Accepted Manuscript

Title: Hot Isostatic Pressing of IN718 Components

Manufactured by Selective Laser Melting

Author: W. Tillmann C. Schaak J. Nellesen M. Schaper M.E.

Aydinöz K.-P. Hoyer

PII: S2214-8604(16)30049-5
DOI: http://dx.doi.org/doi:10.1016/j.addma.2016.11.006
Reference: ADDMA 136

To appear in:

Received date: 30-3-2016

Revised date: 5-10-2016
Accepted date: 26-11-2016

Please cite this article as: {http://dx.doi.org/

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Hot Isostatic Pressing of IN718 Components Manufactured by
Selective Laser Melting
W. Tillmann1, C. Schaak1, J. Nellesen2, M. Schaper3,4, M. E. Aydinöz3,4, K.-P. Hoyer3,4

1 TU Dortmund, Institute of Materials Engineering (christopher.schaak@udo.edu), Dortmund,

Germany
2 RIF e.V., Institut für Forschung und Transfer, Dortmund, Germany

3 Paderborn University, Lehrstuhl für Werkstoffkunde (Materials Science), Paderborn,

Germany
4 Direct Manufacturing Research Center (DMRC), Paderborn, Germany

Abstract
Selective laser melting and other additive manufacturing (AM) techniques have recently
attracted substantial interest of both researchers and the processing industry. The freedom
of design leads to completely new possibilities for constructions and, thus, to entirely new
products. In the selective laser melting (SLM) process, the components are produced layer-
wise using a laser beam. SLM is a powder bed based AM process and is characterized by
the complete melting of the utilized powder material. Employing SLM, complex three-
dimensional parts and light weight structures can be produced directly from 3D CAD data.
However, although SLM is a very promising technology, there are still challenges to solve. In
the present study, a close look is taken at the porosity. Under cyclic loading, pores can act as
stress raisers and lead to premature crack initiations, which reduce the fatigue strength of the
material. Hot isostatic pressing (HIP) offers the possibility to reduce the porosity. HIP
combines high pressure and high temperature to produce materials with superior properties.
The influence of the HIP process parameters on the density and microstructure of IN718
SLM components is investigated by means of micro X-ray computed tomography and
scanning electron microscopy. The results of the experiments show that the majority of pores
can be densified by means of HIP. On the other hand, some pores cannot be densified. The
reason for this is seen in entrapped argon gas from the SLM process.

Keywords
Selective Laser Melting, Additive Manufacturing, Hot Isostatic Pressing, Inconel 718

1. Introduction
The generic term additive manufacturing (AM) is an umbrella term for different manufacturing
processes in which the components are built up layer-by-layer. Near-net-shape components
can be manufactured directly from 3D CAD data and do not require further forming or
process planning. So-called powder bed fusion or powder bed laser fabrication processes,
such as selective laser melting (SLM), belong to the group of AM processes. In the SLM
process, a powder layer (= powder bed) is selectively exposed to a high power laser beam
(see Fig.1). A scanner system moves the focus point of the laser according to the contour of
the desired part. The powder is locally full melted, and the powder particles are bonded to
each other. After this procedure, the build platform is lowered, and a new powder layer is
applied with a recoater system (see Fig. 1). Subsequently, the new powder layer is
selectively exposed to the laser. This procedure is repeated until the complete height of the
part is obtained. The entire process takes place in an argon-filled chamber to avoid an

1
oxygen and nitrogen contamination of the processed metal. The main advantages of SLM
are the wide range of possible geometries (even with undercuts), the high flexibility of
metallic feedstock powders, and the realization of inner structures with additional functions.
Commonly used SLM powder materials are Ni-based superalloys, Fe-alloys, and Ti-alloys [1-
4].

Inconel 718 / IN718 (Special Metals Corporation, Huntington, USA) is one Ni-based
superalloy commonly employed for high temperature applications [5]. Its superior mechanical
properties, such as the high temperature strength and outstanding creep behavior, are
mainly based on solid solution strengthening and precipitation strengthening [6, 7]. SLM
offers new possibilities to manufacture parts from IN718, especially in the aerospace
industry, such as internal cooling structures, light weight structures or integrated sensors,
which are not featured by conventional production methods (e.g. forming and machining).
Thus, researchers, as well as the manufacturing industry are highly interested in this process
[14]. IN718 powder can be successfully processed by SLM. Several researchers and
institutions have already started to analyze the SLM building process with an IN718 powder
feedstock and to investigate the mechanical properties of IN718 components made by SLM
[8-13]. In general, the properties of components made by means of SLM are mainly
determined by the parameters of the SLM process. For IN718, different columnar dendritic
structures and different porosity values were observed, depending on the laser energy
density [9, 13]. High laser energy density leads to slender and uniformly distributed columnar
dendrites and a low porosity [9]. So far, even with optimized parameters, SLM still faces
challenges that need to be solved with regard to the quality of parts. One critical point is the
poor surface quality immediately after manufacturing by means of SLM [15, 16]. The surface
roughness of components made by SLM is typically very high and depends on the SLM
process characteristics. It is commonly known that a high surface roughness influences the
fatigue strength negatively. Another main disadvantage of the SLM process is that
completely dense parts cannot be obtained with this process [8, 9, 14]. Especially under
cyclic loading, the remaining pores act as stress raisers and can lead to premature crack
initiation, eventually reducing the fatigue strength of the component.
To improve the static and dynamic mechanical properties of SLM-parts, different post-
processing methods moved into the focus of numeral investigations. HIP is increasingly
attracting attention as a post-processing technique for SLM. HIP is an established technique,
which is used for master forming, diffusion welding, compaction, and of course traditional
sintering. High pressures and temperatures lead to diffusion and plastic deformation. The
main parameters are temperature, pressure, and time [17-19]. Most literature related to
IN718, combined with HIP processing is about the master forming of components by means
of HIP and does not deal with the post-densification of SLM parts (e.g. [20,21]). Several
studies reveal that the internal porosity of laser consolidated, laser deposited, and SLM
processed parts made out of different powder materials, is strongly reduced by means of HIP
and hence the mechanical properties are enhanced [22-28]. In the experiments of Leuders et
al. [27] the fatigue strength (high-cycle fatigue) of Ti6Al4V fatigue samples, produced by SLM
and compacted by HIP, was clearly improved. This effect was attributed to the reduction of
the pore size and overall porosity, respectively. Some previous studies [23, 25] already
confirmed the positive effect of HIP post-processing for pure titanium samples. Likewise,
CoCr alloy powder, (ASTM F75) processed by SLM and densified with HIP, showed a higher
fatigue strength (high-cycle fatigue) compared to the as-built condition and the as-cast
material [29]. The closure of all remaining pores and a complete recrystallization of the

