Effect of Heat Treatment on the Corrosion Behavior of Weld-Deposited Chromium Carbide-Based Hardfacing Alloys

Abstract

The effects of heat treatment on the microstructure and corrosion behavior of chromium carbide-based hardfacing alloys deposited via gas metal arc welding were investigated. The hardfacing alloy, high chromium white iron (HCWI), containing 27 wt% Cr and 4.8 wt% C, was heat treated at 650 °C and 950 °C for six hours followed by natural cooling to room temperature. Microstructural characterization revealed that heat treatment promoted the transformation of austenite to ferrite and increased carbide precipitation. X-ray diffraction analysis identified the primary carbides as Cr₇C₃, which remained stable during heat treatment. Electrochemical corrosion testing in artificial seawater, including potentiodynamic polarization and electrochemical impedance spectroscopy (EIS), demonstrated progressively improved corrosion resistance with heat treatment temperature. Both techniques confirmed that the specimen treated at 950 °C exhibited superior corrosion resistance compared to the 650 °C treatment and as-deposited condition, with the specimen treated at 950 °C exhibiting the highest charge transfer resistance (4711 Ω·cm²) compared to the 650 °C treatment (2608 Ω·cm²) and as-deposited condition (374.6 Ω·cm²). The enhanced corrosion resistance was attributed to the increased carbide precipitation and optimization of the matrix composition. While heat treatment at both temperatures improved corrosion performance, the 950 °C treatment yielded superior results, suggesting this could be an optimal temperature for enhancing the corrosion resistance of chromium carbide-based hardfacing alloys.

1. Introduction

Hardfacing is often used as a method to improve surface properties of a substrate metal. One common method of hardfacing is by welding deposition onto a substrate. This process improves various surface properties such as of hardness, wear, and corrosion resistance. Australian standards such as AS/NZS 2576:2005 can be used to select appropriate hardfacing materials based on applications. [1].

One type of hardfacing material is based on high chromium white iron (HCWI) weld deposit. This material contains iron with high amounts of chromium content (nominally above >12% and carbon content above 3 wt%) [2,3]. This allows it to form polygonal chromium carbides in the material, surrounded by a eutectic matrix. High chromium alloys are often studied and used in industries such as mining and mineral processing due to their high wear resistance [4]. However, it can be noted that due to its high chromium content, these alloys can also form protective Cr₂O₃ films that can passivate against further corrosion, with increased corrosion resistance seen in materials containing chromium contents above 20% [5].

Heat treatment can be utilized to modify the properties of hardfacing layers. This may be used to relieve not only stress but also increase the toughness of materials and modify the microstructure of alloys [6,7]. Control of the form and type of carbides present within the material can optimize the performance of materials. For instance, it is speculated that cast heat treatment-induced spheroidization may lead to carbides being more globular in shape, enhancing the toughness of the material [3,8].

The effects of heat treatment on microstructure and hardness are well-studied for HCWI. A review for HCWI describes the as-deposited state as typically consisting of austenitic dendrites, with austenite and M₇C₃ carbides, with the carbide volume fraction determined by both chromium and carbon content [3]. Heat treatment has been found to decrease the stability of austenite due to precipitation of carbides, reducing the carbon and chromium content in the austenite. This leads to the transformation of austenite into martensite; however, austenite can still be retained, with an increased austenite proportion found at higher destabilization temperatures [9]. Chromium content is also expected to affect the form of carbides precipitated, with one study finding 20%Cr only forming M₇C₃ carbides, whilst 27%Cr led to the formation of both M₂₃C₆ and M₇C₃ carbides [10]. Precipitated carbides improved matrix strength, as well as provided support for other carbides [6,11].

While heat treatment is generally agreed to improve properties compared to its as-cast state, some debate remains concerning the temperatures and processes required for the optimization of its properties [12,13]. For instance, one study found that heat treatment over increased holding times (6 h) led to improved wear resistance; however, hardness at the same time also decreased [14]. This effect was attributed by the authors to have been caused by the simultaneous modification of matrix toughness, whilst hardness decreased from carbide coarsening. Lower destabilization temperatures at subcritical levels such as at 600 °C have been associated with greater carbide precipitation and lower retained austenite, leading toward reduced hardness compared to destabilization temperatures such as 1000 °C [3,15]. The precipitation of secondary carbides has also been found to be affected by heating rate and temperatures [16]. In general, however, it is thought that control of carbide size is critical for controlling wear [17,18].

