Materials Science and Engineering A 509 (2009) 98–104

Contents lists available at ScienceDirect

Materials Science and Engineering A

journal homepage: www.elsevier.com/locate/msea

Laser aided direct metal deposition of Inconel 625 superalloy: Microstructural

evolution and thermal stability
G.P. Dinda a,b,∗ , A.K. Dasgupta a , J. Mazumder b
a
Center for Advanced Technologies, Focus: HOPE, Detroit, MI 48238, USA
b
Center for Laser Aided Intelligent Manufacturing, University of Michigan, Ann Arbor, MI 48109, USA

a r t i c l e i n f o a b s t r a c t

Article history: Direct metal deposition technology is an emerging laser aided manufacturing technology based on a
Received 30 October 2008 new additive manufacturing principle, which combines laser cladding with rapid prototyping into a solid
Received in revised form 2 January 2009 freeform fabrication process that can be used to manufacture near net shape components from their CAD
Accepted 6 January 2009
files. In the present study, direct metal deposition technology was successfully used to fabricate a series
of samples of the Ni-based superalloy Inconel 625. A high power CO2 laser was used to create a molten
Keywords:
pool on the Inconel 625 substrate into which an Inconel 625 powder stream was delivered to create a 3D
Laser deposition
object. The structure and properties of the deposits were investigated using optical and scanning electron
Inconel 625
Microstructure
microscopy, X-ray diffraction and microhardness test. The microstructure has been found to be columnar
X-ray diffraction dendritic in nature, which grew epitaxially from the substrate. The thermal stability of the dendritic
Microhardness morphology was investigated in the temperature range 800–1200 ◦ C. These studies demonstrate that
Inconel 625 is an attractive material for laser deposition as all samples produced in this study are free
from relevant defects such as cracks, bonding error and porosity.
© 2009 Elsevier B.V. All rights reserved.

1. Introduction
Inconel 625 (Alloy 625) is a nickel-based superalloy strength-
ened mainly by the solid-solution hardening effect of the refractory
metals, niobium and molybdenum, in a nickel–chromium matrix
[1]. Alloy 625 was originally developed as a solid-solution strength-
ened material. It was soon determined that the alloy is somewhat
precipitation (age) hardenable [2–5]. Inconel 625 exhibits precipi-
tation hardening mainly due to the precipitation of fine metastable
phase ␥ [Ni3 Nb] after annealing over a long period in the tem-
perature range 550–850 ◦ C [3]. Moreover, various forms of carbides
(MC, M6 C and M23 C6 ) can also precipitate depending upon the time
and temperature of ageing. Alloy 625 has found extensive use in
many industries for diverse applications over a wide temperature
range from cryogenic conditions to ultra hot environments over
1000 ◦ C [6–9]. The alloy is endowed with good combination of yield
strength, creep strength, fatigue strength and excellent oxidation
and corrosion resistance in aggressive environments. Moreover, its
good weldability and fabricability have made it the choice for many
diverse applications. Thus, over 50 years, alloy 625 has been widely
used in aerospace, chemical, petrochemical and marine applica-
tions. However, many of the Inconel 625 components are highly
∗ Corresponding author at: Center for Advanced Technologies, 1400 Oakman Blvd.,
Focus: HOPE, Detroit, MI 48238, USA. Tel.: +1 313 494 4452; fax: +1 313 481 6171.
E-mail address: dindag@focushope.edu (G.P. Dinda).
complex shapes that are very expensive to produce due to extensive
machining [10].
Direct metal deposition (DMD) technology [11,12], developed at
the University of Michigan, is a laser aided rapid manufacturing
technology which can be used to fabricate porous or solid metallic
parts directly from CAD model. Manufacturing processes, similar to
DMD, have been developed with different names at various labo-
ratories, such as direct light fabrication (DLF) [13,14] at Los Alamos
National Laboratories, Laser Engineering Net Shaping (LENS) [15,16]
at Sandia National Laboratories, selective laser powder remelting
(SLPR) [17] at Fraunhofer Institute (Germany), etc. The basic prin-
ciples of these technologies are similar that they use a focused
laser beam as a heat source to melt metallic powder and create
a three-dimensional (3D) object. The key characteristics of these
manufacturing processes such as DMD and LENS have a feedback
control system that provides a close-loop control to maintain a uni-
form deposition thickness, thus saving precious post-machining
time.
Like rapid prototyping, the DMD process starts with a CAD
design. First, the CAD model of the component is sliced into a series
of parallel layers with a build height typically 1/4th to 1/3rd of
the laser beam diameter. Then, the tool path is generated to fill
each layer and subsequently the tool path output data is then post-
processed, converting the tool path data into conventional CNC G
and M codes. Next, the post-processed data is downloaded into
the DMD machine. In the DMD process, a high-energy laser beam
is focused onto the substrate or a previously deposited layer to

