University of Wollongong

Research Online
Faculty of Engineering and Information Sciences -
Faculty of Engineering and Information Sciences
Papers: Part A

2016

Fabricating superior NiAl bronze components

through wire arc additive manufacturing
Donghong Ding
University of Wollongong, donghong@uow.edu.au

Zengxi Stephen Pan

University of Wollongong, zengxi@uow.edu.au

Stephen van Duin

University of Wollongong, svanduin@uow.edu.au

Hui Jun Li
University of Wollongong, huijun@uow.edu.au

Chen Shen
University of Wollongong, cs395@uowmail.edu.au

Publication Details
Ding, D., Pan, Z., van Duin, S., Li, H. & Shen, C. (2016). Fabricating superior NiAl bronze components through wire arc additive
manufacturing. Materials, 9 (8), 652-1-652-12.

Research Online is the open access institutional repository for the University of Wollongong. For further information contact the UOW Library:
research-pubs@uow.edu.au
Fabricating superior NiAl bronze components through wire arc additive
manufacturing
Abstract
Cast nickel aluminum bronze (NAB) alloy is widely used for large engineering components in marine
applications due to its excellent mechanical properties and corrosion resistance. Casting porosity, as well as
coarse microstructure, however, are accompanied by a decrease in mechanical properties of cast NAB
components. Although heat treatment, friction stir processing, and fusion welding were implemented to
eliminate porosity, improve mechanical properties, and refine the microstructure of as-cast metal, their
applications are limited to either surface modification or component repair. Instead of traditional casting
techniques, this study focuses on developing NAB components using recently expanded wire arc additive
manufacturing (WAAM). Consumable welding wire is melted and deposited layer-by-layer on substrates
producing near-net shaped NAB components. Additively-manufactured NAB components without post-
processing are fully dense, and exhibit fine microstructure, as well as comparable mechanical properties, to as-
cast NAB alloy. The effects of heat input from the welding process and post-weld-heat-treatment (PWHT) are
shown to give uniform NAB alloys with superior mechanical properties revealing potential marine
applications of the WAAM technique in NAB production.

Keywords
bronze, nial, superior, fabricating, components, wire, manufacturing, arc, additive

Disciplines
Engineering | Science and Technology Studies

Publication Details
Ding, D., Pan, Z., van Duin, S., Li, H. & Shen, C. (2016). Fabricating superior NiAl bronze components
through wire arc additive manufacturing. Materials, 9 (8), 652-1-652-12.

This journal article is available at Research Online: http://ro.uow.edu.au/eispapers/6079

 materials
Article
Fabricating Superior NiAl Bronze Components
through Wire Arc Additive Manufacturing
Donghong Ding, Zengxi Pan *, Stephen van Duin, Huijun Li and Chen Shen
School of Mechanical, Materials, and Mechatronics Engineering, Faculty of Engineering and Information Science,
University of Wollongong, Northfield Ave, Keiraville, NSW 2500, Australia; donghong@uow.edu.au (D.D.);
svanduin@uow.edu.au (S.v.D.); huijun@uow.edu.au (H.L.); cs395@uowmail.edu.au (C.S.)
* Correspondence: zengxi@uow.edu.au; Tel.: +61-2-42215498

Academic Editor: Guillermo Requena

Received: 9 July 2016; Accepted: 1 August 2016; Published: 3 August 2016

Abstract: Cast nickel aluminum bronze (NAB) alloy is widely used for large engineering components
in marine applications due to its excellent mechanical properties and corrosion resistance. Casting
porosity, as well as coarse microstructure, however, are accompanied by a decrease in mechanical
properties of cast NAB components. Although heat treatment, friction stir processing, and fusion
welding were implemented to eliminate porosity, improve mechanical properties, and refine the
microstructure of as-cast metal, their applications are limited to either surface modification or
component repair. Instead of traditional casting techniques, this study focuses on developing
NAB components using recently expanded wire arc additive manufacturing (WAAM). Consumable
welding wire is melted and deposited layer-by-layer on substrates producing near-net shaped NAB
components. Additively-manufactured NAB components without post-processing are fully dense,
and exhibit fine microstructure, as well as comparable mechanical properties, to as-cast NAB alloy.
The effects of heat input from the welding process and post-weld-heat-treatment (PWHT) are shown to
give uniform NAB alloys with superior mechanical properties revealing potential marine applications
of the WAAM technique in NAB production.

Keywords: NiAl bronze; additive manufacturing; arc welding; mechanical properties; microstructures

1. Introduction
Cast nickel aluminum bronze (NAB) alloys are widely used for marine applications due to their
excellent mechanical properties and corrosion resistance. It also has good abrasion resistance, which
makes NAB alloy an ideal material for landing gear bushings and bearings. As summarized by
Brezina [1], NAB is a copper-based quaternary alloy that typically contains 8–12 wt % aluminum,
3–6 wt % nickel and iron. It has a two-phase microstructure, an equilibrium fcc α phase and a
high-temperature bcc β phase. NAB has enhanced mechanical properties due to the addition of nickel
and iron to copper-aluminum alloys, which increase its mechanical properties through the precipitation
of several intermetallics κ among the α and the β phases. In addition, the low content of manganese
(max 3.5 wt %) in commercial NAB alloys improves its casting ability and acts as a microstructure
stabilizer. Apart from mechanical properties, NAB alloys are characterized by their excellent corrosion
behavior, partially because Ni and Fe extend the terminal α phase field and suppress the γ2 phase
formation that occurs in binary Cu-Al alloys. Culpan and Rose [2] claimed that the absence of the
γ2 phase is preferable as it is a hard and brittle phase containing high Al content that impairs the
material’s corrosion resistance. Additionally, Yu et al. [3] found the formation of a tough aluminum
oxide film (Al2 O3 ) gives NAB its excellent corrosion and erosion cavitation resistance.
In marine applications however, Ferrara and Caton [4] summarized that cast components usually
have thick sections, and the resulting slow cooling rates (typically 10´3 to 10´2 ˝ C/s) contribute to