2
microstructure were identified as causes for this effect in [29]. Qiu et al. [28] showed for
Ti6Al4V alloys that in addition to the lower porosity, the quasi-static mechanical properties
were changed due to heat treatment during the HIP process as well. In the as-built state, the
tensile strength is high and the ductility is low due to fine martensitic needles in a typical SLM
microstructure. After the HIP process, the tensile strength is reduced and the ductility is
enhanced [28]. There exists a detailed study concerning the post-processing of IN718 made
by means of selective electron beam melting (SEBM) [30] as well as a brief investigation of
the HIP post-processing of IN718 made by means of direct laser deposition (DLD) [31]. The
HIP post-processing of IN718 DLD samples reduced the anisotropy and improved the
interlayer adhesion [31]. For SEBM microstructural findings were reported and a grain growth
after HIP processing was recognized [30]. The investigation of the hardness after HIP
showed that the HIP processing reduced the hardness when compared to its state directly
after SEBM manufacturing [30]. The porosity before and after HIP processing was not
quantified and discussed in both studies.
A fundamental problem not often taken into account is that the SLM process chamber is filled
with an inert gas (e.g. argon) during the building processing and it is possible that the
protective gas is encapsulated in cavities. Consequently, the complete closure of these pores
would be hard to achieve by means of HIP. Furthermore, for an effective compression, the
component should have a closed porosity to avoid pressure equalization inside the pores,
which would prevent the pores from closing [17-19]. In our own studies [8], the sporadic
presence of small pores which were open to the surface was shown in IN718 components,
made by means of SLM. Also, it was shown that these pores could not be compacted by
means of HIP. Hence, the sample must be encapsulated [8].
Overall, two important effects were identified that are responsible for the mechanical
properties after post-processing the SLM components by HIP. The first is the reduction of
pores. The second is the carefully controlled heat treatment.

2. Material and Methods

For the conducted densification experiments, a simple geometric shape was chosen for the
specimens. Flat rectangular specimens (26 mm x 7 mm x 1.5 mm) were processed vertically,
using a SLM 250HL machine (SLM Solutions GmbH, Germany) equipped with a 400 W fiber
laser and SLM AutoFab software (Marcam Engineering GmbH, Germany). Inconel 718
powder was employed with an average particle size of approximately 30 µm. Figure 2 shows
a SEM image of the IN718 powder used and the particle size distribution. The particle size
distribution was measured with a Mastersizer 2000 (Malvern Instruments Ltd, United
Kingdom) system. This system uses laser diffraction to measure the size of particles.

Table 1 shows the composition of the IN718 samples used in the experiments. The
composition was determined by X-ray fluorescence spectroscopy. This analysis was
conducted at the Revierlabor (Chemische Laboratorien für Industrie und Umwelt GmbH,
Essen, Germany). The results show that the composition listed in Table 1 is within the
standard specification. The applied and optimized SLM process parameters are summarized
in Table 2. The parameters were optimized to achieve a low porosity. Furthermore, fatigue
experiments certified a good fatigue behavior of the samples produced with these
parameters [32]. The skin-core strategy was used to produce the samples. Hence, the
process parameters of the contour and the volume area differ from each other (Tab. 2). For
the better connection between the skin and core area, beam offset (compensation) for the
3
contour was used. Without this, large pores were found in this area. This fact would lead to a
significant higher porosity.