Limited studies regarding the corrosion behavior of heat treatment for cast HCWI alloys appear to have been conducted. For instance, Sarac et al. investigated the effects of heat treatment on cast HCWI alloys in alkaline and acidic conditions [19]. Their results for heat treatment at 900 °C found refinement of primary carbides and precipitation of excessive carbides, with further tempering at 240 °C resulting in more homogenous carbide dispersion. These heat treatment effects were attributed to improving corrosion resistance compared to the as-cast sample. Generally, it has been found that heat treatment reduces the corrosion of as-cast samples; however, it has also been noted that heat treatment at subcritical temperatures produced better corrosion resistance [10,20].

In contrast to as-cast alloys, research into the corrosion effects of heat-treated chromium carbide-based weld hardfaced deposits is still to be investigated, at least to the best of the author’s knowledge. In this study, the effects of heat treatment of chromium carbide-based weld hardfaced deposits were investigated [21].

2. Experimental Procedure

2.1. Material Preparation

Commercially available Supershield CrC (Hyundai, sourced from Sydney, Australia) was deposited onto a base mild steel substrate by gas metal arc welding. As per its datasheet, this alloy can be described as iron-containing chromium carbides, as designed for use cases against severe abrasion/erosion and moderate impacts [22]. This alloy is also designed to be self-shielding, not requiring a shielding gas under regular conditions for ease of portable deposition. The expected composition of this welding alloy is shown in

Table 1. Expected composition of Supershield CrC.

Cr (wt%)	C (wt%)	Mn (wt%)	Si (wt%)	Fe (wt%)
27	4.8	1.6	0.4	Balanced

below.

This weld metal was manually deposited using a Cigweld MIG welding machine (sourced from Sydney, Australia). A mean voltage of 32 V and current of 350 A was also used, with a wire feed speed of 11.5 m/min. A minimum of four welding layers was also deposited to reduce the effects of dilution, as per AS/NZS 2576:2005 [1], with a total deposited height of 1.5 cm.

The topmost layer of three different samples was cut off to expose a flat 2 × 2 cm surface. The heat treatment of samples was undertaken toward target temperatures as listed in

Table 2. Targeted heat treatment temperatures.

Name	Heat Treatment Temperature
Sample H0	(No heat treatment)
Sample H650	650 °C
Sample H950	950 °C

. This was conducted in a furnace from room temperature up to the target temperature in the air, which was then held for six hours once the target temperature was reached. Once completed, samples were then left in the furnace to cool back to room temperature. Samples are nominally labeled as H0 to describe the as-deposited state H650 for the sample undergoing heat treatment at 650 °C and H950 for the sample with heat treatment at 950 °C. Earlier studies have found that heat treatment at 650 °C would produce the precipitation of carbides while heat treatment at 950 °C resulted in the dissolution of the carbides [21].

Before each surface characterization or electrochemical test, samples were polished to a 1 µm finish. These samples were then etched using 5% nital.

Artificial seawater was also prepared for analysis as the main electrolytic test medium for exposing prepared metal samples. This was created as per ASTM D1141-98, with compositional ingredients as shown in

Table 3. Artificial seawater composition.

Compound	Concentration (g/L)
NaCl	24.53
MgCl2	5.20
Na2SO4	4.09
CaCl2	1.16
KCl	0.695
NaHCO3	0.201
KBr	0.101
H3BO3	0.027
SrCl2	0.025
NaF	0.03

below [23].

2.2. Surface Characterisation and Compositional Analysis

Deposited and heat-treated samples were characterized using scanning electron microscopy (SEM). A PhenomXL (Thermofisher, Adelaide, Australia) scanning electron microscope was used to achieve these goals, taking images of the microstructural surface both before and after electrochemical testing. Pre-corrosion images were prepared via polishing under up to 1 µm diamond paste, as further detailed under the material preparation in Section 2.1. Further analysis of SEM images was conducted using ImageJ software (version 1.54). The PhenomXL machine was also used to simultaneously perform Energy Dispersive X-ray Spectrocopy (EDS) measurements to further characterize the chemical composition of microstructural features.

X-ray Diffraction (XRD) was performed using a PANanalytical Empyrean (Malvern Panalytical, Sydney, Australia) to gauge the composition of the non-heat treated sample. This utilized a copper target under a range of 2θ = 20–90°, with a step size of 0.0394° and a time of 300 s per step.

