- Email: [email protected]
Supersolidus liquid-phase sintering of ultrahigh-boron high-carbon steels for wear-protection applications
Materials Science and Engineering A 532 (2012) 511–521
Contents lists available at SciVerse ScienceDirect
Materials Science and Engineering A journal homepage: www.elsevier.com/locate/msea
Supersolidus liquid-phase sintering of ultrahigh-boron high-carbon steels for wear-protection applications A. Röttger a,∗ , S. Weber a,b , W. Theisen a a b
Lehrstuhl Werkstofftechnik, Ruhr-Universität Bochum, 44801 Bochum, Germany Helmholtz-Zentrum Berlin für Materialien und Energie GmbH, 14109 Berlin, Germany
a r t i c l e
i n f o
Article history: Received 15 September 2011 Received in revised form 27 October 2011 Accepted 29 October 2011 Available online 11 November 2011 Keywords: Boron Steels Wear resistance SLPS HIP
a b s t r a c t Powder metallurgy (PM) represents an alternative to conventional casting processes for the production of wear-resistant materials. PM hard alloys for wear-protection applications feature both higher strength and fracture toughness compared to cast hard alloys due to their more finely grained microstructure. However, densification by hot-isostatic pressing (HIP), the conventional PM-compaction method, is relatively expensive and thus partially counteracts low-cost processing. To increase the economic efficiency of the processing route, supersolidus liquid-phase sintering (SLPS) was investigated. In addition, expensive Ni- and Co-base hard alloys were substituted by boron-rich Fe-base hard-facing alloys. In this study, three ultrahigh-boron hard-facing alloy powders were densified by SLPS and HIP. The sintering temperatures were optimized by means of sintering experiments that were supported by thermodynamic calculations. Both densification states were investigated and compared with respect to the microstructure and the tribological and mechanical properties of the compacted hard-facing alloys. It was shown that the mechanical and tribological properties are strongly influenced by the microstructure. Although the microstructure is affected by the chemical composition, it can also be adapted by the densification process. SLPS-densified hard-facing alloys have a coarse microstructure that imparts not only a high wear resistance but also a detrimental effect on the mechanical properties. © 2011 Elsevier B.V. All rights reserved.
1. Introduction White cast irons are widely used for wear protection applications in the mining, mineral processing, and cement industries owing to their high hardness, high wear resistance, low price, and easy production [1–3]. These characteristics arise from a microstructure consisting of primary and/or eutectic hard phases that are finely distributed in an austenitic, pearlitic, or martensitic matrix [4,5]. White cast iron with added boron has recently been receiving more attention owing to the higher hardness of borides compared to carbides and the lowering effect of boron on the solidus temperature of steels. This means that steels containing sufficient boron can be processed at lower temperatures. One main attribute of boron-containing steels is the low solubility of boron in ␥-Fe and ␣-Fe and the formation of hard Fe-rich borides of the type M2 B, M3 (C,B), and M23 (C,B)6 [5–7]. As a result, it is not necessary to add hard-phase forming elements such as Mo, V, and Nb to obtain an Fe-base alloy with a large amount of hard phases.
∗ Corresponding author. Tel.: +49 0234 32 22366; fax: +49 0234 32 14104. E-mail addresses: [email protected] (A. Röttger), [email protected] (S. Weber), [email protected] (W. Theisen). 0921-5093/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.msea.2011.10.118
Only a few authors have dealt with boron additions higher than 1 mass%. This is due to the fact that an increasing boron content decreases the fracture toughness [8–11]. The microstructure of ultrahigh-boron, high-carbon alloys is characterized by primary solidified metal cells (hypoeutectic) or hard phases (hypereutectic) that are surrounded by a low-melting and almost brittle eutectic. This fine eutectic, in particular, promotes crack propagation because its fracture toughness is lower than that of the metal matrix, which results in insufficient mechanical properties. Godfrey et al. investigated Ti-base MMCs with added boron that were processed by hot-isostatic pressing. Their results indicated spheroidization of the hard phases during HIP treatment, which promotes an increase in mechanical strength [5,12,13]. Furthermore, boron addition inhibits strong coarsening of the microstructure due to precipitation of borides at the grain boundaries [12]. If a high wear resistance and good mechanical properties are required, the HIP process seems to be a promising processing route for boron-containing PM alloys. The addition of a small amount of boron to PM steels has been reported to have a major effect on the densification behavior during supersolidus liquid-phase sintering (SLPS) [14–16]. Sercombe investigated the effect of a small boron addition on the sintering behavior of a freeformed maraging PM steel. He showed that
512
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
the optimized sintering temperature (OST) could be decreased by increasing the boron content from 0 mass% to 0.4 mass% [17]. Boron forms a low-melting eutectic with Fe and therefore promotes the formation of a liquid phase at lower temperatures. However, boron forms stable phases with Mo, Ti, and Cr. This alters the phase composition of the steel matrix, which may have an effect on the mechanical and tribological properties after heat treatment. Based on the results of Sercombe and Lal et al., boron-containing PM steels are suitable materials for SLPS densification to achieve wearresistant materials with a low porosity [17–20]. The aim of the present paper is to study the densification behavior of three ultrahigh-boron, high-carbon gas-atomized Febase powders by means of supersolidus liquid-phase sintering and hot-isostatic pressing. The SLPS process was used to achieve fast densification due to the presence of low-melting, boron-rich eutectics. In addition, Weber et al. mentioned the possibility of influencing the size and shape of the hard phases in a ledeburitic cold-work tool steel by means of SLPS processing with the aim of adjusting the diameter of the hard phase to the wear system [21,22]. In contrast, spheroidization of hard phases due to a decrease in the surface energy during HIP is well known. It increases the mean free path of the metal matrix between two adjacent hard phases, which thus improves the mechanical properties. Therefore, this work investigated the influence of the densification process on the microstructure and on the mechanical and tribological properties. 2. Experimental procedure 2.1. Materials Three ultrahigh-boron, high-carbon steels were used as gas-atomized powders. Two of these were commercially available hypereutectic high-boron steels, X200CrNiBMo10-4-3-3 and X360NiBCrCu4-2, and the third was a self-developed hypoeutectic high-boron steel X60CrB2-2. The chemical composition of the bulk samples after HIP was measured by optical emission spark spectroscopy (QSG750 spectrometer OBLF) (Table 1). Prior to SLPS and HIP densification, the gas-atomized steel powders were analyzed with respect to their microstructure, particle size, and particle distribution by means of laser diffraction. 2.2. Thermodynamic calculations Determination of the optimized sintering and HIP temperatures was supported by thermodynamic equilibrium calculations using the Calphad method. Software package Thermo-Calc version R (Thermo-Calc AB, Stockholm, Sweden) and database TCFE6.2 were used to calculate the solidification path, phase diagrams, and the chemical composition of each phase in the equilibrium state. The calculations were performed with a phase set reduced to LIQUID, BCC A2, FCC A1, CEMENTITE, M7C3, M2B tetr, Cr2B orth, M23C6, and M6C. 2.3. Thermal analysis, sintering, and HIP Thermal analysis was performed to confirm the Thermo-Calc results concerning the sintering window. Differential thermal analysis (DTA) was performed by heating small amounts of gasatomized steel powder (∼100 mg) in alumina crucibles at a heating rate of 10 K min−1 in an argon gas atmosphere. The optimized sintering temperatures were investigated experimentally by sintering at different temperatures above the solidus temperature (TSOL ). The metal powder was filled into alumina preforms and precompacted to tap density. Sintering was performed in a batch furnace (∅ = 50 mm, L = 120 mm) under a vacuum (0.5 Pa)
with a temperature accuracy of ±5 K. The specimens were first heated to 300 ◦ C to degas the metal powder and to drive out residual vapor. They were then heated to the sintering temperature at a rate of 20 K min−1 . The isothermal holding time at the sintering temperature was 10 min for all specimens. SLPS was followed by a slow furnace cooling at a rate of approx. 15 K min−1 . After sintering at the different temperatures, the specimens were characterized by means of metallographic analysis, hardness measurements, and determination of the density by the Archimedes method. Besides SLPS-densification, the gas-atomized powders were compacted by hot-isostatic pressing (HIP) for 2 h in an argon atmosphere at a temperature of 1000 ◦ C and a pressure of 100 MPa. Ferritic steel capsules were filled with gas-atomized steel powder, which was then precompressed to tap density by vibration. The capsules were evacuated for 30 min and finally sealed by TIG welding.
2.4. Heat treatment To achieve a martensitic microstructure, the HIP- and SLPSdensified specimens were austenitized at 850–1050 ◦ C, quenched in water, and double-tempered in a temperature range from 100 to 600 ◦ C for 2 h.
2.5. Metallography and microscopy Microstructural examinations were carried out using optical and scanning electron microscopy (SEM). Specimens were sectioned, ground on abrasive paper, and polished with a 1 m diamond suspension. If necessary, the specimens were etched with 3% Nital. X-ray diffraction was used for phase analysis. Measurements were performed with a Siemens-D500 X-ray diffractometer using chromium K␣ radiation and a step size of 0.02◦ 2 in a range of 40–165◦ 2. The diffractograms were evaluated by the program ADM and the JCPDS database.
2.6. Hardness measurements and abrasive wear tests Macro- and microhardness measurements were carried out on the cross-section of the respective specimen using a Vickers indenter and a load of 294.3 N or 49.15 N. The wear resistance of the heat-treated specimens was investigated by a pin-on-paper test against the abrasives flint (SiO2 ) and corundum (Al2 O3 ). This type of test involves rotating a cylindrical specimen with a diameter of ∅ = 6 mm against an abrasive paper with an average particle size of 80 mesh (∼200 m) or 220 mesh (∼66 m). The dimensionless value of the wear resistance was calculated on the basis of the weight loss m of the specimen during testing, the specimen’s contact surface A, the coating density COAT , and the distance travelled L (Eq. (1)). −1 Wab =
A · L · COAT m
(1)
2.7. Four-point bending tests The mechanical properties were analyzed by four-point bending tests using specimens with a length of 70 mm and a cross-section of 5 mm × 5 mm. The testing method was based on DIN EN 53452 with a cross-head speed of 0.5 mm min−1 . The deflection of the specimens was measured by cross-head displacement and an extensometer.
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
513
Table 1 Chemical composition of the steel powders measured by optical emission spark spectrometry. Steel
C
B
Cr
Ni
Mo
Si
Mn
(P + S)
Cu
Fe
X200CrNiBMo10-4-2-2 X360NiBCrCu4-2 X60CrB2-2
2.12 3.60 0.56
3.54 2.40 1.73
11.96 1.20 2.02
3.95 4.30 0.21
3.28 – –
2.01 0.40 0.65
0.15 1.40 0.82
0.12 0.23 0.04
1.17 1.80 0.01
Bal. Bal. Bal.
