Conference PaperPDF Available

Fatigue performance of micro-crack free Hastelloy X produced by selective laser melting (SLM)

Authors:

Abstract and Figures

The production of micro-crack free Hastelloy X by selective laser melting (SLM) process without subsequent hot isostatic pressing (HIP) has proven to be challenging. Hastelloy X is shown to be susceptible to solidification cracking during the SLM process, decreasing its mechanical performance especially its tensile ductility. In addition, the presence of micro-cracks does not allow this material to be used in high performance applications. Recent works have reported on composition modifications reducing the amounts of trace elements such as Si, C or Mn in order to avoid solidification cracking in this material during SLM. Besides these promising results, no work has been reported on validating its dynamic mechanical performance. The current work investigates the fatigue life behaviour of micro-crack free SLM processed Hastelloy X. The use of a modified and commercial Hastelloy X powder along with optimised process parameters, resulted in the production of micro-crack free, high density specimens. Axial tension-compression fatigue tests were conducted on fatigue specimens with axes oriented parallel and perpendicular to the build direction, in order to validate the dynamic behaviour of this modified Hastelloy X variant. 3D microstructural characterization was performed to investigate its effect on potential anisotropic behaviour due to the presence of crystallographic/morphological texture that can be present in SLM processed materials. Fractography analysis was carried out in order to better understand the fatigue performance of the material. Moreover, the influence of a conventional heat treatment (1177 °C for 2h with subsequent (argon) air cooling) on the fatigue performance was studied and the material behaviour was compared to that of material tested in the as-built condition.
Content may be subject to copyright.
ESIAM19, 9-11 September 2019, Trondheim, Norway
1
Fatigue performance of micro-crack free Hastelloy X produced by
selective laser melting (SLM)
M.L. Montero-Sistiaga 1,2, N.M. Dhansay 3, L. Bautmans 4, S. Nardone 2, T.H. Becker 3, J-P
Kruth 5, K. Vanmeensel 1
1 KU Leuven, Department of Materials Engineering, Kasteelpark Arenberg 44, Heverlee 3001,
Belgium
2 Engie-Laborelec, Rodestraat 125, Linkebeek 1630, Belgium
3 University of Stellenbosch, Department of Mechanical and Mechatronic Engineering,
Stellenbosch 7602, South Africa
4 Oerlikon Eldim (NL) BV, AD Lomm 5943, The Netherlands
5 KU Leuven, Department of Mechanical Engineering, Member of Flanders Make, Celestijnenlaan
300B, Heverlee 3001 Belgium
ABSTRACT
The production of micro-crack free Hastelloy X by selective laser melting (SLM) process without subsequent
hot isostatic pressing (HIP) has proven to be challenging. Hastelloy X is shown to be susceptible to
solidification cracking during the SLM process, decreasing its mechanical performance especially its tensile
ductility. In addition, the presence of micro-cracks does not allow this material to be used in high performance
applications. Recent works have reported on composition modifications reducing the amounts of trace
elements such as Si, C or Mn in order to avoid solidification cracking in this material during SLM. Besides
these promising results, no work has been reported on validating its dynamic mechanical performance.
The current work investigates the fatigue life behaviour of micro-crack free SLM processed Hastelloy X. The
use of a modified and commercial Hastelloy X powder along with optimised process parameters, resulted in
the production of micro-crack free, high density specimens. Axial tension-compression fatigue tests were
conducted on fatigue specimens with axes oriented parallel and perpendicular to the build direction, in order
to validate the dynamic behaviour of this modified Hastelloy X variant. 3D microstructural characterization
was performed to investigate its effect on potential anisotropic behaviour due to the presence of
crystallographic/morphological texture that can be present in SLM processed materials. Fractography
analysis was carried out in order to better understand the fatigue performance of the material. Moreover,
the influence of a conventional heat treatment (1177 °C for 2h with subsequent (argon) air cooling) on the
fatigue performance was studied and the material behaviour was compared to that of material tested in the
as-built condition.
