The Effect of Quenching and Partitioning (Q&P) Processing on the Microstructure, Hardness, and Corrosion Resistance of SAE 9254 Spring Steel

Abstract

In the present work, the effect of quenching and partitioning cycles on the microstructure, hardness, and corrosion behavior of SAE 9254 spring steel was investigated. Initially, the critical phase transformation temperatures were analyzed by dilatometry. The samples were then treated by four routes of quenching and partitioning in a dilatometer with quenching stop temperatures of 250 and 220 °C. The partitioning temperatures were 300 and 400 °C. The partitioning time was 480 s. Quantitative characterization of austenite and martensite volume fractions was carried out by X-ray diffraction. Qualitative characterization was carried out by optical microscopy and scanning electron microscopy in addition to quantitative assessments of the chemical composition of segregations by EDS. The formation of martensite, retained austenite, and bainite was observed. The dilatometric curves displayed the occurrence of volumetric expansion in the partitioning step, indicating the formation of secondary martensite (fresh martensite) during the final cooling process (final quenching). The mechanical properties were evaluated by Vickers microhardness and nanoindentation tests. There was heterogeneity of hardness inside and outside the banding regions. The electrochemical properties were evaluated by electrochemical impedance spectroscopy and potentiodynamic polarization tests in a 0.1 M H₂SO₄ solution. The best corrosion resistance was achieved for samples quenched at 250 °C and partitioned at 400 °C due to the higher volume fraction of retained austenite when compared to the other heat treatment conditions.

1. Introduction

The pressure on reducing greenhouse gas emissions has driven the development of new materials for the automotive industry. In this context, the manufacturing of several different structural components has experienced many innovative steps in the past decade [1,2]. In this context, suspension components have gained attention, as they are crucial to the safety and ride comfort of vehicles [3]. Helical spring steels are known for their high strength and durability that enable proper absorption of impacts and vibrations [4]. SAE 9254 spring steel is an example of such a high-performance structural component. It is conventionally commercialized with a tempered martensite microstructure that is carefully designed to withstand the severe mechanical loadings during usage of vehicles [5].

Microstructural control is an effective route to tailor the mechanical properties of SAE 9254 steel to meet the growing demands for high strength and lightweight structural automotive components. In this context, heat treatments can be designed to produce multiphase microstructures with specific technical requirements [6]. Hasan et al. [7] optimized the balance between strength and ductility of SAE 9254 spring steel by exploring different austenitization and tempering treatments. The best mechanical properties were obtained for samples austenitized at 910 °C and tempered at 310 °C due to a favorable combination of retained austenite fraction and small prior austenite grain size (PAGS). Santos et al. [8] investigated the effect of the cooling rate on the martensitic transformation kinetics of SAE 9254 spring steel and PAGS. Depending on the cooling rate, the formation of a microstructure consisting of bainite and martensite with a PAGS as small as 24 µm was obtained.

In an attempt to further improve the mechanical properties of spring steels, several researchers have dedicated efforts to develop novel heat treatment routes based on the quenching and partitioning approach (Q&P). The development of advanced technologies such as Q&P processing has become a powerful tool for the formation of multiphase microstructures with martensite, lower bainite, and retained austenite, resulting in excellent tensile strength, higher ductility, and good impact toughness [9]. The mechanical properties of spring steels must be carefully designed so that these materials are able to withstand the high mechanical loads to which they are typically subjected in service. In this way, a balance between tensile strength and high ductility is possible in Q&P processing. The improved mechanical behavior of the steels is attributed to an increase in the retained austenite (RA) fraction in the martensite matrix [10]. The presence of RA is sufficient to promote the TRIP effect (transformation-induced plasticity), raising the work-hardening rate and postponing necking, thus leading to a good balance between strength and ductility [11]

In the Q&P treatment, the steel is subjected to four different steps: (i) austenitization; (ii) quenching to promote the formation of a mixed martensitic–austenitic microstructure; (iii) partitioning at the quenching temperature or at a higher temperature, either above or below the initial temperature of the martensitic transformation (Ms); (iv) final cooling [12]. The final microstructure exhibits a multiphase constitution, comprised of bainite, martensite, and carbon-enriched RA [13]. The main mechanism responsible for the increased ductility of steels subjected to Q&P processing is carbon partitioning from the supersaturated martensite to RA during isothermal treatment. This mechanism is essential for the stabilization of RA and increases its relative fraction, resulting in the combination of high strength and ductility [14].