3. Results 3.1. Powder characterization The microstructure of the steel powders in the gas-atomized state is shown in Fig. 1. The hypereutectic steels exhibit hard phases on the surfaces of the powder particles. The microstructure of the cross-section of steel powder X220CrNiBMo10-4-3-3 in the gasatomized condition features primary solidified Cr–Fe-rich M2 B and
Cr-rich M7 C3 in addition to Mo-rich ␥-Fe-M3 B2 and Fe-rich ␥-FeM3 (C,B) eutectics (Fig. 1a and b). The microstructure of carbon-rich hypereutectic steel X360NiBMoCr4-2 contains globular and primary solidified M3 (C,B), which are surrounded by a ␥-Fe-M3 (C,B) eutectic (Fig. 1c and d). In contrast, a dendritic microstructure on the surfaces of the powder particles can be seen for hypoeutectic steel X60CrB2-2 (Fig. 1e and f). The microstructure of the crosssection is characterized by ␥-Fe dendrites that are surrounded by a ␥-Fe-M2 B eutectic.
Fig. 1. Microstructure of the particle surface and of the cross-section of the steel particles: (a) and (b) steel X200CrNiBMo10-4-3-3; (c) and (d) steel X360NiBCrCu4-2; (e) and (f) steel X60CrB2-2.
514
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
Fig. 2. Sum distribution and probability density functions of steel powders X200CrNiBMo10-4-3-3, X360NiBCrCu4-2, and X60CrB2-2.
The steel powder particles were analyzed with regard to particle size and distribution using laser diffraction. The probability density functions of the steel powders are given in Fig. 2. The sum distribution functions of the steel powders are characterized by a multimodal function featuring two or three inflection points. Thus, several particle maxima are present. The median particle size of the hypereutectic steel powders is almost uniform and was measured to be x50,r = 70–71 m. In contrast, the median particle size of the hypoeutectic powder was found to be x50,r = 142 m. 3.2. Estimation of optimized sintering temperatures Thermodynamic calculations were performed to obtain an initial assessment regarding the sintering window. The melting sequence and the amount of liquid phase formed were calculated as a function of the temperature. However, these thermodynamic calculations describe the behavior in the equilibrium state, which is not the case in reality. DTA measurements were performed to estimate the differences between the calculated thermodynamic values and their experimental counterparts. TSOL and TLIQ were determined as the main signals, and the values were compared with the calculated results (Table 2). As shown in Table 2, thermodynamic calculations can be used for an initial estimation of the overall sintering interval. However, a higher solidus temperature was found by DTA measurements. The differences in temperature between the calculated and the experimental TSOL and TLIQ were 30–100 ◦ C. To obtain sintered specimens with a dense and fine-grained microstructure without a distinctive distortion, the optimized sintering parameters must be determined within a narrow temperature window. Therefore, sintering experiments were performed at different sintering temperatures. Furthermore, the sintered specimens were investigated with respect to their density and microstructure. curves for alloys X200CrNiBMo10-4-3-3, Sintering X360NiBCrCu4-2, and X60CrB2-2 are depicted in Figs. 3–5. The sintering curves confirm a minor densification of the powder particles at sintering temperatures below TSOL . The density of the sintered specimens correlates with the tap density of the powders before sintering. Thus the sintering mechanisms in the solid state led to slight skeletonization due to the formation of sintering bridges. A high degree of densification was not observed. On reaching TSOL , the density of the material rapidly increased. In this case, only the M3 (C,B) phase was observed in the microstructure of hard-facing alloy X360NiBCrCu4-2 (Fig. 3). As a result, eutectic M3 (C,B) generated a liquid phase, which surrounded the previously primary solidified hard phase M3 (C,B). The amount
Fig. 3. Sintering temperature versus density curve of steel X360NiBCrCu4-2.
of liquid phase formed by eutectic M3 (C,B) was calculated to be 22 vol.%. A further increase in sintering temperature led to complete melting of the primary M3 (C,B), which occurred within a T of 47 ◦ C Therefore, an intentional adjustment of the amount of liquid phase to approx. 30–40 vol.% cannot be achieved due to temperature fluctuations of ±8.5 ◦ C at the nominal temperature of 1140 ◦ C. Powder of X200CrNiBMo10-4-3-3 was completely densified by SLPS processing. Although the measured porosity after sintering at 1120 ◦ C was less than 1 vol.%, a higher sintering temperature of 1150 ◦ C was chosen to achieve a higher matrix hardness (Fig. 4). The sintering temperature versus density curve of steel X60CrB2-2 is characterized by a small gradient with an increase
Fig. 4. Sintering temperature versus density curve of steel X200CrNiBMo10-4-3-3.
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
515
Table 2 Calculated and experimental TSOL and TLIQ . Scheil calculations
Steel
TSOL (◦ C) X60CrB2-2 X220CrNiBMo10-4-3-3 X360NiBCrCu4-2
1092 1017 984
Equilibrium calculations
DTA measurements
TLIQ (◦ C)
TSOL (◦ C)
TLIQ (◦ C)
TSOL (◦ C)
TLIQ (◦ C)
1328 1223 1122
1095 1047 1053
1328 1223 1122
1143 1074 1112
– 1153 1177
Table 3 Optimized sintering parameters and hardness of the investigated steels. Steel
Optimized sintering temperature (OST) (◦ C)
Hardness (HV30)
Porosity (%)
X60CrB2-2 X360NiBCrCu4-2 X200CrNiBMo10-4-3-3
1220 1140 1150
635 ± 5 848 ± 18 919 ± 11
0.53 0.02 0.39
in density with increasing sintering temperature, although a low solidus temperature of TSOL = 1143 ◦ C was measured by DTA and the optimized sintering temperature was estimated to be TOST = 1220 ◦ C (Fig. 5). An overview of the optimized sintering parameters is given in Table 3. 3.3. Microstructural investigations The microstructure of the compacted steel powders in the SLPS, HIPed, and heat-treated state are depicted in Figs. 6–8. Phase analysis of the hard-facing alloys in the SLPS and heat-treated states was carried out by XRD measurements (Fig. 9). In the case of steel X200CrNiBMo10-4-3-3 in the SLPS state, the metal matrix contains mainly squared and lamellar hard particles of type M2 B and M7 C3 . In addition, EDX and X-ray diffraction analysis showed eutectic Mo-rich (M3 B2 ) and Fe-rich (M3 (C,B)) hard phases in an austenitic microstructure (Fig. 9b). The microstructure of steel X200CrNiBMo10-4-3-3 in the HIPed, quenched, and tempered state exhibited the same hard-phase composition in a martensitic–austenitic microstructure. The microstructure of the SLPS and HIPed state showed differences with respect to the shape
and size of the hard phases. The hard phases (M2 B, M7 C3 , M3 B2 and M3 (C,B)) in the HIPed state have a lamellar-to-globular shape and a higher amount of eutectic phases. In the case of steel X360NiBCrCu4-2 in the SLPS state, the lamellar-to-globular hard phases of type M3 (C,B) are distributed in the microstructure and surrounded by a fine lamellar boron cementite (M3 (C,B)) eutectic (Fig. 9c). The HIPed microstructure is characterized by smaller and globular M3 (C,B) hard-phase grains. Furthermore, eutectic boron cementite was detected at the triple points of grain boundaries of globular M3 (C,B). The hypoeutectic steel X60CrB2-2 in the SLPS and heat-treated state is characterized by globular metal cells that are surrounded by a Fe-rich M2 B network. XRD measurements indicate that the metal matrix has a martensitic microstructure (Fig. 9a). The microstructure in the HIPed and heat-treated state shows globular Fe-rich hard phases of type M2 B that are finely distributed in a mainly martensitic microstructure. Furthermore, a eutectic microstructure of the phases M23 (C,B)6 and ␣-Fe is observed. 3.4. Heat treatment The mechanical and tribological properties of Fe-base alloys can be adjusted by a heat treatment. In the case presented, quenching and tempering were performed to increase the strength and hardness of the metal matrix. Tempering curves of the hypereutectic steels are depicted in Fig. 10. Steel X200CrNiBMo104-3-3 possesses a high post-quenching hardness of 950 ± 56 HV30. Higher tempering temperatures decreased the hardness from 864 ± 45 HV30 at 100 ◦ C to 792 ± 10 HV30 at 400 ◦ C. The hardness increased again when the tempering temperature exceeded 400 ◦ C (peak of secondary hardness). Subsequently, SLPS as well as HIPed specimens were tempered at 520 ◦ C, at which temperature a secondary peak hardness of 859 ± 18 HV30 was measured. Steel X360NiBCrCu4-2 has a lower post-quenching hardness of 768 ± 34 HV30 compared to steel X200CrNiBMo10-4-3-3. Higher tempering temperatures continuously decreased the hardness of steel X360NiBCrCu4-2. The optimized tempering temperature was estimated to be 200 ◦ C. Hypoeutectic steel X60CrB2-2 was austenitized at TAUS = 850 ◦ C, 950 ◦ C, and 1050 ◦ C to investigate the influence of the solute carbon content in ␥-Fe on the post-quenching hardness. Increasing the austenitizing temperature increased the hardness (Fig. 10). In addition, the hardness of steel X60CrB2-2 decreased continuously with increasing tempering temperature. The hardness as well as the optimized heat treatment parameters for different densification treatments are summarized in Table 4. 3.5. Mechanical and tribological properties
Fig. 5. Sintering temperature versus density curve of steel X60CrB2-2.
3.5.1. Hardness measurements Owing to their high metalloid content in conjunction with a large amount of hard-phase-forming elements, the hard alloys investigated in this work are suitable for wear-protection applications. A high wear resistance requires a high matrix hardness as well as a large amount of hard phases dispersed in the metal
516
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
Fig. 6. Microstructure of steel X220CrNiBMo10-4-3-3 in the (a) SLPS and (b) HIPed state.
Fig. 7. Microstructure of steel X360NiBCrCu4-2 in the (a) SLPS and (b) HIPed state.
