- Email: [email protected]
Surface hardening of steels with carbon by non-vacuum electron-beam processing
Surface & Coatings Technology 242 (2014) 164–169
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Surface hardening of steels with carbon by non-vacuum electron-beam processing I.A. Bataev a,⁎, M.G. Golkovskii b, A.A. Bataev a, A.A. Losinskaya a, R.A. Dostovalov a, A.I. Popelyukh a, E.A. Drobyaz a a b
Novosibirsk State Technical University, 630092 Novosibirsk, Karl Marks prospect, 20, Russia Budker Institute of Nuclear Physics, Siberian Branch of the Russian Academy of Sciences, 630090 Novosibirsk, Akademika Lavrentieva prospect, 11, Russia
a r t i c l e
i n f o
Article history: Received 23 October 2013 Accepted in revised form 13 January 2014 Available online 27 January 2014 Keywords: Carburizing Low carbon steel Electron-beam Microstructure Mechanical properties
a b s t r a c t Surface layers containing ~1.57–2.55% (wt) carbon were produced by atmospheric electron-beam cladding of low carbon steel plates with an iron–graphite powder mixture. The main process parameter determining the thickness, structure, and mechanical properties of the hardened layer was the electron-beam current. As the beam current was increased from 20 to 26 mA, the thickness of the cladding layers increased from 1.2 to 2.6 mm and the hardness decreased from 5.7 to 4.5 GPa. In friction tests against fixed abrasive particles, the wear-resistance of the cladding layers was close to the wear resistance of pack-carburized specimens. In electron-beam cladding of steel plates 10 mm thick with the powder mixture with electron-beam scanning over the plate surface, the cooling rate of the surface layer was less than the critical value, which made it impossible to obtain a martensitic structure. The main structural components in the cladding layers were ledeburite, secondary Widmanstätten cementite, and pearlite. To produce a high-carbon martensitic structure directly during cladding by enhancing the heat transfer to the colder volume of the workpiece, it is necessary to increase the thickness and mass of the workpiece or reduce the thickness of the hardened layer. Saturation with carbon and quenching of the cladding layer can be performed successively using the same electron accelerator. © 2014 Elsevier B.V. All rights reserved.
1. Introduction One of the simplest, most economical and effective method of hardening of steel parts is the diffusion saturation of their surface layers with carbon [1]. In the technical literature, this process is called carburizing. There are various methods for introducing carbon into the surface layers of steel articles under factory conditions. Carburizing is carried out in various solid, liquid and gaseous media [1] which act as carbon sources. Articles subjected to carburizing are usually heated above the point AC3 (in the temperature range from 910 to 930 °C, in some cases, to 1050 °C). At these temperatures, steel is in the austenite state. Saturation of the surface layers of articles with carbon occurs by a diffusion mechanism. Usually, carburizing is combined with subsequent quenching and low-temperature tempering. Carburizing is used for low-carbon steels containing about 0.1–0.2% C, characterized by high ductility and low hardness, strength, and durability. The mechanism of steel hardening by carburizing and subsequent quenching involves the formation of a high-carbon martensite structure with high hardness in the surface layers. A hypereutectoid structure is formed (pearlite and secondary cementite) during the carburizing in the layer located closer to the workpiece surface. After quenching and low-temperature tempering, a ⁎ Corresponding author. Tel.: +7 913 913 2956. E-mail address: [email protected] (I.A. Bataev). 0257-8972/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.surfcoat.2014.01.038
tempered martensite with globular cementite inclusions is formed in this layer. This structure improves the wear resistance of the material under different wear conditions. In this case, the core of the steel article has increased toughness while remaining low-carbon. Despite the advantages noted above, the carburizing process also has disadvantages. One of these is low productivity and high energy consumption of the process. The formation of hardened layer 0.5–2 mm thick requires carburizing for 6–8 h or more. Another drawback is that the technological capabilities of carburizing methods are limited by the dimensions of the thermal equipment. Many large articles requiring hardening cannot be placed in existing furnaces. The design and operation of carburizing equipment for large articles are not economically rational. An effective solution to this problem is the cladding of the surface of an article with a hard material. There are many methods for cladding with high-wear-resistant materials. These are, first of all, laser [2–8], electron-beam [9–12], plasma [13,14], and electric-arc [15] cladding processes. The technology of surface hardening of steels using electron beams extracted into air is worthy of special mention. This technology, implemented at powerful industrial accelerators produced at the Budker Institute of Nuclear Physics (Novosibirsk), provides high-performance cladding of steel products with durable materials [10,11] and surface hardening of steel [16,17]. The powder mixtures used for cladding typically contain carbides, nitrides, boron and various borides. This makes it possible to produce surface layers of complex compositions, hardened with high-strength
I.A. Bataev et al. / Surface & Coatings Technology 242 (2014) 164–169
165
1
particles. Both published data and the results of our studies show that carbon can be used as a cladding material in these processes. Saturation of the surface layers of steels with carbon is carried out using laser [2–8] and plasma cladding [14] and electrocontact thermochemical treatment [18]. Some of these methods are also often called laser surface alloying [3,5,7,8] or selective case hardening [14]. Carbon for cladding is commonly used in the form of graphite [2–8,14]. However, a number of studies have demonstrated the possibility of using carbon nanotubes for alloying [6,19], which has some advantages but is impractical due to the high cost of nanotubes. The aim of this work was to study the surface hardening of low carbon steel plates by remelting a powder mixture containing graphite using electron beams extracted into the atmosphere.
2 3
50 V
4
2. Materials and methods
5
6
Fig. 1. Schematic diagram of non-vacuum electron-beam cladding of steel plate with the iron–graphite powder mixture. 1—Outlet, 2—electron beam, 3—electromagnetic scanning device, 4—powder mixture of graphite, iron, and flux, 5—cladding layer, 6—steel substrate.
