Accepted Manuscript

Title: Main defects observed in aluminum alloy parts

produced by SLM: from causes to consequences

Authors: Cassiopée Galy, Emilie Le Guen, Eric Lacoste,

Corinne Arvieu

PII: S2214-8604(17)30322-6
DOI: https://doi.org/10.1016/j.addma.2018.05.005
Reference: ADDMA 371

To appear in:

Received date: 26-7-2017

Revised date: 16-2-2018
Accepted date: 3-5-2018

Please cite this article as: Galy C, Le Guen E, Lacoste E, Arvieu C, Main defects
observed in aluminum alloy parts produced by SLM: from causes to consequences,
Additive Manufacturing (2010), https://doi.org/10.1016/j.addma.2018.05.005

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.
Main defects observed in aluminum alloy parts produced by SLM: from causes to consequences

Cassiopée GALY, Emilie LE GUEN, Eric LACOSTE*, Corinne ARVIEU

Univ. Bordeaux, I2M, UMR 5295, F-33400 Talence, France.

*Corresponding author:

T
IP
Eric LACOSTE

R
Present address: I2M, UMR 5295 - Site IUT - 15, rue Naudet - CS 10207 - 33175 - Gradignan Cedex

SC
- FRANCE Tel.: (33)5 56 84 58 65 – Fax: (33)5 56 84 58 43 – email: eric.lacoste@u-bordeaux.fr

HIGHLIGHTS:

U
N
 Defects of aluminium alloys SLMed parts are described from causes to consequences
A
 New prospects for studies concerning SLM and Al- alloys are highlighted
 Aluminium alloys studied by SLM are reviewed
M
ED

Abstract
PT

In recent years, the SLM process has been studied for the production of aluminum alloy parts, as these

alloys demonstrate significant potential for the future, notably due to their low density which allows a
E

considerable reduction in mass. The aim of this bibliographical study is to identify and classify the
CC

parameters and phenomena which influence the appearance of defects in aluminum alloy parts

produced using the SLM process and hence the final properties of these parts. To do this, a cause tree
A

diagram was created. For each defect or consequence identified (porosities, defects linked with hot

cracking phenomena, anisotropy in the material and surface quality), we revealed the potential sources

of the appearance of this defect, going back to the initial causes.

Keywords

1
Selective Laser Melting (SLM) – Aluminum alloys – Defects

1. Introduction

Additive manufacturing is not a single process, but a set of techniques with one principle in common,

that of shaping parts by the addition of material [1]: depositing layer by layer, depositing by spraying

matter, depositing by fused filament fabrication, etc. The basic principle used in these techniques is

therefore the opposite of traditional shaping which involves the removal of material (machining

T
processes). Additive manufacturing processes complement existing shaping processes (machining,

IP
welding, molding, injection, forming, etc.) by making it possible to produce new types of parts with

R
more complex geometries. Depending on the processes, the materials used are mainly in the form of

SC
liquids, powders, filaments or sheets and all families of materials can be used with this process,

polymers, metals, ceramics, composites, biomaterials, etc.

U
The Selective Laser Melting (SLM) is an additive manufacturing process using a powder bed where a
N
quasi-finished product is produced from metal powders and a 3D CAD file. A great deal of research
A
has been done into the SLM process applied to steels [2], Nickel-Chrome alloys [3] or titanium alloys
M

[4,5]. Studies have also tended to focus also on aluminum alloys. Aluminum and its alloys do indeed

have great potential for the future, due in particular to their low density which can result in significant
ED

reductions in volume. Due to the fast heating and cooling rates during SLM building during SLM

building, the microstructures obtained with SLM and and consequently the mechanical behavior are
PT

different compared to traditionally fabricated Al–12Si parts [6-8]. Moreover, despite being studied for

more than twenty years, the final properties of aluminum alloy parts manufactured by SLM are still
E

difficult to control due to the numerous defects generated, porosities, hot cracking and consequences,
CC

poor surface state or anisotropy.

Because of interactions between the laser and the metal powders, the SLM process involves
A

combinations of physical and thermal phenomena (with phase changes linked with the melting and

solidification of metal powders), metallurgical phenomena (diffusion), mechanical phenomena

(appearance of residual stresses due to temperature gradients during the process) and hydrodynamic

phenomena (convection in the melt pool). In the case of the manufacture of aluminum alloy parts,

2
macrographs and analysis of the microstructure show the formation of oxides at the surface of the melt

pool which may be responsible for the appearance of defects such as porosities and the trapping of

unmelted powders. The material health of parts produced from aluminum is therefore particularly

related to the size of the melt pools and the ability of the laser to break the oxide films that have

formed, which are themselves directly linked to the operational parameters of the process. In addition,

aluminum is a good heat conductor compared with other metal alloys used but it has a lower

T
absorption rate of the laser beam energy, hence a very large proportion of the laser energy is reflected.

IP
The aim of this bibliographic article is to investigate the origins of the main defects observed in

aluminum alloy parts manufactured by SLM, in order to identify and classify the parameters and

R
phenomena involved which influence the final properties and hence the appearance of defects. A well-

SC
known risk analysis was used, the root cause analysis tree diagram. The principle is simple: for each

U
defect identified, potential sources are revealed, going back to the initial causes, also called the root

causes.
N
A

2. Defects in SLM production of aluminium alloy parts

M

2.1. Principle of the SLM process

ED

In the literature, SLM is often associated with SLS (Selective Laser Sintering) [9], however, these two

additive manufacturing processes are quite distinct, even though their basic principle is the same.
PT

Figure 1 shows a system where powder is fed via a tank located above the level of the construction

zone. Its distribution in the recoater is made by gravity. Then, powder is deposited on the build
E

substrate thanks to the recoater movement. Another system deposits powder by an “elevator” powder
CC

reservoir [10]. Layer after layer, the lateral reserve is raised to deliver the required amount of powder.

In both cases, the powder is then spread in a thin uniform layer across the build surface using a
A

recoater mechanism that may be a rake, which may be flexible or hard, or a roller. The deposited

powder is between 10 and 100 µm thick. In the case of aluminum alloys, it is generally around 30 to

50 µm [11-14]. The thickness of the powder deposit depends on the granulometry of the powder used;

it corresponds to a slice of the part being produced, a slice that will then be sintered (SLS) or melted

3
(SLM) depending on the process employed. The energy required to change the state of the material is

provided by a laser.

