Page 76 - MSAM-1-1

Materials Science in Additive Manufacturing A ML model for AM PSP of Ti64

to predict the relationship between cutting conditions, cutting energy. The success of this approach depends on
temperature, grain size, grain fraction, and hardness data the critical definition of a suitable reduced-order form
on Ti-6Al-4V for generic microstructures derived from of descriptors for heterogeneous grain morphology
conventional manufacturing . Hence, high efficiency and across a wide range of microstructures for given material
[22]
validated PSP linkages for metal AM processing, which composition. Figure 1 presents the aims, scope, and
capture the AM part machining behavior and material methodology of this study, which are detailed in subsequent
properties, need to be developed. sections.

This study builds on the prior work described. Gong
et al. (2020) showed statistically significant differences in the 2. Methodology
machining behavior of Ti-6Al-4V across build directions 2.1. Sample preparation
in EB-PBF specimens, and as-AM and after heat-treated
L-PBF specimens (21% lower specific cutting power in In the case of Ti-6Al-4V EB-PBF specimens, 25 × 25 ×
[23]
L-PBF specimens) . In addition, Ren et al. (2019) found 50 mm Ti-6Al-4V blocks were fabricated in an Arcam
that visual evaluation of material characterization data A2 electron beam melting machine with 50 µm layer
showed textured differences in microstructure, residual thickness using standard Ti-6Al-4V 50 µm preheat and
stress, and crystal graphic information among different PFB melt parameters provided by Arcam. In the case of L-PBF
parts . Recent study by Goh et al. (2021) has identified specimens, Ti-6Al-4V blocks of similar dimensions were
[24]
the need to establish standards for sharing large dataset fabricated in an EOSINT M280 system using a fiber laser
of AM processing conditions to accelerate advancements power of 200 W and spot size of 80 µm, and power density
2
[25]
in ML applications to improve AM . A recent review by reach ~40 kW/mm using standard EOS Ti-6Al-4V
Nasiri and Khosravani (2021) presented opportunities for parameters with raster scanning and a hatch distance of
applying ML methods to understand fracture behavior 100 µm. All specimens had built orientations along the
of AM parts . In addition, a recent report by Sing et al. Z-axis, which is the direction of the smallest dimension.
[26]
(2021) established opportunities to integrate ML methods Additional L-PBF samples in the same build direction
in both upstream (i.e., part design and file preparation) were fabricated for heat treatment and for residual stress
and downstream (i.e., in-process monitoring) . relief. As per AM standards, samples were heated under
[27]
vacuum to a temperature within the range of 899~927 ±
It is evident that there is a need for a systematic
framework to quantify the heterogeneity in Ti-6Al-4V 14°C (1650~1700 ± 25°F), held for 2–4 h and argon cooled
material structures processed through varied AM and to below 427°C (800°F), then heated again to 538 ± 14°C
post-AM conditions to better understand the PSP linkages. (1000 ± 25°F) for 4 h in vacuum followed by Argon cooling
In this study, statistical functions are used to represent to room temperature. To eliminate the potential effects
scanning electron microscopy (SEM), electron backscatter of part location across build plates, specimen locations
diffraction (EBSD) microstructure information, and were randomized. Furthermore, to eliminate the effects of
residual stress captured from X-ray diffraction (XRD). proximity to build plate, samples for characterization were
According to Chen and Guestrin (2016), a novel ML harvested from the center of the part from the topmost
tool was developed to construct a reusable S-P linkage to layer (Figure 3).
predict the machining behavior of as-AM and heat-treated Representative AM Ti-6Al-4V specimens were
PBF Ti-6Al-4V . In addition to the novel PSP linkage, sectioned from the build platform using wire EDM.
[28]
this study employed a comprehensive dataset to generate Samples representing the surface parallel and vertical to
an aggregate database that is reflective of all PBF processing build orientation on all EB-PBF, as-AM L-PBF, and heat-
of Ti-6Al-4V alloys, as shown in Figure 1: (1) 200 SEM treated L-PBF specimens were mounted in the epoxy
images per material, that is, L-PBF with and without heat model for mechanical polishing to achieve 0.5 µ grading
treatment – HT, EB-PBF per surface, that is, parallel-XY surface finish. A final ion milling was applied to prepare
and along-XZ build orientation; (2) 3 XRD per material the sample surfaces for EBSD testing using a Thermo
per surface; and (3) 3 EBSD per material per surface. Scientific Apreo SEM with an Oxford Instruments
TM
Measurements were conducted from three different EBSD detector. Kroll’s reagent was then applied for 30 s
samples from the same processing conditions. to etch all samples, revealing all grain boundaries for
In summary, the overall goal of the study is to provide SEM microstructure observation. Other samples from
a novel framework for AM Ti-6Al-4V machining by the same batch of fabrication representing all the surfaces
developing PSP linkages to link microstructures to the described above were also used in X-ray residual stress
corresponding machining behavior, based on the specific measurements.

Volume 1 Issue 1 (2022) 4 https://doi.org/10.18063/msam.v1i1.6