Fig. 8. Microstructure of steel X60CrB2-2 in the (a) SLPS and (b) HIPed state.
matrix. The hardness of the hard phases was measured by the Vickers indentation method, and the volume fraction of the hard phases was calculated by the Calphad method (Table 5). The chromiumrich carbide of type M7 C3 exhibited the highest hardness of
1823 ± 40 HV0.05 with respect to the solute chromium/iron ratio. The hardness of the M7 C3 phase decreased with increasing iron content. Furthermore, chromium-rich M2 B (HV0.05) exhibited a slightly lower hardness, which was estimated to be
Table 4 Hardness and optimized heat treatment parameters of the steels for different densification treatments. Material
Densification route
Heat treatment
Hardness (HV30)
X60CrB2-2 X60CrB2-2 X360NiBCrCu4-2 X360NiBCrCu4-2 X200CrNiBMo10-4-3-3 X200CrNiBMo10-4-3-3
HIP SLPS HIP SLPS HIP SLPS
TAUS = 1050 ◦ C, 2 × 2 h, 200 ◦ C TAUS = 1050 ◦ C, 2 × 2 h, 200 ◦ C TAUS = 1000 ◦ C, 2 × 2 h, 200 ◦ C TAUS = 1000 ◦ C, 2 × 2 h, 200 ◦ C TAUS = 1000 ◦ C, 2 × 2 h, 520 ◦ C TAUS = 1000 ◦ C, 2 × 2 h, 520 ◦ C
663 605 951 805 870 949
± ± ± ± ± ±
3 5 1 4 5 3
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
517
Table 5 Hardness and volume fraction (calculated by Calphad at T = 950 ◦ C) of the various phases in the investigated steels. Phase
M7 C3 M2 B M3 (C,B) M23 (C,B)6
X60CrB2-2
X360NiBCrCu4-2
X200CrNiBMo10-4-3-3
Vol. fraction
Hardness (HV0.05)
Vol. fraction
Hardness (HV0.05)
Vol. fraction
Hardness (HV0.05)
– 18.85 – 12.66
– 1654 ± 73 – 1275 ± 82
– 1.16 93.24 –
– – 1044 ± 38 –
13.25 43.25 11.59 –
1823 ± 40 1579 ± 81 905 ± 84 –
1654 ± 73 HV0.05 in steel X60CrB2-2 and 1579 ± 81 HV0.05 in steel X200CrNiBMo10-4-3-3. The lowest hardness of 905 ± 84 HV0.05 (eutectic) to 1044 ± 38 HV0.05 (primary) was measured in the boron cementite. 3.5.2. Abrasive wear tests The wear resistance of the steels was investigated by means of a pin-on-paper test using abrasives SiO2 (1000–1200 HV0.05) and Al2 O3 (2000–2250 HV0.05) with different particle sizes (220 mesh (66 m) and 80 mesh (200 m)). The results indicated a decrease in wear resistance with increasing abrasive particle size and abrasive hardness (Fig. 11). Abrasive Al2 O3 , in particular, had a higher hardness than the hard phases in the metal matrix. As a result, the abrasives are able to scratch and groove the hard particles. In the case of larger abrasive particle sizes, chipping and ploughing wear mechanisms became more active. In addition, the specimens in the SLPS state had a higher wear resistance than the HIP-densified specimens, except for steel X360NiBCrCu4-2. The highest wear resistance was exhibited by steel X200CrNiBMo10-4-3-3 in the SLPS and HIPed state. In the SLPS state, hypoeutectic steel X60CrB2-2 had a higher wear resistance compared to hypereutectic steel X360NiBCrCu4-2. In the HIPed state, steels X60CrB2-2 and X360NiBCrCu4-2 exhibited inverse behavior of the wear resistance. 3.5.3. Four-point bending test The bending strength of the hard-facing alloys in the quenched and tempered state was characterized by means of a four-point bending test. The results are listed in Table 6. The measured
Fig. 9. Phase identification by means of X-ray diffraction of steels (a) X60CrB22, (b) X200CrNiBMo10-4-3-3, and (c) X360NiBCrCu4-2 in the SLPS, quenched and tempered state.
bending strength was higher in the HIP-densified specimens, apart from alloy X200CrNiBMo10-4-3-3. The highest bending strength (1141 ± 109 MPa) and deflection of the specimens (2.58 ± 0.23%) was found in hypoeutectic steel X60CrB2-2 in the HIPed state. In general, the values of the bending strength and deflection of the hypereutectic steel specimens were low. The fracture surface of the hard alloys after the four-point bending test is depicted in Fig. 12. SLPS-densified alloy X60CrB2-2 is mainly characterized by cleavage fracture across the M2 B network structure (Fig. 12b). In contrast, HIP-densified X60CrB2-2 is characterized by a more ductile fracture with a honeycomb structure inside the metal matrix (Fig. 12a). The fracture surface of the hypereutectic steels in the SLPS state shows a brittle fracture across the coarse hard phases of type M3 (C,B), M7 C3 , and M2 B without a recognizable ductile fracture in the metal matrix (Fig. 12d and f). In the case of hard alloy X200CrNiBMo10-4-3-3 in the HIPed state, crack propagation occurred along the former powder particle surfaces, which is why separate particles can be recognized in the fractured surface (Fig. 12e). 4. Discussion 4.1. Powder compaction by SLPS processing SLPS requires the formation of a sufficient amount of liquid phase in order to reach a dense microstructure [23]. If the sintering temperature is too low, the amount of liquid phase is insufficient and partial powder compaction takes place instead. In contrast, excessively high temperatures promote distortion and strong growth of carbides and grains that have a negative impact on the material’s strength. Furthermore, a wide solidification interval is required to reach a reproducible and stable process with regard to temperature fluctuations that affect the amount of
Fig. 10. Tempering curves of the hypereutectic steels X200CrNiBMo10-4-3-3 and X360NiBCrCu4-2 as well as the hypoeutectic steel X60CrB2-2 austenitized at different temperatures.
518
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
Fig. 11. Abrasive wear resistance of the investigated steels against coarse abrasives (flint and corundum) in the (a) SLPS state and the (b) HIPed state.
liquid phase formed. Enough liquid phase must be formed to ensure sufficient decomposition of the powder particles (fast compaction) and complete wettability of the decomposed particles. Therefore, knowledge of the material-specific TSOL and the solidification path constitute key parameters for achieving fast and full powder compaction along with high mechanical and tribological properties. The first liquid phase forms above TSOL . The low-melting eutectic was found to be M3 (C,B) and ␥-Fe in all alloys. As the temperature increases, the evolution of the formed liquid phase depends on the chemical composition of the hard alloys, which affects both the type and the melting sequence of the phases. The type of eutectic formed depends on the boron/carbon ratio as well as additional alloying elements such as Cr, Mo, Mn, and Si. The influence of metalloids and their stabilizing effect on Fe-rich borides and carboborides can be described by the boron/carbon ratio [B/(B + C)]. If a low B/(B + C) ratio is present in the steel, a M3 (C,B)-type phase is stabilized. Exceeding a B/(B + C) ratio of 0.5–0.65, stabilizes the -phase (M23 (C,B)6 ) and carboborides of type M3 (C,B). The highest B/(B + C) ratios (>0.7) lead to the formation of borides of type M2 B. In addition, alloying elements such as Cr and Mo have a stabilizing effect on the M2 B phase due to the formation of borides with a similar crystallographic structure. In contrast, the elements Si and Mn possess a high solubility in the M3 (C,B) phase and inhibit formation of M2 B. In the case of hypoeutectic alloy X60CrB2-2, chromium-iron-rich borides of type M2 B dissolve in the temperature range between 1096 and 1140 ◦ C. Above a temperature of 1140 ◦ C, the liquid phase is only provided by the ␥-Fe phase. Due to the low carbon/boron ratio (B/(B + C) ∼ 0.4) of alloy X360NiBCrCu42, only carboborides of the type M3 (C,B) are stabilized. Therefore, the microstructure mainly consists of a M3 (C,B) eutectic that surrounds the primary M3 (C,B). The eutectic M3 (C,B) starts to melt above TSOL . At higher temperatures, primary M3 (C,B) melts, thus promoting the formation of the liquid phase. However, there is a narrow melting interval (TLIQ − TSOL = 47 K), which is attributed to the presence of only the hard phase of type M3 (C,B). The hard alloy X220CrNiBMo10-4-3-3 possesses a high boron/carbon ratio of 0.63 and a large amount of hard-phase-forming elements such as Cr and Mo. The high boron/carbon ratio in combination with a high chromium content stabilizes the Cr–Fe-rich phases M2 B and
M7 C3 . The low-melting eutectic was found to be boron cementite along with a Mo-rich M3 B2 eutectic. The larger amount of different hard phases leads to a wider melting interval of TLIQ − TSOL = 79 ◦ C. This more distinct melting interval increases the reproducibility of the sintering process and thus there are smaller differences in the amount of liquid fraction as a result of temperature fluctuations. Owing to measuring errors when using thermocouples, an almost constant amount of liquid phase must be formed over a wide range of temperatures. The amount of formed liquid phase should be selected in dependence on strong capillary forces, decomposition of the powder particles, and prevention of material distortion by viscous material flow. On the basis of these requirements, appropriate steels for SLPS-densification are hypoeutectic alloys with a sufficient amount of low-melting eutectics such as Fe-rich M3 (C,B), M2 B, and Cr-rich M7 C3 . In hypoeutectic steels, the amount of formed liquid phase can be easily modified by adapting the amount of low-melting eutectics by increasing the content of carbon and boron metalloids. In the case of hypereutectic alloys, the required wide solidification interval can be expanded by adapting the alloy composition by means of a successively built-up solidification path of different phases that melt at different temperatures. In the case presented, steel X200CrNiBMo10-4-3-3 features a wider solidification path than steel X360NiBCrCu4-2. This means that greater temperature fluctuations can be tolerated. As shown in Table 3, optimized sintering temperatures for alloy X60CrB2-2 were found to be about 80 ◦ C above TSOL . The required higher temperatures can be explained by the dendritic microstructure. A considerable amount of liquid phase is produced within the interdendritic spaces. The locally bound liquid phase within the interdendritic space does not support powder compaction. Thus a larger amount of liquid phase is required to obtain fully compacted specimens. As the temperature increases, the dendrites start to decompose, thus leading to a redistribution of the individual dendrite fragments. 4.2. Heat treatment The hardness, toughness, and content of retained austenite in the metal matrix influence the tribological and mechanical
Table 6 Bending strength B (MPa) and specimen deflection p (%) of hard-facing alloys X60CrB2-2, X200CrNiBMo10-4-3-3, and X360NiBCrCu4-2. Material
State HIP SLPS
X60CrB2-2
X200CrNIBMo10-4-3-3
X360NiBCrCu4-2
B (MPa)
p (%)
B (MPa)
p (%)
B (MPa)
p (%)
1142 ± 109 519 ± 24
2.58 ± 0.23 1.21 ± 0.01
219 ± 31 279 ± 14
0.50 ± 0.06 0.61 ± 0.08
381 ± 43 238 ± 14
0.91 ± 0.11 0.59 ± 0.04
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
519
Fig. 12. Influence of the processing route on the fracture behavior: (a) X60CrB2-2 HIP, (b) X60CrB2-2 SLPS, (c) X360NiBCrCu4-2 HIP, (d) X360NIBCrCu4-2 SLPS, (e) X200CrNiMoB10-4-3-3 HIP and (f) X200CrNIBMo10-4-3-3 SLPS.
properties. After SLPS- and HIP-densification, additional quenching and tempering are necessary to adjust the material’s properties to achieve a high hardness and a sufficient toughness and thus a high wear resistance. The highest post-quenching hardness of 950 ± 56 HV30 was measured in hard alloy X200CrNiBMo10-4-3-3. A lower hardness of 768 ± 34 HV30 was measured in alloy X360NiBCrCu4-2, although it had a higher metalloid content and therefore a higher hard phase content. The lower overall hardness can be explained by a higher hardness of the chromium- and ironrich hard phases of the type M7 C3 (1823 ± 40 HV0.05) and M2 B (1654 ± 73 HV0.05) in steel X220CrNiBMo10-4-3-3 compared to the M3 (C,B) phase (1044 ± 38 HV0.05), which was only present in steel X360NiBCrCu4-2. The lowest hardness was measured in hypoeutectic steel X60CrB2-2, irrespective of the chosen austenitizing and tempering temperatures. However, the hardness increased with increasing austenitizing temperature. This can be explained by the higher solute carbon content in the ␥-Fe phase. As shown in
Fig. 13, an insufficient amount of carbon (0.295 mass%) was present in the metal matrix for an austenitizing temperature of TAUS = 850 ◦ C. Increasing the austenitizing temperature raised the solute carbon content in the ␥-Fe phase (TAUS950 = 0.408 mass%; TAUS1050 = 0.540 mass%), as calculated by Calphad method. Thus an insufficient carbon content was present in the ␥-Fe phase for austenitizing temperatures less than TAUS < 1000 ◦ C. As a result, an austenitizing temperature of 1050 ◦ C was selected, which gave a high hardness of 684 ± 7 HV30. A comparison of the tempering curves of alloys X60CrB2-2 and X360NiBCrCu4-2 shows a stronger decrease in hardness for the hypoeutectic steel with increasing tempering temperature. The greater loss of hardness can be attributed to a lower volume fraction of hard phases, and the metal matrix becomes noticeably softer. To achieve the required hardness and sufficient toughness, low tempering temperatures of 200 ◦ C were determined to be the best compromise. Alloy X200CrNiBMo10-4-3-3 is characterized by an additional increase in hardness for a tempering temperature above 400 ◦ C.
