The effects of additive elements on the sliding wear behavior of Fe-base hardfacing alloys

The effects of additive elements on the sliding wear behavior of Fe-base hardfacing alloys

Wear 255 (2003) 481–488 The effects of additive elements on the sliding wear behavior of Fe-base hardfacing alloys Kown-yeong Lee a,∗ , Sung-hoon Lee...

531KB Sizes 1 Downloads 44 Views

Wear 255 (2003) 481–488

The effects of additive elements on the sliding wear behavior of Fe-base hardfacing alloys Kown-yeong Lee a,∗ , Sung-hoon Lee a , Yangdo Kim b , Hyun Seon Hong a , Young-min Oh a , Seon-jin Kim a b

a Division of Materials Science and Engineering, Hanyang University, Seoul 133-791, South Korea Division of Materials Science and Engineering, Pusan National University, Pusan 609-735, South Korea

Abstract The present research is aiming at the development of new Fe-base wear resistance alloy, which has been expected to the substitute for Stellite 6 in nuclear power industry. The effects of Cr and C addition on the wear resistance were investigated. The sliding wear tests were performed in air at the temperature range 25–300 ◦ C, and also in pressurized water at 25–250 ◦ C. The contact stress was 103 MPa in accordance with current nuclear power plant component designs. The composition of Fe-base alloy was optimized to improve the wear resistance. The sliding wear behaviors of the optimized Fe-base alloy and Stellite 6 were compared in the present study. The weight loss of new Fe-base alloy after 100 cycles sliding wear test was found to be smaller than that of Stellite 6. © 2003 Published by Elsevier Science B.V. Keywords: Hardfacing alloy; Sliding wear; Strain-induced phase transformation

1. Introduction Co-base Stellite alloys have been used as hardfacing materials for nuclear power plant valves due to their superior corrosion and wear resistance under sliding conditions [1]. However, the need to avoid the use of Stellite alloy has emerged since it is the main source of Co, which is the largest contributor to the occupational radiation exposure [2]. As a most effective way to reduce Co contaminations, many Co-free hardfacing alloys, such as Fe-base and Ni-base alloys, have been investigated to replace Stellite [3–5]. Most of the Fe-base hardfacing alloys developed to replace Stellite are austenitic stainless steels containing a large volume fraction of eutectic or non-eutectic carbides [5]. Among their micro-structural features, the austenitic matrix is primary importance under sliding conditions because the galling resistance is generally dependent on the matrix structure [6]. Ohriner et al. [3] reported that the wear resistance of the hardfacing alloy, NOREM 02, is based on the low stacking fault energy (SFE) of the matrix. The low SFE suppresses cross-slip of dislocations, resulting in an increased work hardening rate [1,6]. Thus, it prevents severe plastic deformation at asperity contacts which increases the resistance to galling. In general, however, the SFE increases with ∗

Corresponding author.

0043-1648/03/$ – see front matter © 2003 Published by Elsevier Science B.V. doi:10.1016/S0043-1648(03)00155-8

temperature [7–10]. As the SFE increases, the distance of separation of two partial dislocations decreases and the recombination of partial dislocations necessary for cross-slip becomes easier. This implies that the wear resistance of alloys based on a low SFE can be lowered significantly as the temperature increases. In case of the NOREM alloy, the wear losses increased with temperature and galling occurred above 200 ◦ C. The decrease of the wear resistance of NOREM alloy at about 200 ◦ C has been reported to be caused by the absence of strain induced martensitic transformation above Md temperature [11]. Wear resistance of Fe-base hardfacing alloy can be improved by not only reducing SFE of matrix, but also providing lots of carbides and enhancing the strain induced martensitic transformation. Especially, strain induced martensitic transformation has been reported to be an important wear resisting mechanism in an iron-base hardfacing alloy below the Md temperature [3,11], which is the maximum strain induced martensitic transformation occurrence temperature [10]. So, wear surface is hardened and repressed to be deformed plastically. As a result, the occurrence of galling can be prohibited. Therefore, excellent wear resistance can be achieved by raising the Md temperature above its work temperature. The sliding wear resistance under severe conditions is also required to replace Stellite that is used for nuclear power plant valves seats. The maximum stress at the normal