The Q&P treatment is often applied to Si-Mn steels with carbon contents up to 0.6 wt.% [15,16]. Manganese is known as an inhibitor of austenite decomposition, improving its stability during the Q&P steps. Silicon, in turn, is used to inhibit the precipitation of carbides in the martensite matrix, avoiding excessive hardening during the partitioning stage [17]. The Q&P treatment has been applied to spring steels in order to regulate its mechanical properties. Partitioning temperatures adopted by some authors ranged from 250 °C to 400 °C, depending on the phase transformation temperatures identified by dilation curves [14,15,18]. Masoumi et al. [18], for instance, proposed an innovative two-step Q&P treatment of SAE 9254 spring steel to generate a high-strength, multiphase, tough microstructure, consisting of a mixture of tempered martensite, lower bainite, nanoscale carbide precipitates, and approximately 5% RA.

While the effect of Q&P treatments on the mechanical behavior of spring steels has received much attention, the corrosion properties have been little explored. The need for conducting studies on the corrosion behavior of Q&P steels has been highlighted by Yang et al. [19]. They investigated the effect of Q&P treatment on the corrosion resistance of Fe–0.4C–1.5Mn–1.5Si steel in a 3.5 wt.% NaCl solution. When compared to samples subjected to conventional quenching and tempering heat treatment, the Q&P samples exhibited superior corrosion resistance due to lower residual stresses and the presence of retained austenite. Conversely, Mehner et al. [20] observed an increase of the corrosion rate of a Q&P advanced high-strength steel when compared to a conventionally quenched and tempered sample after immersion in a 0.25 mol·L⁻¹ H₂SO₄ solution. The formation of microgalvanic cells between the anodic carbon-depleted martensitic phase and the cathodic retained austenite accelerated the corrosion kinetics of the Q&P steel. The current literature lacks information on the effect of Q&P treatments on the corrosion behavior of SAE 9254 spring steel. To fill this gap, the aim of the present work was to study the correlation between the microstructure and corrosion properties of SAE 9254 spring steel in a 0.1 M H₂SO₄ solution.

2. Materials and Methods

2.1. Material and Q&P Treatments

A commercial SAE 9254 spring steel bar was used in the present work, and its chemical composition is given in

Table 1. Chemical composition (in wt.%) of the SAE 9254 spring steel.

C	Si	Mn	P	S	Cr	Mo	Ni	Al	Cu	Ti	N	Fe
0.544	1.338	0.699	0.009	0.003	0.658	0.003	0.127	0.226	0.150	0.005	0.003	Balance

. The material was kindly provided by Thyssenkrupp Brasil Ltd.a. (São Paulo-SP, Brazil) in the form of a round bar with a diameter of 11.2 mm.

Heat treatments were carried out in a quenching dilatometer (TA Instruments, DIL 805 A/D, DE, USA) using cylindrical specimens of 10 mm lengths cut from the original round bar using water-cooled aluminum oxide blades in a cut-off machine (Teclago, CM 120, São Paulo, Brazil). The cooling steps of the heat treatments were carried out using helium gas to avoid damage to the specimen surface. The Q&P conditions are shown in Figure 1. Briefly, the specimen was austenitized at 900 °C for 720 s then quenched at a cooling rate of 100 °C/s at holding temperatures of 220 °C and 250 °C for 600 s. Two partitioning temperatures were set, 300 °C and 400 °C, for 480 s. Next, the specimens cooled down to room temperature at a cooling rate of 50 °C/s. Based on these conditions, four different samples were obtained, as described in

Table 2. Sample code of the Q&P SAE 9254 spring steel.

Samples	Quenching Stop Temperature (TQ) (°C)	Partitioning Temperature (TP) (°C)
250–400	250	400
250–300	250	300
220–400	220	400
220–300	220	300

.