520
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
for alloys X200CrNiBMo10-4-3-3 and X60CrB2-2. Sufficient wear resistance requires a minimum diameter of the hard phase, which should be equal to the scratch width of the acting abrasives. If not, the hard phase will be removed together with the metal matrix by a chipping mechanism. Therefore, the higher wear resistance of specimens in the SLPS state can be discussed in terms of a larger hard phase diameter. The coarser microstructure acts more effectively against larger abrasives and therefore prevents removal of the metal matrix and the hard phases by chipping. 4.4. Mechanical properties
Fig. 13. Solute carbon content in mass% in ␥-Fe as a function of the temperature.
This secondary peak hardness can be attributed to the alloying elements chromium and molybdenum, which promote precipitation of secondary carbide. In the case of X360NiBCrCu4-2, the hardness decreased with increasing tempering temperature due to the absence of secondary carbide-forming elements. 4.3. Tribological properties of hard-facing alloys Hard alloys are used to produce wear-resistant coatings on substrate materials. In the case presented, the investigated alloys contained a larger amount of hard phases finely distributed in the metal matrix, thus protecting the metal matrix against grooving, scratching, and chipping by the abrasives. Key factors influencing the resistance to abrasive wear are the type, size, and amounts of the hard phases. As shown in Fig. 11, the highest wear resistance against the Al2 O3 and SiO2 abrasives was exhibited by alloy X200CrNiBMo104-3-3 in both densification states. This behavior is attributed to the simultaneous presence of large volume fractions of M2 B and Cr7 C3 phases that have a high hardness. Both hard phases had a higher hardness than the SiO2 abrasive, which resulted in sufficient wear resistance. The hardness of the Al2 O3 abrasive meets or exceeds the value of the hard phases of type M2 B and M7 C3 . Thus, the materials showed a lower resistance to abrasive wear by Al2 O3 . In addition, not only the volume fraction of hard phases promotes sufficient wear resistance, as shown by the results of hard-facing alloy X360NiBCrCu4-2. This alloy contained the largest amounts of hard phases; however, their hardness was lower than that of the abrasives and thus insufficient, so that these hard phases M3 (C,B) were grooved and scratched by the abrasives. Otherwise, a high wear resistance was achieved with steel X360NiBCrCu42 in the HIPed state. This behavior can be attributed to the size and type of the hard phase. In the HIPed state, the microstructure of alloy X360NiBCrCu4-2 contained mainly primary and coarse M3 (C,B) that resisted ploughing and detachment. Hard-facing alloy X360NiBCrCu4-2 had a softer microstructure in the SLPS state as a result of a larger amount of fine eutectics. With the exceptions of the alloy X360NiBCrCu4-2, a higher abrasive wear resistance in the SLPS state compared to the HIPed state was measured. This is attributed to the smaller particle size of the hard phases dissolved in the metal matrix in the HIPed state
The results of the four-point bending test clearly show that the amount of hard phases, their distribution, size, and hardness affect the mechanical strength under a bending load. The highest bending strength was exhibited by hypoeutectic alloy X60CrB2-2 in the HIPed state, in which the hard phases possess a globular shape. The lower bending strength in the SLPS state can be attributed to a network-like structure of the hard phases of the type M2 B, and thus crack propagation is promoted by the brittle hard phases (Fig. 12b). In the case of HIP-densification, the globular shape of the hard phases results in longer distances between two adjacent hard phases and thus cracks are propagated through the more ductile metal matrix (Fig. 12a). In the case of the hypereutectic steels, lower bending strengths were measured due to the presence of a large amount of hard phases in the metal matrix. Furthermore, alloy X200CrNiBMo10-43-3 exhibited a lower bending strength in the HIPed state compared to the SPLS state. This is attributed to an inadequate densification state, and thus crack propagation mainly occurs along the former particle surfaces (Fig. 12e). Therefore, more pronounced diffusion processes are necessary to increase adhesion to the former gasatomized powder particles. Greater adhesion between the powder particles during HIP processing can be achieved using a higher temperature, pressure, or a longer processing time. However, the use of higher temperatures is limited due to the presence of the low-melting M3 (C,B)/M2 B eutectics in the ternary Fe–C–B system. Therefore, only higher pressures and a longer processing time are feasible. 5. Outlook This work has shown that the densification route influences the microstructure and therefore the mechanical and tribological properties of hard-facing alloys. The results of the four-point bending test, in particular, indicate a more ductile fracture of the hypoeutectic steels in the HIPed state. Compared to specimens with SLPS-densification, fracturing was shifted to higher bending stresses due to spheroidization of the hard phases, which is associated with an increase in the size of the mean free path across the metal matrix between two adjacent hard phases. Further research is focusing on the spheroidization of hard phases and the development of wear-resistant hypoeutectic steels with a high mechanical strength based on an alloying strategy and a material-adapted heat treatment. Spheroidization and a more uniform distribution of hard phases can be achieved by injecting the melt with small amounts of hard-phase-forming elements that act as crystallization nuclei for low-melting hard phases. Furthermore, the melting behavior of hard-facing alloys can be influenced by alloy designing to achieve a dense and fine-grained microstructure by means of SLPS-densification. It may be possible to increase spheroidization of the hard phases after SLPS-densification by high-temperature annealing. Preliminary tests confirm spheroidization of the hard phases with simultaneous development of a fine-grained microstructure after
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
521
Fig. 14. Microstructure of hard-facing alloy X60CrB2-2 in (a) SLPS state and (b) after annealing at 950 ◦ C for 4 h.
extended annealing at the austenitization temperature (Fig. 14). In addition, specimens can be quenched from the annealing temperature to achieve a martensitic microstructure. 6. Conclusion The present paper focuses on the microstructural evolution and the mechanical as well as tribological properties of three ultrahighboron, high-carbon steels that were densified by means of SLPS and HIP processing. The main results are: 1. Hypereutectic as well as hypoeutectic Fe–C–B steels can be fully densified by SLPS and HIP processing. 2. Sintering experiments at temperatures slightly higher than TSOL confirmed a large increase in density and hardness, which is attributed to the formation of a liquid phase. 3. SLPS processing produced specimens with a density close to the bulk value. The amount of liquid phase, which is necessary for fast densification without material distortion, was estimated by means of thermodynamic calculations and sintering experiments. Hypereutectic alloys exhibited a narrower interval between TSOL and TLIQ compared to hypoeutectic steel X60CrB2-2. 4. Sintering temperatures 50–70 ◦ C higher than TSOL must be chosen to achieve a dense microstructure and a high hardness, simultaneously. Differences between calculated and experimental TSOL and TLIQ were determined to be 30–100 ◦ C. 5. In contrast to literature data, a larger amount of liquid phase of 30–50 vol.% was required to reach fully compacted specimens. 6. Steel X200CrNiBMo10-4-3-3 exhibited a secondary hardness peak due to the presence of the elements chromium and molybdenum. In the case of steels X60CrB2-2 and X360NiBCrCu4-2, the hardness decreased on increasing the tempering temperature. 7. Specimens in the SLPS state had a higher wear resistance than HIP-densified specimens. This is attributed to a larger hard particle size that counteracts chipping and ploughing by abrasives more effectively.
8. The highest bending strength found for hypoeutectic steel X60CrB2-2 in the HIPed state. This is attributed to the globular shape of the hard phase of type M2 B. References [1] C. Xiang, L. Yanxiang, Materials Science and Engineering A 528 (2010) 770–775. [2] L. Zhongli, L. Ynxiang, C. Xiang, H. Kaihua, Materials Science and Engineering A 486 (2008) 112–116. [3] H. Berns, Wear 254 (1–2) (2003) 47–54. [4] A.H. Kasama, A.J. Mourisco, C.S. Kiminami, W.J. Botta Fo, C. Bolfarini, Materials Science and Engineering A 375–377 (2004) 589–594. [5] H. Berns, W. Theisen, Ferrous Materials – Steel and Cast Iron, Springer, Berlin, Heidelberg, 2008. [6] J.A. Jiminez, G. Frommeyer, P. Acosta, O.A. Ruano, Materials Science and Engineering A 202 (1995) 94–102. [7] P. Acosta, J.A. Jimenez, G. Frommeyer, O.A. Ruano, Materials Science and Engineering A 206 (1996) 194–200. [8] H.W. Jin, Y.M. Rhyim, S.G. Hong, C.G. Park, Materials Science and Engineering A 304–306 (2001) 1069–1074. [9] A. Saltykova, H. Berns, Materials Science Forum 672 (2011) 91–98. [10] H. Berns, A. Saltykova, A. Röttger, D. Heger, Steel Research International 82 (7) (2011) 786–794. [11] B. Rajasekaran, G. Mauer, R. Vaßen, A. Röttger, S.L. Weber, W. Theisen, Proceedings of International Thermal Spray Conference (ITSC-2010), May 3–5, Singapore, 2010. [12] T.M.T. Godfrey, A. Wisbey, P.S. Goodwin, K. Bagnall, C.M. Ward-Close, Materials Science and Engineering A 282 (2000) 240–250. [13] A. Röttger, Entwicklung neuer Schichtverbunde auf Fe-Basis gegen Abrasion, Ph.D. Thesis, Ruhr-Universität Bochum, Chairof Materials Technology, 2011. [14] T.B. Sercombe, G.B. Schaffer, Materials Science and Engineering A 528 (2010) 751–755. [15] K.S. Narasimhan, Materials Chemistry and Physics 67 (2001) 56–65. [16] R. Bollina, R.M. German, Proceedings of EURO PM 2004-Sintered Steels, 2004. [17] T.B. Sercombe, Materials Science A 363 (2003) 242–252. [18] A. Lal, R.G. Iacocca, R.M. German, Metallurgical and Materials Transaction A 30 (8) (1999) 2201–2208. [19] R. Bollina, S. Park, R.M. Germanr, Powder Metallurgy 53 (1) (2010) 20–26. [20] R.M. German, Metallurgical and Materials Transaction A 28 (7) (1997) 1553–1567. [21] S. Weber, W. Theisen, Powder Metallurgy 49 (4) (2006) 355–362. [22] S. Weber, Ph.D. Thesis, Ruhr-Universität Bochum, Institut für Werkstoffe, Fortschritts-Bericht VDI, Reihe 5, No. 720, VDI-Verlag, Düsseldorf, 2005. [23] C. Hu, T.N. Baker, Materials Science and Engineering A 190 (1995) 125–129.