For densification experiments, a HIP facility featuring a maximum temperature of 2000 °C

and a maximum gas pressure of 200 MPa was used (QIH-3, Asea/Avure Technologies,
Franklin (TN), USA). Argon (ALPHAGAZ 1 ARGON, Air Liquide GmbH, Germany) was
utilized as an inert process gas. This gas is normally utilized for chemical analysis devices
and shows a high purity (purity ≥ 99,995 %). Hence, a contamination with oxygen and other
gases is avoided. First, the influence of the HIP process parameters was investigated; the
experimental design is summarized in Table 3. The holding time was kept constant (120 min)
whereas the temperature and pressure was varied (Tab. 3). The temperatures ranged from
500 °C to 1300 °C and the pressures ranged from 500 MPa to 150 MPa. The parameters
were varied over a width range. The parameter sets with a HIP temperature of 1300 °C is
above the solidus temperature of IN718. The solidification temperature during heating ranges
between 1320 °C and 1240 °C depending on the literature [21]. In most of the literature, the
solidification temperature is slightly above 1300 °C. At this temperature, a liquid phase
occurs that could support the post-densification during HIP processing (compare liquid phase
sintering). The porosity was measured after HIP processing, and the values were correlated
with the process temperature and pressure. Figure 3 shows the HIP profile of the conducted
experiments. Before starting the HIP process, the pressure chamber was evacuated, flooded
with argon, and then evacuated again. Subsequently, the pressure chamber was heated until
the set temperature was reached. When the temperature has reached the target value, the
pressure was increased fast (5 MPa per min). The holding time was initiated when both
values were stable. After the holding time, the samples freely cooled down in the furnace
under pressure. The pressure decreases due to the cooling.

The porosity was investigated with a micro X-ray computed tomography (μCT) in the as-built
state and after densification by means of HIP. The μCT was conducted with a commercial
tomography scanner phoenix v|tome|x | 240 (GE Sensing & Inspection Technologies GmbH,
Wunstorf, Germany). Tomograms were reconstructed from the X-ray projections, using
software which employs the Feldkamp algorithm. The obtained voxel edge length was 8 μm.
The μCT technique was used to determine the overall as well as the local porosity of the
SLM specimens. Compared to other methods which calculate the porosity from a
metallographic cross-section, μCT allows analysis of the porosity of a large volume in 3D and
thus, has a higher accuracy. Furthermore, the preferred orientations and agglomerations of
pores can be identified by means of μCT. For comparison, the porosity was also measured
by a reflected-light optical microscope (Axiophot; Carl Zeiss GmbH, Jena, Germany) on the
basis of metallographic cross-sections. The analysis software (Axiovision 4.6, Carl Zeiss
GmbH, Jena, Germany) uses a grayscale differentiation to identify and to count pores. To
check if the pores contain entrapped argon, an argon-analysis was conducted by Bodycote
HIP GmbH (Haag-Winden, Germany). For the measurement, a specific sample mass was
evaporated and analyzed by means of a gas-phase chromatograph type GC 600 (AGC
Instruments Ltd, Ireland).

4
Investigations of the microstructure were accomplished by means of scanning electron
microscopy (SEM, JEOL JSM-7001F; Joel GmbH, Eching, Germany). Some samples were
polished and etched with the Beraha III (color etching) and Kalling 2 (micro etching) etching
method to obtain a better look at grains and precipitation after the building process and after
post-densification [33]. The indentation hardness was measured for the as-built state and
after post-densification with the Vickers hardness tester M-400 (Leco Instrumente GmbH,
Germany). A number of samples were solution-annealed at 1000 °C for 1 h. This process
was followed by a slow cooling in the furnace. Other samples were precipitation heat treated
as follows: the samples were first solution annealed and subsequently held at 720 °C for 8 h,
before they were cooled down with 50 °C/h to 621 °C, and held at 621 °C for 8 h followed by
air cooling. The heat treated samples were used to measure the hardness and to compare
these hardness values with the hardness values after HIP processing.

3. Results and Discussion

3.1 Microstructure in the as-built state

Directly after manufacturing, and without post-treatment, the IN718 components made by
means of SLM show a microstructure typical for this manufacturing technique. The as-built
microstructure is strongly affected by the SLM processing parameters [1, 4]. Figure 4 shows
a light microscope image of the cross-section of an IN718 sample. The shown cross-section
is perpendicular to the building direction. With low magnification the material seems to be
homogenous with a high density, however, some sporadic pores and defects occur. The
sample surface is very rough with sporadic stuck, un-melted powder particles. This effect is
typical for powder bed fusion processes [34]. In this state, the average hardness is
273.2 ± 11.9 HV0.1.

First, the investigation of etched cross- and longitudinal-sections reveals the characteristic
microstructure of SLM processed parts (Fig. 5). In the etched state, the solidified melt pools
and scan tracks are visible. Hence, the scanning strategy can be read from the images. In
this case, the layers were melted using an alternating bi-directional strategy (Fig. 5a). A high
density can be guaranteed with this building strategy [34]. Figure 5b shows that the used
SLM process parameters lead to a partially re-melting depth greater than the layer thickness
of one powder layer (≈ 30 µm), so a good metallurgical bonding is guaranteed between
vicinal layers. Furthermore, the scan tracks show a sufficient overlapping (Fig. 5a). This
sufficient overlapping leads to a gap free macro-structure.
The grain boundaries become visible after etching with the Kalling 2 method, so the grain
structure can be recognized (Fig. 6). The grains are larger than the melt pools and surpass
the melt pool boundaries. The majority of grains are elongated in the building direction, but in
areas close to the surface, the grains are elongated perpendicularly. This effect can be
observed in the cross-section and the longitudinal-section (Fig. 6) and is due to the heat flux
direction. During the SLM process, the heat flux mainly proceeds through the powder bed
and the building platform.