The main difference between sintering and laser fusion is the power of the laser used. In SLS, the

powder is sintered. The temperature to be reached varies according to the grade of the aluminum alloy,

but is generally between 595°C and 625°C [15]. For SLM, on the other hand, the temperature to be

reached is at least the melting temperature of aluminum (T ≥ 660°C). As a result, the physical

T
phenomena involved are different. In the case of SLM, the mechanics of the melt fluids play a major

IP
part in the final quality of the parts. The sintering process is complicated to put into operation on

aluminum powders because of the natural presence of a thin layer of alumina [16].

R
Using a laser to melt the material gives the SLM process some similarities to laser assembly processes.

SC
Basically, a part made by SLM could be represented as a pile of laser weld beads. However, these 2

U
processes differ in some important ways and the results obtained for welded parts in terms of

microstructures and defects in particular cannot be transposed to the SLM process. The dimensions of
N
laser weld beads are a minimum of a few mm, whereas in the case of SLM, laser thicknesses are
A

around a hundred microns maximum. In addition, the scan speeds of the laser are much lower during
M

welding than in SLM and hence the cooling speeds of the alloy are also very different as they are
ED

extremely fast in SLM (~ 106 K.s-1). The thermal history during the process is also more disrupted in

the case of SLM: a layer of powders that melts as the laser passes across at a given moment will
PT

undergo variations in temperature as the laser passes over to make the upper layers (figure 2) [17]. The

environment of the melt pool is also very different: in welding, melting occurs all the time within one
E

or more materials in the solid state (parts for assembly) (figure 3) [18] whereas in the case of SLM on
CC

a powder bed, depending on the geometry of the part being produced and the laser strategy used, the

weld bead is either in contact with the powder or with the continuous material in the solid state, or
A

with a combination of the two, as illustrated in figure 4.

2.2. Parameters of the SLM process

To master the SLM process and thus account for the appearance of a particular defect, it is essential to

know and understand all the operational factors that can influence the final properties of the

4
manufactured product. The key parameters of SLM are presented in detail by Spears [10]. Spears

classifies the parameters into 4 categories: laser parameters, initial powder properties, powder bed

properties and recoating, and build environment parameters. Parameters defined as “controlled” are

those that can be easily modified with current SLM technology. Of the 50 parameters listed, only a

dozen are considered to be controllable, i.e. directly modifiable during the process.

T
To these fifty parameters identified by Spears, it is essential to add further parameters which can

IP
impact on the quality of the final part:

- the flowability of the powders: poor flowability can lead to problems with spreading the

R
powder;

SC
- the hygrometry of the powders: high humidity can also lead to problems with spreading the

-
powders;
U
the recoating technique used to spread the powders (roller or rake): some machines control the
N
compacting of the powders and others spread the powders on both the outward sweep and the
A

return sweep.
M

In view of the very large number of parameters and following studies that have focused on the
ED

parameters of the laser and the scanning, 4 of these parameters were studied through the specific laser

energy Ψ, defined according to equation (1) [19,20]:

𝑃
PT

𝛹 = 𝜈.ℎ.𝑑 (1)

where P is the power of the laser (W), v the scan speed of the laser (mm/s), h the inter-bead distance
E

(mm) and d the thickness of the layer of powder deposited (mm). Thus the energy density of the
CC

process increases if the laser power increases, or if the scan speed, the inter-bead distance or the

thickness of the layer decreases. There is a basic optimal specific energy, but its value seems to be
A

dependent on the overall environment and hence on the SLM machine and the alloy used [20].

However, this equation is debatable. Indeed, it does not take into account a number of important

factors such as the laser diameter and its offsets at the surface of the melt, the direction of the gas flow,

the hatch style and so on [21]. Prashanth et al claims that an extensive modification of the equation has

5
to be made involving the hatch parameters and the material properties in order to effectively employ

this in the SLM parameter optimization process. By comparing samples build using different process

parameters at constant energy density by varying the laser power and laser speed combination and at

constant speed and varying laser, their study suggests that the laser power is one of the most

influencing process parameter to be considered.

T
2.3. SLM process applied to aluminum alloys

IP
Manufacturers of metal additive manufacturing machines have chosen to use lasers with a wavelength

of 1µm (1.064µm) because this allows optimal absorption for all metal materials (figure 5) [22].

R
Nevertheless, for aluminum, laser absorption for a wavelength of 1064 nm is low, varying between 5

SC
and 15%. In the case of wavelengths close to 10µm (CO2 laser) the oxide layer Al2O3 naturally present

U
on the surface of the powder particles probably absorbs the laser energy better in this wavelength

range (figure 6) [23].

N
In the literature, all laser sources used emit in continuous mode. A few rare studies have looked into
A

the possibility of using pulsed sources to allow time for the heat to diffuse into the material between 2
M

pulses [24]. Chou et al. demonstrated that the change in laser emission mode significantly modifies the
ED

geometry of the microstructure obtained in the case of an AlSi12 alloy and therefore influences the

final properties of the material.

PT

2.4. Aluminum alloys produced by SLM

E

Overall, the most commonly studied aluminums in SLM belong to the group of alloys called structural
CC

hardening alloys. Table 1 summarizes the principal aluminum alloys that have already been studied in

SLM giving the references to bibliographic articles dealing with each of the alloys. The presence of
A

alloying elements such as Cu, Mg, Si and Zn make it possible to improve the mechanical properties by

a sequence of heat treatments. The majority of these alloys are reputed to be easily weldable, including

with a high energy process and rapid kinetics such as laser welding.

We observed that most studies were carried out using the 4XXX series (40,000 casting alloys). These

alloys are easy to process due to the small difference between liquidus and solidus temperature. In

6
addition the fact that these alloys have high silicon content gives a better laser absorption, avoids

cracking and gives the material good flowability and weldability properties. Today we easily obtain

parts with densities exceeding 99% using high laser powers of 200W and above. Moreover, taking the

example of AlSi10Mg, which is without a doubt the most studied aluminium alloy studied by SLM,

the mechanical properties of parts produced by SLM are higher than the one obtained by casting. It is

explained due to the fine cellular-dendritic microstructure and fine distribution of Si phase developed.