0921-5093/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.msea.2009.01.009
 G.P. Dinda et al. / Materials Science and Engineering A 509 (2009) 98–104 99

form a melt pool; metal powders are simultaneously delivered into Table 1
Experimental design based on L9 Taguchi matrix.
the melt pool by a specially designed coaxial nozzle. The nozzle is
designed such that the powder streams converge at the same point Run Laser Scan speed Powder feed Hardness
on the focused laser beam. A CNC system is used to control the noz- order power (W) (mm/min) rate (g/min) (VHN)
zle and beam focusing optics as per the tool path generated from L1 600 300 8 255
the CAD model. Thus, a three-dimensional object is formed layer by L2 600 375 10 257
layer. L3 600 450 12 263
L4 750 300 10 254
The DMD technology has many potential applications, includ-
L5 750 375 12 260
ing production of near net shaped components, surface coatings, L6 750 450 8 262
part repairing, and fabricating functionally graded materials. How- L7 900 300 12 248
ever, the production rate of the DMD process is low compared to L8 900 375 8 254
L9 900 450 10 260
the conventional manufacturing processes. Hence, DMD process is
especially suitable for repairing or fabricating high-value parts with
low production volume. As a particular application, DMD is very
attractive to the aerospace and biomedical industries for produc-
ing complex-shaped metallic components, which is very difficult to Ni-based superalloy clad by epitaxial growth onto a single crystal
produce by conventional manufacturing processes. Due to the very Ni-based superalloy substrate. However, very little work is reported
high cooling rate of the laser deposition process, the microstruc- on laser aided manufacturing of Inconel 625 in the open literature
ture of the resulting part is different from equilibrium structure. [20]. In this paper, we present the microstructural evolution, ther-
Recent studies by Gaumann et al. [18,19] using a laser metal form- mal stability and mechanical response of laser deposited Inconel
ing technique indicated that it is possible to deposit a single crystal 625.

