Content uploaded by Sabine Bodner
Author content
All content in this area was uploaded by Sabine Bodner on Apr 28, 2020
Content may be subject to copyright.
Inconel-Steel Multilayers by Liquid Dispersed Metal Powder Bed Fusion :
Microstructure, Residual Stress and Property Gradients
S.C. Bodner,1 , ∗B. van de Vorst,2J. Zalesak,3J. Todt,3J.F. Keckes,3V. Maier-Kiener,4
B. Sartory,5N. Schell,6J.W. Hooijmans,7J.J. Saurwalt,7and J. Keckes1
1Department of Materials Science – Chair of Materials Physics,
Montanuniversit¨at Leoben, A-8700 Leoben, Austria
2TNO, 5656 AE Eindhoven, the Netherlands
3Erich Schmid Institute of Materials Science, Austrian Academy of Sciences, A-8700 Leoben, Austria
4Department of Materials Science, Chair of Physical Metallurgy and Materials Testing,
Montanuniversit¨at Leoben, A-8700 Leoben, Austria
5Materials Center Leoben GmbH, A-8700 Leoben, Austria
6Helmholtz Zentrum Geesthacht, Centre for Materials and Coastal Research, D-21502 Geesthacht, Germany
7Admatec Europe BV, 5051 DV Goirle, the Netherlands
Synthesis of multi-metal hybrid structures with narrow heat affected zones, limited residual
stresses and secondary phase occurrence represents a serious scientific and technological challenge.
In this work, liquid dispersed metal powder bed fusion was used to additively manufacture a mul-
tilayered structure based on alternating Inconel 625 alloy (IN625) and 316L stainless steel (316L)
layers on a 316L base plate. Analytical scanning and transmission electron microscopies, high-energy
synchrotron X-ray diffraction and nanoindentation analysis reveal sharp compositional, structural
and microstructural boundaries between alternating 60 µm thick alloys’ sub-regions and unique
microstructures at macro-, micro- and nano-scales. The periodic occurrence of IN625 and 316L sub-
regions is correlated with a cross-sectional hardness increase and decrease and compressive stress
decrease and increase, respectively. The laser scanning strategy induced a growth of elongated
grains separated by zig-zag low-angle grain boundaries, which correlate with the occurrence of zig-
zag cracks propagating in the growth direction. A sharp h110ifiber texture within the 316L regions
turns gradually into a h100ifiber texture in the IN625 regions. The occurrence of the C-like stress
gradient with a pronounced surface tensile stress of about 500 MPa is interpreted by the temper-
ature gradient mechanism model. Chemical analysis indicates a formation of reinforcing spherical
chromium-metal-oxide nano-dispersoids and demonstrates a possibility for reactive additive manu-
facturing and microstructural design at the nanoscale, as a remarkable attribute of the deposition
process. Finally, the study shows that the novel approach represents an effective tool to combine
dissimilar metallic alloys into unique bionic hierarchical microstructures with possible synergetic
properties.
Keywords: Powder Bed Fusion; Microstructure and Stress; Multi-Metal Material; Stainless Steel; Inconel
I. INTRODUCTION
Multi-material hybrid structures in nature and tech-
nology combine properties of their individual con-
stituents throughout complex microstructures to achieve
synergistic effects, like improved mechanical properties
and oxidation resistance, which go beyond the proper-
ties of their monophasic counterparts [1–3]. Typically,
multilayered, composite and hierarchical microstructures
are used to achieve improved functionality by combining
materials with different physical properties [2,4–7]. Sim-
ilarly, functionally graded materials (FGMs) are designed
to comply with spatially variable functional requirements
like strength, oxidation resistance or toughness by adopt-
ing gradients of dissimilar phases, microstructure and/or
residual stresses [8–10]. However, joining of metals into
multi-metal hybrid structures with diffuse or abrupt in-
∗sabine.bodner@unileoben.ac.at
terfaces between the individual phases or alloys and nar-
row heat affected zones resulting in limited the first-or-
der residual stress, macroscopic distortion and secondary
phase formation represents still a serious technological
challenge [11].
By their principle of a sequential layer-by-layer deposi-
tion, additive manufacturing technologies (AM) are ide-
ally suited to manufacture parts with complex external
and internal geometries [3,12–14]. Additionally, pow-
der and wire feed material supplies have been known to
allow for an adjustment of volume fraction of metallic
components [15–17]. In particular, the directed energy
deposition (DED) approach, in which powder is fed into
the melt-zone and molten at every layer by a laser, can
be used to fabricate components with a variable layer-
by-layer composition and unique microstructures [18].
Within this contribution, however, a novel laser powder
bed fusion-related technology, based on liquid dispersed
metal powder, is used to produce a multi-metal hybrid
structure.
2
The fabrication of FGMs using AM technologies allows
for generating property-specific part areas by matching
the process parameters to the localized functionality [19].
It is however common that combining dissimilar metals
using AM technology may result in the formation of com-
positional and microstructural inhomogeneities as well as
in secondary phases and stress concentrations, which may
decisively deteriorate part’s functional properties [20–22].
A significant effort has been devoted to the devel-
opment of multi-material additive manufacturing (MM-
AM) of hybrid structures, reviewed recently in Ref. [3].
Exemplary AM systems equipped with several powder
feeders were used to prepare graded Ti-V [23,24], Ti-
Mo [24], V-Ti6Al4V alloy [25], stainless steel-Ti6Al4V
alloy [26] and Ti-6Al-4V/Ti-6.5Al-3.5Mo-1.5Zr-0.3Si [27]
materials with novel refined microstructural features,
aiming at aircraft and aero-engine applications. Simi-
larly, bimetallic structures like Inconel 718-copper alloy
GRCop-84 [28] and stainless steel 304L-Invar 36 [25] as
well as gradient structures like Ti-TiO2[29] and yttria-
stabilized zirconia-steel [30] were fabricated.
In the field of nuclear and steam power plants,
aerospace and/or repair applications, especially high
strength, corrosion and oxidation resistance and/or creep
and fatigue resistance are required [31,32]. For this rea-
son, MM-AM of stainless steel (SS) and Inconel have
been extensively explored [33,34]. Carroll et al. [35] re-
ported on fabrication of Inconel-steel FGM with diffuse
compositional, structural and microstructural bound-
aries by applying gradually varying mixtures of Inconel
625 and grade 304L SS powders. The observed forma-
tion of cracks with a length of several hundred microns
in a region with 79 wt% SS 304L and 21 wt% IN625
was attributed to the presence of secondary phases of
transition metal carbides in the form of (Mo,Nb)C. Sim-
ilarly, aiming at nuclear fission reactor applications, Hi-
nojos et al. [36] studied the joinability of Inconel 718
and 316L SS and vice versa utilizing electron beam melt-
ing (EBM) and reported abrupt compositional, structural
and microstructural boundaries as well as low concentra-
tion of typical welding features. Cracking observed at
the 316L/IN718 interface was attributed to the deforma-
tion constrains imposed by IN718. Additionally, using
finite-element numerical analysis, Hofmann et al. [9]
showed that a gradient transition from SS 304L to IN625
across an automobile valve stem are expected to exhibit
ten times lower stress at the transitioning zone at 1000K
compared to a friction stir-welded joint of the same ma-
terials.
In general, all previous studies on FGMs combining
nickel-based alloys and steel indicated complex process-
microstructure-stress-properties relationships [35,36].