Materials 2016, 9, 652; doi:10.3390/ma9080652 www.mdpi.com/journal/materials

Materials 2016, 9, 652 2 of 12

coarse microstructures and reduced mechanical properties. This is coupled with the crucial problem of
casting porosity which further reduces mechanical properties and service performance of cast NAB [5].
Several methods for improving the microstructure and mechanical properties of NAB e.g.,
heat treatment, friction stir processing (FSP), and fusion welding, are reported in the literature.
The microstructures and corresponding hardness of NAB material are strongly dependent on the
applied heat treatment. Wu et al. [6] investigated that different peak temperatures and cooling rates
form various phases (i.e., Widmanstätten α, bainitic α, or even martensitic β) resulting in different
mechanical properties, as well as corrosion behavior. However, in marine application, some large
cast components with varied thicknesses may not be heat-treatable. FSP was extensively performed
to selectively modify the near-surface layers of cast NAB components. Fuller et al. [5] achieved
microstructure refinement and homogenization, as well as closure of porosity of as-cast NAB materials
through FSP. Nevertheless, this modification method is only available for near-surface layers rather
than for through-thickness components due to the shallow pin depth of the FSP tools [7]. Fusion
welding was often implemented to repair damaged NAB components. Li et al. [8] investigated the
effect of welding parameters, such as current, voltage, wire-feed rate and travel speed, on the formation
of the weld bead. It is found that due to the rapid cooling rate (around 1 ˝ C/s), α with a Widmanstätten
morphology, as well as bainitic and martensitic products of β phase may become apparent, which
exhibit both higher strength and ductility of the material than the slowly cooled cast metal.
Recently, wire arc additive manufacturing (WAAM) techniques, using either the gas metal
arc welding (GMAW) or gas tungsten arc welding (GTAW) processes, have been developed for
layer-by-layer manufacturing of functional metal components with full density [9–11]. Ding et al. [12]
reviewed the technologies and developments of wire-feed additive manufacturing of metal components
and revealed that WAAM becomes a promising alternative for fabricating components with medium to
large size due to its productivity, cost-competitiveness, energy efficiency, and unlimited build envelop.
For aerospace applications, wall structures of titanium alloys have been built using WAAM, and
their mechanical properties were demonstrated to be comparable to cast or wrought materials [13].
In comparison to casting technology, firstly the additive manufacturing (AM) technique offers a great
potential for time and cost savings, especially for newly-designed components. Secondly, it is able to
fabricate complex shapes owing to its paradigm shift through layer-by-layer manufacturing. Lastly,
as demonstrated by Shen et al. [14], AM techniques make it possible to build intermetallic materials
and functional graded materials, which poses challenges using traditional casting technology.
At present, there has not been any work reported for manufacturing of NAB alloys using the
WAAM process. Investigation of the WAAM of NAB may provide an alternative production method to
significantly reduce manufacturing lead times as well as overcoming common defects trapped inside
cast components. This study focuses on the WAAM of NAB components. Firstly, thin wall samples are
fabricated to investigate the feasibility of the WAAM of NAB. Then, microstructure and mechanical
properties of additively manufactured NAB wall structures are compared to as-cast NAB. Finally,
the effects of heat inputs and post-weld-heat-treatment (PWHT), on the properties of weld metal (WM),
are analyzed and discussed. Other properties, like erosion-corrosion, wear and high temperature
behavior of the NAB alloy, are beyond the scope of this paper. However, corrosion behavior was
intensively investigated and can be found elsewhere [15,16].

2. Experimental Procedure
An experimental robotic WAAM system was developed as shown in Figure 1. A computer
interface (1) is used to program the experimental processes. The robot controller (2) is used to
coordinate both the welding power source (3) and the robot motions (4). Either a GMAW (5) or GTAW
torch (6) could be used to deliver the wire arc depending on experimental requirements. An example
of an AM work-piece is shown in (7).
Materials 2016, 9, 652 3 of 12
Materials
Materials2016,
2016,9,9,652
652 3 3ofof1212

Figure
Figure1.1.
Figure 1.Schematic
Schematicdiagram
Schematic diagramof
diagram ofthe
of thedeveloped
the developedexperimental
developed experimentalWAAM
experimental WAAMsystem.
WAAM system.
system.

InInthis
In thisstudy,
this study,GMAW
study, GMAWprocess
GMAW processis
process isisperformed
performeddue
performed dueto
due totoits
itshigh
its highdeposition
high depositionrate
deposition rate(up
rate (upto
(up toto5 kg/h
kg/hin
55kg/h inthe
in the
the
present
present study).
study). AAsynergic
synergic pulse
pulse spray
spray transfer
transfer mode
mode and
and argon
argon shielding
shielding gas
gas
present study). A synergic pulse spray transfer mode and argon shielding gas were used for welding.
were
were used
used for
for welding.
welding.
Figure
Figure2a
Figure
2a shows
2ashows
shows theschematic
thethe
schematic
schematic diagram
diagram
diagram
ofoflayer-by-layer
layer-by-layer
of layer-by-layer
additive
additiveadditive manufacturing
manufacturing
manufacturing
process
process process
using
using GMAW. using
GMAW.
GMAW. A molten
A molten metal pool is sustained when an electric arc forms between a consumable wire
A molten metal poolmetal pool iswhen
is sustained sustained whenarc
an electric anforms
electric arc forms
between between awire
a consumable consumable
electrode wire
and
electrode
electrode and
and work-piece
work-piece metal
metal (base
(base metal
metal ororprevious-deposited
previous-deposited weld
weld metal).
metal). Welding
Welding wire
wire isisfed
fed
work-piece metal (base metal or previous-deposited weld metal). Welding wire is fed into the molten
into
intothe
the molten
molten pool
pool under
under the
theprotection
protection ofofthe
theshielding
shielding gas.
gas.When
When moving
moving the
the welding
welding torch
torchwith
with
pool under the protection of the shielding gas. When moving the welding torch with controlled travel
controlled
controlled travel speeds along
alongpredefined orientations and paths, geometrical metal objects
objectsare
speeds alongtravel speedsorientations
predefined predefined
and paths, orientations
geometricaland paths,
metal geometrical
objects metal
are fabricated layer-by-layer. are
fabricated
fabricated layer-by-layer.
layer-by-layer. Figure
Figure 2b
2bshows
shows the
the setup
setup ofofthe
the robotic
robotic GMAW
GMAW process.
process.
Figure 2b shows the setup of the robotic GMAW process.

Figure
Figure2.2.
Figure 2.(a)
(a)Schematic
(a) Schematicdiagram
Schematic diagramof
diagram oflayer-by-layer
of layer-by-layerdeposition
layer-by-layer depositionusing
deposition usingGMAW
using GMAWprocess;
GMAW process;and
process; and(b)
and (b)the
(b) thesetup
the setup
setup
of
ofthe robotic
the robotic GMAW
GMAW process.
process.
of the robotic GMAW process.