482

K.-y. Lee et al. / Wear 255 (2003) 481–488

operating condition is 103 MPa and the present experiment was conducted under this stress level [2], however, it can be exceeded by any unstable conditions. This study dealt with the effect of Cr, C, which can affect the formation of austenite phase and wear resistance. The new Fe-base alloy was compared with Stellite 6 under several conditions to decide the optimized composition and the possibility for commercial applications.

2. Experimental procedure 2.1. Specimen Alloy blocks were arc melted under argon atmosphere using a water-cooled copper crucible and a non-consumable tungsten electrode. The iron used for making alloys was 99.98% pure. Other alloying elements, such as Cr, Si and C, were added to from three master alloys of which chemical compositions were Fe–59Cr–0.03C, Fe–58.9Cr–6.36C and Fe–74.2Si. Every block was melted eight times, and inverted between meltings to ensure homogeneity. The chemical composition of alloys was measured using an ARD 3460 optical emission spectrometer, and their nominal chemical compositions are summarized in Table 1. Stellite 6, used as a reference material, was deposited on a 12 mm thick AISI 304 plate by gas tungsten arc welding (GTAW). The GTAW condition has been reported elsewhere [12]. After the arc melting and deposition, specimens were machined by wire cutting into wear test specimens as presented in Fig. 1. Table 1 Nominal chemical and measured compositions of alloys (wt.%) Alloys

Element

Fe–16Cr–1.3C–1Si Fe–20Cr–1.3C–1Si Fe–24Cr–1.3C–1Si Fe–20Cr–0.6C–1Si Fe–20Cr–1.7C–1Si Fe–20Cr–2.0C–1Si Fe–20Cr–1.3C–0Si Fe–20Cr–1.3C–2Si

Fe

Cr

C

Si

P

S

bal. bal. bal. bal. bal. bal. bal. bal.

15.65 19.91 23.88 19.99 20.11 19.71 18.93 20.02

1.302 1.288 1.295 0.608 1.686 2.01 1.287 1.304

0.940 1.001 0.997 0.964 0.986 1.006 0.001 1.985

0.005 0.007 0.008 0.007 0.008 0.006 0.006 0.007

0.008 0.009 0.009 0.005 0.012 0.011 0.008 0.009

Fig. 2. Apparatus for sliding wear test in pressurized water.

Wear test surfaces were polished to a roughness value Ra < 0.02 ␮m by grinding with 2000 grit SiC abrasive paper. The hardfacing thickness of as-weld specimens was about 3.5–4 mm and it was reduced to about 2–2.5 mm in wear test specimens after final grinding. 2.2. Sliding wear test The block-on-block type sliding friction machine supplied by Plint & Partners Ltd. was used for these high load sliding tests. The self-mated tests were performed in air at the temperature range from room temperature to 300 ◦ C under an applied normal contact stress of 103 MPa (15 ksi). The sliding wear test was performed in pressurized water at 250 ◦ C using autoclave. Autoclave wear test machine is shown Fig. 2. The sliding speed and stroke were 3 mm/s and 9 mm, respectively. The total weight losses of the moving disc and the fixed plate were measured after 100 cycles of sliding. More than three tests were performed to obtain reliable data. 2.3. Examination of the surfaces The microstructure of the surfaces was examined using scanning electron microscope (SEM), energy dispersive spectroscopy (EDS), X-ray diffractometer (XRD) and transmission electron microscope (TEM). Microhardness of the matrix below the worn surfaces were measured to evaluate the work hardening. Microhardness measurements were conducted more than five times at each specimen by a Vickers microindentation hardness tester under a load of 25 g. 3. Results and discussion

Fig. 1. Geometry of sliding wear test specimens.