At higher magnification and with backscattered electron contrast (BSE contrast), the
anisotropic microstructure can be seen. The microstructure differs, depending on the
5
sectional direction. The cross-section image (Fig. 7) shows mainly fine columnar and cellular
dendritic structures with different scales. However, the longitudinal-section image (Fig. 7)
reveals primarily structures which have a preferentially elongated shape and some dendrites
that traverse several layers. These structures are preferentially elongated along the building
direction. Consequently, the microstructure depends on the building direction. This typical
microstructure is formed by the high cooling rates and fast solidification of the melt pool
during the SLM process and the repeated heat flux through the already solidified material.
The micro-structural and macro-structural results correspond with the findings reported in
literature for Inconel alloys, processed by means of SLM [12, 35].

As mentioned before, despite the optimized parameters, imperfections were found in the
produced components. These flaws reduce the relative density of the produced SLM part
and could further influence the static and dynamic mechanical properties. Figure 8 shows an
example of two types of defects which can be found in the sample body. Some defects have
a round, spherical shape (Fig. 8a). Other defects are shapeless and look like cracks (Fig.
8b). These voids can have different causes: One reason may be an insufficient or
inhomogeneous melting of the powder feedstock or failures in the re-coated powder layer
itself which are due to the stochastic nature of the powder bed [36]. Furthermore, the surface
tension of the melted metal material leads to a contraction of the melt pool, which results in a
fluctuation of the width of the scan track and hence leads to gaps between the scan tracks
[36]. Also, pores and defects can occur due to contraction and shrinkage whilst cooling. This
effect is comparable to the cooling process during conventional casting processes.
Investigations of the powder feedstock showed that pores are already in the powder particles
(Fig. 9). Different sized spherical pores were identified. In some extreme cases (Fig. 9b) up
to 38.6 vol.-% of the particle is hollow. Furthermore, Gerling et al. [37] verified the existence
of pores in gas atomized TiAl powders. They proved that these pores were filled with argon
from the gas atomizing process and influence the powder metallurgical processing (HIP) [37].
It is conceivable that the pores in the feedstock powder for SLM also can cause
imperfections in the final SLM part. Due to the rapid melting (heating) and solidification
(cooling) during SLM, the entrapped argon has no time to outgas and hence, consequently,
remains in the part. This theory is supported by the findings of Sames et al. [38].

Despite the existing discontinuities, the conducted density measurement showed a very low
porosity and hence, a good building result of IN718 parts. The relative density is higher than
99.8% (Tab. 4). It seems that the cavities are randomly distributed over the cross-section as
in Figure 4. Also, it seems that there are no preferred directions or accumulations of pores
(Fig. 4). In contrast, the µCT tomograms reveal that most cavities in the samples can be
found in the boundary region, between the shell and core region (Fig. 10, transitional area).
But the number of cavities is very low and their size is small. These defects could be caused
by the skin-core strategy that was used. Because of the successively melting of skin and
core layer, the skin area is already cooled and solidified when the core layer was melted.
Due to different thermal gradients, resulting in different shrinkages, a pore formation in this
area is most probable.

6
The center area of the sample shows a smaller amount of pores (Fig. 10, center area). The
areas close to the surface are free of pores (Fig. 10, near to surface area). It is noteworthy
that this pore distribution cannot be gathered from a few of metallographic cross-sections.
The detected pores generally have a size of a few micrometers. The largest recorded defect
has a radius of about 26.8 µm compared to the smallest measured pore, which has a radius
of 2.13 µm. The average pore radius is approximately 9.83 ± 7.2µm. These values were
calculated from light microscope image as shown in Figure 4. In Table 4, the measured
overall 2D porosity values from light microscopy investigations are compiled and compared
to 3D porosity values calculated from 3D tomograms.

The illustrated results of the porosity measurement (Tab. 4) show a significant difference
between the measurement methods that were used. The overall porosity calculated from
light microscopic images is higher than the one measured by µCT (Tab. 4). The relative
density is about 99.87 %-99.91 % calculated by light microscope and 99.917 % calculated by
µCT.
The benefit of µCT is a non-destructive investigation of the entire volume, rather than of a cut
view as performed with a conventional light microscope. The resolution of the µCT
tomograms is given approximately by the voxel edge length, which amounts in the conducted
investigations to 8 µm. This fact means that a defect must have a minimum diameter of 8 µm
or larger to be detected by means of µCT. Hence, it is an obvious disadvantage of µCT that
pores smaller than 8 µm cannot be detected. Yet, the investigations showed that pores below
8 µm exist. These pores were not considered in the µCT measurement and hence, the
porosity value is lower (Tab. 4). Overall, the porosity determined by light microscopy seems
to be more reliable. A disadvantage of this method is the large effort for the metallographic
preparation and evaluation, especially for a large number of samples.
Since the SLM process takes place under an argon atmosphere, cavities (like in Fig. 8, 9)
can be filled with process gas. Entrapped gas could hinder a perfect post-densification by
means of HIP. Furthermore, entrapped gas can lead to internal stresses due to thermal
expansion in high temperature applications. Thus, only gas-free shrinkage cavities could be
perfectly closed by HIP. Argon-analyses were conducted to obtain more information about
the gas content of the present cavities [39]. The results of the Ar-measurements show a
slightly increased average Ar-content of 0.3 ppm in the IN718 samples produced by SLM.
For comparison, conventionally manufactured C45 samples show a lower average Ar-
content of 0.049 ppm. Both values are clearly below the allowed limit value of 2 ppm for hot
isostatic pressed parts. Nevertheless, it could be verified that Ar is entrapped in the pores.
Reasons could be the Ar-atmosphere during the SLM process or Ar-inclusions in the
feedstock powder. First, the measured value seems to be very low, but even low Ar-contents
entrapped in the structure can have a significant influence on the post-treatment and
mechanical properties [40]. Under high temperature conditions, as during HIP processing,
machining or in different applications, the entrapped Ar expands (gas laws) and causes
stresses that can lead to elastic or plastic deformations. In the case of HIP processing the
direction of these stresses is directed towards the stresses caused by outer pressures. The
pressure during HIP processing should support the sintering process, respectively the
densification. However, in this case the internal stresses prevent a perfect densification.
Furthermore, in applications under cycling loading and high temperature, like in gas turbines
(turbine blades), the stresses could act as a crack initiator and hence, reduce the life time of
the component.