T
More and more studies are focused on the Al-Cu alloys (2XXX series). These high strength aluminium

IP
alloys represent a great potential for automotive, aerospace and defense industries. Actually, with a

higher strength-to-weight ratio than aluminium-silicon alloys, the expectation of reduction in mass is

R
increased. The high risk of failures to create dense part with classical Al-Cu alloys – due to the well-

SC
known susceptibility to hot cracking in welding - explains why it was first necessary to develop the

U
SLM process with aluminium alloy easy to process, as 4XXX series. The results indicate that these Al-

alloys display fine microstructure and considerably improved mechanical properties. Similar findings
N
can be associated to Al-alloys from 7XXX series.
A

Concerning 5XXX series, it should be noted that one particular alloy was developed by the Airbus
M

Group Innovation in order to find a compromise between the lightness of aluminum alloys and
ED

titanium alloy resistance: the Scalmalloy®. Due to its specific composition, this AlMgScZr alloys

offers excellent material properties in terms of weldability, corrosion resistance and mechanical
PT

properties.
E

2.5. Main defects and consequences identified on aluminum parts produced by SLM
CC

As in smelting and welding, four main defects and consequences were identified in the aluminum

alloy parts produced by SLM: porosities, defects related to hot cracking phenomena, anisotropy
A

induced in the material and surface quality. From this set of defects, we can start to build the cause

tree diagram (figure 7). For each case, we searched for all the causes for their appearance and grouped

all this information together to complete the tree shown in Figure 7, while systematically giving, for

each cause, the bibliographic references for where they were studied. It can be seen that many causes

are common to several defects.

7
3. Porosities related to the production of aluminum alloy parts by SLM

3.1. Causes of the appearance of porosities

Porosities are considered as one of the major defects in parts produced by SLM [67]. The density of a

part is an indicator of its material health and hence of the final performance that can be expected [68].

The size and the shape of a porosity are generally characteristics of the phenomenon that has caused it

T
[69]. Thus, we can classified the porosities of a SLM part into 2 categories: metallurgical pores which

IP
originate either from the absorption of the surrounding gas (N, O or H), or from the evaporation of

certain alloying elements (e.g. Mg), and parameter based pores which are mainly the result of a

R
process fault.

SC
Also, the size of these porosities is a revealing indicator of material health which affects the

performance of the final part. Thus “macroporosities” are considered more harmful than
U
microporosities. Indeed, if they are not perfectly spherical, macroporosities can be the source of
N
cracks. Consequently, the more macroporosities there are, the more fragile the material will be.
A

“Initial” microporosities may become critical, especially after heat treatment. During such operations
M

it can be seen that microporosities have a tendency to coalesce, forming macroporosities that are
ED

acicular in shape and therefore likely to generate the problems already discussed.

Among the causes of the appearance of porosities are those related to the process parameters, to the
PT

presence of impurities on the base material (the powder), to the low absorption of laser energy by the

aluminum, to problems of wettability of the solid aluminum by the liquid aluminum, to the
E

atmospheric conditions in the combustion chamber, or to phenomena of evaporation of the elements

CC

that make up the composition of the alloy. We built the tree for each of these causes, showing the

studies in the literature (Figure 8).

A

It should be recalled that all the studies dealing with the density of the parts produced by SLM are

dependent on the experimental means used. When the porosity level of a part is measured, the

dimension of the porosities being considered is usually not specified. Thus, depending on the method

used, determining the porosity level can give varying results.

8
3.2. Influence of the process on the appearance of porosities

The majority of studies on reducing the rate of porosity within a material produced by SLM focus on

the process, and especially on the optimization of the energy density factor, presented previously (1).

The parameters recognized as being the most influential for porosity levels are the laser scan speed

and laser power [12,20,24,26,30,36,43,45]. For instance, Kimura et al studied the influence of the

level of energy density on the kind of porosities [12]. By optimizing the laser irradiation conditions, a

T
relative density of 99.8% is obtained on A356.0 specimens. On as-build part, they showed that the

IP
relative density of the part decreased at both higher and lower energy density. In low-energy density

specimen, porosities were attributable to incomplete melting of the powder layer due to insufficient

R
energy supply from laser irradiation. It results porosities with an irregularly shaped and contained

SC
residual powder particles. In high-energy density case, they find a lot of spherical porosities with the

U
morphological features of gas pores. This is attributed either to a trapping of the argon from the

combustion chamber due to the active flow of the molten metal, or to the dissolved hydrogen having
N
degassed at the time of fusion.
A

Buchbinder et al. produced parts from aluminum alloy Al Si10 Mg with a volume of 5 mm3 and a
M

relative density of around 99.5% using a laser power of 250 W [70,71]. They showed that by simply
ED

increasing the laser power (~1000W), combined with an optimized laser speed and inter-bead distance,

the volume of the manufactured parts could be significantly increased with the same relative density
PT

(99.5%), while the build rate increased from 5 mm3/s to 21 mm3/s [43]. In this case, it has been

showed that low density of samples is also caused by spattering and process instabilities, which
E

produced imperfections. The gas flow at the surface of the lasered area is also a parameter that
CC

influences the porosity level [72]. Depending on the position and flow rate of the gas, slag (powder

that has already melted) may be present in the powder bed which leads to melting defects during
A

lasering and hence porosities.

3.3. Influence of other parameters on the appearance of porosities

In order to limit the appearance of porosities associated with the presence of hydrogen during the SLM

process, one method is to reduce the presence of humidity at the surface of the powder bed. To do this,

9
Weingarten et al. proposed a double scan system of the powder bed: a first scan is carried out with a

low power laser resulting in the sintering of the powder; the second scan is the one that melts the

powder [44]. This initial laser application reduces porosity density by 90% when producing an

AlSi10Mg alloy. Other authors have shown that porosity can be considerably reduced by drying the

powder before placing it in the SLM machine [46].