Fig. 1. The transverse-section images of nine as-deposited samples produced with different combination of laser deposition parameters as per L9 Taguchi matrix (Table 1).
100 G.P. Dinda et al. / Materials Science and Engineering A 509 (2009) 98–104
Fig. 2. The optical micrographs of the longitudinal section of the as-deposited samples (a) L1 and (b) L1A.
2. Experimental
Commercially available, gas atomized Inconel 625 powder
(Praxript Inc.) was used for this study. The powder had a com-
position of 64.61 Ni, 21.25 Cr, 8.45 Mo, 4.65 Nb and 1.06 Fe (in
wt.%) with a powder size distribution of 45–135 ␮m diameter.
The substrate was an Inconel 625 rolled plate with dimensions of
150 mm × 100 mm × 7 mm. There are many variables in the DMD
process that can affect the microstructure and mechanical prop-
erties of the resulting part, such as laser power, powder feed rate
and laser scanning speed. In the present study, a statistical design
of experiments based on L9 orthogonal array of Taguchi method
was employed to assess the effect of different processing parame-
ters on the microstructure and mechanical properties of the laser
deposited samples.
Laser deposits were made using a DMD system, developed at
the University of Michigan. A 6 kW CO2 laser unit with a laser beam
diameter of 0.5 mm was used in this experiment. To prevent oxi-
dation, helium gas was used as a shielding gas as well as powder
carrier gas. A combination of the following DMD parameters—laser
power 600, 750 and 900 W, laser scanning speed 300, 450 and
600 mm/min, and powder feed rate of 8, 10, and 12 g/min were
used as per the L9 orthogonal array of Taguchi matrix (Table 1).
Nine different thin wall samples of dimensions 50 mm × 8 mm lay-
ers, with thickness varying between 0.6 and 1.2 mm according to
the process condition, were produced using the DMD machine
with a single laser pass in a helium atmosphere. The Z-increment
between layers was 0.25 mm in all the samples examined in this
paper. Samples were fabricated on a substrate of Inconel 625 rolled
plates.
The microstructures of the longitudinal (x–z), transverse (y–z)
and horizontal (x–y) sections of the as-deposited samples were
investigated by optical microscopy (Nikon eclipse ME 600) and
scanning electron microscopy (Philips XL30 FEG SEM). The crys-
tal structure and the lattice parameter of the as-deposited samples
were examined using a Rigaku rotating anode X-ray diffractometer
by Cu K␣ radiation at 40 kV and 100 mA. All samples were scanned
in the standard –2 geometry from 30◦ to 100◦ with a 0.01◦ step
size and 1-s dwell time. A mixture of 10 ml HNO3 , 15 ml HCl and
10 ml CH3 –COOH etchant was used for the optical and scanning
electron microscopy. The hardness of the sample was measured
using a Vickers microhardness tester (Clark, CM-400 AT) using a
500 gm load for a dwell time of 10 s. The final hardness value was
calculated by averaging data from five indentations.
To study the thermal stability of the as-deposited structure, a
set of samples were annealed at 700, 800, 900, 1000, 1100 and
1200 ◦ C for 1 h followed by cooling in still air. The heat treatment
was performed in argon gas atmosphere in a horizontal cylindrical
furnace.
3. Results and discussion
Fig. 1 shows the transverse-section images of nine deposits pro-
duced with different combination of laser deposition parameters as
per the L9 Taguchi matrix (Table 1). The primary goal of the design of
experiments was to optimize laser deposition parameters for pro-
ducing samples with minimum porosity and maximum deposition
rate. However, it should be noted that all combination of deposi-
tion parameters selected in this study (power: 600–900 W, speed:
300–450 mm/min, and feed rate: 8–12 g/min), produced samples
with no relevant defects such as cracks, bonding error at the inter-
face between deposit and substrate or pores in the deposits. This
observation revealed that Inconel 625 is a suitable material for laser
deposition with a wide process window. It can be seen from Fig. 1
that the maximum deposition rate was observed in the sample
L7 that was produced with maximum power (900 W), minimum
scanning speed (300 mm/min) and maximum powder feed rate