Microhardness values were also taken for each sample using a DuraScan 70 machine (Struers, sourced from Milton, Australia). This utilized a load of 0.05 kg for indentation on the carbides, as well as a eutectic carbide/matrix mix. Microhardness had to be conducted on a eutectic carbide/matrix mix due to its small size preventing hardness of the matrix and eutectic carbides from being measured independently.

2.3. Electrochemical Testing

Electrochemical testing was conducted on the metal samples using a Gamry 3000 potentiostat (sourced from Menai, Australia). A three-electrode setup was used, with the metal samples acting as the working electrode, artificial seawater as the electrolyte, Ag/AgCl reference electrode, and a platinum-coated counter electrode.

Potentiodynamic tests were performed on the samples. Before each potentiodynamic test, samples were polished with up to 1 µm diamond paste as described via the material preparation in Section 2.1. These samples were then left exposed to the artificial seawater medium for one hour to allow the system to normalize. Open Circuit Potential (OCP) was then run over ten minutes before beginning potentiodynamic testing. The range of potential values used was −0.2 V below OCP to 1.3 V, with a scan rate of 0.167 mV/s. Similarly, electrochemical impedance spectroscopy (EIS) was also run over the range of 0.1–10,000 Hz after polishing the samples and leaving them exposed to artificial seawater after an hour. An amplitude of ±10 mV was used.

3. Results

3.1. Microstructural Characterisation and Hardness

SEM images of Sample H0, H650, and H950 are shown in Figure 1, Figure 2 and Figure 3. These images display a mix of larger carbides and smaller eutectic carbides and austenite microstructure, as annotated in Figure 1b. It can be seen in the higher magnification images that heat treatment resulted in a greater proportion of smaller carbides (or secondary carbides) being precipitated than in the as-deposited sample. Sample H950 also exhibited tiny carbide flecks as labeled in Figure 3b. Limited spheroidization may have occurred in the H950 sample, as carbides seem to have slightly rounded sides compared to the samples in Figure 2a,b.

Quantitative analysis using ImageJ software was conducted using three equal-sized images enumerating the number of carbides (

Table 4. Table of Enumerated Carbides (primary, seconday, and eutectic).

Alloy	Average Carbide Size (µm2)	Max Carbide Found (µm2)	Total No. of Carbides
H0	2.56	207	674
H650	1.49	167	1012
H950	1.06	160	1444

). This confirmed a significant increase in the total number of carbides due to precipitation, increasing from 674 in Sample H0 to 1012 and 1444 in Sample H650 and H950, respectively. Some refinement of the carbides also occurred, with the maximum carbide size decreasing from 207 µm² in Sample H0 to 167 µm² and 160 µm² in Sample H650 and H950, respectively.

The hardness of the samples was measured using a 0.05 kgf load for the carbides and a eutectic carbide/mix (recalling as per the experimental section that the eutectic carbides/matrix could not be measured independently due to small size); the average of three separate points was found. These results are provided in

Table 5. Hardness values for carbides and the eutectic carbide/matrix mix.

Area	Alloy	Hardness (HV0.05)
		Point 1	Point 2	Point 3	Average
Carbide	H0	1622	1415	1567	1536
	H650	1258	1389	1400	1349
	H950	2226	1654	1944	1941
Eutectic Carbide/Matrix	H0	1009	1058	1023	1030
	H650	687	672	761	707
	H950	869	963	1016	949

and Figure 4. Notably, the hardness was greatest in the eutectic carbide/matrix mix for the as-deposited sample (1030 HV_0.05) than in the heat-treated samples (707 HV_0.05 and 949 HV_0.05 for Sample H650 and H950, respectively). Meanwhile, heat treatment at 650 °C resulted in a slight decrease in primary carbide hardness compared to the as-deposited sample (1349 HV_0.05 and 1536 HV_0.05, respectively), whilst Sample H950 displayed the highest carbide hardness at 1536 HV_0.05).

3.2. EDS and XRD Results

EDS results are shown in

Table 6. EDS results of austenite/ferrite microstructure.

Alloy	Carbon (wt%)	Oxygen (wt%)	Silicon (wt%)	Chromium (wt%)	Iron (wt%)
H0	7.6	1.1	0.9	14.3	76.1
H650	6.89	0.1	0.9	12.9	79.2
H950	5.31	0.1	0.6	14.2	79.8

,

Table 7. EDS results of primary carbide microstructure.