Studies have shown that decreasing the oxygen content in the machine results in a decrease in porosity

T
associated with the presence of oxygen [60]. In addition, Louvis et al. recall difficulties in producing

IP
aluminum alloy parts by SLM because of the combined effects of the very high reflectivity and

thermal conductivity of these materials [36].

R
SC
4. Other defects in aluminum alloy parts produced by SLM

4.1. Hot cracking defects

U
Another defect encountered when laser welding aluminum alloys is hot cracking [73]. As there is a
N
large solidification range, we observe a lack of supply of liquid to the interdendritic spaces of the weld
A

metal zone during solidification, which causes cracking.

M

One method to avoid these cracks is to modify the composition of the welded alloy to increase the
ED

amount of eutectic. In particular, the addition of Silicon to the alloy reduces the melting temperature

and the solidification range, thus leading to a decrease in the level of hot cracking [74]. Adding Nickel
PT

to the alloy can also reduce the appearance of cracks [64].

To limit these cracking problems when producing aluminum parts by SLM, most studies were carried
E

out using what are called casting aluminum alloys (AlSi10Mg, AlSi12, AlSi7Mg0.3), which are less
CC

susceptible to the risk of hot cracking because of the high Silicon content (figure 9) [75] and their

limited solidification range (ΔT ~50°C).

A

To understand the reasons why this defect appears, we looked in particular at the phenomena present

during laser welding (figure 10).

In his work on 6XXX series aluminum alloy laser welding, Fabrègue identified 5 factors that influence

hot cracking: the composition of the alloy, the microstructure, the dissolved gas content, the

wettability of the solid by the liquid and the process [76]. Sanchez et al. showed that the susceptibility

10
of an aluminum alloy to cracking depends on both its composition and the laser parameters used

(power, speed) [77]. It exits similarities between these phenomena and those observed in on SLM

process. Zhang et al. found that, in the case of an Al-Cu-Mg alloy, which is known to be sensitive to

cracking because of the presence of copper and magnesium, the reduction in laser speed leads to a

decrease in the presence of defects, including cracks [56]. Montero et al. successfully manufactured

parts out of alloy 7075 by SLM with no cracking by doping the composition with silicon [59].

T
Bartkowiak et al don’t observed cracks in their SLM samples manufactured with AlCu, ALZn or

IP
AlSi10Mg [29].

R
4.2. Anisotropy of manufactured parts

SC
Another consequence identified when producing aluminum parts by SLM concerns the anisotropy of

U
mechanical properties. This may appear to be secondary when compared with the two defects

mentioned previously, but studies show that mechanical properties depend on the build direction of the
N
part (Figure 11). Figure 12 summarizes the causes identified as being responsible for the anisotropic
A

nature of the aluminum parts produced by SLM.

M

There are two causes that may explain this microstructure orientation. The first concerns the build
ED

orientation of the part: because of the complex thermal history that the part has undergone, we observe

epitaxial grain growth [78]. During the study of the overall texture of AlSi10Mg produced by SLM
PT

Thijs et al have demonstrated that a more anisotropic or isotropic part can be obtained by choosing the

applied scanning strategy [51].

E

During their studies concerning the impact of processing parameters on tensile properties, Prashanth et
CC

al have showed that some combination of parameters may lead to anisotropic properties in the material

due to not homogeneous throughout the length of the AlSi12 samples produced by SLM [79]. This
A

criterion must therefore be taken into account when choosing the orientation and the support of the

part during the CAD stage (pre-manufacture).

Some authors have observed anisotropy in mechanical properties about samples manufactured by

SLM: longitudinal fracture strains were greater than those observed in the transversal direction in an

AlSi7Mg0.3 alloy produced by SLM in optimal conditions [12]; the elongation for parts built in

11
transversal direction is almost 2% higher than the elongation in the longitudinal direction with an

AlSi10Mg alloy [41] and the hardness observed along the build direction is higher than the

perpendicular one with the same alloy [42]. Karg et al have also observed anisotropy in the elongation

at break with an AlCu6Mn alloy while yield strength and ultimate tensile strength were isotropic [31]

and anisotropy for elongation and ultimate tensile strength with AlCu2Mg1.5Ni [61]. Other authors

have observed isotropic properties on their specimens on AlSiMg [48] or very low anisotropy on

T
Scalmalloy [26].

IP
4.3. Study of surface conditions

R
An optimal use of the SLM process would be the possibility of producing an operational part using

SC
SLM, where no post-processing machine finishing is required. To date, this goal has not yet been

U
achieved as surface conditions obtained after SLM do not fulfill the specifications for certain

applications, particularly aeronautics. Studying this consequence is therefore a means to improve the
N
process, especially in the case of aluminum alloys. Figure 13 shows the various avenues to explore to
A

bring about improvement in this criterion.

M

Few studies so far have looked into this problem. Currently, the solution usually employed to improve
ED

surface condition is to vary the laser parameters locally on the contours of the parts. However, this

change in the operational parameters may lead to the formation of porosities in the parts. The laser
PT

parameters used will directly affect the stability of the melt pool obtained and hence the uniformity of

the final bead [36, 45]. The phenomenon of balling, where coarse molten metal droplets are formed, is
E

also observed in SLM. This phenomenon depends on processing parameters and acts to the surface
CC

quality: Olakanmi has showed the influence of laser power and scanning rate about surface

morphology [47] ; Louvis et al have observed that low scan speeds generate roughness greater than the
A

set layer thickness with AlSi12 [36] ; Calignano et al studied the influence of laser and process

parameters like laser beam, power, spot size scanning speed, hatching distance and layer thickness

about the surface quality of AlSi10Mg produced by SLM [80]. In addition of these parameters, the

surface quality depends also on scan strategies such as contour scans and skywriting scans: the surface

12
roughness can be improved by means of a contour scan to alleviate surface inhomogeneity, and by

skywriting to facilitate scanning with more uniform energy density [81].