(12 g/min). It is obvious that a high laser energy melts large amounts
of powder and eventually results in a higher deposition rate.
The optical micrographs of the longitudinal section of the
deposits L1 and L1A are shown in Fig. 2. Both samples were pro-
duced with the same laser deposition parameters (power: 600 W,
speed: 300 mm/min, and feed rate: 8 g/min), however, laser scan
direction was different as schematically shown in Fig. 2. For sample
L1A, the laser scan direction for all layers was identical (from right to
left), whereas for sample L1, the scan direction of each layer changed
by 180◦ with the direction of previously deposited layer. Microstruc-
ture of longitudinal section revealed the pattern of layer-by-layer
deposition. It should be pointed out that the microstructure mostly
consists of columnar dendrites, which grew epitaxially from the
substrate. Moreover, the growth direction of the columnar den-
drites was tied to the laser scanning direction. From Fig. 2(b), it can
be seen that the primary dendrites have almost same orientation
 G.P. Dinda et al. / Materials Science and Engineering A 509 (2009) 98–104
Fig. 3. Transverse-section microstructures at different locations of the sample L1.
throughout the sample. In contrast, Fig. 2(a) shows that the growth
direction of the columnar dendrites changes by 90◦ in each layer.
It should be also noted that the epitaxial growth direction of the
primary dendrites was at an angle of around 60◦ with the substrate
of the sample that was produced with single laser scan direction.
When laser scan direction changes at each layer, the growth angle of
primary dendrites changes to 45◦ with the substrate. In laser depo-
sition, the solid (substrate and deposit) acts as a heat sink. Hence,
during solidification of the melt pool, cooling mostly occurs via the
substrate and the deposit. This leads to the directional growth of
the grains counter to the heat flux direction and subsequently the
formation of the columnar grains. The columnar dendritic growth
is caused by epitaxial growth of primary dendrite from the par-
tially remelted grains of the previously deposited layer, which acts
as pre-nuclei for directional growth of crystal. It has been reported
in the literature [18,19] that the dendrite growth direction is mostly
perpendicular to the substrate of the laser deposited samples. How-
ever, in the present investigation we observed that this angle is
about 60◦ and 45◦ with the substrate depending upon the laser
scanning direction. As described previously, the growth direction
of the dendrites is opposite to the resultant heat flux direction. If
the cooling of the melt pool fully occurs via the substrate, then the
growth direction of the crystal will be vertical (perpendicular to the
substrate). However, during laser deposition, cooling mostly occurs
via the substrate or previously deposited layer and partially via the
adjacent solidified deposit layer which grew from the trailing end of
the laser melt pool. Since the temperature of the deposit at the back
of the melt pool is quite high compared to the substrate; the resul-
tant heat flux direction will be at an angle of more than 45◦ and less
than 90◦ with the substrate. In the present investigation, this angle
appeared to be 60◦ which demonstrates that the vertical heat flux
via the substrate was twice intense than the horizontal heat flux via
the deposit towards to trailing side of the melt pool). However, same
laser deposition parameters produced a primary dendrite growth
angle of 45◦ with the substrate when the laser scan direction was
101
changed after each layer. In this case both heat flux and dendrite
growth morphology play equal role in determining the angle of
dendritic growth. The cooling direction of the melt pool changes
with the direction of laser scan. As a result, the heat flux direction
becomes very close to the direction of secondary dendrites of the
previously deposited layer as secondary dendrites are near perpen-
dicular to the primary dendrites. Therefore, during laser deposition
for this strategy, the secondary dendrites of the previous deposit
layer act as a growth site for the primary dendrites of the top layer.
