Features of Increasing the Wear Resistance of 90CrSi Tool Steel Surface under Various Electrophysical Parameters of Plasma Electrolytic Treatment

Abstract

The paper investigates the feasibility of plasma electrolytic treatment (PET) of 90CrSi tool steel to enhance hardness and wear resistance. The influence of electrophysical parameters of PET (polarity of the active electrode, chemical-thermal treatment, and polishing modes) on the composition, structure, morphology, and tribological properties of the surface was studied. Tribological tests were carried out under dry friction conditions according to the shaft-bushing scheme with fixation of the friction coefficient and temperature in the friction contact zone, measurements of surface microgeometry parameters, morphological analysis of friction tracks, and weight wear. The formation of a surface hardened to 1110–1120 HV due to the formation of quenched martensite is shown. Features of nitrogen diffusion during anodic PET and cathodic PET were revealed, and diffusion coefficients were calculated. The wear resistance of the surface of 90CrSi steel increased by 5–9 times after anodic PET followed by polishing, by 16 times after cathodic PET, and up to 32 times after subsequent polishing. It is shown that in all cases, the violation of frictional bonds occurs through the plastic displacement of the material, and the wear mechanism is fatigue wear during dry friction and plastic contact.

1. Introduction

Tool steels are materials with increased performance characteristics. However, to address specific or niche problems, they often undergo additional processing to improve certain properties. This paper discusses the application of plasma electrolytic treatment (PET) technology to enhance the hardness and wear resistance of surfaces of products made from 90CrSi tool alloy steel (equivalent to DIN 150Cr14 and 90CrSi5). These steels are utilized for manufacturing critical components that require high wear resistance; fatigue strength under bending, torsion, and contact loading; and elastic properties. Common applications include drills, reamers, taps, dies, and cutters. To improve the surface properties of these materials, various techniques have been employed, including laser treatment (hardening) [1,2,3], electrical discharge machining [4] and heat treatment [5]. These methods enhance strength properties, increase wear resistance, and reduce surface roughness. If we consider technologies aimed at changing the structure of surface layers, laser hardening allows for an increase of the wear resistance of tool steel under dry friction by two times [2]. The increase in hardness after laser hardening of CrWMn steel, which is close to 90CrSi steel in many characteristics, can reach 1017 HV [3]. Heat treatment has shown a positive result in the processing of C80W1 and 90CrSi5 tool steels, but cryogenic treatment applied after hardening reduces wear resistance due to a decrease in the proportion of residual austenite [5].

PET addresses similar challenges and is designed to enhance the complex surface properties of processed products made from steels and non-ferrous alloys. Several types of PET include plasma electrolytic oxidation, aimed at forming protective ceramic-like coatings [6,7] and utilized for polishing metal surfaces [8,9,10], and chemical-thermal treatment. Plasma electrolytic chemical-thermal treatment consists of heating the product to high temperatures and diffusion saturation of the surface with atoms of light elements: carburizing [11], nitriding [12,13], carbonitriding [14,15], nitrocarburizing [16], boriding, and other combinations of elements [17]. This technology has been applied to enhance the performance properties of titanium alloys [18,19], serving as an alternative to plasma electrolytic oxidation [20,21]. However, the primary purpose of PET is the processing of steels, both carbon [22,23] and alloyed types, such as 42CrMo [24,25], 40Cr [26], 30CrMnSiA [27], AISI H13 [28], T8 [29], 38CrMoAl [30], 20CrMnTi [31], 30CrMnSi [32], and others.

As a result of such treatment, changes occur in the structure, phase, and elemental composition of the surface layers. For example, after plasma electrolytic carburizing of H13 steel, Fe₃C, FeC, and Fe_1.88C_0.12 carbides were detected [28]. Fe₅C₂ carbide was formed after carburizing of 30CrMnSiA steel [33]. A feature of PET in aqueous solutions is the formation of oxides. FeO, Fe₃O₄, Fe₂O₃, and Fe₃C phases were found in the oxide layer after carburizing of T8 steel [34]. FeN_0.076 iron nitride and Fe₃O₄ oxide were detected after the nitriding of 316L steel [35]. The main phase during nitrocarburizing of 316L steel is nitrogenous austenite [36]. When processing alloy steels, oxides of NiFe₂O₄ and FeCr₂O₄ [37], Fe₂O₃, Fe₃O₄ and Cr₂O₄ [38], Fe(Fe,Cr)₂O₄ [39,40], nitrides of Fe₃N [38], CrN, and Cr₂N, carbides of Cr₃C₂ and Cr₇C₃ [40], and silicon dioxide [38] can be formed.

The formation of a modified surface on steel products contributes to enhanced performance properties. For instance, the friction coefficient and wear rate decrease after nitrocarburizing 30CrMnSiA steel. This improvement is attributed to the formation of finely dispersed secondary phases of FeN, (Fe,Cr)₃C, and Me₂₃(C,N)₆, along with an increase in hardness by 2.5–3.3 times [27]. Nitriding of 34CrNiMo alloy steel reduces the intensity of adhesive wear of steel by 35% during dry friction of a corundum ball compared to untreated steel [41]. A reduction in adhesive wear is also observed after nitriding 38CrMoAl steel due to the formation of a needle-shaped microstructure containing nitrides, small cementite grains, and martensite [30].