Additionally, FGMs produced with diffuse and abrupt
interfaces comprise usually unique microstructures with
secondary phases and precipitates like (Mo,Nb)C,
M23C6, NbC [35,36]etc. and there were cracks with
lengths up to several 100 µm observed systematically
in the transition regions [35]. These findings indicate
that there is a significant potential to further opti-
mize the functional properties of the particular nickel-
based alloys-steel FGMs by knowledge-based microstruc-
tural design, including a development of novel deposition
routes and recipes.
For this work, the non-precipitation hardening alloy
Inconel 625 (IN625) and stainless stee grade 316L (316L)
are selected to fabricate a multi-metal hybrid structure.
A relatively novel deposition technique based on liquid
dispersed metal powder bed fusion is used to form a com-
plex microstructure with multilayered morphology, zig-
zag grain boundaries, multiscale interfaces and ceramic
nanoparticle reinforcement. The three-fold work objec-
tives encompass (i) exploring the possibilities and ad-
vantages of the liquid powder bed fusion technology, (ii)
analyzing gradients of phases, microstructure, composi-
tion and mechanical properties at multiple length-scales
using correlative cross-sectional micro-analytics and (iii)
additive manufacturing of multi-metal hybrid materials.
Additionally, reactive additive manufacturing as an inte-
gral part of the liquid dispersed metal powder bed fusion
process is introduced in order to indicate the possibility
to tailor mechanical properties of the grown structures
by integrating homogeneously dispersed nanoparticles.
II. MATERIALS AND METHODS
For the synthesis of a multi-layered IN625-316L struc-
ture on a 3 mm thick 316L stainless steel base plate,
a prototype system LASERFLEX Conflux based on liq-
uid dispersed metal powder bed fusion (Fig. 1) was
used. As input material, AISI 316L (specified with
∼[65,12,18,2] % of Fe, Ni, Cr and Mo, respectively, as
well as other balancing elements with less than ∼1 % con-
centration) and IN625 (specified with ∼[58,20,8,4] % of
Ni, Cr, Mo, Fe and Nb, respectively, as well as other
balancing elements with less than ∼1 % concentration)
metal particles with d50-diameters of ∼4.2 and ∼3.4µm,
respectively, were applied. In order to produce two pow-
der suspensions, the particles were dispersed in an aque-
ous solution enriched with an ethanol-based micropoly-
mer glue. Individual IN625 and 316L layers were pro-
duced by (i) lowering the building platform by ∼30 µm,
(ii) depositing the particular suspension, (iii) wiping the
metal suspension across the building platform, (iv) wait-
ing for a few seconds in order to dry and densify the
liquid powder bed in the inert nitrogen atmosphere and
(v) melting the dried metal powder in the nitrogen at-
mosphere, using a laser source (cf. Fig. 1). The laser
system and its optics provides radiation with a wave-
length of 1064 nm and a spot size of ∼12 µm. A laser
power of 137 W, a scan speed of 500 mm/s, a hatch dis-
tance of 90 µm and a hatch rotation angle of 66 degrees
between the layers were used. These process parameters
were identical for both alloys. The temperature of the
base plate was kept at ∼25℃. After the laser melting
the described procedure from (i) to (v) was repeated to
3
produce the next layer. For further microscopy analysis,
the structure’s cross-section was mechanically polished
using a standard colloidal silica polishing suspension by
Struers with a particle size of 0.04 µm (O-PS polishing
suspension).
FIG. 1. A schematic description of the liquid dispersed metal
powder bed fusion process. Within the first step, a suspension
layer with a thickness of ∼30 µm containing metallic 316L or
IN625 particles is applied on the 316L base plate or on already
existing build-up. Then, the suspension dries in a nitrogen
flow and densifies to a thickness of ∼25 µm. Finally, a laser
source is used to fuse the particles together and to produce a
new metallic layer of ∼20 µm in thickness, which is reinforced
with dispersed nanoceramic particles (cf. Sec. I II F).
The approach from Fig. 1was used to grow two multi-
layered IN625-316L plate structures, identical in their
morphology, with overall dimensions of ∼(15×2×7) mm
each, on a 316L SS base plate with dimensions of ∼
(40 ×40 ×3) mm (Fig. 2a). Both structures possessed a
complex irregular cross-sectional morphology introduced
in Fig. 2b. The aim behind selecting this specific mor-
phology was primarily to evaluate the accuracy of the
deposition approach to fabricate relatively thin mono-
alloy layers, to obtain information on the intermixing of
IN625 and 316L alloys, to analyze microstructural evo-
lution at and near inter-alloy interfaces and to evalu-
ate cross-sectional residual strain across the structure.
Further on, experimental results obtained only from one
structure will be presented.
Optical microscopy (OM) characterization was per-
formed using an Olympus BX51 system. Scanning elec-
tron microscopy (SEM), electron backscatter diffraction
(EBSD) and energy dispersive X-ray analysis (EDX)
were performed using a Gemini SEM 450 and an
AURIGA CrossBeam systems from Carl Zeiss. A
cross-sectional transmission electron microscopy (TEM)
lamella with a thickness of ∼50 nm was fabricated us-
ing focused ion beam (FIB) machining within the Zeiss
Auriga system by applying an acceleration voltage of
30 kV and currents in the range from 20 nA to 50 pA.
The lamella was extracted from a sample region contain-
ing an interface between the two alloys. Microstructural
and chemical characterization of the lamella was per-
formed using scanning TEM (STEM) mode in a probe-
corrected FEI Titan Themis platform operated at 200 kV
and equipped with a Gatan Enfinium ER spectrometer
as well as FEI Super-X EDX four quadrant detectors.
The collected EDX signal was treated using Bruker Es-
prit software applying built-in standards.
Nanoindentation experiments were conducted using a
platform Nanoindenter G200 with a Berkovich diamond
tip in strain-rate controlled mode. A sinusoidal load sig-
nal was superimposed during the continuous sample load-
ing in order to record the contact stiffness. Tip shape and
frame stiffness calibrations were performed on a regular
base according the Oliver and Pharr method [37]. Ar-
rays of indents with an indentation depth of ∼1µm and
a minimum distance of ∼20 µm between the indents were
positioned over the R1–R4 regions. Hardness and inden-
tation moduli were evaluated for the four regions as well
as for bulk IN625 and 316L.
Synchrotron characterization of the multi-layered sam-
ple was performed at the high energy materials sci-
ence (HEMS) beamline P07B of PETRA III at DESY
in Hamburg, which is operated by Helmholtz Zen-
trum Geesthacht. The experimental setup is presented
schematically in supplementary data. The sample char-
acterization was performed using a monochromatic beam
of 87.1 keV photon energy and a cross-section of ∼
20 ×500 µm, which was directed approximately parallel
to the interfaces between IN625 and 316L. To perform
a scanning cross-sectional X-ray micro-diffraction exper-
iment (CSmicroXRD), the sample was moved along the
growth direction with an increment of 20µm. Almost 400
diffraction patterns, each with multiple Debye-Scherrer
(DS) rings, were recorded from the sample’s cross-section
in transmission geometry using a Perkin Elmer two-
dimensional (2D) flat panel detector of 2048 ×2048 pix-
els with a pixel pitch of 200 µm. The sample-detector
distance was set to ∼1.3 m. A dark current correc-
tion was applied to each exposure. The exact geometry
and distance of the detector with respect to the sam-
ple’s position was determined using a LaB6 calibration
standard powder. Diffraction data were treated using
the pyFAI software package developed at the European
Synchrotron Radiation Facility [38] and evaluated using
custom scripts written in Python.
