 achieved by HT400, while HT360 resulted in the maximum hardness of Ni–P/SiC coatings, 1240 ± 130 HV0.02.

**Figure 6.** Microhardness valued of Ni–P and Ni–P/SiC coatings before and after heat treatments. Error bars represent standard deviation.

#### *3.3. Wear*

Coefficient of friction (COF) trends and wear rate of as-plated samples are represented in Figure 7. The COF evolution (Figure 7a) differed significantly with the presence of SiC particles. The average COF value for the Ni–P coating was 0.4, and it increased to 0.6, with the presence of SiC particles. Figure 7a shows one curve for each condition, the closest to the average, representing all tested samples. The wear rates (Figure 7b) did not differ significantly in the same condition, as all values are ~10 <sup>−</sup><sup>5</sup> mm3/Nm for the pure coating and ~2 × 10 <sup>−</sup><sup>5</sup> mm3/Nm for the composite ones. Figure 7b gives the correspondent P and Si content on the pin tip for each wear rate value, the changes were not significant to influence the wear rate.

**Figure 7.** As-plated samples: (**a**) coefficient of friction (COF) evolution; (**b**) wear rate and P and Si content.

Representative wear tracks of the as-plated samples are depicted in Figure 8a for the Ni–P coatings, and Figure 8b for the composite ones. The yellow arrow in Figure 8a points at the residual layers of Fe oxides, coming from the antagonist, that adhere to the wear track. The representative wear track in Ni–P coating (Figure 8a) has a diameter of ~600 µm, smaller than the one of Ni–P/SiC coating (Figure 8b) with a diameter of ~800 µm. This observation mirrors the comparison of wear rate values in Figure 7b. In all cases, the substrate was not reached during the wear test.

**Figure 8.** Wear tracks of the as-plated samples: (**a**) Ni–P; (**b**) Ni–P/SiC.

Figure 9 depicts the COF evolution and wear rate of heat-treated samples. For HT300 samples (Figure 9a), the presence of SiC particles determined a slight decrease in the average COF, from ~0.65 to ~0.5. On the contrary, the COF curves reached the stable state at the same value for pure and composite coating in HT360 (Figure 9b) and HT400 (Figure 9c) treatments. Figure 9d gives the correspondent P and Si content on the pin tip for each sample as even small changes could influence the precipitation during the HT.

**Figure 9.** *Cont*.

**Figure 9.** Heat-treated samples: (**a**) COF evolution of HT300; (**b**) COF evolution of HT360; (**c**) COF evolution of HT400; (**d**) wear rate and P content.

Figure 10 depicts the representative wear tracks for pure and composite coatings in each heat-treated condition: HT300 (Figure 10a,b), HT360 (Figure 10c,d) and HT400 (Figure 10e,f). In some cases the substrate steel was reached during wear, as visible for example in Figure 10a; in others, the wear track presented a mixed surface, characterised by substrate, coating residuals and Fe oxides originated from the antagonist (Figure 10d).

No specific trends were evident, as summarised in Table 3, although the substrate was reached more times with the pure coatings than with the composite ones. The different occurrence can be ascribed to local features, like the local thickness of the coating.

**Figure 10.** Wear tracks, after acetone rinsing, of the heat-treated samples: for HT300 treatment, (**a**) Ni–P and (**b**) Ni–P/SiC; for HT360 treatment, (**c**) Ni–P and (**d**) Ni–P/SiC; for HT400 treatment, (**e**) Ni–P and (**f**) Ni–P/SiC.



#### **4. Discussion**

#### *4.1. As-Plated Coatings*

Addition of SiC particles reduced the P content of the coating by 10%, although this reduction was not seen in the previous study [25]. This decrease can be related to the significant amount of co-deposited SiC, around 11 wt.% in this study. It has been reported that for Ni plating at low pH, SiC particles enhance the hydrogen (H2) evolution by absorbing protons on their surfaces [26,27] and consequently, results in increased hydrogen evolution. Therefore, the nascent H<sup>2</sup> required for reducing P source to phosphine (according to the indirect reactions) [28] is decreased, hence reducing the P content in the coating. Besides, a CE reduction is expected by the P decrease in the coatings due to the H<sup>2</sup> catalysis by SiC particles. In this study, the CE for Ni–P and Ni–P/SiC coatings were comparable, as seen in Table 2.

The wear mechanism involved in these coatings consisted of abrasion and tribooxidation. Abrasive wear was dominant in the early stages of contact. Then, the formation of Fe-based oxides led to tribo-oxidation. According to EDS measurements, the oxide layers were present in both pure and composite coatings, with no significant difference in the composition. The presence of the Cr signal confirmed that these oxides formed at expenses of the 100Cr6 antagonist. The COF evolution stabilised at a higher value for the composite coatings than the pure ones, as represented in Figure 7a. These results were mirrored by the wear rates in Figure 7b and suggested that the SiC particles determined an increase of friction in the system. Comparing the two wear tracks in Figure 8a,b, the SiC particles also acted as the third body in abrasive wear. They determined a rough appearance of the wear track. The SiC agglomerations are exposed by coating wear and can be easily disrupted during the sliding under load, leading to a spread of particles between the coating and the antagonist. In this case, the presence of Fe oxides mitigated the abrasive action of the ceramic particles, without fully compensating it. The steady-state COF values did not align with the results from Aslanyan et al. [21] that obtained similar COF trends for both Ni–P and Ni–P/SiC coatings in a system with corundum (Al2O3) as the antagonist. Nevertheless, the difference between these results can be expected, given the significant role of the antagonist. Similar observations hold for the work by Aghaie et al. [29], that tested as-plated Ni–P/SiC against copper (Cu). The authors reported a COF in the 0.5–0.7 range for the initial 200 m, with fluctuation peaks, and these values align with the present study despite the different wear system. They also observed the pull-out of SiC particles from the matrix as wear proceeded, and those particles determined abrasion at expenses of the coating. Chang et al. [19] compared Ni–P and Ni–P/SiC coatings against zirconium oxide (ZrO2), observing a higher volume loss for the composite coating than the pure one, as well as a higher COF. These results are comparable with what reported in the present study with a different wear system. In summary, different coupling systems can lead to different wear mechanisms, and in the present study the formation of Fe oxides coming from the antagonist influenced the response of coatings to wear.

#### *4.2. Heat-Treated Coatings*

The hardness values in Figure 6 support the hypothesis that crystallisation and precipitation happened at a lower temperature in Ni–P/SiC coating than Ni–P one. Since all the tested samples had similar composition (similar P and SiC wt.%), the hardness of the coatings in Figure 6 was related to the different heat treatments. At 400 ◦C, after reaching the maximum hardness, grain growth started, and so hardness was reduced. Hence, SiC particles did not influence the grain boundaries migration in this case.

The wear mechanisms consisted of abrasion and tribo-oxidation for the heat-treated coatings, similarly to the as-plated coatings. The presence of SiC particles determined a slight difference in the COF evolution for the HT300 samples (Figure 9a). At the same time, it did not influence the COF values for HT360 (Figure 9b) and HT400 (Figure 9c) samples. Nevertheless, the composite coatings presented noisier curves because of the presence of reinforcing particles acting as third body once exposed and pulled-out from the matrix. For what concerns wear rates, it is necessary to distinguish the samples with wear tracks that reached the substrate (solid circles in Figure 9d) from the ones that presented a mixed surface (empty circles in Figure 9d).

The wear tracks of HT300 pure coatings showed the substrate presented a wear rate of ~3.5 × 10−<sup>6</sup> mm3/Nm. Micro-plough traces were evident on the surfaces, and due to the fragmentation of coating during sliding (Figure 10a). The lubricant role of Fe oxides did not compensate for the abrasive action of such fragments on the substrate. In the cases with substrate exposed, the thickness of the coating was ~26 and ~34 µm, respectively. These values were estimated from the dimension of the annular section of the coating that characterised the wear track. The sample that presented a mixed surface also presented a slightly lower P than the other samples. For the HT300 composite coatings, the wear track presented a mixed surface, with the simultaneous presence of substrate, Fe oxides, and residual coating in the centre (Figure 10b). The higher hardness of composites (Figure 6) and SiC particles' presence as the third body was mitigated by the Fe oxides. The one sample that showed the substrate had a thickness of ~27 µm. The substrate surface was smooth and regular, resulting from a homogeneous abrasion due to the action of SiC particles. It was also the sample that showed the higher content of P with respect to the other ones. These observations align well with the COF trends of Figure 9a since the presence of Fe oxides did not prevent micro-ploughing mechanisms in pure coatings but showed a beneficial effect for the composite coatings. Kong et al. [30] also reported similar COF values for electroless Ni–P coatings treated at 300 ◦C and observed abrasive wear mechanisms, although the antagonist material was not stated.

