Elsevier

Additive Manufacturing

Volume 36, December 2020, 101431
Additive Manufacturing

Research Paper
Quantitatively accounting for the effects of surface topography on the oxidation kinetics of additive manufactured Hastelloy X processed by electron beam melting

https://doi.org/10.1016/j.addma.2020.101431Get rights and content

Abstract

The effects of surface roughness on the high-temperature corrosion of nickel-based superalloy X produced through electron-beam melting (EBM) additive manufacturing were studied. Surface area of samples in the rough, as-fabricated condition was measured using a 2D cross-section image analysis technique and a 3D image analysis technique. To verify the accuracy of surface area measurements, oxidation testing of samples in the as-fabricated and polished conditions was performed; polished sample oxidation curves were then compared to the as-fabricated sample oxidation curves before and after being adjusted using obtained surface area measurements. It was found that the 2D technique accurately measured surface area and accounted for the overlapping features present on EBM sample surfaces, whereas the 3D technique did not. Further, it was also observed that oxidation kinetics were moderately faster on the alloy X samples in the as-fabricated condition than for the polished condition.

Introduction

The manufacturing industry is increasingly looking to use metal additive manufacturing (AM) to produce complex shapes difficult to achieve with conventional casting techniques [1], [2]. Sectors such as gas turbine manufacturers are particularly interested in AM parts. The high temperatures used in these environments (800 C) necessitate the use of nickel-based superalloys for their high temperature tensile strength, creep strength, and oxidation resistance [3].

One AM process of interest is electron beam melting (EBM). EBM is a powder bed fusion (PBF) AM process in which components are produced by selectively melting regions of powder and allowing them to solidify in a layer-by-layer fashion. Compared to other AM PBF techniques such as selective laser melting (SLM), EBM processes have the potential to better control the microstructures formed, have lesser internal residual stress, and have a lower likelihood of thermal shock-induced cracking [4], [5], [6], [7], [8], [9]. After fabrication, rough surface finishes (attributed to a “balling effect”) are often observed on the surface of SLM and EBM parts [10], [11], [12], [13], [14], [15], [16], [17], [18]. The roughness of SLM Ti-6Al-4V was found to increase as a result of melt flow behavior (associated with higher laser scan speeds) and powder layer thickness [10]. High intensity lasers (used to provide near full density parts) [13] and process parameters such as beam current and scan strategies were also found to affect roughness of EBM Ti-6Al-4V [17], SLM Inconel 625 [11], Fe-based steel [12], and stainless steels powders [13]. The powder size was also observed to affect the surface roughness of steel powder [16]. Large powder sizes 45–105 μm, often chosen for EBM processes to limit electrostatic charging, were found to result in higher roughness than for smaller SLM particles (10–60 μm) [18].

Surface roughness can influence fluid dynamics, heat transfer efficiency and frictional behavior. The effect of surface deformation on transient oxidation behavior of several materials was also studied [19], [20], [21], [22], [23], [24]. Surface modification (such as electropolishing, sandblasting or shot-peening) was found to promote the formation of a protective Cr2O3 scale instead of the development of Fe-rich oxides in electropolished Fe-28Cr after exposure in oxygen between 800 and 1200 C [19] and sandblasted 9–20 wt.% Cr steels oxidized at 600 C in air [24] and in water vapor [22]. In contrast, more Fe-rich nodules were formed on rougher surfaces of Fe-5Cr-10%Al during the early stage of oxidation at 1000 C in 1 atm O2 [20]. For Ni based alloys, laser surface treatment was found to enhance Cr supply towards the formation of Cr2O3 scale in Ni-10Cr oxidized at 1025 C in 1 atm O2 but had no effect on Ni-15Cr, which had sufficient Cr in the bulk to sustain Cr2O3 development [21]. However, the enhanced Cr diffusion towards the surface could breakdown after longer exposure times and at higher temperatures due to subsurface recrystallization, as observed for 304H alloy after 5000 h exposure at 650 C in steam [25] and for Ni-based superalloy C263 after oxidation in dry air between 700 and 1100 C [26]. The effect of surface roughness on the oxidation behavior was also reported. Spinel-oxide formation, thinner Al2O3 scale and Y-Al internal oxidation zones were formed on convex surface areas than on concave areas of rough (spark eroded) as-sprayed NiCoCrAlY compared to 1200 grit polished specimen exposed up to 1200 h between 900 and 1100 C in air [23]. Finally, the effect of rough surface finish was observed to lead to a greater under-estimation of the surface area of rough EBM 718 alloy than on SLM 718 after up to 1000 h exposure at 850 C in dry air and of rough Ti-6Al-4V after up to 500 h exposure between 500 and 600 C in air [15], [27].