7
3.2 Microstructure after Post-Densification by HIP

As mentioned, densification experiments were conducted to gain information about the

influence of the HIP parameters. The holding time was kept constant (120 min) in the
experiments while the temperature and pressure were varied (see Tab. 3). After HIP
processing, the porosity was measured by means of µCT and linked to the process
parameters. Figure 11 shows the results of the analysis.

It can be seen that the porosity is not significantly reduced for temperatures lower than
≈ 900 °C (Fig. 11). Below ≈ 900 °C, the porosity can mainly be reduced by increasing the
pressure. Hence, in a temperature range below 900 °C, the pressure is the main influencing
factor. This effect is similar to the densification effects during cold isostatic pressing. A higher
densification first occurs above 900 °C. The solvus temperatures of the phases (γ’’-phase:
T = 940 °C; δ-phase: T = 1025 °C) in IN718 are within this temperature range and
recrystallization occurs [41]. The solid solution strengthening and precipitation strengthening
phases are more or less dissolved. Therefore, the resistance against deformations is
decreased. At this point, the thermally induced diffusion processes are clearly supported by
plastic deformations due to the gas pressure. The tendency that a higher gas pressure
improves the densification is still visible at higher temperatures (Fig. 11). The worst porosity
values were measured for densifications with a pressure of 50 MPa. A pressure of 100 MPa
and 150 MPa lead to similar densification results at temperatures above 900 °C. The porosity
values reach almost stable values above a temperature of ≈ 1150 °C for 100 MPa and 150
MPa. In this case, the porosity is between 0.011 and 0.015 vol.-%. Thus, the relative density
of IN718 SLM components is about 99.985 %─99.989 % after HIP processing. A HIP
temperature slightly above the solidus temperature of IN 718 (THIP = 1300 °C) did not further
improve the density. These findings match the results in literature, which showed that the
porosity of SLM components can be reduced by means of HIP [22-28].
Based on the results in Figure 11 and taking economic considerations into account, optimal
HIP parameters were selected. As an optimal solution, the parameter set 1150 °C and
100 MPa was chosen. The results showed that a pressure of 100 MPa is adequate for
densification. Higher pressure (150 MPa) does not lead to clear advantages. Additionally,
above 1150 °C the improvement of the porosity is low. For a further improvement of the
porosity, the holding time was extended to 4 hours. Post-densification experiments with
IN718 SLM components were conducted with the chosen parameters and showed promising
results. In Figure 12, a SEM image of a cross-section after HIP processing is depicted.

After HIP processing with extended time (4 h), the porosity is nearly unverifiable. The
grayscale differentiation measuring technique by means of light microscope calculated zero
percent porosity. As shown in Figure 12, the cross-section is free from imperfections and the
inner structure is very homogenous. Some pores are sporadically recognizable only by
means of µCT, since the entire volume is analyzed. In Figure 13, a 2D slice extracted from a
3D tomogram is rendered in which such a pore can be spotted. The average porosity was
calculated as 0.0019 ±0.001 %. The porosity is extremely low, but it must be noted that there
are indeed pores in the component.