Alloy	Carbon (wt%)	Oxygen (wt%)	Chromium (wt%)	Manganese (wt%)	Iron (wt%)
H0	9.2	1.2	45.4	2.4	41.8
H650	10.5	1.10	43.6	2.20	42.5
H950	8.19	1.40	43.0	1.90	45.6

and

Table 8. Eutectic carbide microstructure composition via EDS.

Alloy	Carbon (wt%)	Oxygen (wt%)	Chromium (wt%)	Manganese (wt%)	Iron (wt%)
H0	9.92	0.9	34	2.1	53.1
H650	9.41	0.9	35.2	2.3	52.2
H950	11.5	1.3	37	2.6	47.6

for the carbides, eutectic carbides, and austenite/ferrite in Samples H0, H650, and H950.

As per Table 7, the chromium content in the primary carbides appeared to marginally decrease with increasing heat treatment (45.4 wt% to 43.6 wt% to 43.0 wt% for H0, H650, and H950, respectively). The chromium content in the austenite/ferrite decreases slightly from 14.3 wt% in Sample H0 to 14.2 wt% in H950 and 12.9 wt% in H650. The loss of chromium and carbon from the primary carbides and eutectic austenite resulted in the precipitation of secondary carbides with an increase in chromium content from 34 wt% in Sample H0 to 35.2 wt% and 37 wt% in Sample H650 and H950, respectively. The total number of carbides has increased from 674 to 1012 to 1444 for H0, H650, and H950, respectively, as shown in Table 4.

The results from XRD testing are shown in Figure 5. Notably, the non-heat-treated Sample H0 displayed primary carbides of the form Cr₇C₃ and austenite. Heat treatment does not appear to have transformed Cr₇C₃ to any other form of carbides. However, the H650 and H950 samples appear to display the transformation of austenite into ferrite/martensite. This behavior is in line with other studies for HCWI alloys, which describe transformation to ferrite [24,25] and martensite [26] upon cooling due to chromium depletion from the matrix.

3.3. Potentiodynamic and Potentiostatic Results

Potentiodynamic tests were undertaken for Samples H0, H650, and H950 using a three-electrode setup to investigate the corrosion behavior. The results from this testing are as shown in Figure 6, with corrosion parameters E_corr and I_corr as shown in

Table 9. Corrosion parameters from the potentiodynamic plot.

Alloy	Ecorr (mV/cm2)	Icorr (µA/cm2)
Sample H0	−545	11.5
Sample H650	−528	3.32
Sample H950	−492	3.24

as found from Tafel plot analysis in Gamry Echem Analyst software (ver 7.9.0). From this, it appears that H650 had only a slightly better corrosion potential E_cor than Sample H0 (−545 mV and −528 mV for H0 and H650, respectively), whilst heat treatment at 950 °C displayed the best corrosion resistance with an E_cor of −492 mV.

Notably, after open circuit potential has been passed, all samples appear to display two different major points of inflection on the curve; one around −0.2 V and another at approximately past 1.0 V. Using this as a guide, potentiostatic points were chosen for etching at different potentials for 15 min to further observe corrosion behavior within these sections. These points are labeled in Figure 7 for Sample H0, with a table of these values as shown in

Table 10. Potentiostatic points as used for Samples H0, H650, and H950.

Etched Point	Etched Voltage (mV)
Point A	−330
Point B	800
Point C	1100

.

Samples were etched at these potentials using the electrolyte shown in Table 3, and SEM pictures after etching are shown in Figure 8, Figure 9 and Figure 10. The EDS results of carbides and the matrix are shown in

Table 11. Matrix EDS composition at varying potentiostatic points.

Alloy	Pot. Point (mV)	Element (wt%)
		Carbon	Oxygen	Silicon	Chromium	Iron	Chlorine	Sodium
H0	−330	2.3	0.9	1	12.3	82.7	0.8	-
	800	4.2	18.5	1.3	16.8	49.5	5	4.6
H650	−330	3.7	0.4	0.5	12.3	82.8	0.3	-
	800	3.7	12.6	1	17.9	53.1	7.7	4
H950	−330	4.8	0.4	0.6	11.5	87.1	0.4	-
	800	4.4	15.9	2.3	18.3	49.5	7.5	2.108

and

Table 12. Carbide EDS composition at varying potentiostatic points.