For engineering dimension tolerances, particular surface finishes are often required and thus machining of these rough surfaces is a topic of interest [28], [29]. In order to standardize sample conditions to allow accurate comparison of results between different oxidation studies, specimens are usually polished to a 600 or 1200 grit surface finish. However, machining of AM parts is not always possible due to the complexity of AM component geometry, and is sometimes intentionally foregone to reduce cost. Previous work by Sanviemvongsak et al. [15] noted that higher roughness likely leads to greater absolute mass change due to an increase in surface area. To account for this, the authors of [15] measured the standard roughness and waviness parameters of rough as-fabricated EBM and SLM AM 718 alloys in an attempt to accurately compare oxidation kinetics of rough samples with polished samples. The surface area of the specimens was modeled with the use of three different periodic functions (sinusoidal, crenation, spherical) to demonstrate the effect of surface roughness on the oxidation behavior of the specimens. However, to verify this method, it was mentioned that characterization and evaluation of the experimental surface topography of as-fabricated EBM AM parts was needed.

There has been much recent work focusing on surface metrology of additive manufactured materials [30], [31]. Areal techniques (sometimes referred to as ‘3D’ techniques) performed normal to a sample surface are the most widely used. Common areal topography techniques used to measure AM surfaces include tactile profilometry [32], focus variation microscopy [33], confocal microscopy [34], coherence scanning interferometry [35] and 3D structured light imaging [36]. Atomic force microscopy (AFM), a particularly high-precision technique, is rarely used because the depths of as-fabricated EBM features exceed the capability of most AFM instrumentation. A high likelihood of AFM stylus damage is another noteworthy deterrent [30]. Nonetheless, line-of-sight areal techniques are significantly limited in their ability to describe EBM components, as they do not account for the overlapping features [37] commonly present on EBM surfaces. One areal technique, however, has been shown to capture such re-entrant features: non-destructive X-ray computed tomography (XCT) [38]. Indeed, XCT has been shown to be useful for surface metrology of SLM AlSi10Mg parts [39]. An approach to compute the areal surface texture of freeform surfaces and its limitations were described [39], [40]. Experimental validations would be of value to evaluate the applicability on rough EBM components.

In this paper, a method for measuring the real surface area of EBM parts using 2D cross-section analysis was developed and compared to well-established 3D structured light imaging. This method was applied to study the oxidation behavior of rough EBM Hastelloy® X (HX) samples: HX samples in the as-fabricated and polished conditions were exposed to a flowing air atmosphere at 800 C in 100-h cycles for total exposure times ranging from 100 to 500 h. Oxidation rates of the HX samples were then adjusted using obtained surface area measurements. The adjusted oxidation rates of the HX samples were examined and oxide scale morphologies (thickness, nature) were characterized to evaluate the accuracy of the two surface area measurement techniques.

Section snippets

Sample preparation

Alloy X, a commercial solution strengthened nickel-based superalloy widely used due to its oxidation resistance at high temperatures [41], was the material of interest in this study. An EBM-cut HX powder provided by Solar Turbines Incorporated was utilized with a particle distribution of 105 ± 45 μm; the composition of the powder (which falls within the nominal composition range of HX as provided by the manufacturer [42]) is reported in Table 1. Long thin wall HX plates 100 mm by 20 mm by 2 mm

Simulating surface area measurement techniques on a modeled sinusoidal surface

A model surface was used to validate the accuracy of both the 2D cross-section light imaging and 3D structured light methods described above. The surface area measurement techniques proposed were tested on a sinusoidal surface (Fig. 5) defined by Eq. (7):h(x,y)=B(cos(x)+cos(y));0x2π;0y2πThe calculations were performed for amplitudes, B, ranging from 0.5 to 10. To mimic the imperfect resolutions of each technique, the sinusoidal surface was grid-sampled with N=100 points per period along

Comparison of surface area measurement techniques employed

Prior to use on EBM samples, the 3D-SLI and 2D-CSLI techniques were used to measure the surface area of modeled sinusoidal surfaces with varying amplitudes (Section 3.1.1). As shown in Fig. 6, the 3D-SLI technique estimated the model's surface area accurately at all amplitudes. On the other hand, the 2D-CSLI technique estimated the model's surface area accurately for sinusoidal surface amplitudes below 8, with errors between 0% and 16%. Deviation at surface amplitudes above 8 were induced

Conclusions

As-fabricated EBM components have significant roughness that, if not accounted for, caused large overestimations of oxidation kinetics. An example of this phenomenon was provided in this study using nickel-based superalloy Alloy X oxidized at 800 C in dry air. As-fabricated and polished EBM samples underwent cyclic oxidation testing for up to 500 h and oxidation kinetics were determined. Two surface metrology techniques-3D structured light imaging (3D-SLI) and the proposed 2D cross-section

Author contributions

Matthew Kuner: Conceptualization, data curation, formal analysis, investigation, methodology, writing – original draft, writing – review & editing, visualization, validation. Marie Romedenne: Conceptualization, data curation, investigation, writing – original draft, writing – review & editing, supervision. Patxi Fernandez-Zelaia: Resources – provision of study materials, software – development of image analysis code, writing – review. Sebastien Dryepondt: Conceptualization, writing – review &

Declaration of Competing Interest

The authors report no declarations of interest.