Contents lists available at SciVerse ScienceDirect
Materials Science and Engineering A journal homepage: www.elsevier.com/locate/msea
Supersolidus liquid-phase sintering of ultrahigh-boron high-carbon steels for wear-protection applications A. Röttger a,∗ , S. Weber a,b , W. Theisen a a b
Lehrstuhl Werkstofftechnik, Ruhr-Universität Bochum, 44801 Bochum, Germany Helmholtz-Zentrum Berlin für Materialien und Energie GmbH, 14109 Berlin, Germany
a r t i c l e
i n f o
Article history: Received 15 September 2011 Received in revised form 27 October 2011 Accepted 29 October 2011 Available online 11 November 2011 Keywords: Boron Steels Wear resistance SLPS HIP
a b s t r a c t Powder metallurgy (PM) represents an alternative to conventional casting processes for the production of wear-resistant materials. PM hard alloys for wear-protection applications feature both higher strength and fracture toughness compared to cast hard alloys due to their more finely grained microstructure. However, densification by hot-isostatic pressing (HIP), the conventional PM-compaction method, is relatively expensive and thus partially counteracts low-cost processing. To increase the economic efficiency of the processing route, supersolidus liquid-phase sintering (SLPS) was investigated. In addition, expensive Ni- and Co-base hard alloys were substituted by boron-rich Fe-base hard-facing alloys. In this study, three ultrahigh-boron hard-facing alloy powders were densified by SLPS and HIP. The sintering temperatures were optimized by means of sintering experiments that were supported by thermodynamic calculations. Both densification states were investigated and compared with respect to the microstructure and the tribological and mechanical properties of the compacted hard-facing alloys. It was shown that the mechanical and tribological properties are strongly influenced by the microstructure. Although the microstructure is affected by the chemical composition, it can also be adapted by the densification process. SLPS-densified hard-facing alloys have a coarse microstructure that imparts not only a high wear resistance but also a detrimental effect on the mechanical properties. © 2011 Elsevier B.V. All rights reserved.
1. Introduction White cast irons are widely used for wear protection applications in the mining, mineral processing, and cement industries owing to their high hardness, high wear resistance, low price, and easy production [1–3]. These characteristics arise from a microstructure consisting of primary and/or eutectic hard phases that are finely distributed in an austenitic, pearlitic, or martensitic matrix [4,5]. White cast iron with added boron has recently been receiving more attention owing to the higher hardness of borides compared to carbides and the lowering effect of boron on the solidus temperature of steels. This means that steels containing sufficient boron can be processed at lower temperatures. One main attribute of boron-containing steels is the low solubility of boron in ␥-Fe and ␣-Fe and the formation of hard Fe-rich borides of the type M2 B, M3 (C,B), and M23 (C,B)6 [5–7]. As a result, it is not necessary to add hard-phase forming elements such as Mo, V, and Nb to obtain an Fe-base alloy with a large amount of hard phases.
∗ Corresponding author. Tel.: +49 0234 32 22366; fax: +49 0234 32 14104. E-mail addresses: [email protected] (A. Röttger), [email protected] (S. Weber), [email protected] (W. Theisen). 0921-5093/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.msea.2011.10.118
Only a few authors have dealt with boron additions higher than 1 mass%. This is due to the fact that an increasing boron content decreases the fracture toughness [8–11]. The microstructure of ultrahigh-boron, high-carbon alloys is characterized by primary solidified metal cells (hypoeutectic) or hard phases (hypereutectic) that are surrounded by a low-melting and almost brittle eutectic. This fine eutectic, in particular, promotes crack propagation because its fracture toughness is lower than that of the metal matrix, which results in insufficient mechanical properties. Godfrey et al. investigated Ti-base MMCs with added boron that were processed by hot-isostatic pressing. Their results indicated spheroidization of the hard phases during HIP treatment, which promotes an increase in mechanical strength [5,12,13]. Furthermore, boron addition inhibits strong coarsening of the microstructure due to precipitation of borides at the grain boundaries [12]. If a high wear resistance and good mechanical properties are required, the HIP process seems to be a promising processing route for boron-containing PM alloys. The addition of a small amount of boron to PM steels has been reported to have a major effect on the densification behavior during supersolidus liquid-phase sintering (SLPS) [14–16]. Sercombe investigated the effect of a small boron addition on the sintering behavior of a freeformed maraging PM steel. He showed that
512
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
the optimized sintering temperature (OST) could be decreased by increasing the boron content from 0 mass% to 0.4 mass% [17]. Boron forms a low-melting eutectic with Fe and therefore promotes the formation of a liquid phase at lower temperatures. However, boron forms stable phases with Mo, Ti, and Cr. This alters the phase composition of the steel matrix, which may have an effect on the mechanical and tribological properties after heat treatment. Based on the results of Sercombe and Lal et al., boron-containing PM steels are suitable materials for SLPS densification to achieve wearresistant materials with a low porosity [17–20]. The aim of the present paper is to study the densification behavior of three ultrahigh-boron, high-carbon gas-atomized Febase powders by means of supersolidus liquid-phase sintering and hot-isostatic pressing. The SLPS process was used to achieve fast densification due to the presence of low-melting, boron-rich eutectics. In addition, Weber et al. mentioned the possibility of influencing the size and shape of the hard phases in a ledeburitic cold-work tool steel by means of SLPS processing with the aim of adjusting the diameter of the hard phase to the wear system [21,22]. In contrast, spheroidization of hard phases due to a decrease in the surface energy during HIP is well known. It increases the mean free path of the metal matrix between two adjacent hard phases, which thus improves the mechanical properties. Therefore, this work investigated the influence of the densification process on the microstructure and on the mechanical and tribological properties. 2. Experimental procedure 2.1. Materials Three ultrahigh-boron, high-carbon steels were used as gas-atomized powders. Two of these were commercially available hypereutectic high-boron steels, X200CrNiBMo10-4-3-3 and X360NiBCrCu4-2, and the third was a self-developed hypoeutectic high-boron steel X60CrB2-2. The chemical composition of the bulk samples after HIP was measured by optical emission spark spectroscopy (QSG750 spectrometer OBLF) (Table 1). Prior to SLPS and HIP densification, the gas-atomized steel powders were analyzed with respect to their microstructure, particle size, and particle distribution by means of laser diffraction. 2.2. Thermodynamic calculations Determination of the optimized sintering and HIP temperatures was supported by thermodynamic equilibrium calculations using the Calphad method. Software package Thermo-Calc version R (Thermo-Calc AB, Stockholm, Sweden) and database TCFE6.2 were used to calculate the solidification path, phase diagrams, and the chemical composition of each phase in the equilibrium state. The calculations were performed with a phase set reduced to LIQUID, BCC A2, FCC A1, CEMENTITE, M7C3, M2B tetr, Cr2B orth, M23C6, and M6C. 2.3. Thermal analysis, sintering, and HIP Thermal analysis was performed to confirm the Thermo-Calc results concerning the sintering window. Differential thermal analysis (DTA) was performed by heating small amounts of gasatomized steel powder (∼100 mg) in alumina crucibles at a heating rate of 10 K min−1 in an argon gas atmosphere. The optimized sintering temperatures were investigated experimentally by sintering at different temperatures above the solidus temperature (TSOL ). The metal powder was filled into alumina preforms and precompacted to tap density. Sintering was performed in a batch furnace (∅ = 50 mm, L = 120 mm) under a vacuum (0.5 Pa)
with a temperature accuracy of ±5 K. The specimens were first heated to 300 ◦ C to degas the metal powder and to drive out residual vapor. They were then heated to the sintering temperature at a rate of 20 K min−1 . The isothermal holding time at the sintering temperature was 10 min for all specimens. SLPS was followed by a slow furnace cooling at a rate of approx. 15 K min−1 . After sintering at the different temperatures, the specimens were characterized by means of metallographic analysis, hardness measurements, and determination of the density by the Archimedes method. Besides SLPS-densification, the gas-atomized powders were compacted by hot-isostatic pressing (HIP) for 2 h in an argon atmosphere at a temperature of 1000 ◦ C and a pressure of 100 MPa. Ferritic steel capsules were filled with gas-atomized steel powder, which was then precompressed to tap density by vibration. The capsules were evacuated for 30 min and finally sealed by TIG welding.
2.4. Heat treatment To achieve a martensitic microstructure, the HIP- and SLPSdensified specimens were austenitized at 850–1050 ◦ C, quenched in water, and double-tempered in a temperature range from 100 to 600 ◦ C for 2 h.
2.5. Metallography and microscopy Microstructural examinations were carried out using optical and scanning electron microscopy (SEM). Specimens were sectioned, ground on abrasive paper, and polished with a 1 m diamond suspension. If necessary, the specimens were etched with 3% Nital. X-ray diffraction was used for phase analysis. Measurements were performed with a Siemens-D500 X-ray diffractometer using chromium K␣ radiation and a step size of 0.02◦ 2 in a range of 40–165◦ 2. The diffractograms were evaluated by the program ADM and the JCPDS database.
2.6. Hardness measurements and abrasive wear tests Macro- and microhardness measurements were carried out on the cross-section of the respective specimen using a Vickers indenter and a load of 294.3 N or 49.15 N. The wear resistance of the heat-treated specimens was investigated by a pin-on-paper test against the abrasives flint (SiO2 ) and corundum (Al2 O3 ). This type of test involves rotating a cylindrical specimen with a diameter of ∅ = 6 mm against an abrasive paper with an average particle size of 80 mesh (∼200 m) or 220 mesh (∼66 m). The dimensionless value of the wear resistance was calculated on the basis of the weight loss m of the specimen during testing, the specimen’s contact surface A, the coating density COAT , and the distance travelled L (Eq. (1)). −1 Wab =
A · L · COAT m
(1)
2.7. Four-point bending tests The mechanical properties were analyzed by four-point bending tests using specimens with a length of 70 mm and a cross-section of 5 mm × 5 mm. The testing method was based on DIN EN 53452 with a cross-head speed of 0.5 mm min−1 . The deflection of the specimens was measured by cross-head displacement and an extensometer.
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
513
Table 1 Chemical composition of the steel powders measured by optical emission spark spectrometry. Steel
C
B
Cr
Ni
Mo
Si
Mn
(P + S)
Cu
Fe
X200CrNiBMo10-4-2-2 X360NiBCrCu4-2 X60CrB2-2
2.12 3.60 0.56
3.54 2.40 1.73
11.96 1.20 2.02
3.95 4.30 0.21
3.28 – –
2.01 0.40 0.65
0.15 1.40 0.82
0.12 0.23 0.04
1.17 1.80 0.01
Bal. Bal. Bal.
3. Results 3.1. Powder characterization The microstructure of the steel powders in the gas-atomized state is shown in Fig. 1. The hypereutectic steels exhibit hard phases on the surfaces of the powder particles. The microstructure of the cross-section of steel powder X220CrNiBMo10-4-3-3 in the gasatomized condition features primary solidified Cr–Fe-rich M2 B and
Cr-rich M7 C3 in addition to Mo-rich ␥-Fe-M3 B2 and Fe-rich ␥-FeM3 (C,B) eutectics (Fig. 1a and b). The microstructure of carbon-rich hypereutectic steel X360NiBMoCr4-2 contains globular and primary solidified M3 (C,B), which are surrounded by a ␥-Fe-M3 (C,B) eutectic (Fig. 1c and d). In contrast, a dendritic microstructure on the surfaces of the powder particles can be seen for hypoeutectic steel X60CrB2-2 (Fig. 1e and f). The microstructure of the crosssection is characterized by ␥-Fe dendrites that are surrounded by a ␥-Fe-M2 B eutectic.
