The galling wear resistance of new iron-base hardfacing alloys: a comparison with established cobalt- and nickel-base alloys

The galling wear resistance of new iron-base hardfacing alloys: a comparison with established cobalt- and nickel-base alloys

XiiliFACE COATIN6S ELSEVIER 1"[6#110101 Surfaceand Coatings Technology76-77 (1995)456-461 The galling wear resistance of new iron-base hardfacing ...

482KB Sizes 8 Downloads 470 Views

XiiliFACE

COATIN6S

ELSEVIER

1"[6#110101 Surfaceand Coatings Technology76-77 (1995)456-461

The galling wear resistance of new iron-base hardfacing alloys: a comparison with established cobalt- and nickel-base alloys H. O c k e n Electric Power Research Institute, PO Box 10412, Palo Alto, CA 94303, USA

Abstract

Iron-base wear-resistant alloys, designated [email protected], were developed to address concerns unique to the nuclear power industry. Laboratory evaluations of the galling wear resistance of candidate NOREM compositions and other iron- and nickelbase alloys were performed using a pin-on-plate specimen geometry, and these results were compared with those obtained from the long-established cobalt-base [email protected] alloys. The hardfacing alloy was typically deposited as powder using gas tungsten arc welding or as wire using automatic gas tungsten arc welding. Attempts were made to correlate the results from the galling wear tests to factors such as alloy compositions, microstructure and deposited hardness. The galling wear resistance of NOREM and other iron-base alloys matched or exceeded that of the cobalt-base standard, but the factors responsible for good resistance to galling wear could not be identified. Nickel-base alloys typically showed much higher values of surface damage in these tests. Keywords: Wear-resistant alloys; Galling wear

1. Introduction

The high and highly varying price of cobalt, the key constituent of many wear-resistant alloys, has intermittently provided an incentive to develop similar alloys that use less costly metals such as iron or nickel. Another drawback of using cobalt is encountered in the nuclear industry where the largest contributor to occupational radiation exposure of plant maintenance personnel is the radioisotope 6°Co. This concern led the Electric Power Research Institute (EPRI) to support a comprehensive program to develop an iron-base alloy that could serve as a generic substitute for the cobalt-base hardfacing alloys. The need for such a program was strengthened by results from sliding wear tests performed on hardfacing alloys available in the early 1980s; their performance was clearly inferior to the cobalt-base [email protected] [' 1"1. The primary need was for a high performance hardfacing alloy that could be used in valves, and the key attribute desired was outstanding resistance to adhesive or galling wear. Budinski ['21 describes such wear as being due to localized bonding between contacting solid surfaces, which leads to material loss or material transfer between two surfaces. This is the type of wear damage that can occur when valve components, such as discs and seats, make contact. The EPRI program led ElsevierScienceS.A.

to the iron-base hardfacing alloys, designated [email protected] Laboratory evaluations of cast products showed some compositions with the desired attributes [31, and production of welding consumables and the development of welding procedures followed. Seats and discs of small gate valves were hardfaced with three galling-resistant iron-base hardfacing alloys, and another was hardfaced with a cobalt base standard. These valves were subjected to stroke cycling under conditions simulating those found in operating reactors. These tests confirmed the outstanding performance of both NOREM and Everit 50 [4,51. Subsequent studies aimed to improve the "weldability" of NOREM consumables needed in the field [61. Success here has spurred utility acceptance of NOREM, which now has been used in valves installed in about a dozen US commercial nuclear reactors. Over the years this program has evaluated the galling wear of a significant number of NOREM variants and other commercially available and experimental hardfacing alloys. This information is compiled and reviewed in this paper. This assessment permits one to comment on the role of primary alloying elements, deposition technique, alloy microstructure, and hardness on galling wear resistance. Results from some of the more recent galling wear tests have been reported by Vikstr6m [7].

H. Ocken/Smface and Coatings Technology 76-77 (J995) 456-461

2. Experiments

457

brushed to remove any loose debris. Surface profilometry was used to measure the average peak-to-valley surface roughness. The surface profile was measured on the plate specimen in directions both parallel and perpendicular to the original grinding marks. The reported surface damage value is the difference between the post-test and pre-test values. Hardness values were determined using either the 10 kgf Vickers scale or the Rockwell C scale. Microstructures were determined by optical microscopy of polished and etched specimens. Nominal compositions of the NOREM alloys that were investigated are given in Table 1. Compositions of the iron-base alloys, nickel-base alloys, and cobalt-base alloys are given in Tables 2, 3 and 4.