INTRODUCTION
Hastelloy X is commonly used for petrochemical applications and combustion-zone components
due to its high corrosion resistance and strength at elevated temperatures. Wang [1] first produced
Hastelloy X by SLM and reported the presence of cracks. The author investigated the influence of
process parameters on the crack occurrence. Afterwards, several investigations reported on the
modification of the alloy composition in order to avoid cracking. Harrison et al. [2] suggested
increasing the ultimate tensile strength (UTS) by solid-solution strengtheners in order to decrease
the cracking susceptibility. On the other hand, Tomus et al. [3,4] attributed cracking to the presence
of Si and C. Lastly, Marchese et al. [5] found carbides at grain boundaries and interdendritic
regions, suggesting they make the material brittle and cause the cracking. Recently Sanchez-Mata
ESIAM19, 9-11 September 2019, Trondheim, Norway
2
et al. [6] have published crack free Hastelloy X parts by SLM using a commercial powder and they
claim the powder did not have compositional modifications. On the other hand, Han et al. [7] have
reported on the fatigue behaviour of Hastelloy X in as-built condition, containing micro-cracks, and
of hot isostatic pressed (HIP) samples without cracks.
In current work, the mechanical behaviour of micro-crack free Hastelloy X in tensile and fatigue
was studied using an intentionally modified commercial Hastelloy X powder. The samples were
teste in as-built and heat-treated conditions. In addition, vertical and horizontal samples were tested
to evaluate the anisotropy induced by the SLM process.
METHODS
MetcoAdd HX-D powder, a modified composition of Hastelloy X supplied by Oerlikon, was used as
feedstock. The powder composition is shown in Table 1. The samples were produced using a
SLM280hl machine equipped with a Gaussian laser, 70 µm beam diameter (1/e2) and a maximum
laser output of 400 W. A wide range of process parameters was varied at the optimisation step. For
simplicity, the volumetric energy density (Ev=P · (hs·t·v)-1 [J/mm3]) was used where P is the laser
power, hs is the hatch spacing, t is the layer thickness and v is the scan speed.
Table 1 Composition of MetcoAdd HX-D powder supplied by Oerlikon.
Weight Percent (nominal)
Ni
Cr
Fe
Mo
Co
W
C
Mn
Balance
20.50-23.00
17-20
8-10
0.50-2.50
0.20-1.00
0.05-0.15
< 1.00
Si
Cu
Al
Ti
P
S
B
< 1.00
< 0.50
< 0.50
< 0.15
< 0.04
< 0.03
< 0.01
As-built (AB) and heat-treated (HT) samples were analysed to observe the effect of the heat
treatment on the tensile and fatigue behaviour. Heat-treatment was performed at 1177 °C for 2 h
followed by forced air (argon) cooling at a rate of > 40 °C/min. In addition, samples were built both
in vertical (V) and horizontal (H) direction to analyse possible anisotropy. Samples were labelled
as AB-H, AB-V, HT-H and HT-V, referring to the AB or HT condition followed by the building
direction.
Figure 1 Axial tension-compression specimen dimensions according to ASTM E466 07.
Five flat tensile samples were tested per condition according to ASTM E8. 60 cylindrical samples
110 mm long and 9 mm in diameter were built in horizontal and vertical direction. All samples were
machined with a continuous radius between ends according to ASTM E466 07 standard as shown
in Figure 1. After machining the Ra roughness was 0.6 µm, so the samples were manually ground
to Ra < 0.1 µm using a lathe and SiC grinding paper up to 4000 grid size. Axial tension-compression
fatigue tests (R=-1) were performed on a hydraulic Schenk machine at 20 Hz frequency. Five stress
levels were tested from 40% to 80% in steps of 10%. The stress level percentage was calculated
ESIAM19, 9-11 September 2019, Trondheim, Norway
3
according to the tensile yield strength for each condition; 530 MPa for AB and 350 MPa for HT
samples. Microstructure and fractography of the AB and HT samples was conducted using a light
optical (LOM) in a Leica DMILM HC microscope and scanning electron microscope (SEM) in a FEI
Nova NanoSEM 450 microscope.
RESULTS AND DISCUSSION
The process parameters were optimised to achieve high density and crack-free parts. Scan speed
and laser power were varied leading to a wide range of Ev. The results are shown in Figure 2(a),
where the effect of the volumetric energy density is depicted with respect to the relative density.
When increasing the energy input, an increase in relative density is observed. Ev below 45 J/mm3
lead to lack of fusion porosities which are caused by insufficient energy supplied to melt the powder
layer and partially re-melt the underlying layer. On the other hand, energies above 45 J/mm3 result
in fully dense parts up to 150 J/mm3 energies. This behaviour differs from previous studies where
energies above 100 J/mm3 lead to keyhole porosities, decreasing the relative density of the parts
[8], possible because composition differences for trace elements.