2.2. Microstructural Characterization

The microstructure of the SAE 9254 steel samples was examined by optical metallography (Zeiss, AxioVert A1, Oberkochen, Germany). Firstly, the specimens were embedded in phenolic resin. Surface preparation was comprised of grinding with waterproof silicon carbide papers up to grit 600 followed by polishing with alumina suspension up to grit 1 µm. Next, the samples were etched with Nital 3% for 5 s. Additional characterization of the microstructures was carried out by scanning electron microscopy (Tescan, Vega3, Waltham, MA, USA) coupled to an energy dispersive X-ray spectroscopy (EDS) detector for elemental mapping.

2.3. X-Ray Diffraction

The volume of retained austenite was determined by X-ray diffraction using a Stresstech Xstress 3000 G2R (Vaajakoski, Finland) instrument equipped with a Cr anode (Cr-Kα radiation with a wavelength of 2.29 Å) using a vanadium filter. The measurement diffraction angle was 156.45°. XTronic software was used for data processing to obtain the integrated peaks of austenite (200 and 220) relative to ferrite (200 and 211). Then, the volume of retained austenite (V_γ) was calculated based on Equation (1):

$${\mathrm{V}}_{\mathsf{\gamma }}=\left(1-{\mathrm{V}}_{\mathrm{c}}\right)\left(1/\mathrm{q}\sum _{\mathrm{j}=\mathrm{l}}^{\mathrm{q}}{(\mathrm{I}}_{\mathsf{\gamma }\mathrm{j}}/{\mathrm{R}}_{\mathsf{\gamma }\mathrm{j}})\right)/\left[\left(1/\mathrm{p}\sum _{\mathrm{i}=\mathrm{l}}^{\mathrm{p}}{\mathrm{I}}_{\mathsf{\alpha }\mathrm{i}}/{\mathrm{R}}_{\mathsf{\alpha }\mathrm{i}}\right)+\left(1/\mathrm{q}\sum _{\mathrm{j}=\mathrm{l}}^{\mathrm{q}}{\mathrm{I}}_{\mathsf{\gamma }\mathrm{j}}/{\mathrm{R}}_{\mathsf{\gamma }\mathrm{j}}\right)\right]$$

(1)

In Equation (1), V_c is the volume of carbides, q is the number of austenite peaks, I_γj is the integrated intensity of a specific austenite peak, R_γj is the parameter of theoretical integrated intensity for austenite, p is the number of ferrite peaks, I_αi is the integrated intensity of a specific ferrite peak, and R_αi is the parameter of the theoretical integrated intensity for ferrite.

2.4. Vickers Microhardness and Nanoindentation Tests

Vickers microhardness measurements were made with specimens previously embedded and prepared as described in Section 2.2 using a Mitutoyo HM-210 (Kanagawa, Japan) hardness tester. The testing load was 2.94 N, and the dwell time was 15 s. Five indentations were made on each specimen.

Nanoindentation tests were conducted using an Anton Paar NHT² instrument (Graz, Austria) equipped with a Berkovich indenter. The maximum load was 50 mN, and the loading and unloading rates were 30 mN/min. Before the measurements, the specimens were subjected to chemical etching using a two-step procedure to clearly distinguish between different microconstituents. Firstly, a Picral 4% solution was used, and right after, the specimen was etched with a sodium metabisulfite 10% solution for 15 s.

2.5. Electrochemical Tests

The tests were conducted in a potentiostat/galvanostat Aubolab M101 (Utrecht, The Netherlands). The specimens were connected to a copper wire using a conductive silver colloidal paste. Next, they were cold mounted with epoxy resin and ground with silicon carbide waterproof papers up to grit 600. A conventional three-electrode cell configuration was used for all tests, with the SAE 9254 specimen as the working electrode, Ag/AgCl (KCl, 3.0 M) as the reference electrode, and a platinum wire as the counter electrode. The electrolyte consisted of a 0.1 M H₂SO₄ solution at room temperature. Initially, the open circuit potential (OCP) was monitored for 3600 s. Subsequently, electrochemical impedance spectroscopy (EIS) measurements were conducted at the OCP. The amplitude of the sinusoidal perturbation signal was ±10 mV (rms), with an acquisition rate of 10 points per decade, in the frequency range of 100 kHz to 10 mHz. Potentiodynamic polarization tests were conducted right after the EIS measurements at a sweep rate of 1 mV·s⁻¹ in the potential range of −250 mV versus the OCP up to +1.0 V versus the OCP.