Galling wear was measured using a [email protected] specimen geometry. Pins and plates were hardfaced with the same alloy, typically the most demanding test condition. Although the earliest wear measurements on NOREM alloys were performed on castings, deposition of most of the hardfacing alloys investigated in this program typically was effected by welding. Weld consumables in the form of powder were deposited by plasma transfer arc welding (PTAW); wire was deposited by gas tungsten arc welding (GTAW). A few specimens evaluated were prepared by laser deposition or from powder consolidated by hot isostatic pressing (HIP). After deposition, wear test surfaces were finished to a roughness value Ra < 0.4 gm, measured normal to the grinding direction. This finish was obtained by grinding with 320 grit SiC abrasive paper using water as a lubricant. Testing was conducted by loading the flat surface of the pin against the plate using a modified Brinell hardness tester. The applied loads were 9800, 19600 and 29 400 N, which corresponds to normal applied stresses of 140, 275 and 415 MPa (20, 40 and 60 klbf in-2). The pin specimen was rotated about its vertical axis back and forth ten times thorough 120 ° against the plate. Duplicate tests were performed at each applied stress. The test environments were air and deionized water at ambient temperature. After the test the plate was lightly

Table 3 Nominal compositions of nickel-base hardfacing alloys Alloy

Amount (wt.%) Ni

Deloro 40 Deloro 50 Deloro 25~ Haynes 711 TribaloyT700

Fe

Balance 2.5 Balance 4 Balance 0.3 Balance 23 Balance

C

Cr

0.45 0.60 0.04 2.7 0.08

10 13

Si

Mo

2.3 4 2.4 27 1 15.5 3.4

Other

2.5 B 3B 20 1.3 B 8 12 Co, 3 W 32.5 < 3 C o + F e

a Mixture of Deloro 25 and Mo powder. Table 1 Compositions of NOREM iron-base hardfacing alloys Heat

01 04 M2 A 02

Table 4 Nominal compositions of cobalt-base hardfacing alIoys

Amount (wt.%) Fe

C

Mn

Cr

Si

Ni

Mo

N

Balance Balance Balance Balance Balance

1.18 1.20 0.90 1.25 1.25

9.89 10.94 4.99 5.5 4.5

24.99 24.98 24.61 26.0 25.0

3.55 4.60 2.96 3.3 3.3

4.38 8.07 8.0 5.0 4.0

1.56 2.03 2.06 2.0 2.0

0.11 0.24 0.i4 0.16 0.16

AiIoy

Amount (wt.%)

Stellite 6 Stellite 156 Stellite 21

Co

Fe

Ni

Balance Balance Balance

5 0.4 0.75 3 2.8

C

Cr

1.i 28 1.6 28 0.25 27

Si

Mo

1.2

1 5

Table 2 Nominal compositions of other iron-base hardfacing alloys Alloy

Cenium Z20 Everit 50 Everit 50 So Antinit Dur 300 EB 5183 Nelsit Elmax APM 2311 Tristelle 1 Tristelle 2

Amount (wt.%) Fe

Ni

C

Cr

Si

Balance Balance Balance Balance Balance Balance Balance Balance Balance Balance

18

0.3 2.5 2.0 0.12 2 0.03 1.7 2.0 1 2

27 25 25 21.0 20 18 17 26 30 35

<0.5 0.4 5.0 4.5 7 0.4 i 5 5

8.0 9 10

10 10

Mo 9 3.2 3.5

Mn

<1.0 0.9 6.5

Other 2W 0.5 V 0.5 V 7 Nb

3 1

2 0.3 0.5

3.0 V 12 Co 12 Co

Mn

W

2 1

4 4.6

~ Ocken/Surface and Coatings Technology 76-77 (1995) 456-461

458

3. Results

4. Discussion

Results from the galling wear tests are provided in Tables 5-8, which also indicate the technique used to deposit the hardfacing alloy. The duplicate entries for Stellites 6 and 21 were obtained from specimens prepared by different organizations and where different organizations performed the galling wear tests. These data suggest that the reported surface damage measurements are reproducible to within _+0.5 gm. The microstructures of these alloys as determined by optical microscopy are indicated in Table 9, together with representative hardness values of the deposit before wear testing.

