Influence of a Plasma Nitriding Treatment on the Corrosion Behavior of API 5L X70 Steel in Simulated Soil Solution

Abstract

In this work, plasma nitriding was carried out to improve the corrosion resistance of API 5L X70 steel. The process was conducted at different treatment times, 4, 6, 8, and 10 h, to determine which one provides greater resistance to corrosion. The conditions under which the nitriding was carried out were as follows: a mixture of 20% N₂ and 80% H₂ at 3 torr pressure, a current of 2.6 × 10⁻⁶ A, a voltage of 360 V, and the temperature inside the plasma chamber was 550 °C. The blank and nitrided materials were characterized using dispersive energy spectroscopy and scanning microscopy to study their morphology and chemical composition. In addition, open potential circuit, electrochemical impedance spectroscopy, and potentiodynamic polarization curves in simulated soil solution were performed to evaluate the materials’ corrosion resistance. The treatment achieved at 10 h presented the greatest corrosion resistance, reducing the corrosion current density up to three orders of magnitude. The thickness reached 678.75 µm for this condition.

1. Introduction

The oil industry is one of the most studied fields in corrosion. The main materials used in oil ducts [1] are steels (HSLA), which contain elements (V, Ti, Nb) at different concentrations and a low carbon content (0.08–0.12 wt %) [2,3,4,5]. Corrosion affects pipeline networks, leading to high economic losses and a potential impact on the environment if a leak occurs [6,7]. The main aggressive species found in pipelines are NaCl and corrosive gases such as CO₂ and H₂S [8,9,10,11,12,13]. Additionally, given the working environmental conditions, these steels are susceptible to stress corrosion cracking [14,15,16,17,18,19,20]. Most leaks in pipelines are due to failures from both internal and external corrosion (33% and 12%, respectively) [7,20,21]. External corrosion depends on the degree of aggressiveness of the soil in which the pipes are buried [22,23,24,25,26,27,28]. Over time, different soil samples have been analyzed, revealing various degrees of aggressivity, which were classified as follows: NS1 to NS4, NS4 being the highest. This solution is also referred to as a simulated soil solution [29]. For the corrosion protection of buried pipelines, different methods exist—including coatings and/or cathodic protection—to prevent leaks and ruptures [30,31,32]. Other methods for improving the corrosion resistance in different areas are coatings of titanium nitride (TiN) or carburizing with plasma electrolytic treatment [33,34]. However, currently, plasma nitriding treatment is mainly used in areas such as machine tool manufacturing, the automotive industry, etc. Since this surface heat treatment significantly improves resistance to wear, fatigue, and corrosion resistance, one of its main advantages is that it is more environmentally friendly compared to other nitriding technologies [35,36,37,38,39]. In the thermochemical treatment of plasma nitriding, several parameters can be controlled, depending on the application; some of these parameters include the type of gases, their mixture, the treatment time, and the temperature [40,41,42,43,44,45,46,47]. This treatment is easy to apply on surfaces; however, it can be challenging on geometrically complex surfaces [35,41].

In this study, how plasma nitriding treatment influences corrosion resistance at different treatment times is analyzed. It aims to determine which thermal treatment duration provides the best protection against corrosion and record the behavior of the material over time. Additionally, the study aims to understand how nitrides affect the corrosion of API 5L X70 steel when it is in contact with a simulated corrosive soil solution.

2. Materials and Methods

2.1. Samples Preparation

The API 5L X70 steel was cut into cubes of 1 cm per side with a disk cutter at low revolutions to achieve a clean and precise cut. The samples were polished with silicon carbide sandpaper, progressing from 80 to 2000 grain size. After that, the surface was polished with alumina until it developed a mirror-like finish. After polishing, the samples were washed with acetone and alcohol before being stored in a desiccator for later use [42,43,44,45,46,47,48,49,50,51].

2.2. Plasma Surface Modification

The discharge chamber was made of glass, and the base was made of aluminum. The power supply had an alternating current capacity of 360 volts and 0.3 amps. The mechanical vacuum pump was a Leibold (Trivac model D10) with a pumping capacity of 6 dm³ per second, reaching a vacuum pressure of 1 × 10⁻³ torr. The Pirani MKS vacuum gauge (model 945) had a measurement range from 1 × 10⁻³ torr to atmospheric pressure. Electrical variables were measured with an Agilent multimeter (model 344401A) with a capacity of 0–700 V and 1 A. The pressure was controlled using a Matheson flowmeter through a needle valve.