The workpieces were 100 × 50 × 10 mm plates of low carbon steel (0.19% C, 0.47% Mn, 0.20% Si, 0.009% P, 0.042% S, 0.15% Ni, 0.15% Cu). The powder mixture for cladding consisted of graphite, carbonyl iron and a flux which protected the melt from oxygen. Magnesium fluoride (MgF2 ) was used as a flux. In the powder mixture, the flux content was 50% (wt), and the concentrations of iron and graphite were 25% (wt) each. The iron powder mixed with graphite provided the preservation of carbon and its more uniform distribution in the surface layer. Before cladding, the powder mixture was uniformly spread over the surface of the steel workpiece in an amount of 0.2 g per 1 cm2. The cladding process was performed in an ELV-6 electron accelerator [9–12]. The electron-beam energy was 1.4 MeV. The beam current was varied in the range of 20–26 mA. The electron beam was extracted into the atmosphere through a diaphragm of 1 mm diameter. The distance from the beam outlet to the workpiece was 90 mm. Under these conditions, the Gaussian diameter of the electron beam on the specimen surface was 12 mm. The longitudinal speed of the specimen relative to the outlet was 10 mm/s. The main modes of cladding are presented in Table 1. The beam was scanned with an electromagnetic scanner to increase the performance of the treatment (modes 1–4). The peak-to-peak value of the scanning electron beam was the same as the sample width and was equal to 50 mm. The high scanning frequency (50 Hz) provided uniform exposure of the sample area to the beam (Fig. 1). One of the workpieces was processed with an electron beam without scanning the surface to reduce the amount of molten material, reduce the amount of heat introduced into the workpiece, and increase the cooling rate of the cladded layer (mode 5). The workpiece was moved in the longitudinal direction at a speed of 10 mm/s relative to the beam, and the beam current was equal to 6 mA. Chemical composition of the cladded layers was determined on an ARL 3460 optical emission spectrometer. The structure of the materials was studied using optical metallography (Carl Zeiss Axio Observer Z1m microscope), scanning electron microscopy (Carl Zeiss EVO 50 XVP microscope), transmission electron microscopy (Tecnai G2 20 TWIN microscope), and X-ray diffraction (ARL X'TRA θ–θ diffractometer). Diffraction patterns were taken using Cu Kα radiation. The specimen surfaces were scanned in a step-scan mode (Δ2θ = 0.02°) with a dwell time of 5 s per point. Metallographic microsections were made by mechanical grinding and subsequent polishing. The structure of
alloys was etched using a 3% solution of nitric acid in ethanol. The preparation of foils for transmission electron-microscopic studies included cutting of flat specimens, grinding of sample to a thickness of 0.1 mm, and dimple grinding on a Gatan Dimple Grinder and ion thinning on a Gatan PIPS 659 ion mill. Microhardness was measured on cross sections using a Wolpert Group 402 MVD microhardness tester at an indenter load of 0.98 N. To estimate the embrittlement of steel by the cladding layers the Charpy V-notch test was carried out. The scheme of the test was previously used in Ref. [20]. The wear resistance of the cladding materials was investigated in friction tests against fixed abrasive particles (GOST 17367-71). The test specimens had a cylindrical shape 2 mm in diameter and 10 mm high. During the tests, the specimens were pressed with a power of 3 N against an abrasive cloth rotating at a rate of 100 rpm and were simultaneously moved in the radial direction. The trajectory of the specimens with respect to the abrasive cloth had the shape of the Archimedes spiral. Neighboring tracks did intersect each other. The abrasive material was silicon carbide with particle sizes of 80 to 100 μm. The duration of the tests was 35 s. 3. Results and discussion The melting of the materials in an oxidizing medium, low density of graphite and the high rate of its heating suggested that under a brief exposure to a high-power electron beam, the carburizing process may not be sufficiently effective. Expected problems were related to the possible oxidation of carbon and ejection of the graphite powder with the formation of a gas cloud during rapid heating of the material. Nevertheless, chemical analysis and structural studies showed that the electron-beam melting of the powder mixture and the substrate material resulted in the formation of surface layers with increased carbon content. According to atomic-emission spectrometry data, the carbon content in the layers melted by the electron beam was 1.57–2.55% (wt) (Table 1).
Table 1 Modes of non-vacuum electron-beam cladding (beam electron energy—1.4 MeV, diameter of the electron spot on the surface—12 mm, speed of movement—10 mm/s). Mode number
Beam current, mA
Transverse scanning
Specific surface energy kJ/cm2
Thickness of the cladding layer, mm
Maximum carbon content, % (wt)
Weight loss of powder mixture, including the weight of slag (flux),%
1 2 3 4 5
20 22 24 26 6
Yes
5.6 6.2 6.7 7.3 7.0
1.2 1.3 2.0 2.6 –
2.55 2.27 2.19 1.57 –
77 75 73 75 –
No
I.A. Bataev et al. / Surface & Coatings Technology 242 (2014) 164–169
CL
C2
1
P
C2
2
P
3
L
4
The loss of the powder mixture during the electron-beam melting was 73–75% by weight (Table 1). The expected losses caused by the removal of the flux (due to scattering and burnout during cladding and mechanical removal of the slag) were 50%. Thus, the loss of the iron–carbon mixture accounted for 25% of the total loss of the initial powder weight. Elemental analysis of the cladding layer suggested that carbon was lost to the greatest extent during the processing. The results of metallographic studies indicated good quality of the cladding layers. In the specimens obtained with electron-beam scanning, microcracks were not found, indicating a reduced level of residual mechanical stresses. Single equiaxed pores with a size of ~ 5–50 μm were observed near the surface. The cladding layer structure revealed by etching is shown in Fig. 2a. A cross section of a specimen with a cladding layer obtained with electron-beam scanning is shown schematically in Fig. 3. In the cladding layer with maximum carbon content produced at a beam current of 20 mA, it was possible to distinguish zones of hypoeutectic cast iron (zone 1) in Figs. 2b and 3 and narrow zones corresponding to hypereutectoid (zone 2), eutectoid (zone 3), and hypoeutectoid (zone 4) steels. The zone of the substrate bordering the cladding material contained a coarse ferrite–pearlite structure with Widmanstätten ferrite crystals. Its formation was due to the overheating of steel by the heat transferred from the melted material. Features of the structure and carbon content in the abovementioned zones were determined by the process conditions of the electron-beam cladding of the iron–graphite mixture. The main process parameter that determined the structure and mechanical properties of the cladding layers was the electron-beam current. For the same values of the electron-beam energy, speed of movement of steel workpieces, and other parameters, an increase in the beam current led to an increase in the depth of the metal melting. This resulted in a reduction in the concentration of carbon in the cladding layer. After processing of the powder mixture with an electron beam at I = 20 mA, the maximum carbon content in the surface layer (2.55%) corresponded to the hypoeutectic cast iron. In this case, the
FW F P
5
166
Fig. 3. Diagram of the surface layer produced by non-vacuum electron-beam cladding. 1—Hypoeutectic white cast iron layer; 2—layer of hypereutectoid steel with secondary Widmanstätten cementite; 3—layer with a pearlite structure; 4—heataffected zone with pearlite and ferrite structures predominantly of Widmanstätten type; 5—initial ferrite–pearlite structure of low carbon steel. CL—ledeburite cementite, C2—secondary cementite, P—pearlite, Fw—Widmanstätten ferrite, F—polyhedral ferrite, L—ledeburite.
Fig. 2. Structure of surface layers produced using a scanning electron beam. a—I = 20 mA; b—I = 20 mA; numerical symbols correspond to the numbers given in Fig. 3; c—I = 24 mA; and d—I = 24 mA.
I.A. Bataev et al. / Surface & Coatings Technology 242 (2014) 164–169
8
Microhardness, GPa
a decrease in microhardness from 5–6 GPa to 1.8–1.9 GPa occur in a layer with thickness of about 10–15% of the total thickness of the hardened layer (Fig. 4). The type of structure that formed during the cladding is largely determined by the cooling rate of the material in the temperature range from the point Ar1 to the temperature of the beginning of the martensitic transformation. In turn, this rate depends on the mass of the steel slab (substrate thickness), beam current, speed of motion of the sample relative to the beam, and the degree of overlapping of the heated zones during beam scanning. Increasing the amount of heat introduced into the processed workpiece reduces the cooling rate of the surface layer material. In the process of cladding of 10 mm thick plates with an electron beam current I = 20–26 mA and a beam scanning frequency of 50 Hz, the cooling rate of the alloy was less than the critical value. For this reason, in the specimens obtained under these conditions, a martensitic structure was not observed. Structural analysis revealed only ledeburite, secondary cementite, and pearlite. X-ray phase analysis revealed only the presence of α-iron and cementite in the cladding layers (Fig. 5a). It should be noted that other studies using beam scanning also have not found martensite, which may due to the small size of the substrate. During cladding, the substrate was highly heated, and the rate of heat transfer from it was lower than that required for quenching. In Ref. [14], the specimen was placed in a tank with water to obtain a martensitic structure in the cladding layer. During the cladding without beam scanning, the hardened layer had the form of a track ~ 1.2–1.3 mm deep and ~ 12 mm wide. During this
a
Intensity a.u.