To
Toillustrate
illustratethethefabrication
fabricationofofNAB
NABalloy,
alloy,Figure
Figure333shows
showsthree
threetesting
testingwall
wallstructures
structuresdeposited
deposited
To illustrate the fabrication of NAB alloy, Figure shows three testing wall structures deposited
onon as-cast
as-cast base
base metals
metals using
using WAAM.
WAAM. The
Thecomposition
composition of
ofNAB
NAB base
basemetal
metal (BM)
(BM) and
and welding
welding wire
wireare
are
on as-cast base metals using WAAM. The composition of NAB base metal (BM) and welding wire are
listed
listedininTable
Table 1,1,showing
showing the
the similar
similar composition
composition ofofmaterials,
materials, specifically
specificallythe
thecontent
content ofofaluminum,
aluminum,
listed in Table 1, showing the similar composition of materials, specifically the content of aluminum,
nickel,
nickel,andandiron.
iron.The
Theas-cast
as-castBM
BMwaswasmade
madeby byVeem
VeemLtd.,
Ltd.,Perth,
Perth,Australia
Australiaaccording
accordingto tothe
theAustralia
Australia
nickel, and iron. The as-cast BM was made by Veem Ltd., Perth, Australia according to the Australia
Defense
Defense Standard
Standard NES
NES 747
747specifications.
specifications.Commercial
Commercial 1.2 mm
1.2 mm ALUNOX
ALUNOX AX-CuAl8Ni6
AX-CuAl8Ni6 welding
welding wire
wire
Defense Standard NES 747 specifications. Commercial 1.2 mm ALUNOX AX-CuAl8Ni6 welding
was used. The welding parameters used are listed in Table 2, in which three different
was used. The welding parameters used are listed in Table 2, in which three different wire-feed rates wire-feed rates
were
wereapplied
appliedtototest testthe
theeffects
effectsofofheat
heatinputs
inputson onmicrostructures
microstructuresand andmechanical
mechanicalproperties
propertiesofof
Materials 2016, 9, 652 4 of 12

wire was used. The welding parameters used are listed in Table 2, in which three different wire-feed
rates were
Materials 2016,applied
9, 652 to test the effects of heat inputs on microstructures and mechanical properties 4 of of
12
additively manufactured NAB components. No BM preheating was conducted prior to commencing
additively
the depositionmanufactured NAB components.
process; therefore, the cooling No
rateBM preheating
of the was conducted
first deposited layer wasprior to commencing
maximum. With the
the deposition
increasing heightprocess;
of thetherefore,
depositedthe cooling
wall, rate of the
the cooling ratefirst deposited
decreases. layer
The was maximum.
deposited With the
wall components
increasing
(Figure height of have
3) typically the deposited wall,
a length of 100the
mmcooling
and arate decreases.
height The With
of 40 mm. deposited wall components
the increasing of heat
(Figure
input 3) typically
from 653 J/mm have a length
to 1114 J/mmof 100 mm and
in Table 2, thea height of 40 of
thicknesses mm.theWith
wallsthe increasing
increased fromof~10
heatmm
inputto
frommm.
~14 653 J/mm to 1114 J/mm in Table 2, the thicknesses of the walls increased from ~10 mm to ~14 mm.

Figure 3.
Figure 3. Three
Three additively-manufactured
additively-manufactured NABNAB components.
components. The deposited wall
The deposited wall components
components were
were
wire-cut from the middle in the lengthwise direction with one half of the walls further processed with
wire-cut from the middle in the lengthwise direction with one half of the walls further processed with
post-weld-heat-treatment.
post-weld-heat-treatment.

Table 1. Composition of NAB-based metal and welding wire (wt %).

Table 1. Composition of NAB-based metal and welding wire (wt %).
Element Cu Al Ni Fe Mn Si Zn Sn Pb Mg Cr
Element
Base Metal Cu
Rem Al
8.8 Ni
5.2 Fe
4.4 Mn
1.1 Si
0.07 Zn
0.02 Sn
0.02 Pb
<0.01 Mg
<0.01 Cr
<0.01
Base Metal
Welding wire Rem
Rem 8.8
8.5–9.5 5.2
4–5.5 4.4
3–5 1.1
0.6–3.5 0.07
0.1 0.02
0.1 0.02
0.1 <0.01
0.02 <0.01 <0.01
Welding Wire Rem 8.5–9.5 4–5.5 3–5 0.6–3.5 0.1 0.1 0.1 0.02

Table 2. Welding parameters of three tests.

Table 2. Welding parameters of three tests.
Wire Feed Rate Travel Speed Average Average Heat Input
Test No.
(m/min)
Wire Feed (mm/min)
Travel Speed Current
Average (A) Voltage
Average (V) (J/mm)
Heat Input
1Test No. 5.4
Rate (m/min) 400
(mm/min) 175.5
Current (A) Voltage (V)24.8 (J/mm)653
2 6.7 400 218.3 26.7 874
1 5.4 400 175.5 24.8 653
3 8 400 256.1 29.0 1114
2 6.7 400 218.3 26.7 874
3 8 400 256.1 29.0 1114
Figure 4 shows the sample preparation procedures. Each wall was wire-cut into two sub-walls
with one half post-weld-heat-treated for comparison, as shown in Figure 4a. Annealing treatment of
Figure 4 shows the sample preparation procedures. Each wall was wire-cut into two sub-walls
675 °C for six hours, followed by air cooling, was used to release the induced residual stresses from
with one half post-weld-heat-treated for comparison, as shown in Figure 4a. Annealing treatment of
the welding process. For each sub-wall, the bulk tensile sample was cut (see Figure 4b), and then
675 ˝ C for six hours, followed by air cooling, was used to release the induced residual stresses from
more than 10 specimens were sliced and numbered from the bottom to the top covering the region of
the welding process. For each sub-wall, the bulk tensile sample was cut (see Figure 4b), and then more
BM and WM, as shown in Figure 4c. A total of around 80 miniature specimens cut from six sub-walls
than 10 specimens were sliced and numbered from the bottom to the top covering the region of BM
were obtained and performed tensile testing. The detailed dimensions of the specimens used are
and WM, as shown in Figure 4c. A total of around 80 miniature specimens cut from six sub-walls were
shown in Figure 4d. Cross-sections of each sub-wall were cut out with a 2 mm wire saw for
obtained and performed tensile testing. The detailed dimensions of the specimens used are shown
microstructure observations and hardness testing. Microhardness was tested for all cross-sectioned
samples along the vertical centerline and the heat-affected zones just under the fusion line, as
indicated in Figure 4e.
Materials 2016, 9, 652 5 of 12

in Figure 4d. Cross-sections of each sub-wall were cut out with a 2 mm wire saw for microstructure
observations and hardness testing. Microhardness was tested for all cross-sectioned samples along the
vertical centerline
Materials and the heat-affected zones just under the fusion line, as indicated in Figure
2016, 9, 652 5 of 124e.