Alloy	Pot. Point (mV)	Element (wt%)
		Carbon	Oxygen	Chromium	Iron	Sodium
H0	−330	5.3	0.1	48.1	45.8	-
	800	4.8	2.4	47.6	44	1.2
	1100	4.5	1.3	45.3	45.6	3.3
H650	−330	4.6	0.8	45.4	48.5	-
	800	4.3	4.7	45.8	44.2	1.0
	1100	3.7	1.2	48.1	44.5	2.5
H950	−330	6.1	0.9	46.9	46.1	-
	800	4.1	3.2	45.1	46.7	0.9
	1100	4.2	1.0	48.0	45.8	1.0

. Carbides did not appear to be particularly affected by etching at different potentials, with all chromium content lying between a 45 and 48 wt% range. However, increases of at least 4 wt% can be observed for matrix (ferrite/austenite) chromium when increasing the etching potential from −330 mV to 800 mV for all samples (e.g., 12.3 wt% -> 16.8 wt% for Sample H0, 12.3 wt% -> 17.9 wt% for Sample H650 and 11.5 wt% -> 18.3 wt% for Sample H950). Meanwhile, the iron content in the matrix decreased by ~30 wt% for all samples when increasing the potential from −330 mV to 800 mV (e.g., 82.7 wt% -> 49.5 wt% for Sample H0, 82.8 wt% -> 53.1 wt% for Sample H650 and 87.1 wt% -> 49.5 wt% for Sample H950). This may suggest that whilst the iron matrix preferentially corroded, chromium remained mostly untouched. Furthermore, the oxygen content in the matrix increased between the potentiostatic points of −300 and 800 mV (e.g., 0.9 wt% -> 18.5 wt% for Sample H0), suggesting that the loss of iron was due to the formation of iron oxides.

3.4. Electrochemical Impedance Spectroscopy Results

Electrochemical impedance spectroscopy (EIS) was carried out using a three-electrode setup in artificial seawater. Nyquist and Bode plots are shown in Figure 11, Figure 12 and Figure 13.

Equivalent circuit models can be fit over this data. For coatings, these are typically expected to be in the form R_soln[C_c[R_po[C_corR_cor]] [27]. In this model, R_soln represents solution resistance, C_c represents the capacitance of the overall coating, R_po represents the resistance of coating pores, C_cor represents the double-layer capacitance, and R_cor represents charge transfer resistance. It was also found that these models became more accurate when modeling the capacitors as complex phase elements (Q), likely due to non-homogenous elements in the system, and as such may be better described by the model R_soln[Q_c[R_po[Q_corR_cor]], as seen in Figure 14. Parameters as obtained via this model are as shown in

Table 13. Table of parameters found for EIS analysis.

Alloy	Rsoln (Ω.cm2)	Qc (Ω−1cm−2sn)	nc	Rpo (Ω.cm2)	Qcor (Ω−1cm−2sn)	ncor	Rcor (Ω.cm2)	Chi2
H0	34.11	0.00358	0.7153	1381	0.00375	0.8741	374.6	8.10 × 10−5
H650	25.23	0.00021	0.7951	73.82	0.00024	0.7642	2608	0.00105
H950	25.1	5.30 × 10−5	0.9283	38.75	0.00061	0.6252	4711	0.00118

.

From Table 13, it was found that all models had a chi² value of ~1 × 10⁻³ or less, indicating a good value of fit.

The EIS analysis reveals significant differences in the corrosion behavior between the as-deposited (H0) and heat-treated (H650 and H950) samples of the chromium carbide-based hardfacing alloy. The Nyquist plots (Figure 11) and corresponding EIS parameters (Table 12) provide valuable insights into the electrochemical characteristics and corrosion resistance of these samples.

The most striking observation is the substantial increase in charge transfer resistance (Rcor) following heat treatment. The H0 sample exhibits the lowest Rcor value of 374.6 Ω.cm², while H650 and H950 show dramatically higher values of 2608 Ω.cm² and 4711 Ω.cm², respectively. This progressive increase in Rcor indicates enhanced corrosion resistance with heat treatment, particularly at the higher temperature of 950 °C. The larger Rcor values suggest that the heat-treated samples form more stable passive films, which act as better barriers against charge transfer processes at the metal–solution interface.