-Fe Fe3 C
30
40
50
60
70
80
90
100
2θ deg
b -Fe -Fe Fe3 C
Intensity a.u.
thickness of the cladding layer was 1.2 mm. Almost all carbon in the cladding layer was in the form of cementite. The structure resulting from cladding with the powder mixture at a beam current I = 20– 24 mA was hypoeutectic white iron (Fig. 2c). As the beam current was increased to 26 mA, melting of steel occurred to a depth of 2.6 mm. The carbon concentration in the layer reduced to a level corresponding to hypereutectoid steel (1.57% C). Metallographic microscope images show pearlite and laminar Widmanstätten cementite (Fig. 2d). The formation of cementite of this form is due to heating of steel to high temperature [21,22]. Similar changes in structure with increasing power of the source were reported in a number of papers, for example Refs. [7,18]. In the case of non-vacuum electron-beam cladding, a special mention should be made of the great thickness of the resulting coatings (up to 2.6 mm, Table 1), which is much greater than the thickness of coatings obtained by laser alloying [2–8] and even by the PTA technology [14]. Furthermore, the performance of electron-beam cladding is much higher compared to the abovementioned processes. In our case, the rate of cladding was 500 mm2/s, while an additional increase in the beam current was still possible. The high cladding rate attained in this work is due to the high power of the ELV-6 electron accelerator and features of the interaction of electrons with the material. The 1.4 MeV electron beam used in this work has a great depth of electron penetration in metals. For iron, it is 0.8 mm, i.e., comparable with the required thickness of the layer formed. Thus, the cladding layer is heated to a considerable depth almost instantaneously. In the laser and plasma cladding processes, only the surface of the workpiece is heated due to the interaction with the source, and the heating of the entire cladding layer due to heat transfer processes requires more time. As noted above, an increase in the beam current led to an increase in the thickness of cladding layers, which ultimately affected their microhardness (Fig. 4). Initially, the microhardness of steel was ~2 GPa. The highest microhardness (~5.7 GPa) was reached after cladding with the iron–graphite powder mixture at an electron beam current of 20 mA. Increasing the beam current to 24 mA resulted in a decrease in microhardness to ~ 5.1 GPa. The average microhardness of the surface layer deposited at I = 26 mA was 4.5 GPa. The hardness of the cladding layers is consistent with the results of Refs. [7,14] and others in the cases where laser or plasma coating had hypoeutectic or eutectic structures. Carbon was distributed fairly uniformly across the thickness of the cladding layers. Only after cladding at a beam current of 24 mA, zones depleted in carbon were found within a layer with the structure of hypoeutectic cast iron. Within these zones, Widmanstätten cementite crystals were present and primary cementite crystals were absent. Analysis of the microhardness distribution across the thickness of the hardened layer shows that the width of the transition zones within which the microhardness decreases to that of the substrate is ~ 200– 250 μm. Thus, the most significant structural changes associated with
167
6
4 3
2 1
2
30 0
600
1200
1800
2400
3000
Distance, μm Fig. 4. Microhardness distribution across the thickness of the layer deposited using a scanning electron beam: 1—I = 20 mA, 2—I = 24 mA, and 3—I = 26 mA.
40
50
60
70
80
90
100
2θ deg Fig. 5. Diffraction patterns of specimens after electron-beam cladding. a—Process with electron-beam scanning, I = 20 mA; b—track cladding process without beam scanning, I = 6 mA.
I.A. Bataev et al. / Surface & Coatings Technology 242 (2014) 164–169
treatment, the volume of the molten material was significantly less than that in the processes with beam scanning. In this case, the rate of cooling of the material of the surface layer ensured the formation of a complex mixed structure (Fig. 6). Results of optical metallography (Fig. 6a, b, c), scanning and transmission electron microscopic (Fig. 6d) studies showed that the main structural components of the cladding layer were martensite, Widmanstätten cementite lamellas, retained austenite, and a small fraction of pearlite. In many regions, the Widmanstätten ferrite crystals produced by overheating of the top of the steel substrate were arranged in parallel to the martensite crystals formed at the bottom of the cladding layer. The parallelism of these crystals can be judged from the light dotted line drawn in Fig. 6c. This can be explained by the fact that in these regions, both the martensite or Widmanstätten ferrite crystals occurred within the same austenite grain. In this case, the orientation relations between ferrite (F) and austenite (A) and between martensite (M) and austenite are of the same type: {111}A||{110}F, b110 NA||b111 NF and {111}A||{110}M, b 110NA||b111NM. Retained austenite in the surface layer, along with α-iron and cementite, was determined by X-ray diffraction (Fig. 5b). On micrographs retained austenite looks like light zones (marked by digit 3 in Fig. 6a, b, c), within martensite crystals and Widmanstätten cementite lamellas are located. The formation of retained austenite was favored by the high heating temperature of the material and the high carbon content in the layer. It should be noted that the contribution of retained austenite to the mechanical properties of the cladding layers is controversial. Retained austenite reduces the hardness of the coating [7,14]. In addition, its decay leads to a change in the geometric dimensions, which may be essential to the fabrication and operation of precision parts. At the same time, it has been reported [23] that increasing the volume fraction of retained austenite leads to an increase in the corrosion resistance and a significant improvement in the cavitation resistance of cladding layers. The γ → α′ transformation absorbs the part of the energy that in the absence of retained austenite may lead to the destruction of the coating material.
10
Microhardness, GPa
168
8
6
4
2
0
600
1200
1800
2400
Distance, μm Fig. 7. Microhardness distribution across the thickness of the layer deposited by track cladding without beam scanning, I = 6 mA.
The formation of martensite during the cladding increased the microhardness of the material to ~8–9 GPa (Fig. 7). This example demonstrates that there are significant differences between the cladding processes with and without beam scanning. Impact tests performed with V-notched specimens indicated that the cladding with high-carbon layers had an embrittling effect on lowcarbon steels (Fig. 8). In this case, the impact toughness correlated with the thickness rather than the carbon content of the hardened layer. Specimens with a 2.6 mm thick coating produced by electronbeam cladding with a beam current of 26 mA had minimum toughness. Reduction in the thickness of the cladding layer to 1.2 mm was accompanied by an increase of 60 kJ/m2 in the impact toughness. The fracture behavior of the cladding layer was fundamentally different from that of the substrate material. The casting origin of the surface layer and the high content of cementite in it led to embrittlement of the material.