4
FIG. 2. Synthesis and geometry of the IN625-316L structure. (a) Various stages of the fabrication process to produce two
identical multilayered plates. (b) On a 316L base plate, ∼1 mm thick layer of 316L was grown followed by four regions R1–R4
including IN625, which are always separated by ∼1 mm of 316L. The deposition was finalized by ∼1 mm of IN625 on top. As
indicated in (b) left, the R1–R4 regions consist of 3 (R1) and 5 (R2) layers of IN625, alternating IN625 and 316L layers in R3
and alternating groups of three IN625 and three 316L layers (R4). The thickness of every individual layer was set to ∼20 µm
(cf. Fig. 1).
III. RESULTS
A. Cross-sectional morphology
The morphology of the polished IN625-316L structure
cross-section was analysed using OM and SEM. The OM
micrograph in Fig. 3shows the base plate, the five ∼
1 mm thick regions of 316L, the regions R1–R4 and a
∼1 mm thick IN625 region on top (cf. also Fig. 2b).
The interfaces between the layers appear to propagate
parallel to the base plate surface and indicate that the
microstructure is laterally homogeneous. The interface
between the multi-layered structure and the base plate
does not show any features of intermixing or dilution.
A variety of cracks were found at the structure’s cross-
section with a very typical zig-zag morphology. A rep-
5
resentative example of one zig-zag crack is shown in the
inset of Fig. 3(cf. also OM and SEM micrographs in sup-
plementary data). Comparable zig-zag cracks were found
in the regions R1–R4 as well as in the top-most ∼1 mm
thick IN625 region. Their occurrence can be correlated
with the presence of IN625. The cracks propagated across
the layers of both alloys within regions R1–R4 and ap-
proximately in the structure’s growth direction, their ori-
gins will be discussed in Secs. III B and III C.
FIG. 3. OM micrograph showing the multi-layered IN625-
316L structure cross-section with ∼3 mm base plate, five
∼1 mm thick 316L regions separated by the regions R1–
R4 with IN625 layers and layer-groups, respectively, and an
IN625 region on top. The base plate’s surface exhibits a con-
cave bending. The inset at the right shows the cross-sectional
morphology of a typical crack (cf. supplementary data).
The OM micrograph in Fig. 3shows that the upper
surface of the base plate shows a slight concave bending,
which indicates the presence of compressive and tensile
residual stress states within the top and bottom regions
of the plate, respectively [39]. Consequently, it can be
expected that the near-base plate region of the struc-
ture possesses tensile stresses which change gradually into
compressive at the structure’s surface. The stress forma-
tion is a result of the structure fabrication and its origins
will be discussed in the next section. Fig. 3as well as all
SEM and TEM micrographs presented along the text are
shown in full resolution in supplementary material.
B. CSmicroXRD analysis
In order to obtain grain-averaged information on the
cross-sectional distribution of crystalline phases, tex-
ture and strain across the multi-layered structure, CSmi-
croXRD was carried out and volume-averaged data from
the 2 mm thick sample (Fig. 2a) (along the X-ray beam
direction) were collected. XRD data obtained by an az-
imuthal integration of the complete Debye-Scherrer rings
in Figs. 4a and b show a cross-sectional evolution of
IN625 and 316L reflections. Since both alloys possess
face-centered cubic (fcc) lattice and lattice parameters of
stress-relieved IN625 matrix and 316L steel differ only
a few percent (depending on the actual alloy composi-
tion), fcc reflection positions (modified by the residual
stress gradient) in Fig. 4a vary only marginally. More-
over, no other reflections from secondary intermetallic
phases like γ0,γ00 and δin IN625 [40] or ferrite and car-
bide phases in 316L [41] can be identified in this phase
plot. It should be however noted that quantitative XRD
analysis is not sensitive to secondary phases with volume
fractions smaller than 3–10 %, depending on the crystal-
lite size. The evolution of IN625 and 316L 311-reflections
in Fig. 4b reveals also the presence of the individual R1–
R4 layers, the top IN625 region as well as the base plate.
The data in Fig. 4b allowed resolving also the presence
of five groups of layers within region R4, which correlate
with the multiple alternation of three IN625 and three
316L layers.
The layer-by-layer laser melting process, which is ac-
companied by the generation of localized high thermal
loads during the metal powder bed fusion (Fig. 1), results
in the formation of complex multiaxial residual stress dis-
tributions [42] whose tensile stress concentrations may
significantly influence the mechanical stability of the fab-
ricated parts. In the case of multi-metal hybrid struc-
tures like in Fig. 3, (i) thermal stresses are expected to
be formed as a result of the mismatch of respective coef-
ficients of thermal expansion (CTEs) of ∼15 ×10−6K−1
[43] and ∼18 ×10−6K−1[44] at 100 ℃between IN625
and 316L regions. Additionally, (ii) growth stresses are
formed as a result of complex cross-sectional microstruc-
ture and stress evolution during the fabrication pro-
cess accompanied by cyclic elasto-plastic deformation.
As the presence of residual stress was indicated by the
base plate’s bending (Fig. 3), scanning CSmicroXRD was
used to assess cross-sectional strain-stress distributions
in the structure. The motivation was to obtain quali-
tative data and assess the role of particular microstruc-
tural features in the formation of stress. IN625 and 316L
311-Debye-Scherrer rings collected from the individual
cross-sectional structure positions were evaluated with
regard to their ellipticity. The obtained data were used
to estimate depth gradients of X-ray elastic strains and
residual stresses. For the evaluation, it was supposed
that the shear X-ray elastic strains ε311
ij and shear resid-
ual stresses σij can be neglected for simplicity. In this
case, the measured ellipticity is proportional to the dif-
ference of in-plane and out-of-plane residual stresses of
σ11 −σ33 and X-ray elastic strains ε311
11 −ε311
33 , which
were evaluated using the Hill grain interaction model ap-
plying appropriate X-ray elastic constants [45,46]. Since
the out-of-plane stresses σ33 can be supposed to be rela-
tively small, due to the free sample surface, the evaluated
σ11 −σ33 values from Fig. 4c can be therefore used to de-
scribe an evolution of the in-plane stresses σ11 across the
structure’s cross-section.
The surface region of the base plate is in compression
[39] and the stress profile across the multi-layered struc-
ture exhibits a complex oscillatory stresses behaviour,
which is superimposed on a dominant C-shaped stress
depth dependence (Fig. 4c). According to the tempera-
ture gradient mechanism (TGM) model [39,42,47], the
6
FIG. 4. Results from CSmicroXRD on the multi-layered structure. (a) The phase plot reveals only the presence of fcc reflections
of IN625 and 316L alloys with very similar lattice parameters. (b) The evolution of IN625 and 316L 311 reflections shows the
positions of R1–R4 regions, the top IN625 layer and the base plate. (c) The strain/stress plot indicates a dominant C-shaped
stress depth dependence with compressive stress relaxations at the positions of R1–R4 regions. (d–g) The texture plots show
the distributions of diffraction intensities along Debye-Scherrer rings for IN625 and 316L 220, 111, 200 and 311 reflections.