Figure
Figure 4. Sample
4. Sample preparation:(a)
preparation: (a)wall
wall structure
structure and
andthe
themiddle
middlecutting trajectory;
cutting (b) bulk
trajectory; tensile
(b) bulk tensile
sample cut from the sub-wall; (c) tensile specimens after slicing bulk sample; (d) dimensions
sample cut from the sub-wall; (c) tensile specimens after slicing bulk sample; (d) dimensions of of
miniature tensile specimens in mm; and (e) wire-cut cross-section for microstructure observations and
miniature tensile specimens in mm; and (e) wire-cut cross-section for microstructure observations and
hardness testing.
hardness testing.
The metallographic specimens were prepared using standard procedures for NAB. A mixed
The metallographic
solution specimens
of 5 vol % ferric chloride, were
25 volprepared using acid,
% hydrochloric standard
and 70procedures for NAB.
vol % distilled water A
was mixed
solution of 5 as
selected volthe
% ferric
etchingchloride, 25 vol
agent. The % hydrochloric
microstructure of theacid, and 70were
specimens vol % distilledusing
examined waterscanning
was selected
electron
as the etching microscopy
agent. The (SEM, JEOL JSM-6490LA,
microstructure of the Tokyo, Japan).
specimens A GBC
were MMA using
examined X-ray diffractometer
scanning electron
(XRD) (GBC Scientific Equipment, Chicago, IL, USA) with CuKα radiation
microscopy (SEM, JEOL JSM-6490LA, Tokyo, Japan). A GBC MMA X-ray diffractometer (λ = 1.5418 Å) was used(GBC
(XRD)
to identify
Scientific the phaseChicago,
Equipment, constitutions. Vickers
IL, USA) microhardness
with profiles(λwere
CuKα radiation measured
= 1.5418 at a load
Å) was usedofto1 identify
kg
(HV1) by a DuraScan Hardness Tester. Tension tests were performed on a MTS370 (MTS Systems
the phase constitutions. Vickers microhardness profiles were measured at a load of 1 kg (HV1) by a
Corporation, Minneapolis, MN, USA) load unit at a strain rate of 1.8 × 10−3 s−1 at room temperature.
DuraScan Hardness Tester. Tension tests were performed on a MTS370 (MTS Systems Corporation,
Minneapolis,
3. ResultsMN, USA) load unit at a strain rate of 1.8 ˆ 10´3 s´1 at room temperature.
and Discussion
Fully-dense
3. Results NAB components are obtained, as the macro photography of the cross-section
and Discussion
shows in Figure 5a. The regions of BM, heat affected zone (HAZ), and WM are found from the bottom
Fully-dense
to NAB
the top of the components
cross-section. are region
In the obtained, as the
of WM, macro
a series of photography
layer bands can ofbe
the cross-section
readily observed.shows
in Figure
This 5a. The regions
phenomenon is of BM,documented
well heat affectedinzone (HAZ),ofand
a number WM are
studies on found
differentfromAM theprocesses.
bottom to the
top of the cross-section.
Microstructures In theregions
of different region are
of WM,
showna inseries
Figureof 5b–d
layerand bands
the can be are
details readily observed.
provided in the This
following is
phenomenon paragraphs.
well documented in a number of studies on different AM processes. Microstructures
Figure
of different 5b shows
regions the representative
are shown in Figure 5b–dmicrostructure
and the details of are
WM. The as-deposited
provided WM contains
in the following paragraphs.
predominantly Widmanstätten α and very fine martensite (dark areas). It was fairly fine and uniform.
Figure 5b shows the representative microstructure of WM. The as-deposited WM contains
The layer band of reheated weld metal (area between two dashed lines in the Figure 5b) is very
predominantly Widmanstätten α and very fine martensite (dark areas). It was fairly fine and uniform.
narrow, with the thickness of ~200 µm, and the distance between each area of reheated weld metal is
The layer band of reheated weld metal (area between two dashed lines in the Figure 5b) is very narrow,
~2 mm. The number of layer bands in additively manufactured NAB is found to be exactly equal to
with the
thenumber
thickness of ~200 µm,
of deposition andThis
passes. the means
distancethebetween eachthickness
average layer area of reheated weld metal
is approximately 2 mmisand ~2 mm.
The number
the heightof layer
of thebands
wallinisadditively manufactured
approximately 40 mm. NABThis isisfound to be exactly
consistent with theequal to the number
experimental
of deposition passes.
measurements. TheThis
finalmeans the average
layer thickness layer thickness
is approximately 3 mm is approximately
indicating that the2 mm and the
previous height
layer is of
partially
the wall re-melted by the
is approximately 40deposition
mm. Thisofisthe precedingwith
consistent layer.
theWithout the further
experimental re-melting process,
measurements. The final
layerthe thickness
thickness is of the last layer is3 larger
approximately than the average
mm indicating that thelayer thickness.
previous layerLayer bands are
is partially a kind ofby the
re-melted
HAZ, which results from the sub-melting-point reheating treatment that
deposition of the preceding layer. Without the further re-melting process, the thickness was detailed by Liu of
et al.
the last
[17].
layer is larger than the average layer thickness. Layer bands are a kind of HAZ, which results from the
sub-melting-point reheating treatment that was detailed by Liu et al. [17].
Materials 2016, 9, 652 6 of 12
Materials 2016, 9, 652 6 of 12

Figure
Figure 5. Microstructure
5. Microstructure of the
of the deposited
deposited NABNAB components:
components: (a) macro
(a) macro photography
photography of cross-section
of the the cross-
of section
the wallofstructure;
the wall structure; (b) representative
(b) representative microstructure
microstructure in the
in the weld weld(c)
metal; metal; (c) morphology
morphology of
of the HAZ;
the HAZ; (d) representative microstructure in the base metal; (e) formation of transformed products
(d) representative microstructure in the base metal; (e) formation of transformed products in NAB
in NAB alloys; (f) typical microstructure of the HAZ after PWHT; and (g) typical microstructure of
alloys; (f) typical microstructure of the HAZ after PWHT; and (g) typical microstructure of the weld
the weld metal after PWHT.
metal after PWHT.