Moving to the pure coatings after HT360 treatment, the substrate emerged in the wear track for two samples, and the wear rate values are in the 4.5× 10−6–5.8 × 10−<sup>6</sup> mm3/Nm range. The centre of wear tracks appeared uniform and regular (Figure 10c), thanks to the Fe oxide's presence that mitigated the damaging effect of coating fragments. In these cases, the coating thickness was estimated at ~35 and ~48 µm. One sample presented a mixed surface. In composite coatings, all samples presented a mixed surface on the wear track, with the co-presence of residual coating and Fe oxides. In all these cases, the wear rate was ~10−<sup>5</sup> mm3/Nm, higher than the average value obtained for the pure ones. These results accord well with the hardness values, significantly higher for the Ni–P/SiC coating than the pure ones after the HT360 treatment (Figure 6). Nevertheless, these results are not mirrored by the COF trends in Figure 9b, comparable for both pure and composite coatings. Kong et al. [30] investigated electroless Ni–P coatings and in particular the influence of the heat-treatment on the wear response. They reported that the coatings treated at 350 ◦C present the maximum hardness of ~6100 MPa. The friction coefficient is 0.8. This is also the condition that determines the change from abrasive wear, for temperatures below 350 ◦C, to adhesive wear for temperatures higher than 350 ◦C.

The results for the HT400 treated samples were different. The Ni–P coatings presented a mixed surface on the wear track in all cases, as depicted in Figure 10e. The wear rate values are lower than ~10−<sup>5</sup> mm3/Nm, and values in the same range were found for the composite coatings, as represented in Figure 10f. One sample presented the substrate, and the estimated thickness was ~38 µm. The appearance is comparable with the other samples that reached the substrate because similar Fe oxides originated from the antagonist. The mechanism is still abrasive, and this result differs from what reported by Kong et al. [30] on electroless Ni–P treated at 400 ◦C. Prabu Ram et al. [17] also tested electroless Ni–P after heat treatment performed at 400 ◦C against AISI 440C steel. They observed the formation of an iron oxide layer that adheres on the coatings and has a strong influence on the wear mechanism. Delamination of the coating was not observed, and these results are in line with the present study. The wear system is different, but the similarities are due to the nature of the Fe-based antagonist materials. The overlapping of wear rate values for the two conditions mirrors the COF trends in Figure 9c and the similar hardness values measured for the HT400 samples (Figure 6). The presented results for the HT400 treatment agree with Aslanyan et al. [21] despite the different coupling system. They reported similar COF trends

for both pure and Ni–P/SiC coatings after 1 h at 420 ◦C. Using corundum as the antagonist, the authors reported a wear volume loss higher for the composite coatings than the pure ones. They attributed it to the pull out of SiC particles during wear, that contributed to the abrasion of the material. The same observations apply to the non-abrasive coupling of the present study with 100Cr6 steel. On the other hand, Chang et al. [19] reported different results with zirconium oxide as antagonist. After the heat treatment carried out at 400 ◦C for 1 h, the pure coating presented a higher volume loss than the composite one. Similarly, the average COF was slightly higher for the Ni–P coating. The authors attributed these results to the formation of Ni5Si<sup>2</sup> during heat treatment, that increases ductility due to the small angle phase boundary. A similar behaviour was not observed with the samples of the present study.

The presence of reinforcing particles did not alter the COF values for HT360 and HT400 conditions; however, it affected the wear rate, particularly for the HT360 samples. It is critical to underline that the influence of heat treatment is determined by the P content within the high-P family. Ahmadkhaniha et al. [23] recently clarified that small variations in the P content, within 10 wt.%–16 wt.%, led to different mechanical properties after heat treatment. Thus, the P content is the dominant variable, yet it cannot be controlled during the production process, either electrodeposited or electroless. Nevertheless, the samples of the present study are of interest because they represent the real condition in production systems, with P content in the 10 wt.%–16 wt.% range. The difficulties in optimising specific heat treatments stem from the variation of composition obtained in real production.

#### **5. Conclusions**

The present work investigated the wear behaviour of Ni–P and Ni–P/SiC coatings in the as-plated condition and after different heat treatments. The antagonist was a 100Cr6 steel, representative of non-abrasive systems. The interest of the present study was the representation of the real industrial production of Ni–P coatings and the related variations in P content. The aim of the present work was to differentiate the contributions of SiC addition and heat treatment on the wear properties of Ni–P coatings, to find the optimum condition that minimizes the wear rate.

The presence of SiC particles significantly influenced the COF and the wear rate of the as-plated coatings, despite the limited influence in increasing hardness. The presence of Fe oxides, originated in the tribo-oxidation wear mechanisms on the antagonist side, did not completely mitigate the ceramic particles' abrasive action in the composite coatings. These presented higher COF and wear rate than the pure ones.

After heat treatment, the maximum hardness of pure coatings was after 1 h at 400 ◦C (HT400), while the maximum hardness of composite coatings was after 2 h at 360 ◦C (HT360). During wear tests, the COF trends reported a difference between pure and composite coatings only in the HT300 condition (300 ◦C for 2 h). In other cases, the COF stabilised at similar values. The wear mechanisms involved were mainly abrasion and tribo-oxidation, with the formation of lubricant Fe oxides at expenses of the antagonist steel. Micro-plough also occurred for the pure HT300 samples, and Fe oxides combined with SiC particles mitigated it in the composite coating. The COF trends were not mirrored by hardness and wear rates in the HT360 condition, that shown distinct values.

**Author Contributions:** Conceptualisation, C.Z. and M.M.; methodology, C.Z., D.A., L.L., and A.F.; resources, C.Z. and M.M.; data curation, D.A., L.L., and F.B.; writing—original draft preparation, D.A., L.L., and F.B.; writing—review and editing, A.F., C.Z., and M.M.; supervision, C.Z. and M.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on reasonable request from the corresponding author. The data are not publicly available due to privacy.

**Acknowledgments:** No acknowledgements are needed.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### *Article* **Microcracks Reduction in Laser Hardened Layers of Ductile Iron**

**Eduardo Hurtado-Delgado \*, Lizbeth Huerta-Larumbe , Argelia Miranda-Pérez and Álvaro Aguirre-Sánchez**

Corporación Mexicana de Investigación en Materiales, COMIMSA, Ciencia y Tecnología 790, Saltillo 25290, Coahuila, Mexico; lahuertal1990@gmail.com (L.H.-L.); argelia.miranda@comimsa.com (A.M.-P.); alvaroits@hotmail.com (Á.A.-S.)

**\*** Correspondence: eduardoehd@alitmac.com; Tel.: +52-844-411-3200

**Abstract:** A study of surface hardening of Ductile Iron (DI) with and without austempering heat treatment was developed. The chemical composition of the material contains alloying elements such as Cu and Ni, that allow to obtain a Ductile Iron Grade 120-90-02, based on ASTM A536, which was heat treated to be transformed to Austempered Ductile Iron (ADI). Specimens of 10 × 10 × 5 mm<sup>3</sup> were obtained for application of surface hardening by Nd:YAG UR laser of 150 W maximum power. The parameters used were advance speed of 0.2 and 0.3 mm/s and power at 105, 120, 135 and 144 W; the departure microstructures were fully pearlitic in the samples without heat treatment, and ausferrite for austempered samples. Microstructural characterization of hardened samples was performed were analyzed and martensite and undissolved carbides were identified in the pearlitic samples, while in ausferrite samples it was found finer martensite without carbides. The depth of hardened surface to different conditions and their respective microhardness were measured. The results indicate that the surface hardening via laser is a suitable method for improving wear resistance by means of hardness increment in critical areas without compromising the core ductility of DI components, but the surface ductility is enhanced when the DI is austempered before the laser hardening, by the reduction of surface microcracks.

**Keywords:** laser hardening; ausferrite; austempered ductile iron; nodular iron; heat treatment

#### **1. Introduction**

*1.1. Characteristics of the Ductile Iron and the Austempered Ductile Iron*

Ductile Iron (DI) is commonly used in many engineering applications, like sheet forming dies and rolling mills, as reported in literature [1,2]. Their high manufacturability and machinability represent an excellent combination of economic application performances [1,3,4]. By subjecting the DI to heat treatment, it transforms to Austempered Ductile Iron (ADI), which is essentially a spheroidal graphitic iron with ausferrite microstructure comprising mainly low carbon ferrite (α) and high carbon retained austenite (γ). Because of an excellent combination of strength, ductility, toughness and fatigue resistance [5], ADI is now being increasingly used in key automobile components like crank shafts, steering parts, camshafts and gears [6–8], sometimes substituting steel parts [9].

#### *1.2. Laser Surface Hardened Melting*

Despite the good properties of DI and ADI, under some operating conditions such as erosive and corrosive environments its performance is limited by their relative low hardness [10–12]. This problem can be overcome by improving the surface properties of DI. High-power laser treatment (Nd: YAG, CO2) is found as a significant technique to enhance the mechanical properties of ductile iron according to [13,14], including multipass and surface alloying using different powders [15,16], as reported by some authors who proposed some general guidelines for this process. Nevertheless, the presence of microcracks in the hardened case and other surface defects, as reported by [17], constrain

**Citation:** Hurtado-Delgado, E.; Huerta-Larumbe, L.; Miranda-Pérez, A.; Aguirre-Sánchez, Á. Microcracks Reduction in Laser Hardened Layers of Ductile Iron. *Coatings* **2021**, *11*, 368. https://doi.org/10.3390/ coatings11030368

Academic Editor: Mattia Merlin

Received: 15 February 2021 Accepted: 18 March 2021 Published: 23 March 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

its applicability. Zheng et al. in 2013 proposed a novel technique in order to avoid crack formation in multiple overlapping laser tracks that represent a potential problem that must be reduced as possible [18].

