Tribological Properties of Carbon Tool Steel after Plasma Electrolytic Nitrocarburizing

Abstract

The effect of plasma electrolytic nitrocarburizing on the wear resistance of carbon tool steel in friction couples with hardened steel and lead-tin bronze is considered in order to study the mechanism and type of wear, as well as the influence of structural and morphological characteristics of the surface on them. The microgeometry of friction tracks and its change with an increasing duration of friction tests are analyzed. The equilibrium roughness is determined, which is optimal for the friction couple and ensures minimal wear. The optimal values of the plasma electrolytic nitrocarburizing parameters, which provide the lowest values of the friction coefficient and wear rate, have been determined. The phase and elemental composition of the surface layer was studied using X-ray diffraction analysis and EDX analysis. The relationship of the microstructure of the nitrocarburized layer of tool steel with the friction coefficient and weight wear is established.

1. Introduction

Tool carbon steels are characterized by low hardness and good machinability by cutting and pressure in the annealed state. One of the applications of such steels is the manufacturing of cutting tools. The steel must have a high hardness in the cutting edge, much higher than the hardness of the material being machined and high wear resistance. Wear resistance is necessary to maintain the size and shape of the cutting edge during cutting. Increasing the wear resistance of the surfaces of products made of tool steels is relevant and promising in various industries [1,2,3,4,5,6,7].

One of the technological solutions to increase hardness and wear resistance is nitrocarburizing [8,9,10]. This surface modification method has several advantages. The simultaneous diffusion of nitrogen and carbon atoms in austenite accelerates the diffusion of the latter. In this regard, nitrocarburized layers have a higher hardenability than carburized ones. In addition, the introduction of nitrogen atoms into the surface layer leads to an increase in the content of retained austenite, which inhibits the formation of fatigue damage and increases fatigue strength.

The disadvantages of traditional nitrocarburizing include the need to strictly maintain the carburizing and nitriding capacity of the gaseous medium within the required limits. In conventional ovens, a surface concentration of saturating elements of 1% is reached in a few hours. The same result is obtained in a few seconds in the case of plasma electrolytic treatment [11,12,13]. Hence, the prospects of using plasma electrolytic nitrocarburizing (PENC) for the intensification of chemical-thermal treatment are obvious. It was previously shown that PENC of low-carbon steel in a solution of triethanolamine, glycerol and sodium carbonate, due to an increase in hardness, eliminates abrasive wear [14]. Similar results were obtained after PENC of the same steel in a carbamide electrolyte, when friction was carried out in tandem with a counter body made of high-strength steel [15]. Because of PENC of N13 steel in the same electrolyte, the type of wear changes from abrasive to adhesive [16]. The same results were shown after PENC in a solution of carbamide with ammonium chloride and sodium carbonate of 316L steel [17]. In the above and other articles, wear resistance is considered as one of the surface characteristics that changes under the influence of processing. At the same time, a more in-depth study of the mechanism and type of wear, depending on the structural features and morphology of the surface, as well as the material of the counter body, has not been carried out.

The purpose of this work is to study the effect of PENC on the wear resistance of carbon tool steel, to determine the type and mechanism of wear, to establish the effect of the obtained structure of the nitrocarburized layer on wear resistance characteristics, and to calculate the equilibrium roughness in tribocontact. Research is carried out in different friction pairs: PENC treated steel vs. (1) hardened tool alloy steel, and vs. (2) tin-lead bronze. The combination of heat-treated steel in a friction pair with bronze is widely used for various interfaces in industry. For example, for sliding bearings of various types, worm gears, screw-nut interfaces, bushings, and guides for injection molding machines. The friction of hardened steel on hardened steel is also widespread, for example, in bearings and rolling guides, machine tool slideways, gears, and so on. The choice of widely used materials for friction interfaces determines the relevance of the study.

2. Materials and Methods

The cylindrical samples (a diameter of 10 mm, a height of 15 mm) of tool carbon steel were used for PENC (wt.%: 0.76–0.83 C; 0.17–0.33 Si; 0.17–0.33 Mn; ≤0.25 Ni; ≤0.028 S; ≤0.03 P; ≤0.2 Cr; ≤0.25 Cu and balance Fe).