5. Conclusion

Studies of the SLM process cover very many different grades of aluminium alloy. SLM creates

complex functional products by selectively melting powder particles of a powder bed layer after layer

T
using a high-energy laser beam. The high-energy density applied to the material with SLM processing,

IP
the fast heating and cooling rates during SLM building and the additive character of the process result

in a material which microstructure and properties are different from those of materials manufactured

R
with traditional process. Like any other process, SLM applied to the aluminum alloy parts involves

SC
specific defects (porosities, defects related to hot cracking phenomena) and other consequences

U
concerning anisotropy induced in the material and surface quality. From each set of defects, according

to the literature studies, we have proposed a cause tree diagram which detailed all causes responsible
N
for the occurrence of defects.
A

The parameters of the process, the wettability of aluminium by melted aluminium, the low absorption
M

of laser energy and the presence of impurities have been identified as the main causes of the presence
ED

of porosities and hot cracking. It should be added the evaporation of the elements of alloys

composition and the atmosphere of building room for the porosities and the alloy composition and the
PT

dissolved gas content for hot cracking.

The parameters of the process and the bailing effect the surface quality while the building direction
E

and plate preheating are the two principles parameters responsible of the presence of anisotropy.
CC

Each of the causes of each defect has been then detailed in a cause tree diagram with literature

references corresponding.
A

Future research should focus on defining new grades of aluminium alloy that can limit defects such as

porosities and cracking. Furthermore, the SLM process involves complex coupled phenomena

associated with interactions between the laser and the material, initially in the form of powders. In

order to study and fully understand defects in such a complex process, instrumentation needs to be

developed which will be able to monitor changes in physical quantities during the process and modify

13
certain parameters while the process is underway. These studies can be developed from SLM

machines but also from instrumented test benches which can reproduce physical phenomena that are

representative of the SLM process.

Acknowledgment

This work is carried out within the framework of the FUTURPROD research project in collaboration

T
with the region of Nouvelle-Aquitaine (France), ENSAM, ArianeGroup, Stelia Aerospace, PolyShape,

IP
and AGB.

R
References

SC
[1] AFNOR, NF E 67-001 “Fabrication additive- vocabulaire” (2011).

[2] C.Y. Yap, C.K Chua, Z.L. Dong, Z.H. Liu, D.Q. Zhang, L.E. Loh, S.L. Sing, Review of selective

U
laser melting: materials and applications, Applied Physics Reviews 2, 041101(2015).
N
[3] K.N. Amato, S.M. Gaytan, L.E. Murr, E. Martinez, P.W. Shindo, J. Hernandez, S.Collins, F.
A
Medina, Microstructures and mechanical behaviour of Inconel 718 fabricated by selective laser
M

melting, Acta Materialia 60 (2012) 2229-2239.

[4] M. Simonelli, Y.Y. Tse, C. Tuck, Effect of the build orientation on the mechanical properties and
ED

fracture modes of SLM Ti-6Al-4V, Materials Science & Engineering A 616 (2014) 1–11.

[5] V. Cain, L. Thijs, J. Van Humbeeck, B. Van Hooreweder, R. Knutsen, Crack propagation and
PT

fracture toughness of Ti6Al4V alloy produced by selective laser melting, Additive Manufacturing 5

(2015) 68–76.
E

[9] E.O. Olakanmi, R.F. Cochrane, K.W. Dalgarno, A review on selective laser sintering/melting
CC

(SLS/SLM) of aluminium alloy powders: Processing, microstructure, and properties, Progress in

Materials Science (2015) http://dx.doi.org/10.1016/j.pmatsci.2015.03.002.

A

[10] T.G. Spears, S.A. Gold, In-procress sensing in selective laser melting (SLM) additive

manufacturing, Integrating materials and Manufacturing Innovation (2016) 5:2.

14
[11] D. Buchbinder, W Meiners, K. Wissenbach, R. Poprawe, Selective laser melting of aluminium

die-cast alloy – Correlations between process parameters, solidification conditions, and resulting

mechanical properties, Journal of Laser Applications 27, S29205 (2015); doi: 10.2351/1.4906389.

[12] T. Kimura, T. Nakamoto, Microstructures and mechanical properties of A356 (AlSi7Mg0.3)

aluminum alloy fabricated by selective laser melting, Mater. and Design 89 (2016) 1294-1301.

[13] D. Manfredi, F. Calignano, E.P. Ambrosio, M. Krishnan, R. Canali, S. Biamino, M.Pavese, E.

T
Atzeni, L. Luliano, P. Fino, C. Badini, Direct metal laser sintering: an additive manufacturing

IP
technology ready to produce lightweight structural parts for robotic applications, La metallurgia

Italiana 105 (2013) 15-24.

R
[14] P. Ma, K.G. Prashanth, S. Scudino, Y. Jia, H. Wang, C. Zou, Z. Wei, J. Eckert, Influence of

SC
Annealing on Mechanical Properties of Al-20Si Processed by Selective Laser Melting, Metals 4

(2014), 28-36; doi:10.3390/met4010028.

U
[15] ASM specialty handbook, Aluminium and aluminium alloys, sixth printing, p.720, 2007.
N
[16] Rapport République Française Ministère de l’Economie et des Finances – République Française,
A

Futur de la fabrication additive: Focus sur les alliages d’aluminium, ISBN 978-2-11-151552-9, 2017.
M

[17] F. Wang, S. Williams, P. Colegrove, A.A. Antonysamy, Microstructure and mechanical

ED

properties of Wire and Arc Additive Manufactured Ti-6Al-4V, Metallurgical and materials

transactions 44A (2013) 968-977.

PT

[18] K. Abderrazak, W. Ben Salem, H. Mhiri, G. Lepalec, M. Autric, Modelling of CO2 laser welding

of magnesium alloys, Optics & Laser Technology 40 (2008) 581–588.

E

[19] A. Simchi, H. Pohl, Effects of laser sintering processing parameters on the microstructure and
CC

densification of iron powder, Materials and Engineering A359 (2003) 119-128.

[20] N. Read, W. Wang, K. Essa, M.M. Attalah, Selective laser melting of AlSi10Mg alloy: Process
A

optimisation and mechanical properties development, Materials and Design 65 (2015) 417-424.

[21] K.G. Prashanth, S. Scudino, T. Maity, J. Das, J. Eckert, Is the energy density a reliable parameter

for materials synthesis by selective laser melting?, Materials Research Letters (2017) 1-5.

[22] L. Quintino, R. Miranda, U. Dilthey, D. Iordachescu, M. Banasik, S. Stano, Laser Welding of

Structural Aluminium, Adv. Struct. Mater 8 (2012) 33–57.