The solution resistance (Rsoln) shows a slight decrease from 34.11 Ω.cm² for H0 to approximately 25 Ω.cm² for both heat-treated samples. This minor variation in Rsoln suggests that the electrolyte conditions remained relatively consistent during testing, allowing for reliable comparison of other parameters.

A particularly interesting trend is observed in the porous resistance (Rpo) values. The as-deposited H0 sample shows a relatively high Rpo of 1381 Ω.cm², which drastically decreases to 73.82 Ω.cm² for H650 and further drops to 38.75 Ω.cm² for H950. This trend correlates well with the microstructural changes observed, particularly the increase in carbide count and decrease in carbide size with heat treatment. The higher number of smaller carbides likely creates more interfaces and potential pathways through the passive film, resulting in lower Rpo values.

The constant phase element parameters (Qc and nc) for the coating layer show a consistent trend toward more ideal capacitive behavior with heat treatment. The nc value increases from 0.7153 (H0) to 0.9283 (H950), approaching unity, which indicates the formation of a more uniform and stable passive film. Simultaneously, the Qc values decrease by almost two orders of magnitude from H0 (0.00358) to H950 (5.30 × 10⁻⁵), suggesting the formation of a thicker or more compact passive layer.

These EIS results align well with the potentiodynamic findings, where both heat-treated samples showed significantly lower corrosion current densities (Icorr) compared to the as-deposited condition. The improvement in corrosion resistance can be attributed to several factors revealed in the microstructural analysis. (a) The redistribution of chromium during heat treatment, particularly in the matrix and eutectic regions, likely contributes to the formation of a more effective passive film. (b) The increased number of finer carbides (1444 carbides for H950 vs. 674 for H0) creates a more uniform distribution of chromium-rich phases, potentially leading to more consistent passive film formation. (c) The changes in matrix composition, particularly the optimization of chromium content (14.2 wt% in H950), appear to provide an ideal balance for corrosion resistance.

Analysis of the Bode plot (Figure 12) provides complementary insights into the frequency-dependent electrochemical behavior of the hardfacing alloy samples, revealing distinct phase angle responses across the frequency spectrum (0.1 Hz to 10 kHz).

The phase angle profiles demonstrate characteristic differences between the as-deposited and heat-treated conditions. At lower frequencies (0.1–1 Hz), H650 exhibits the most negative phase angle maximum (approximately −55°), followed by H950 (−50°), while H0 shows a notably less negative maximum (around −48°). This behavior in the low-frequency region corresponds to the diffusion-controlled processes and indicates that H650 achieves the most effective capacitive behavior in this range, suggesting a more stable passive film formation.

In the intermediate frequency range (1–100 Hz), there is a clear shift in the phase angle maxima positions. The H0 sample shows its maximum phase angle at a higher frequency compared to both heat-treated samples, indicating faster charge transfer kinetics but potentially less stable passive film behavior. This observation aligns with the previously noted lower charge transfer resistance (Rcor) for the H0 sample.

The high-frequency region (>1000 Hz) shows convergence of phase angles toward zero for all samples, though with subtle differences in their decay patterns. The H950 sample maintains slightly more negative phase angles in this region, suggesting better retention of capacitive behavior even at higher frequencies. This characteristic correlates well with its highest nc value (0.9283) among the three conditions, indicating the near-ideal capacitive behavior of its passive film. A notable feature is the broader phase angle peak for H950 compared to both H0 and H650, suggesting a wider range of effective capacitive behavior. This broadening of the frequency response indicates a more complex and potentially more effective barrier against corrosion, consistent with the microstructural refinement observed in the H950 sample. The gradual transition in phase angles across the frequency spectrum for heat-treated samples, particularly H950, indicates a more homogeneous electrochemical response. This homogeneity can be attributed to the refined carbide distribution and optimized matrix composition achieved through heat treatment, resulting in more uniform corrosion protection.

The Bode magnitude plot (Figure 13) provides additional insights into the impedance behavior of the hardfacing alloy samples across the frequency spectrum. The impedance magnitude (|Z|) versus frequency representation reveals several distinctive characteristics that complement our previous electrochemical analyses.