As expected, the measured surface damage in the galling wear tests increases with increasing applied stress. For a given applied stress, damage is less in tests run in deionized water compared with tests performed in air, suggesting that the water serves as a lubricant to reduce damage. The question of how well these tests reflect expected performance in the field is problematic. The highest applied stress of 415 MPa (60 klbf in-z) is higher than calculated average contact stresses between components such as a valve disc and seat, but edge loading effects can produce higher local contact stresses in valves. Also, the sliding speed obtained by rotating the pin over

Table 5 Galling wear of NOREM iron-base hardfacing alloys Alloy

Surface damage (lain) at the folIowing stresses Tests in air

NOREM NOREM NOREM NOREM NOREM NOREM NOREM NOREM NOREM

01, cast 04, cast 01, PTAW 04, PTAW 01, HIP 04, HIP M2, GTAW A, PTAW 02, GTAW

Tests in water

140 MPa 20 klbf in -z

275 MPa 40 kIbf in -2

415 MPa 60 kIbf in -2

140 MPa 20 klbf in -2

275 MPa 40 klbf in -2

415 MPa 60 klbf in -2

2.0 1.2 1.1 2.3 0.2 0.9 1.0 0.5 0.5

7.2 1.8 1.7 2.3 0.3 1.3 1.1 0.9 1.4

20.5 4.2 4.i 3.1 0.6 1.4 1.2 0.5 2.0

0.9 0.5 0.5 1.0 0.1 0.4 NT NT 0.3

2.8 0.7 0.7 0.9 0.1 0.5 NT NT 1.0

7.3 1.5 1.5 1.1 0.2 0.5 NT NT 1.3

NT, not tested.

Table 6 Galling wear of other iron-base alloys Alloy

Surface damage (Nn) at the following stresses Tests in air

Cenium Z 20, PTAW Everit 50, GTAW Everit 50, So, GTAW Antinit Dur 300, GTAW EB 5183, PTAW NeIsit, PTAW Elmax, PTAW Tristelle 1 Tristelle 2 Haynes 711 APM 2311, HIP NT, not tested.

Tests in water

140 MPa 20 klbf in-2

275 MPa 40 klbf in-2

415 MPa 60 klbf in -z

140 MPa 20 klbf in -2

275 MPa 40 klbf in -2

415 MPa 60 klbf in-2

2.7 0.4 1.1 1.7 1.4 0.5 0.0 2.2 0.8 59 0.1

11.9 0.5 1.7 7.2 2.9 0.8 0.2 1.8 0.3 7O 0.5

17.9 0.9 1.4 10.1 i8.4 0.8 0.5 7.0 2.2 69 0.4

2.4 0.2 NT 1.6 0.5 NT NT 0.9 0.0 1.6 NT

3.5 0.4 NT 1.8 0.7 NT NT 1.9 2.8 42 NT

5.8 0.4 NT 6.0 1,0 NT NT 2.3 1.1 45 NT

459

H O&en/Surface and Coatings Technology 76-77 (!995) 456-461

Table 7 Galling wear of nickel-base alloys Alloy

Surface damage (gin) at the followingstresses Tests in air

Deloro 40, GTAW Deloro 50, GTAW Deloro 25, laser welding Tribaloy T700, PTAW

Tests in water

140 MPa 20 klbf in -2

275 MPa 40 klbf in -2

415 MPa 60 klbf in -2

140 MPa 20 klbf in -2

275 MPa 40 klbf in -2

415 MPa 60 klbf in -2

37.1 38.6 19.5 23

41.3 NT 16.6 39

50 76 25.6 27

16.9 17.5 NT 12.8

15.9 NT NT 11.8

17.9 27.1 NT 11.7

NT, not tested. Table 8 Galling wear of cobalt-base alloys Alloy

Surface damage (gm) at the followingstresses Tests in air

Stellite 6, PTAW Stellite 6, PTAW Stellite 156 PTAW Stellite 156, PTAW+laser Stellite 21, GTAW Stellite 21, GTAW

Tests in water

140 MPa 20 "Idbfin -2

275 MPa 40 klbf in -z

415 MPa 60 klbf in -~

140 MPa 20 klbf in -2

275 MPa 40 klbf in -2

415 MPa 60 klbf in -a

1.9 2.2 NT 0.9 0.9 1.3

2.1 2.6 NT 1.3 1.7 1.9

2.0 2.8 2.9 2.0 1.2 2.4

NT 1.1 NT 0.3 0.4 0.5

NT 1.7 NT 1.0 0.9 1.0

NT 1.6 0.9 1.2 1.8 1.5

NT, not tested. the stationary plate is higher than that encountered in many field applications where hardfacing alloys are used to provide resistance to galling wear. These factors suggest that the conditions imposed in the laboratory tests are likely to be more demanding than those typically encountered in most field applications. Accordingly, the test results are best used as a means to provide information about the relative performance of different alloys. These results suggest than any number of iron-base alloys (Elmax, A P M 2311, N O R E M , Nelsit and Everit 50 and 50 So) match or exceed the performance of the long-established cobalt-base alloys. There is little incentive to use alloys with reduced cobalt contents, such as the Tristelles, in nuclear applications. Typically, the performance of nickel-base alloys was inferior to those of iron- and cobalt-base alloys. Haynes 711, which contains a modest amount of both cobalt and nickel, showed significant surface damage in the galling wear tests. In spite of the poorer performance of the nickel-base alloys in this test, the field performance of valves hardfaced with such alloys is generally acceptable. This suggests that the use of other alloys with extremely high galling wear resistance provides an added margin of