We have made austenitic Fe-base hardfacing alloys whose wear resistance is caused by strain induced martensitic

K.-y. Lee et al. / Wear 255 (2003) 481–488

483

Fig. 3. Schaeffler diagram.

transformation (γ → α or ε). Austenite phase formation can be predicted by Schaeffler diagram [13]. The diagram in Fig. 3 shows the effects of alloy element on the matrix structures in Cr–Ni stainless steel. The x and y axis are equivalents of Cr and Ni, respectively. As shown in the Schaeffler diagram, the alloys were expected to have austenite phase. C was added as austenite stabilizer. Cr and Si were added for corrosion resistance and fluidity, respectively. XRD patterns of new Fe-base alloys are presented in Fig. 4. The XRD peaks of austenite phase were observed in every new Fe-base alloy except Fe–24Cr–1.3C–1Si. The Schaeffler diagram was constructed based on the statistical values of phase stability within a limited composition range. Therefore, additional alloying elements might alter its boundaries [14]. So, Fe–24Cr–1.3C–1Si alloy sufficiently differs in composition from the Schaeffler diagram. Fig. 5 shows the microstructure of new Fe-base alloys. The new Fe-base alloys showed a typical microstructure of hypoeutectic system, which has austenitic dendrite and Cr7 C3 /austenite lamella structures. The lamella structure was believed to be formed after the formation of primary austenite [15]. The ratio of primary austenite phase to eutectic phase is presented in Table 2. The wear losses of the present alloys as a function of temperature are shown in Fig. 6. Fig. 6a shows the effect of Cr addition on the wear resistance of Fe–xCr–1.3C–1Si (x = 16, 20, 24 wt.%) alloys. The wear resistance of Fe–16Cr–1.3C–1Si and Fe–20Cr–1.3C–1Si specimens was superior compare to the Fe–24Cr–1.3C–1Si specimen.

However, the wear resistance of Fe–16Cr–1.3C–1Si and Fe–20Cr–1.3C–1Si specimens were not sufficient enough to replace Stellite. To investigate the effect of C, specimens having various carbon amounts were also examined as shown in Fig. 6b. Wear losses of Fe–20Cr–1.7C–1Si and Fe–20Cr–2.0C–1Si were found to be small. Si additions did not have any significant effect on the wear resistance as shown in Fig. 6c. If Cr and C contents increase higher than eutectic point, hypereutectic structure is formed after formation of primary phase. Hypereutectic structure has a lot of carbides and coarse microstructures. It shows different microstructure Table 2 The ratio of primary austenite phase to eutectic phase of new Fe-base alloy with variations of (a) Cr, (b) C and (c) Si additions Alloys

Primary austenite

Eutectic phase

(a) Fe–xCr–1.3C–1Si 16Cr 20Cr 24Cr

78 68 –

22 32 –

(b) Fe–20Cr–yC–1Si 0.6C 1.3C 1.7C 2.0C

91 68 64 55

9 32 36 45

(c) Fe–20Cr–1.3C–zSi 0Si 71 1Si 68 2Si 65

29 32 35

484

K.-y. Lee et al. / Wear 255 (2003) 481–488

Fig. 4. XRD patterns of new Fe-base alloys with variations of (a) Cr, (b) C and Si additions.

Fig. 5. Optical micrographs of alloys.

K.-y. Lee et al. / Wear 255 (2003) 481–488

485

Fig. 6. Wear losses of new Fe-base alloys with variations of (a) Cr, (b) C and (c) Si additions tested in air up to 300 ◦ C.