Figure 2 (a) Relative densities of parts built with increasing volumetric energy density. Microstructures parallel to the
building direction (BD) using LOM of (b) AB and (d) HT and using SEM of (c) AB and (e) HT.
The optimal Ev was chosen to obtain the highest densities, crack free parts and also maximise the
productivity. For the latter one, a combination of scan settings with the highest scan speed was
chosen. Therefore, an Ev of 62.5 J/mm3 was selected as the optimal parameter. These parameters
were used for building tensile and fatigue specimens.
In Figure 2 (b-e) the microstructure of AB and HT samples is depicted by LOM and SEM. For AB,
wide and shallow melt pools are observed with grains growing epitaxially across several melt pools
parallel to the building direction (BD). The shallow melt pool nature explains the absence of keyhole
porosities, within the tested Ev range when increasing Ev with respect to previous results from
literature. At sub-grain level, sub-micron size cells are observed, similar to the results found in
literature [2,3,5,8]. Cell boundaries consist of high density dislocation walls, with lower dislocation
density at the cell core [8]. After the heat treatment no sub-grain structure is observed by SEM,
although, as seen in a previous study [8], it is known that the sub-grain microstructure is maintained
after the heat treatment. This sub-grain microstructure contains low angle boundaries (< 4°), with
ESIAM19, 9-11 September 2019, Trondheim, Norway
4
lower dislocation density as compared to the AB condition [8]. It is important to note that no cracks
were found in any of the samples neither in AB nor in HT condition. This confirms that the
intentionally tuned composition results in micro-crack free Hastelloy X by SLM using MetcoAdd HX-
D powder.
In order to define the stress levels for the axial tension-compression fatigue tests, tensile tests were
conducted for the four conditions: AB-H, AB-V, HT-H and HT-V. The results are shown in Figure 3
where a single representative sample is shown per condition for the sake of clarity. AB samples
show a significantly higher yield tensile strength (YTS) values than HT samples, with YTS above
500 MPa and around 300 MPa for HT samples. For HT samples higher ductility values are observed
with elongations close to 50%. Slight anisotropy between the horizontally and vertically built
samples is observed for both conditions, AB and HT. Horizontal samples, in general, present
slightly higher strength and lower elongation values. Tomus et al. [9] also observed similar
anisotropy although the strain values were significantly lower for AB and HT values. Those authors
obtained similar strain levels after HIP and a consequent heat treatment. Based on the tensile
results, the fatigue stress levels were defined for AB and HT samples, as a percentage of the YTS,
being 530 MPa and 350 MPa for AB and HT samples, respectively.
Figure 3 Strain stress curves for a representative sample of each testing condition: AB-H, AB-V, HT-H and HT-V.
Axial tension-compression fatigue results for all testing conditions are summarised in Figure 4. Five
stress levels and 3 samples per stress level were tested for each condition. In AB condition, both
vertically and horizontally built samples reach the fatigue limit at 107 cycles at a stress level of 40%
(212 MPa). In addition, a bigger spread is observed for samples tested at 60%. It is difficult to
address the fatigue performance differences between horizontal and vertical samples, due to the
big spread observed for AB-H and AB-V samples.
Due to the high amount of samples exceeding 107 cycles, an extra stress level was added for HT-
V, at 90%. No sample was tested at 40%. As depicted in Figure 4(b), all samples at 50% and 70%
did not break, while 2 samples broke before 107 cycles at 60%. At 80% stress level, one sample
also reached 107 cycles. For HT-V samples, a big spread is observed as well as an increase in
fatigue life behaviour compared to AB samples. The spread observed can only be understood by
fractography of the initiation zones, which will be analysed. In contrast, for HT-H samples, only
samples tested at 40% exceeded the fatigue limit, showing a lower fatigue strength with respect to
HT-V samples. The spread for HT-H is higher at higher stress levels.
ESIAM19, 9-11 September 2019, Trondheim, Norway
5
Figure 4 Axial tension-compression fatigue behaviour for samples built in vertical (V) and horizontal (H) direction in
as-built condition (AB).
Figure 5 Fractography images of (a-d) AB samples and (e-g) HT samples. (a) full fracture surface of AB-V sample
(265 MPa (50%), 273,197 cycles), where (b) is the ductile fracture and (c) is the initiation. (d) and (e) with white arrows
show the striations for samples in AB and HT condition respectively. (f) and (g) show some type of building defects for
HT samples built in V and H direction respectively (yellow arrows and dashed square). (I) Initiation-propagation and
(II) final fracture regions.