3. Results and Discussion

3.1. Dilatometric Analysis

Figure 2a–c show the variation in length as a function of temperature in the Q&P-treated steel (250–400/250–300/220–400/220–300) and quenched steel, justifying the selection of these specific temperatures for the Q&P process. In Figure 2a,b, dilatometric curves obtained during heating exhibit temperature increases up to Ac1. In this part of the thermal cycle, the length of the specimen increases. However, at a certain point, a contraction occurs. This contraction indicates the transformation from BCC ferrite to FCC austenite between Ac1 and Ac3 in the range of 785 to 810 °C. As shown in the dilatation curves, the lengths of the specimens initially increase with the increase in temperature. During the cooling stage, at certain specific regions of the dilatometric curves, the length remains constant over time. Inflections in cooling curves indicate the start temperature of the martensitic transformation (Ms = 286 °C). After each expansion stage is completed, as shown in Figure 2c, both specimens under the partitioning processes at 300 and 400 °C take more easy carbon partitioning from the saturated martensite into the surrounding austenite. Subsequently, during the final cooling process (final quenching), secondary martensite is formed. This microconstituent is different from the primary martensite obtained during the initial cooling process (initial quenching). The secondary martensite is an extremely hard phase, because it is highly enriched with carbon, while carbon partitioning into austenite results in very low carbon content in the primary martensite. Subsequently, during the cooling process, the final microstructure consists of martensite matrix and retained austenite, although competitive processes such as carbide precipitation and bainite formation can also occur [20,21].

The expansion and contraction behaviors of all the Q&P specimens are exhibited in Figure 3. Inflections showing the expansion stage are seen in the dilatation vs. time curves in ranges from 824 s to 1117 s during the quenching temperature (TQ) at 220 and 250 °C, as indicated in Figure 4. The possible explanation for the expansion is related to the decomposition of some carbon-enriched austenite to untempered high-carbon martensite [14,22]. The cooling after austenitizing to below Ms cooling and during the quenching temperature (TQ) promotes the carbon partitioning. For the quenching temperature of 250 °C, a higher expansion is noticed when compared to the quenching temperature of 220 °C, possibly due to higher carbon diffusion at 250 °C.

The gradual expansion during the partitioning stages at 300 and 400 °C in ranges between 1120 and 1600 s also shows expansions in dilatation vs. time curves. It is noticed that increasing the partitioning temperature (TP) from 300 to 400 °C results in higher expansion during isothermal treatment. This expansion behavior is a complex discussion, which can be affected by different competing mechanisms, such as the volume expansion due to the partitioning of carbon from supersaturated martensite to austenite and also the isothermal decomposition of austenite resulting in isothermal martensite formation or the possibility of bainitic transformation in the microstructure [14,22,23].

3.2. Transformed Phase Fraction

The volume fraction of the martensitic phase was determined by applying the lever rule to the dilatometry curve, based on the volume changes identified from the slope segments corresponding to each phase transformation, as shown in Figure 5, following the method proposed in the literature [8]. L(T) represents the specimen length at temperature T and L_a(T) and L_b(T) correspond to the measured lengths from the slope curves, assuming a fully austenitic structure and the maximum martensite fraction at temperature T, respectively.

The volumetric fractions of martensite and austenite for a specific quenching temperature were calculated using Equation (2) [8]. The Koistinen–Marburger (K-M) model, based on displacive transformation mechanisms, was developed to characterize the kinetics of martensitic transformation. Nevertheless, the Koistinen–Marburger (K-M) model does not give precise values in the very beginning, which can promote uncertainties in its application [22,23].

f_(T)= 1 − exp (−α_m(T_KM − T))

(2)

In this context, f_(T) denotes the fraction of martensite formed, α_m is the transformation rate parameter, and T_KM refers to the theoretical martensite start temperature (Ms). Van Bohemen and Sietsma [22] suggested that TKM can be estimated using an empirical equation for Ms based on chemical composition. Consequently, experimental data were examined to determine the quenching temperatures for the design of a multicomponent microstructure. Following quenching temperatures at 220 °C and 250 °C, the primary martensite fractions were estimated from the cooling curves through linear analysis in conjunction with the lever rule, as illustrated in Figure 6.