To improve the wear resistance of the ductile iron, laser surface modification, without remelting, has been used in industrial applications, as it prevents failure by propagation of surface cracks [19,20]. In those cases when the iron has a ferrite-pearlite or even only a ferrite matrix, however, it is necessary to use laser hardening by melting the surface layer. This procedure creates a thin, microstructurally modified surface layer with a higher hardness. This layer consists of two parts: a melted zone and a heat hardened zone. The depth and width of the modified layer depend directly on the energy distribution and laser-beam diameter on the workpiece surface, the laser beam speed with respect to the workpiece, and the physical properties of the working material. The temperature and hence the properties of the surface can be controlled by the power density and the scan speed (in case of line hardening) or the interaction time (in case of single-shot hardening) of the beam [21].

Laser surface hardened melting (LSM) offers several advantages over other surfacemodification techniques. This process is a non-equilibrium method of surface modification reaching cooling rates around 103–108 K/s, which are considerably high. The resultant microstructures, some of them metastable phases, are mainly composed of unique properties that are only obtained with this process and not with conventional ones [22,23]. Laser hardening is usually constrained to low heat inputs in order to avoid surface microcracks in ductile irons, resulting in shallow hardened layers [24].

#### *1.3. Microstructure and Typical Hardness*

During laser hardening process, the surface of the irradiated material is heated in order to transform the microstructure of the heat affected zone into austenite. The surrounding material acts as an efficient heat sink, quickly cooling the material likely below the martensite start temperature [19].

The present microstructure after treatment LSM depends on the parameters used in the process as well as the initial microstructure of the workpiece. In this way, Benyonius [10] reported that if the microstructure in the DI is ferritic, after LSM treatment, eventually it will be formed a microstructure of dendrites made of austenite, surrounded by continuous networks of Fe3C and some martensite needles within the austenite islands. Alabeedi [23] presented an LSM treatment in a ferritic DI and showed that the laser melting led to complete dissolution of the graphite nodules which on solidifying created an interdendritic network of ledeburite eutectic with a very fine structure, good homogeneity and high hardness (650 HV). In another paper presented by Fernández [2], it was studied the effects of laser surface treatment on the microstructure, crackability and stresses generated on laser hardened layers produced in several ductile cast iron materials; in the study, two principal types of spheroidal graphite were selected. Considerable cracking by thermal stresses was produced on both irons, pearlitic and acicular bainitic, the energy densities achieved was above 40 J/mm<sup>2</sup> . It was observed that lower energy densities refrain cracking but only in the pearlitic ones, this was achieved by the excessive austenite retention that controlled the generation of transformational stresses. Grum [25] reported that in case of having an initial pearlitic matrix in the ductile iron, after a laser surface hardening, the resultant and predominant structure produced in the surface is martensite.

Regarding ADI that presents ausferrite microstructure initially, Roy [26] observed that the structure of the laser surface melted area was mainly austenitic, while a higher microhardness of more than 1000 HV happened with a martensitic microstructure. Furthermore, LSM produced more compressive residual stresses and enhanced significantly the wear resistance of the austempered ductile iron [19]. Putatunda [27], who applied laser hardening techniques, carried out an investigation on ADI. They used manganese phosphate coatings and colloidal graphite to achieve more uniform hardness. The hardness

values reached were around 700 HV and the microstructure of a thin hardened martensitic layer improved the mechanical properties of the material.

Amirsadeghi [28] studied the microhardness and wear resistance of different microstructures formed by tungsten inert gas (TIG) surface melting and chromium surface alloying (using ferrochromium) of ADI. Surface melting resulted in the formation of a ledeburitic structure in the melted zone, and this structure has hardness up to 896 HV, as compared to 360 HV in that of ADI. The results also indicated that surface melting reduced the wear rate of the ADI by approximately 37%. Finally, in a work presented by Grum [25], it was studied the laser surface-melt hardening in gray and nodular irons, and found that the melting produced by low-power laser beam can obtain an adequately modified hardened layer, which results in an increment of the surface wear resistance. Material properties play a dominant role in determining the interaction between the laser beams and engineering materials. Many material properties change with temperature. The mechanical properties of many engineering materials may be favorably modified by application of a suitable heat treatment, which can be full or superficial [29]. One of the most important superficial treatments of metals has been the laser transformation hardening of steel [29,30], but this treatment can also be applied successfully to ductile irons.

In the present study, an attempt has been made to enhance surface hardness and wear resistance of DI with and without austempering heat treatment. The aim of this work is to show that the austempered heat treatment before laser hardening of ductile iron is effective in reducing the amount of surface microcracks in a wide range of heat inputs.

#### **2. Materials and Methods**

#### *2.1. Ductile Iron*

The nodular iron utilized for these experiments corresponds to 120-90-02 grade, under the ASTM A536 [31] standard and it has a chemical composition that is typical for this type of irons. Cu and Ni were added to increase the amount of pearlite in the as-cast microstructure. The chemical composition is shown in Table 1.


**Table 1.** Chemical composition (wt. %) of the Ductile Iron (DI).

The carbon equivalent for this ductile iron was of C<sup>e</sup> = 3.82, which is defined as hypereutectic iron. Besides this, eutectic saturation was calculated as S<sup>c</sup> = 1.05 considering Si, Mn and P, according to the equation presented in [32].

#### *2.2. Austempered Ductile Iron*

The DI samples were fully austenitized at 900 ◦C for 120 min and austempered in an isothermal salt bath at 340 ◦C to 360 ◦C for 60 min followed by cooling in air at room temperature. The salt bath was 60% KNO<sup>3</sup> and 40% NaNO<sup>2</sup> a schematic diagram of the austempering heat treatment process is shown in Figure 1.

**Figure 1.** Schematic heat treatment process.

#### *2.3. Experiment Design and Laser Parameters*

In this study, DI samples with and without austempering heat treatment were surface hardened by UR LaserTechnologie Nd:YAG laser of 150 W maximum power. Diameter of laser spot was about a half of the bead width, which can be observed as the length of the transversal fusion zone near the surface. It is difficult to assign a defined value because there is a power density reduction from the center to de periphery, following a Gaussian function. However, the effective spot size, in this case can be considered as that which produces the fusion the metal in the axial direction; it is approximately the full length of the fusion zone at the half depth, shown in the micrographs in results section. The specimen dimensions were 10 × 10 × 5 mm. The selected parameters for the surface treatment were the following:


Four samples were used for the experiment: two DI and two ADI samples. For each sample, four laser-melted beads were produced, one for each power level, according to the Figure 2 schemes.

**Figure 2.** Design of experiments for the laser surface hardening.

#### *2.4. Laser Surface Melting-Hardening of DI and ADI*

Cross sections of the samples were cut for metallographic examination. The microstructural characterization was consisted in grinding (using 120, 240, 320, 600, 800 SiC paper) and polishing with a one µm diamond paste, using Nital at 1% as reactive etchant for 5 s. The samples were inspected in the different seams weld zones with an optical microscope (OM) Nikon Eclipse MA200 and electron microscope Tescan Mira 3; besides this, hardness examinations in each zone were performed using 300 g for Vickers indentation. Microhardness evaluation, using a Wilson hardness Tukon 2500 equipment, was performed in order to compare the parameters effect in the weld beads and to find a proper

combination with the higher hardness without cracks. Wear resistance is favored with these characteristics [33]. Figure 3 shows each zone area where RZ and HZ correspond to the re-melted zone and hardened zone (affected by the heat), respectively. All the indentations for microhardness profile are shown as well: 3 indentations for each position (H1, H2, H3, H4) in order to obtain the average and more reliable results.

**Figure 3.** Current zones in the weld seam of the laser surface hardened melting (LSM).

#### **3. Results and Discussion**

#### *3.1. Base Material*

The original microstructures for DI samples before the surface hardening are shown in Figure 4. Microstructure consist mainly of pearlite and graphite nodules, around 30 µm diameter, with a composition of 84% and 12%, respectively, and the difference may indicate segregation zones predominantly Ni and Cr (Figure 4a). The microstructure was measured by image analysis with Image Pro Software and NIS Element coupled to the OM; for the ADI samples (Figure 4b), it was present ausferrite, graphite nodules and austenite islands, and the segregated zones disappeared with the austempering. The compositions of these phases were 83%, 11% and 6%, respectively, measured by image analysis.

 **Figure 4.** Optical micrograph, (**a**) DI (Perlite + Nodules + Segregations), (**b**) Austempered Ductile Iron (ADI; Ausferrite + Nodules + Austenite).

On average hardness in the DI was 295 HB and heat treatment increased the ADI to 314 HB.

#### *3.2. Dimensions and Morphology of the Melted Zone*

As the iron castings are a mixture of phases of iron and graphite, when the metal is re-melted by the laser it dissolves all the free graphite in the liquid and in the subsequent and fast cooling it results in an oversaturated carbon alloy, mainly formed by martensite, some retained austenite and iron carbides. The proportion of these phases depends upon the maximum temperature reached, the holding time at this temperature and the cooling rate. The first variable depends on the laser power, and the last two variables depend on the advanced speed of the laser beam and the thermal metal properties.