Fig. 6. Structure of the surface layers and transition zones formed by electron-beam cladding of low carbon steel without electron-beam scanning at a current of 6 mA. a, b, and c—optical metallography, d—transmission electron microscopy. 1—Pearlite, 2—martensite, 3—retained austenite.
I.A. Bataev et al. / Surface & Coatings Technology 242 (2014) 164–169
1,2
Relative wear resistance , ε
Impact strength, J/cm2
180
169
160
140
120
100
1,1 1,0 0,8 0,6 0,4 0,2
80 20
22
24
1
26
2
3
4
5
Beam current, mA Fig. 8. Impact toughness of specimens versus electron-beam current.
Because of this, the zones of the substrate and the cladding material had markedly different fracture surfaces. Results of tribological tests of the analyzed materials are presented in Fig. 9. The reference material was low carbon steel after pack carburizing to a depth of 0.8 mm, with subsequent quenching and tempering at 200 °C. The relative wear resistance of this material was taken as unity. There was no significant difference in the wear resistance between the materials produced by the cladding technique and the pack carburizing. Specimens with the maximum thickness of the cladding layer and the minimum carbon content showed the lowest wear resistance during the test. Specimens produced without electronbeam scanning (with the martensite structure in the surface layer) showed 20% higher wear-resistance than that of pack carburized specimens. Thus, electron-beam cladding of low carbon steel with the iron– graphite powder mixture provides high wear resistance commensurate with that of the carburized steel. The results of our study show that the technology of electron-beam cladding of low-carbon steel with the iron–graphite powder mixture provides various opportunities for improving the hardness and the wear resistance of materials: 1. Cladding with a layer of white cast iron. Additional heat treatment of the material is not required. Wear resistance of the surface layer is ensured by ledeburite and primary cementite. 2. Cladding with a layer of hypereutectoid steel with pearlite and secondary cementite structures. In this case, additional heat treatment, including surface hardening and low-temperature tempering, is appropriate. Both cladding and subsequent heat treatment can be performed using the same electron accelerator [16,17]. Thus, the cladding of steel with the iron–graphite mixture allows a wide variation in the structural, mechanical, and tribological properties of surface layers. 4. Conclusion
Fig. 9. Relative wear resistance of specimens with carbon-hardened surface layers under conditions of friction against fixed abrasive particles. 1—Pack carburizing to a depth of 0.8 mm, quenching, and low-temperature tempering, 2—electron-beam cladding with electron-beam scanning at I = 26 mA; 3—I = 24 mA; 4—I = 22 mA; 5—electron-beam cladding without electron-beam scanning at I = 6 mA.
was lower than the critical value, which made it impossible to obtain a martensitic structure. The main structural components in the cladding layers were ledeburite, secondary Widmanstätten cementite, and pearlite. To produce a high-carbon martensitic structure directly during cladding by enhancing the heat transfer to the colder volume of the workpiece, it is necessary to increase the thickness and mass of the workpiece or reduce the thickness of the hardened layer. Saturation with carbon and quenching of the cladding layer can be performed successively using the same electron accelerator. Acknowledgments The authors gratefully acknowledge financial support from the Russian Ministry of Education and Science (research task #2014/138, project #257). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]
Surface layers containing ~ 1.57–2.55% carbon were produced by atmospheric electron-beam cladding of low carbon steel plates with the iron–graphite powder mixture. The role of iron in the powder mixture was to provide a uniform carbon distribution and a homogeneous structure across the thickness of the cladding layer. The main process parameter determining the thickness, structure, and mechanical properties of the hardened layer was the electron-beam current. As the beam current was increased from 20 to 26 mA, the thickness of the cladding layers increased from 1.2 to 2.6 mm and the hardness decreased from 5.7 to 4.5 GPa. In the cladding process with electron-beam scanning over the surface of 10 mm thick steel plates, the rate of cooling of the surface layer
[13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
S. Lampman, ASM Handbook, Heat Treating, volume 04, 1990, pp. 259–267. A. Walker, D.R.F. West, W.M. Steen, Met. Technol. 11 (1984) 399–406. D.K. Das, Mater. Charact. 38 (1997) 135–141. A.I. Katsamas, G.N. Haidemenopoulos, Surf. Coat. Technol. 115 (1999) 249–255. A.I. Katsamas, G.N. Haidemenopoulos, Surf. Coat. Technol. 139 (2001) 183–191. J.H. Yao, C.N. Ma, M.X. Gao, F.Z. Kong, Q.L. Zhang, Surf. Coat. Technol. 201 (2006) 2854–2858. J.H. Abboud, K.Y. Benyounis, A.G. Olabi, Hashmi M. St, J. Mater. Process. Technol. 182 (2007) 427–431. I.Y. Khalfallah, M.N. Rahoma, J.H. Abboud, K.Y. Benyounis, Opt. Laser Technol. 43 (2011) 806–813. S.N. Fadeev, M.G. Golkovski, A.I. Korchagin, N.K. Kuksanov, A.V. Lavruhin, S.E. Petrov, et al., Radiat. Phys. Chem. 57 (2000) 653–655. I.A. Bataev, A.A. Bataev, M.G. Golkovsky, A.Y. Teplykh, V.G. Burov, S.V. Veselov, Surf. Coat. Technol. 207 (2012) 245–253. A. Teplykh, M. Golkovskiy, A. Bataev, E. Drobyaz, S. Veselov, E. Golovin, et al., Adv. Mater. Res. 287-290 (2011) 26–31. M. Golkovsky, T. Zhuravina, I. Bataev, A. Bataev, S. Veselov, V. Bataev, et al., Adv. Mater. Res. 314–316 (2011) 23–27. S.S. Samotugin, N.K. Solyanik, A.V. Puiko, Weld. Int. 9 (1995) 489–493. J. Nikolaou, L. Bourithis, G. Papadimitriou, J. Mater. Sci. 38 (2003) 2883–2891. M. Eroglu, N. Ozdemir, Surf. Coat. Technol. 154 (2002) 209–217. E.A. Bataeva, I.A. Bataev, V.G. Burov, L.I. Tushinskii, M.G. Golkovskii, Met. Sci. Heat Treat. 51 (2009) 103–105. V.A. Bataev, A.A. Bataev, M.G. Golkovskii, P.I. Ostromenskii, B.V. Korotaev, Met. Sci. Heat Treat. 44 (2002) 510–513. Y.G. Gurevich, Powder Metall. Met. Ceram. 50 (2012) 619–624. J.H. Yao, Q.L. Zhang, M.X. Gao, W. Zhang, Appl. Surf. Sci. 254 (2008) 7092–7097. M.G. Golkovski, I.A. Bataev, A.A. Bataev, A.A. Ruktuev, T.V. Zhuravina, N.K. Kuksanov, R.A. Salimov, V.A. Bataev, Mater. Sci. Eng. A Struct. 578 (2013) 310–317. A.A. Bataev, V.A. Bataev, S.A. Kotorov, P.V. Ilyushenko, A.I. Popelyukh, Phys. Met. Metallogr. 85 (1998) 678–681. G. Spanos, M.V. Kral, Int. Mater. Rev. 54 (2009) 19–47. C.T. Kwok, H.C. Man, F.T. Cheng, Surf. Coat. Technol. 126 (2000) 238–255.