The Koistinen–Marburger (K-M) model parameters were obtained by the fitted experimental curve represented by a red dashed line, indicating the volume fraction of martensite. The values of parameters α_m and T_KM were 274.7 °C and = 0.0336 K⁻¹, respectively. Therefore, the estimated martensite fractions for the quenching temperatures (QT) at 220 °C and 250 °C were 64.5 and 19%, respectively.

3.3. Microstructural Examination

Optical micrographs of the SAE 9254 spring steel are displayed. As can be seen in Figure 7a–d, the microstructures of the Q&P-treated samples (250–400/250–300/220–400/220–300) reveal very similar microstructural features. The microstructures consisted of fully martensitic matrix (brown phase), indicating the presence of proeutectoid ferrite (PF), and a white phase located at the boundaries of the martensite (M). As mentioned by Yihao and colleagues [24], 51CrMnV spring steel after quenching at 120° C for 180 s and partitioning at 350 °C for 1200 s revealed a microstructure that basically consisted of martensite, proeutectoid ferrite, and retained austenite.

The SEM micrographs of the Q&P-treated samples are exhibited in Figure 8a–d. The partitioning temperature plays an important role in the morphological aspects of the microconstituents of the Q&P steels, changing in particular the shape of the martensite and retained austenite.

All Q&P-treated samples exhibited the formation of a multiphase microstructure with similar features, mainly consisting of PF, martensite–austenite islands (M/RA), and retained austenite film (RA film) [14]. The PF preferentially precipitates at prior austenite grain boundaries (PAGBs) that coincide with the martensite boundaries after the quenching process [24]. The M/RA island formation is derived from the conversion of unstable RA to martensite during the final quenching, resulting in the formation of a mixed M/RA structure. The retained austenite present in the microstructure exhibits a film-like morphology along the interface boundary [25]. Dany et al. [26] reported that carbon partitioning plays a fundamental role in the formation of a complex microstructure containing PF, pearlite, and lower bainite during heat treatments of SAE 9254 steel. The authors highlighted the importance of the PF at grain boundaries, contributing to the excellent steel ductility. Other authors [12,25] also reported the formation of multiphase microstructure in Q&P steels, highlighting that the retained austenite cannot easily be distinguished by SEM micrographs, even though the Q&P-treated steels exhibit more areas in which austenite is likely present, exhibiting different morphologies.

3.4. XRD Results

The volume fractions of RA in the different Q&P samples were estimated by X-ray diffraction, following the procedure described in Section 2.3. The results are shown in Figure 9. The result obtained for the quenched and tempered sample (Q&T) is included for comparison. The RA volume fractions for the Q&P-treated samples at 250–400/250–300/220–400/220–200 were 20.9%, 14.4%, 19.7%, and 13.9%, respectively.

Statistical data indicated that the volume fraction of RA in the Q&P spring steel increased with the partitioning temperature. This is because during the Q&P process, carbon is partitioned to RA, decreasing its concentration in the martensitic matrix and stabilizing the austenite. Carbon diffusion is facilitated at higher partitioning temperatures [25]. Considering the 220 and 250 °C quenching temperatures, it is easily observed that there are no significant changes in the retained austenite values. In a study with low-alloy 42SiCr steel heat-treated by the quenching and partitioning process, Ivo et al. [27] reported an increase in the retained austenite fraction of the treated steel by the Q&P process (11.5%) when compared to quenched and tempered steel (2.9%), which is close to the samples subjected to the partitioning temperature of 300 °C in the present work.

3.5. Vickers Microhardness and Nanoindentation Measurements

The Vickers microhardness values are shown in

Table 3. Results of the RA volume fraction (%) and elasto-plastic material properties obtained from the nanoindentation tests (nanohardness H and Young’s modulus E).