15
[23] Y. Arata, I. Miyamoto, Some fundamental properties of high power laser beam as a heat source

(report 2) – CO2 laser absorption characteristics of metal, Transactions of the Japan Welding Society 3

(1972) 152-162.

[24] R. Chou, J. Milligan, M. Paliwal, M Brochu, Additive manufacturing of alloy via pulsed selective

laser melting, JOM 67 (2015) 590-596.

[25] A.B. Spierings, K. Dawson, T. Heeling, P.J. Uggowitzer, R. Schäublin, F. Palm, K. Wegener,

T
Microstructural features of Sc- and Zr-modified Al-Mg alloys processed by selective laser melting,

IP
Mater. and Design 115 (2017) 52-63.

[26] K. Schmidtke, F. Palm, A. Hawkins, C. Emmelmann, Process and Mechanical Properties:

R
Applicability of a Scandium modified Al-alloy for laser Additive Manufacturing, Phys. Procedia 12

SC
(2011) 369-374.

U
[27] R Gehm, High-strength aluminum powder developed for additive manufacturing in aerospace,

automotive, http://articles.sae.org/14175/, 2015 (accessed 18.04.17).

N
[28] D. Buchbinder, K. Wissenbach, Additive manufacturing of high-strength aluminium components,
A

https://www.generativ.fraunhofer.de/, 2015 (accessed 13.03.17).

M

[29] K. Bartkowiak, S. Ullrich, T. Frick, M. Schmidt, New developments of laser processing

ED

aluminium alloys via additive manufacturing technique, Physics Procedia 12 (2011) 393-401.

[30] M. Karg, B. Ahuja, S. Kuryntsev, A. Gorunov, M. Schmidt, Processability of high strength

PT

Aluminium-Copper alloys AW-2022 and 2024 by Laser Beam Melting in Powder Bed, sff symposium

(2014) 420-436.
E

[31] M. Karg, B. Ahuja, S. Wiesenmayer, S. Kuryntsev, M. Schmidt, Effect of the process conditions
CC

on the mechanical behaviour of aluminium wrought alloy EN AW-2219 (AlCu6Mn) additively

manufactured by laser beam melting in powder bed, Micromachines (2017) 8, 23,

A

doi:10.3390/mi8010023.

[32] S. Dadbakhsh, L. Hao, Effect of layer thickness in selective laser melting on microstructure of

Al/5wt%Fe2O3 powder consolidated parts, The Scientific World Journal (2014)

http://dx.doi.org/10.1155/2014/106129.

16
[33] L. Zheng, Y. Liu, S. Sun, H. Zhang, Selective laser melting of Al-8,5Fe-1,3V-1,7Si alloy:

investigation on the resultant microstructure and hardness, Chinese journal of aeronautics (2015) 564-

569.

[34] S. Sun, L. Zheng, Y. Liu, J. Liu, H. Zhang, Characterisation of Al-Fe-V-Si heat-resistant

aluminum alloy components fabricated by selective laser melting, Journal of materials research 30

(2015) 1661-1669.

T
[35] P. Ma, Y. Jia, K.G. Prashanth, S. Scudino, Z. Yu, J. Eckert, Microstructure and phase formation

IP
in Al-20Si-5Fe-3Cu-1Mg synthesized by selective laser melting, Journal of Alloys and Compounds

657 (2016) 430-435.

R
[36] E. Louvis, P. Fox, C.J. Sutcliffe, Selective laser melting of aluminium components, Journal of

SC
Materials Processing Technology 211 (2011) 275–284.

U
[37] P.Vora, K. Mumtaz, I. Todd, N. Hopkinson, AlSi12 in-situ alloy formation and residual stress

reduction using anchorless selective laser melting, Additive Manufacturing 7 (2015) 12-19.
N
[6] X.P Li, X.J Wang, M. Saunders, A. Suvorova, L.C. Zhang, Y.J. Liu, M.H. Fang, Z.H. Huang, T.B.
A

Sercombe, A selective laser melting and solution heat treatment refined Al-12Si alloy with a
M

controllable ultrafine eutectic microstructure and 25% tensile ductility, Acta Materialia 95 (2015) 74-
ED

82.

[7] K.G. Prashanth, S. Scudio, H.J. Klauss, K.B. Surreddi, L. Löber, Z. Wang, A.K.Chaubey, U.
PT

Kühn, J. Eckert, Microstructure and mechanical properties of Al-Si12 produced by selective laser

melting: Effect of heat treatment, Materials Science & Engineering A 590 (2014) 153-160.
E

[38] N.T. Aboulkhair, I. Maskery, C. Tuck, I. Ashcroft, N. M. Everitt, The microstructure and
CC

mechanical properties of selectively laser melted AlSi10Mg: The effect of a conventional T6-like heat

treatment, Materials Science & Engineering A 667 (2016) 139-146.

A

[39] F. Trevisan, F. Calignano, M. Lorusso, J. Pakkanen, A. Aversa, E. P. Ambrosio, M. Lombardi, P.

Fino, D. Manfredi, On the selective laser melting (SLM) of the AlSi10Mg alloy: Process,

Microstructure, and Mechanical Properties, Materials 10 (2017) doi:10.3390/ma10010076.

17
[40] D. Buchbinder, W. Meiners, N. Pirch, K. Wissenbach, Investigation in the reducing distortion by

preheating during manufacture of aluminium components using selective laser melting, Journal of

laser applications 26 (2014) http://dx.doi.org/10.2351/1.4828755.

[41] K. Kempen, L. Thijs, J. Van Humbeeck, J.P. Kruth, Mechanical properties of AlSi10Mg

produced by Selective Laser Melting, Physics Procedia 39 (2012) 439-446.

[42] N.T. Aboulkhair, A. Stephens, I. Maskery, C Tuck, I. Ashcroft, N. M. Everitt, Mechanical

T
properties of selective laser melted AlSi10Mg: nano, micro, and macro properties, In proceeding of

IP
the solid freeform fabrication symposium, Austin, TX, USA (2015) 1026-1035.