Plasma electrolytic chemical-thermal treatment, which is based on an electrochemical process, is categorized by the polarity of the processed product-electrode into anodic [42] and cathodic [34] treatments. This paper discusses the use of both PET options for processing 90CrSi alloy tool steel. As a diffusion saturation method, nitriding was performed, followed by hardening from the saturation temperature. Additionally, after plasma electrolytic nitriding (PEN), plasma electrolytic polishing (PEP) was conducted to address issues that arose during the processing of carbon steels: removing the outer loose oxide layer after anodic PET and smoothing microroughness on the surface after cathodic PET. Previous studies have demonstrated the positive effect of combining chemical-thermal treatment and polishing in plasma electrolysis using carbon steels, as an example [43].

This work examines the possibility of modifying the surface layers of 90CrSi tool steel through plasma electrolytic nitriding (PEN) followed by plasma electrolytic polishing (PEP) to enhance the performance characteristics of products made from this material. It also investigates the influence of the electrophysical parameters of PET on the structure, morphology, and tribological properties of the steel surface.

2. Materials and Methods

2.1. Materials

In this study, cylindrical samples of 90CrSi tool steel with a diameter of 10 mm and a height of 15 mm (

Table 1. Chemical composition of 9CrSi steel samples.

Element	C	Cr	Si	Mn	Ni	S	P	Mo	W	V	Ti	Cu	Fe
wt.%	0.85–0.95	0.95–1.25	1.2–1.6	0.3–0.6	≤0.4	≤0.03	≤0.03	≤0.2	≤0.2	≤0.15	≤0.03	≤0.3	balance

) were used. Prior to treatment, the samples were polished using sandpaper to achieve a surface roughness R_a of 1.0 ± 0.1 μm and then washed with acetone in an ultrasonic bath.

2.2. Processing

PEN and PEP were conducted in an electrolyzer equipped with forced electrolyte circulation and a heat exchanger to maintain a constant temperature (Figure 1). PEN was performed under both anodic and cathodic polarities. In cathodic nitriding (CPEN), the sample was connected to the negative pole of a DC source, while the working chamber was connected to the positive pole. Conversely, in anodic nitriding (APEN), the connections were reversed. PEP was consistently performed with the anodic polarity of the sample. The electrolyte was cyclically supplied from below through a nozzle, flowing longitudinally around the sample before draining into a pan, from where it was pumped back into the heat exchanger. The electrolyte temperature was monitored by a thermocouple placed at the bottom of the electrolyzer, maintaining (30 ± 2) °C during PEN and (70 ± 2) °C during PEP.

PEN was conducted in a solution of 5% ammonium chloride and 5% ammonia at 750 °C for 5 min (APEN) and 10 min (CPEN). A notable feature of APEN is the formation of a continuous vapor-gas envelope around the anode sample, accompanied by a characteristic glow. In contrast, during CPEN, strong microdischarges exhibiting a blue glow were visually observed on the surface (Figure 2). After PEN, the samples were quenched in an electrolyte from the saturation temperature by turning off the voltage.

Subsequent PEP was carried out at a voltage of 325 V for 1 and 2 min in a solution of 3% ammonium chloride with 5% glycerol and a solution of 5% ammonium sulfate. During this process, a transition from film boiling around the anode sample (as seen in APEN) to nucleate boiling was observed, evidenced by the “pulling” of the electrolyte onto the sample, along with accompanying microdischarges (Figure 2).

Voltage and current were measured using a DP6-DV voltmeter and a DP6-DA ammeter. The sample temperature was monitored with a MY-K2 thermocouple connected to an APPA109N multimeter, which has an accuracy of 3% within the temperature range of 400–1000 °C. The thermocouple was positioned in the sample hole at a distance of 2 mm from the end.

2.3. Study of Morphology, Structure, and Element and Phase Composition

X-ray diffraction (XRD) analysis was conducted to determine the phase composition of the samples. The XRD patterns were obtained using a PANalytical Empyrean X-ray diffractometer (Malvern Panalytical, Malvern, UK) with CoKα radiation in theta-2theta mode, employing a step size of 0.026° and a scanning rate of 4.5°/min. Phase composition analysis utilized the PANalytical High Score Plus software (v. 2.0.0) [44] and the ICCD PDF-2 and COD databases [45].

Scanning electron microscope (SEM) was performed with a Tescan Vega 3 (Tescan, Brno, Czech Republic) using secondary electrons (SE) detectors for topological contrast, and backscattered electrons (BSE) detectors for compositional contrast was used to study PET samples. The elemental analysis was conducted using the X-Act energy dispersive analysis (EDX) detector (Oxford Instruments, Abingdon, UK) at an accelerating voltage of 20 kV to identify the characteristic lines of the main elements.

Surface morphology and microstructure of the steel samples’ cross-section were examined using a Micromed MET optical metallographic microscope (Micromed, St. Petersburg, Russia) with digital image visualization.

2.4. The Microhardness Measurement

The microhardness of the cross-sections of the treatment sample was measured using a Vickers microhardness tester (Falcon 503, Innovatest Europe BV, Maastricht, The Netherlands) with a 0.1 N load. The average microhardness value was calculated based on five measurements.