Figure 5c shows the microstructure of the HAZ. The width of the HAZ is very narrow (<1 mm)
forFigure 5c shows
all welding the microstructure
parameters. The microstructureof the underwent
HAZ. The significant
width of changes
the HAZinisthe very
HAZ.narrow
A
(<1transformation
mm) for all welding parameters.
of eutectic structure toThe microstructure
martensite occurred, underwent
in particular significant
in the areachanges
close to thein the HAZ.
fusion
A transformation of eutecticinstructure
line. The peak temperature this area towasmartensite
above the occurred,
eutectic point,in particular
resulting in inthe
theformation
area close ofto
a βthe
fusion line.
phase. The
This peak
was temperature
retained with rapidin this areaand
cooling wasthe
above the eutectic
subsequent point,ofresulting
formation in the
a martensitic formation
structure.
of The changing
a β phase. temperature
This was retainedgradient, with distance
with rapid cooling toand thethefusion line, ledformation
subsequent to regions of of aamartensitic
partially
transformed
structure. The eutectic
changing structure. The martensitic
temperature gradient,structure is harder
with distance tothan the α phase,
the fusion line, but
led are also more
to regions of a
brittle and susceptible to corrosion and cavitation erosion, making it undesirable.
partially transformed eutectic structure. The martensitic structure is harder than the α phase, but are
also moreFigure 5d and
brittle shows a typical to
susceptible microstructure
corrosion andofcavitation
as-cast BM consisting
erosion, makingof αitphase and various κ
undesirable.
intermetallic phases.a During
Figure 5d shows typical slow cooling (~10
microstructure
−3 °C∙s ) BM
of as-cast
−1 to room temperature,
consisting of α phase the andNAB alloy κ
various
undergoes several microstructure changes
intermetallic phases. During slow cooling (~10 as shown
´ 3 in
˝ C¨s ) to room temperature, the NAB bcc
Figure
´ 1 5e. The microstructure is entirely alloy
β phase at high temperature. At the temperature
undergoes several microstructure changes as shown in Figure around 1030 °C,5e. theTheprimary fcc Cu-rich is
microstructure α phase
entirely
starts to form within the β phase. When the temperature falls until 930 °C, the dendritic (variously
bcc β phase at high temperature. At the temperature around 1030 ˝ C, the primary fcc Cu-rich α phase
called globular or rosette)-shaped κII phase begins to form and is normally distributed at the α/β
starts to form within the β phase. When the temperature falls until 930 ˝ C, the dendritic (variously
interface. The κII precipitates are based on the Fe3Al (DO3 structure) and typically around 5 to 10 µm.
called globular or rosette)-shaped κII phase begins to form and is normally distributed at the α/β
In NABs with a high iron content of greater than ~5% and/or a high Fe to Ni ratio, the large dendritic
interface. The κII precipitates are based on the Fe3 Al (DO3 structure) and typically around 5 to 10 µm.
shaped κI phase, typically 20 to 50 µm in diameter, would form and mostly locate in the centers of α
In grains
NABs as with a high iron
described content
by Hasan of greater
et al. than ~5%
[18]. A small amount and/orof κI aprecipitates
high Fe to wereNi ratio,
foundtheinlarge dendritic
the present
shaped κI metal as shown in Figure 5d since the weight of iron is close to 5%. This phase is also in of
cast base phase, typically 20 to 50 µm in diameter, would form and mostly locate in the centers a α
grains
formas ofdescribed
Fe3Al (DO3 bystructure),
Hasan et al. but[18].
much A coarser
small amount
than theofκIIκIphase.
precipitates
Aroundwere found in thebelow
the temperature present
cast base metal as shown in Figure 5d since the weight of iron is close to 5%. This
860 °C, the much finer κIV phase begins to form in the α phase since the solubility of iron is exceeded. phase is also in a
form
Theofcuboidal
Fe3 Al (DO3 structure),
or cruciform but κmuch
shaped coarser than
IV precipitates thebased
are also κII phase.
on theAround the temperature
Fe3Al (DO3 structure). Atbelowthe
860 ˝ C, the much finer κIV phase begins to form in the α phase since the solubility of iron is exceeded.
Materials 2016, 9, 652 7 of 12
Materials 2016, 9, 652 7 of 12

temperature
The cuboidal of or 800 °C, theshaped
cruciform remaining β phase is transformed
κIV precipitates are also based intoonthe
theintermetallic
Fe3 Al (DO3κstructure).
III phase through
At the
temperature 800 ˝ C,βthe
the eutectoidofreaction → remaining
α + κIII. Theβκphase
III phase, which is based
is transformed intoontheNiAl (B2 structure),
intermetallic typically
κIII phase has
through
a lamellar morphology and formed at the α/β boundary. Once the cast NAB
the eutectoid reaction β Ñ α + κIII . The κIII phase, which is based on NiAl (B2 structure), typically has alloy is slowly cooled to
aroom temperature,
lamellar morphology the and
alloyformed
entirely atcontains
the α/β the α phaseOnce
boundary. and the several
cast NAB κ intermetallic
alloy is slowly phases.
cooled to
roomPWHT relievesthe
temperature, thealloy
residual stresses
entirely contains andthehomogenizes
α phase and thethemicrostructure. The BM phases.
several κ intermetallic contained a
stable casting
PWHT structure;
relieves therefore,
the residual littleand
stresses to no change occurred
homogenizes with PWHT.
the microstructure. The TheBMHAZ, where
contained a
martensite
stable castingwas present,therefore,
structure; was refined little as shown
to no change in occurred
Figure 5f.with ThePWHT.
HAZ retained
The HAZ, large
where sections of α
martensite
phase,
was however,
present, closer astoshown
was refined the fusion
in Figure line5f. appeared partiallylarge
The HAZ retained decomposed.
sections of Additionally,
α phase, however, the
martensite
closer to thewas replaced
fusion with a fine
line appeared graindecomposed.
partially α phase andAdditionally,
κ precipitates. theWM has a structure
martensite with
was replaced
Widmanstätten
with a fine grainααand phasea small
and κamount
precipitates.of martensite,
WM has whilst the PWHT
a structure decomposed these
with Widmanstätten α andleaving
a smalla
fine homogenized
amount microstructure
of martensite, whilst the PWHT of α phase and κ precipitates
decomposed these leavingas shown in Figure 5g. The
a fine homogenized refinement
microstructure
that
of α occurred
phase andinκboth the HAZasand
precipitates shown the WM was favorable.
in Figure A fine microstructure
5g. The refinement that occurred provides
in bothadditional
the HAZ
strength
and the WMto the material
was favorable.and improved corrosion resistance.
A fine microstructure provides additional strength to the material and
In order
improved to identify
corrosion the phase structure, an XRD analysis was conducted in the cross-section of
resistance.
the components. A comparison
In order to identify the phase of structure,
the XRD data an XRDfromanalysis
as-cast was
BM,conducted
heat treated in BM, as-depositedof
the cross-section WM,
the
components. A comparison of the XRD data from as-cast BM, heat treated BM, as-deposited WM, anda
and heat treated WM is shown in Figure 6. All the diffraction patterns are found to contain
prominence
heat treated WMof α(Cu) phase
is shown in and a certain
Figure 6. All theamount of NiAl
diffraction or Fe3Al
patterns areprecipitates.
found to containIn comparison
a prominence to the
of
BM, the
α(Cu) volume
phase and afraction of precipitates
certain amount of NiAldecreases significantlyInincomparison
or Fe3 Al precipitates. the WM due to BM,
to the the suppressed
the volume
eutectoidofreaction
fraction β→α
precipitates + κIII by the
decreases rapid cooling
significantly rate.
in the WMWithdue thetoPWHT, there is little
the suppressed change
eutectoid in BM,
reaction
βhowever,
Ñ α + κIIImore
by the intermetallic
rapid cooling phases
rate. With are theprecipitated
PWHT, there in the WM,
is little which
change is consistent
in BM, however, morewith
microstructure observations.
intermetallic phases are precipitated in the WM, which is consistent with microstructure observations.

Figure 6.
Figure 6. Comparison
Comparison of
of phase
phase constitutions
constitutions of
of as-cast
as-cast BM,
BM, as-deposited
as-deposited WM,
WM, heat-treated
heat-treated BM,
BM, and
and
heat-treated WM.
heat-treated WM.