In Figure 5(a) the fracture surface of an AB-V sample is shown. It was tested at 265 MPa (50%
stress level) and suffered from an early fracture. Looking at the initiation point, a lack of fusion pore
is observed (Figure 5(c)) of more than 100 µm in diameter along with some unmolten powder
particles. Nevertheless, in the final fracture surface, a ductile type fracture is observed. Several
samples for AB condition present similar lack of fusion defects which were the cause of the crack
initiation. AB and HT samples, in both directions, present striations in the propagation zone, as
shown in Figure 5(d,e) depicted with white arrows.
After HT, some of the samples present aligned lack of fusion zones as shown in Figure 5(f,g) for
vertical and horizontal samples, respectively. Figure 5(f) shows parallel scan tracks (shown with
the yellow rectangle and arrows) at the same layer which were not attached to the layer that was
ESIAM19, 9-11 September 2019, Trondheim, Norway
6
scanned next. For horizontal samples (Figure 5 (g)) a similar effect is observed as shown in the
depicted with the arrow. A line defect is observed with no cohesion between subsequently scanned
layers. This suggests that there might have been a recoating problem whithin those layers with
non-homogeneous powder depositions, leading to slightly thicker powder layers. As seen in the
optimization, the powder used in this work does not create deep melt pools. Therefore, the shallow
melt pool are more sensitive to powder layer thickness variations. It should be noted that these
layer defects are only observed after HT, which suggests that these porosities become more
detrimental after the HT. Nevertheless, more research should be done to understand better the
effect of the HT.
The parameter optimization was performed for cubes of 1 cm3 which showed close to 100%
densities. Fatigue samples were built in very charged platforms, which can lead to more spatter
formation, due to no lamellar gas flow, and powder bed instabilities during deposition, hence lack
of fusion porosities. Therefore, higher laser powers or lower layer thicknesses are recommended
to increase the Ev and avoid lack of fusion porosities.
CONCLUSIONS
This work confirms the printability of micro-crack free Hastelloy X parts by SLM using the optimised
alloy composition, provided by Oerlikon. The tensile samples show promising results, with high
strain values after the heat treatment and competitive strength values when comparing to
commercial Hastelloy X.
In AB condition, the fatigue limit, at 107 cycles, was achieved at 212 MPa (40%) for both V and H
samples. The spread on the results can be attributed to the lack of fusion porosities found on the
fracture surfaces. After the heat treatment, the effect of these porosities on the fatigue behaviour
was found to be more detrimental, especially in horizontal samples.
Spatter formation and powder deposition instabilities may have contributed to the porosities found.
These phenomena seem to be more sensitive for this powder composition with shallow melt pools.
When building a high number of samples per build plate, small adjustments should be performed
with respect to the process parameters.
REFERENCES
[1] F. Wang, Mechanical property study on rapid additive layer manufacture Hastelloy® X alloy by selective laser melting
technology, Int. J. Adv. Manuf. Technol. 58 (2012) 545551.
[2] N.J. Harrison, et al., Reduction of micro-cracking in nickel superalloys processed by Selective Laser Melting: A
fundamental alloy design approach, Acta Mater. 94 (2015) 5968.
[3] D. Tomus, et al., Controlling the microstructure of Hastelloy-X components manufactured by Selective Laser Melting,
Phys. Procedia. 41 (2013) 823827.
[4] D. Tomus, et al., Effect of minor alloying elements on crack-formation characteristics of Hastelloy-X manufactured by
selective laser melting, Addit. Manuf. 16 (2017) 6572.
[5] G. Marchese, et al., Study of the Microstructure and Cracking Mechanisms of Hastelloy X Produced by Laser Powder Bed
Fusion, Materials (Basel). 11 (2018) 106.
[6] O. Sanchez-Mata, et al., Fabrication of crack-free nickel-based superalloy considered non-weldable during laser powder
bed fusion, Materials (Basel). 11 (2018) 19.
[7] Q. Han, et al., Laser powder bed fusion of Hastelloy X: Effects of hot isostatic pressing and the hot cracking mechanism,
Mater. Sci. Eng. A. 732 (2018).