The DI can be hardened by a laser beam because of the great amount of carbon contained in the microstructure which can be dissolved and form martensite after the fast fusion of the metal, taking advantage of the high-density power of the laser beam [34]. This can be realized using even a low power equipment, of only 150 W, like that used in this study. In Figure 5 shows the morphology of the left half beads produced with the laser at its highest power, 144 W and their dimensions. DI beads are wider and deeper than their corresponding ADI seams with slower speed. They also have more and larger surface cracks. DI cracks are observed as larger as the hardened layer, as in S1P1 where its value corresponds to 371 µm, crossing thoroughly the melted zone. Cracks in the hardened iron are in the diagonal and vertical direction. As reported for surface alloyed carbon steels [35,36], the presence of cracks is due to hot cracking. Since susceptibility for hot cracking is determined by the alloy plasticity and solidification temperature range (∆T); in the surface area, a composition near the eutectic point (3.4–4.5 C wt.%) is expected, so ∆T is small, resulting in some plasticity and fine dendritic structure at laser temperatures. Therefore, once the metal is partially solid, crack appearance depends on the thermal contraction of the remaining molten metal.

**Figure 5.** Seams dimensions of DI: Sample 1 (S1) and ADI: Sample 3 (S3), with 0.2 mm/s speed.

If the advance speed of the laser source is changed from 0.2 mm/s to 0.3 mm/s, DI beads are wider than their corresponding ADI beads, in all cases, but not necessarily deeper; this can be observed in Figure 6, where the depth of ADI increases except for the highest power. Moreover, due to their greater depths of melted zone, 0.2 mm/s speed presented greater nodule dissolution.

**Figure 6.** Beads dimensions of DI: Sample 2 (S2) and ADI: Sample 4 (S4), with 0.3 mm/s speed.

The results shown in Tables 2 and 3 indicate that the power input density was between 325 J/mm<sup>2</sup> and 579 J/mm<sup>2</sup> for speed of 0.2 mm/s; furthermore, the power density was between 248 J/mm<sup>2</sup> and 385 J/mm<sup>2</sup> for 0.3 mm/s. This outcome is far from the power density of 40 J/mm<sup>2</sup> , reported by [2]. Nevertheless, there is no evidence of crack appearance at the lowest power used in the experiments of this work, so the later reference value is likely conservative.

**Table 2.** Bead measures in different samples (0.2 mm/s speed).



**Table 3.** Bead measures in different zones (0.3 mm/s speed).

#### *3.3. Microstructure*

3.3.1. Base Metal

Figure 7a shows the pearlitic initial microstructure of the DI samples; the pearlite is constituted of Fe3C lamellae with ferrite, and the hardness value is around 320 HV. In Figure 7b, the ausferritic microstructure that is characteristic of ADI is presented; this structure has a hardness of approximately 410 HV and consists of high carbon austenite plates with ferrite. Ausferrite microstructure refinement is dependent of the austempering temperature. In this case, the ausferrite is not as fine as could be under lower austempering temperatures [37].

 **Figure 7.** Microstructure, (**a**) DI: Pearlite, (**b**) ADI: Ausferrite.

#### 3.3.2. High Power Remelted Zone

On the fusion zone (H1) from DI samples, sample 1 (S1) presented in Figure 8a has a large amount of refined acicular Fe3C (AFC), high carbon martensite (HCM) at the bottom and some retained austenite (γr); for that reason, the reached hardness in this area was >1000 HV. On the other hand, sample 2 (S2) obtained a lower hardness (731 HV). As can be seen, the hardness reduction from S1 to S2 is due to the increase of speed, Fe3C is present in lower quantities and is coarser (ACC), the presence of HCM increased as well as γr, and the carbides decreased and some of them were separated from martensite to be grouped and form a platelike constituent (Figure 8b).

 **Figure 8.** Microstructure of the melting zone (H1) to 150 W (P1), (**a**) Sample 1, (**b**) Sample 2, (**c**) Sample 3 and (**d**) Sample 4.

Regarding the ADI, the reached temperatures on the fusion zone were lower than those for the DI; for that reason, the dissolved carbon amount was minor. Figure 8c shows the H1 zone microstructure which corresponds to sample 3 (S3); it is constituted of HCM islands surrounded by Fe3C fine plates (PFC) and small amounts of γr, with almost 968 HV. This microstructure was formed due to the lower laser speed and the higher temperature reached; the γ from the sample was not rich in carbon, which prevented its stabilization at room temperature and, from the rapid cooling, the microstructure transformed mostly to HCM [34,38] and the surrounded liquid to Fe3C.

At higher advance speed, the maximum temperature in the sample is lower and the arising γ dissolves more carbon that stabilizes it at room temperature. Figure 8d exhibits sample 4 (S4), on the H1 zone, and presents large quantities of γ<sup>r</sup> in form of islands surrounded by coarser and greater Fe3C plates (PCC). Some zones, wherein γ could not dissolve too much carbon, transformed into HCM with rapid cooling. The higher amount of γ<sup>r</sup> in this sample is the cause of the hardness decay (864 HV) compared to S3.

Using the diagram in Figure 9, the microstructures from different samples were deducted. In this diagram [39] the carbon concentration curves as a function of the cooling and heating rates are presented, as an example. The amount and type of resultant microstructures depend on the maximum temperature reached, as well as the heating and cooling rates, which in turn depend on the power and advance speed of the laser beam. This can be visualized using the Fe-C-Si phase diagram, the heating and cooling cycles superimposed for a high speed-high power, and a low speed-low power beams, both acting upon a surface of nodular iron with a fully pearlitic matrix (i.e., 0.8% C). In the first case, fusion zone, a high temperature is reached very fast, but there is no time to dissolve a great amount of carbon from the graphite spheroids; after the fast heating, a fast cooling is followed and the result is an austenite with low carbon content which is transformed to a mixture of martensite and some γr, surrounded by Fe3C, which arose from the molten metal, as the final microstructure. In the second case (low speed-low power), the reached temperature is lower, but there is more time to dissolve carbon and, according to the Fe-C-Si diagram, carbon has more solubility in austenite at lower temperatures above the eutectic; after heating, the cooling is faster but, because the greater content of carbon, less austenite

transforms to martensite, since M<sup>s</sup> (martensite start transformation temperature) is lower, and the austenite is more stable at ambient temperature in this condition. The total amount of molten metal can be greater, at lower speed (and lower T), but the liquid volume fraction is lower because, at lower temperature, more austenite can coexist with the liquid; the carbon dissolution has the effect to lower the liquidus temperature in the Fe-C-Si system. Consequently, the volume fraction of cementite formed at the end is greater when higher advance speed is used, even when the temperature had been lower. It is important to know this, because the number and extension of surface microcracks depend on the amount of martensite and cementite in the microstructure.

**Figure 9.** Schematic Fe-C-Si diagram [39].

3.3.3. Low Power Remelted Zone

γ Due to the low temperature reached on the H1 zone (P4) in S1 sample the microstructure formed (Figure 10a) consisted mostly in γ<sup>r</sup> with HCM islands surrounded by PFC; the amount of carbides decreased compared to the P1 power, and for that reason the hardness decay >1000 HV (P1) to 633 HV (P4). In Figure 10b are shown the microstructures obtained in the H1 zone from S2 sample; unlike the previous one, more presence of PFC is evident since the required liquidus temperature in this zone was greater than S1 at the same power, due to the lower carbon dissolution.

γ

 **Figure 10.** Melting zone (H1) microstructure 70 W (P4), (**a**) Sample 1, (**b**) Sample 2, (**c**) Sample 3, (**d**) Sample 4.

Instead, for ADI at P4 power with 0.2 mm/s the maximum temperature barely melted the metal; it exceeded the liquidus line but did not get to dissolve a considerable amount of nodules; when cooling begins, the material transforms to austenite with high carbon content and a small quantity of liquid, at certain intermediate temperature this small quantity of liquid transforms to ledeburite and high carbon austenite. A slight amount of Fe3C that encloses the γ<sup>r</sup> grains arises from ledeburite. The sample at higher speed S4\_P4 did not reach the liquidus line when it started to cool; the liquid that forms, as well as the austenite, has lower carbon content than the maximum achievable, which corresponds to the eutectic, just at the maximum solubility of austenite; at high cooling rate the liquid rapidly reaches an inferior temperature than the eutectic, starting the cementite formation. Most of the prior austenite, with insufficient carbon to prevail as a metastable phase at low temperature, transforms to martensite.

#### 3.3.4. HAZ

As observed in Figure 11, the heat affected zone is very similar for both materials and both speeds; it consists mainly of martensite and retained austenite as mentioned in [16]. The structure is finer for the DI samples because it reached lower temperature and the carbon homogenization is faster in the pearlitic condition than in the ausferritic.

**Figure 11.** Microstructure of the HAZ (H2) to 70 W (P4), (**a**) Sample 1, (**b**) Sample 2, (**c**) Sample 3, (**d**) Sample 4.

#### *3.4. Cracks and Microhardness*

#### 3.4.1. High Hardness

Table 4 shows microhardness results obtained at different positions of all samples. The beads of DI that presented higher hardness values were those with the lower speed and higher power (P1), this is due to most of the microstructure being Fe3C in S1\_P1; the Fe3C was in acicular form and martensite in small amounts. For that reason, high hardness was obtained >1000 HV, so that it caused embrittlement at cooling producing a great number of cracks (Figure 5). On the other hand, S2\_P1 presented a coarser Fe3C with larger martensite volume fraction than S1\_P1; therefore, the hardness resulted lower (731 HV) and the cracks presence was reduced, hence the sample was less fragile, as pointed out in [2].

**Table 4.** Microhardness at different positions (HV).


At the same P1 in the ADI samples, S3 achieved hardness values of 968 HV that are close to S1\_P1, the crack appearance was abruptly reduced due to the microstructure but graphite flotation was evident on the surface. In S4\_P1, the hardness value was 867 HV since γ<sup>r</sup> transformation, compared to S3\_P1, the cracks were smaller and graphite flotation was removed.