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Surface hardening of steels with carbon by non-vacuum electron-beam processing I.A. Bataev a,⁎, M.G. Golkovskii b, A.A. Bataev a, A.A. Losinskaya a, R.A. Dostovalov a, A.I. Popelyukh a, E.A. Drobyaz a a b
Novosibirsk State Technical University, 630092 Novosibirsk, Karl Marks prospect, 20, Russia Budker Institute of Nuclear Physics, Siberian Branch of the Russian Academy of Sciences, 630090 Novosibirsk, Akademika Lavrentieva prospect, 11, Russia
a r t i c l e
i n f o
Article history: Received 23 October 2013 Accepted in revised form 13 January 2014 Available online 27 January 2014 Keywords: Carburizing Low carbon steel Electron-beam Microstructure Mechanical properties
a b s t r a c t Surface layers containing ~1.57–2.55% (wt) carbon were produced by atmospheric electron-beam cladding of low carbon steel plates with an iron–graphite powder mixture. The main process parameter determining the thickness, structure, and mechanical properties of the hardened layer was the electron-beam current. As the beam current was increased from 20 to 26 mA, the thickness of the cladding layers increased from 1.2 to 2.6 mm and the hardness decreased from 5.7 to 4.5 GPa. In friction tests against fixed abrasive particles, the wear-resistance of the cladding layers was close to the wear resistance of pack-carburized specimens. In electron-beam cladding of steel plates 10 mm thick with the powder mixture with electron-beam scanning over the plate surface, the cooling rate of the surface layer was less than the critical value, which made it impossible to obtain a martensitic structure. The main structural components in the cladding layers were ledeburite, secondary Widmanstätten cementite, and pearlite. To produce a high-carbon martensitic structure directly during cladding by enhancing the heat transfer to the colder volume of the workpiece, it is necessary to increase the thickness and mass of the workpiece or reduce the thickness of the hardened layer. Saturation with carbon and quenching of the cladding layer can be performed successively using the same electron accelerator. © 2014 Elsevier B.V. All rights reserved.
1. Introduction One of the simplest, most economical and effective method of hardening of steel parts is the diffusion saturation of their surface layers with carbon [1]. In the technical literature, this process is called carburizing. There are various methods for introducing carbon into the surface layers of steel articles under factory conditions. Carburizing is carried out in various solid, liquid and gaseous media [1] which act as carbon sources. Articles subjected to carburizing are usually heated above the point AC3 (in the temperature range from 910 to 930 °C, in some cases, to 1050 °C). At these temperatures, steel is in the austenite state. Saturation of the surface layers of articles with carbon occurs by a diffusion mechanism. Usually, carburizing is combined with subsequent quenching and low-temperature tempering. Carburizing is used for low-carbon steels containing about 0.1–0.2% C, characterized by high ductility and low hardness, strength, and durability. The mechanism of steel hardening by carburizing and subsequent quenching involves the formation of a high-carbon martensite structure with high hardness in the surface layers. A hypereutectoid structure is formed (pearlite and secondary cementite) during the carburizing in the layer located closer to the workpiece surface. After quenching and low-temperature tempering, a ⁎ Corresponding author. Tel.: +7 913 913 2956. E-mail address: [email protected] (I.A. Bataev). 0257-8972/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.surfcoat.2014.01.038
tempered martensite with globular cementite inclusions is formed in this layer. This structure improves the wear resistance of the material under different wear conditions. In this case, the core of the steel article has increased toughness while remaining low-carbon. Despite the advantages noted above, the carburizing process also has disadvantages. One of these is low productivity and high energy consumption of the process. The formation of hardened layer 0.5–2 mm thick requires carburizing for 6–8 h or more. Another drawback is that the technological capabilities of carburizing methods are limited by the dimensions of the thermal equipment. Many large articles requiring hardening cannot be placed in existing furnaces. The design and operation of carburizing equipment for large articles are not economically rational. An effective solution to this problem is the cladding of the surface of an article with a hard material. There are many methods for cladding with high-wear-resistant materials. These are, first of all, laser [2–8], electron-beam [9–12], plasma [13,14], and electric-arc [15] cladding processes. The technology of surface hardening of steels using electron beams extracted into air is worthy of special mention. This technology, implemented at powerful industrial accelerators produced at the Budker Institute of Nuclear Physics (Novosibirsk), provides high-performance cladding of steel products with durable materials [10,11] and surface hardening of steel [16,17]. The powder mixtures used for cladding typically contain carbides, nitrides, boron and various borides. This makes it possible to produce surface layers of complex compositions, hardened with high-strength
I.A. Bataev et al. / Surface & Coatings Technology 242 (2014) 164–169
165
1
particles. Both published data and the results of our studies show that carbon can be used as a cladding material in these processes. Saturation of the surface layers of steels with carbon is carried out using laser [2–8] and plasma cladding [14] and electrocontact thermochemical treatment [18]. Some of these methods are also often called laser surface alloying [3,5,7,8] or selective case hardening [14]. Carbon for cladding is commonly used in the form of graphite [2–8,14]. However, a number of studies have demonstrated the possibility of using carbon nanotubes for alloying [6,19], which has some advantages but is impractical due to the high cost of nanotubes. The aim of this work was to study the surface hardening of low carbon steel plates by remelting a powder mixture containing graphite using electron beams extracted into the atmosphere.
2 3
50 V
4
2. Materials and methods
5
6
Fig. 1. Schematic diagram of non-vacuum electron-beam cladding of steel plate with the iron–graphite powder mixture. 1—Outlet, 2—electron beam, 3—electromagnetic scanning device, 4—powder mixture of graphite, iron, and flux, 5—cladding layer, 6—steel substrate.