[8] M.L. Montero-Sistiaga, et al., Microstructure and mechanical properties of Hastelloy X produced by HP-SLM (high power
selective laser melting), Mater. Des. 165 (2019) 107598.
[9] D. Tomus, et al., Influence of post heat treatments on anisotropy of mechanical behaviour and microstructure of Hastelloy-
X parts produced by selective laser melting, Mater. Sci. Eng. A. 667 (2016) 4253.
... However, one aspect which has drawn considerable attention is the correlation of microstructure with mechanical properties of as-built Hastelloy X 3D parts viz. forming (Rosenthal et al., 2019), tensile (Keshavarzkermani et al., 2019a,b;Han et al., 2018Han et al., , 2019Esmaeilizadeh et al., 2020;Harrison et al., 2015;Kong et al., 2019;Tomus et al., 2016;Ni et al., 2019;Montero-Sistiaga et al., 2019a, fatigue (Han et al., 2018;Montero-Sistiaga et al., 2019a;Saarimäki et al., 2018;Calmunger et al., 2019) and corrosion . It is known that Hastelloy X exhibits hot cracking during LPBF process. ...
... However, one aspect which has drawn considerable attention is the correlation of microstructure with mechanical properties of as-built Hastelloy X 3D parts viz. forming (Rosenthal et al., 2019), tensile (Keshavarzkermani et al., 2019a,b;Han et al., 2018Han et al., , 2019Esmaeilizadeh et al., 2020;Harrison et al., 2015;Kong et al., 2019;Tomus et al., 2016;Ni et al., 2019;Montero-Sistiaga et al., 2019a, fatigue (Han et al., 2018;Montero-Sistiaga et al., 2019a;Saarimäki et al., 2018;Calmunger et al., 2019) and corrosion . It is known that Hastelloy X exhibits hot cracking during LPBF process. ...
Article
In the present study, grain size and grains exhibiting hot cracks of laser powder bed fusion (LPBF) processed Hastelloy X as-built parts under two scanning strategies i.e., 90° and 45° rotation vectors have been characterized using electron back scattered diffraction (EBSD) technique. EBSD orientation maps revealed relatively finer grain size in 45° scanning strategy as compared to 90° scanning strategy. Laser scanning speed was found to be the governing parameter for variation in grain size as compared to laser power and energy density. Maximum crack density was observed at the highest value of energy density and it appears that energy density has more dominant effect on hot cracking tendency as compared to laser power. At the maximum value of the energy density under 45° scanning strategy, the preferred texture type changes from <001> to <111> orientation. EBSD orientation maps revealed that a specific combination of grains, one of which is stiffer as compared to other, is more susceptible for hot cracks. They are either Goss ({110}<001>), Brass ({110}<112>) or Copper ({112}<111>) oriented grains. It is proposed that these stiffer grains, with high elastic modulus and low capacity of tolerating higher magnitude of thermal strains, paves the way for relieving locked-in thermal stresses during LPBF process.
... Figure 1: Printed platforms used for the creep teesting. Already removed samples were used for a separate study on fatigue resistance [9]. ...
Article
Additive manufactured components are increasingly being applied in operational equipment. If these components are going to be used at higher temperatures, their creep behaviour needs to be verified. In the ECCC workgroup 3C an exploratory study of the short-term creep behaviour of the HASTELLOY® X alloy produced by Selective Laser Melting was carried out, with the aim of characterising the material behaviour and collecting input for the heat treatment optimisation, if necessary. Hastelloy X is a nickel-chromium-iron-molybdenum alloy for applications where a combination of oxidation resistance and high-temperature strength is required together with good fabricability. Creep test results up to about 3000 h were compared with creep results of conventionally produced Hastelloy X. Two orientations were tested: parallel and perpendicular to the building direction. In addition, the microstructure of the ruptured creep specimens was investigated to determine the fracture location/mechanism and the main microstructural features. The obtained results can be considered promising and give useful indications in the perspective of LPBF material application at high temperatures.
... Figure 1: Printed platforms used for the creep teesting. Already removed samples were used for a separate study on fatigue resistance [9]. ...