C-shaped stress profile across the multi-layered structure
can be interpreted by the effect of the cyclic heating of the
mechanically restrained preceding metallic layers, which
results in the formation of compressive stresses and ma-
terial plastification. During cooling down, tensile stresses
are formed in the plastified surface zone (like in the top-
most ∼1 mm thick IN625 region in Fig. 4c), which are
balanced by compressive stresses in the build-up zone be-
neath and in the base plate. Additionally, the oscillatory
stress behaviour in Fig. 4c indicates that the growth of
316L layers results systematically in an increase of com-
pressive stresses at all five cross-sectional positions in the
structure. Conversely, the deposition of IN625 results in
a systematic relaxation of compressive stresses. The re-
sulting compressive stress oscillations can be interpreted
by a higher degree of plastic strain generation in 316L
compared to IN625, which is due to a lower yield strength
of 316L, as well as by a stress relaxation in IN625 man-
ifested by the formation of zig-zag cracks, as visible in
Fig. 3. It is clear, however, that this qualitative explana-
tion of the stress gradient formation is very simplified and
the formation depends on a variety of process and ma-
terial parameters like heat source path and heat transfer
to the underlying layers, temperature distributions, lo-
cal thermal conductivity, local yield strength and local
CTEs. Moreover, due to their unique microstructures,
materials’ parameters are temperature- and, thus, laser
power-dependent [17].
7
The plots in Figs. 4d–g show azimuthal distributions of
diffraction intensities along IN625 and 316L 220, 200, 111
and 311 Debye-Scherer rings and document thus a qual-
itative evolution of crystallographic texture across the
structure. Since the angle 0 and 90 degrees in Figs. 4d–g
correspond to the out-of-plane and in-plane orientations
of the diffraction vectors, respectively, the results indi-
cate the presence of h110ifibre texture, which evolves
gradually in the first 316L layer and saturates within
∼0.4 mm above the base plate. Thereafter, it is pre-
served in all subsequent 316L layers (Fig. 3d). This is
documented for instance by the strong azimuthal max-
ima of 220, 200 and 111 reflections at ∼0°,∼45°and
∼35°, respectively. These maxima correspond to the an-
gles of ∼45°and ∼35°between (110) and (100) as well
as (110) and (111) crystallographic planes, respectively.
The occurrence of azimuthal maxima broadening at the
positions of R2, R3 and R4 regions indicates that the
addition of IN625 weakens the h110ifibre texture, which
practically disappears or even turns into a h100ifibre
texture within the top IN625 sublayer. The results also
reveal that the base plate is not free from texture, but
there is a certain type of a weak biaxial texture. This is
represented in Figs. 4d, e by the occurrence of weak max-
ima at ∼0 and ∼90 degrees for 220 and 111 reflections,
respectively.
C. Local texture analysis
In Fig. 5, EBSD micrographs show the cross-sectional
evolution of crystallite orientations and microtexture
along the in-plane and out-of-plane sample axes [100] and
[001], respectively, at the interface between the base plate
and the first 316L layer (Fig. 5e, f) and across the regions
R1 (Fig. 5c, d) as well as R4 (Fig. 5a, b). The data in
Fig. 5e document that the apparently homogenous mi-
crostructure of the base plate with globular grains with
an average size of ∼12 µm in diameter rapidly turns into
a textured microstructure with elongated grains and an
average grain length of ∼108 µm within the first 316L
layers. In case of all five ∼1 mm thick 316L layers,
the EBSD data indicate the presence of a relatively pro-
nounced h110ifibre texture with random in-plane crystal-
lite orientations (e.g. along [100] and sample direction),
in agreement with the data from Figs. 4d–g.
The region R1 (Figs. 5c, d) shows a very localized
grain refinement and a small change in the crystallite
orientation, which however seemingly does not influ-
ence the cross-sectional microstructure significantly. In
other words, the insertion of the three IN625 layers (cf.
Figs. 2b, 3) did not influence the epitaxial overgrowth of
the individual layers resulting in the preservation of the
columnar grain microstructure as well as texture. This
is in agreement with the XRD results from Figs. 4d–g,
where also no significant texture disruption was observed
within the region R1. The behaviour of the preserved
crystallite microtexture can be interpreted by very simi-
lar lattice parameters of both alloys, differing only a few
percent, which clearly promoted a heteroepitaxial over-
growth.
The EBSD data from the region R4, however, indicate
changes in the crystallite orientation (Figs. 5a, b) caused
by the addition of IN625 sublayers, namely the h110ifi-
bre texture was gradually disrupted as can be seen espe-
cially in Fig. 5b. This observation is in agreement with
the XRD data from Figs. 3d–g, which also showed that
the deposition of multiple IN625 layers induces a gradual
texture randomization.
Moreover, the EBSD results from Fig. 5document a
zig-zag cross-sectional morphology of elongated grains
(and low-angle grain boundaries with a misorientation
angle less than about 15 degrees), which is obtained as a
result of the applied scanning strategy. Grain-boundary
maps are also presented in the supplementary material.
This zig-zag morphology can be correlated to the zig-zag
cracks patterns observed using SEM and OM especially
at the cross-sectional positions of the regions R1–R4 (cf.
inset in Fig. 3). The zig-zag cracks are expected to orig-
inate from the particular shape of the grain boundaries
and/or in-plane residual stresses induced by CTEs mis-
match between the sublayers.
D. Chemical analysis
SEM and EDX analyses were performed in order to
characterize microstructure and chemistry of both alloys
within the individual layers. The results indicate a ho-
mogenous distribution of elements in both phases and,
additionally, no microscopic precipitation resulting in a
localized elemental enrichment or depletion, in agreement
with the CSmicroXRD data (Fig. 4). Moreover, no for-
mation of microscopic dendritic regions was observed at
the microscale. This is in contrast to the results from sim-
ilar studies on the IN718-304L systems, where the forma-
tion of microscopic enrichments in Nb and Mo, with a few
shared areas containing C, was observed inter- and intra-
granularly in IN718 and IN625 [35,36]. This discrepancy
can be interpreted by the different thermal treatments
applied in Refs. [35,36] and in the present process.
EDX concentration mapping of the region R4 was per-
formed to examine the intermixing of the elements be-
tween IN625 and 316L sub-regions during the multilayer
structure fabrication. In Fig. 6, distributions of major el-
ements indicate the presence of ten sub-regions, each con-
sisting of three layers of IN625 or 316L. Both deposited
alloys contain Fe, Ni, Cr and Mo elements, whereas Nb
is initially present only in IN625. In the Nb concentra-
tion map in Fig. 6, it is possible to resolve five Nb-rich
sub-regions representing IN625 and four 316L sub-regions
with a smaller amount of Nb. Moreover, there is evidence
of a higher concentration of Nb in the ∼1 mm thick 316L
region above region R4, compared to the Nb (EDX back-
ground) concentration level in the ∼1mm thick 316L
region, below region R4. These observations indicate
8
FIG. 5. EBSD data show the orientation of the crystallites with respect to the in-plane and out-of-plane sample axes [100] and
[001], respectively, at the interface between the base plate and the first 316L layer (e, f) and across the regions R1 (c, d) as
well as R4 (a, b). These microtexture data indicate the presence of a columnar grain morphology and a h110ifibre texture in
both alloys (cf. also supplementary data). The zig-zag patterns in the microstructure are indicated by black, dotted lines and
white arrows (a, c, d). The dark arrow in (a) indicates the growth direction.
that after the deposition of IN625 layers, the elements
of this alloy were distributed along the growth direction
upwards as a result of out-diffusion into the melt-pool.
Probably, the concentration of the IN625 alloying ele-
ments decreases exponentially with distance from IN625
layers.