2.5. Surface Roughness and Weight of Samples Measurement

The surface roughness was measured using a TR-200 profilometer (Beijing TIME High Technology Ltd., Beijing, China). The average value of the roughness indicators was calculated based on ten measurements. Weight changes of the samples were determined using a Citizen CY224C electronic analytical balance (ACZET, Mumbai, India) with an accuracy of ±0.0001 g after washing the samples in distilled water to remove salt traces and subsequently drying them.

2.6. Study of Tribological Properties

Tribological tests were carried out on the side surface of a cylindrical sample according to the shaft-bushing scheme (patent RU 213 413 U1) (Figure 3). The counterbody was constructed of hardened carbon tool steel. The friction path covered was 1 km, with a sliding speed of 1.555 m/s and a load of 10 N.

The friction track profile and surface microtopology parameters were measured using a TR-200 profilometer (Beijing TIME High Technology Ltd., Beijing, China). The friction contact temperature was recorded at the friction track directly at the exit of the contact zone using a digital infrared thermometer MLX90614 (Melexis Electronic Technology, Shanghai, China).

2.7. Determination of the Microgeometry Characteristics and Type of Violation of Friction Bonds during Friction

The characteristics of friction and wear largely depend on the properties of frictional contact, realized in individual areas forming the actual contact area A_r. The friction force is proportional to this contact area, while wear is influenced by its size. The diameter of the contact areas affects the duration of their interaction during friction. The destruction of surface layers and the formation of adhesive bonds depend on the actual pressure P_r on the frictional contact. These contact properties significantly impact the rigidity of butt joints in machine parts.

The type of contact (elastic or plastic) and wear mechanism can be determined by quantitative assessment of the relative penetration of surface irregularities. The relative penetration h/r is the ratio of absolute penetration h to the radius of a single roughness r.

After PET and subsequent tribological tests, roughness protrusions are formed on the friction tracks, which are modeled by bodies of double curvature:

$$r={\displaystyle \frac{9R_a^2{S}_m^2}{128{\left(5.5R_a-R_p\right)}^3}},$$

(1)

where R_a is the arithmetic mean deviation of the profile; R_p is the smoothing height or distance from the line of protrusions to the center line within the base length; and S_m is the average pitch of irregularities.

To determine the absolute penetration of friction surfaces, it is necessary to determine the distribution function of roughness protrusions on the surface along their height. The curve of the supporting surface, representing the ratio of the area of actual contact between the sample and the counterbody to the contour area, can be expressed by the Demkin function:

$$\eta =l_m{\left({\displaystyle \frac{y}{R_p}}\right)}^{\nu }=b{\left({\displaystyle \frac{y}{R_{\max }}}\right)}^{\nu }=b{\epsilon }_{\max }^{\nu }={\displaystyle \frac{A_r}{A_c}}={\displaystyle \frac{P_c}{P_r}}={\displaystyle \frac{n_r}{n_c}},$$

(2)

where y is the level of the profile section, measured from the line of protrusions; R_max is the maximum height of irregularities; ε = y/R_max is the relative profile height; A_r is the actual contact area; A_c is the contour contact area, P_r is the average actual pressure on the friction contact; P_c is the contour pressure; n_r is the number of contacting protrusions on the actual contact area; and n_c is the number of roughness protrusions on the contour area.

Expression (2) represents the statistical law of distribution of the modified sample material along the height of the rough layer; it includes the parameters ν and b, determined in direct measurements with a profilometer on the friction tracks (Figure 4) [46]:

$$\nu =2l_m\left({\displaystyle \frac{R_p}{R_a}}\right)-1,$$

(3)

$$b=l_m{\left({\displaystyle \frac{R_{\max }}{R_p}}\right)}^{\nu },$$

(4)

l_m is the relative reference length of the profile at the level of the centerline:

$$l_m={\displaystyle \frac{{\displaystyle \sum _1^n\Delta l_i}}{l}},$$

(5)

where l is the base length, and Δl_i is the length of the segments cut off by the centerline in the profile. The parameter l_m is determined by direct measurements with a profilometer on the friction paths of not only the sample under study but also the counterbody. These quantities describe the surface of a single rough body and can be used to describe the contact of this body with an idealized smooth surface at a given time.

To calculate the actual pressure P_r at the tops of microcontacts, it is necessary to determine the type of deformation at the frictional contact: elastic or plastic. To assess the nature of deformations, the Greenwood–Williamson criterion is used:

$$K_p={\displaystyle \frac{\mathsf{\Theta }}{HB}}\sqrt{{\displaystyle \frac{R_p}{r}}},$$

(6)

where Θ is the reduced modulus of elasticity:

$$\mathsf{\Theta }={\left({\displaystyle \frac{1-{\mu }_1^2}{E_1}}+{\displaystyle \frac{1-{\mu }_2^2}{E_2}}\right)}^{-1},$$

(7)

and μ_i and E_i are Poisson’s ratios and elastic moduli of interacting bodies, and r is the radius of microroughness (1).

The dimensionless plasticity index K_p describes the deformation properties of a rough surface. According to the Greenwood criterion, the deformations of asperities in contact with a flat surface will be predominantly plastic if K_p is greater than unity.

For elastic contact, the deformation of individual protrusions can be calculated according to the classical Hertzian contact problem. With a plastic contact, the average stress on the contact is numerically equal to the microhardness P_r ≈ HB.

