Volume 12 - Fractography Part 12 docx

Sulfidation at the surface of the filler metal 65% Ni led to formation of the Ni 3 S 2 -Ni eutectic, which melts at 635 °C 1175 °F, well below the operating temperature of the coupling

Figure 817 shows a slow-bend fracture of a weld specimen made by gas tungsten-arc butt welding two 19-mm ( -in.) thick plates of 18% Ni maraging steel having a yield strength of 1724 MPa (250 ksi) A specimen 48 mm 2 × 180 mm (0.75 in 2 × 7 in.) was cut across the weld After the weld was ground flush, the specimen was notched and then fatigue-cracked in the center of the weld and parallel to the length

of the weld After being aged at 490 °C (915 °F), the specimen was broken at room temperature in slow, three-point bending Fracture surface shows steps caused by splitting A portion of one of these steps is shown at higher magnification in Fig 818; the surface here exhibits a network of high- reflectivity area, one example being identified by the arrows The remaining areas of the surface were relatively dull See also Fig 819, 820, 821, and 822 Fig 817: Actual size Fig 818: 8×

Figures 819 and 820 are TEM views, at different magnifications, taken from one of the reflectivity areas in Fig 818 The shiny aspect was due to flat platelets resulting from cleavage of particles of unidentified composition Note that in both views there are tear dimples between the reflecting facets Figure 821 reveals that the structure in the low-reflectivity area in this fracture surface consists of rather flat equiaxed dimples Figure 822 is from a companion 18% Ni maraging steel weld specimen to the one in Fig 817, 818, 819, 820, and 821; specimen here was hydrogen embrittled and then fractured by sustained loading at room temperature This fracture surface shows dimples, together with quasi-cleavage facets that contain characteristic steps, river patterns, and tear ridges Fig 819, 821, and 822: TEM p-c replicas, all at 2000× Fig 820: TEM p-c replica, 5000×

high-Failure due to sulfidation of a coupling for a hot-gas sampling line used in a coal-gasification pilot plant The coupling was made of the iron-nickel-base superalloy RA-330 (UNS N08330), joined using Inconel alloy 182 filler metal (AWS A5.11, class ENiCrFe-3) It failed after 8500 h of service at a temperature of 870 to 980 °C (1600 to 1800 °F) in gas having this typical composition (in vol%): 56.9%

H 2 O, 16.9% H 2 , 16.2% CO 2 , 8.0% CO, 1.9% CH 4 , and 0.1% H 2 S Higher H 2 S levels may have been present on occasion Sulfidation at the surface of the filler metal ( 65% Ni) led to formation of the

Ni 3 S 2 -Ni eutectic, which melts at 635 °C (1175 °F), well below the operating temperature of the coupling This liquid phase penetrated the interdendritic regions of the weld metal and then the grain boundaries of the base metal Fig 824: Backscattered electron image of ENiCrFe-3 weld metal

showing severe interdendritic attack SEM, 65× Fig 825: Sulfur map of field in Fig 824 reveals high level of the element in interdendritic regions Technique used was energy-dispersive x-ray spectroscopy (EDS) 65× Fig 826: Backscattered electron image of superalloy base metal adjacent to filler metal shows intergranular attack by Ni 3 S 2 -Ni (at B), most of which transformed to more stable chromium sulfide (at A) SEM, 130× Fig 827: Sulfur map of field in Fig 826 reveals high levels of the element in grain boundaries 130× (D.R Diercks, Argonne National Laboratory)

Hydrogen-embrittlement fracture of a single-cyrstal nickel-base superalloy (CMSX-2) developed for gas turbine engine applications Composition: 4.6% Co, 8.0% Cr, 0.6% Mo, 7.9% W, 5.6% Al, 0.9%

Ti, 5.8% Ta, <0.1% Hf, 0.01% Zr, 0.005% C, remainder Ni Heat-treated notched samples were stress relieved and then tensile tested in either helium (control samples) or hydrogen at 35 MPa (5 ksi) and room temperature Cross-head speed, 2.12 μ/s; notched strength ratio H 2 /He, 0.14 Fig 828: Fracture surface of control sample tested in helium 4.6× Fig 829: Higher-magnification view of fracture surface in Fig 828 Note ductile failure mode SEM, 95× Fig 830: Fracture surface of samples tested

in hydrogen 4.6× Fig 831: Higher-magnification view of fracture surface in Fig 830 Note brittle cyrstallographic cracking SEM, 95× (R.J Schwinghamer, NASA Marshall Space Flight Center)

Fatigue and creep fractures of case and hot isostatically pressed IN-738 tested in a salt environment (40% MgSO 4 , 59% Na 2 SO 4 , 1% NaCl) at 705 °C (1300 °F) Samples heat treated 2 h at 1120 °C (2050