[43] D. Buchbinder, H. Schleifenbaum, S. Heidrich, W. Meiners, J. Bültmann, High power selective

R
laser melting (HP SLM) of aluminium parts, Physics Procedia 12 (2011) 271–278.

SC
[44] C. Weingarten, D. Buchbinder, N. Pirch, W. Meiners, K. Wissenbach, R.Poprawe, Formation and

U
reduction of hydrogen porosity during laser melting of AlSi10Mg, Journal of Materials Processing

Technology 221 (2015) 112-120.

N
[45] K. Kempen, L. Thijs, E. Yasa, M. Badrossamay, W; Verheecke, J.P. Kruth, Process optimization
A

and microstructural analysis for selective laser melting of AlSi10Mg, International Solid Freeform
M

fabrication Symposium 22 (2011) 484-495.

ED

[8] W. Li, S. Li, A. Zhang, Y. Zhou, Q. Wei, C. Yan, Y. Shi, Effect of heat treatment on AlSi10Mg

alloy fabricated by selective laser melting: Microstructure evolution, mechanical properties and
PT

fracture mechanism, Materials Science & Engineering A 663 (2016) 116-125.

[46] X.P. Li, K.M. O’Donnell, T.B. Sercombe, Selective laser melting of Al-12Si alloy: enhanced
E

densification via powder drying, Additive Manufacturing 10 (2016) 10-14.

CC

[47] E.O. Olakanmi, Selective laser sintering/melting (SLS/SLM) of pure Al, Al-Mg, and Al-Si

powders: effect of processing conditions and powder properties, Journal of Materials Processing
A

Technology 213 (2013) 1387-1405.

[48] D. Manfredi, F. Calignano, M. Krishnan, R. Canali, E.P. Ambrosio, E. Atzeni, From powders to

dense metal parts, characterization of a commercial AlSiMg alloy processed through Direct Metal

Laser Sintering, Materials 6 (2013), 856-869.

18
[49] E.O. Olakanmi, K.W. Dalgarno, R.F. Cochrane, Laser sintering of blended Al-Si powders, Rapid

Prototyping Journal 18 (2012) 109-119.

[50] E.O. Olakanmi, R.F. Cochrane, K.W. Dalgarno, Densification mechanism and micro-structural

evolution in selective laser sintering of Al-12Si powder, Journal of Materials Processing Technology

211 (2011) 113-121.

[51] L. Thijs, K. Kempen, J.P. Kruth, J. Van Humbreek, Fine-structured aluminium products with

T
controllable texture by selective laser melting of pre-alloyed AlSi10Mg powder, Acta Materialia 61

IP
(2013) 1809–1819.

[52] S. Sun, L. Zheng, Y. Liu, J.Liu, H. Zhang, Selective laser melting of Al-Fe-V-Si heat-resistant

R
aluminium alloy powder: modelling and experiments, Int. J. Adv. Manuf. Technol. 80 (2015) 1787-

SC
1797.

U
[53] Y. Li, D. Gu, Parametric analysis of thermal behaviour during selective laser melting additive

manufacturing of aluminium alloy powder, Materials & design 63 (2014) 856-867.

N
[54] H. Zhang, H. Zhu, X. Nie, T. Qi, Z. Hu, X. Zeng, Fabrication and heat treatment of high strength
A

Al-Cu-Mg alloy processed using selective laser melting, Proc. Of SPIE (2016) doi:
M

10.1117/12.2211362.
ED

[55] H. Zhang, H. Zhu, T. Qi, Z. Hu, X. Zeng, Selective laser melting of high strength Al-Cu-Mg

alloy: Processing, microstructure and mechanical properties, Materials Science & Engineering A 656
PT

(2016) 47-54.

[56] H. Zhang, H. Zhu, X Nie, J Yin, Z Hu, X Zeng, Effect of Zirconium addition on crack,
E

microstructure and mechanical behaviour of selective laser melted Al-Cu-Mg alloy, Scripta Materialia
CC

134 (2017) 6-10.

[57] R. Casati, J.N. Lemke, A.Z. Alarcon, M. Vedani, Aging Behavior of High-Strength Al Alloy
A

2618 Produced by Selective Laser Melting, Metellurgical and Materials Transactions A 48 (2016)

575-579.

[58] D.K. Kim, W. Woo, J.H. Hwang, K. An, S.H. Choi, Stress partitioning behaviour of an

AlSi10Mg alloy produced by selective laser melting during tensile deformation using in situ neutron

diffraction, Journal of Alloys and Compounds 686 (2016) 281-286.

19
[59] M.L. Montero Sistiaga, R. Mertens, B. Vrancken, X. Wang, B. Van Hooreweder, J.P. Kruth, J.

Van Humbeeck, Changing the alloy composition of Al7075 for better processability by selective laser

melting, Journal of Materials Processing Technology 238 (2016) 437-445.

[60] D.M. Bauer, K. Dietrich, M. Walter, P. Forêt, G. Witt, Effect of process gas and powder quality

on aluminium alloys processed by laser based powder bed melting process, Solid Freeform Fabrication

2016: Proceeding on the 27th annual international solid freeform fabrication symposium – An Additive

T
Manufacturing Conference (2016) 419-425.

IP
[61] M. Karg, B. Ahuja, A. Schaub, J. Schmidt, M. Sachs, A. Mahr, S. Wiesenmayer, L. Wigner,

K.E. Wirth, W. Peukert, M. Merklein, M. Schmidt, Effect of process conditions on the mechanical

R
behavior of aluminium wrought alloy EN AW-2618 additively manufactured by laser beam melting in

SC
powder bed, Lasers in Manufacturing Conference (2015).

U
[62] A. B. Spierings, K. Dawson, K.Kern, F.Palm, K. Wegener, SLM-processed Sc- and Zr- modified

Al-Mg alloy: Mechanical properties and microstructural effects of heat treatment, Materials Science &
N
Enginnering A 701 (2017) 264-273.
A

[63] Y.D. Jia, P. Ma, K.G. Prashanth, G. Wang, J. Yi, S. Scudino, F.Y. Cao, J.F. Sun, J. Eckert,
M

Microstructure and thermal expansion behaviour of Al-50Si synthesized by selective laser melting,
ED

Journal of Alloys and Compounds 699 (2017) 548-553.