The actual contact area is determined by the ratio of normal load to actual pressure:

$$A_r={\displaystyle \frac{N}{P_r}}.$$

(8)

By relating the equation that approximates the reference curve in the form (2) to the ratio of the contour and actual pressures at the level y = h, the following expression for the absolute convergence of the contacting surfaces is obtained:

$$h=R_{\max }{\left({\displaystyle \frac{P_c}{b\cdot P_r}}\right)}^{{\displaystyle \frac{1}{\nu }}}=R_{\max }{\left({\displaystyle \frac{N}{b\cdot P_r}}\right)}^{{\displaystyle \frac{1}{\nu }}}.$$

(9)

If, during plastic contact, the radius of a single roughness protrusion is modeled by a body of double curvature (1), then the following expression for relative penetration will be obtained:

$${\displaystyle \frac{h}{r}}={\displaystyle \frac{128R_{\max }{\left(5.5R_a-R_p\right)}^3}{9R_a^2{S}_m^2}}\cdot {\left({\displaystyle \frac{N}{b\cdot HB}}\right)}^{{\displaystyle \frac{1}{\nu }}}.$$

(10)

The obtained characteristics can be utilized to distinguish between types of contact: elastic, plastic, and microcutting. The characteristic can also be applied to analyze the results of experimental data. Bond failure due to elastic displacement of metal is observed during dry friction on steels with values of h/r < 0.01. In this case, failure typically occurs due to fatigue phenomena in the near-surface layer after a large number of test cycles. Plastic deformation of the metal is observed at h/r > 0.1 and manifested on the contact surface as residual deformation. At h/r > 0.1, destruction of surfaces begins due to microcutting and the formation of microchips [47].

For a comprehensive assessment of the quality of the sample surface after PET, the Kragelsky-Kombalov criterion is calculated:

$$\Delta ={\displaystyle \frac{R_{\max }}{r{b}^{\frac{1}{\nu }}}}={\left({\displaystyle \frac{100}{l_m}}\right)}^{{\displaystyle \frac{1}{\nu }}}\cdot \left({\displaystyle \frac{R_p}{r}}\right).$$

(11)

Average roughness R_a is considered the most common surface characteristic. However, significant variations in pitch characteristics, vertex radii, and profile height may exist among different profiles. These differences considerably affect the friction and wear of surfaces. For an objective assessment of roughness, a complex parameter that accounts for not only the average roughness but also other height characteristics must be used. The bearing capacity of a surface in a tribological contact is dependent on the value of Δ: the smaller it is, the more efficiently the surface copes with the load [48].

3. Results and Discussion

3.1. XRD Analysis of Samples Surface

As a result of XRD analysis, the phase composition of the surface of 90CrSi steel after APEN and CPEN followed by PEP under various conditions was determined (Figure 5 and Figure 6). Regardless of the processing method, phases such as martensite and retained austenite, resulting from hardening, along with iron carbide (cementite) as a component of hypereutectoid steel’s phase composition, were identified on the surface. The formation of iron oxides of various compositions on the surface occurred as a result of high-temperature oxidation in aqueous electrolyte vapor at diffusion saturation (sample temperature was 750 °C):

Fe + x/2H₂O → FeO_x/2 + xH⁺ + xe⁻

(12)

Oxides remain on the surface after PEP. The considerable thickness of the oxide layer hinders the identification of the phases underneath. Iron nitrides were detected in sufficient quantities following APEN and subsequent PEP for 1 min in sulfate electrolyte. This likely occurred after the partial removal of the oxide layer during PEP. Increasing the PEP duration to 2 min did not result in the detection of nitrides, probably due to the electrochemical dissolution of the surface layer of steel containing nitrides:

Fe → Fe²⁺ + 2e⁻

(13)

Iron nitrides were not detected after PEP in a solution of ammonium chloride and glycerol due to intense anodic dissolution. In contrast, during PEP in sulfate electrolyte, the anodic dissolution of iron was restrained by surface passivation:

2H₂O – 4e⁻ → 4H⁺ + O₂

(14)

xFe + y/2O₂ → Fe_xO_y

(15)

After CPEN, the phase composition of the steel surface is similar to that after APEN. Notably, a higher intensity of martensite phases and a lower intensity of iron oxide phases were observed. It is likely that the oxide layer is partially destroyed during cathodic treatment under the influence of microdischarges. Iron nitrides were not detected; thus, the nitrogen potential (nitrogen concentration on the metal surface) during cathodic treatment is lower compared to APEN.

3.2. SEM and EDX Analysis of Samples Cross-Section

According to SEM analysis of the cross-section, an oxide layer is revealed on the surface of the samples after APEN and CPEN, exhibiting morphological inhomogeneities such as cracks, chips, and various irregularities (Figure 7 and Figure 8). EDX analysis of the oxide layer showed an increased concentration of alloying components (Si and Cr) with a non-stoichiometric ratio of iron and oxygen concentrations (Figure 7). This allows the oxide layer to be considered a mixture of iron oxides and a solid solution of oxygen subtraction in the original matrix. A small amount of oxygen is detected under the oxide layer at a depth of up to 5 µm due to diffusion (Figure 7). The concentrations of alloying components in the material’s structure beneath the oxide layer have the initial values (Figure 7 and Figure 8).