Fig. 1. Microstructure of the particle surface and of the cross-section of the steel particles: (a) and (b) steel X200CrNiBMo10-4-3-3; (c) and (d) steel X360NiBCrCu4-2; (e) and (f) steel X60CrB2-2.
514
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
Fig. 2. Sum distribution and probability density functions of steel powders X200CrNiBMo10-4-3-3, X360NiBCrCu4-2, and X60CrB2-2.
The steel powder particles were analyzed with regard to particle size and distribution using laser diffraction. The probability density functions of the steel powders are given in Fig. 2. The sum distribution functions of the steel powders are characterized by a multimodal function featuring two or three inflection points. Thus, several particle maxima are present. The median particle size of the hypereutectic steel powders is almost uniform and was measured to be x50,r = 70–71 m. In contrast, the median particle size of the hypoeutectic powder was found to be x50,r = 142 m. 3.2. Estimation of optimized sintering temperatures Thermodynamic calculations were performed to obtain an initial assessment regarding the sintering window. The melting sequence and the amount of liquid phase formed were calculated as a function of the temperature. However, these thermodynamic calculations describe the behavior in the equilibrium state, which is not the case in reality. DTA measurements were performed to estimate the differences between the calculated thermodynamic values and their experimental counterparts. TSOL and TLIQ were determined as the main signals, and the values were compared with the calculated results (Table 2). As shown in Table 2, thermodynamic calculations can be used for an initial estimation of the overall sintering interval. However, a higher solidus temperature was found by DTA measurements. The differences in temperature between the calculated and the experimental TSOL and TLIQ were 30–100 ◦ C. To obtain sintered specimens with a dense and fine-grained microstructure without a distinctive distortion, the optimized sintering parameters must be determined within a narrow temperature window. Therefore, sintering experiments were performed at different sintering temperatures. Furthermore, the sintered specimens were investigated with respect to their density and microstructure. curves for alloys X200CrNiBMo10-4-3-3, Sintering X360NiBCrCu4-2, and X60CrB2-2 are depicted in Figs. 3–5. The sintering curves confirm a minor densification of the powder particles at sintering temperatures below TSOL . The density of the sintered specimens correlates with the tap density of the powders before sintering. Thus the sintering mechanisms in the solid state led to slight skeletonization due to the formation of sintering bridges. A high degree of densification was not observed. On reaching TSOL , the density of the material rapidly increased. In this case, only the M3 (C,B) phase was observed in the microstructure of hard-facing alloy X360NiBCrCu4-2 (Fig. 3). As a result, eutectic M3 (C,B) generated a liquid phase, which surrounded the previously primary solidified hard phase M3 (C,B). The amount
Fig. 3. Sintering temperature versus density curve of steel X360NiBCrCu4-2.
of liquid phase formed by eutectic M3 (C,B) was calculated to be 22 vol.%. A further increase in sintering temperature led to complete melting of the primary M3 (C,B), which occurred within a T of 47 ◦ C Therefore, an intentional adjustment of the amount of liquid phase to approx. 30–40 vol.% cannot be achieved due to temperature fluctuations of ±8.5 ◦ C at the nominal temperature of 1140 ◦ C. Powder of X200CrNiBMo10-4-3-3 was completely densified by SLPS processing. Although the measured porosity after sintering at 1120 ◦ C was less than 1 vol.%, a higher sintering temperature of 1150 ◦ C was chosen to achieve a higher matrix hardness (Fig. 4). The sintering temperature versus density curve of steel X60CrB2-2 is characterized by a small gradient with an increase
Fig. 4. Sintering temperature versus density curve of steel X200CrNiBMo10-4-3-3.
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
515
Table 2 Calculated and experimental TSOL and TLIQ . Scheil calculations
Steel
TSOL (◦ C) X60CrB2-2 X220CrNiBMo10-4-3-3 X360NiBCrCu4-2
1092 1017 984
Equilibrium calculations
DTA measurements
TLIQ (◦ C)
TSOL (◦ C)
TLIQ (◦ C)
TSOL (◦ C)
TLIQ (◦ C)
1328 1223 1122
1095 1047 1053
1328 1223 1122
1143 1074 1112
– 1153 1177
Table 3 Optimized sintering parameters and hardness of the investigated steels. Steel
Optimized sintering temperature (OST) (◦ C)
Hardness (HV30)
Porosity (%)
X60CrB2-2 X360NiBCrCu4-2 X200CrNiBMo10-4-3-3
1220 1140 1150
635 ± 5 848 ± 18 919 ± 11
0.53 0.02 0.39
in density with increasing sintering temperature, although a low solidus temperature of TSOL = 1143 ◦ C was measured by DTA and the optimized sintering temperature was estimated to be TOST = 1220 ◦ C (Fig. 5). An overview of the optimized sintering parameters is given in Table 3. 3.3. Microstructural investigations The microstructure of the compacted steel powders in the SLPS, HIPed, and heat-treated state are depicted in Figs. 6–8. Phase analysis of the hard-facing alloys in the SLPS and heat-treated states was carried out by XRD measurements (Fig. 9). In the case of steel X200CrNiBMo10-4-3-3 in the SLPS state, the metal matrix contains mainly squared and lamellar hard particles of type M2 B and M7 C3 . In addition, EDX and X-ray diffraction analysis showed eutectic Mo-rich (M3 B2 ) and Fe-rich (M3 (C,B)) hard phases in an austenitic microstructure (Fig. 9b). The microstructure of steel X200CrNiBMo10-4-3-3 in the HIPed, quenched, and tempered state exhibited the same hard-phase composition in a martensitic–austenitic microstructure. The microstructure of the SLPS and HIPed state showed differences with respect to the shape
and size of the hard phases. The hard phases (M2 B, M7 C3 , M3 B2 and M3 (C,B)) in the HIPed state have a lamellar-to-globular shape and a higher amount of eutectic phases. In the case of steel X360NiBCrCu4-2 in the SLPS state, the lamellar-to-globular hard phases of type M3 (C,B) are distributed in the microstructure and surrounded by a fine lamellar boron cementite (M3 (C,B)) eutectic (Fig. 9c). The HIPed microstructure is characterized by smaller and globular M3 (C,B) hard-phase grains. Furthermore, eutectic boron cementite was detected at the triple points of grain boundaries of globular M3 (C,B). The hypoeutectic steel X60CrB2-2 in the SLPS and heat-treated state is characterized by globular metal cells that are surrounded by a Fe-rich M2 B network. XRD measurements indicate that the metal matrix has a martensitic microstructure (Fig. 9a). The microstructure in the HIPed and heat-treated state shows globular Fe-rich hard phases of type M2 B that are finely distributed in a mainly martensitic microstructure. Furthermore, a eutectic microstructure of the phases M23 (C,B)6 and ␣-Fe is observed. 3.4. Heat treatment The mechanical and tribological properties of Fe-base alloys can be adjusted by a heat treatment. In the case presented, quenching and tempering were performed to increase the strength and hardness of the metal matrix. Tempering curves of the hypereutectic steels are depicted in Fig. 10. Steel X200CrNiBMo104-3-3 possesses a high post-quenching hardness of 950 ± 56 HV30. Higher tempering temperatures decreased the hardness from 864 ± 45 HV30 at 100 ◦ C to 792 ± 10 HV30 at 400 ◦ C. The hardness increased again when the tempering temperature exceeded 400 ◦ C (peak of secondary hardness). Subsequently, SLPS as well as HIPed specimens were tempered at 520 ◦ C, at which temperature a secondary peak hardness of 859 ± 18 HV30 was measured. Steel X360NiBCrCu4-2 has a lower post-quenching hardness of 768 ± 34 HV30 compared to steel X200CrNiBMo10-4-3-3. Higher tempering temperatures continuously decreased the hardness of steel X360NiBCrCu4-2. The optimized tempering temperature was estimated to be 200 ◦ C. Hypoeutectic steel X60CrB2-2 was austenitized at TAUS = 850 ◦ C, 950 ◦ C, and 1050 ◦ C to investigate the influence of the solute carbon content in ␥-Fe on the post-quenching hardness. Increasing the austenitizing temperature increased the hardness (Fig. 10). In addition, the hardness of steel X60CrB2-2 decreased continuously with increasing tempering temperature. The hardness as well as the optimized heat treatment parameters for different densification treatments are summarized in Table 4. 3.5. Mechanical and tribological properties
Fig. 5. Sintering temperature versus density curve of steel X60CrB2-2.
3.5.1. Hardness measurements Owing to their high metalloid content in conjunction with a large amount of hard-phase-forming elements, the hard alloys investigated in this work are suitable for wear-protection applications. A high wear resistance requires a high matrix hardness as well as a large amount of hard phases dispersed in the metal
516
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
Fig. 6. Microstructure of steel X220CrNiBMo10-4-3-3 in the (a) SLPS and (b) HIPed state.
Fig. 7. Microstructure of steel X360NiBCrCu4-2 in the (a) SLPS and (b) HIPed state.
Fig. 8. Microstructure of steel X60CrB2-2 in the (a) SLPS and (b) HIPed state.