Microhardness testing
Microhardness testing was
was performed
performed to to understand
understand thethe effect
effect of
of fusion
fusion welding
welding and
and PWHT
PWHT on on
the hardness of the samples. The microhardness profile of the sample before PWHT,
the hardness of the samples. The microhardness profile of the sample before PWHT, as the function as the function
of distance
of distance from
from the
the HAZ
HAZ along
along thethe wall
wall build
build direction,
direction, is
is shown
shown in in Figure
Figure 7a.
7a. The
The BM
BM (distance
(distance
from −5
from ´5toto0 mm)
0 mm)showed
showed little variation
little in readings
variation with with
in readings the mean value of
the mean 188 HV1.
value of 188This
HV1.is expected
This is
due to the uniform microstructure of the BM. The HAZ (at 0 mm) in samples
expected due to the uniform microstructure of the BM. The HAZ (at 0 mm) in samples showed higher showed higher values
of hardness
values due todue
of hardness thetoformation
the formationof martensite.
of martensite.Large variations
Large variations were
wereobserved
observedas as there exist
there exist
partially-transformed martensitic products and the width of HAZ is generally slightly
partially-transformed martensitic products and the width of HAZ is generally slightly larger than the larger than the
diagonal of indention left by Vickers Hardness Test. It should be noted that the largest
diagonal of indention left by Vickers Hardness Test. It should be noted that the largest variations of variations of
hardness were
hardness wereshown
shownininthetheregion
regionofofthe
the first
first deposited
deposited layer
layer (at(at distance
distance approximately
approximately 1 mm)1 mm) as
as the
the microstructure in this region underwent significant change due to the different heat input and
cooling rate. The remainder region of as-deposited WM (distance >2 mm) shows very uniform
Materials 2016, 9, 652 8 of 12

microstructure in this region underwent significant change due to the different heat input and cooling
Materials 2016, 9, 652 8 of 12
rate. The remainder region of as-deposited WM (distance >2 mm) shows very uniform hardness
hardness (averaged
(averaged to 181 HV1) to 181 HV1)iswhich
which is also comparable
also comparable to the to
BM. theWith
BM. With
PWHT, PWHT, as shown
as shown in Figure
in Figure 7b,
7b, hardness
the the hardness in BM in shows
BM shows littlelittle change,
change, whilewhile
in theinHAZ,
the HAZ,
shiftsshifts
to thetosame
the same
valuevalue
as theasBM.theIn BM.WM, In
Materials 2016, 9, 652 8 of 12
WM,
the the hardness
hardness is improved
is improved around around 15% from
15% from 181 to181
210toHV1
210 HV1
whichwhich is favorable.
is favorable. ThisThis
is due is due
to the to
the refined
refined grains
hardness grains (as shown
(as shown
(averaged toin
181 in Figure
Figure
HV1) is5g)
5g) and
which and
more
also more precipitations
precipitations
comparable to theof ofPWHT,
intermetallic
intermetallic
BM. With inasαshowninin(as
phase αFigure
phase
shown(as in
shown
Figure in
6). Figure 6).
7b, the hardness in BM shows little change, while in the HAZ, shifts to the same value as the BM. In
WM, the hardness is improved around 15% from 181 to 210 HV1 which is favorable. This is due to
the refined grains (as shown in Figure 5g) and more precipitations of intermetallic in α phase (as
shown in Figure 6).

Figure 7.7.Microhardness
Figure Microhardness profile
profile in cross-section
in the the cross-section
as theas the function
function of the distance
of the distance from
from HAZ. HAZ.
(a) Before
(a) Before
PWHT; (b)PWHT; (b) After PWHT.
After PWHT.
Figure 7. Microhardness profile in the cross-section as the function of the distance from HAZ.
(a) Before PWHT; (b) After PWHT.
The room
The roomtemperature
temperature tensile
tensile properties
properties of as-deposited
of the the as-deposited
NAB are NAB are shown
shown in Figurein8.Figure
Samples 8.
Samples
from from
The
three threetemperature
room
deposited deposited
walls, usingwalls, using
tensiledifferent different
properties of thewelding
welding parameters,
as-deposited
parameters, NAB
are are are
named named
shown“1”, “1”, and
in “2”,
Figure “2”, and
8. “3”,
“3”, respectively
Samples from as shown
three in
depositedFigure 8.
walls, A uniform
using yield
different strength
welding (YS),
parameters,ultimate
respectively as shown in Figure 8. A uniform yield strength (YS), ultimate tensile strength (UTS),are tensile
named strength
“1”, “2”, and (UTS),
“3”, respectively
and elongation
elongation as
areshown
(%) are found in in
Figure 8. A uniform
all three
three testingyield
wallsstrength
with the(YS),
the ultimate tensile
variations strength
in average
average (UTS),being
values
and (%) found in all testing walls with variations in values being
and elongation
approximately26 (%)
26MPa,
MPa,35 are found
35MPa,
MPa,and in all
and2%, three testing
2%,respectively.walls
respectively. with
This the variations
indicates that in average values being
approximately This indicates that thethe effect
effect of heat
of heat inputinput on
on the
the approximatelyproperties
mechanical 26 MPa, 35 of MPa, the and 2%, respectively.
additively This indicates
manufactured that the effect
component is of heat input on
insignificant. This is
mechanical properties
the mechanical of the additively
properties manufactured
of the additively component
manufactured is insignificant.
component This is desirable
is insignificant. This is as a
desirable as
widedesirable a
range of as wide
welding range of welding parameters can be used for the WAAM of NAB while
a wideparameters
range of welding can beparameters
used for the canWAAM
be usedofforNAB thewhile
WAAM maintaining
of NAB while uniform
maintainingproperties.
mechanical uniform mechanical properties.
maintaining uniform mechanical properties.

Figure 8. Effects of heat input on mechanical properties of as-deposited WM. Each data point
Figure 8. Effects of heat input on mechanical properties of as-deposited WM. Each data point represents
represents the average of more than 10 specimens from one of three welding heat inputs.
the average
Figure of more
8. Effects ofthan
heat10input
specimens from one ofproperties
on mechanical three welding heat inputs. WM. Each data point
of as-deposited
represents
As thethe average
walls wereofdeposited
more than 10 specimens from
layer-upon-layer, it isone of three welding
interesting heat inputs.
to understand the variation in
mechanical properties along the wall build direction. Figure 9 shows the mechanical properties of
As the walls
samples wereone
taken from deposited layer-upon-layer,
of the as-deposited it is interesting
walls at different to understand
vertical locations. the variation
As expected, the BM in
mechanical properties along the wall build direction. Figure 9 shows the mechanical properties of
samples taken from one of the as-deposited walls at different vertical locations. As expected, the BM
Materials 2016, 9, 652 9 of 12