Conference Paper
Full-text available
Additive manufactured components are increasingly being applied in operational equipment. If these components are going to be used at higher temperatures, their creep behaviour needs to be verified. In the ECCC workgroup 3C an exploratory study of the short-term creep behaviour of the HASTELLOY® X alloy produced by Selective Laser Melting was carried out, with the aim of characterising the material behaviour and collecting input for the heat treatment optimisation, if necessary. Hastelloy X is a nickel-chromium-iron-molybdenum alloy for applications where a combination of oxidation resistance and high-temperature strength is required together with good fabricability. Creep test results up to about 3000h were compared with creep results of conventionally produced Hastelloy X. Two orientations were tested: parallel and perpendicular to the building direction. In addition, the microstructure of the ruptured creep specimens was investigated to determine the fracture location/mechanism and the main microstructural features. The obtained results can be considered promising and give useful indications in the perspective of SLM material application at high temperatures.
... They reported that HIPed samples have superior fatigue performance compared to the as-built samples, due to the closure of internal porosities and residual stress relief. Montero-Sistiaga et al. [42] tested the micro-crack free Hastelloy X flat machined samples under reverse loading condition and observed a large scatter in the experimental fatigue lives due to presence of lack of fusion defects. ...
Article
The mechanical performance of parts, made by laser powder-bed fusion (LPBF), under quasi-static and cyclic loadings can be altered significantly by changing the process parameters. To better understand the related changes, tensile and fatigue LPBF Hastelloy X coupons were manufactured at high and low laser scanning speeds (LSS) in the nearly full-dense processing window. The experimental results demonstrate that, although the microstructure and surface roughness of the parts varies by altering the LSS, the porosity or density of the printed samples are not changing significantly within the range studied here. High LSS samples show higher tensile strength and higher rates for strain hardening at the primary stages of the deformation. Statistical analysis of variance indicates that LSS is a significant factor affecting the fatigue life. The S-N diagram shows Basquin lines intersection at the midlife fatigue region. In the low cycle fatigue (LCF) region high LSS samples show higher cycles to failures due to hysteresis stabilization at lower strains and hence less relative pre-strain and cyclic damage. Since stress risers are important features at high cycle fatigue (HCF) due to stress concentration and associated plasticity, the low LSS samples tolerate a greater number of cycles at lower stress levels and show a higher fatigue limit obtained by the step-loading method. These findings suggest that the process parameters can be optimized to achieve the best possible performance based on the desired application in LCF or HCF for AM parts.
... Fig. 16(a-c) shows the room temperature tensile properties such as yield strength, ultimate tensile strength (UTS) and elongation to failure of PBF-EB and PBF-LB HX specimens, compared with that of the requirement from the ASTM B572-06 standard. The tensile tests were performed only on vertically built PBF-EB and PBF-LB specimens as the vertical build orientation were reported to have comparable [43] or inferior mechanical properties [44,45] than horizontally build specimens. It can be seen that the tensile strength of as-fab and stress-relieved PBF-LB HX are significantly higher than that of PBF-EB while the elongation is lower. ...
Article
Understanding the microstructure and mechanical properties associated with different metal additive manufacturing techniques is crucial for its wide-scale acceptance in industries. In this work, we aim to understand the microstructural variations and mechanical properties of Hastelloy X (HX) manufactured by electron beam powder bed fusion (PBF-EB) and laser-based powder bed fusion (PBF-LB) process. The size and shape of melt pool and grains were analysed and correlated to final microstructure. The elevated powder bed temperature during PBF-EB was found to promote the formation of grain-boundary precipitates. The higher thermal gradient and cooling rate in PBF-LB resulted in higher tensile residual stress (σmax = 447 ± 10 MPa) within the parts in comparison to PBF-EB parts (σmax = 16 ± 13 MPa). The as-fabricated PBF-EB HX specimens showed a lower tensile strength of 590 MPa but a higher elongation of 60%, whereas its PBF-LB counterparts demonstrated a significantly higher tensile strength of 825 MPa but a lower elongation of 38%. This notable difference in the mechanical behaviour of PBF-EB and PBF-LB built HX was attributed to the columnar microstructure,<100> crystallographic texture and underlying strengthening mechanisms. Furthermore, the mechanical properties of PBF-EB and PBF-LB built HX specimens were predicted using multiple strengthening mechanisms, which demonstrated a good agreement with that of experimentally measured.