The experimental parameters were the following: the discharge was carried out 360 V, the current was 2.6 × 10⁻⁶ A, and electrodes were separated 5 mm apart. The gas mixture consisted of 80% hydrogen and 20% nitrogen at a pressure of 3.0 Torr, conditions previously used successfully in this laboratory and by other authors [37,39,43,52]. A vacuum of approximately 10⁻³ Torr was also created before treating the pieces. Previous research has explored nitriding treatments at various durations to determine the optimal time, considering factors such as temperature, current, and gas concentration [40,43]. The thermochemical treatment was conducted for 4, 6, 8, and 10 h to determine the optimal treatment time and its influence on the thickness of the nitrided layer [36,38]. All samples were prepared in triplicate to verify reproducibility.

2.3. SEM and EDS

The electrode surface of X70 steel and materials having undergone thermochemical treatments at different times were evaluated using a scanning electron microscope on a cross-section to determine the nitride layer’s thickness after 10 h of treatment. The EDS were also analyzed to obtain qualitative information about the elemental composition of the surface of the blank samples and materials under different conditions.

2.4. X-Ray Diffraction (XDR)

The spectrum was recorded by a Bunker 2D PHASER diffractometer (35 kV, 20 mA, Japan) with Cu Ka radiation ($\lambda =1.548 \AA$) at a scan rate of 2°/min, in the 2θ angle range of 10° to 90°.

2.5. Corrosion Protection Measurements

Corrosion tests were performed using electrochemical techniques: open circuit potential (OCP), potentiodynamic polarization curves, and electrochemical impedance spectroscopy, using an ACM Instruments Gill AC potentiostat and an electrochemical cell with three electrodes—a working electrode (X70 steel and X70 nitrided steel), an Ag-AgCl reference electrode, and an auxiliary graphite electrode. Simulated soil solution was used as the corrosive medium, the chemical composition of which is provided in

Table 1. Chemical composition of simulated soil solution [11,12,13,14,16,17,18,19,20].

Components	Concentration (mg/L)
KCl	122
NaHCO3	483
CaCl22H2O	181
MgSO47H2O	131

.

OCP measurements were recorded for 20 min until the potential stabilized. Electrochemical impedance spectroscopy measurements were performed by applying AC signal with an amplitude of ±10 mV over a frequency range from 10 khz to 0.01 Hz. Potentiodynamic polarization curves were obtained by polarizing the working electrode from ±1000 mV relative to its corrosion potential at a scan rate of 60 mV/min. All tests were conducted in triplicate to verify reproducibility and were based on other authors’ tests and ASTM standards (G03, G59, and G106) [2,6,7,8,53,54].

2.5.1. Corrosion Inhibition Efficiency (%η)

The corrosion inhibition efficiency (%η) was calculated from the corrosion density (i_corr) using the following relation:

$$\%\mathsf{\eta }={\displaystyle \frac{i_{corr blank}-i_{corr treatment}}{i_{corr blank}}}\times 100$$

(1)

where (i_corr) is the corrosion current density without treatment and (i_{corr treatment}) is the corrosion current density with thermochemical treatment [6,49,50,51].

2.5.2. Calculation of the Electrochemical Double Layer C_dl

Based on literature, impedance values were used to determine the electrochemical double layer capacitance (C_dl), obtained using Equation (2):

$$C_{dl}={\displaystyle \frac{1}{2\pi f_{max}{R}_{ct}}}$$

(2)

where $f_{max}$ is the frecuency at which imaginary component of the impedance is maximum, and R_ct is the charge transfer resistance [49,50,51].

3. Results and Discussion

3.1. SEM and EDS Analysis

3.1.1. Blank (Steel X70)

EDS analysis of the blank steel X70 steel sample showed a carbon content of 14 at.% (Figure 1 and

Table 2. The results show the percentage of the elements present in the blank sample.