8
Next to the fact that open porosity hinders a perfect densification [8], the conducted results
reveal a second problem for post-densification of SLM parts by HIP processing. It is
supposed that the detected entrapped argon hinders a perfect densification. During the HIP
process, the heat treatment leads to the expansion of the entrapped argon. Hence, the
pressure in the pore increases. This inner pressure counteracts the densification by HIP
processing. This fact evidences that a full density (100%) cannot be achieved.
After HIP processing with optimized parameters (1150 °C / 100 MPa / 4 h), the
microstructure was investigated. As expected, the investigation showed that the
microstructure is affected by the heat treatment during the post-densification process. In
Figure 14 light microscopic images of an etched cross-section and longitudinal-section of
one IN718 SLM sample are visualized. The typical SLM microstructure (Fig. 5 ─ Fig. 6)
cannot be observed after post-densification. After HIP processing, the grains are larger than
in the as-built state. The preferential orientation of the grains in the building direction is lost
after post-densification (Fig. 14). Hence, it can be concluded that the samples show better
isotropic material characteristics in this condition. However, the yield strength is reduced and
the tensile elongation and the fracture stress increase with higher HIP temperatures due to
the larger grains [20]. This correlation was detected by Chang et. al [20] for IN 718 powder
compacted by means of HIP. From THIP = 1140 °C to THIP = 1260 °C the fracture stress
increased about 20,6 % [20].
After post-densification, the average Vickers hardness value of the IN718 samples is
258.8 ± 8.2 HV0.1. Thus, the hardness is almost similar to the solution annealed state. The
samples in the as-built state exhibit a slightly higher hardness value (Tab. 5). In comparison,
the Vickers hardness constitutes 482.7 ± 21.4 HV0.1 after precipitation hardening. Detailed
investigations of the microstructure by means of electron backscatter diffraction (EBSD) and
transmission electron microscopy (TEM) after different states (after SLM processing, after
HIP processing, after aging and so on) can be found in [32]. In this study, it was shown that
HIP processing leads to recrystallization. Hence, the typical SLM structures are eliminated.
Furthermore, fatigue tests at room temperature (RT) showed that HIP processing leads to a
decrease of fatigue life. On the other hand, HIP processing increases ductility under quasi-
static loading at RT. These findings correspond with the measured hardness values for
different states (Tab. 5). Overall, the heat treatment and cooling during HIP processing do
not lead to an age hardening of the samples.

4. Summary and Conclusion

In the present study, the influence of the HIP process parameters on the porosity of IN718
SLM components was investigated by means of micro X-ray computed tomography and light
microscopy. The microstructure was compared before (as-built state) and after HIP
densification. The main findings are summarized in the following points:

- The IN718 samples showed a microstructure, which characteristically depends on the

SLM process. The relative density is high (about 99.917 %). Additionally, some
sporadic cavities occur. Most cavities can be found in the boundary section between
the shell and the core region. Argon measurements showed an Ar-content of
0.3 ppm. It was concluded that the cavities are in parts filled with argon. The argon
could be entrapped during the SLM process or embedded in the powder particles.

9
 - For post-densification by means of HIP, the process temperature has a greater
influence than the process pressure (the holding time is kept constant). For IN718
processed with SLM, the temperature should be higher than 1150 °C and the
pressure should be above 100 MPa to obtain a significant reduction of cavities. The
relative density can be improved to 99.985 % - 99.989 % by means of HIP.

- Even after HIP processing with an enhanced holding time (4 h), some sporadic
cavities are still detectable. The relative density slightly increased to 99.998%. It was
concluded that even with increased HIP parameters, 100 % dense parts cannot be
achieved. Entrapped argon is expected as a cause for an imperfect densification. The
argon counteracts the pressure applied during HIP processing.

- After HIP processing, the characteristic SLM microstructure disappears. The grain
size is enlarged due to the heat-treatment which occurs during HIP processing.

Overall, it was shown that the HIP process improves the density of IN718 parts made by
means of SLM. Also, it seems that it is impossible to obtain a 100 % density by means of HIP
because small amounts of entrapped argon gas hinder the densification. Hence, additive
manufacturing processes conducted under vacuum should show clear benefits. In this case,
no process gas would be entrapped.

Acknowledgment

The financial support by the German Research Foundation (DFG) (Ti343/70-1 and Ni1327/4-
1) is gratefully acknowledged.

References

[1] I. Gibson, D.W. Rosen, B. Stucker, Additive manufacturing technologies. Rapid

prototyping to direct digital manufacturing, New York, 2010.
[2] A. Gebhardt, Generative Fertigungsverfahren: Rapid prototyping - rapid tooling - rapid
manufacturing, 3rd ed., Hanser, München, 2007.
[3] VDI 3404: Additive fabrication - Rapid technologies (rapid prototyping) -
Fundamentals, terms and definitions, quality parameters, supply agreements, 2009.
[4] K. Osakada, M. Shiomi, Flexible manufacturing of metallic products by selective laser
melting of powder, Int. J. Mach. Tool Manu. 46 (2006) 1188-1193.
[5] J. Warren, D.Y. Wei, The cyclic fatigue behavior of direct age IN 718 at 149, 315, 454
and 538 °C, Mat. Sci. Eng. A 428 (2006) 1–2 106–15.
[6] G. Korth, Creep rupture of alloy 718 after thermal aging, JOM 52 (2000) 1 40-42.
[7] M. Miller, S. Babu, M. Burke: Comparison of the phase compositions in Alloy 718
measured by atom probe tomography and predicted by thermodynamic calculations,
Mat. Sci. Eng. (202) 84–88.
[8] W. Tillmann, C. Schaak, J. Nellesen, M. Schaper, M.E. Aydinöz, T. Niendorf,
Functional Encapsulation of Laser Melted Inconel 718 by Arc-PVD and HVOF for Post
Compacting by Hot Isostatic Pressing, Powder Metall. 58 (2015) 4 259-64.
[9] Q. Jia, D. Gu, Selective laser melting additive manufacturing of Inconel 718
superalloy parts: Densification, microstructure and properties, J. Alloy Compd. 585
(2014) 0 713-21.