Sample	RA (%)	Hardness (HV)	H (GPa)	E (GPa)
250–400	20.9 ± 1.7	523 ± 9	5.13 ± 0.32	222.72 ± 18.03
250–300	14.4 ± 2.1	602 ± 12	5.27 ± 0.54	209.69 ± 13.70
220–400	19.7 ± 0.2	567 ± 11	5.22 ± 0.23	198.38 ± 06.59
220–300	13.9 ± 1.3	529 ± 8	5.24 ± 0.43	218.38 ± 30.51

, along with the retained austenite volume fraction determined by XRD, the nanohardness values, and the Young modulus obtained by the nanoindentation tests.

As can be seen, the Q&P-treated samples (250–400/250–300/220–400/220–300) presented hardnesses from 523.1 to 601.8 HV. Load–depth curves obtained from the nanoidentation tests and the corresponding indentation markings are displayed in Figure 10.

The hardness measurements show that the combination of the higher quenching and partitioning temperatures reduced the hardness values. This is due to the faster carbon diffusion from martensite to unstable austenite, which results in a lower relative amount of the hard-phase martensite and a higher fraction of the soft-phase RA in the microstructure. Marques et al. [28] reported that higher quenching temperatures tend to reduce the hardness values of martensitic stainless steel (0.2% C and 12% Cr) after quenching and partitioning (Q&P). The Q&P-treated steel achieved a hardness value of 569.5 HV.

As seen in Figure 10, the load–depth curves of the Q&P SAE 9254 spring steel exhibited different evolutions, depending on the region in which the indentation was made. This behavior can be related to the multiphase microstructure of the steel. Each curve contains the loading step, the dwell time with the maximum loading, and the unloading step. The average nanohardness values of the Q&P-treated samples were from 5.13 to 5.27 GPa, and the Young’s modulus values were from 198.3 to 222.7 GPa. Observing the size of the indentations in each distinct region in the microstructures, it is noticed that the hardness values varied for all the Q&P-treated samples. Lower hardness values were obtained for the samples subjected to a higher partitioning temperature, which could be explained due to partial carbide dissolution and higher carbon partitioning, resulting in higher volume fractions of RA. This behavior shows the different microstructural characteristics associated with the material properties obtained from the nanoindentation tests for the hardest and the softest microconstituents, which possibly correspond to untempered martensite and ferrite, respectively [29].

It is noteworthy that factors such as different grain orientations, non-uniform chemical distribution, grain boundaries, and dislocation density within grains also contribute to the nanomechanical behavior of the sample [30]. During the plastic deformation generated in a multiphase microstructure with RA and martensitic matrix, in the loading step, the stress is transferred to the phase with higher hardness. When the transferred stress is large enough, the martensite (hard phase) starts to deform plastically to reach its elastic limit. Therefore, the volume fraction of RA can influence the high initial strain hardening of martensite, which is possibly the main factor that leads to the high stress increment observed for the SAE 9254 spring steel in the present work. Mohtadi-Bonab et al. [20] studied the quenching and partitioning (Q&P) process of commercial pearlitic carbon–silicon steel. The authors reported that partial nanocarbide dissolution and higher carbon partitioning at a high partitioning temperature (Q&P-220-375) promoted lower hardness (6.36 ± 0.25 GPa) when compared to the sample treated at a lower partitioning temperature (Q&P-220-325), exhibiting average hardness values of 6.97 ± 0.30 GPa.