The concentration of nitrogen in the surface layer resulting from diffusion during PEN is particularly significant. The depth of nitriding for APEN and CPEN is approximately the same, not exceeding 25 µm (Figure 7 and Figure 8). The nitrogen concentration after APEN is nearly three times higher than after CPEN. At the same nitriding temperature, nitrogen diffusion occurs more intensely with the anodic polarity of the sample. The higher nitrogen potential during the anodic process leads to the formation of iron nitrides, confirmed by XRD analysis (Figure 5d). Conversely, at low nitrogen concentrations, nitrides are not formed during the cathodic process (Figure 6). It can be assumed that the higher nitrogen potential with APEN results from the anodic dissolution of the base material. According to EDX analysis, iron, and alloying additives comprise about 2/3 of the surface oxide layer, and upon anodic dissolution, they migrate outward, creating a significant number of vacancies in the crystal lattice and pores through which nitrogen diffuses from the surface through the porous oxide layer into the steel. In the case of CPEN, anodic dissolution does not occur, resulting in a denser oxide layer that inhibits nitrogen penetration from the surface through the oxide layer into the steel.

Subsequent PEP leads to a change in the morphology of the surface layer. After APEN, PEP in a solution of ammonium chloride with glycerol removes almost the entire oxide layer, along with the etching of some surface areas (Figure 7). After polishing in an ammonium sulfate solution for 1 min, a morphologically homogeneous oxide layer is formed, which is associated with passivation of the surface (reaction Equations (14) and (15)). This occurs simultaneously with the removal of irregularities and weak areas of the oxide layer. An increase in the duration of PEP results in the etching of the steel substrate, similar to the effect observed when using a solution of ammonium chloride with glycerol.

After CPEN followed by PEP, a more uniform removal of the oxide layer occurs, likely due to the formation of a denser structure of the oxide layer with the cathodic polarity of treatment (Figure 8). This assumption is supported by the absence of anodic dissolution of metals during CPEN. In this case, ion migration does not occur, resulting in an oxide layer with greater density and strength and fewer defects compared to that formed during APEN.

3.3. Calculation of Nitrogen Diffusion Coefficients

It is known that nitrogen can form interstitial solid solutions in both austenite and ferrite. The formation of such solutions is based on the fact that the radius of the dissolved atoms is only slightly larger than the vacant spaces between atoms in the base metal lattice. According to Hagg’s rule, an interstitial solid solution can form in a densely packed atomic lattice with a coordination number of 8 or 12 if the radius of the atoms of the interstitial element is no more than 0.59 of the actual radius of the base atoms. The actual radius of an iron atom is 12.8 nm, while that of a nitrogen atom is 7.1 nm. The ratio of the actual radius of nitrogen to that of iron is 0.55, indicating a high solubility of nitrogen in the Fe–N system.

Unlike substitutional solid solutions, where different diffusion mechanisms are possible, the diffusion mechanism of interstitial solid solutions is uniform. Nitrogen consistently occupies part of the interatomic octahedral sites in the cubic face-centered lattice of austenite. During the diffusion process, a particle with energy higher than the activation energy moves into an adjacent vacant cavity.

Unlike substitutional solid solutions, in interstitial ones, diffusion is assumed to involve only one diffusing element. This scenario is described by the mathematical model below.

The differential equation of diffusion in a semi-infinite body is solved:

$${\displaystyle \frac{\partial C(x,\tau )}{\partial \tau }}=D{\displaystyle \frac{{\partial }^{2}C(x,\tau )}{\partial x^2}},$$

(16)

where C is the concentration of the diffusing element (at. %); x is the horizontal coordinate (m); τ is the time (s); and D is the diffusion coefficient (m²/s). Under the following boundary conditions:

$$\begin{array}{l}C(x,0)=C^0\\ C(0,\tau )=C^S\\ C(\infty ,\tau )=C^0\end{array},$$

(17)

where C⁰ is the initial concentration of the element in the material, and C^S is the concentration of the element on the surface of the material.

The solution to Equation (16) under the indicated conditions is known:

$$C(x,\tau )=C^S-(C^S-C^0)erf{\displaystyle \frac{x}{2\sqrt{D\tau }}}.$$

(18)

To determine the unknown diffusion coefficients D_j of nitrogen after nitriding, the least squares method is used.

The coefficients D_j are selected to ensure the smallest sum of squared deviations between the theoretical concentration values according to Equation (16) and the experimental values. Thus, the following condition must be met:

$$F(D_j)={\displaystyle \sum _{i=1}^n{\left[C_i^S-(C_i^S-C_i^0)erf\left(\frac{x_i}{2\sqrt{D_j\tau }}\right)-C_i(x,\tau )\right]}^2=\min }.$$

(19)

To determine the minimum value of the functional F(D_j), it is necessary to equate its partial derivative with respect to the variable D_j to zero.