The workpieces were 100 × 50 × 10 mm plates of low carbon steel (0.19% C, 0.47% Mn, 0.20% Si, 0.009% P, 0.042% S, 0.15% Ni, 0.15% Cu). The powder mixture for cladding consisted of graphite, carbonyl iron and a flux which protected the melt from oxygen. Magnesium fluoride (MgF2 ) was used as a flux. In the powder mixture, the flux content was 50% (wt), and the concentrations of iron and graphite were 25% (wt) each. The iron powder mixed with graphite provided the preservation of carbon and its more uniform distribution in the surface layer. Before cladding, the powder mixture was uniformly spread over the surface of the steel workpiece in an amount of 0.2 g per 1 cm2. The cladding process was performed in an ELV-6 electron accelerator [9–12]. The electron-beam energy was 1.4 MeV. The beam current was varied in the range of 20–26 mA. The electron beam was extracted into the atmosphere through a diaphragm of 1 mm diameter. The distance from the beam outlet to the workpiece was 90 mm. Under these conditions, the Gaussian diameter of the electron beam on the specimen surface was 12 mm. The longitudinal speed of the specimen relative to the outlet was 10 mm/s. The main modes of cladding are presented in Table 1. The beam was scanned with an electromagnetic scanner to increase the performance of the treatment (modes 1–4). The peak-to-peak value of the scanning electron beam was the same as the sample width and was equal to 50 mm. The high scanning frequency (50 Hz) provided uniform exposure of the sample area to the beam (Fig. 1). One of the workpieces was processed with an electron beam without scanning the surface to reduce the amount of molten material, reduce the amount of heat introduced into the workpiece, and increase the cooling rate of the cladded layer (mode 5). The workpiece was moved in the longitudinal direction at a speed of 10 mm/s relative to the beam, and the beam current was equal to 6 mA. Chemical composition of the cladded layers was determined on an ARL 3460 optical emission spectrometer. The structure of the materials was studied using optical metallography (Carl Zeiss Axio Observer Z1m microscope), scanning electron microscopy (Carl Zeiss EVO 50 XVP microscope), transmission electron microscopy (Tecnai G2 20 TWIN microscope), and X-ray diffraction (ARL X'TRA θ–θ diffractometer). Diffraction patterns were taken using Cu Kα radiation. The specimen surfaces were scanned in a step-scan mode (Δ2θ = 0.02°) with a dwell time of 5 s per point. Metallographic microsections were made by mechanical grinding and subsequent polishing. The structure of
alloys was etched using a 3% solution of nitric acid in ethanol. The preparation of foils for transmission electron-microscopic studies included cutting of flat specimens, grinding of sample to a thickness of 0.1 mm, and dimple grinding on a Gatan Dimple Grinder and ion thinning on a Gatan PIPS 659 ion mill. Microhardness was measured on cross sections using a Wolpert Group 402 MVD microhardness tester at an indenter load of 0.98 N. To estimate the embrittlement of steel by the cladding layers the Charpy V-notch test was carried out. The scheme of the test was previously used in Ref. [20]. The wear resistance of the cladding materials was investigated in friction tests against fixed abrasive particles (GOST 17367-71). The test specimens had a cylindrical shape 2 mm in diameter and 10 mm high. During the tests, the specimens were pressed with a power of 3 N against an abrasive cloth rotating at a rate of 100 rpm and were simultaneously moved in the radial direction. The trajectory of the specimens with respect to the abrasive cloth had the shape of the Archimedes spiral. Neighboring tracks did intersect each other. The abrasive material was silicon carbide with particle sizes of 80 to 100 μm. The duration of the tests was 35 s. 3. Results and discussion The melting of the materials in an oxidizing medium, low density of graphite and the high rate of its heating suggested that under a brief exposure to a high-power electron beam, the carburizing process may not be sufficiently effective. Expected problems were related to the possible oxidation of carbon and ejection of the graphite powder with the formation of a gas cloud during rapid heating of the material. Nevertheless, chemical analysis and structural studies showed that the electron-beam melting of the powder mixture and the substrate material resulted in the formation of surface layers with increased carbon content. According to atomic-emission spectrometry data, the carbon content in the layers melted by the electron beam was 1.57–2.55% (wt) (Table 1).
Table 1 Modes of non-vacuum electron-beam cladding (beam electron energy—1.4 MeV, diameter of the electron spot on the surface—12 mm, speed of movement—10 mm/s). Mode number
Beam current, mA
Transverse scanning
Specific surface energy kJ/cm2
Thickness of the cladding layer, mm
Maximum carbon content, % (wt)
Weight loss of powder mixture, including the weight of slag (flux),%
1 2 3 4 5
20 22 24 26 6
Yes
5.6 6.2 6.7 7.3 7.0
1.2 1.3 2.0 2.6 –
2.55 2.27 2.19 1.57 –
77 75 73 75 –
No
I.A. Bataev et al. / Surface & Coatings Technology 242 (2014) 164–169
CL
C2
1
P
C2
2
P
3
L
4
The loss of the powder mixture during the electron-beam melting was 73–75% by weight (Table 1). The expected losses caused by the removal of the flux (due to scattering and burnout during cladding and mechanical removal of the slag) were 50%. Thus, the loss of the iron–carbon mixture accounted for 25% of the total loss of the initial powder weight. Elemental analysis of the cladding layer suggested that carbon was lost to the greatest extent during the processing. The results of metallographic studies indicated good quality of the cladding layers. In the specimens obtained with electron-beam scanning, microcracks were not found, indicating a reduced level of residual mechanical stresses. Single equiaxed pores with a size of ~ 5–50 μm were observed near the surface. The cladding layer structure revealed by etching is shown in Fig. 2a. A cross section of a specimen with a cladding layer obtained with electron-beam scanning is shown schematically in Fig. 3. In the cladding layer with maximum carbon content produced at a beam current of 20 mA, it was possible to distinguish zones of hypoeutectic cast iron (zone 1) in Figs. 2b and 3 and narrow zones corresponding to hypereutectoid (zone 2), eutectoid (zone 3), and hypoeutectoid (zone 4) steels. The zone of the substrate bordering the cladding material contained a coarse ferrite–pearlite structure with Widmanstätten ferrite crystals. Its formation was due to the overheating of steel by the heat transferred from the melted material. Features of the structure and carbon content in the abovementioned zones were determined by the process conditions of the electron-beam cladding of the iron–graphite mixture. The main process parameter that determined the structure and mechanical properties of the cladding layers was the electron-beam current. For the same values of the electron-beam energy, speed of movement of steel workpieces, and other parameters, an increase in the beam current led to an increase in the depth of the metal melting. This resulted in a reduction in the concentration of carbon in the cladding layer. After processing of the powder mixture with an electron beam at I = 20 mA, the maximum carbon content in the surface layer (2.55%) corresponded to the hypoeutectic cast iron. In this case, the
FW F P
5
166
Fig. 3. Diagram of the surface layer produced by non-vacuum electron-beam cladding. 1—Hypoeutectic white cast iron layer; 2—layer of hypereutectoid steel with secondary Widmanstätten cementite; 3—layer with a pearlite structure; 4—heataffected zone with pearlite and ferrite structures predominantly of Widmanstätten type; 5—initial ferrite–pearlite structure of low carbon steel. CL—ledeburite cementite, C2—secondary cementite, P—pearlite, Fw—Widmanstätten ferrite, F—polyhedral ferrite, L—ledeburite.
Fig. 2. Structure of surface layers produced using a scanning electron beam. a—I = 20 mA; b—I = 20 mA; numerical symbols correspond to the numbers given in Fig. 3; c—I = 24 mA; and d—I = 24 mA.
I.A. Bataev et al. / Surface & Coatings Technology 242 (2014) 164–169
8
Microhardness, GPa
a decrease in microhardness from 5–6 GPa to 1.8–1.9 GPa occur in a layer with thickness of about 10–15% of the total thickness of the hardened layer (Fig. 4). The type of structure that formed during the cladding is largely determined by the cooling rate of the material in the temperature range from the point Ar1 to the temperature of the beginning of the martensitic transformation. In turn, this rate depends on the mass of the steel slab (substrate thickness), beam current, speed of motion of the sample relative to the beam, and the degree of overlapping of the heated zones during beam scanning. Increasing the amount of heat introduced into the processed workpiece reduces the cooling rate of the surface layer material. In the process of cladding of 10 mm thick plates with an electron beam current I = 20–26 mA and a beam scanning frequency of 50 Hz, the cooling rate of the alloy was less than the critical value. For this reason, in the specimens obtained under these conditions, a martensitic structure was not observed. Structural analysis revealed only ledeburite, secondary cementite, and pearlite. X-ray phase analysis revealed only the presence of α-iron and cementite in the cladding layers (Fig. 5a). It should be noted that other studies using beam scanning also have not found martensite, which may due to the small size of the substrate. During cladding, the substrate was highly heated, and the rate of heat transfer from it was lower than that required for quenching. In Ref. [14], the specimen was placed in a tank with water to obtain a martensitic structure in the cladding layer. During the cladding without beam scanning, the hardened layer had the form of a track ~ 1.2–1.3 mm deep and ~ 12 mm wide. During this
a
Intensity a.u.