Article
The tensile deformation behaviour of a crack-free additively manufactured (AM) Ni base superalloy and its wrought (W) counterpart has been investigated by in-situ synchrotron X-ray diffraction. In order to understand the evolution of dislocation character, dislocation density (ρ), coherent scattering domain size (D) and dislocation arrangement parameter (M), the diffraction data corresponding to the first six{hkl} reflections have been analyzed using modified Williamson-Hall and modified Warren-Averbach methodologies. The ρ was found to be relatively higher in AM superalloy as compared to W superalloy during both the regimes of elastic and plastic deformation. In the plastic deformation regime, while the fraction of edge and screw dislocations was found to be nearly equal in proportion in W superalloy, the fraction of screw dislocations remained higher in AM superalloy. In view of the relatively high initial ρ and inefficacy of D in AM superalloy, heat treatment in as-built parts is suggested.
Article
There is a growing demand to produce parts from Hastelloy-X materials, which exhibit a high unit cost per weight, by the selective laser melting (SLM) process. However, when the Hastelloy-X material was employed, solidification crack occurred during the SLM process. This cracking can be reduced or eliminated by optimizing the process parameters, such as the laser power and scan speed, although the optimization is time- and cost-intensive. In this study, we evaluated the possibility of exploring the change in the plastic strain, obtained by thermal elastoplastic analysis, during the solidification process, as an indicator of the occurrence of solidification cracking. The experimental observations of solidification cracking in the single-track formation tests for various laser conditions, and the degree of change in the plastic strain during the solidification process obtained by the thermal elastoplastic analysis, were compared. The single-track formation part was the foundational part of the parts made by SLM process. There was a clear correlation between the maximum change in the plastic strain in the melted area and the existence of solidification cracking. The maximum change in the plastic strain was less than 0.0041 when no crack was found, whereas it was in the range larger than 0.0048 when the cracks were observed. These results indicated the possibility of assessing the occurrence of solidification cracking, in terms of the maximum change in the plastic strain in the melted area.
Article
Full-text available
Crack-free Hastelloy X fabricated through laser powder bed fusion (LPBF) from powder with a standard chemical composition is reported. Electron backscatter diffraction (EBSD) analysis evidenced columnar grains parallel to the building direction. The typical LPBF columnar dendrite microstructure was found to be finer than reported elsewhere. Mo-enriched carbides (~50 nm), presumed to play an important role in the cracking behavior of the alloy, were confirmed along interdendritic regions. Crack-free condition was maintained after heat treatment at 1177 • C for 1 h followed by water quenching, and the resulting microstructure was analyzed.
Article
Full-text available
Hastelloy X (HX) is a Ni-based superalloy which suffers from high crack susceptibility during the laser powder bed fusion (LPBF) process. In this work, the microstructure of as-built HX samples was rigorously investigated to understand the main mechanisms leading to crack formation. The microstructural features of as-built HX samples consisted of very fine dendrite architectures with dimensions typically less than 1 µm, coupled with the formation of sub-micrometric carbides, the largest ones were mainly distributed along the interdendritic regions and grain boundaries. From the microstructural analyses, it appeared that the formation of intergranular carbides provided weaker zones, which combined with high thermal residual stresses resulted in hot cracks formation along the grain boundaries. The carbides were extracted from the austenitic matrix and characterized by combining different techniques, showing the formation of various types of Mo-rich carbides, classified as M₆C, M12C and MnCm type. The first two types of carbides are typically found in HX alloy, whereas the last one is a metastable carbide probably generated by the very high cooling rates of the process.
Article
Full-text available
The Selective Laser Melting (SLM) process generates large thermal gradients during rapid melting of metallic powdered feedstock. During solidification certain alloys suffer from thermally induced micro-cracking which cannot be eliminated by process optimisation. An alloy’s crack susceptibility may reduce by increasing its Thermal Shock Resistance (TSR), potentially achieved through an increase in tensile strength. This hypothesis is investigated with Hastelloy X, a common nickel-base superalloy of known high crack susceptibility when processing SLM. It is demonstrated that through consideration of the imposed rapid solidification conditions, Hastelloy X can be made to form a supersaturated solid solution in the as deposited state. The fundamental solid solution strengthening (SSS) effect is exploited to generate an increase in lattice stress, by increasing the most potent SSS elements present within the alloy, whilst maintaining specification composition. The modified alloy displayed a 65% reduction in cracking and an increase in elevated temperature tensile strength, lending support to the initial hypothesis and identifying a possible approach for developing further SLM crack resistant versions of well-known alloy compositions.