The result is:

$${\displaystyle \frac{\partial F(D_j)}{\partial D_j}}=2{\displaystyle \sum _{i=1}^n\begin{array}{l}\left[C_i^S-(C_i^S-C_i^0)erf\left(\frac{x_i}{2\sqrt{D_j\tau }}\right)-C_i(x,\tau )\right]\\ \times \exp \left(-\frac{x_i^2}{4D_j\tau }\right)\cdot (C_i^S-C_i^0)\cdot \frac{x_i}{2\sqrt{D_j^3\tau }}.\end{array}}$$

(20)

From this, an equation is obtained for determining the unknown nitrogen diffusion coefficients D_j:

$${\displaystyle \sum _{i=1}^n\left(C_i^S-C_i(x,\tau )\right)\cdot x_i\cdot \exp \left(-\frac{x_i^2}{4D_j\tau }\right)-(C_i^S-C_i^0){\displaystyle \sum _{i=1}^n{x}_i\exp \left(-\frac{x_i^2}{4D_j\tau }\right)erf\left(\frac{x_i}{2\sqrt{D_j\tau }}\right)}}=0.$$

(21)

Based on the results of a graphical solution of Equation (10), based on EDX analysis data on the nitrogen content in the surface layer of steel (Figure 7 and Figure 8), the values of nitrogen diffusion coefficients for APEN and CPEN of 90CrSi steel were obtained (

Table 2. Values of nitrogen diffusion coefficients in 90CrSi steel.

APEN	CPEN
2.50 μm2/s	0.30 μm2/s

).

3.4. Metallographic Analysis and Microhardness of Samples Cross-Section

Metallographic analysis after etching a thin section in a 3% solution of nitric acid in ethyl alcohol (nital) revealed a correlation between the phase and elemental composition of the surface layers and their structure (Figure 9). In the case of APEN, a dark outer nitride layer with reduced microhardness was observed beneath the oxide layer (Figure 10a), which, according to XRD analysis, contains an increased content of austenite and nitrides (Figure 5). After CPEN, no nitride layer is observed, and the microhardness at the edge decreases slightly (Figure 10b). The maximum value of microhardness in both cases is achieved not at the boundary of the metal substrate with the oxide layer but slightly deeper from the surface (after APEN, this occurred along the boundary of the nitride layer) and amounts to 1110–1120 HV. The results of microhardness measurements indicated the preservation of the strengthened layer after PEP for all considered processing cases (Figure 10).

3.5. Surface Morphology and Roughness

The surface morphology of PEN is determined by the polarity of the processed sample and the process’s physicochemical conditions. As a result of APEN, the risks of mechanical grinding were removed from the surface due to anodic dissolution (Figure 11a) with the formation of a homogeneous porous surface of the oxide layer (Figure 11b). Electrochemical dissolution of irregularities leads to a decrease in roughness by 2.7 times in R_a (

Table 3. Values for the loss of the samples’ weight during treatment Δm, the surface roughness R_a and R_z, the average friction coefficient μ for the last 100 m of the distance, the average temperature in the friction contact zone T_fr for the last 100 m of the distance, the loss of samples weight during friction Δm_fr, the relative introduction of surface roughness h/r and the Kragelsky-Kombalov criterion Δ samples from 90CrSi steel before and after APEN and subsequent PEP for 1 and 2 min in a solution of ammonium chloride with the addition of glycerol and in a solution of ammonium sulfate.

Electrolyte for PEP	PEP Time (min)	Δm (mg)	Ra (μm)	Rz (μm)	μ	Tfr (°C)	Δmfr (mg)	h/r	Δ
Untreated			1.00 ± 0.10	7.22 ± 0.51	0.361	51	28.40	0.080 ± 0.002	0.30 ± 0.01
APEN		32.5	0.37 ± 0.07	2.12 ± 0.43	0.464	102	3.16	0.071 ± 0.001	0.42 ± 0.01
NH4Cl + C3H8O3	1	25.8	3.19 ± 0.56	20.20 ± 3.03	0.532	111	5.29	0.072 ± 0.001	0.44 ± 0.01
	2	67.4	3.84 ± 0.98	24.21 ± 4.91	0.506	104	5.07	0.078 ± 0.002	0.47 ± 0.01
(NH4)2SO4	1	18.9	0.38 ± 0.05	1.96 ± 0.13	0.511	125	5.02	0.072 ± 0.001	0.48 ± 0.01
	2	34.4	0.54 ± 0.24	5.42 ± 3.15	0.426	106	3.11	0.075 ± 0.002	0.45 ± 0.01

).

After CPEN, a high-relief oxide layer is formed under the influence of electric discharges, resulting in a slight increase in roughness in R_a and a decrease in R_z (

Table 4. Values for the loss of the samples’ weight during treatment Δm, the surface roughness R_a and R_z, the average friction coefficient μ for the last 100 m of the distance, the average temperature in the friction contact zone T_fr for the last 100 m of the distance, the loss of samples weight during friction Δm_fr, the relative introduction of surface roughness h/r and the Kragelsky-Kombalov criterion Δ samples from 90CrSi steel before and after CPEN and subsequent PEP for 1 and 2 min in a solution of ammonium chloride with the addition of glycerol and in a solution of ammonium sulfate.

Electrolyte for PEP	PEP Time (min)	Δm (mg)	Ra (μm)	Rz (μm)	μ	Tfr (°C)	Δmfr (mg)	h/r	Δ
Untreated			1.00 ± 0.10	7.22 ± 0.51	0.361	51	28.4	0.080 ± 0.002	0.30 ± 0.01
CPEN		101	1.16 ± 0.11	6.94 ± 0.57	0.300	99	1.8	0.048 ± 0.001	0.21 ± 0.04
NH4Cl + C3H8O3	1	54.1	1.13 ± 0.25	6.92 ± 0.91	0.309	92	1.6	0.048 ± 0.001	0.21 ± 0.04
	2	82.1	0.68 ± 0.07	3.45 ± 0.26	0.263	79	1.8	0.046 ± 0.001	0.22 ± 0.04
(NH4)2SO4	1	47.6	1.67 ± 0.32	9.80 ± 1.78	0.169	130	0.9	0.041 ± 0.001	0.20 ± 0.04
	2	93.1	0.68 ± 0.16	3.56 ± 0.74	0.244	78	1.4	0.045 ± 0.001	0.22 ± 0.04

), which is associated with the removal of large irregularities by discharges. This is reflected in the formation of a developed morphology of the surface oxide layer, on which traces of chips and melting from the action of discharges are visible (Figure 12b).