safety. Also, other factors that bear on the selection of hardfacing alloys for field applications, such as the ability to deposit sound defect-free welds and the corrosion resistance expected in service, should also be taken into consideration. Various techniques were used to fabricate N O R E M specimens (castings, PTAW and GTAW overlays, and HIP). Galling wear measurements on castings typically showed higher surface damage values than those on weld overlays of the same composition. This trend suggests that processing techniques yielding finer microstructures are those with the highest resistance to galling wear. However, there are limits to the improvement in galling wear resistance than can be achieved by forming finer microstructures. Laser melting of PTAW-deposited Stellite 156 resulted in a significant reduction in grain size and a more uniform distribution of alloying dements in the cobalt-rich matrix and the carbide precipitates, but the galling wear resistance of the laser-melted specimens was unchanged from the values measured on the as-deposited specimens [8]. The wide range of alloy microstructures and deposited hardness suggest that outstanding resistance to galling wear can be achieved by diverse routes. A number of the iron-base alloys with outstanding resistance to

460

H. Ocken/Surface and CoatOzgsTechnology 76-77 (1995) 456-461

Table 9 Hardfacing alloy microstructures and deposited hardnesses Alloy

Microstructure

hardness Rockwell C (HRC)

hardness Vickers (HV 10)

NOREM

Eutectic M7C3 and non-eutectic carbides (M6C and M3C) in an austenitic matrix Eutectic carbides and a G phase in an austenitic matrix Interdendritic Cr carbides in an austenitic matrix Auste±tic-ferritic duplex structure Hard interdendritic carbides and randomly dispersed globular carbides in a soft dendritic matrix Two lamellar Laves phases dispersed in a third phase V and Cr carbides in an austenitic matrix Finely distributed carbides in a tempered martensite matrix Carbides and borides dispersed in a solid solution strengthened matrix Two hard intermetallic Laves phases dispersed in a sorer solid solution matrix Cr borides in ferritic stainless steel matrix Borides and carbides in Ni-Cr matrix Cr carbides dispersed in a solid-solutionstrengthened austenitic matrix Cr carbides dispersed in a solid-solutionstrengthened austenitic matrix Cr carbides (M7C3) and tungsten carbides (M6C) dispersed in a solid-solutionstrengthened matrix

39-45

360

44

415

Cenium Z 20 Everit 50, 50 So Antinit DUR 300 EB 5183 Nelsit Elmax APM 2311 Deloro 25, 40, 50 Tribaloy 700 Fe-base Ni-base Haynes 711 Tristelle 1 and 2 Stellite 6 and 21

galling wear are characterized by solid-solutionstrengthened austenitic matrices and overall silicon additions of a few per cent. The availability of a large number of austenitic alloys with good resistance to galling wear marks a significant change since this program began. At that time, austenitic stainless steels typically did not exhibit high galling wear resistance; the sole exception was Nitronic 60 (Fe-18wt.%Cr-Swt.%Mn-4wt.%Si0.12wt.%N). Schumacher [-9] attributed the high galling resistance of this alloy to its low stacking fault energy, high work-hardening rate, and its ability to form lubricating oxide layers at the wear surface. Anthony [i0] and Bhansali and Miller [11] have discussed the factors that contribute to galling wear resistance of alloys that crystallize in f.c.c, crystal structures. Their view is that fracture at asperity junctions (i.e. local adhesion sites), rather than deformation, results in low friction and greater freedom from galling. Alloys in which the dislocation mobility is low and cleavage can occur should not gall; ductile alloys, where it is easy for dislocations to cross-slip over more than one plane, should be galling prone. In f.c.c, structures the stacking fault energy is considered to be a good indicator of cross-slip tendencies. High stacking fault energy indicates a low number of impeding stack faults and an increased tendency to cross-slip and gall. Thus low stacking fault