PENC was carried out in an aqueous electrolyte of the composition: ammonium chloride NH₄Cl (10%) and carbamide (NH₂)₂CO (20%). Carbamide is both a source of nitrogen and carbon, characterized by low cost and environmental safety. The treatment temperature was 650, 800, 850, and 900 °C. The duration of the PENC was 5 min. Quenching was carried out from a single-phase austenitic region. Samples after PENC at 800, 850, and 900 °C were quenched from the diffusion saturation temperature. PENC sample at 650 °C was quenched from a temperature of 800 °C. When the voltage was turned off, the vapor-gas envelope condensed. This led to the resumption of contact between the cooled electrolyte and the sample and the hardening of the latter. Two types of control samples were used in the work: raw, untreated and heated in an oven to 800 °C and then quenched in oil.

PENC was carried out in a working setup (Figure 1). The electrolyte was fed into the working chamber using a pump, and the flow rate was controlled by a rotameter. The electrolyte temperature at the outlet of the heat exchanger was recorded with multimeter using a chromel–alumel thermocouple. The samples were connected to the positive pole, and the working chamber was connected to the negative pole. Next, voltage was applied to the system and the samples were immersed in the electrolyte.

The structure of the PENC layers and the nitrogen concentration in them were determined using a Phenom G2 pro scanning electron microscope (SEM) (Phenom-World B.V., Eindhoven, The Netherlands).

The microhardness of the PENC layers was measured using a Vickers microhardness tester (Falcon 503, Innovatest Europe BV, Maastricht, The Netherlands) under a load of 0.1 N.

The geometric characteristics of the surface were measured using a Hommel tester t8000 profilometer (Jenoptik, Jena, Germany).

Weight measurements were performed on a CitizonCY224C electronic analytical balance (ACZET (Citizen Scale), Mumbai, India).

The phase composition of PENC layers was determined using a PANalytical Empyrean X-ray diffractometer (XRD) (Malvern Panalytical, Malvern, UK) with CoKα radiation using PANalytical High Score Plus software, program [18] and ICCD PDF-2 and COD [19].

Tribological tests were carried out according to the shaft-block scheme (Figure 2). The cylindrical sample was rotated at the required frequency. The counter body was a 2 mm thick rectangular plate. For the manufacture of counter bodies, two types of materials were used: tool alloy steel (wt.%: 0.9–1.2 Cr, 1.2–1.6 W, 0.8–1.1 Mn, 0.9–1.05 C and balance Fe) and tin-lead bronze (wt.%: 4–6 Pb; 4–6 Zn; 4–6 Sn and balance Cu). There was a semicircular cutout on one of the edges of the plate. The cutout tightly covered the surface of the sample. This kept the contact area of the friction pair constant. The pressing force was created using a pneumatic cylinder. The required value of the pressing force was set by the value of the air pressure in the cylinder. The design of the friction machine ensured that the alignment of the sample and the counter body was maintained despite the wear of the friction pair during testing. The measurement of the values of the forces required to calculate the friction coefficient was carried out using load cells. The temperature in the interface zone of the friction couple was measured using an infrared thermometer. Load cells and an infrared thermometer were interrogated using a microcontroller with a period of 0.3 s with the transmission of measurement results to a PC. Friction tests were carried out in dry friction mode: a load of 10 N, a sliding speed of 1.555 m/s, a friction distance of 1 km.

3. Results

3.1. Energy Characteristics of PENC

Figure 3 shows the current-voltage and temperature-voltage characteristics of the PENC. The energy of the power source is released in the zone of greatest resistance, that is, in the near-electrode region of the electrolyte. Therefore, heat will go from the electrolyte to the electrode. The formation of a vapor-gas envelope is a consequence of local boiling of the electrolyte due to internal heat sources due to the passage of an electric current [20,21,22]. The current-voltage characteristic has a falling character, since the thickness of the envelope increases with increasing voltage, and its electrical resistance increases. The temperature-voltage characteristic increases due to an increase in the power introduced into the system. The sample temperature in the range of 400–1000 °C is regulated by voltage, which allows hardening.