Hence, the angles of the primary dendrites of the subsequent layers
become 90◦ .
Fig. 3 shows the microstructure of the transverse section of the
sample L1. It can be seen from Fig. 3 that the microstructure of the
each layer is composed of two regions. Bottom part of each layer
is mostly a columnar structure which often shows no secondary
dendrites associated with it. The top part of the layer shows typical
fine dendritic structure with classical secondary dendrite arms. In
laser deposition process, the cooling rate of the melt pool and con-
sequently the solidification velocity is very high at the bottom part
of the layer. The upper part of the layer cools down slowly compared
to the bottom part. Due to the very high solidification velocity of the
bottom part of the melt pool, secondary dendrites could not grow.
As a result bottom part of each layer consisted of mainly primary
dendrites. Because of the progressive decrease of the cooling rate
from the bottom to the top part of the melt pool, a gradual transi-
tion of microstructure from fully columnar to dendritic transition is
observed. The average primary dendrite spacing at the bottom part
the layer is about 5 ␮m. The secondary dendrite arm spacing at the
upper part of each layer varies between 1.5 and 2.5 ␮m.
The microstructure of the horizontal section, i.e., cross-section
perpendicular to the build direction is shown in Fig. 4. The mor-
phology of the microstructure of this section is typically equiaxed
cellular structure. Since the as-deposited microstructure is near
vertical in orientation, therefore, it appears as equiaxed structure
in the section perpendicular to the build direction of the deposit.
102 G.P. Dinda et al. / Materials Science and Engineering A 509 (2009) 98–104
Fig. 4. The microstructure of the horizontal section, i.e., cross-section perpendicular
to the build direction.
As described previously, the microstructure of the transverse sec-
tion varies at different location of each layer depending upon the
local cooling rate. A detailed microstructural morphology of the as-
deposited sample from bottom to top layer can be seen in Fig. 5.
It shows isometric (three-dimensional) view of microstructure at
Fig. 5. Showing three-dimensional view of microstructures at three location of the deposit: (a) first layer, (b) middle of the sample (between first and top layer), and (c) lop
layer.
three locations of the deposit: (a) first layer, (b) middle of the sample
(between first and top layer), and (c) lop layer. As can be seen from
the three dimensional view that the microstructural elements are
mostly columnar dendrites from bottom to top layers of the deposit.
However, an orientation transition from horizontal dendrites to the
vertical dendrites was observed at the upper part of the top layer.
This dendrite orientation change was observed at the uppermost
part (∼20–50 ␮m thick) of each layer. Due to the partial remelting
of the upper part of the previously deposited layer, this transition
layer remelted during subsequent layers deposition. Hence, only
topmost layer shows this orientation transition. This transforma-
tion is due to the changes in the heat flux direction at the topmost
part of the melt pool. At the top surface of the melt pool, the mag-
nitude of heat flux to the back of the melt pool is higher compared
to the substrate due to the hemispherical shape of the melt pool. As
a result, at the very top part of each layer, dendrite grew from the
back of the laser melt pool which appears as horizontal grain in the
deposit.
To study the thermal stability of the dendritic structure, i.e.,
effect of post-deposition heat treatment on microstructure and
mechanical properties, the as-deposited samples were annealed at
700, 800, 900, 1000, 1100, and 1200 ◦ C for 1 h followed by cooling in
air. Fig. 6 shows the microstructures of the laser deposited sample
annealed at different temperatures. It can be seen from Fig. 6 that
the dendritic structure was stable up to 1000 ◦ C. There is little indi-
cation of crystallization near the interface between substrate and
 G.P. Dinda et al. / Materials Science and Engineering A 509 (2009) 98–104 103