The Ni and Fe concentration maps in Fig. 6indicate
very clearly the presence of five IN625 and five 316L
layer-groups in region R4. In both types of sub-regions
a simultaneous occurrence of Fe and Ni was evaluated.
This effect can be attributed to an intermixing of the
two elements between the deposited sub-regions and to
the (intrinsic) simultaneous occurrence of the elements
in both alloys. The following effects are remarkable how-
ever: The top-most sub-region of three layers of 316L
in region R4 (cf. Fig. 2b) shows lower and higher con-
centrations of Fe and Ni, respectively, compared to the
neighboring top-most ∼1 mm thick 316L region (Fig. 6).
Similarly, also the bottom-most sub-region of three 316L
layers in region R4 (following the first IN625 layer, cf.
Fig. 2b) shows lower and higher concentrations of Fe
and Ni, respectively, compared to the underlying ∼1 mm
thick 316L region, which appears to be almost free of Ni
(Fig. 6). These observations also indicate that growth
of the IN625 and 316L sub-regions was accompanied be
cyclic remelting of the alloys and an intermixing of the el-
ements between the layer being deposited and preceding
layers.
Since IN625 contains an amount of Mo that is approx-
imately two times higher than that in 316L steel, Mo
enrichments in the concentration map in Fig. 6can be
correlated with the occurrence of IN625 layers in region
9
FIG. 6. EDX concentrations maps show the distribution of elements in the area of region R4. The arrow in the SEM micrograph
indicates the building direction.
R4. The homogenous distribution of Cr across the region
R4 (Fig. 6) correlates well with ∼20 % amount of Cr in
both alloys (cf. also Sec. II I E).
An attempt was made to relate the occurrence of the
cross-sectional cracks (Fig. 3) with the chemical gradients
across the multilayered sample [35], but no such correla-
tion was observed. In other words, a local fluctuation of
the elemental concentrations was not found to be respon-
sible for the zig-zag crack growth behavior, in contrast to
the results of Carroll et al. [35], where microscopic Nb-
and Mo- enrichment was observed within the cracked re-
gions in IN625-304L system.
E. Mechanical characterization
Nanoindentation measurements were performed across
the R4 region in order to evaluate the spatial distribu-
tion of hardness and reduced elastic modulus (the latter
is presented in supplementary material). In Fig. 7, the
cross-sectional distribution of hardness in the range of
∼(3–4) GPa is superimposed on an occurrence of Ni de-
termined using EDX analysis (from Fig. 6). The data
indicates that the multiple hardness increase can be cor-
related with the occurrence of Ni within the five groups
of IN625 layers. In four sub-regions, each consisting of
three 316L layers, as well as on the borders of region
R4, the hardness decreases systematically. The regions
R1–R3 showed the same behavior, where the insertion of
IN625 interlayers caused an increase in hardness.
Hardness measurements were performed also on the
sample (bulk) regions which consist exclusively of IN625
and 316L, where hardness values of 3.95 ±0.08 and
2.81 ±0.08 GPa were found, respectively, in relatively
good agreement with the literature values of ∼4.3 GPa
[48] and ∼3.5 GPa [49]. This means that the bulk IN625
and 316L regions possess lower and higher hardness than
the IN625 and 316L sub-regions in region R4, respec-
10
FIG. 7. Cross-sectional distribution of hardness across the
region R4 superimposed on the elemental distribution of Ni
from Fig. 6. The hardness increase correlates with the posi-
tions of five IN625 sub-regions.
tively. This finding can be interpreted by the elemental
intermixing (cf. Fig. 6) and/or by different chemical com-
positions of CrMOxdispersoids formed within region R4
(cf. Sec. II I F).
F. TEM characterization
TEM characterization was performed in order to ana-
lyze the local microstructure and chemical composition
across an exemplary interface, between the upper part
of the ∼1 mm thick 316L region and the topmost IN625
region (Figs. 2b, 3). Bright field and high-angle annu-
lar dark-field (HAADF) micrographs revealed the pres-
ence of morphologically sharp interface between both al-
loys with a large amount of spherical nanodispersoids
distributed throughout the interface region and an oc-
currence of heavy elements primarily at the IN625 grain
boundaries, as documented by the HAADF micrograph
in Fig. 8.
However, EDX analysis of the interface indicated grad-
ual changes in the concentrations of Fe and Ni at a
length scale of several ∼100 nm across the interface
(Fig. 8). At the IN625 side, grain boundaries were re-
vealed to contain enrichments of Nb, Mo and C as result
of nano-segregation, which is usually observed after ther-
mal treatments of Inconel alloys [39]. At the 316L side,
the segregation of Nb, Mo and C at the grain boundaries
was observed to a much smaller extent and Nb as well as
Si enrichments within the dispersoids were detected (cf.
Fig. 8).
Also a detailed EDX analysis of the spherical disper-
soids with a diameter of ∼20–100 nm was performed
(Fig. 9) at both sides of the interface. As documented
by EDX maps in Fig. 9, the dispersoids were identified
to be chromium-metal-oxide nanoparticles CrMOx en-
riched with Si and exhibiting Nb-rich shells. Generally,
the dispersoids were found in both alloys across the whole
structure at high concentrations and their actual compo-
sition depends on the chemistry of the particular alloy’s
matrix. This indicates that the dispersoid’s formation is
a characteristic attribute of the particular fusion process
and resembles the formation of protective oxide scale rich
in Cr in stainless steels with a relatively large fraction
of Cr [50]. In other words, the liquid dispersed metal
powder bed fusion induces an oxidation of Cr present
in both alloys, resulting in the formation of dispersoids
with some traces of Si and Nb (Fig. 9). This phenomenon
will be denoted as reactive additive manufacturing. The
ceramic dispersoids represent nanoscopic alloy reinforce-
ments, which are expected to be favorable for the me-
chanical properties of the structure. Therefore reactive
additive manufacturing as a part of the liquid dispersed
metal powder bed fusion can be used as an effective tool
to homogeneously incorporate nanoscopic particles across
deposited structures in order to tailor mechanical prop-
erties.
IV. DISCUSSION
This study demonstrates the feasibility of fabricating
multi-layered IN625-316L structures (Figs. 2b, 3) with
sharp structural and microstructural boundaries between
the two alloys, deposited using liquid dispersed metal
powder bed fusion (Fig. 1), where the alloys alternation
was achieved at meso- and micro-scales (Figs. 3,6). The
implementation of regions with periodically varying me-
chanical properties is expected to hinder crack propaga-
tion and may results in the synthesis of damage-resistant
materials [51,52]. Due to the out-diffusion of alloying
elements in the growth direction, however, an elemen-
tal intermixing was observed at length scales of several
100 nm (Fig. 8). The results from the multilayered re-
gion R4 (Figs. 3–6) document that the approach allows
for the deposition of spatially alternating alloys with the
resolution of ∼60 µm in the growth direction, which con-
sequently exhibits an oscillating hardness depth profile
with a period of ∼100 µm (Fig. 7).
The 316L layers possess a h110ifiber texture, which
saturates already in the first ∼1 mm thick 316L region
at a distance of ∼0.4 mm (Fig. 4e) above the base
plate. However, the deposition of IN625 alloy induces
a gradual disruption of the h110itexture and the for-
mation of a much weaker h100ifiber texture like in Ref.
[36], which is present especially in the top-most IN625
sublayer (Figs. 4d–g). The deposition of IN625 induces
moreover a refinement of the microstructure, which is
visible already in region R1 (Figs. 5c, d).