3.2. Composition, Structure and Microhardness of the Surface Layer

Quenching after PENC provides high cooling rates, on the order of hundreds of degrees per second [23]. This cooling rate is higher than the critical one for this steel. Therefore, the microstructure of all PENC samples is an acicular martensite (Figure 4) with a characteristic high hardness (Figure 5).

EDX analysis showed surface saturation with nitrogen at all temperatures (

Table 1. EDX analysis results.

PENC temperature (°C)	650	800	850	900
Nitrogen concentration (wt.%)	4.24	4.19	3.72	3.20

). The largest amount of nitrogen on the surface was recorded for the sample after PENC at 650 °C and decreases with its increase.

According to XRD data, the surface layer of samples nitrocarburized at a temperature of 650 °C contains iron nitrides and carbides (Figure 6). X-ray analysis of samples after PENC at higher temperatures revealed less intense peaks of iron nitrides. With an increase in the PENC temperature, the peaks corresponding to iron oxides become the most intense and developed in the X-ray patterns. The formation of oxides is a consequence of the high-temperature oxidation of iron in the gas-vapor envelope since the main component of the envelope is water vapor.

3.3. Tribological Properties of Treated Surface

Figure 7 shows the dependence of the friction coefficient on the sliding distance of PENC samples. As control samples, raw and quenched into oil after heating in a furnace to a temperature of 800 °C are presented. The counter body is made of hardened steel. The running in of PENC samples occurred earlier than that of the control ones. For a raw sample and a sample after traditional oil quenching, the friction coefficient decreases markedly after the first 250 m of the test. For PENC samples, steady wear occurs earlier, after about 100 m of testing. After the first 100 m of the path, the friction coefficient of PENC samples changes insignificantly.

The smallest value of the friction coefficient of 0.179 is shown by a sample nitrocarburized at a temperature of 650 °C and hardened from 800 °C (Figure 8). The minimum value of the friction coefficient is 2.8 times less than that of a raw untreated sample and 2.2 times less than that of a sample after traditional hardening. With an increase in the PENC temperature, the friction coefficient gradually increases to a value of 0.256 at temperatures of 850–900 °C. The maximum value of the friction coefficient for samples after PENC is 2 times lower than for raw and 1.5 times lower than for quenched in the traditional way.

The results of measurements of the friction coefficient correlate with the dependences of the weight loss of the sample during friction tests (Figure 9). The smallest weight loss is recorded for the sample with the minimum friction coefficient after PENC at 650 °C with quenching from 800 °C. In this case, the weight loss due to friction is 9.7 times less than that of the raw one and 7.5 times less, compared to the sample after traditional hardening. An increase in the PENC temperature leads to an increase in weight friction losses in accordance with an increase in the friction coefficient (Figure 8 and Figure 9). At all PENC temperatures, the weight loss during friction in PENC samples is lower than in two control samples.

The smallest weight loss of the hardened steel counter body (Figure 9) is achieved after working in tandem with the sample after PENC at 650 °C, which is characterized by the lowest friction coefficient during the entire test (Figure 7 and Figure 8). The operation of this friction pair provides 36% less weight loss of the counter body than the pair with the untreated control sample, and 24% less than when the counter body works with the sample after traditional hardening. Increasing the PENC temperature to 800 °C increases the weight loss of the hardened steel counter body. In the PENC temperature ranges from 650 to 800 °C, the weight loss of the counter body in the friction pair remains smaller than in friction pairs with both control samples. PENC at 850 and 900 °C leads to weight loss of the counter body that exceeds the control samples. At the PENC temperature of up to 800 °C, the oxide layer has a dense structure and is strongly bonded to the metal substrate. Increasing the temperature of PENC to 850 and 900 °C weakens the bond between oxides and the base metal, and oxides easily exfoliate during friction. After the peeling of the oxides, friction occurs on a metal substrate with high hardness (Figure 5). At lower temperatures, PENC oxides, that were firmly bonded to the substrate, play the role of a lubricant during friction, providing less weight loss of the counter body.