10
[10] F. Liu, X. Lin, G. Yang, M. Song, J. Chen, W. Huang, Microstructure and residual
stress of laser rapid formed Inconel 718 nickel-base superalloy, Opt. Laser. Technol.
43 (2011) 1 208–13.
[11] Q, Jia, D., Gu, Selective laser melting additive manufactured Inconel 718 superalloy
parts: High-temperature oxidation property and its mechanisms, Opt. Laser. Technol.
62 (2014) 0 161–71.
[12] Z. Wang, K. Guan, M. Gao, X. Li, X, Chen, X. Zeng, The microstructure and
mechanical properties of deposited-IN718 by selective laser melting. J. Alloy Compd.
513 (2012) 0 518-23.
[13] K.N. Amato, S.M. Gaytan, L.E. Murr, E. Martinez, P.W. Shindo, J. Hernandez, S.
Collins, F. Medina, Microstructures and mechanical behavior of Inconel 718 fabricated
by selective laser melting, Acta Mater. 60 (2012) 5 2229-39.
[14] A. Angrish, A critical analysis of additive manufacturing technologies for aerospace
applications, Aerospace Conference 2014 IEEE 2014.
[15] L.W. Lherbier, D.J. Novotnak, The Pluses and Minuses of Additive Manufacturing,
Proceedings Euro PM2014 (2014).
[16] G. Strano, L. Hao, R.M. Everson, K.E. Evans, Surface roughness analysis, modelling
and prediction in selective laser melting, J. Mater. Process. Tech. 213 (2013) 4, 589-
597.
[17] H.-D. Steffens, J. Lebküchner-Neugebauer, B. Wielage, Hot Isostatic Pressing.
Heißisostatisches Pressen. Mat.-wiss. u. Werkstofftech. 21 (1990) 28-31.
[18] E.K.H. Li, P.D. Funkenbusch, Hot isostatic pressing (HIP) of powder mixtures and
composites: Packing, densification, and microstructural effects, Metall. Mater. Trans.
A 24 (1993) 6 1345-1354.
[19] H.V. Atkinson, S. Davies, Fundamental aspects of hot isostatic pressing: An overview,
Metall. Mater. Trans. A 31 (2000) 12 2981-3000.
[20] L. Chang, W. Sun, Y. Cui, R. Yang, Influences of hot-isostatic-pressing temperature
on microstructure, tensile properties and tensile fracture mode of Inconel 718 powder
compact, Mat. Sci. Eng. A 599 (2014) 186–95.
[21] T. Antonsson, H. Fredriksson, The effect of cooling rate on the solidification of
INCONEL 718. In: Metall. Mater. Trans. B 36 (2005) 1, 85-96.
[22] P.L. Blackwell, The mechanical and microstructural characteristics of laser-deposited
IN718, J. Mater. Process. Tech. 170 (2005) 1-2 240-246.
[23] K. Osakada, M. Shiomi, Flexible manufacturing of metallic products by selective laser
melting of powder, Int. J. Mach. Tool Manu. 46 (2006) 11, 1188-93.
[24] S. Dadbakhsh, L. Hao, Effect of hot isostatic pressing (HIP) on Al composite parts
made from laser consolidated Al/Fe2O3 powder mixtures, J. Mater. Process. Tech.
212 (2012) 11 2474-83.
[25] F. Abe, E.C. Santos, Y. Kitamura, K. Osakada, M. Shiomi, Influence of forming
conditions on the titanium model in rapid prototyping with the selective laser melting
process, J. Mech. Eng. Sci. 217C (2003) 119-126.
[26] J.L. LaGoy, (Bodycote HIP N.A.): A Brief Explanation of Hot Isostatic Pressing (HIP)
as it Pertains to AM, Metallic Materials for Additive Manufacturing – CIMP3D,
Pennsylvania State University, USA, 2013.
[27] S. Leuders, M. Thöne, A. Riemer, T. Niendorf, T. Tröster, H.A. Richard, H.J. Maier,
On the mechanical behaviour of titanium alloy TiAl6V4 manufactured by selective
laser melting: Fatigue resistance and crack growth performance, Int. J. Fatigue 48
(2013) 300-07.