Fig. 6. The microstructures of the laser deposited sample annealed at different temperatures as shown in figure.

deposit at 1000 ◦ C. As shown in Fig. 6(a) obtained from specimen Therefore, as can be seen from Fig. 7, the decrease of hardness with
annealed at 1000 ◦ C, some new grain boundaries were formed in increasing annealing temperature is attributed to the formation of
the dendritic structure near the substrate–deposit interface. How- stable ␦ phase between 700 and 900 ◦ C followed by coarsening of
ever, most of the sample shows fully dendritic structure. A mixture grains as well as precipitates and finally dissolution of ␦ precipitate
of recrystallized (equiaxed) and dendritic microstructures can be into the ␥-matrix above 1000 ◦ C.
seen in the sample annealed at 1100 ◦ C. The edges of the sam- The X-ray diffraction patterns of the laser deposited sample
ple show fully recrystallized grains. When the as-deposited sample annealed at different temperatures are shown in Fig. 8. It can be seen
was annealed at 1200 ◦ C, the microstructure has no dendritic char- that the as-deposited as well as annealed samples show the pres-
acteristic. Fig. 5(c) shows that fully equiaxed grain structure was ence of only ␥-Ni (fcc) solid solution. At elevated temperature some
developed in the laser deposited sample post-heated at 1200 ◦ C. A intermetallic phases such as ␥ [Ni3 Al], ␥ [Ni3 Nb] and ␦ [Ni3 Nb]
significant grain growth occurred at 1200 ◦ C. Moreover, the grain may be precipitated in ␥-Ni matrix. However, these intermetallic
size distribution is inhomogeneous. The center part of the sam- phases were not detected by XRD due to the very fine nature of these
ple shows very coarse grains of average diameter about 250 ␮m, precipitates. Without prolonged annealing and coarsening of the
whereas, edges show relatively finer grain structure of about precipitate, XRD scan cannot detect it. However, lattice parameter
100 ␮m. In addition, the laser deposited sample that was annealed of the ␥-matrix calculated from XRD scan can indirectly indicates
at 1200 ◦ C shows annealing twin (Fig. 3c). The formation of twin in
the annealed sample indicates that the as-deposited sample con-
tains large amount of residual stress which is partially released by
forming twin grains during post-deposition annealing.
Microhardness measurement of the as-deposited nine (L1–L9)
samples revealed that with wide variation of laser processing
parameters, the hardness varies within a narrow window (254 ± 6
VHN). Since the cooling rate of the DMD process is very high (103
to 105 K/s), most of the strengthening elements, i.e., Nb and Mo
remain in the solid solution of ␥-Ni matrix. The little variation
of the hardness of the as-deposited sample is attributed to the
variation of supersaturation of these refractory elements in the ␥-
matrix. Although, In-625 was initially developed as a solid-solution
strengthened alloy, sooner it was found that the alloy is somewhat
age hardenable [2–5]. Fig. 7 shows the ageing behavior of laser
deposited In-625 that was annealed at 700, 800, 900, 1000, 1100
and 1200 ◦ C for 1 h. The sample annealed at 700 ◦ C shows some
increase of hardness due to the precipitation of metastable coher-
ent ␥ [Ni3 Nb] in the ␥-matrix. However, ␥ transforms to stable
incoherent ␦ [Ni3 Nb] phase above 700–750 ◦ C, depending upon
the aging time. The strengthening effect of ␦ phase is lower than
␥ phase due to the incoherent nature of ␦ phase in the ␥-matrix. Fig. 7. Variation of hardness with ageing temperature of laser deposited In-625.
104 G.P. Dinda et al. / Materials Science and Engineering A 509 (2009) 98–104
Fig. 8. The X-ray diffraction patterns of the laser deposited sample annealed at
different temperatures as shown in the figure.
Table 2
Variation in lattice parameter of the laser deposited In-625 with post-deposition
heat treatment.
Treatment Lattice parameter (pm)
As-deposited (AD) 360.01
AD + 800 ◦ C/1 h 360.22
AD + 1000 ◦ C/1 h 360.09
AD + 1100 ◦ C/1 h 360.34
the nature of precipitation. The lattice parameter of the ␥-Ni matrix
of the as-deposited and annealed sample was determined by XRD
and is shown in Table 2. In the as-deposited sample, it is expected
that all strengthening elements remain in solid solution of ␥ matrix
because of the very high cooling rate of the laser deposition process.
The thermal history of the as-deposited sample varies at different