°F), air cool; 14 h at 845 °C (1550 °F), air cool Fig 832: Fatigue fracture surface Test conditions: 483 MPa (70 ksi) ± 207 MPa (30 ksi); 102,000 cycles at failure Fractograph shows region of transgranular fracture initiation with characteristic "thumbprint" and river patterns pinpointing origin at surface (bottom) Failure was salt corrosion assisted SEM, 29× Fig 833: Fatigue fracture surface of salt- coated sample Test conditions: 483 MPa (70 ksi) ± 345 MPa (50 ksi); 5000 cycles at failure Fracture is transgranular with multiple surface origins SEM, 13× Fig 834: Fracture surface after stress-rupture testing at 896 MPa (130 ksi) Mechanism: salt corrosion-assisted intergranular fracture Note grain- boundary, decohesion SEM, 23× (E.A Schwarzkopf, J Stefani, and J.K Tien, Columbia University)

Figure 843 shows the surface of a dendritic stress-rupture fracture in a cast specimen of IN-100 base alloy that was annealed at 1175 °C (2150 °F) and loaded at 980 °C (1800 °F) to a tensile stress of

nickel-97 MPa (14 ksi) The specimen broke after 49 h of testing Figure 844 is a TEM p-c replica of an area

at the arrow in Fig 843, showing some evidence of intergranular separation and some facets of what appear to be cleaved particles (as at A) of intermetallic compounds Figure 845 is a TEM p-c replica of another area at the arrow in Fig 843, showing evidence of glide-plane decohesion (at A's) as well as smooth areas of stretching (as at B) Some surface oxidation apparently occurred after fracture Fig 843: 8× Fig 844 and 845: 6500×

∆

∆

Simultaneous metallographic-fractographic evaluation of an Inconel alloy 718 (UNS N07718) toughness test sample via selective-area electropolishing technique The nickel-base alloy was conventionally heat treated Fig 847: Schematic of selective-area electropolishing Use of method enables simultaneous study of both the fracture surface topography and the underlying microstructure Fig 848: Metallographic-fractographic profile (fracture surface at bottom left) SEM, 460× Fig 849: Higher-magnification view of field in Fig 848 These profiles reveal a duplex microvoid coalescence morphology Growth of the large primary microvoids, which nucleated at ruptured (Ni,Ti)C inclusions, was preempted as many smaller microvoids nucleated at δ-phase precipitates SEM, 950× (J.E Nolan, Westinghouse Hanford Company)

fracture-Creep fracture in Inconel alloy MA754, a mechanically alloyed nickel-base superalloy stabilized by yttria Fig 850: Fracture surface of longitudinal creep specimen tested at 760 °C (1400 °F) and 207 MPa (30 ksi) for 84 h Fracture was intergranular Note delamination along longitudinal and transverse grain boundaries SEM, 190× Fig 851: Fracture surface of long-transverse creep specimen tested at 980 °C (1800 °F) and 48 MPa (7 ksi) for 54 h Fracture occurred along longitudinal grain boundaries (perpendicular to the applied stress) SEM, 50× (J.K Tien, T.E Howson, and J.E Stulga, Columbia University)

Fatigue fracture of Nimonic 115 at room temperature and 845 °C (1550 °F) Thermal treatment: solution treat 24 h at 1230 °C (2250 °F); furnace cool to 1000 °C (1830 °F) to nucleate precipitation; air

cool; precipitation age for 96 h at 1000 °C (1830 °F); air cool Test conditions: R = -1, constant strain

rate ramps Fig 853: Fracture surface of sample tested at room temperature Failure is initiated stage I with stage I propagation until overload 5.7× Fig 854: Substructure of sample in Fig

surface-853 consists of dense, homogeneous dislocation tangles Note well-defined slip bands that have sheared

γ' precipitates TEM thin foil, 12,400× Fig 855: Fracture surface of sample tested at 845 °C(1550 °F) Failure due to crack initiation in a stage II mode around most of the circumference 5.8× Fig 856: Substructure of sample in Fig 855 Note similarity to substructure of sample tested at room temperature (Fig 854) For example, γ' shearing due to coarse slip bands is evident However, dislocation tangles are less dense TEM thin foil, 11,400× (J.K Tien and L Fritzmeir, Columbia University)

Effect of thermal cycling on fatigue fracture of single-crystal PWA 1480 with NiCoCrAlY applied by the low-pressure plasma spray process Fig 857: Fracture surface of coated sample tested at 1050 °C (1920 °F) in isothermal low-cycle fatigue The NiCoCrAlY coating was very ductile at the test temperature and did not crack The well-protected superalloy failed via multiple internal cracking originating at microporosity (arrow) SEM, 12.5× Fig 858: High-magnification view of micropore at arrow in Fig 857 SEM, 280× Fig 859: Fracture surface of coated sample cycled between 650 and