11
[28] C. Qiu, J.E. Nicholas, M.M. Attallah, Microstructure and tensile properties of
selectively laser-melted and of HIPed laser-melted Ti–6Al–4V, Mat. Sci. Eng. A 578
(2013) 0 230-239.
[29] J. Haan, M. Asseln, M. Zivcec, J. Eschweiler, K. Rademacher, C. Broeckmann, Effect
of Subsequent Hot Isostatic Pressing on Mechanical Properties of ASTM F75 Alloy
Produced by Selective Laser Melting, Proceedings Euro PM2014, 2014, Salzburg,
Austria.
[30] J.W. Sames, Additive Manufacturing of Inconel 718 using electron beam
melting: processing, post-processing, mechanical properties, Dissertation 2015,
Texas A&M University.
[31] P. L. Blackwell, The mechanical and microstructural characteristics of laser-deposited
IN718, J. Mater. Process. Tech. 170 (2005) 1–2, 240-246.
[32] M. E. Aydinöz, F. Brenne, M. Schaper, C. Schaak, W. Tillmann, J. Nellesen, T.
Niendorf, On the microstructural and mechanical properties of post-treated additively
manufactured Inconel 718 superalloy under quasi-static and cyclic loading. In:
Materials Science and Engineering A 669 (2016), 246-258.
[33] G. Petzow, Metallographisches, keramographisches, plastographisches Ätzen,
Nachdr. der 6., vollst. überarb. Aufl. Berlin 2006.
[34] J.P. Kruth, M. Badrossamay, E. Yasa, J. Deckers, L. Thijs, J. van Humbeeck, Part
and material properties in selective laser melting of metals, Proceedings of the 16th
International Symposium on Electromachining, 2010 Shanghai, China.
[35] P. Kanagarajah, F. Brenne, T. Niendorf, H.J. Maier, Inconel 939 processed by
selective laser melting: Effect of microstructure and temperature on the mechanical
properties under static and cyclic loading. Mat. Sci. Eng. A 588 (2013) 188-195.
[36] S.A. Khairallah, A. Anderson, Mesoscopic simulation model of selective laser melting
of stainless steel powder, J. Mater Process. Tech. 214 (2014) 11 2627-36.
[37] R. Gerling, R. Leitgeb, F. Schimansky, Porosity and argon concentration in gas
atomized TiAl powder and hot isostatically pressed compacts. Mat. Sci. Eng. A 252
(1998) 2, 239-47.
[38] W. J. Sames, F. Medina, W. H. Peter, S. S. Babu, and R. R. Dehoff, Effect of Process
Control and Powder Quality on Inconel 718 Produced Using Electron Beam Melting,
Proceedings 8th Superalloy 718 and Derivatives, Pittsburgh, PA, 2014.
[39] W. Tillmann, C. Schaak, M. Schaper, M.E. Aydinöz, Process Gas Infiltration in Inconel
718 samples during SLM Processing, RTeJournal – Fachforum für Rapid
Technologie, Vol. 2016, Iss.1.
[40] B. Hofer, K.-U. Volker, F. Bigolin, The effect of Argon on HIP parts, permissible values
and concentration determination, Proceedings Euro PM2012, 2012, Basel,
Switzerland.
[41] M.S. Lewandowski, R.A. Overfelt, High temperature deformation behavior of solid and
semi-solid alloy 718, Acta Mat. 47 (1999) 18 4695-4710.

12
Figure 1: SLM process sketch (source: DMRC)

Figure 2: Particle size distribution and SEM-image of the IN718 powder.

Figure 3: HIP profile for the conducted experiments. First, the temperature was increased to
the target temperature before the pressure was increased to the target temperature.

13
Figure 4: Light microscope image of a polished sample cross-section (perpendicular to the
building direction) with low magnification directly after production (as-built state).

Figure 5: Light microscope images of etched (Kalling 2) SLM samples directly after
production. The solidified melt pools are clearly visible.

Figure 6: Light microscope images of etched (Beraha III) SLM samples directly after
production. The grain structure is clearly visible and elongated in the building direction.

14
Figure 7: SEM images (BSE) of the microstructure with higher magnification.

Figure 8: SEM images of a single located pore and a bonding defect in one SLM sample (as-
built condition), recompiled from [39].

Figure 9: SEM images of powder particles embedded in resin and polished (cross-section).
The particles already exhibit pores.

15
Figure 10: 2D sectional view (µCT image) parallel to the building direction at different depths
after manufacturing. The black dots are cavities.

16
Figure 11: Porosity measured after HIP processing of IN718 (SLM) with different
temperatures and pressures. The temperature field of solidification is based on [37].

Figure 12: SEM image of the cross-section of an IN718 sample after post-densification with
HIP (1150 °C / 100 MPa / 4 h).

17
Figure 13: µCT image after post-densification with HIP (1150 °C / 100 MPa / 4 h). The cycle
marks a single cavity in the volume.

Figure 14: Light microscope images of etched (Kalling 2) SLM samples after post-
densification (1150 °C / 100 MPa / 4 h). A homogenous grain structure is clearly visible.

18
Table 1: Composition of the IN718 samples used in the experiment.
chem.
element Mn Si Ni Cr Cu Fe Mo V Ti Co Al Nb Ta C
Wt.% 0.04 0.12 balance 18.9 0.04 18.9 3.07 0.03 0.97 0.09 0.58 5.06 <0.03 0.03

Table 2: SLM process parameters used in this study for the IN718 powder.
process parameter volume contour volume area
laser power [W] 100 175
laser scanning speed [mm/s] 540 620
hatch distance [mm] - 0.12
layer thickness [µm] 30 30

Table 3: HIP process parameter sets for parameter study

No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
T [C°] 500 700 900 1100 1300 500 700 900 1100 1300 500 700 900 1100 1300
p
[MPa] 50 50 50 50 50 100 100 100 100 100 150 150 150 150 150

Table 4: Results of the density measurement with different measurement methods

light microscopy cross-section longitudinal-section
(average value) 0.13 ± 0.06 % 0.09 ± 0.07 %
µCT full volume
(average value) 0.069 +/- 0.012 vol.-%.

Table 5: The hardness values of SLMed IN718 measured after different processing steps.
Condition Vickers
hardness HV0,1
as-built 273.2 ± 11.9
solution annealed 262.7 ± 6.8
precipitation hardened 482.7 ± 21.4
HIPed 258.8 ± 8.2

19