Element	Mass [%]	Normalized Mass [%]	Atoms [%]
Carbon	3.32	3.42	14.08
Aluminum	0.08	0.08	0.15
Silicon	0.18	0.19	0.33
Chrome	0.23	0.23	0.22
Manganese	0.63	0.65	0.58
Iron	92.59	95.31	84.52
Nickel	0.13	0.13	0.11
	97.15	100.00	100.00

), along with other alloying elements such as Al, Si, Mn, Cr, and Ni. This is consistent with previously reported compositions [15,17,25,26,27,28].

3.1.2. EDS of Nitrided Steel with 4 h of Treatment

The results of the EDS analysis demonstrated the main elements presented in nitrided steel after 4 h of treatment (see Figure 2 and

Table 3. The results show the percentage of the elements present in the sample after 4 h of treatment.

Element	Mass [%]	Normalized Mass [%]	Atoms [%]
Carbon	2.01	2.18	8.10
Nitrogen	0.35	0.36	1.34
Oxygen	1.70	1.73	1.04
Aluminum	0.25	0.26	0.49
Silicon	0.24	0.25	0.36
Chrome	0.27	0.27	0.28
Manganese	0.66	0.68	0.64
Iron	91.23	93.90	87.43
Nickel	0.35	0.37	0.32
	97.06	100.00	100.00

). The analysis reveals the presence of nitrogen due to the plasma nitriding process, which is absent in the blank sample.

3.1.3. EDS of Nitrided Steel with 6 h of Treatment

EDS analysis revealed a higher nitrogen content in the nitrided steel after 6 h of treatment compared to 4 h (Figure 3 and

Table 4. The results show the percentage of the elements present in the sample after 6 h of treatment.

Element	Mass [%]	Normalized Mass [%]	Atoms [%]
Carbon	2.21	2.31	8.83
Nitrogen	0.52	0.55	1.80
Oxygen	4.00	4.19	12.02
Aluminum	0.84	0.88	1.49
Silicon	0.29	0.30	0.49
Chrome	0.42	0.44	0.39
Manganese	0.66	0.69	0.56
Iron	85.37	89.36	73.42
Nickel	1.22	1.28	1.00
	95.53	100.00	100.00

). This indicates that a longer treatment time leads to increased nitrogen incorporation.

3.1.4. EDS of Nitrided Steel with 8 h of Treatment

EDS analysis showed that the nitrogen concentration in the nitrided steel did not increase further after extending the treatment time from 6 to 8 h (Figure 4 and

Table 5. The results show the percentage of the elements present in the sample after 8 h after treatment.

Element	Mass [%]	Normalized Mass [%]	Atoms [%]
Carbon	3.32	3.37	13.68
Nitrogen	0.52	0.55	1.91
Aluminum	0.24	0.25	0.45
Silicon	0.20	0.21	0.37
Chrome	0.25	0.26	0.24
Manganese	0.69	0.72	0.64
Iron	90.06	94.21	82.35
Nickel	0.42	0.44	0.36
	95.60	100.00	100.00

).

3.1.5. EDS of Nitrided Steel with 10 h of Treatment

EDS analysis revealed that the nitrided steel treated for 10 h exhibited the highest nitrogen concentration compared to other treatment durations (Figure 5 and

Table 6. The results show the percentage of the elements present in the sample after 10 h of treatment.

Element	Mass [%]	Normalized Mass [%]	Atoms [%]
Carbon	2.70	2.75	9.86
Nitrogen	0.56	0.57	1.75
Oxygen	7.00	7.14	19.19
Aluminum	0.15	0.15	0.24
Silicon	0.12	0.13	0.19
Chrome	0.28	0.28	0.23
Manganese	0.73	0.74	0.58
Iron	86.21	87.91	67.70
Nickel	0.32	0.33	0.24
	98.08	100.00	100.00

).