39 650 495 765 40

525

43

770

42

440

energy favors high resistance to galling wear. Bhansali and Miller showed that substituting nickel for cobalt in a cobalt-base alloy increased the stacking fault energy (as measured using X-ray diffraction techniques), reduced the strain hardening rate and lowered the stress at which dislocation cross-slip occurred, thereby increasing plasticity. Resistance to galling wear decreased as the nickel content increased. Some confirming evidence that outstanding resistance to galling wear is favored by the ability of contacting surfaces to fracture rather than deform plastically is suggested by Vikstrom's [12] investigation of post-test wear surfaces of galling-resistant iron-base alloys using optical microscopy and scanning electron microscopy (SEM). Using optical microscopy, the most gallingresistant alloys showed evidence only of burnishing where the surface was characterized by a smoother and shinier appearance than the original ground surface. SEM of the most heavily damaged surface regions showed evidence of small cracks in NOREM, Nelsit, Elmax, Everit 50 So and APM 2311. In addition to the outstanding galling wear resistance found in the NOREM alloys (Table 5), Table 6 shows that similar performance was found in some alloys that also contained modest amounts of silicon as an alloying constituent. The beneficial effects of silicon additions in

H. Ocken/Smface and Coat#zgs Technology 76-77 (1995) 456-461

improving the wear resistance of austenitic alloys is also suggested by Laroudie et al. [13] who found that the sliding wear resistance of AISI 316L stainless steel was improved by laser surface alloying, where Ti and SiC powders were dissolved in the melt.

46i

Acknowledgments The galling wear measurements reported here were performed by staff at the AMAX Research Center and at the EPRI Non-destructive Evaluation Center. Support for this program has been provided by the EPRI under RP 1935.

5. Conclusions

References

A number of iron-base alloys with resistance to galling wear matching or exceeding those of the long-established cobalt-base hardfacing alloys were found using the pinon-plate specimen geometry. Surface damage in these alloys and in cobalt-base standards as measured by profilometry was found to be minimal (less than about 3 ~tm) in the most demanding test condition of 415 MPa (60 k/bf in-2) in air. Some alloys showed surface damage values less than 1 ~tm under this condition. Most other nickel-base alloys showed damage values of about 50-75 gm at this condition. Because these alloys have been found to perform satisfactorily in valves installed in both fossil and nuclear power plants, these results suggests other factors, such as the ability to achieve defect-free weld overlays and corrosion resistance, should play a significant role in the choice of hardfacing alloys used in critical applications. Good resistance to galling wear was found in iron-base alloys with different microstructures and different deposited hardnesses. Many of the most galling-resistant iron-base alloys were found to consist of precipitated carbides and/or borides in an austenitic matrix, which contained modest additions of silicon as well as the major alloying constituents.

I-1] E.I. Landerman, D.J. Boes, P. Bowen and M.J. Huck, EvaIuation of low-cobalt alloys for hardfacing applications in nuclear components, Rep. EPRI NP-3446, August 1984 (Electric Power Research Institute). [2] K. Budinski, Swface Engineering for Wear Resistance, PrenticeHall, Englewood Cliffs, NJ, 1988, p. 30. [-3] E.K. Ohriner, T. Wada, E.P. Whelan and H. Ocken, in H.D. Merchant and K.J. Bhansali (eds.), Metal Transfer and Galling Wear in Metallic Systems, MetalIurgicaI Society of AIME, Warrendale, PA, 1987, p. 247. [4] E.V. Murphy and I. Inglis, Endurance tests of valves with cobalt-free hardfacing alloys: PWR phase final report, Rep. EPRI TR-I0060I, May 1992 (Electric Power Research Institute). [5] E.V. Murphy and I. Inglis, Endurance tests of valves with cobalt-free hardfacing alloys: BWR phase final report, Rep. EPRI TR-I01847, January 1993 (Electric Power Research Institute). [6] M.K. PhilIips and S. Findlan, Maintenance and Repair Welding in Power Plants, American Welding Society, Miami, FL, 1991, p. 30. [7] J. Vikstr6m, Wear, I79 (1994) 143. 1-8] S.C. Agarwal and H. Ocken, Wear, 140 (1990) 223. [-9] W. Schumacher, Mach. Des., 55 (i986) 87. [10] K.C. Anthony, J. Met., 35 (1983) 52. [11] K.J. Bhansali and A.E. Miller, Wear, 75 (1982) 241. [-i2] J. Vikstr6m, Galling wear of cobalt-free hardfacing aIIoys, Rep. EPRI TR-103845, May 1994 (Electric Power Research Institute). ['13] F. Laroudie, C. Tassin, M. Pons and L. Lelait, Proc. Int. Conf. on Metallurgical Coatings on Thin Films, I995, Paper E2.07.