Subsequent PEP of nitrided samples leads to changes in surface morphology and roughness. Significant differences in roughness measurements were observed based on the electrolyte compositions and the initial roughness. After APEN, polishing in a chloride-glycerol electrolyte results in an increase in roughness by an order of magnitude, while PEP in a sulfate electrolyte allows for the maintenance of the low roughness value achieved after APEN (Table 3). These results are also reflected in the surface morphology. After PEP in a chloride-glycerol electrolyte, the formation of a non-uniformly developed surface is observed due to uneven etching with the release of areas of the metal substrate (Figure 11c,d). After PEP in the sulfate electrolyte, the surface profile is leveled (Figure 11e), followed by a brightening of the surface due to the uniform removal of the oxide layer (Figure 11f). With both electrolytes, increasing PEP duration results in increased roughness.

After CPEN, polishing in both electrolytes for 2 min results in a reduction of roughness by 1.7 times in R_a and two times in R_z (Table 4). In this case, the surface morphology becomes more uniform. After PEP in a sulfate electrolyte, a uniform oxide layer is observed (Figure 12e,f), and after PEP in a chloride-glycerol electrolyte, the surface is etched with visualization of the metal substrate (Figure 12c,d). These results are fully consistent with SEM analysis of cross-sections of machined surfaces (Section 3.2).

3.6. Tribological Properties of Samples

Tribological tests of samples made of 90CrSi steel after APEN, as well as subsequent PEP, resulted in an increase in the coefficient of friction by 1.3–1.5 times and the temperature in the friction contact zone by two or more times, while a decrease in weight wear by 5–9 times was observed (Table 3). The running-in period of friction pairs ranged from 200 to 400 m of path (Figure 13a). The lowest friction coefficient and weight wear were revealed after PEP of a nitrided sample in a sulfate electrolyte for 2 min. Calculation of surface microgeometry parameters showed an increase in the value of the Kragelsky-Kombalov criterion of the treated samples by approximately 1.4–1.5 times. Consequently, the bearing capacity of the rough profile for these samples decreased, indicating that the maximum friction load, beyond which plastic flow begins, decreased.

After CPEN and subsequent PEP, other features were identified (Table 4). A decrease in the friction coefficient was observed after nitriding by 1.2 times, and a further reduction of more than two times was noted after polishing in a sulfate electrolyte for 1 min. Similarly, weight wear was reduced by a factor of 16 after CPEN and by a factor of 32 after PEP under the existing conditions. The running-in period of friction pairs, in contrast to APEN, has been reduced to 100–200 m (Figure 13b). In contrast to APEN, cathodic treatment and subsequent PEP led to a decrease in the value of the Kragelsky-Kombalov criterion by 1.4–1.6 times, indicating that the surface has greater wear resistance than the untreated sample.

On the friction tracks of samples after APEN and subsequent PEP (Figure 14c–f), the oxide layers shielding the metal surface are practically invisible. Each passage of a microprotrusion during sliding is characterized by residual deformation of the surface. After the number of cycles implemented in the tests, stripes in the sliding direction without sharp boundaries are visible in the images of the friction tracks. In light areas lacking oxide films, the formation of plastically deformed microvolumes of metal occurs, resulting in the formation of pre-fracture zones. These zones, with repeated exposure, lead to the formation of wear particles and separation of the material. The deformation extends to a greater depth than CPEN. The relative penetration of the sample and counterbody surfaces in friction tests after APEN can be up to two times greater in comparison with samples after CPEN (Table 3 and Table 4). An increase in R_z after APEN and PEP in a chloride electrolyte with glycerol leads to a decrease in the radii of curvature of the tips of microroughnesses. This contributes to deeper penetration into the volume of the deformed material and an increase in both the deformation component of the friction coefficient and the total friction coefficient (Table 3, Figure 13a).

A significant amount of oxides is observed on the friction tracks of the samples after CPEN (Figure 15b) and CPEN followed by PEP (Figure 15c–f). The shear strength of the surface oxide layer is less than that of the modified iron base material underneath. This results in the oxide layer being removed from the surface of a nitrided sample more easily than the underlying modified layer during friction. The brittle oxide layer (layer 1 in Figure 9b), when destroyed during sliding, blocks the destruction in the modified layer (layer 3 in Figure 9b) and, thus, plays the role of a lubricant during friction. An outer layer of reduced strength is necessary for external friction. The lower the strength of the outer oxide layer, the further the threshold of external friction moves away. Under such conditions, all deformations are concentrated in a thin surface oxide layer, which is evident from the small depth of relative penetration of the sample into the surface of the counterbody (Table 4). Relative embedding after CPEN is reduced by 1.7-fold compared to the untreated control. CPEN, followed by PEP for 1 min in a sulfate electrolyte, reduces the absolute penetration of rubbing surfaces by two times.