As the walls were deposited layer-upon-layer, it is interesting to understand the variation in

mechanical
Materials properties
2016, 9, 652 along the wall build direction. Figure 9 shows the mechanical properties 9 of of
12
samples taken from one of the as-deposited walls at different vertical locations. As expected, the BM
has
has uniform
uniform mechanical
mechanical properties
properties withwith the
the average
average values
values ofof YS,
YS, UTS,
UTS, and
and elongation
elongation being
being 332
332 MPa,
MPa,
689
689 MPa,
MPa, and
and 25.3%,
25.3%, respectively.
respectively. In In the
the HAZ,
HAZ, thethe higher
higher YS YS of
of 394
394 MPa
MPa andand higher
higher UTS
UTS of of 711
711 MPa,
MPa,
while
while aa lower
lower elongation
elongation of of 19.6%
19.6% was was found,
found, which
which is is predictable.
predictable. The
The values
values ofof UTS
UTS andand YSYS in
in the
the
WM
WM span 339–342 MPa and 630–653 MPa, respectively. There is no significant variation (less than
span 339–342 MPa and 630–653 MPa, respectively. There is no significant variation (less than
3.6%)
3.6%) in
in tensile
tensile strength
strength and
and yield
yield strength
strength inin WM
WM which
which is is plausible.
plausible. The
The average
average elongation
elongation in in WM
WM
is 27.8%. The microstructure in the HAZ is predominantly partially transformed
is 27.8%. The microstructure in the HAZ is predominantly partially transformed eutectoid structure, eutectoid structure,
in
in particular, in the
particular, in the area close to
area close to the
the fusion
fusion line.
line. The
The heat
heat input
input isis high
high enough
enough to to raise
raise the
the temperature
temperature
above
above the eutectic point, and therefore forming the β phase. This was retained with rapid cooling,
the eutectic point, and therefore forming the β phase. This was retained with rapid cooling,
resulting
resulting ininaamartensitic
martensiticstructure
structure that
that is harder
is harder thanthan
the the α phase,
α phase, but less
but also alsoductile
less ductile
and moreandbrittle.
more
brittle. The mechanical
The mechanical properties
properties across the across
weldthe weld
metal aremetal are expected
expected to be reasonably
to be reasonably consistent.consistent.
From the
From the above data analysis, it is revealed that the mechanical properties of
above data analysis, it is revealed that the mechanical properties of as-deposited WM are comparableas-deposited WM are
comparable to those
to those of as-cast BM. of as-cast BM.

Figure
Figure 9.
9. Location
Location effect
effect on
on mechanical
mechanical properties
properties of
of the
the NAB
NAB component
component produced by WAAM.

To investigate the effects of PWHT on mechanical properties of both as-cast and as-deposited
To investigate the effects of PWHT on mechanical properties of both as-cast and as-deposited NAB,
NAB, a comparison of each tensile experiment was performed with the results shown in Figure 10.
a comparison of each tensile experiment was performed with the results shown in Figure 10. The data
The data from different heat inputs were collected together with each data point in Figure 10
from different heat inputs were collected together with each data point in Figure 10 representing the
representing the average value of samples from three different welding parameters. Figure 10 shows
average value of samples from three different welding parameters. Figure 10 shows that: (1) mechanical
that: (1) mechanical properties of as-deposited WM are comparable to those of as-cast BM; and (2)
properties of as-deposited WM are comparable to those of as-cast BM; and (2) with the PWHT, there is
with the PWHT, there is little change in cast NAB components, while there is an improvement in
little change in cast NAB components, while there is an improvement in tensile strength and a decrease
tensile strength and a decrease in elongation of PWHT WM. The improved mechanical properties of
in elongation of PWHT WM. The improved mechanical properties of PWHT WM are mainly due to
PWHT WM are mainly due to the refinement in grain size and precipitations in the α phase during
the refinement in grain size and precipitations in the α phase during the heat treatment. These were
the heat treatment. These were confirmed by XRD results and microstructural observations.
confirmed by XRD results and microstructural observations. Therefore, the WAAM technique is
Therefore, the WAAM technique is able to produce NAB components with comparable or even
able to produce NAB components with comparable or even superior mechanical properties than
superior mechanical properties than casting technology. It should be noted that the cast plate has the
casting technology. It should be noted that the cast plate has the thickness of 30 mm that would
thickness of 30 mm that would not induce severe casting porosity. For marine applications, potential
not induce severe casting porosity. For marine applications, potential porosities trapped in large
porosities trapped in large components would reduce mechanical properties, while that would not
components would reduce mechanical properties, while that would not be an issue in wire arc
be an issue in wire arc additively-manufactured components.
additively-manufactured components.
Materials 2016, 9, 652 10 of 12
Materials 2016, 9, 652 10 of 12

Materials 2016, 9, 652 10 of 12

FigureFigure 10. Effects

10. Effects of heat
of heat treatment
treatment on mechanical
on mechanical properties
properties ofof as-castBM
as-cast BMandandas-deposited
as-deposited WM.
WM. Each
Each data point represents the average of samples from three different welding parameters,
data point represents the average of samples from three different welding parameters, corresponding
to thecorresponding
Figure
three-wall to the
10. Effects three-wall
of heat
samples weresamples
thattreatment that were produced.
on mechanical
produced. properties of as-cast BM and as-deposited WM.
Each data point represents the average of samples from three different welding parameters,
Representative
corresponding tofracture surface
the three-wall of thethat
samples tensile
were samples
produced.are shown in Figure 11a–d. The fracture
Representative fracture
morphologies in as-cast BM surface
and heatof treated
the tensile samples
BM after tensileare shown
tests in Figure
are quite similar.11a–d. The fracture
In all cases, the
morphologies
fracture in as-cast
Representative
surfaces BMa and
fracture
exhibit heatof
surface
mixture oftreated BMsamples
the tensile
brittle aftermodel
cleavage tensile tests
are shown
and inare quite
Figure
microvoid similar.
11a–d. In allallcases,
The with
coalescence fracture
morphologies
elongations
the fracture in as-cast
greater
surfaces than BM
exhibit aand
20%. Theheat treated
as-deposited
mixture BM
of brittle after
WM tensilemodel
contains
cleavage tests are quite
predominantly
and similar. Incoalescence
all cases,
Widmanstätten
microvoid the with
α. The
fracture surfaces
surface exhibit
exhibits a
the mixture
feature ofof brittle
lath cleavage
structure, model
indicating and
the microvoid
fracture
all elongations greater than 20%. The as-deposited WM contains predominantly Widmanstätten coalescence
preferentially with
propagates all
elongations
along the greater
lath than 20%.The
boundaries. Theheat
as-deposited
treatment WM contains
relieves predominantly
residual stresses Widmanstätten
and homogenizes α. The
the
α. The fracture surface exhibits the feature of lath structure, indicating the fracture preferentially
fracture surfacein
microstructure exhibits
the weldthe feature of lath structure,
metal. Figure indicating
5 clearly shows thatthe
thefracture preferentially
heat treatment propagates
decomposed the
propagates along the lath boundaries. The heat treatment relieves residual stresses and homogenizes
along the
basket lath boundaries.
weave-like WidmanstättenThe heat treatment
structure, leavingrelieves
a fine, residual
homogenized stresses and homogenizes
microstructure of α phasethe
the microstructure
microstructure
in the weld
in the
metal. Figure5 5clearly
clearly shows that the heat treatment decomposed the
and κ precipitates. Theweld metal. Figure
fine dimples shows that
on the fracture surface the heat treatment
are consistent decomposed the
with the microstructure.
basket weave-like
basket Widmanstätten
weave-like Widmanstättenstructure,
structure,leaving
leaving aa fine,
fine,homogenized
homogenized microstructure
microstructure of α of α phase
phase
and κandprecipitates. TheThe
κ precipitates. finefine
dimples
dimples ononthethefracture
fracture surface
surface areareconsistent
consistent with
with thethe microstructure.
microstructure.