locations because of the nature of layer-by-layer deposition pro-
cess. During deposition of subsequent layers, a heat affected zone
may form locally in the previously deposited layer. As a result some
intermetallic phases such as ␥ , ␥ and ␦ may be precipitated in
the previously deposited layer depending upon the temperature.
Although the precipitation kinetics of these intermetallic phase for-
mation is very slow, some precipitation is expected due to the very
high temperature near the melt pool and solid interface. This is
further supported by the lattice parameter of the as-deposited sam-
ple (360.01 pm) which is considerably lower than that of sample
annealed at 1100 ◦ C. The lattice parameter of the sample annealed
at 800 ◦ C was increased most probably due to the partial dissolution
of ␥ phase in the ␥-matrix. Above 800 ◦ C, the lattice parameter of
the matrix decreases because of the precipitation of order ␦ [Ni3 Nb]
phase in the matrix. During precipitation of ␦ phase, Nb comes out
from the matrix to form incoherent ␦ precipitate, and as a result
the lattice parameter of the matrix decreases. Above 1000 ◦ C, the ␦
precipitates again dissolve in the matrix and consequently increase
the lattice parameter of the matrix.
4. Conclusions
A series of In-625 samples have been successfully fabricated by
DMD process that is free from relevant defects like crack, bonding
error or porosity. The as-deposited microstructure mostly consisted
of columnar dendrites, which grew epitaxially from the substrate.
However, the growth direction of the columnar dendrites was tied
to the direction of laser scan. The columnar dendritic structure was
stable up to 1000 ◦ C (∼3/4 Tm ). Fully recrystallized equiaxed struc-
ture was visible at around 1200 ◦ C. The hardness of the deposit
was very high (254 ± 6) due to the very fine microstructure with
higher degree of supersaturation of strengthening elements. The
present results revealed that directionally solidified components
can be prepared or repaired by DMD when appropriate processing
strategy is followed, especially the laser scanning direction should
be identical for the deposition of all layers.
Acknowledgements
The authors would like to thank the Office of Naval Research for
financial support of this work (Grant No. N00014-07-C0282). Tech-
nical inputs from the Naval Undersea Warfare Center, Keyport WA
are gratefully acknowledged. The authors would also like to thank
their colleagues from CLAIM-University of Michigan and the Engi-
neering Department at Focus: HOPE, Detroit, MI for their support
in completing this work.
References
[1] Special Metals Corporation Products, INCONEL® alloy 625, www.specialmetals.
com/products.
[2] H. Bohm, K. Ehrlich, K.H. Kramer, Metall 24 (1970) 139–144.
[3] H.K. Kohl, K. Peng, J. Nucl. Mater. 101 (1981) 243–250.
[4] W.E. Quist, R. Taggart, D.G. Polonis, Metall. Trans. 2 (1971) 825–832.
[5] M. Sundararaman, P. Mukhopadhyay, S. Banerjee, Metall. Trans. A 19 (1988)
453–465.
[6] T. Charles, Int. J. Press. Vessels Piping 59 (1994) 41–49.
[7] V. Shankar, K. Bhanu Sankar Rao, S.L. Mannan, J. Nucl. Mater. 288 (2001)
222–232.
[8] L.E. Shoemaker, in: E.A. Loria (Ed.), Superalloys 718, 625, 706 and Various
Derivatives, TMS, Warrendale, PA, 2005, pp. 409–418.
[9] G.H. Gessinger, Powder Metallurgy of Superalloys, Butterworth & Co., London,
1984, pp. 3–15.
[10] J.J. Valencia, J. Spirko, R. Schmees, in: E.A. Loria (Ed.), Superalloys 718, 625, 706
and Various Derivates, TMS, Warrendale, PA, 1997, pp. 753–762.
[11] J.L. Koch, J. Mazumder, Proceedings of ICALEO, 1993, pp. 556–565.
[12] J. Mazumder, J.D. Dutta, N. Kikuchi, A. Ghosh, Opt. Lasers Eng. 34 (2000)
397–414.
[13] R. Mah, Adv. Mater. Process. 151 (1997) 31–33.
[14] G.K. Lewis, R. Nemec, J. Milewski, D.J. Thoma, D. Cremers, M. Barbe, ICALEO
(1994) 17–26.
[15] D.M. Keicher, W.D. Miller, J.E. Smugeresky, J.A. Romero, Proceedings of the TMS
Annual Meeting, San Antonio, TX, 1998, pp. 369–377.
[16] D.M. Keicher, W.D. Miller, Met. Powder Rep. 53 (1998) 26–126.
[17] W. Meiners, K. Wissenbach, R. Poprawe, Proceedings of ICALEO, 1998, pp. 31–37,
Section-E.
[18] M. Gaumann, S. Henry, F. Cleton, J.D. Wagniere, W. Kurz, Mater. Sci. Eng. A 271
(1999) 232–241.
[19] M. Gaumann, C. Bezencon, P. Canalis, W. Kurz, Acta Mater. 49 (2001) 1051–
1062.
[20] C.P. Paul, P. Ganesh, S.K. Mishra, P. Bhargava, J. Negi, A.K. Nath, Opt. Laser Technol.
39 (2007) 800–805.