3.2. X Ray Diffraction (XRD)

Figure 6 shows the XRD patterns of the base X70 steel and the plasma-nitrided samples at different treatment times. The base material exhibits peaks corresponding to the BCC [110], [200], and [211] planes at 45°, 65°, and 82°, respectively, and the FCC [220] plane at 97°. This is consistent with reported XRD patterns for X70 and other HSLA steels [26,27,28,29,30,31]. Following plasma nitriding, new peaks appear at approximately 42° (γ’-Fe₄N and ε-Fe_2-3N), 48° (γ’-Fe₄N), 69° (γ-Fe and γ’-Fe₄N), 82°, and 88° (γ-Fe and ε-Fe₃N), in agreement with previous studies [22,39,42,43,55]. The intensity of these nitride peaks increases with nitriding time, consistent with observations in other nitrided alloys [43,44,45]. CrN peaks, observed in some other steels [39,40,42,44], were not detected in the present study.

3.3. Electrochemical Reactions

The material’s electrochemical behavior is represented for the following reactions. The third reaction represents the dissolution of the iron, the fourth and five are the cathodic reaction, with the primary reaction being the evolution of hydrogen, and the sixth and seven reactions are the species that contributed to the process of stress corrosion cracking (SCC). Reaction eight is due to corrosion products formed on the steel surface [16,27,29,31].

$$Fe\to {Fe}^{2+}+{2e}^{-}$$

(3)

$$O_2+{4H}^{+}+{4e}^{-}\to {2H}_{2}O$$

(4)

$${2H}^{+}+{2e}^{-}\to H_2$$

(5)

$${2HCO}_3^{-}+{2e}^{-}\to H_2+{2CO}_3^{-}$$

(6)

$$2H_{2}C{O}_3+{2e}^{-}\to H_2+{2HCO}_3^{-}$$

(7)

$${Fe}^{2+}+{2OH}^{-}\to Fe{\left(OH\right)}_2$$

(8)

3.4. Potentiodynamic Polarization Curves

Figure 7 presents the potentiodynamic polarization curves of the blank X70 steel and plasma-nitrided samples in a simulated soil solution. The blank sample exhibited an open-circuit potential (E_corr) of approximately −707 mV, consistent with reported values for X70 and similar steel grades in near-neutral solutions (−700 to −780 mV), which can vary depending on surface finish and pH [11,12,26]. The anodic branch of the blank sample showed active dissolution between −509 mV and −195 mV, with current fluctuations likely due to the formation and subsequent breakdown of corrosion products (oxides and oxyhydroxides), potentially leading to pitting corrosion [16,17,18,24,48]. The cathodic branch indicated a mass-transfer-controlled oxygen reduction process. Hydrogen evolution, the primary cathodic reaction for X70 steel in this environment, was consistent with an activation-controlled anodic reaction and was a significant factor in stress corrosion cracking (SCC) [26,27].

The nitrided samples showed a positive shift in E_corr compared to the blank samples, with the 8 h treatment exhibiting the noblest E_corr. However, the 10 h treatment showed a slightly more active E_corr than the 4, 6, and 8 h treatments. The cathodic branches of the nitrided samples revealed a decrease in the cathodic slope with increasing nitriding time, suggesting that the treatment hinders hydrogen reduction kinetics. This behavior, attributed to the formation of a nitrided layer containing Fe₃N, indicates improved corrosion resistance due to the plasma nitriding treatment [41,42].

Table 7. Electrochemical parameters obtained from the polarization curves of the blank and nitrided steel with different treatment times.

Condition	Ecorr (mV)	${\mathit{\beta }}_{\mathit{a}}$ (mV/Dec)	${\mathit{\beta }}_{\mathit{c}}$ (mV/Dec)	Icorr (mA/cm2)	$\mathit{\eta }\%$
Blank	−707	126	287	1.55 × 10−2	-
4 h	−512	188	210	7.07 × 10−3	54.35
6 h	−504	109	169	1.29 × 10−3	91.64
8 h	−440	79	99	8.98 × 10−4	94.20
10 h	−525	75	60	1.52 × 10−4	99.01

summarizes the electrochemical parameters, averaged from three measurements. Generally, the corrosion current density (i_corr) decreased with increasing treatment time. The efficiency (η%), calculated using Equation (1), also increased with treatment time. The sample nitrided for 10 h exhibited the highest corrosion resistance, with a corrosion rate two orders of magnitude lower than that of the blank material. This improvement in corrosion resistance with increased nitriding time, reflected in the decrease in i_corr, is consistent with previous findings for similar steels [36,40,41,46].