#### 3.4.2. Low Hardness

At lower power and speeds, the hardness decreased significantly from 1022 HV (S1\_P1) to 633 HV (S1\_P4), and crack presence was eliminated due to the resulted microstructure. At higher speed hardness, decay was not evident for beads P1 to P4, since it passes from 731 HV to 622 HV. In the ADI at lower power a similar behavior was maintained as in DI samples, since hardness decrease was more noticeable at lower speeds (968 HV a 542 HV); this behavior was also reported in [19].

#### **4. Conclusions**

The parameters which demonstrate improved performance regarding hardness, dimensions and crack formations were:


Surface hardening by laser treatment is reliable, but cracks are generated during the solidification if not properly applied.

ADI re-melted beads are more narrow than their corresponding DI beads.

The highest hardness was 1145 HV obtained from the DI condition, without austempering heat treatment.

The DI samples presented more and larger cracks in all experimental conditions because the contraction of cementite during cooling.

ADI is less prone to crack formation than DI, because it contains less cementite and more martensite, and because ausferrite is more heat-conductive than pearlite.

Concisely, this work has demonstrated with no doubt that the ADI is a better option for laser hardening than DI, because the former can dissipate the heat input faster and more evenly, due to the thermal characteristics of both materials, identical in chemical composition, but not in phase composition. ADI has a metal conductive matrix with carbon saturated austenite and graphite nodules, while DI has a mixture of metal and ceramic (pearlite) matrix and graphite nodules which is less heat-conductive. Microcracks are related to the excessive accumulation of heat, which produces higher thermal gradients and formation of greater amounts of carbides.

**Author Contributions:** E.H.-D. conceived and planned all experiments. L.H.-L. and Á.A.-S. performed all experimentations. A.M.-P. and L.H.-L. wrote the manuscript, and all authors participated in results analysis and discussion. All authors have read and agreed to the published version of the manuscript.

**Funding:** The present work was funded by CONACyT Mexico under the project Proinnova 216536.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data discussed in this contribution is available on request from the corresponding author.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

1. Nêmeˇcek, S. Surface of cast iron after laser hardening. *Adv. Mater. Res.* **2013**, *685*, 92–96. [CrossRef]


## *Article* **Microstructural and Erosive Wear Characteristics of a High Chromium Cast Iron**

**Annalisa Fortini \* , Alessio Suman , Alessandro Vulpio, Mattia Merlin and Michele Pinelli**

Department of Engineering, University of Ferrara, 44122 Ferrara, Italy; alessio.suman@unife.it (A.S.); alessandro.vulpio@unife.it (A.V.); mattia.merlin@unife.it (M.M.); michele.pinelli@unife.it (M.P.) **\*** Correspondence: annalisa.fortini@unife.it

**Abstract:** Surface material loss due to erosive wear is responsible for the increased cost of maintenance and downtime in industries. Hence, hardfacing is one of the most valuable and effective techniques employed to improve the wear resistance of heavy-duty components. The present paper investigates the microstructural and erosive wear characteristics of a hypereutectic high-chromium cast iron, considering the erosion resistance, resulting from the impact of micro-sized particles, of both as-received and heat-treated conditions. Micro-sized particles involve the erosion-resistant characteristics of carbide and matrix, contemporary. Due to this, the enhancement of the matrix strength could improve the mechanical support to withstand cracking deformation and spalling. Accordingly, the effect of a destabilization heat treatment on the microstructure was firstly investigated by hardness tests, X-ray diffraction analyses, optical and scanning electron microscopy. Specifically designed erosive tests were carried out using a raw meal powder at an impingement angle of 90◦ . The resulting superior wear resistance of the heat-treated samples was relayed on the improved matrix microstructure: consistent with the observed eroded surfaces, the reduced matrix/carbides hardness difference of the heat-treated material is pivotal in enhancing the erosion resistance of the hardfacing. The present results contribute to a better understanding of the microstructure–property relationships concerning the erosive wear resistance.

**Keywords:** hardfacing; high chromium cast iron; heat treatment; erosion tests; wear resistance

#### **1. Introduction**

A widespread solution in industrial applications to extend the service life of components, in machinery equipment or construction, is found in cladding certain areas of the surface exposed to various severe wear conditions. Within the different surface coating protective and hardening techniques, hardfacing is one of the most adopted due to its low-cost and easy handling characteristics [1,2]. Hardfacing enables enhancing the corrosive, abrasive, and heat resistance properties of a metal workpiece's surface, creating a cladding metal layer with improved features [3]. Iron-based hardfacing alloys are the most widely-used thanks to their good wear resistance and low cost. Among these, due to their superior wear resistance, High Chromium Cast Irons (HCCIs) are broadly employed in both abrasive (e.g., grinding media) and erosive (e.g., slurry, gravel, and dredge pumps) applications [2]. Accordingly, HCCIs are widely used in minerals and mining industries, cement plants, paper and pulp industry, thermal power plants, iron and steel industries, etc.

Considering that hardfacing is usually deposited on the substrate by welding techniques, the microstructure of Fe–Cr–C hardfacing alloys resulting from a non-equilibrium solidification process consists of a Fe–Cr solid solution phase and complex carbides, depending on the Cr and C contents of the alloys. Cr-rich cast irons with hypereutectic structure, i.e., primary M7C<sup>3</sup> carbides surrounded by eutectic austenite and M7C<sup>3</sup> carbides [2,4–6], show high hardness and superior wear resistance to the hypoeutectic ones. The excellent abrasion resistance of hypereutectic alloys stems from the dispersion of the

**Citation:** Fortini, A.; Suman, A.; Vulpio, A.; Merlin, M.; Pinelli, M. Microstructural and Erosive Wear Characteristics of a High Chromium Cast Iron. *Coatings* **2021**, *11*, 490. https://doi.org/10.3390/ coatings11050490

Received: 30 March 2021 Accepted: 19 April 2021 Published: 22 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

hard (1300–1800 HV) M7C<sup>3</sup> eutectic carbides [7,8]. Despite that, the as-cast condition of hypereutectic HCCIs cannot meet the demand of heavy impact conditions due to the difference in hardness between the matrix and the carbides. Hence, while on the one hand hypoeutectic HCCIs are still used in high demanding environments, on the other hand, efforts are devoted to enhancing the wear resistance of both hypo- and hyper-eutectic alloys [9]. HCCIs may be regarded as composite materials as they show a structure composed of large eutectic M7C<sup>3</sup> carbides in a softer iron matrix. Hence, a good combination between matrix and carbides' hardness and toughness should be tuned to enhance the overall resistance [10].

Many studies have been focused on the improvement of wear resistance of HCCIs by the addition of strong carbide forming elements such as W [11,12], V [13], Nb [14,15], Ti [16], Mo [17,18], and B. Alloying elements are added to the melt to promote the precipitation of abrasion-resistance MC carbides (M is the metal and C is the carbon), stronger and harder than M7C<sup>3</sup> ones [19–22]. At the same time, attention has been paid to the size, distribution, and volume fraction of carbide phases that overall affect the wear resistance of the alloy [23].

Moreover, the toughness and the strain hardening behavior of the matrix influence the wear resistance of the hardfacing since it should provide the mechanical support to withstand cracking deformation and spalling [7]. Several attempts have been made to improve the tribological and wear behavior of HCCIs through heat treatments [10,24–26] and mechanical treatments [27]. In the as-cast condition, the matrix is mainly austenite that, upon specific thermal treatments, is transformed into martensite. While the primary carbides formed inside the melt after casting and the eutectic carbides formed at about 1250 ◦C are not altered by heat treatment, the metastable austenite transforms to martensite through proper destabilization and subcritical treatments. To induce austenite to martensite phase transformations and, in turn, the overall hardness increment, heat treatments comprise heating for 1–6 h in the range of 900–1100 ◦C are usually applied [28].

In the last few decades, much research has been devoted to explaining the role of microstructural characteristics in the wear behavior of hardfacing alloys through specific wear tests (i.e., pin-on-disk or wheel, slurry erosion test, impact erosion resistance) [29–33]. Renewed attention has been paid to the impact erosion resistance of white cast irons, investigated through solid particle erosion tests [3,9,24,34–36]. Data from several studies have proved that wear resistance is not mainly influenced by the bulk hardness of the material, but it is a complex phenomenon derived from several factors, like type, volume fraction, size, and morphology of eutectic carbides together with their interaction with the matrix [7,37]. Moreover, the microstructural modifications of the matrix resulting from different heat treatments applied to destabilized the austenite are pivotal in the assessment of the wear resistance performance. Finally, to the authors' knowledge, there is no evidence about research on the wear behavior of HCCIs subjected to impact erosion tests with micrometric particles with diameters in the range of 1–10 µm, commonly used in the industrial process but very different from the standard powder used for erosion tests.