Figure 10 shows the dependence of the friction coefficient on the sliding distance of PENC samples when working in tandem with a counter body made of tin-lead bronze. PENC, at all temperatures, provides lower values of the friction coefficient than that of a raw untreated sample. PENC at temperatures in the range from 650 to 850 °C gives lower friction coefficients in comparison with the sample after traditional hardening. The minimum value of the friction coefficient (Figure 11), as in the case of friction on hardened steel was recorded for the sample after PENC at 650 °C with quenching from 800 °C. The value of the friction coefficient in this pair is 0.274, which is 1.6 times less than a pair with an untreated sample and 1.3 times less than in a friction pair with a sample quenched into oil after heating in a furnace.

The weight loss of samples after PENC in the process of friction on tin-lead bronze in all modes of PENC is less than that of a raw sample (Figure 12). PENC at temperatures not exceeding 800 °C leads to lower weight loss of the sample when compared to conventional hardening. The weight loss during friction in PENC samples increases with an increase in the PENC temperature, which correlates with an increase in the friction coefficient (Figure 11 and Figure 12). The minimum weight loss is achieved with a sample paired with tin-lead bronze with the lowest friction coefficient. PENC at a temperature of 650 °C provides 11.5 times less weight loss of the sample during friction compared to the untreated one and 5 times less than the sample quenched in the traditional way. The weight loss of the bronze counterbody when working on processed samples is less than on raw ones (Figure 12).

3.4. Determination of Equilibrium Roughness and the Type of Violation of Frictional Bonds during Friction

During the running-in period, the roughness of the friction surface undergoes significant changes. The running-in process is considered complete when the initial roughness becomes an operational one [24]. It was experimentally established [25] that at the end of running in, a roughness is formed on the friction surface, which does not depend on the initial roughness obtained during machining but depends only on the wear conditions. This roughness is optimal for a given pair, friction conditions, and provides minimal wear. It can be either less or more than the original [26].

I.V. Kragelsky and V.S. Kombalov [27] introduced the concept of “equilibrium roughness” for the roughness formed in the process of running in and proposed a dimensionless complex parameter for its evaluation. The dimensionless complex parameter of equilibrium roughness is determined from the equation

$$\Delta =\frac{R_{\max }}{r{b}^{\frac{1}{\nu }}},$$

(1)

where R_max is the maximum roughness height; r is the radius of rounding of the tops of microasperity; and ν and b are parameters of the reference Abbott curve.

Modeling the protrusions of a rough sample by bodies of double curvature gives the following expression for the average radius of a single microasperity:

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

(2)

where R_a is the arithmetic mean profile deviation; R_p is the distance from the protrusion line to the center line within the base length; and S_m is the average step of the irregularities. These values are determined by profilometer measurements.

During the period of stationary wear, the equilibrium roughness is reproduced throughout the subsequent process of normal operation of the friction pair. The greater the difference between the initial roughness and the optimal one, the greater the wear during the running-in period, and, consequently, the shorter the durability of the friction pair. This suggests that the friction surface treatment technology has a significant impact on its durability.

Table 2. The values of the Kragelsky–Kombalov criterion Δ for PENC at different temperatures and control samples before and after tribological test.

Treatment Mode	PENC 650 °C	PENC 800 °C	PENC 850 °C	PENC 900 °C	Hardened	Untreated
Before friction	0.392 ± 0.008	0.397 ± 0.008	0.401 ± 0.008	0.410 ± 0.008	0.870 ± 0.007	0.904 ± 0.018
After friction on hardened steel	0.301 ± 0.006	0.307 ± 0.006	0.318 ± 0.006	0.304 ± 0.006	0.586 ± 0.012	0.611 ± 0.012
After friction on tin-lead bronze	0.318 ± 0.006	0.320 ± 0.006	0.317 ± 0.006	0.302 ± 0.006	0.605 ± 0.012	0.703 ± 0.014

presents the values of the Kragelsky–Kombalov criterion Δ for samples after PENC at different temperatures and control ones before and after friction on hardened steel and tin-lead bronze. The surfaces of the sample and the counter body have a roughness of the same order, which follows from Table 2 and

Table 3. The values of the Kragelsky–Kombalov criterion Δ for counter bodies coupled different samples after tribological test. The value of the criterion before the test is given in parentheses.