matrix. The hardness of the hard phases was measured by the Vickers indentation method, and the volume fraction of the hard phases was calculated by the Calphad method (Table 5). The chromiumrich carbide of type M7 C3 exhibited the highest hardness of
1823 ± 40 HV0.05 with respect to the solute chromium/iron ratio. The hardness of the M7 C3 phase decreased with increasing iron content. Furthermore, chromium-rich M2 B (HV0.05) exhibited a slightly lower hardness, which was estimated to be
Table 4 Hardness and optimized heat treatment parameters of the steels for different densification treatments. Material
Densification route
Heat treatment
Hardness (HV30)
X60CrB2-2 X60CrB2-2 X360NiBCrCu4-2 X360NiBCrCu4-2 X200CrNiBMo10-4-3-3 X200CrNiBMo10-4-3-3
HIP SLPS HIP SLPS HIP SLPS
TAUS = 1050 ◦ C, 2 × 2 h, 200 ◦ C TAUS = 1050 ◦ C, 2 × 2 h, 200 ◦ C TAUS = 1000 ◦ C, 2 × 2 h, 200 ◦ C TAUS = 1000 ◦ C, 2 × 2 h, 200 ◦ C TAUS = 1000 ◦ C, 2 × 2 h, 520 ◦ C TAUS = 1000 ◦ C, 2 × 2 h, 520 ◦ C
663 605 951 805 870 949
± ± ± ± ± ±
3 5 1 4 5 3
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
517
Table 5 Hardness and volume fraction (calculated by Calphad at T = 950 ◦ C) of the various phases in the investigated steels. Phase
M7 C3 M2 B M3 (C,B) M23 (C,B)6
X60CrB2-2
X360NiBCrCu4-2
X200CrNiBMo10-4-3-3
Vol. fraction
Hardness (HV0.05)
Vol. fraction
Hardness (HV0.05)
Vol. fraction
Hardness (HV0.05)
– 18.85 – 12.66
– 1654 ± 73 – 1275 ± 82
– 1.16 93.24 –
– – 1044 ± 38 –
13.25 43.25 11.59 –
1823 ± 40 1579 ± 81 905 ± 84 –
1654 ± 73 HV0.05 in steel X60CrB2-2 and 1579 ± 81 HV0.05 in steel X200CrNiBMo10-4-3-3. The lowest hardness of 905 ± 84 HV0.05 (eutectic) to 1044 ± 38 HV0.05 (primary) was measured in the boron cementite. 3.5.2. Abrasive wear tests The wear resistance of the steels was investigated by means of a pin-on-paper test using abrasives SiO2 (1000–1200 HV0.05) and Al2 O3 (2000–2250 HV0.05) with different particle sizes (220 mesh (66 m) and 80 mesh (200 m)). The results indicated a decrease in wear resistance with increasing abrasive particle size and abrasive hardness (Fig. 11). Abrasive Al2 O3 , in particular, had a higher hardness than the hard phases in the metal matrix. As a result, the abrasives are able to scratch and groove the hard particles. In the case of larger abrasive particle sizes, chipping and ploughing wear mechanisms became more active. In addition, the specimens in the SLPS state had a higher wear resistance than the HIP-densified specimens, except for steel X360NiBCrCu4-2. The highest wear resistance was exhibited by steel X200CrNiBMo10-4-3-3 in the SLPS and HIPed state. In the SLPS state, hypoeutectic steel X60CrB2-2 had a higher wear resistance compared to hypereutectic steel X360NiBCrCu4-2. In the HIPed state, steels X60CrB2-2 and X360NiBCrCu4-2 exhibited inverse behavior of the wear resistance. 3.5.3. Four-point bending test The bending strength of the hard-facing alloys in the quenched and tempered state was characterized by means of a four-point bending test. The results are listed in Table 6. The measured
Fig. 9. Phase identification by means of X-ray diffraction of steels (a) X60CrB22, (b) X200CrNiBMo10-4-3-3, and (c) X360NiBCrCu4-2 in the SLPS, quenched and tempered state.
bending strength was higher in the HIP-densified specimens, apart from alloy X200CrNiBMo10-4-3-3. The highest bending strength (1141 ± 109 MPa) and deflection of the specimens (2.58 ± 0.23%) was found in hypoeutectic steel X60CrB2-2 in the HIPed state. In general, the values of the bending strength and deflection of the hypereutectic steel specimens were low. The fracture surface of the hard alloys after the four-point bending test is depicted in Fig. 12. SLPS-densified alloy X60CrB2-2 is mainly characterized by cleavage fracture across the M2 B network structure (Fig. 12b). In contrast, HIP-densified X60CrB2-2 is characterized by a more ductile fracture with a honeycomb structure inside the metal matrix (Fig. 12a). The fracture surface of the hypereutectic steels in the SLPS state shows a brittle fracture across the coarse hard phases of type M3 (C,B), M7 C3 , and M2 B without a recognizable ductile fracture in the metal matrix (Fig. 12d and f). In the case of hard alloy X200CrNiBMo10-4-3-3 in the HIPed state, crack propagation occurred along the former powder particle surfaces, which is why separate particles can be recognized in the fractured surface (Fig. 12e). 4. Discussion 4.1. Powder compaction by SLPS processing SLPS requires the formation of a sufficient amount of liquid phase in order to reach a dense microstructure [23]. If the sintering temperature is too low, the amount of liquid phase is insufficient and partial powder compaction takes place instead. In contrast, excessively high temperatures promote distortion and strong growth of carbides and grains that have a negative impact on the material’s strength. Furthermore, a wide solidification interval is required to reach a reproducible and stable process with regard to temperature fluctuations that affect the amount of
Fig. 10. Tempering curves of the hypereutectic steels X200CrNiBMo10-4-3-3 and X360NiBCrCu4-2 as well as the hypoeutectic steel X60CrB2-2 austenitized at different temperatures.
518
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
Fig. 11. Abrasive wear resistance of the investigated steels against coarse abrasives (flint and corundum) in the (a) SLPS state and the (b) HIPed state.
liquid phase formed. Enough liquid phase must be formed to ensure sufficient decomposition of the powder particles (fast compaction) and complete wettability of the decomposed particles. Therefore, knowledge of the material-specific TSOL and the solidification path constitute key parameters for achieving fast and full powder compaction along with high mechanical and tribological properties. The first liquid phase forms above TSOL . The low-melting eutectic was found to be M3 (C,B) and ␥-Fe in all alloys. As the temperature increases, the evolution of the formed liquid phase depends on the chemical composition of the hard alloys, which affects both the type and the melting sequence of the phases. The type of eutectic formed depends on the boron/carbon ratio as well as additional alloying elements such as Cr, Mo, Mn, and Si. The influence of metalloids and their stabilizing effect on Fe-rich borides and carboborides can be described by the boron/carbon ratio [B/(B + C)]. If a low B/(B + C) ratio is present in the steel, a M3 (C,B)-type phase is stabilized. Exceeding a B/(B + C) ratio of 0.5–0.65, stabilizes the -phase (M23 (C,B)6 ) and carboborides of type M3 (C,B). The highest B/(B + C) ratios (>0.7) lead to the formation of borides of type M2 B. In addition, alloying elements such as Cr and Mo have a stabilizing effect on the M2 B phase due to the formation of borides with a similar crystallographic structure. In contrast, the elements Si and Mn possess a high solubility in the M3 (C,B) phase and inhibit formation of M2 B. In the case of hypoeutectic alloy X60CrB2-2, chromium-iron-rich borides of type M2 B dissolve in the temperature range between 1096 and 1140 ◦ C. Above a temperature of 1140 ◦ C, the liquid phase is only provided by the ␥-Fe phase. Due to the low carbon/boron ratio (B/(B + C) ∼ 0.4) of alloy X360NiBCrCu42, only carboborides of the type M3 (C,B) are stabilized. Therefore, the microstructure mainly consists of a M3 (C,B) eutectic that surrounds the primary M3 (C,B). The eutectic M3 (C,B) starts to melt above TSOL . At higher temperatures, primary M3 (C,B) melts, thus promoting the formation of the liquid phase. However, there is a narrow melting interval (TLIQ − TSOL = 47 K), which is attributed to the presence of only the hard phase of type M3 (C,B). The hard alloy X220CrNiBMo10-4-3-3 possesses a high boron/carbon ratio of 0.63 and a large amount of hard-phase-forming elements such as Cr and Mo. The high boron/carbon ratio in combination with a high chromium content stabilizes the Cr–Fe-rich phases M2 B and
M7 C3 . The low-melting eutectic was found to be boron cementite along with a Mo-rich M3 B2 eutectic. The larger amount of different hard phases leads to a wider melting interval of TLIQ − TSOL = 79 ◦ C. This more distinct melting interval increases the reproducibility of the sintering process and thus there are smaller differences in the amount of liquid fraction as a result of temperature fluctuations. Owing to measuring errors when using thermocouples, an almost constant amount of liquid phase must be formed over a wide range of temperatures. The amount of formed liquid phase should be selected in dependence on strong capillary forces, decomposition of the powder particles, and prevention of material distortion by viscous material flow. On the basis of these requirements, appropriate steels for SLPS-densification are hypoeutectic alloys with a sufficient amount of low-melting eutectics such as Fe-rich M3 (C,B), M2 B, and Cr-rich M7 C3 . In hypoeutectic steels, the amount of formed liquid phase can be easily modified by adapting the amount of low-melting eutectics by increasing the content of carbon and boron metalloids. In the case of hypereutectic alloys, the required wide solidification interval can be expanded by adapting the alloy composition by means of a successively built-up solidification path of different phases that melt at different temperatures. In the case presented, steel X200CrNiBMo10-4-3-3 features a wider solidification path than steel X360NiBCrCu4-2. This means that greater temperature fluctuations can be tolerated. As shown in Table 3, optimized sintering temperatures for alloy X60CrB2-2 were found to be about 80 ◦ C above TSOL . The required higher temperatures can be explained by the dendritic microstructure. A considerable amount of liquid phase is produced within the interdendritic spaces. The locally bound liquid phase within the interdendritic space does not support powder compaction. Thus a larger amount of liquid phase is required to obtain fully compacted specimens. As the temperature increases, the dendrites start to decompose, thus leading to a redistribution of the individual dendrite fragments. 4.2. Heat treatment The hardness, toughness, and content of retained austenite in the metal matrix influence the tribological and mechanical
Table 6 Bending strength B (MPa) and specimen deflection p (%) of hard-facing alloys X60CrB2-2, X200CrNiBMo10-4-3-3, and X360NiBCrCu4-2. Material
State HIP SLPS
X60CrB2-2
X200CrNIBMo10-4-3-3
X360NiBCrCu4-2
B (MPa)
p (%)
B (MPa)
p (%)
B (MPa)
p (%)
1142 ± 109 519 ± 24
2.58 ± 0.23 1.21 ± 0.01
219 ± 31 279 ± 14
0.50 ± 0.06 0.61 ± 0.08
381 ± 43 238 ± 14
0.91 ± 0.11 0.59 ± 0.04
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
519
Fig. 12. Influence of the processing route on the fracture behavior: (a) X60CrB2-2 HIP, (b) X60CrB2-2 SLPS, (c) X360NiBCrCu4-2 HIP, (d) X360NIBCrCu4-2 SLPS, (e) X200CrNiMoB10-4-3-3 HIP and (f) X200CrNIBMo10-4-3-3 SLPS.
properties. After SLPS- and HIP-densification, additional quenching and tempering are necessary to adjust the material’s properties to achieve a high hardness and a sufficient toughness and thus a high wear resistance. The highest post-quenching hardness of 950 ± 56 HV30 was measured in hard alloy X200CrNiBMo10-4-3-3. A lower hardness of 768 ± 34 HV30 was measured in alloy X360NiBCrCu4-2, although it had a higher metalloid content and therefore a higher hard phase content. The lower overall hardness can be explained by a higher hardness of the chromium- and ironrich hard phases of the type M7 C3 (1823 ± 40 HV0.05) and M2 B (1654 ± 73 HV0.05) in steel X220CrNiBMo10-4-3-3 compared to the M3 (C,B) phase (1044 ± 38 HV0.05), which was only present in steel X360NiBCrCu4-2. The lowest hardness was measured in hypoeutectic steel X60CrB2-2, irrespective of the chosen austenitizing and tempering temperatures. However, the hardness increased with increasing austenitizing temperature. This can be explained by the higher solute carbon content in the ␥-Fe phase. As shown in
Fig. 13, an insufficient amount of carbon (0.295 mass%) was present in the metal matrix for an austenitizing temperature of TAUS = 850 ◦ C. Increasing the austenitizing temperature raised the solute carbon content in the ␥-Fe phase (TAUS950 = 0.408 mass%; TAUS1050 = 0.540 mass%), as calculated by Calphad method. Thus an insufficient carbon content was present in the ␥-Fe phase for austenitizing temperatures less than TAUS < 1000 ◦ C. As a result, an austenitizing temperature of 1050 ◦ C was selected, which gave a high hardness of 684 ± 7 HV30. A comparison of the tempering curves of alloys X60CrB2-2 and X360NiBCrCu4-2 shows a stronger decrease in hardness for the hypoeutectic steel with increasing tempering temperature. The greater loss of hardness can be attributed to a lower volume fraction of hard phases, and the metal matrix becomes noticeably softer. To achieve the required hardness and sufficient toughness, low tempering temperatures of 200 ◦ C were determined to be the best compromise. Alloy X200CrNiBMo10-4-3-3 is characterized by an additional increase in hardness for a tempering temperature above 400 ◦ C.