Relative penetration determines the deformation (mechanical) component of the friction coefficient. The decrease in the relative penetration of surfaces after CPEN ensures a decrease in the deformation component of the friction coefficient. In the case of the predominance of plastic deformations on the friction unit, the deformation component has the most noticeable effect on the overall friction coefficient; the friction coefficient weakly depends on molecular interaction. As a result, the overall friction coefficient decreases after CPEN and CPEN, followed by PEP (Table 4, Figure 13b).

According to calculations of the relative penetration of irregularities, the disruption of frictional bonds occurs through the plastic displacement of the material, and the wear mechanism is identified as fatigue wear during dry friction and plastic contact.

The work of friction forces during tribological tests turns into heat, which causes an increase in temperature in the friction unit (Table 3 and Table 4). The temperatures reached at the frictional contact when testing samples after APEN and CPEN, followed by PEP, under conditions of plastic deformation can activate additional oxidation of the friction surfaces in the atmosphere. In the case of APEN followed by PEP, the oxides are sheared along the metal-oxide line on the friction tracks; the time interval between successive destructions of the oxides is not sufficient for the formation of a film of relatively large thickness. Figure 14 shows practically no dark oxidized areas on the friction tracks. During the cathodic process, the outer oxide layer is further oxidized to higher oxide, which is a very effective lubricant and can also provide conditions for external friction, prevent deep tearing, and help reduce the coefficient of friction. Most likely, the denser and stronger oxide layers formed during CPEN will be preserved and replenished with oxidation products during friction. This will contribute to greater wear resistance of the nitrided steel surface in CPEN as opposed to APEN, where the hardened metal substrate is directly involved in the tribocoupling.

In general, the use of PET to increase the wear resistance of tool steels has shown a positive result. The multiple increase in wear resistance exceeded the results of laser hardening [2,3] and quenching followed by cryogenic treatment [5]. This determines the prospects for using PET for processing tools and critical parts.

4. Conclusions

1.

The possibility of increasing the hardness and wear resistance of the surface of 90CrSi tool steel by PEN followed by PEP has been demonstrated. The influence of electrophysical parameters of treatment on the composition, structure, morphology, and tribological properties of the steel surface was studied.

2.

Structural changes in the surface of 90CrSi steel after PEN were revealed. Regardless of the polarity of the workpiece, an oxide layer with different phase composition and morphology is formed on the surface as a result of high-temperature oxidation of steel in electrolyte vapor. The formed oxide layers after APEN have a lower density compared to CPEN, which is explained by the process of anodic dissolution of steel and the formation of vacancies and pores. As a result of anodic dissolution, the surface roughness after APEN is reduced by 2.7 times. Free vacancies and pores in the oxide layer favor nitrogen diffusion: the diffusion coefficient during anodic treatment is 2.50 μm²/s, while after CPEN its value is 0.30 μm²/s. Consequently, after APEN, a nitride layer is formed under the oxide layer, which is not detected after CPEN. In both cases, martensite is formed as a result of quenching, and the microhardness increases to 1110–1120 HV. Due to incomplete martensitic transformation, retained austenite and cementite are identified in the surface layers.

3.

It has been established that as a result of PEP, oxide layers having morphological inhomogeneities in the form of cracks, chips, and various irregularities are partially removed. The preservation of the nitride layer occurs only with PEP in a sulfate electrolyte for 1 min when surface passivation inhibits the electrochemical dissolution of steel. Under other PEP conditions, the nitride layer dissolves, and the roughness increases due to uneven etching. A denser oxide layer after cathodic treatment with PEP is removed more uniformly, which is accompanied by a decrease in roughness by 1.7 times. It was noted that under any PEP conditions after APEN and CPEN, the strengthened layer is preserved.

4.

The features of the tribological behavior of the steel surface after processing at various PET parameters were determined. Wear of the surface of 90CrSi steel after APEN and subsequent PEP is accompanied by the formation of plastically deformed microvolumes of metal, which, upon repeated exposure, lead to the formation of wear particles and separation of the base material. In contrast, after CPEN and subsequent PEP, denser oxide layers take the load and, while slowly deteriorating, act as friction lubricants. This leads to the concentration of strains in a thin surface oxide layer and is reflected in a decrease in the relative penetration of the sample into the surface of the counterbody by 1.7 times after CPEN compared to the untreated sample, while after APEN, this figure decreases by 1.1 times. Similarly, the bearing capacity of a steel surface after CPEN followed by PEP is significantly higher than after APEN followed by PEP: in the first case, the Kragelsky-Kombalov criterion decreases by 1.4–1.5 times, and in the second case, it increases by 1.4–1.6 times compared to an untreated surface. The nature of wear is reflected in an increase in the friction coefficient by 1.3–1.5 times after APEN and subsequent PEP and in a decrease by 1.2 times after CPEN and two times after subsequent PEP. Despite significant differences in wear patterns, all modified samples showed an increase in wear resistance: 5–9 times after APEN followed by PEP, 16 times after CPEN, and up to 32 times after subsequent PEP. In all cases, the disruption of frictional bonds occurs through the plastic displacement of the material, and the wear mechanism is classified as fatigue wear during dry friction and plastic contact.