Article
Full-text available
Selective Laser Melting (SLM) is steadily gaining acceptance across the manufacturing industry. Techniques for manufacturing components with complex geometries layer by layer have proven to be very effective in accelerating product development and hence reducing time to market. To build components by SLM from a nickel based super-alloy requires an understanding of process parameters and how they influence the microstructure, the degree of porosity and the properties. In this work, it was found that the as-fabricated density of parts manufactured from Hastelloy-X by SLM could be increased from 77 to 99% by decreasing the laser scan speed, and that the degree of cracking can be reduced by decreasing the amount of minor alloying additions such as Mn and Si.
Article
In order to increase the production rate during selective laser melting (SLM), a high power laser with a large beam diameter is used to build fully dense Hastelloy X parts. Compared to SLM with a low power and small diameter beam, the productivity was increased from 6 mm³/s to 16 mm³/s, i.e. 2.6 times faster. Besides the productivity benefit, the influence of the use of a high power laser on the rapid solidification microstructure and concomitant material properties is highlighted. The current paper compares the microstructure and tensile properties of Hastelloy X built with low and high power lasers. The use of a high power laser results in wider and shallower melt pools inducing an enhanced morphological and crystallographic texture along the building direction (BD). In addition, the increased heat input results in coarser sub-grains or high density dislocation walls for samples processed with a high power laser. Additionally, the influence of hot isostatic pressing (HIP) as a post-processing technique was evaluated. After HIP, the tensile fracture strain increased as compared to the strain in the as-built state and helped in obtaining competitive mechanical properties as compared to conventionally processed Hastelloy X parts.
Article
Two batches of pre-alloyed Hastelloy-X powder with different Si, Mn and C contents were used to produce specimens by Selective Laser Melting (SLM). Cracks with various morphologies were found in some of the parts. Two major reasons that control crack formation and propagation were considered: (i) internal strain accumulation due to the thermal cycling that is characteristic to SLM processing; (ii) crack formation and propagation during solidification. This phenomenon, known as hot tearing, is frequently found in conventional casting and is dependent on chemical composition. Using thermodynamic software simulation, the temperature vs fraction of solid curves was used to determine hot tearing sensitivity as a function of Si, Mn and C content. It was found that low Si and C contents help in avoiding crack formation whereas cracking propensity was relatively independent of Mn concentration. Hence, the cracking mechanism during SLM is believed to be as follows: crack initiation is mainly induced during solidification and is dependent on the content of minor alloying elements such as Si and C, whereas crack propagation predominantly occurs during thermal cycling. If microstructures free of micro-cracks after solidification can be generated with optimised SLM parameters, these manufactured parts can sustain the internal strain level and, thus, crack formation and propagation can be avoided.
Article
Selective laser melting (SLM) was used to fabricate tensile specimens using Hastelloy-X pre-alloyed powder. Mechanical behaviour at room temperature, normal and parallel to building direction, was investigated. Furthermore, as-fabricated tensile samples were compared with ones post processed by heat treatments (HT), hot isostatic pressing (HIP) and a combination of both (HIP+HT).Yield strength (YS), ultimate tensile strength (UTS) and elongation to failure (εf) were analysed and explained based on the microstructure evolution. Dendrites and molten pool boundaries are mainly responsible for the observed anisotropy in εf of horizontal and vertical samples in the as-fabricated condition. After their dissolution by HT an increase in εf was observed. The columnar grain structure also contributes to the observed anisotropy in εf, inducing more ductile and cleavage like fracture surfaces in vertical and horizontal samples, respectively. The removal of porosity after HIP and HIP+HT yields a positive effect on εf. HIP or HT after SLM reduces the YS due to recovery processes such as dislocation density reduction and rearrangement of these dislocations in subgrain boundaries. Carbides of the type MxCy were partially segregated at the grain boundaries after HIP with detrimental effect on εf.
Article
Tensile mechanical properties of selective laser-melted Hastelloy® X alloy in as-deposited condition and after hot isostatic pressing (HIP) have been studied at ambient and elevated temperatures. Room temperature four-point bending and tension–tension fatigue properties have also been investigated in as-deposited condition and after HIP. The yield strength of the as-deposited selective laser-melted Hastelloy® X specimen is higher than the heat-treated (hot forged) samples. The ultimate strength is also higher than that of the hot forged samples while the elongation property is lower. This can be attributed to its ultrafine microstructure caused by rapid solidification, which is characteristic of the selective laser melting process. It is also found that the mechanical properties (tensile and fatigue) do not vary with samples built in different bed locations.