The occurrence of a C-shaped stress gradient with a
negligible near-base plate stress, the oscillating compres-
sive stresses at a level of ∼−150 MPa and the pronounced
surface tensile stress of ∼500 MPa (Fig. 3c) are inter-
preted by means of the TGM model [39,42,47], suppos-
ing cyclic material plastification and densification dur-
ing the additive deposition. The C-shaped stress profile
11
FIG. 8. EDX concentration maps show distributions of elements at the interface between 316L and IN625 alloys, which is
indicated by arrows in the HAADF micrograph. Besides gradual changes in Fe and Ni concentrations and in the composition
of CrMOxdispersions, Nb-rich precipitates change their shape as a function of the Nb-content in the base material.
is a result of the mechanical equilibrium achieved after
cooling down from the process temperature of compres-
sively and tensilely stressed and mutually restrained sur-
face and underlying bulk structure regions, respectively,
whose stress signs are expected to almost reverse at room
temperature. The deposition of IN625 layers results in
the relaxation of compressive residual stresses (Fig. 3c),
as a result of the supposed lower degree of plastic strain
generation in IN625 and zig-zag cracking, which pene-
trates layers of both alloys (cf. inset in Fig. 3and sup-
plementary data) along zig-zag low-angle-grain bound-
aries. This effect is denoted as “ductility dip crackin”and
appears as a result of IN625’s ductility decrease during
cooling down from the process temperature at straight
or non-tortuous boundaries [36,53].
EDX analysis in SEM (Fig. 6) indicated the absence
of microscopic dendrites or precipitates with metallic en-
richments/depletions zones in both alloys, in agreement
with the CSmicroXRD data (Fig. 4a). Also no cracks
were observed, which could be correlated with local-
ized elemental enrichment or depletion. This is differ-
ent from the results of Hinjos et al. [36] and Carroll et
al. [35], where microscopic precipitates and features were
reported in the same and/or similar alloys and which
were correlated also with the occurrence of microscopic
cracks. At the nanoscale, however, the EDX analysis per-
formed in TEM indicated the presence of Nb, Mo and C
enrichments caused by nano-segregation at IN625 grain
boundaries (Fig. 8). These observations indicate that
the liquid dispersed metal powder bed fusion process is
favorable for the deposition of complex alloys because it
is possible to suppress the formation of secondary phases
at the microscale. Probably, due to the shorter diffusion
lengths predefined by the particular laser design, the seg-
regation of elements was limited and occurred only to a
very small extent at the grain boundaries of IN625, as
shown in Fig. 8.
IN625 regions and the top layer showed the presence of
undesirable zig-zag cracks (Fig. 3), which may obviously
negatively influence mechanical stability of the IN625-
SS316L build-up. Therefore a further optimization of
the process is needed. The possibilities are to include
additives into the powder suspension that will help to
reduce solidification cracking. Another approach would
be to increase the lasers pulse frequency, inspired by the
results of Patterson et al. [54].
12
FIG. 9. EDX concentration maps show distributions of elements within representative dispersoids in IN625 (a) and 316L (b)
alloys near the interface shown in Fig. 8. The spherical CrMOxnanodispersoids possess Nb-rich shells and Si concentration
appears to depend on the matrix alloy (cf. also Fig. 8).
A large concentration of spherical dispersoids with a
diameter of ∼20–100 nm, existing across the whole struc-
ture, was identified as a remarkable attribute of the liquid
dispersed metal powder bed fusion process, which there-
fore includes also reactive additive manufacturing of ce-
ramic nanodispersoids. The dispersoids could be found in
both alloys and were identified mainly as CrMOxprecip-
itates, enriched with Si and Nb, depending on the local
alloy matrix chemistry. Their complex internal morphol-
ogy is presented in Fig. 9. The dispersoid’s formation is
interpreted by the oxidation of Cr and other elements and
resembles the formation of protective Cr2O3oxide layers
in Cr-rich stainless steels [50]. The dispersoids represent
another hierarchical level within the alloy’s microstruc-
ture and actually provide structural reinforcement. Their
formation indicates another interesting feature of AM
technologies, namely the possibility to perform reactive
additive manufacturing and microstructural design at the
nanoscale. In other words, the deliberate formation of ox-
ide nanoparticles during the liquid dispersed metal pow-
der bed fusion can be used as an effective tool to tune
mechanical properties of AM-prepared alloys.
The multilayered structure (Fig. 3) exhibits a variety of
bio-inspired attributes. The hatch strategy was selected
to deposit microstructure with elongated grains and zig-
zag low-angle grain boundaries (Fig. 5). Zig-zagging in-
terfaces between crystallites and microstructural features
are very common in natural materials and serve to en-
hance fracture toughness while preserving strength, for
instance in Saxidomus purpurata shells with zig-zag ori-
ented aragonite platelets. The deposition of the alter-
nating alloys at meso- (Fig. 3) and micro-scales (Fig. 6a)
with different mechanical properties and oscillating stress
gradients [51] at the cross-section (Fig. 3) resembles a
13
nacre-like microstructure [55] and has been known to
improve the mechanical properties of hybrid-materials.
Even though the mismatch in mechanical properties of
IN625 and 316L is not sufficiently high, this work shows
the possibility to tailor multilayer materials using the
liquid dispersed metal powder bed fusion approach. Ad-
ditionally, the incorporation of ceramic nano-dispersoids
(Figs. 8,9) based on Cr oxides can be used as an effec-
tive tool to tune the mechanical properties of the AM
structure by nanoscale design.
In summary, the liquid dispersed metal powder bed
fusion approach offers the possibility to produce hierar-
chical microstructures based on meso- and micro-scale
multilayers, grain boundaries of zig-zag shape, gradually
changing textures, oscillatory stress concentrations and
nanoscopic reinforcements, which together comprise a va-
riety of multi-scale interfaces and can be used to tailor
mechanical properties of AM structures.
A. Conclusion
This study demonstrates the feasibility of synthesiz-
ing a multi-metal hybrid structure based on IN625 and
316L, featuring unique multilayered microstructures at
meso- and micro-scales, zig-zag shaped elongated grains
and grain boundaries, regularly distributed nanoscopic
dispersoids, complex texture evolution with gradual tran-
sitions, oscillating cross-sectional residual stresses and
abrupt variation of mechanical properties across the mul-
tilayer. The zig-zag cracking correlated with the occur-
rence of IN625 (Fig. 2) as well as the pronounced C-
shaped residual stress gradient (Fig. 3c), however, indi-
cate the importance of process optimization in the fab-
rication of multi-metal hybrid materials with crack-free
microstructures.
Although the synthesis of the multilayered structure
(Fig. 3) was inspired by the architectural principles ob-
served in biological materials [56] and the structure pos-
sesses a variety of interesting bionic microstructural fea-
tures, the hardness values of IN625 and 316L do not differ
significantly. Thus, effects like crack deflection toughen-
ing (observed in nacre) at the interfaces between IN625
and 316L layers and regions could not be achieved here.
The fabrication of hybrid systems consisting of materials
with significantly different intrinsic mechanical proper-
ties like metal-ceramic composites has future potential to
design multi-material structures with superior mechani-
cal properties like high strength and toughness.
Finally, the presence of unique microstructures at
meso-, micro- and nano-scales observed within the IN625-
316L multilayer (Fig. 2) indicates the necessity of us-
ing cutting edge analytical techniques operating at mul-
tiple length scale in order to reveal the very partic-
ular process-microstructure-stress-property relationships
in AM structures.