520
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
for alloys X200CrNiBMo10-4-3-3 and X60CrB2-2. Sufficient wear resistance requires a minimum diameter of the hard phase, which should be equal to the scratch width of the acting abrasives. If not, the hard phase will be removed together with the metal matrix by a chipping mechanism. Therefore, the higher wear resistance of specimens in the SLPS state can be discussed in terms of a larger hard phase diameter. The coarser microstructure acts more effectively against larger abrasives and therefore prevents removal of the metal matrix and the hard phases by chipping. 4.4. Mechanical properties
Fig. 13. Solute carbon content in mass% in ␥-Fe as a function of the temperature.
This secondary peak hardness can be attributed to the alloying elements chromium and molybdenum, which promote precipitation of secondary carbide. In the case of X360NiBCrCu4-2, the hardness decreased with increasing tempering temperature due to the absence of secondary carbide-forming elements. 4.3. Tribological properties of hard-facing alloys Hard alloys are used to produce wear-resistant coatings on substrate materials. In the case presented, the investigated alloys contained a larger amount of hard phases finely distributed in the metal matrix, thus protecting the metal matrix against grooving, scratching, and chipping by the abrasives. Key factors influencing the resistance to abrasive wear are the type, size, and amounts of the hard phases. As shown in Fig. 11, the highest wear resistance against the Al2 O3 and SiO2 abrasives was exhibited by alloy X200CrNiBMo104-3-3 in both densification states. This behavior is attributed to the simultaneous presence of large volume fractions of M2 B and Cr7 C3 phases that have a high hardness. Both hard phases had a higher hardness than the SiO2 abrasive, which resulted in sufficient wear resistance. The hardness of the Al2 O3 abrasive meets or exceeds the value of the hard phases of type M2 B and M7 C3 . Thus, the materials showed a lower resistance to abrasive wear by Al2 O3 . In addition, not only the volume fraction of hard phases promotes sufficient wear resistance, as shown by the results of hard-facing alloy X360NiBCrCu4-2. This alloy contained the largest amounts of hard phases; however, their hardness was lower than that of the abrasives and thus insufficient, so that these hard phases M3 (C,B) were grooved and scratched by the abrasives. Otherwise, a high wear resistance was achieved with steel X360NiBCrCu42 in the HIPed state. This behavior can be attributed to the size and type of the hard phase. In the HIPed state, the microstructure of alloy X360NiBCrCu4-2 contained mainly primary and coarse M3 (C,B) that resisted ploughing and detachment. Hard-facing alloy X360NiBCrCu4-2 had a softer microstructure in the SLPS state as a result of a larger amount of fine eutectics. With the exceptions of the alloy X360NiBCrCu4-2, a higher abrasive wear resistance in the SLPS state compared to the HIPed state was measured. This is attributed to the smaller particle size of the hard phases dissolved in the metal matrix in the HIPed state
The results of the four-point bending test clearly show that the amount of hard phases, their distribution, size, and hardness affect the mechanical strength under a bending load. The highest bending strength was exhibited by hypoeutectic alloy X60CrB2-2 in the HIPed state, in which the hard phases possess a globular shape. The lower bending strength in the SLPS state can be attributed to a network-like structure of the hard phases of the type M2 B, and thus crack propagation is promoted by the brittle hard phases (Fig. 12b). In the case of HIP-densification, the globular shape of the hard phases results in longer distances between two adjacent hard phases and thus cracks are propagated through the more ductile metal matrix (Fig. 12a). In the case of the hypereutectic steels, lower bending strengths were measured due to the presence of a large amount of hard phases in the metal matrix. Furthermore, alloy X200CrNiBMo10-43-3 exhibited a lower bending strength in the HIPed state compared to the SPLS state. This is attributed to an inadequate densification state, and thus crack propagation mainly occurs along the former particle surfaces (Fig. 12e). Therefore, more pronounced diffusion processes are necessary to increase adhesion to the former gasatomized powder particles. Greater adhesion between the powder particles during HIP processing can be achieved using a higher temperature, pressure, or a longer processing time. However, the use of higher temperatures is limited due to the presence of the low-melting M3 (C,B)/M2 B eutectics in the ternary Fe–C–B system. Therefore, only higher pressures and a longer processing time are feasible. 5. Outlook This work has shown that the densification route influences the microstructure and therefore the mechanical and tribological properties of hard-facing alloys. The results of the four-point bending test, in particular, indicate a more ductile fracture of the hypoeutectic steels in the HIPed state. Compared to specimens with SLPS-densification, fracturing was shifted to higher bending stresses due to spheroidization of the hard phases, which is associated with an increase in the size of the mean free path across the metal matrix between two adjacent hard phases. Further research is focusing on the spheroidization of hard phases and the development of wear-resistant hypoeutectic steels with a high mechanical strength based on an alloying strategy and a material-adapted heat treatment. Spheroidization and a more uniform distribution of hard phases can be achieved by injecting the melt with small amounts of hard-phase-forming elements that act as crystallization nuclei for low-melting hard phases. Furthermore, the melting behavior of hard-facing alloys can be influenced by alloy designing to achieve a dense and fine-grained microstructure by means of SLPS-densification. It may be possible to increase spheroidization of the hard phases after SLPS-densification by high-temperature annealing. Preliminary tests confirm spheroidization of the hard phases with simultaneous development of a fine-grained microstructure after
A. Röttger et al. / Materials Science and Engineering A 532 (2012) 511–521
521
Fig. 14. Microstructure of hard-facing alloy X60CrB2-2 in (a) SLPS state and (b) after annealing at 950 ◦ C for 4 h.
extended annealing at the austenitization temperature (Fig. 14). In addition, specimens can be quenched from the annealing temperature to achieve a martensitic microstructure. 6. Conclusion The present paper focuses on the microstructural evolution and the mechanical as well as tribological properties of three ultrahighboron, high-carbon steels that were densified by means of SLPS and HIP processing. The main results are: 1. Hypereutectic as well as hypoeutectic Fe–C–B steels can be fully densified by SLPS and HIP processing. 2. Sintering experiments at temperatures slightly higher than TSOL confirmed a large increase in density and hardness, which is attributed to the formation of a liquid phase. 3. SLPS processing produced specimens with a density close to the bulk value. The amount of liquid phase, which is necessary for fast densification without material distortion, was estimated by means of thermodynamic calculations and sintering experiments. Hypereutectic alloys exhibited a narrower interval between TSOL and TLIQ compared to hypoeutectic steel X60CrB2-2. 4. Sintering temperatures 50–70 ◦ C higher than TSOL must be chosen to achieve a dense microstructure and a high hardness, simultaneously. Differences between calculated and experimental TSOL and TLIQ were determined to be 30–100 ◦ C. 5. In contrast to literature data, a larger amount of liquid phase of 30–50 vol.% was required to reach fully compacted specimens. 6. Steel X200CrNiBMo10-4-3-3 exhibited a secondary hardness peak due to the presence of the elements chromium and molybdenum. In the case of steels X60CrB2-2 and X360NiBCrCu4-2, the hardness decreased on increasing the tempering temperature. 7. Specimens in the SLPS state had a higher wear resistance than HIP-densified specimens. This is attributed to a larger hard particle size that counteracts chipping and ploughing by abrasives more effectively.
8. The highest bending strength found for hypoeutectic steel X60CrB2-2 in the HIPed state. This is attributed to the globular shape of the hard phase of type M2 B. References [1] C. Xiang, L. Yanxiang, Materials Science and Engineering A 528 (2010) 770–775. [2] L. Zhongli, L. Ynxiang, C. Xiang, H. Kaihua, Materials Science and Engineering A 486 (2008) 112–116. [3] H. Berns, Wear 254 (1–2) (2003) 47–54. [4] A.H. Kasama, A.J. Mourisco, C.S. Kiminami, W.J. Botta Fo, C. Bolfarini, Materials Science and Engineering A 375–377 (2004) 589–594. [5] H. Berns, W. Theisen, Ferrous Materials – Steel and Cast Iron, Springer, Berlin, Heidelberg, 2008. [6] J.A. Jiminez, G. Frommeyer, P. Acosta, O.A. Ruano, Materials Science and Engineering A 202 (1995) 94–102. [7] P. Acosta, J.A. Jimenez, G. Frommeyer, O.A. Ruano, Materials Science and Engineering A 206 (1996) 194–200. [8] H.W. Jin, Y.M. Rhyim, S.G. Hong, C.G. Park, Materials Science and Engineering A 304–306 (2001) 1069–1074. [9] A. Saltykova, H. Berns, Materials Science Forum 672 (2011) 91–98. [10] H. Berns, A. Saltykova, A. Röttger, D. Heger, Steel Research International 82 (7) (2011) 786–794. [11] B. Rajasekaran, G. Mauer, R. Vaßen, A. Röttger, S.L. Weber, W. Theisen, Proceedings of International Thermal Spray Conference (ITSC-2010), May 3–5, Singapore, 2010. [12] T.M.T. Godfrey, A. Wisbey, P.S. Goodwin, K. Bagnall, C.M. Ward-Close, Materials Science and Engineering A 282 (2000) 240–250. [13] A. Röttger, Entwicklung neuer Schichtverbunde auf Fe-Basis gegen Abrasion, Ph.D. Thesis, Ruhr-Universität Bochum, Chairof Materials Technology, 2011. [14] T.B. Sercombe, G.B. Schaffer, Materials Science and Engineering A 528 (2010) 751–755. [15] K.S. Narasimhan, Materials Chemistry and Physics 67 (2001) 56–65. [16] R. Bollina, R.M. German, Proceedings of EURO PM 2004-Sintered Steels, 2004. [17] T.B. Sercombe, Materials Science A 363 (2003) 242–252. [18] A. Lal, R.G. Iacocca, R.M. German, Metallurgical and Materials Transaction A 30 (8) (1999) 2201–2208. [19] R. Bollina, S. Park, R.M. Germanr, Powder Metallurgy 53 (1) (2010) 20–26. [20] R.M. German, Metallurgical and Materials Transaction A 28 (7) (1997) 1553–1567. [21] S. Weber, W. Theisen, Powder Metallurgy 49 (4) (2006) 355–362. [22] S. Weber, Ph.D. Thesis, Ruhr-Universität Bochum, Institut für Werkstoffe, Fortschritts-Bericht VDI, Reihe 5, No. 720, VDI-Verlag, Düsseldorf, 2005. [23] C. Hu, T.N. Baker, Materials Science and Engineering A 190 (1995) 125–129.