-Fe Fe3 C
30
40
50
60
70
80
90
100
2θ deg
b -Fe -Fe Fe3 C
Intensity a.u.
thickness of the cladding layer was 1.2 mm. Almost all carbon in the cladding layer was in the form of cementite. The structure resulting from cladding with the powder mixture at a beam current I = 20– 24 mA was hypoeutectic white iron (Fig. 2c). As the beam current was increased to 26 mA, melting of steel occurred to a depth of 2.6 mm. The carbon concentration in the layer reduced to a level corresponding to hypereutectoid steel (1.57% C). Metallographic microscope images show pearlite and laminar Widmanstätten cementite (Fig. 2d). The formation of cementite of this form is due to heating of steel to high temperature [21,22]. Similar changes in structure with increasing power of the source were reported in a number of papers, for example Refs. [7,18]. In the case of non-vacuum electron-beam cladding, a special mention should be made of the great thickness of the resulting coatings (up to 2.6 mm, Table 1), which is much greater than the thickness of coatings obtained by laser alloying [2–8] and even by the PTA technology [14]. Furthermore, the performance of electron-beam cladding is much higher compared to the abovementioned processes. In our case, the rate of cladding was 500 mm2/s, while an additional increase in the beam current was still possible. The high cladding rate attained in this work is due to the high power of the ELV-6 electron accelerator and features of the interaction of electrons with the material. The 1.4 MeV electron beam used in this work has a great depth of electron penetration in metals. For iron, it is 0.8 mm, i.e., comparable with the required thickness of the layer formed. Thus, the cladding layer is heated to a considerable depth almost instantaneously. In the laser and plasma cladding processes, only the surface of the workpiece is heated due to the interaction with the source, and the heating of the entire cladding layer due to heat transfer processes requires more time. As noted above, an increase in the beam current led to an increase in the thickness of cladding layers, which ultimately affected their microhardness (Fig. 4). Initially, the microhardness of steel was ~2 GPa. The highest microhardness (~5.7 GPa) was reached after cladding with the iron–graphite powder mixture at an electron beam current of 20 mA. Increasing the beam current to 24 mA resulted in a decrease in microhardness to ~ 5.1 GPa. The average microhardness of the surface layer deposited at I = 26 mA was 4.5 GPa. The hardness of the cladding layers is consistent with the results of Refs. [7,14] and others in the cases where laser or plasma coating had hypoeutectic or eutectic structures. Carbon was distributed fairly uniformly across the thickness of the cladding layers. Only after cladding at a beam current of 24 mA, zones depleted in carbon were found within a layer with the structure of hypoeutectic cast iron. Within these zones, Widmanstätten cementite crystals were present and primary cementite crystals were absent. Analysis of the microhardness distribution across the thickness of the hardened layer shows that the width of the transition zones within which the microhardness decreases to that of the substrate is ~ 200– 250 μm. Thus, the most significant structural changes associated with
167
6
4 3
2 1
2
30 0
600
1200
1800
2400
3000
Distance, μm Fig. 4. Microhardness distribution across the thickness of the layer deposited using a scanning electron beam: 1—I = 20 mA, 2—I = 24 mA, and 3—I = 26 mA.
40
50
60
70
80
90
100
2θ deg Fig. 5. Diffraction patterns of specimens after electron-beam cladding. a—Process with electron-beam scanning, I = 20 mA; b—track cladding process without beam scanning, I = 6 mA.
I.A. Bataev et al. / Surface & Coatings Technology 242 (2014) 164–169
treatment, the volume of the molten material was significantly less than that in the processes with beam scanning. In this case, the rate of cooling of the material of the surface layer ensured the formation of a complex mixed structure (Fig. 6). Results of optical metallography (Fig. 6a, b, c), scanning and transmission electron microscopic (Fig. 6d) studies showed that the main structural components of the cladding layer were martensite, Widmanstätten cementite lamellas, retained austenite, and a small fraction of pearlite. In many regions, the Widmanstätten ferrite crystals produced by overheating of the top of the steel substrate were arranged in parallel to the martensite crystals formed at the bottom of the cladding layer. The parallelism of these crystals can be judged from the light dotted line drawn in Fig. 6c. This can be explained by the fact that in these regions, both the martensite or Widmanstätten ferrite crystals occurred within the same austenite grain. In this case, the orientation relations between ferrite (F) and austenite (A) and between martensite (M) and austenite are of the same type: {111}A||{110}F, b110 NA||b111 NF and {111}A||{110}M, b 110NA||b111NM. Retained austenite in the surface layer, along with α-iron and cementite, was determined by X-ray diffraction (Fig. 5b). On micrographs retained austenite looks like light zones (marked by digit 3 in Fig. 6a, b, c), within martensite crystals and Widmanstätten cementite lamellas are located. The formation of retained austenite was favored by the high heating temperature of the material and the high carbon content in the layer. It should be noted that the contribution of retained austenite to the mechanical properties of the cladding layers is controversial. Retained austenite reduces the hardness of the coating [7,14]. In addition, its decay leads to a change in the geometric dimensions, which may be essential to the fabrication and operation of precision parts. At the same time, it has been reported [23] that increasing the volume fraction of retained austenite leads to an increase in the corrosion resistance and a significant improvement in the cavitation resistance of cladding layers. The γ → α′ transformation absorbs the part of the energy that in the absence of retained austenite may lead to the destruction of the coating material.
10
Microhardness, GPa
168
8
6
4
2
0
600
1200
1800
2400
Distance, μm Fig. 7. Microhardness distribution across the thickness of the layer deposited by track cladding without beam scanning, I = 6 mA.
The formation of martensite during the cladding increased the microhardness of the material to ~8–9 GPa (Fig. 7). This example demonstrates that there are significant differences between the cladding processes with and without beam scanning. Impact tests performed with V-notched specimens indicated that the cladding with high-carbon layers had an embrittling effect on lowcarbon steels (Fig. 8). In this case, the impact toughness correlated with the thickness rather than the carbon content of the hardened layer. Specimens with a 2.6 mm thick coating produced by electronbeam cladding with a beam current of 26 mA had minimum toughness. Reduction in the thickness of the cladding layer to 1.2 mm was accompanied by an increase of 60 kJ/m2 in the impact toughness. The fracture behavior of the cladding layer was fundamentally different from that of the substrate material. The casting origin of the surface layer and the high content of cementite in it led to embrittlement of the material.