At the lowest frequencies (0.1 Hz), both heat-treated samples (H650 and H950) exhibit slightly higher impedance magnitudes (approximately 2000 Ω·cm²) compared to the as-deposited H0 sample (approximately 1500 Ω·cm²). This higher low-frequency impedance correlates with the enhanced barrier properties of the heat-treated surfaces, particularly their improved passive film characteristics. A notable feature is the slope of the impedance curves in the mid-frequency range (1–100 Hz). All three samples show a characteristic linear decrease in impedance with increasing frequency but with subtle differences in their slopes. The heat-treated samples demonstrate a slightly steeper decline in this region, suggesting a more pronounced capacitive behavior of their surface films. In the high-frequency domain (>1000 Hz), the impedance values converge to similar magnitudes (around 25–30 Ω·cm²) for all samples, corresponding to the solution resistance (Rsoln) values previously noted. This convergence indicates that at high frequencies, the electrical response is dominated by the electrolyte properties rather than the surface characteristics of the samples.

The overall shape of the magnitude plots, when considered alongside the previously discussed phase angle responses, suggests a two-time-constant system. This indicates the presence of both a surface film capacitance and a double-layer capacitance, with the heat-treated samples showing a more distinct separation between these time constants as discussed below.

The electrochemical analysis of the two-time constants can be better understood through the quantification of their characteristic frequencies (fc), which can be calculated using the relationship fc = 1/(2π(RQ)^(1/n)), where R represents the respective resistance (Rpo or Rcor), Q is the constant phase element parameter (Qc or Qcor), and n is the corresponding exponential factor (nc or ncor). For the porous layer (first time constant), this calculation yields characteristic frequencies of approximately 185 Hz for H0, increasing to 2.1 kHz for H650 and further to 4.8 kHz for H950. The charge transfer processes (second-time constant) show distinctly lower characteristic frequencies of 1.2 Hz, 0.25 Hz, and 0.08 Hz for H0, H650, and H950, respectively. The first time constant, manifesting at higher frequencies, corresponds to the porous layer characteristics, while the second time constant at lower frequencies represents the charge transfer processes. The evolution of these time constants with heat treatment demonstrates a systematic transformation of the corrosion protection mechanism. In the as-deposited condition (H0), the porous layer exhibits a relatively low nc value of 0.7153, indicating moderate film homogeneity, with a high capacitance (Qc) of 0.00358 Ω⁻¹ cm⁻² s^n. Heat treatment progressively enhances the passive film quality, as evidenced by the increase in nc to 0.9283 for H950, accompanied by a significant decrease in Qc to 5.30 × 10⁻⁵ Ω⁻¹cm⁻²s^n, suggesting the formation of a thicker more compact oxide layer. The characteristic frequencies of these processes show remarkable separation, with the separation factor (ratio of higher to lower characteristic frequencies) increasing from 154 for H0 to approximately 60,000 for H950, indicating increasingly distinct electrochemical processes at the surface and interface. The resistance distribution undergoes a fundamental shift, as indicated by the Rpo/Rcor ratio changing from 3.687 in H0 to 0.008 in H950, signifying a transition from a porous layer-dominated protection mechanism to one predominantly controlled by charge transfer resistance. This transition aligns with the microstructural evolution, where the increase in carbide count from 674 to 1444 and reduction in average carbide size from 2.56 to 1.06 μm² creates a more uniform corrosion barrier. The decreasing ncor values from 0.8741 (H0) to 0.6252 (H950) reflect increased surface heterogeneity at the charge transfer interface, attributable to the refined carbide distribution. The capacitive behavior shows increasingly distinct interfacial processes, with the Qcor/Qc ratio rising from 1.047 in H0 to 11.509 in H950. These quantitative parameters collectively demonstrate that heat treatment, particularly at 950 °C, fundamentally enhances corrosion resistance through the development of a more effective layered protection mechanism characterized by a homogeneous passive film and well-defined electrochemical processes at different interfaces.