In the light of the above, the main objective of the present study is to investigate the microstructural features and wear erosive behavior of a commercially available hypereutectic HCCI. Accordingly, this paper begins with a synopsis of the literature regarding HCCIs and, more specifically Fe–Cr–C hardfacing alloys, to highlight the microstructural features responsible for their tribological behavior. The second section describes the adopted methods to assess the chemical, microstructural and mechanical features of the hardfacing alloy. Moreover, the erosion tests conducted through a dedicated test rig and the effect of a destabilization heat treatment on the microstructure and, in turn, on the erosion resistance, are explained. The third section of the paper describes the microstructural characterization of the hardfacing alloy, of nominal composition Fe–22Cr–4.8C wt.%, brings together the findings of hardness tests, optical microscopy, electron scanning microscopy and X-ray diffraction analyses. Further, considering the role of microstructural features on the wear properties of the alloy, the destabilization of austenite, through a heat treatment at 950 ◦C

for 3 h + oil quenching conducted to promote martensite formation and secondary carbides precipitation, is presented. The wear behavior of the as-received and heat-treated samples is investigated through a dedicated test rig, by simulating the operating condition of the considered HCCI alloy used in a large-sized centrifugal fan [38]. The observed improved erosion resistance is described and discussed. Lastly, the fourth section summarizes the main findings of the research study. The novelty of the present paper is related to the investigation of the wear-resistant behavior of the hardfacing alloy, before and after heattreatment, used to overcome the erosion effects due to the impact of micro-sized particles. This latter characteristic has to be considered not only against the carbide hardness but also on the carbide morphology and matrix characteristics. This experimental study does not engage with a comprehensive evaluation of the destabilization temperature and time parameters that could be tuned to improve the erosive wear resistance of the investigated HCCI alloy. Likewise, it is beyond the scope of this preliminary study to examine the role of carbide spacing against the particle size distribution of the erodent powder.

#### **2. Materials and Methods**

The hardfaced plate analyzed in this study is a layer-composite wear plate, commercially available as EIPA 550 (Eipa Eisen Palmen GmbH, Aachen, Germany), made by the open-arc welding of a flux-cored wire. The HCCI hardfacing electrodes were deposited on a low carbon steel plate. The nominal thickness of the base steel, as well as of the hardfacing, is 5 mm. The chemical composition of the welded layer and the substrate were determined through the Glow Discharge Optical Emission Spectrometry (GD-OES, Spectruma Analitik GDS 650, Hof, Germany) technique. Table 1 reports the chemical composition of the alloys: note that for the HCCI the composition was analyzed in the cross-section considering both an acquisition area near the resistant side (at about 500 µm far from the top surface), named RS, and an acquisition area near the substrate side (at about 4500 µm far from the top surface), named SS. The HCCI is an Nb- and Mo-rich alloy, with a Cr/C ratio of about 5.


**Table 1.** Chemical composition (wt.%) of the hardfacing plate.

The microstructure of the hardfacing alloy was evaluated by sampling longitudinal sections, named L (parallel to the direction of welding), and cross-sections, named T (perpendicular to the direction of welding), following the basic steps for proper metallographic analysis, i.e., cutting, mounting in resin, grinding, final polishing and etching. The latter was conducted by Kalling's No. 2 reagent (5 g CuCl2, 100 mL HCl, 100 mL C2H5OH) to reveal the microstructure: the samples were immersed in the reagent for 5 s, rinsed with ethanol, and air-dried. Metallographic investigations were conducted through a Leica DMi8A (Leica, Wetzlar, Germany) optical microscope (OM) and a Zeiss EVO MA 15 (Zeiss, Oberkochen, Germany) scanning electron microscope (SEM), equipped with an Oxford X-Max 50 (Oxford Instruments, Abingdon-on-Thames, UK) energy dispersive microprobe for semi-quantitative analyses (EDS). The SEM micrographs were recorded in secondary electron imaging (SEI-SEM) and back-scattered electron (BSE-SEM) modes.

Crystallographic phase identification was performed by X-ray diffractometry (XRD) with a Bruker D8 Advance (Bruker, Billerica, MA, USA) diffractometer, equipped with a Cu filament (Kα, 1.5406 Å). All patterns were acquired in the 2θ range of 30◦ to 80◦ with 0.02◦ of step-size and 1 s of step time.

Quantitative metallographic analyses, after preliminary post-processing of the optical micrographs, using the MATLAB® Color Thresholder app and then evaluation through

μ

the Leica LAS (Leica Application Suite) V4.9 software, enabled to examine the Carbide Volume Fraction (% CVF). For each sample, the analyzed region comprised a total of 20 micrographs for the SS and the RS, respectively.

α θ

Bulk hardness measurements were carried out on the polished cross-sections of the samples to evaluate the variation in hardness from the resistant side of the hardfacing to the steel substrate of the wear plate. The Vickers hardness measurements, under 1000 g test load and 15 s loading time (HV1), were carried out by a Future-Tech FM-110 (Future-Tech Corp., Kawasaki, Japan) Vickers indenter, in agreement with the ASTM E92 standard. Moreover, Vickers microhardness on the T cross-sections of the polished and etched hardfacing alloy was also evaluated on both the matrix (test load of 200 g and 15 s loading time, HV0.2) and on the carbides (test load of 50 g and 15 s loading time, HV0.05). In all cases, the mean Vickers hardness was calculated from five indentations.

Erosion tests were carried out using an on-purpose built test rig inspired by the ASTM G76 standard [39]. Since the present analysis was conceived to explore the erosion behavior due to a micro-sized powder, the feeding systems, as well as the nozzle, were modified with respect to the prescribed ones to ensure a constant feeding rate and avoiding clogging phenomenon during the test. The raw meal powder considered in this investigation was experimentally characterized to define particle morphology, size distribution and physical characteristics. The erosion tests were carried out by a raw meal powder commonly used in a cement factory, able to form several agglomerates due to humidity. Digital and SEI-SEM images of a powder sample in the as-received condition are displayed in Figure 1a,b. The overall average density of the powder was evaluated by an AccuPyc II 1340 (Micromeritics Instrument Corporation, Norcross, GA, USA) pycnometer and it results equal to 2700 kg/m<sup>3</sup> . Moreover, a quantitative analysis of particle diameter distribution is carried out to evaluate the presence of different diameter particles (dp) within the powder. The particle size distribution is determined by a Mastersizer 3000 laser diffraction analyzer (Malvern Panalytical, Malvern, UK) and it is depicted in Figure 1c, which reports the number, labeled as N, and mass distributions, labeled as M, of the powder. The raw meal powder was characterized by an average diameter of 4.3 µm (*d*<sup>90</sup> = 9.7 µm). μ

**Figure 1.** (**a**) Digital and (**b**) SEI-SEM images of the powder in the as-received condition, (**c**) number, N, and mass distributions, M, of the raw meal powder.

The feeding system is comprised of two different systems to guarantee the constant dosing rate and the breaking process of the agglomerates. The first part is composed of a hopper, equipped with a calibrated screw able to dosing the powder in the reservoir connect to a Venturi nozzle. The Venturi nozzle is operated by dried and cleaned shop air in order to suck the powder from the secondary line connected to the throat section. By the shear force, the Venturi nozzle allows the breakup of the agglomerates [40,41] ensuring the repeatability of the erosion tests. After the Venturi, a cylindrical nozzle with an internal diameter of 4 mm and a length equal to 32 mm was mounted. The sample overlook the nozzle at a fixed distance equal to 10 mm according to the standard, as reported in Figure 2a and it is held by a sliding and tiltable table to adjust the relative angle between nozzle and specimen surface

as reported in Figure 2b. Samples with dimensions of 50 mm × 25 mm × 10 mm were cut (by abrasive water-jet process) from the wear plate and their top surface was ground, polished (mirror-like finishing up to 1 µm diamond paste), and ultrasonically cleaned in acetone. The erosion test used a constant powder-feeding rate (10 g/min), impingement angle (90◦ ), and particle impact velocity (100 m/s).

**Figure 2.** (**a**) Sample and nozzle setup, (**b**) image of the erosion sample on the sliding and tiltable table.

Three different erosion times, i.e., 30 min, 60 min, and 90 min, were investigated. For each condition, five samples were tested. After each erosion test, the sample was ultrasonically cleaned in acetone to remove any traces of the erodent powder. The erosion resistance was evaluated from the mass loss, computed by weighing each specimen before and after the erosion test. A Kern ABT 100-5NM (Kern, Balingen, Germany) analytical balance, with an accuracy resolution of 0.01 mg was used.

The worn surfaces were then analyzed by both optical and scanning electron microscopy in an attempt to provide a better understanding of the mechanisms of material removal. The investigations were conducted on the worn top surface and the crosssectioned surface. The latter was cut in the center of the erosion crater and polished using the above-described standard metallographic technique.

By using an LTF (Lenton Furnaces and Ovens, Hope, UK) tube furnace, the effects of a destabilization heat treatment on the resulting microstructure and, in turn, on the erosion behavior, were evaluated. Similar to the erosion times investigated in the as-received condition, i.e., 30 min, 60 min, and 90 min, also for the heat-treated condition five samples were tested at the same erosion times. To this end, the furnace was heated at 20 ◦C/min up to 950 ◦C, and then the samples were held at this temperature for 3 h to promote the destabilization of the austenite phase. The final oil quenching to room temperature enabled to precipitate a fine dispersion of the carbides within the matrix. The overall bulk hardness of the hardfacing, before and after the heat treatment, was evaluated under 30 kg load and 15 s loading time (HV30) by VH Metkon (Metkon Instruments Inc., Bursa, Turkey) Vickers hardness tester. The hardness of the welded layer was determined as the average of five indentations to check the reproducibility of the hardness data. Then, both microstructural investigations and erosion tests were carried out according to the above-described methods. Microstructural (OM, SEM/EDS) and crystallographic (XRD) investigations were performed on the heat-treated samples to evaluate the relationship between microstructural features and erosion resistance.