depend on the cooling rates and poor welding characteristics [1]. In hypereutectic structures, if C content is insufficient, C is depleted in the matrix and ferrite can be formed. The work hardening of ferrite matrix is less effective than that of austenite. Severe adhesive wear can occur with ferrite in the matrix. Therefore, 16–20 wt.% Cr and 1.7–2.0 wt.% C appear to be an optimum composition for the improved wear resistance. The wear losses of the Fe–20Cr–1.3C–1Si, Fe–20Cr– 1.7C–1Si and Fe–20Cr–2.0C–1Si alloys were compared with Stellite 6 by sliding wear test. More than three replicate tests were performed on these alloys and the wear losses are presented in Fig. 7. The wear losses of Fe–20Cr–1.7C–1Si and Fe–20Cr–2.0C–1Si were smaller than that of Stellite at the temperature 300 ◦ C. It is concerned about working condition that demands corrosion resistance. So, Fe–20Cr–1.7C–1Si, which has lower C content, is suitable to the nuclear power plants valve. Fig. 8 shows XRD patterns analyzed at wear surface of the Fe–20Cr–1.7C–1Si

Fig. 7. Wear losses of new Fe-base alloy and Stellite 6 tested in air up to 300 ◦ C.

486

K.-y. Lee et al. / Wear 255 (2003) 481–488

Fig. 8. XRD patterns of worn surface of new Fe-base alloy tested in air up to 300 ◦ C.

alloy. The strain induced martensitic transformation was occurred after wear tests as shown in Fig. 8. SEM micrographs of the worn surfaces of Fe–20Cr–1.7C– 1Si alloy tested at various temperatures are shown in Fig. 9. Large portion of worn surfaces was found to be smooth and no scratch mark was observed. Therefore, the major wear mechanism is considered to be the mild adhesive wear. Fur-

thermore, some adhered layers were found on the worn surfaces tested above 200 ◦ C as indicated by arrows in Fig. 9c and d. EDS spectra of worn surfaces tested at 300 ◦ C are presented in Fig. 9e and f. A significant amount of oxygen was detected in the adhered layer. Thus, the adhered layers were considered to be the oxide layers formed on the sliding surface from compaction of wear debris particles that are partially or completely oxidized under the sliding action [16]. The oxide layers prevent the sliding surfaces from direct metal-to-metal contact, and consequently inhibit adhesive wear [17]. In order to investigate the effect of work hardening and strain induced phase transformation, the microhardness variation beneath the worn surfaces of Fe–20Cr–1.7C–1Si and Stellite specimens were measured after sliding test at various temperatures as shown in Fig. 10. The worn surfaces of Fe-base alloy and Stellite were highly hardened up to more than 600, 500 HV and the hardening depth was extended to about 70, 50 ␮m, respectively. Therefore Fe-base alloy was more highly and deeply hardened than Stellite 6. Strain induced martensitic transformation was also confirmed from TEM observations as shown in Fig. 11. The α martensite was observed at the worn surfaces tested at the room temperature. TEM micrograph of the specimen before wear test showed no strain induced martensitic transformation. Therefore, external stress induced martensitic

Fig. 9. SEM micrographs and EDS spectrums of worn surface of new Fe-base alloy tested in air up to 300 ◦ C.

K.-y. Lee et al. / Wear 255 (2003) 481–488

Fig. 10. Microhardness variations beneath the worn surface of (a) new Fe-base alloy and (b) Stellite 6 tested in air under 103 MPa.

Fig. 11. TEM micrographs of worn surface of new Fe-base alloy.

487

488

K.-y. Lee et al. / Wear 255 (2003) 481–488

4. The wear resistance characteristic of the Fe–20Cr–1.7C– 1Si alloy was similar to that of Stellite 6 in water up to 250 ◦ C.

Acknowledgements The authors wish to acknowledge to be financial support from the 2002 Research Program of the Innovative Technology Center for Radiation Safety (ITRS) at Hanyang University. References Fig. 12. Wear losses of new Fe-base alloy and Stellite 6 tested in water up to 250 ◦ C.