Fig. 6. Structure of the surface layers and transition zones formed by electron-beam cladding of low carbon steel without electron-beam scanning at a current of 6 mA. a, b, and c—optical metallography, d—transmission electron microscopy. 1—Pearlite, 2—martensite, 3—retained austenite.
I.A. Bataev et al. / Surface & Coatings Technology 242 (2014) 164–169
1,2
Relative wear resistance , ε
Impact strength, J/cm2
180
169
160
140
120
100
1,1 1,0 0,8 0,6 0,4 0,2
80 20
22
24
1
26
2
3
4
5
Beam current, mA Fig. 8. Impact toughness of specimens versus electron-beam current.
Because of this, the zones of the substrate and the cladding material had markedly different fracture surfaces. Results of tribological tests of the analyzed materials are presented in Fig. 9. The reference material was low carbon steel after pack carburizing to a depth of 0.8 mm, with subsequent quenching and tempering at 200 °C. The relative wear resistance of this material was taken as unity. There was no significant difference in the wear resistance between the materials produced by the cladding technique and the pack carburizing. Specimens with the maximum thickness of the cladding layer and the minimum carbon content showed the lowest wear resistance during the test. Specimens produced without electronbeam scanning (with the martensite structure in the surface layer) showed 20% higher wear-resistance than that of pack carburized specimens. Thus, electron-beam cladding of low carbon steel with the iron– graphite powder mixture provides high wear resistance commensurate with that of the carburized steel. The results of our study show that the technology of electron-beam cladding of low-carbon steel with the iron–graphite powder mixture provides various opportunities for improving the hardness and the wear resistance of materials: 1. Cladding with a layer of white cast iron. Additional heat treatment of the material is not required. Wear resistance of the surface layer is ensured by ledeburite and primary cementite. 2. Cladding with a layer of hypereutectoid steel with pearlite and secondary cementite structures. In this case, additional heat treatment, including surface hardening and low-temperature tempering, is appropriate. Both cladding and subsequent heat treatment can be performed using the same electron accelerator [16,17]. Thus, the cladding of steel with the iron–graphite mixture allows a wide variation in the structural, mechanical, and tribological properties of surface layers. 4. Conclusion
Fig. 9. Relative wear resistance of specimens with carbon-hardened surface layers under conditions of friction against fixed abrasive particles. 1—Pack carburizing to a depth of 0.8 mm, quenching, and low-temperature tempering, 2—electron-beam cladding with electron-beam scanning at I = 26 mA; 3—I = 24 mA; 4—I = 22 mA; 5—electron-beam cladding without electron-beam scanning at I = 6 mA.
was lower than the critical value, which made it impossible to obtain a martensitic structure. The main structural components in the cladding layers were ledeburite, secondary Widmanstätten cementite, and pearlite. To produce a high-carbon martensitic structure directly during cladding by enhancing the heat transfer to the colder volume of the workpiece, it is necessary to increase the thickness and mass of the workpiece or reduce the thickness of the hardened layer. Saturation with carbon and quenching of the cladding layer can be performed successively using the same electron accelerator. Acknowledgments The authors gratefully acknowledge financial support from the Russian Ministry of Education and Science (research task #2014/138, project #257). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]
Surface layers containing ~ 1.57–2.55% carbon were produced by atmospheric electron-beam cladding of low carbon steel plates with the iron–graphite powder mixture. The role of iron in the powder mixture was to provide a uniform carbon distribution and a homogeneous structure across the thickness of the cladding layer. The main process parameter determining the thickness, structure, and mechanical properties of the hardened layer was the electron-beam current. As the beam current was increased from 20 to 26 mA, the thickness of the cladding layers increased from 1.2 to 2.6 mm and the hardness decreased from 5.7 to 4.5 GPa. In the cladding process with electron-beam scanning over the surface of 10 mm thick steel plates, the rate of cooling of the surface layer
[13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
S. Lampman, ASM Handbook, Heat Treating, volume 04, 1990, pp. 259–267. A. Walker, D.R.F. West, W.M. Steen, Met. Technol. 11 (1984) 399–406. D.K. Das, Mater. Charact. 38 (1997) 135–141. A.I. Katsamas, G.N. Haidemenopoulos, Surf. Coat. Technol. 115 (1999) 249–255. A.I. Katsamas, G.N. Haidemenopoulos, Surf. Coat. Technol. 139 (2001) 183–191. J.H. Yao, C.N. Ma, M.X. Gao, F.Z. Kong, Q.L. Zhang, Surf. Coat. Technol. 201 (2006) 2854–2858. J.H. Abboud, K.Y. Benyounis, A.G. Olabi, Hashmi M. St, J. Mater. Process. Technol. 182 (2007) 427–431. I.Y. Khalfallah, M.N. Rahoma, J.H. Abboud, K.Y. Benyounis, Opt. Laser Technol. 43 (2011) 806–813. S.N. Fadeev, M.G. Golkovski, A.I. Korchagin, N.K. Kuksanov, A.V. Lavruhin, S.E. Petrov, et al., Radiat. Phys. Chem. 57 (2000) 653–655. I.A. Bataev, A.A. Bataev, M.G. Golkovsky, A.Y. Teplykh, V.G. Burov, S.V. Veselov, Surf. Coat. Technol. 207 (2012) 245–253. A. Teplykh, M. Golkovskiy, A. Bataev, E. Drobyaz, S. Veselov, E. Golovin, et al., Adv. Mater. Res. 287-290 (2011) 26–31. M. Golkovsky, T. Zhuravina, I. Bataev, A. Bataev, S. Veselov, V. Bataev, et al., Adv. Mater. Res. 314–316 (2011) 23–27. S.S. Samotugin, N.K. Solyanik, A.V. Puiko, Weld. Int. 9 (1995) 489–493. J. Nikolaou, L. Bourithis, G. Papadimitriou, J. Mater. Sci. 38 (2003) 2883–2891. M. Eroglu, N. Ozdemir, Surf. Coat. Technol. 154 (2002) 209–217. E.A. Bataeva, I.A. Bataev, V.G. Burov, L.I. Tushinskii, M.G. Golkovskii, Met. Sci. Heat Treat. 51 (2009) 103–105. V.A. Bataev, A.A. Bataev, M.G. Golkovskii, P.I. Ostromenskii, B.V. Korotaev, Met. Sci. Heat Treat. 44 (2002) 510–513. Y.G. Gurevich, Powder Metall. Met. Ceram. 50 (2012) 619–624. J.H. Yao, Q.L. Zhang, M.X. Gao, W. Zhang, Appl. Surf. Sci. 254 (2008) 7092–7097. M.G. Golkovski, I.A. Bataev, A.A. Bataev, A.A. Ruktuev, T.V. Zhuravina, N.K. Kuksanov, R.A. Salimov, V.A. Bataev, Mater. Sci. Eng. A Struct. 578 (2013) 310–317. A.A. Bataev, V.A. Bataev, S.A. Kotorov, P.V. Ilyushenko, A.I. Popelyukh, Phys. Met. Metallogr. 85 (1998) 678–681. G. Spanos, M.V. Kral, Int. Mater. Rev. 54 (2009) 19–47. C.T. Kwok, H.C. Man, F.T. Cheng, Surf. Coat. Technol. 126 (2000) 238–255.