#### **3. Results and Discussion**

#### *3.1. Microstructural Investigations of the As-Received HCCI*

Figure 3a displays a digital image of the resistant side of the wear plate in the asreceived condition from which the bead pattern, i.e., juxtaposed passes with continuous

overlap, enables to counteract severe wear conditions. Moreover, stress relief cracks, resulting from the relaxation of heat stress in the deposit, develop at right angles to the weld beads and are regularly spaced. To ensure the wear resistance of the hardfacing, these cracks must not be spread to the base metal. Figure 3b shows the 3D isometric optical micrographs of the wear plate in the hardfacing/substrate interface. As can be seen from the cross-section, the crack does not reach the steel substrate, thanks to the buffer layer (the light gray band between the hardfacing and the steel) that acts as a barrier to cracking.

**Figure 3.** Hardfacing plate: (**a**) image of the resistant side of the wear plate; (**b**) 3D isometric optical micrographs of the unetched Fe-Cr-C alloy.

γ X-ray diffraction pattern of the deposit on the resistant-side layer is shown in Figure 4: the presence of MC (M = Nb, Mo), M7C<sup>3</sup> (M = Cr, Fe), and austenite phases can be detected. Per the liquidus projection of the iron corner of the Fe–Cr–C ternary system [42,43] and according to the evaluated chemical composition (Table 1), the present alloy hardface deposit falls in the hypereutectic range. Considering the presence of Nb in the alloy, the formation of MC (M = Nb) carbides precedes the formation of proeuctectic M7C<sup>3</sup> carbides [44–46]. MC carbides precipitate at a high temperature in the melt before the formation of proeutectic M7C<sup>3</sup> carbides, acting as heterogeneous nucleation sites and increasing the nucleation rate promoting the formation of finer proeutectic M7C<sup>3</sup> carbides [44]. As the molten temperature falls to the eutectic point, the residual melt rejects the Cr and C atoms: when the Cr and C concentrations reach the eutectic composition, the (γ + M7C3) eutectic colonies form [44].

**Figure 4.** X-ray Diffraction (XRD) patterns of the hardfacing on the resistant-side layer.

The microstructure of the hardfacing, obtained from the polished and etched crosssection samples, is reported in Figure 5. Figure 5a shows the optical micrograph of the T-section of the alloy on the resistant side: in the upper part, i.e., close to the surface, the

γ

γ

γ

hardfacing presents blade-like proeutectic M7C<sup>3</sup> carbides, with the longer axis perpendicular to the surface and whose orientation is related to the heat flow [47], and the (γ + M7C3) eutectic mixture.

With the increase of the distance from the resistant side, the carbides become finer, rod-like, and uniformly distributed throughout the matrix. The red arrows in Figure 5a highlight the stress-relief crack propagation path. SEM investigations were also conducted on the T cross-section of the hardfacing, at about half of the hardfacing thickness, and at the boundary between two welding passes. As revealed in the BSE-SEM image of Figure 5b, the different microstructure of the two passes, with coarser primary carbides on the left side of the micrograph and finer ones on the right side, is detectable. The dash-dot yellow line separates the passes with coarser and finer rod-like carbides. The optical micrographs of Figure 5c,d compare the microstructure of the SS (Figure 5c) and the RS (Figure 5d) regions of the L-section. Both reveal the presence of γ dendrites (highlighted by red arrows) adjacent to the proeutectic M7C<sup>3</sup> carbides, whose content is greater in the RS region (Figure 5d). This finding has also been recently reported and described in [44]. Moreover, the effect of the solidification conditions is detectable: near the surface, the higher thermal gradient results in primary blade-like M7C<sup>3</sup> carbides that grow along the preferential growth axis [47]. In addition to the high magnification optical micrographs of Figure 5c,d, BSE-SEM analyses were conducted to highlight the microstructural constituents through compositional contrast imaging. Hence, Figure 5e displays the BSE-SEM micrograph of the T-cross section, which provides an overview of the microstructure, while the high magnification micrograph of Figure 5f points out the eutectic carbides network. Consistent with the XRD results (see Figure 4), the BSE-SEM micrographs of the T-section (Figure 5e,f) reveal a microstructure composed of primary M7C<sup>3</sup> carbides (labeled as 1) surrounded by the eutectic (γ + M7C3) structure (labeled as 2). The carbides reveal a rhombohedral/hexagonal cross-section [48]. Nb additions enabled the precipitation of Nb-rich MC carbides displayed as polygonal-shaped structures (labeled as 3), as confirmed by the EDS analyses. Nb- and Mo-rich carbides (labeled as 4) and Mo- and Fe-rich carbides (labeled as 5) have also been detected. From the high magnification BSE-SEM micrograph of Figure 5f, traces of martensite (labeled as 6) could also be observed at the periphery of the carbides network [7,49,50].

Considering the influence of the CVF on the microstructural and wear characteristics of the hardfacing [29,31,51], quantitative metallographic image analyses were carried out. Figure 6 displays examples of the representative microstructures analyzed to evaluate the CVF parameter: Figure 6a,b show the L-section in the RS and SS, respectively, while Figure 6c,d show the T-section in the RS and SS, respectively.

As seen, the RS presents larger carbides with a plate-like morphology, while the SS presents smaller primary carbides with a polygonal shape. It is worth noting that, even within the same distance from the surface, the microstructure was quite heterogeneous, as the result of the solidification process. The estimated mean CVF values were 26.25% and 23.45% for the L-section in the RS and SS, respectively, while 25.98% and 24.40% for the T-section in the RS and SS, respectively. Beyond the negligible differences in the number of carbides between L-section and T-section, attempts were made to evaluate the size of the investigated carbides. Hence, Figure 5e displays the Cumulative Distribution Function (CDF) of the carbides' area, A. These distributions reveal that for the L-section (both for the RS, labeled as L\_RS, and for the SS, labeled as L\_SS) about 90% of the carbides had an area in the range of 4–60 µm<sup>2</sup> while for the T-section (both for RS, labeled as T\_RS and for the SS, labeled as T\_SS) about 90% of the carbides had an area in the range of 2–40 µm<sup>2</sup> . In all cases, carbides with a larger area (from 100 µm<sup>2</sup> up to 900 µm<sup>2</sup> ) were much less with respect to the numerous eutectic carbides, the share of these large carbides amounted to about 10% of the total.

γ **Figure 5.** (**a**) Optical microscope (OM) micrograph of the T-section of the hardfacing captured from the resistant side, red arrows highlight the crack propagation path; (**b**) BSE-SEM image of the Tsection at the boundary between two welding passes: the dash-dot yellow line separates the passes with coarser and finer rod-like carbides; (**c**,**d**) OM micrographs of the L-section in the substrate and resistant side, respectively; (**e**,**f**) BSE-SEM images of the T-section whit primary M7C<sup>3</sup> carbides (labeled as 1), eutectic (γ+ M7C<sup>3</sup> ) structure (labeled as 2), Nb-rich MC carbides (labeled as 3), Nband Mo-rich carbides (labeled as 4), Mo- and Fe-rich carbides (labeled as 5) and traces of martensite (labeled as 6).

**Figure 6.** Representative OM micrographs used for the carbide volume fraction (% CVF) evaluation: L-section in the RS (**a**) and SS (**b**); T-section in the resistant side (RS) (**c**) and substrate side (SS) (**d**). (**e**) The cumulative distribution function (CDF) of the carbides' area A at different locations.

#### *3.2. Hardness of the As-Received HCCI*

The results of hardness measurements, taken on the T cross-section, are reported in Figure 7. Figure 7a shows the Vickers hardness profile measured at increasing distances of 1 mm from the top of the welded hardfacing up to the bottom of the steel substrate. Moreover, the hardness of both the matrix and the primary carbides was evaluated (Figure 7b).

**Figure 7.** Vickers hardness measurements on the T cross-section: (**a**) hardness profile across the hardfacing plate from the resistant side up to the steel substrate; (**b**) microhardness values across the hardfacing for matrix and Cr-based carbides.

#### *3.3. Effect of the Heat Treatment on Microstructure and Erosion Behavior*

The erosive resistance of the investigated HCCI was evaluated by erosion tests conducted by comparing the behavior of the alloy in the as-received and heat-treated conditions. Consecutive tests at 30 min, 60 min, and 90 min of erosion were conducted.

Figure 8 represents the BSE-SEM microstructures of the as-received and heat-treated samples, together with the respective XRD spectra. In the as-received state, the high magnification BSE-SEM micrograph (Figure 8a) displays a microstructure mainly consisted of M7C<sup>3</sup> eutectic carbides, traces of martensite, and MC carbides, in accordance with the identified phases in the respective XRD spectrum (Figure 8c). In the heat-treated condition, the austenite is transformed into martensite, as revealed by the BSE-SEM micrograph (Figure 8b) and by the martensite peak in the XRD spectrum (Figure 8c).

**Figure 8.** Comparison between as-received and heat-treated conditions. (**a**,**b**) BSE-SEM images of the hardfacing microstructure: (**a**) as-received and (**b**) heat-treated at 950 ◦C for 3 h and oil quenched. (**c**) XRD spectra of the as-received and heat-treated conditions.