Acknowledgment
A part of this work was supported by ¨
Osterreichische
Forschungsf¨orderungsgesellschaft mbH (FFG), Project
No. 861496, “CrossSurfaceMech”. Financial support by
the Austrian Federal Government (in particular from
Bundesministerium f¨ur Verkehr, Innovation und Tech-
nologie and Bundesministerium f¨ur Wissenschaft, For-
schung und Wirtschaft) represented by ¨
Osterreichische
Forschungsf¨orderungsgesellschaft mbH and the Styrian
and the Tyrolean Provincial Government, represented
by Steirische Wirtschaftsf¨orderungsgesellschaft mbH and
Standortagentur Tirol, within the framework of the
COMET Funding Programme is gratefully acknowl-
edged. Part of the research leading to this result has been
supported by the project CALIPSOplus under the Grant
Agreement 730872 from the EU Framework Programme
for Research and Innovation HORIZON 2020. Another
part of this work was carried out with the support of
CEITEC Nano Research Infrastructure (ID LM2015041,
MEYS CR, 2016–2019), CEITEC Brno University of
Technology.
[1] K. Martinsen, S. J. Hu, B. E. Carlson, Joining of dis-
similar materials, CIRP Ann. – Manuf. Technol. 64 (2)
(2015) 679–699.
[2] P. Fratzl, R. Weinkamer, Nature’s hierarchical materials
(nov 2007).
[3] A. Bandyopadhyay, B. Heer, Additive manufacturing of
multi-material structures (jul 2018).
[4] W. J. Clegg, K. Kendall, N. M. N. Alford, T. W. Button,
J. D. Birchall, A simple way to make tough ceramics,
Nature 347 (6292) (1990) 455–457.
[5] J. W. Dunlop, P. Fratzl, Biological composites, Annu.
Rev. Mater. Res. 40 (1) (2010) 1–24.
[6] R. Hahn, M. Bartosik, R. Soler, C. Kirchlechner,
G. Dehm, P. H. Mayrhofer, Superlattice effect for en-
hanced fracture toughness of hard coatings, Scr. Mater.
124 (2016) 67–70.
[7] Y. M. Wang, T. Voisin, J. T. McKeown, J. Ye, N. P.
Calta, Z. Li, Z. Zeng, Y. Zhang, W. Chen, T. T.
Roehling, R. T. Ott, M. K. Santala, P. J. Depond, M. J.
Matthews, A. V. Hamza, T. Zhu, Additively manufac-
tured hierarchical stainless steels with high strength and
ductility, Nat. Mater. 17 (1) (2018) 63–70.
[8] A. Mortensen, S. Suresh, Functionally graded metals and
metal-ceramic composites: Part 1 processing, Int. Mater.
Rev. 40 (6) (1995) 239–265.
[9] D. C. Hofmann, J. Kolodziejska, S. Roberts, R. Otis,
R. P. Dillon, J. O. Suh, Z. K. Liu, J. P. Borgonia, Com-
positionally graded metals: A new frontier of additive
manufacturing, J. Mater. Res. 29 (17) (2014) 1899–1910.
[10] Y. Liu, F. Weng, G. Bi, Y. Chew, S. Liu, G. Ma, S. K.
Moon, Characterization of wear properties of the func-
tionally graded material deposited on cast iron by laser-
14
aided additive manufacturing, Int. J. Adv. Manuf. Tech-
nol. 105 (10) (2019) 4097–4105.
[11] Z. Sun, R. Karppi, The application of electron beam
welding for the joining of dissimilar metals: An overview,
J. Mater. Process. Technol. 59 (3 SPEC. ISS.) (1996)
257–267.
[12] D. Herzog, V. Seyda, E. Wycisk, C. Emmelmann, Ad-
ditive manufacturing of metals, Acta Mater. 117 (2016)
371–392.
[13] C. K¨orner, Additive manufacturing of metallic compo-
nents by selective electron beam melting – a review, Int.
Mater. Rev. 61 (5) (2016) 361–377.
[14] D. D. Gu, W. Meiners, K. Wissenbach, R. Poprawe, Laser
additive manufacturing of metallic components: Materi-
als, processes and mechanisms, Int. Mater. Rev. 57 (3)
(2012) 133–164.
[15] W. E. Frazier, Metal additive manufacturing: A review
(apr 2014).
[16] D. Ding, Z. Pan, D. Cuiuri, H. Li, Wire-feed additive
manufacturing of metal components: technologies, de-
velopments and future interests (oct 2015).
[17] M. Yakout, M. A. Elbestawi, S. C. Veldhuis, A study of
thermal expansion coefficients and microstructure during
selective laser melting of Invar 36 and stainless steel 316L,
Addit. Manuf. 24 (2018) 405–418.
[18] T. DebRoy, H. L. Wei, J. S. Zuback, T. Mukherjee, J. W.
Elmer, J. O. Milewski, A. M. Beese, A. Wilson-Heid,
A. De, W. Zhang, Additive manufacturing of metallic
components – Process, structure and properties (mar
2018).
[19] J. J. Lewandowski, M. Seifi, Metal Additive Manufac-
turing: A Review of Mechanical Properties, Annu. Rev.
Mater. Res. 46 (1) (2016) 151–186.
[20] J. J. Sobczak, L. Drenchev, Metallic functionally graded
materials: A specific class of advanced composites, J.
Mater. Sci. Technol. 29 (4) (2013) 297–316.
[21] C. Jang, J. Lee, J. Sung Kim, T. Eun Jin, Mechanical
property variation within Inconel 82/182 dissimilar metal
weld between low alloy steel and 316 stainless steel, Int.
J. Press. Vessel. Pip. 85 (9) (2008) 635–646.
[22] T. Mukherjee, W. Zhang, T. DebRoy, An improved pre-
diction of residual stresses and distortion in additive
manufacturing, Comput. Mater. Sci. 126 (2017) 360–372.
[23] R. Banerjee, P. C. Collins, D. Bhattacharyya, S. Baner-
jee, H. L. Fraser, Microstructural evolution in laser de-
posited compositionally graded α/βtitanium-vanadium
alloys, Acta Mater. 51 (11) (2003) 3277–3292.
[24] P. C. Collins, R. Banerjee, S. Banerjee, H. L. Fraser,
Laser deposition of compositionally graded titanium-
vanadium and titanium-molybdenum alloys, Mater. Sci.
Eng. A 352 (1-2) (2003) 118–128.
[25] D. C. Hofmann, S. Roberts, R. Otis, J. Kolodziejska,
R. P. Dillon, J. O. Suh, A. A. Shapiro, Z. K. Liu, J. P.
Borgonia, Developing gradient metal alloys through ra-
dial deposition additive manufacturing, Sci. Rep. 4 (1)
(2015) 1–8.
[26] H. Sahasrabudhe, R. Harrison, C. Carpenter, A. Bandy-
opadhyay, Stainless steel to titanium bimetallic structure
using LENS™, Addit. Manuf. 5 (2015) 1–8.
[27] S. Q. Wang, J. H. Liu, D. L. Chen, Tensile and fatigue
properties of electron beam welded dissimilar joints be-
tween Ti-6Al-4V and BT9 titanium alloys, Mater. Sci.
Eng. A 584 (2013) 47–56.