Acknowledgments

The authors are thankful for the technical support and collaboration of M. Stephens, V. Cox, T. Lowe, T.D.B. Jacobs, C. Stephens, P. Stack, and S. Uwanyuze. The authors would also like to thank R. Pillai, M.J. Lance, and B.A. Pint for reviewing the manuscript. This research was sponsored by: the U.S. Department of Energy, Fossil Energy Crosscutting Research Program and the Office of Energy Efficiency & Renewable Energy's Advanced Manufacturing Office; Solar Turbines Incorporated; and the U.S.

References (64)

  • N. Sheng et al.

    Surface recrystallization and its effect on oxidation of superalloy C263

    Corros. Sci.

    (2017)
  • O. Oyelola et al.

    Machining of additively manufactured parts: implications for surface integrity

    Procedia CIRP

    (2016)
  • A. Bordin et al.

    Analysis of tool wear in cryogenic machining of additive manufactured Ti6Al4V alloy

    Wear

    (2015)
  • A. Townsend et al.

    Surface texture metrology for metal additive manufacturing: a review

    Precis. Eng.

    (2016)
  • M. Launhardt et al.

    Detecting surface roughness on SLS parts with various measuring techniques

    Polym. Test.

    (2016)
  • L. Newton et al.

    Areal topography measurement of metal additive surfaces using focus variation microscopy

    Addit. Manuf.

    (2019)
  • B. Zhang et al.

    In situ surface topography of laser powder bed fusion using fringe projection

    Addit. Manuf.

    (2016)
  • A. Townsend et al.

    Areal surface texture data extraction from X-ray computed tomography reconstructions of metal additively manufactured parts

    Precis. Eng.

    (2017)
  • L. Pagani et al.

    Towards a new definition of areal surface texture parameters on freeform surface

    Measurement

    (2017)
  • L. De Chiffre et al.

    Quantitative characterisation of surface texture

    CIRP Ann. Manuf. Technol.

    (2000)
  • J. Salvi et al.

    A state of the art in structured light patterns for surface profilometry

    Pattern Recognit.

    (2010)
  • M.M. Romedenne et al.

    High temperature air oxidation behavior of Hastelloy X processed by Electron Beam Melting (EBM) and Selective Laser Melting (SLM)

    Corros. Sci.

    (2020)
  • L. Pagani et al.

    Towards a new definition of areal surface texture parameters on freeform surface: re-entrant features and functional parameters

    Measurement

    (2019)
  • W.E. Frazier

    Metal additive manufacturing: a review

    J. Mater. Eng. Perform.

    (2014)
  • T. DebRoy et al.

    Scientific, technological and economic issues in metal printing and their solutions

    Nat. Mater.

    (2019)
  • T.M. Pollock et al.

    Nickel-based superalloys for advanced turbine engines: chemistry, microstructure and properties

    J. Propuls. Power

    (2006)
  • W.J. Sames et al.

    The metallurgy and processing science of metal additive manufacturing

    Int. Mater. Rev.

    (2016)
  • M. Ramsperger et al.

    Microstructure of the nickel-base superalloy CMSX-4 fabricated by selective electron beam melting

    Metall. Mater. Trans. A

    (2016)
  • P.K. Gokuldoss et al.

    Additive manufacturing processes: selective laser melting, electron beam melting and binder jetting-selection guidelines

    Materials

    (2017)
  • Y.S. Lee et al.

    Asymmetric cracking in Mar-M247 alloy builds during electron beam powder bed fusion additive manufacturing

    Metall. Mater. Trans. A

    (2018)
  • K. Mumtaz et al.

    Top surface and side roughness of Inconel 625 parts processed using selective laser melting

    Rapid Prototyp. J.

    (2009)
  • E. Yasa et al.

    The investigation of the influence of laser re-melting on density, surface quality and microstructure of selective laser melting parts

    Rapid Prototyp. J.

    (2011)
  • Cited by (24)

    View all citing articles on Scopus

    This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).

    1

    Present address: School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, GA 30332, USA.

    View full text