Sample Type	PENC 650 °C	PENC 800 °C	PENC 850 °C	PENC 900 °C	Hardened	Untreated
Hardened steel counter body (0.807 ± 0.016)	0.315 ± 0.006	0.308 ± 0.006	0.312 ± 0.006	0.317 ± 0.006	0.598 ± 0.012	0.604 ± 0.012
Tin-lead bronze counter body (0.752 ± 0.015)	0.322 ± 0.006	0.315 ± 0.006	0.315 ± 0.006	0.315 ± 0.006	0.602 ± 0.012	0.656 ± 0.013

.

It follows from the obtained results that the roughness obtained by the samples after PENC is close to the values of the optimal roughness both in the friction pair with hardened steel and in the friction pair with bronze. The maximum difference between the values of the initial and optimal roughness is recorded for raw and traditionally hardened samples during friction against steel. In this case, the longest and most pronounced running-in period is observed (Figure 7). The work of raw and hardened samples on bronze gives a smaller difference between the initial and steady-state roughness. The running-in period becomes shorter and less pronounced on the dependence of the friction coefficient on the sliding distance (Figure 10).

Due to the surface roughness of the sample and the counter body, their interaction during friction is discrete, the contact occurs in separate areas, and the totality of which constitutes the actual contact area. Rough surfaces under normal load interpenetrate or flatten out, and corresponding stresses and deformations arise in the area of actual contact spots. During friction, a fixed volume of material lying in the near-surface layer experiences many repeated impacts. When the microasperity of the sample slides over the counter body, a frontal bead of the deformable material appears, which is under the influence of compressive stresses. Behind microasperity, due to friction forces, the material is stretched. As a result, an alternating cycle of material loading occurs. Each such cycle does not pass without a trace for the material; damage accumulates in it, weakening it, which ultimately leads to the destruction of the material and the separation of wear particles. Such destruction, according to [24,27], is usually called fatigue.

The variety of changes occurring in the contact layer leads to various types of wear. Therefore, the type of wear cannot be defined in one term. The mechanism of destruction of the surface layer can be different.

To determine the type of violation of frictional bonds in the process of friction, the relative penetration of the sample and the counter body was calculated similarly [28,29,30]:

$$\frac{h}{r}=\frac{\left(R_{\max }\right)}{r}\cdot {\left(\frac{N}{b\cdot HB}\right)}^{\frac{1}{\nu }},$$

(3)

where h is the absolute penetration of the sample into the counter body surface, and HB is the microhardness of the sample surface.

This characteristic makes it possible to distinguish between elastic contact, plastic contact, and microcutting. Friction bonds are destroyed due to elastic displacement during dry friction of steels when h/r < 0.01. Ploughing (plastic deformation) of the metal takes place at h/r between 0.01 and 0.10, while the destruction of friction surfaces as the result of microcutting begins at h/r > 0.10. The results of calculating the value (3) are presented in

Table 4. The values of the relative penetration of the sample into the counter body surface h/r for PENC at different temperatures and control samples after tribological test.

Treatment Mode	PENC 650 °C	PENC 800 °C	PENC 850 °C	PENC 900 °C	Hardened	Untreated
After friction on hardened steel	0.03	0.03	0.05	0.04	0.15	0.20
After friction on tin-lead bronze	0.04	0.04	0.04	0.06	0.08	0.16

. The relative error of all results does not exceed 2%.

It is shown that for all samples after PENC in tribocontact with both hardened steel and bronze, a plastic regime is realized. There is no adhesion and cohesion in the tribocontact zone. In raw and traditionally hardened specimens, frictional bonds are broken by microcutting when rubbing against steel. During friction on bronze, the mechanism of breaking frictional bonds in a raw sample is preserved, while in a hardened one it is replaced by a softer one—plastic deformation.

A more severe mode of destruction of friction bonds during friction of a raw sample against steel and bronze and a hardened sample against steel affects the high values of the friction coefficient and significant weight losses of these samples during friction (Figure 8, Figure 9, Figure 11 and Figure 12).