3.6. Corrosion Behavior

3.6.1. Electrochemical Impedance Spectroscopy (EIS)

The EIS diagrams of the Q&P-treated SAE 9254 steel are shown in Figure 11. The results were obtained after 1 h of immersion in the 0.1 M H₂SO₄ solution at room temperature. The Nyquist plots (Figure 11a) are characterized by one flattened capacitive loop for all samples, independently of the Q&P condition, indicating that the corrosion mechanism was not altered by the different Q&P processing parameters. Conversely, the diameter of the semicircle, which is associated with the corrosion resistance of the electrode [31], was dependent on the Q&P condition. In this respect, samples 250–400 and 220–400 exhibited the biggest capacitive loops, thereby presenting higher corrosion resistance when compared to the samples subjected to the lower partitioning temperature (250–300 and 220–300). The experimental EIS data were simulated with an equivalent electrical circuit (EEC) in order to give a quantitative interpretation of the results. The circuit shown in Figure 12 was used to simulate the data of all samples. A constant phase element (CPE) was used instead of an ideal capacitor to account for the roughness and heterogeneities of the electrode [32]. The impedance of a CPE (Z_CPE) is given in Equation (3), where Q is the CPE magnitude, j is the complex operator, ω is the angular frequency, and n is the CPE exponent (n = 1 for an ideal capacitor and n = 0 for a resistor) [33]. In this circuit, R₁ is the electrolyte resistance, Q₁ is associated with the double layer capacitance, and R₂ is the charge transfer resistance. This model was also adopted by other authors to simulate the EIS response of uncoated carbon steels in acidic solutions [34,35]. The fitting results are shown in

Table 4. EIS fitting parameters of the Q&P-treated SAE 9254 spring steel after 1 h of immersion in the 0.1 M H₂SO₄ solution at room temperature.

Sample	R1 (Ω⋅cm2)	Q1 (10−4⋅F⋅cm−2⋅sn−1)	R2 (Ω⋅cm2)	n1
250–400	5.5	1.39	506	0.88
250–300	4.8	2.07	157	0.89
220–400	8.5	1.61	329	0.89
220–300	5.6	1.62	243	0.87

.

$${\mathrm{Z}}_{\mathrm{C}\mathrm{P}\mathrm{E}}={\left[\mathrm{Q}{\left(\mathrm{j}\mathsf{\omega }\right)}^{\mathrm{n}}\right]}^{-1}$$

(3)

The values of charge transfer resistance shown in Table 4 are compatible with those reported by other authors for carbon steels immersed in acidic solutions [35]. Bode plots are displayed in Figure 11b,c. In Figure 11b, the phase angle plots are characterized by one peak, associated with one time constant due to the charge transfer reactions at the solution–electrode interface, freely corroding in the acidic medium. The phase angles dropped off at low frequencies, indicating the loss of the capacitive character and the onset of corrosion processes. The samples subjected to the highest partitioning temperature of 400 °C (250–400 and 220–400) exhibited higher capacitive phase angles, with the maximum peak reaching −70° for the 250–400 condition. The phase angles of the 250–300 and 220–300 conditions were lower, indicating their lower capacitive response and higher susceptibility to corrosion. In Figure 11c, the capacitive response of the samples can be observed between approximately 10³ and 10 Hz, as denoted by the −1 slope in this frequency region.

For lower frequencies, the impedance modulus assumed a resistive behavior, being independent on the frequency, which is associated with the corrosion processes taking place at the solution–electrode interface [36]. The magnitude of the impedance modulus at a low frequency can be used as a measure of the corrosion resistance [37]. In this regard, the highest impedance moduli were obtained for the 250–400 and 220–400 conditions, confirming their relatively high corrosion resistance when compared to the samples subjected to the lowest partitioning temperature (250–300 and 220–300).

3.6.2. Potentiodynamic Polarization Curves

From the previous section, it can be inferred that the partitioning temperature affected the corrosion resistance of the SAE 9254 spring steel. To gain further understanding of the influence of the Q&P processes on the corrosion properties of the SAE 9254 samples, additional characterization of the electrochemical behavior of the Q&P-treated samples was carried out by potentiodynamic polarization tests. Potentiodynamic polarization curves of the Q&P-treated SAE 9254 spring steel samples after 1 h of immersion in the 0.1 M H₂SO₄ solution at room temperature are shown in Figure 13. The vales of corrosion potential (E_corr) and corrosion current density (i_corr) were determined from these curves using the Tafel extrapolation method, considering only the cathodic branches. The results are displayed in

Table 5. Electrochemical parameters determined from the potentiodynamic polarization curves shown in Figure 8.

Sample	Ecorr (mV vs. Ag/AgCl)	icorr (µA⋅cm−2)
250–400	−485	58
250–300	−491	151
220–400	−462	74
220–300	−458	107

.