3.5. Influence of Time Treatment in Current Density and Efficiency

Figure 8 presents a graph of the current density and efficiency vs. plasma nitriding treatment time which was obtained using the values of the potentiodynamic polarization.

Table 8. Comparison of treatment time, concentration of nitrogen, and Icorr in the pieces studied at different treatment times.

Time (h)	Nitrogen (%wt)	Icorr (mA/cm2)
0	0	1.55 × 10−2
4	0.35	7.07 × 10−3
6	0.52	1.29 × 10−3
8	0.52	8.98 × 10−4
10	0.56	1.52 × 10−4

summarized the behavior of material under different treatments time, the percentage of nitrogen obtained and how it influences their current density. The lowest current density was achieved after 10 h of plasma nitriding treatment, demonstrating the effectiveness of longer treatment times in reducing current density and increasing the efficiency in electrochemical tests [36,41,43,46].

3.6. Electrochemical Impedance Spectroscopy

Figure 9a show the Nyquist diagrams of the blank X70 steel and the plasma-nitrided samples at different treatment times, measured in a simulated soil solution. The data represents the average of three measurements. Prior to the electrochemical impedance spectroscopy (EIS) measurements, the Nyquist diagram of the blank sample exhibits a decrease in the real impedance as a function of frequency, related to adsorption of corrosive species. The nitrided samples treated for 4, 6, and 8 h show a decrease in diameter in the capacitive loop, potentially indicating increased dissolution. In contrast, the 10 h nitrided sample exhibits a larger capacitive loop, suggesting the formation of a protective and adherent corrosion product layer, leading to increased corrosion resistance [36,40].

The Bode plots (Figure 9b) were analyzed based on frequency regions: low frequencies (<1–10 Hz) corresponding to surface processes, intermediate frequencies (10–1000 Hz) to charge transfer processes, and high frequencies (>1000 Hz) to electrolyte resistance [40,49]. The impedance modulus plot of the blank X70 steel shows a high-frequency plateau, indicating the absence of a significant surface layer. In contrast, the sample nitrited for 10 h exhibits a higher impedance modulus at high frequencies, suggesting the presence of a surface layer, likely composed of metal hydroxide-type compounds. At low and intermediate frequencies, all samples exhibit two linear regions in the log|Z| vs. log|f| plot, indicative of two-time constants and corresponding to the two capacitive loops observed in the Nyquist diagrams [48].

From the Bode diagram in its phase angle format, in the high frequency region, it was viewed that the phase angle tends to zero at frequencies greater than 1000 Hz. In the intermediate frequency region, the evolution of a first-time constant was observed whose maximum angle was 18° at 10 h of plasma nitriding treatment. The presence of the first time constant for the blank conditions was visualized in the low frequency region, with 4, 6, and 8 h of plasma nitriding treatment. The shift from left to right and the increase in the phase angle were associated with an increase in its capacitive properties and the formation of protective and adherent corrosion products [39,48].

Table 9. Electrochemical parameters used to fit the EIS data and C_dl for the blank and plasma-nitrided material at different treatment times. The average error was less than 5% for the simulated equivalent electrical circuits.

Conditions	Rs (Ω·cm2)	Rf (Ω·cm2)	CPEf		Rct (Ω·cm2)	CPEdl		Cdl
			Y0 (Ω−1·cm−2sn)	nf		Y0 (Ω−1·cm−2sn)	n
Blank	254.2	280.9	1.79 × 10−3	0.56	483	1.63 × 10−3	0.61	2.086 × 10−4
4 h	248.8	245.2	1.02 × 10−3	0.59	545	1.87 × 10−3	0.45	1.132 × 10−4
6 h	294.7	282.1	1.30 × 10−3	0.53	376	1.35 × 10−3	0.43	7.810 × 10−5
8 h	293.7	288.6	6.63 × 10−4	0.71	342	1.16 × 10−3	0.6	1.594 × 10−4
10 h	304.1	393.3	1.78 × 10−4	0.48	902	9.56 × 10−5	0.62	3.129 × 10−6