4. Discussion

4.1. Effect of Heat Treatment on Microstructure

When hardfacing material with high carbon and chromium contents is deposited by welding, the fast cooling rates experienced lead toward the undiluted top layer consisting of Cr₇C₃ primary carbides and austenite. This is reflected in studies such as by Atamert and Bhadeshia [24], who have conducted in-depth microstructure and stability studies of similar hardfacing alloys. These researchers found in their study that when weld deposited, the top undiluted layer consists of M₇C₃ and an eutectic mixture of more M₇C₃ carbides with austenite. Austenite is stable at temperatures above 1150 °C. When returning from these high temperatures, some austenite is retained due to fast cooling, imparting some toughness. The remainder of the non-austenite equilibrium constituents should be ferrite and a higher volume fraction of M₇C₃ carbides, as shown via XRD. On heating to temperatures below 1150 °C, the most likely transformation of austenite would be to ferrite [24] [25]. As ferrite dissolves much less carbon than austenite, the excess carbon should produce more carbides. This was found in this study, as the samples were heated to 650 and 950 °C, the total number of carbides increased as shown in Table 3. The number of carbides precipitated after 950 °C is much higher than the number of carbides after 650 °C. Fontana and Greene, when describing knife-line attacks, state that at around 950 °C, M₂₃C₆ carbides in stainless steels dissolve and will reform when slowly cooled [21]. Whilst no M₂₃C₆ carbides were found in this study, Atamert and Bhadeshia [25] found that the secondary carbides that form when cooling from heat treatment are initially M₂₃C₆ carbides due to their good lattice matching relationship with austenite, before transforming to the more stable M₇C₃ carbides. Thus, it is possible that similar dissolution and reformation of carbides may have occurred in this study.

The austenite decomposition occurred more fully after the heat treatment at 950 °C and hence the matrix ferrite content is lower than that of 650 °C heat treated samples. Other researchers have found that in casting high chromium white irons, the dissociation of austenite after heat treatment produced martensite [3,6]. Speculatively, the reason for this difference could be the coarseness of the microstructure produced by casting compared to weld hardfacing. The XRD results suggest the presence of martensite or ferrite, which is difficult to distinguish due to similar crystal structures and lattice parameters [28]. Given ferrite’s limited carbon solubility, which may be speculated to not align with the matrix containing up to 5.3% carbon after heat treatment at 950 °C, more investigation is needed to confirm the presence of martensite.

4.2. Effect of Heat Treatment on Corrosion

Figure 8, Figure 9 and Figure 10 show that the matrix corroded in preference to the carbides. One reason is that the carbide numbers increased after heat treatment, as shown in Table 3, which resulted in the matrix volume decreasing. It was found that heat treatment after 950 °C produced the greatest number of carbides and the least amount of matrix ferrite. As the matrix was galvanically corroding in preference to carbides, 950 °C heat treatment produced the best corrosion performance.

In the study by Varmaa and Kannoorpatti [29], it was found that by using a pH 7 solution containing NaCl, the matrix was corroded in preference to the carbide. They explained this behavior by superimposing the Pourbaix diagrams of Cr₇C₃ and Fe on each other to show that the potentials of 1.1 V (vs. Ag/AgCl) would fall in the active or corrosion regions for the Fe and Cr. M7C3 carbides contain about 43% Fe on average and this was also found by other investigators [24]. The matrix contains, on average, about 77% Fe as shown in Table 6. It was also found in other investigations that the corrosion of high chromium white irons is dominated by the formation of FeO [30], with Salasi et al. hypothesizing that the presence of chloride irons disrupts passive layers and initiates dissolution from the carbide/matrix interface [31]. Given the large differences in the chromium contents of the carbides and the matrix, it is likely that this compositional difference resulted in the corrosion of the matrix. However, when the pH is high and iron is more stable than chromium, it is possible that carbides would instead dissolve in preference to the matrix, as shown by other studies through the use of Pourbaix diagrams [29,32].

While the corrosion resistance was improved after the heat treatment at 950 °C, it is possible that the toughness of the hardfacing may be degraded due to the dissociation of austenite to ferrite and carbides, as also reflected by the reduction in matrix hardness in Figure 4. However, in certain applications where higher corrosion resistance is required, heat treatment would be beneficial. More work is needed to understand the phase transformation of the hardfacing alloys due to heat treatment.

5. Conclusions

This study sought to investigate the effects of heat treatment on high chromium white iron alloys and their electrochemical corrosion behavior as deposited via welding. A comparison was made from as-deposited samples toward those treat treated in a normal atmosphere at 650 °C, as well as 950 °C. Overall, it was found that

• Heat treatment led to the transformation of austenite to ferrite;
• Heat treatment at both 650 °C and 950 °C led to carbide precipitation within the matrix, with 950 °C also leading to the formation of small flecks;
• Galvanic corrosion was found to occur with preferential corrosion of the matrix with respect to the carbides;
• Heat treatment led to improved corrosion resistance compared to the as-deposited sample, with heat treatment at 950 °C displaying the best performance.