transformation and affected the wear behavior. The wear resistance is proportioned to the hardness and the strain rate sensitivity in ferrite steels. However, in austenite steels, the wear resistance is related to twin, SFE, and phase transformations. If the initial SFE is lowered, the wear resistance is increased. The γ → α phase transformation improved the wear resistance more than the predicted value by mechanical deformation. In other words, Martensite, formed by phase transformation is nucleated in cross-sections of stacking faults, has high work hardening rate and statistic elongation to be maintained at the high stress [18]. Accordingly, the improvement of wear resistance can be achieved. The wear tests were also conducted in pressurized water at 250 ◦ C to simulate the operating conditions for nuclear power plant. The results are shown in Fig. 12. Analogous to the tests in air, the Fe–20Cr–1.7C–1Si showed similar wear losses to Stellite 6. The study is a part of the author’s continuing researches on the development at improved hardfacing alloys for the use in the nuclear reactors, more theoretical analysis and experimental work on cavitation erosion resistance, corrosion resistance and welding characteristics are considered to be needed. 4. Conclusions In order to develop Fe-base hardfacing alloy as a substitute for the conventional hardfacing alloys, Stellite, the wear tests were conducted and the following results were obtained. 1. The best wear-resistant composition was found to be Fe–20Cr–1.7C–1Si. 2. The high wear resistance of the Fe–20Cr–1.7C–1Si specimen is believed mainly due to the strain induced martensitic transformation. 3. The matrix hardening due to the strain induced martensitic transformation was extended to about 70 ␮m.

[1] K.C. Antony, Wear resistant cobalt-base alloys, J. Metal (1983) 52–60. [2] H. Ocken, Reducing the cobalt inventory in light water reactors, Nuclear Tech. 68 (1985) 18–28. [3] E.K. Ohriner, T. Wada, E.P. Whelan, H. Ocken, The chemistry and structure of wear-resistant, iron-base hardfacing alloys, Metal. Trans. A 22A (1991) 983–991. [4] J. Vikstrom, Galling resistant of hardfacing alloys replacing Stellite, Wear 179 (1994) 143–146. [5] H. Ocken, The galling wear resistance of new iron-base hardfacing alloys: a comparison with established cobalt- and nickel-base alloys, Surface Coating Tech. 76–77 (1995) 456–461. [6] S. Atamert, J. Stekly, Microstructure, wear resistance, and stability of cobalt based and alternative iron based hardfacing alloys, Surface Eng. 9 (1993) 231–240. [7] P.C.J. Gallagher, The influence of alloying, temperature, and related effects on the stacking fault energy, Metal. Trans. 1 (1970) 2429– 2461. [8] R.M. Latanision, A.W. Ruff Jr., The temperature dependence of stacking fault energy in Fe–Cr–Ni alloys, Metal. Trans. 2 (1971) 505–509. [9] F. Lecroisey, A. Pineau, Martensitic transformations induced by plastic deformation in the Fe–Ni–Cr–C system, Metal. Trans. 3 (1972) 387–396. [10] F. Abrassart, Stress-induced γ → α martensitic transformation in two carbon stainless steels. Application to TRIP steels, Metal. Trans. 4 (1973) 2205–2216. [11] J.K. Kim, S.J. Kim, The temperature dependence of the wear resistance of iron-base NOREM 02 hardfacing alloy, Wear (2000) 217–222. [12] EPRI TR-105816, NOREM Applications Guidelines, Electric Power Research Institute, 1995. [13] Sir R. Honeycombe, H.K.D.H. Bhadeshia, Steels: Microstructure and Properties, 2nd ed., Edward Arnold, 1995. [14] V. Raghavan, D.P. Antia, The chromium equivalents of selected elements in austenitic stainless steels, Metal. Trans. 25A (1994) 2675–2681. [15] L.E. Svensson, B. Gretoft, B. Ulander, H.K.D.H. Bhadeshia, Fe–Cr–C hardfacing alloys for high-temperature applications, J. Mater. Sci. 21 (1986) 1015. [16] H. So, The mechanism of oxidational wear, Wear 184 (1995) 161– 167. [17] J. Jiang, F.H. Stott, M.M. Stack, A mathematical model for sliding wear of metals at elevated temperatures, Wear 181–183 (1995) 20– 31. [18] C.J. Heathcock, B.E. Protheroe, Cavitation erosion of stainless steels, Wear 81 (1982) 311.