After destabilization, the secondary carbides precipitated during the heat treatment appear as fine granular particles distributed in the matrix (Figure 8b) [52]. These carbides are detectable from the BSE-SEM images resulting from the deep-etched samples. Figure 9 reports the comparison between the microstructure of the as-received and the heat-treated samples after etching for 24 h in a solution of 10% HCl in methanol, as suggested by [7]. From the BSE-SEM image of Figure 9b, the small and uniformly distributed secondary carbides are detectable, in the framework of the matrix structure.

Accordingly, the overall bulk hardness is increased by the thermal treatment: the as-received sample shows 793 ± 35 HV30 while the heat-treated sample results in 950 ± 52 HV30. Figure 10 exhibits the appearance of a representative worn top surface of an erosion crater on the as-received sample. From the digital image (Figure 10a) it can be seen the modification of the substrate surface due to the particle impact. The erosion pattern is characterized by a circular spot with two concentric regions (commonly known as a crater). The internal matt-grey region (with a diameter equal to about 5 mm) overlooks the nozzle: in this region, particles impact the substrate for the first time at 90◦ with the nominal velocity determining the greater erosion damage. The external circular light-grey region (with a diameter equal to about 9 mm) is generated by the secondary impacts characterized by lower impact angle (almost tangential impact) and lower impact velocity. Hence, the

analyzed eroded area of the worn top surface of the sample, as depicted in the BSE-SEM image of Figure 10b, was a circle of about 3 mm diameter in the central region of the erosion crater.

**Figure 9.** Deep-etched BSE-SEM images of (**a**) as-received and (**b**) heat-treated samples.

**Figure 10.** Images of the top surface after erosion: (**a**) digital image; (**b**) BSE-SEM image.

The worn surface was analyzed through SEM observation of the top surface after the erosion process. As depicted in Figure 11, which displays the worn surface of an erosion crater on the as-received sample after 30 min of erosion, SEM images revealed both the topography of the worn surface (Figure 11a) and the distribution of the chemical elements, detected by the X-ray elemental maps (Figure 11b,c). Due to the erosion, the surface appears rough, with the Cr-rich carbides protrusions resulting from the selective wear of the matrix [31].

Previous research has established the tricky phenomena associated with the wear behavior of HCCIs, deeply investigating the role of matrix, primary and eutectic carbides on the wear resistance [31]. The cross-section of the erosion crater, in the as-received condition, was thus investigated by SEM: Figure 12 depicts the comparison between 30 min and 60 min of erosion of two different regions of the same sample. As can be seen, the cross-section analysis in the center of the erosion crater does not provide information about the erosion mechanisms that occurred. Time-wise evolution of the erosion and wear phenomena are not detectable only by the cross-section evaluation. After 30 min of erosion, Figure 12a, primary M7C<sup>3</sup> carbides appear emerging from the surrounding matrix, which results in some damage by the erodent particles. Conversely, after 60 min of erosion, the center of the erosion crater appears uniformly eroded, with no evident carbide protrusions (Figure 12b). Such behaviors rely on the mutual interdependence between microstructural features, of both matrix and carbides (i.e., size, volume fraction, distribution of primary carbide and eutectic carbides), and fracture behavior (i.e., fracture toughness and hardness). It has been reported [31] that hard primary carbides near the surface could be spalled off as they are cracked during the wear process, despite the high hardness. At the same time, the microstructure, mainly composed of eutectic carbides that show a lower hardness because

of the decrease of the volume fraction of the M7C3, reveals a better wear resistance thanks to the uniform distribution of such carbides. Such harder and uniformly distributed carbides ensure an increased wear resistance since abrasives cannot effectively penetrate into the matrix and carbides are not easily separate from it [53,54]. According to the literature, relatively homogeneous wear is observed (Figure 12b).

**Figure 11.** SEM images of the worn surface, center of the crater after 30 min of erosion on an asreceived sample: (**a**) SEI-SEM image, (**b**) layered image of X-ray maps, and (**c**) maps of the elemental distribution.

**Figure 12.** BSE-SEM micrographs of the cross-sectioned erosion crater: (**a**) 30 min of erosion, (**b**) 60 min of erosion.

In the light of the above reported microstructural findings, to exclude the influence of the microstructural changes associated with the chemical and solidification conditions, erosion tests were performed on the same area of the sample. The erosion results are reported in Figure 13, as the weight loss against the exposure time: it follows that the wear behavior of the heat-treated samples toward erosion is better than the as-received ones. Regardless of the condition, the weight loss by the sample increases with the erosion time with a slightly lower slope for the heat-treated sample. It is worth noting the weight loss decrease for the heat-treated condition, promoted by the destabilization treatment and, in turn, by the superior erosion resistance of the obtained microstructure. Conversely, the weight loss of the as-received sample increases linearly with erosion time. Note that the

uncertainty band associated with the weight loss measurements has the same size as the markers on the charts (extended uncertainty is ±0.06 mg).

**Figure 13.** Erosion tests result for the as-received and heat-treated hardfacing samples: weight loss against erosion time.

The microstructural investigations were performed on the top worn surfaces by both OM to evaluate the CVF, and by SEM to study the topographical changes associated with the erosion phenomena. Since the morphologies of carbides, together with the microstructure of the matrix, are the predominant characteristics influencing the hardness and wear behavior of the alloy, it is of high relevance to consider their quantitative analysis. Table 2 summarizes the CVF values and the respective standard deviations in parentheses for the as-received and the heat-treated conditions.

**Table 2.** CVF values and respective standard deviations (in parenthesis) in the as-received and heat-treated conditions with respect to erosion time. The initial condition before the erosion test is 0 min.


As for the initial condition, indicated as 0 min, it can be stated that for the as-received and the heat-treated samples the CVF values are comparable. Conversely, from the comparison between the CVF values in the initial condition and after 30 min, 60 min, and 90 min of erosion it appears that, aside from the condition, the CVF is reduced to half its initial value and it seems not significantly affected by the erosion time. These experimental findings shed light on the time-wise evolution of weight loss (see Figure 13).

To further investigate the efficacy of the heat treatment, SEM analyses of the worn top surfaces were conducted. Figure 14 presents the topography in the center of the crater after erosion through both SEI-SEM and BSE-SEM images.

From the comparison between the worn surfaces of the as-received (Figure 14a,b) and the heat-treated (Figure 14c,d) samples after 30 min of erosion, it can be detected the different wear behavior. The surface of the as-received sample shows wide carbide protrusion resulting from the selective wear of the matrix while the heat-treated sample reveals the simultaneous wear of matrix and carbides. Indeed, the reduced hardness difference between matrix and carbides in the heat-treated sample, 850 ± 52 HV0.2, and 1620 ± 122 HV0.05 respectively, compared with the as-received sample, 692 ± 58 HV0.2 and 1477 ± 147 HV0.05 respectively, increases the wear resistance of the alloy.

**Figure 14.** SEI-BSE images of the center of the crater: (**a**,**b**) SEI-BSE images of the as-received hardfacing after 30 min of erosion; (**c**,**d**) SEI-BSE images of the heat-treated hardfacing after 30 min of erosion; (**e**,**f**) SEI-BSE images of the heat-treated hardfacing after 60 min of erosion.

The improved wear resistance of the heat-treated sample is further confirmed by the surface topography appearance after 60 min of erosion (Figure 14e,f), which is somewhat similar to the condition of the as-received sample after 30 min of erosion (Figure 14a,b). Note that the weight loss of these two samples is nearly the same. In the light of that, it can be inferred the increased capability of the higher hardness matrix in providing better support for the carbides, compared with the austenitic one, in agreement with findings in [52].

These findings suggest that the proposed heat treatment is effective in improving the erosion resistance of the investigated HCCI alloy, which is strongly affected by the matrix/carbides hardness difference.

#### **4. Conclusions**

The microstructural characteristics and the erosive wear behavior of a hypereutectic HCCI hardfacing alloy were experimentally investigated by considering the effect of a destabilization heat treatment. The experimental findings indicate that:

• according to microstructural and X-ray diffraction analyses, the investigated HCCI consists of a mixture of MC and M7C<sup>3</sup> carbides dispersed in a metastable austenite

matrix containing a high Cr concentration, with traces of martensite at the carbides' periphery. Nb-rich and Mo-rich carbides were also detected;


With the objective of providing a better understanding of the microstructure-property relationships concerning the erosive wear resistance of a Fe–Cr–C hardfacing alloy, the experimental findings of this study give some remarkable hints, useful for optimized exploitation of HCCIs in heavy-duty applications. The key takeaways of the research improve the knowledge of the erosive behavior of hardfacing alloy due to the impact of micro-sized particles and would thus reflect towards the increase of their performance and, in turn, of their application ranges.

**Author Contributions:** Conceptualization, A.F. and A.S.; Data curation, A.V.; Investigation, A.V.; Methodology, A.F. and A.S.; Supervision, M.M. and M.P.; Writing—original draft, A.F.; Writing review & editing, A.S. and M.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** The authors owe thanks to Eng. Paolo Saccenti of Boldrocchi S.r.l. (Biassono, Monza-Brianza, Italy) for the technical support in this research. The authors wish to gratefully acknowledge Luca Marchetti and Matteo Seno for their support in the experimental campaign. Thanks are also due to Iuri Boromei for his contribution to the GD-OES analyses.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Coatings* Editorial Office E-mail: coatings@mdpi.com www.mdpi.com/journal/coatings

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com ISBN 978-3-0365-1803-9