[28] B. Onuike, B. Heer, A. Bandyopadhyay, Additive manu-
facturing of Inconel 718—Copper alloy bimetallic struc-
ture using laser engineered net shaping (LENS™), Addit.
Manuf. 21 (2018) 133–140.
[29] V. K. Balla, P. D. DeVasConCellos, W. Xue, S. Bose,
A. Bandyopadhyay, Fabrication of compositionally and
structurally graded Ti-TiO2 structures using laser engi-
neered net shaping (LENS), Acta Biomater. 5 (5) (2009)
1831–1837.
[30] V. K. Balla, P. P. Bandyopadhyay, S. Bose, A. Bandy-
opadhyay, Compositionally graded yttria-stabilized zir-
conia coating on stainless steel using laser engineered net
shaping (LENS™), Scr. Mater. 57 (9) (2007) 861–864.
[31] G. A. Knorovsky, M. J. Cieslak, T. J. Headley, A. D.
Romig, W. F. Hammetter, INCONEL 718: A solidifica-
tion diagram, Metall. Trans. A 20 (10) (1989) 2149–2158.
[32] N. Nadammal, S. Cabeza, T. Mishurova, T. Thiede,
A. Kromm, C. Seyfert, L. Farahbod, C. Haberland, J. A.
Schneider, P. D. Portella, G. Bruno, Effect of hatch
length on the development of microstructure, texture and
residual stresses in selective laser melted superalloy In-
conel 718, Mater. Des. 134 (2017) 139–150.
[33] S. J. Wolff, Z. Gan, S. Lin, J. L. Bennett, W. Yan, G. Hy-
att, K. F. Ehmann, G. J. Wagner, W. K. Liu, J. Cao, Ex-
perimentally validated predictions of thermal history and
microhardness in laser-deposited Inconel 718 on carbon
steel, Addit. Manuf. 27 (2019) 540–551.
[34] K. Shah, I. ul Haq, A. Khan, S. A. Shah, M. Khan, A. J.
Pinkerton, Parametric study of development of Inconel-
steel functionally graded materials by laser direct metal
deposition, Mater. Des. 54 (2014) 531–538.
[35] B. E. Carroll, R. A. Otis, J. P. Borgonia, J. O. Suh,
R. P. Dillon, A. A. Shapiro, D. C. Hofmann, Z. K. Liu,
A. M. Beese, Functionally graded material of 304L stain-
less steel and inconel 625 fabricated by directed energy
deposition: Characterization and thermodynamic mod-
eling, Acta Mater. 108 (2016) 46–54.
[36] A. Hinojos, J. Mireles, A. Reichardt, P. Frigola, P. Hose-
mann, L. E. Murr, R. B. Wicker, Joining of Inconel 718
and 316 Stainless Steel using electron beam melting ad-
ditive manufacturing technology, Mater. Des. 94 (2016)
17–27.
[37] W. Oliver, G. M. Pharr, An improved technique for de-
termining hardness and elastic modulus using load and
displacement sensing indentation experiments, J. Mater.
Res. 7 (6) (1992) 1564–1583.
[38] J. Kieffer, D. Karkoulis, PyFAI, a versatile library for
azimuthal regrouping, J. Phys. Conf. Ser. 425 (20) (2013)
202012.
[39] S. Timoshenko, Analysis of Bi-Metal Thermostats, J.
Opt. Soc. Am. 11 (3) (1925) 233.
[40] S. K. Rai, A. Kumar, V. Shankar, T. Jayakumar, K. B. S.
Rao, B. Raj, Characterization of microstructures in In-
conel 625 using X-ray diffraction peak broadening and
lattice parameter measurements, Scr. Mater. 51 (1)
(2004) 59–63.
[41] C. Yan, L. Hao, A. Hussein, P. Young, D. Raymont,
Advanced lightweight 316L stainless steel cellular lattice
structures fabricated via selective laser melting, Mater.
Des. 55 (2014) 533–541.
[42] C. Li, Z. Y. Liu, X. Y. Fang, Y. B. Guo, Residual Stress
in Metal Additive Manufacturing, in: Procedia CIRP,
Vol. 71, Elsevier B.V., 2018, pp. 348–353.
15
[43] D. Seo, M. Sayar, K. Ogawa, SiO2and MoSi2formation
on Inconel 625 surface via SiC coating deposited by cold
spray, Surf. Coatings Technol. 206 (11–12) (2012) 2851–
2858.
[44] M. Yakout, M. A. Elbestawi, S. C. Veldhuis, A study of
thermal expansion coefficients and microstructure during
selective laser melting of Invar 36 and stainless steel 316L,
Addit. Manuf. 24 (2018) 405–418.
[45] P. Van Houtte, L. De Buyser, The influence of crystallo-
graphic texture on diffraction measurements of residual
stress, Acta Metall. Mater. 41 (2) (1993) 323–336.
[46] Z. Wang, E. Denlinger, P. Michaleris, A. D. Stoica,
D. Ma, A. M. Beese, Residual stress mapping in Inconel
625 fabricated through additive manufacturing: Method
for neutron diffraction measurements to validate ther-
momechanical model predictions, Mater. Des. 113 (2017)
169–177.
[47] P. Mercelis, J. P. Kruth, Residual stresses in selective
laser sintering and selective laser melting, Rapid Proto-
typ. J. 12 (5) (2006) 254–265.
[48] T. Dai, J. Lippold, Characterization of the Interface of
an Alloy 625 Overlay on Steels Using Nanoindentation,
J. Mater. Eng. Perform. 27 (7) (2018) 3411–3418.
[49] C. Tromas, J. C. Stinville, C. Templier, P. Villechaise,
Hardness and elastic modulus gradients in plasma-
nitrided 316L polycrystalline stainless steel investigated
by nanoindentation tomography, Acta Mater. 60 (5)
(2012) 1965–1973.
[50] T. Jonsson, S. Karlsson, H. Hooshyar, M. Sattari,
J. Liske, J. E. Svensson, L. G. Johansson, Oxidation Af-
ter Breakdown of the Chromium-Rich Scale on Stainless
Steels at High Temperature: Internal Oxidation, Oxid.
Met. 85 (5–6) (2016) 509–536.
[51] O. Kolednik, J. Predan, F. D. Fischer, P. Fratzl, Bioin-
spired Design Criteria for Damage-Resistant Materials
with Periodically Varying Microstructure, Adv. Funct.
Mater. 21 (19) (2011) 3634–3641.
[52] P. Fratzl, H. Gupta, F. Fischer, O. Kolednik, Hindered
Crack Propagation in Materials with Periodically Vary-
ing Young’s Modulus – Lessons from Biological Materials,
Adv. Mater. 19 (18) (2007) 2657–2661.
[53] J. DuPont, J. Lippold, S. Kiser, Welding Metallurgy and
Weldability of Nickel-Base Alloys, John Wiley & Sons,
2009.
[54] R. A. Patterson, J. O. Milewski, GTA weld cracking-alloy
625 to 304L, Weld. J. (Miami, Fla) 64 (1985) 227–231.
[55] F. Barthelat, C. M. Li, C. Comi, H. D. Espinosa, Me-
chanical properties of nacre constituents and their im-
pact on mechanical performance, J. Mater. Res. 21 (8)
(2006) 1977–1986.
[56] U. G. Wegst, H. Bai, E. Saiz, A. P. Tomsia, R. O.
Ritchie, Bioinspired structural materials, Nat. Mater.
14 (1) (2015) 23–36.