Figure 11. SEM fractographs of (a) as-cast BM; (b) heat treated BM; (c) as-deposited WM; and (d) heat
treated WM.
Figure 11. SEM fractographs of (a) as-cast BM; (b) heat treated BM; (c) as-deposited WM; and (d) heat
Figure 11. SEM fractographs of (a) as-cast BM; (b) heat treated BM; (c) as-deposited WM; and (d) heat
treated WM.
treated WM.
Materials 2016, 9, 652 11 of 12

4. Conclusions
The intent of this work is to demonstrate that the WAAM technique has the potential to fabricate
functional NAB alloy. We have shown that fully dense NAB components were successfully produced
using WAAM with various welding parameters. The variation of mechanical properties among all
weld metals was insignificant. The as-deposited materials exhibits relatively fine microstructures,
as well as comparable mechanical properties, with respect to as-cast NAB alloy. PWHT further refined
the microstructure, improved tensile strength and hardness in the weld metal, revealing that WAAM
is a promising alternative for manufacturing superior NAB components. Extending this work would
involve investigating corrosion behavior of the additively-manufactured NAB alloy and the fabrication
of near-net-shaped NAB components for marine applications.

Acknowledgments: This research was carried out at the Materials Research Lab, University of Wollongong.
The work was supported by Defence Materials Technologies Centre (DMTC), which was established and is
supported by the Australia Government’s Defence Future Capability Technology Centre (DFCTC) initiative.
Author Contributions: D.D. designed the research, performed all experiments and wrote the manuscript, Z.P.,
H.L., and S.v.D. designed the research and edited the paper, C.S. contributed the X-ray diffraction and optical
microscopy investigation.
Conflicts of Interest: The authors cite no conflict of interest.

References
1. Brezina, P. Heat treatment of complex aluminium bronzes. Int. Mater. Rev. 1982, 27, 77–120. [CrossRef]
2. Culpan, E.; Rose, G. Microstructural characterization of cast nickel aluminium bronze. J. Mater. Sci. 1978, 13,
1647–1657. [CrossRef]
3. Yu, H.; Zheng, Y.; Yao, Z. Cavitation erosion corrosion behaviour of manganese-nickel-aluminum bronze in
comparison with manganese-brass. J. Mater. Sci. Technol. 2009, 25, 758–766.
4. Ferrara, R.; Caton, T. Review of dealloying of cast aluminum bronze and nickel-aluminum bronze alloys in
sea water service. Mater. Performance 1982, 21, 30–34.
5. Fuller, M.; Swaminathan, A.; Zhilyaev, A.; McNelley, T. Microstructural transformations and mechanical
properties of cast NiAl bronze: Effects of fusion welding and friction stir processing. Mater. Sci. Eng. A 2007,
463, 128–137. [CrossRef]
6. Wu, Z.; Cheng, Y.F.; Liu, L.; Lv, W.; Hu, W. Effect of heat treatment on microstructure evolution and
erosion–corrosion behavior of a nickel–aluminum bronze alloy in chloride solution. Corros. Sci. 2015, 98,
260–270. [CrossRef]
7. Ma, Z. Friction stir processing technology: A review. Metall. Mater. Trans. A 2008, 39, 642–658. [CrossRef]
8. Li, H.; Grudgings, D.; Larkin, N.; Norrish, J.; Callagan, M.; Kuzmikova, L. Optimization of welding
parameters for repairing NiAl bronze components. Mater. Sci. Forum 2012, 706, 2980–2985. [CrossRef]
9. Almeida, P.S.; Williams, S. Innovative process model of Ti-6Al-4V additive layer manufacturing using cold
metal transfer (CMT). In Proceedings of the Twenty-First Annual International Solid Freeform Fabrication
Symposium, Austin, TX, USA, 9–11 August 2010; pp. 25–36.
10. Mok, S.H.; Bi, G.; Folkes, J.; Pashby, I.; Segal, J. Deposition of Ti–6Al–4V using a high power diode laser and
wire, part II: Investigation on the mechanical properties. Surf. Coat. Technol. 2008, 202, 4613–4619. [CrossRef]
11. Brandl, E.; Palm, F.; Michailov, V.; Viehweger, B.; Leyens, C. Mechanical properties of additive manufacturing
titanium (Ti-6Al-4V) blocks deposited by a solid-state laser and wire. Mater. Des. 2011, 32, 4665–4675.
[CrossRef]
12. Ding, D.; Pan, Z.; Cuiuri, D.; Li, H. Wire-feed additive manufacturing of metal components: Technologies,
developments and future interests. Int. J. Adv. Manuf. Tech. 2015, 81, 465–481. [CrossRef]
13. Wang, F.; Williams, S.; Rush, M. Morphology investigation on direct current pulsed gas tungsten arc welded
additive layer manufactured Ti6Al4V alloy. Int. J. Adv. Manuf. Tech. 2011, 57, 597–603. [CrossRef]
14. Shen, C.; Pan, Z.; Cuiuri, D.; Roberts, J.; Li, H. Fabrication of Fe-FeAl Functionally Graded Material Using
the Wire-Arc Additive Manufacturing Process. Metall. Mater. Trans. B 2016, 47, 763–772. [CrossRef]
Materials 2016, 9, 652 12 of 12

15. Sabbaghzadeh, B.; Parvizi, R.; Davoodi, A.; Moayed, M.H. Corrosion evaluation of multi-pass welded
nickel–aluminum bronze alloy in 3.5% sodium chloride solution: A restorative application of gas tungsten
arc welding process. Mater. Des. 2014, 58, 346–356. [CrossRef]
16. Wharton, J.A.; Barik, R.C.; Kear, G.; Wood, R.J.K.; Stokes, K.R.; Walsh, F.C. The corrosion of nickel–aluminium
bronze in seawater. Corros. Sci. 2005, 47, 3336–3367. [CrossRef]
17. Liu, C.M.; Tian, X.J.; Tang, H.B.; Wang, H.M. Microstructural characterization of laser melting deposited
Ti–5Al-5Mo–5V–1Cr–1Fe near β titanium alloy. J. Alloys Compd. 2013, 572, 17–24. [CrossRef]
18. Hasan, F.; Jahanafrooz, A.; Lorimer, G.W.; Ridley, N. Morphology, crystallography, and chemistry of phases
in as-cast nickel-aluminium bronze. Metal. Trans. A 1982, 13, 1337–1345. [CrossRef]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).