1050 °C (1200 and 1920 °F) in a thermomechanical fatigue test The NiCoCrAlY coating was placed in tension during the low-temperature portion of the test, which caused it to crack in a few cycles The nickel-base superalloy now exposed to the atmosphere failed at multiple surface locations (arrow) rather than at internal micropores (as in Fig 857 and 858) SEM, 21× Fig 860: High-magnification view of fracture origin at arrow in Fig 859 SEM, 260× (R.V Miner and S.L Draper, NASA Lewis Research Center)

Fatigue fracture of Udimet 720 tested in a salt environment at 705 °C (1300 °F) Sample heat treatment and salt composition same as in Fig 861 Test conditions: 827 MPa ± 69 MPa (120 ksi ± 10 ksi), 5000 cycles to failure Fig 863: Fracture surface shows intergranular origin at grain-boundary triple point (near bottom) with subsequent transgranular propagation Because origin is near the surface, corrosion may have had an effect on fracture initiation River patterns in transgranular portion of fracture reveal crack path SEM, 23× Fig 864: Higher-magnification view of fatigue fracture origin in Fig 863 SEM, 82× (E.A Schwarzkopf, J Stefani, and J.K Tien, Columbia University)

μ

Fatigue fracture of a cast Vitallium (Co-30Cr-7Mo) surgical implant (side plate of Jewett nail) due to improper insertion A tool used by the surgeon to contour the plate prior to its insertion into the patient also accidentally damaged the surface of the implant, introducing microcracks that led to fracture Fig 881: Tool damage on surface of implant near fracture surface 3× Fig 882: The Jewett nail broke across a screw hole Radial marks on the fracture surface (shown here) reveal the fatigue crack origin near the top of the hole 10× (C.R Brooks and A Choudhury, University of Tennessee)

Fatigue fracture of cast ASTM F75 alloy (Co-28Cr-6Mo) Material was solution treated prior to testing

on a cantilever-beam machine Fig 883: "Stair step" fracture surface indicative of stage I fatigue SEM, 515× Fig 884: Photomicrograph reveals stage I fatigue cracking FeCl 3 electrolytic etch, 200× (R Abrams, Howmedica, Division of Pfizer Hospital Products Group, Inc.)

These stereo pairs of SEM fractographs, as well as those in Figures 892, 893, 894, 895, 896, and 897, are

of fracture surfaces produced in specimens of high-purity copper (99.999% Cu) by creep testing to rupture at temperatures ranging from 425 to 642 °C (797 to 1188 °F) and at tensile stress ranging from 7

to 20 MPa (1 to 3 ksi) The material was continuously cast into 19-mm ( -in.) diam rod, swaged down to 14-mm (0.55-in.) diam, then machined into buttonhead specimens with gage sections 38 mm (1 in.) long and 6.4 mm ( in.) in diameter Before being tested, each specimens was heat treated in the creep- test furnace, first for 1 h at 400 °C (750 °F) in hydrogen and then for 16 h at 800 °C (1470 °F) in purified helium The testing also was done in purified helium At low temperature or low stress, or both, the main mechanism of fracture was intergranular separation with accompanying microvoid coalescence, and the separated-grain facets oriented at high angles relative to the tensile axis At high temperature or high stress, or both, the fracturing was transgranular and highly plastic At intermediate combinations of test temperature and stress, the main mechanism of fracture was the same as at low temperature or low

stress (intergranular separation with accompanying microvoid coalescence), but the fracture surface was fairly smooth and had a fine texture The specimen in Fig 886, 887, and 888 was tested at 425 °C (797 °F) and 20 MPa (3 ksi) and attained to steady-state strain rate at 2.0 × 10 -6 /min, fracturing in 76 h, 54 min These views show a fine texture of dimples, which resulted from microvoid coalescence, on fairly smooth separated-grain facets The area at α in the center of Fig 886 is shown at higher magnification in Fig 887; note the uniform size of the dimples Figure 888, which is of an area of the fracture different from that in Fig 886, shows dimples with shapes that vary depending on orientation of the grain facets relative

to the tensile axis The specimen shown in Fig 889 and 890 was tested at 440 °C (824 °F) and 15.5 MPa (2.25 ksi) and attained a steady-state strain rate of 7.0 × 10 -7 /min, fracturing in 308 h, 30 min These fractographs show two different areas of the fracture surface The area shown in Fig 889 displays separated-grain facets oriented at high angles relative to the tensile axis; the area shown in Fig 890 (at higher magnification) clearly shows the small dimples on the grain facets

α

The specimen in Fig 891 was tested at 525 °C (977 °F) and 10.3 MPa (1.5 ksi) and attained a steady-state strain rate 5.0 × 10 -7 /min, fracturing in 300 h, 6 min; its fracture surface has larger and smoother facets than those in Fig 889 and 890 At top in this view is a region that fractured by transgranular rupture accompanied by plastic flow The specimen in Fig 892, 893, and 894 was tested at 540 °C (1004 °F) and