Thus, the optimal mode of the PENC, which provides the highest wear resistance under friction both coupled with bronze and hardened steel, is processed at a temperature of 650 °C for 5 min with hardening from 800 °C. According to morphological analysis, the structure of the hardened layer is a needle-like martensite of high hardness. X-ray diffraction analysis of the surface layer of these samples shows the presence of iron nitrides. The iron oxides formed at such a processing temperature are firmly bound to the metal base and play the role of lubricant in the friction process. The increased wear resistance characteristics of the samples after PENC at a temperature of 650 °C are explained by the total effect of the oxide layer and the high hardness of the martensitic base with nitrides included in it.

Profilometric analysis together with mathematical modeling allowed us to establish the type of wear mechanism of the samples. For all nitrocarburized samples, the nature of the interaction of surfaces in tribocontact is mechanical. There is no adhesion and cohesion in the friction zone. Friction bonds during friction are destroyed as a result of plastic displacement of the sample material by the counter body material. There is a residual deformation of the sample surface after the passage of counter body microasperities through it. A complex stress state occurs in the surface layer of the sample: there is a compression zone in front of the microasperity, and a stretching zone behind it. As a result, there is an alternating cycle of loading of the material. As a result, low-cycle friction fatigue occurs in the surface layer of the sample. Fatigue cracks begin to spread in a thin near-surface layer. Their development leads to the separation of wear particles.

Thus, the wear mechanism of samples made of carbon alloy steel after PENC at various temperatures during friction against hardened steel and bronze is fatigue wear during friction without lubrication and plastic contact. For raw and hardened samples, friction on steel takes place in a more severe microcutting mode. A hardened sample in a friction pair with bronze has a wear mechanism similar to samples after PENC.

4. Conclusions

(a)

PENC of samples from carbon tool steel for 5 min at various temperatures with quenching from a single-phase austenitic region reduces the friction coefficient and weight loss during friction on hardened steel. The best result was achieved for the sample after PENC at a temperature of 650 °C with quenching from 800 °C. The friction coefficient is 2.8 times less than that of a raw untreated sample and 2.2 times less than that of a sample after traditional hardening. Weight loss during friction is 9.7 times less than that of the raw one and 7.5 times less in comparison with the sample after traditional hardening. High wear resistance is associated with the resulting martensitic structure, saturated with iron carbides and nitrides.

(b)

PENC at temperatures of 650 and 800 °C during friction against hardened steel also provides a reduction in counter body weight loss during testing compared to a raw sample and a sample after traditional hardening. A friction pair from a sample after PENC at a temperature of 650 °C and hardened steel loses 36% less mass of the counter body than a pair with a control untreated sample, and 24% less than when working with a counter body with a sample after traditional hardening. An increase in the PENC temperature to 850 and 900 °C leads to the exfoliation of oxides from the surface of the sample during friction. At lower temperatures, PENC oxides firmly bonded to the substrate play the role of a lubricant during friction, providing less weight loss of the counter body.

(c)

PENC samples of carbon tool steel at any temperature when working on tin-lead bronze reduces the friction coefficient and frictional weight loss in comparison with the raw sample. Samples nitrocarburized at temperatures of 650 and 800 °C are characterized by lower friction coefficients and weight losses in tests compared to the sample after traditional hardening. The weight loss of the bronze counter body when working on samples after PENC and hardened has the same order and is 3.5 times less than when working on raw.

(d)

The wear mechanism of samples after PENC at all temperatures is fatigue wear during dry friction and plastic contact. For raw and hardened samples, when rubbing against steel, and for raw samples, when rubbing against bronze, a more severe friction mode is observed—microcutting.

(e)

Equilibrium roughness was estimated for samples after PENC and controls. The roughness obtained by the samples after PENC is close to the values of the optimal roughness both in the friction pair with hardened steel and in the friction pair with bronze. The maximum difference between the values of the initial and optimal roughness is recorded for raw and traditionally hardened samples during friction against steel. In this case, the longest and most pronounced period of running-in is observed.