presents the electrochemical parameters extracted from the EIS data. The constant phase element (CPE) parameters, CPE_dl (double-layer capacitance), and CPE_f (film capacitance), provide insights into the surface characteristics. Higher CPE values generally indicate a more protective surface with a higher concentration of adsorbed protective species. Conversely, lower CPE values suggest increased dissolution and the presence of aggressive ions. The CPE values are related to the corresponding resistance values, R_ct (charge transfer resistance) and R_f (film resistance). For the 4 h treatment, R_f decreased, while for longer treatment times, R_f increased, likely due to the higher nitrogen concentration in the nitrided layer. R_ct decreased for the 6 and 8 h treatments but increased for the 4 and 10 h treatments, correlating with the observed corrosion rates [7]. The solution resistance (R_s) increased with treatment time, particularly for the 6, 8, and 10 h treatments, possibly due to the formation of reaction products in the solution resulting from the thermochemical treatment.

The CPE exponents, n and n_f, provide information about surface roughness. Values close to 0.5 suggest high surface roughness and a significant contribution from mass transport processes through pores in the corrosion product layer [7]. Conversely, values close to 1 indicate lower surface roughness and reduced metal dissolution. In this study, the lowest n and n_f values were observed for the 4 and 6 h treatments, indicating higher surface roughness, while the highest values were observed for the 10 h treatment, suggesting a smoother, less actively dissolving surface.

The double-layer capacitance (C_dl), calculated using Equation (2), decreased with increasing plasma nitriding time, consistent with the formation of a protective film. Similar trends in capacitance and resistance values have been reported for corrosion inhibitors [49,50,51].

Figure 10 presents the equivalent circuit used to model the electrochemical impedance spectroscopy (EIS) data of the blank and nitrided steel samples at different treatment times in the simulated soil solution. The circuit consists of the solution resistance (R_s), the charge transfer resistance (R_ct), in parallel with a constant phase element (CPE_dl) representing the double-layer capacitance, and the film resistance (R_f), in parallel with a second constant phase element (CPE_f) representing the capacitance of the nitride film. A CPE was used instead of a pure capacitor to account for surface inhomogeneities, such as roughness and non-uniform current distribution, which result in depressed semicircles in the Nyquist plots [30,38,39,48,49,50,51]. The impedance and phase angle are frequency-dependent, as described by the following equations:

$$Z_{CPE}={\displaystyle \frac{1}{Y({jw)}^n}}$$

(9)

where Y is the admittance of the system, a factor that combines properties related to the surface of electroactive species and is independent of the frequency, j is the imaginary unit ($\sqrt{-1}$), w is the angular frequency (2$\pi$f), f is the frequency, and n is related to the slope of log|Z| vs. log|f| [14]. If n is equal to 1, the CPE is an ideal capacitor, where Y is equal to the capacitance. However, if 0.5 < n < 1, then the CPE describes a distribution of dielectric relaxation times in the frequency range [13,38,39,48,49,50,51]. The calculated parameters obtained for the EIS data fitting for blank and nitrided steel at different treatment times are summarized in Table 9.

For the 10 h treatment, R_ct is much higher than other conditions, indicating enhanced corrosion resistance due to protective corrosion products, corroborated by polarization curves. When the n parameter is close to 1, indicating low metal surface roughness. All conditions had similar n values attributed to similar surface roughness.

3.7. Cross-Section Analysis

The nitrided layer thickness after 10 h of treatment was determined to be 678.75 µm from SEM micrography of the cross-section (Figure 11). This thickness measurement is supported by EDS, XRD, and electrochemical (CPP and EIS) analyses, which also indicate the formation of a nitrided layer. The observed relationship between treatment time and nitrided layer thickness is consistent with previous studies [4,38,40,44].

4. Conclusions

• Correct formation of the nitride layer and its behavior with varying treatment times was confirmed.
• Formation of the iron nitrides ε-Fe_2-3N and γ-Fe₄N indicates a clear formation of the nitrided layer. Increasing the plasma nitriding treatment time increases the intensity of the nitride peaks.
• Plasma nitriding significantly reduces corrosion density by up to three orders of magnitude compared to the blank, with 10 h of treatment providing the best results. The variation in Ecorr and Icorr were mainly due to the reaction kinetics in the formation of corrosion products on the nitrided surface.
• According to the current density graph against treatment time, the 10 h plasma nitriding treatment provides the best corrosion resistance and efficiency, making a long treatment unnecessary.