As seen in Figure 13, all samples exhibited active corrosion behavior in the sulfuric acid solution, as indicated by the continuous increase in the anodic current densities with the applied potential. The corrosion potentials (E_corr) varied from −491 to −458 mV_Ag/AgCl, depending on the Q&P condition. These values agree with the results obtained by other authors for carbon steels in sulfuric acid solutions [38,39]. It is noteworthy that the values of i_corr of the 250–400 and 220–400 conditions were lower compared to those obtained for the samples subjected to the lowest partitioning temperature. This result agrees well with the EIS data described in Section 3.6.1. Therefore, the best corrosion resistance was observed for the samples subjected to the highest partitioning temperature (400 °C).

As shown in Section 3.2, the main structural effect of the Q&P processing conducted in the present work was related to the relative volume fraction of retained austenite (RA), depending on the partitioning temperature. From Figure 11, it is possible to identify a strong effect of this parameter on the RA fraction of the Q&P-treated SAE 9254 spring steel. The samples subjected to the lowest partitioning temperature (250–300 and 220–300) exhibited RA fractions of 14.4% and 13.9%, respectively. By raising the partitioning temperature to 400 °C, the RA fractions increased to 20.9% and 19.7% for the 250–400 and 220–400 samples, respectively. Mehner et al. [20] observed that the corrosion rate of high-strength steel in a diluted H₂SO₄ solution increased after Q&P processing due to a higher fraction of RA. Because of the different electrochemical activities of the cathodic carbon-rich RA and the anodic carbon-depleted martensitic matrix, microgalvanic cells were formed, accelerating the corrosion kinetics. Conversely, the results obtained in the present work point to a beneficial effect of the RA volume fraction on the corrosion behavior of the Q&P-processed SAE 9254 spring steel. Therefore, another mechanism should dominate over the microgalvanic effects described by Mehner et al. [20]. Indeed, according to the literature [40,41], the corrosion behavior of carbon steels subjected to Q&P processing can be improved, depending on the treatment parameters. This effect can be explained by the volume fraction of retained austenite. For example, in steels subjected to regular quenching and tempering heat treatments, retained austenite is absent or typically present in very low fractions (below 1%). In this case, carbon is dissolved in the martensitic matrix. In Q&P-treated steels, a different carbon distribution occurs due to the increased fraction of retained austenite. The martensitic matrix becomes depleted in carbon after the Q&P processing. Consequently, lattice distortions are minimized, reducing the internal residual stresses in the microstructure. Moreover, as the volume fraction of RA increases, internal residual stresses are further decreased, favoring the reduction in the corrosion rate [19]. The lower i_corr values of the 250–400 and 220–400 conditions can be associated with their higher RA fractions. The partitioning temperature played a major role in the microstructure of the SAE 9254 spring steel which, in turn, influenced its electrochemical behavior.

4. Conclusions

The effect of quenching and partitioning processing on the microstructure and corrosion behavior of SAE 9254 spring steel in a 0.1 M H₂SO₄ solution was examined in the present work.

Based on dilatometric studies, the partitioning temperatures were selected as 300 °C and 400 °C. The partitioning time was 480 s for both temperatures. The microstructure of the treated samples was comprised of a multiphase structure, consisting of martensite, proeutectoid ferrite, and retained austenite. The partitioning temperature exhibited a remarkable effect on the volume fraction of the retained austenite. The samples subjected to the highest partitioning temperature (400 °C) showed higher volume fractions of RA (up to 20.9%) when compared to those subjected to the lowest partitioning temperature (14.4%).

The combination of the higher quenching and partitioning temperatures (Q&P-250-400) exhibited a lower hardness value (5.13 GPa). This reduction revealed a lower amount of hard-phase martensite and a higher soft-phase austenite fraction due to the faster carbon diffusion from martensite to unstable austenite.

The corrosion behavior in the 0.1 M H₂SO₄ solution was affected by the partitioning temperature. The charge transfer resistance of the samples treated at the highest partitioning temperature was increased. In the same regard, the corrosion current density (i_corr) was lower at this condition, slowing down the corrosion kinetics, likely due to a reduction in lattice distortion and internal residual stresses in the samples with higher volume fractions of RA.