[64] A. Aversa, M. Lorusso, G. Cattano, D. Manfredi, F. Calignano, E.P Ambrosia, S. Biamino, P.

PT

Fino, M. Lombardi, M. Pavese, A study of the microstructure and the mechanical properties of an Al-

Si-Ni alloy produced via selective laser melting, Journal of Alloys and Compounds 695 (2017) 1470-
E

1478.
CC

[65] P. Wang, H.C. Li, K.G. Prashanth, J. Eckert, S. Scudino, Selective laser melting of Al-Zn-Mg-

Cu: Heat treatment, microstructure and mechanical properties, Journal of Alloys and Compounds 707
A

(2017) 287-290.

[66] K.G. Prashanth, H. Shakur Shahabi, H. Attar, V.C. Srivastava, N. Ellendt, V. Uhlenwinkel, J.

Eckert, S. Scudino, Production of high strength Al85Nd8Ni5Co2 alloy by selective laser melting,

Additive Manufacturing 6 (2015) 1-5.

20
[67] J-P. Kruth, M. Badrossamay, E. Yasa, J. Deckers, L. Thijs, J. Van Humbeeck, Part and material

properties in selective laser melting of metal, 16th International Symposium on Electro machining

(ISEM XV) (2010).

[68] V.T. Morgan, The effect of porosity on some of the physical properties of powder-metallurgy

components, Powder metallurgy (2014) 72-86.

[69] W.E. Frazier, Metal Additive Manufacturing: A review, ASM International (2014) 1059-1928,

T
doi: 10.1007/s11665-014-0958-z.

IP
[70] D. Buchbinder, W. Meiners, K. Wissenbach, K. Müller-Lohmeier, E. Brandl, Rapid

Manufacturing of Aluminum Parts for Serial Production via Selective Laser Melting (SLM). In:

R
Aluminum Alloys, Wiley Vol. 2 (2008) p. 2394-2400.

SC
[71] D. Buchbinder, W. Meiners, K. Wissenbach, K. Müller-Lohmeier, E. Brandl, N. Skrynecki,

U
Rapid manufacturing of aluminum parts for serial production via Selective Laser Melting (SLM). In:

4. International Conference on Rapid Manufacturing, 8th & 9th July 2009. Loughborough : University
N
(2009) 1-12.
A

[72] A.B. Anwar, Q-C Pham, Selective laser melting of AlSi10Mg: Effects of scan direction, part
M

placement and inert gas flow velocity on tensile strength, Journal of Materials Processing Technology
ED

240 (2017) 388–396.

[73] M. Rappaz, J.M Drezet, M Gremaud, A new Hot-Tearing Criterion, Metallurgical and materials
PT

transactions A 30 (1999) 449-455.

[74] J.E. Hatch, Aluminium: Properties and Physical Metallurgy, ASM International, 1984.
E

[75] J. Verö, The Hot-Shortness of Aluminum Alloys. The Metal Industry 48 (1936) 431-494.
CC

[76] D. Fabrègue, Microstructure et fissuration à chaud lors du soudage laser d’alliages d’aluminium

6000, thèse INP Grenoble, (2016)

A

[77] J.M. Sánchez-Amaya, T. Delgado, L. González-Rovira, and F.J. Botana, Laser welding of

aluminum alloys 5083 and 6082 under conduction regime. Appl. Surf. Sci. 255(23) (2009) 9512–9521.

[78] D. Dai, D. Gu, R. Poprawe, M. Xia, Influence of additive multilayer feature on thermodynamics,

stress and microstructure development during laser 3D printing of aluminium-based material, Science

Bulletin 62 (2017) 779-787.

21
[79] K.G. Prashanth, S. Scudino, J. Eckert, Defining the tensile properties of Al-12Si parts produced

by selective laser melting, Acta Materialia 126 (2017) 25-35.

[80] F. Calignano, D. Manfredi, E. P. Ambrosio, L. Iuliano, P. Fino, Influence of process parameters

on surface roughness of aluminum parts produced by DMLS, Int J Adv Manuf Technol 67 (2013)

2743–2751.

[81] Y. Tian, D. Tomus, P. Rometsch, X. Wu, Influences of processing parameters on surface

T
roughness of HastelloyX produced by selective laser melting, Additive Manufacturing 13 (2017) 103–

IP
112.

R
Figure

SC
U
N
A
M
ED
E PT

Figure 1: Principle of SLM processes.

CC
A

22
 T
IP
Figure 2: Thermal history of a Ti-6Al-4V alloy manufactured in WAAM, calculated from the

R
finite element model during the last ten layers [17]

SC
U
N
A
M
ED

Figure 3: Physical phenomena during laser welding [18]

E PT
CC
A

23
 T
R IP
Figure 4: Different thermal transfer scenarii according to the geometry and the scanning

SC
strategy for the SLM process [10]

U
N
A
M
ED
E PT
CC

Figure 5: Laser radiation absorption as a function of wavelength: dependence of the material

[22]
A

24
Figure 6: Absorption of laser energy CO2 as a function of the thickness of alumina [23]

T
R IP
SC
U
N
A
M

Figure 7: Root of the cause tree diagram

ED
E PT
CC
A

25
 T
R IP
SC
U
N
A
M
ED
EPT
CC
A

Figure 8: Tree of causes - development of the branch "porosity"

26
 T
R IP
Figure 9: Influence of alloying elements on hot cracking [76]

SC
U
N
A
M
ED
EPT
CC
A

27
 T
R IP
SC
U
N
A
M
ED
E PT
CC
A

Figure 10: Tree of causes - development of the branch "hot cracking"

28
 T
Figure 11: EBSD observation of the microstructure of an AlSi10Mg part manufactured by SLM

IP
[51]

R
SC
U
N
A
M

Figure 12: Tree of causes - development of the "anisotropy" branch

ED
E PT
CC
A

Figure 13: Tree of causes - development of the "surface quality" branch

29
 Table 1: Summary of aluminum alloys studied in SLM

Feasibility by SLM
Susceptibility to
Part density
cracking
Family Alloys + : not susceptible + : d rel ≥ 99% References
0 : susceptible 0 : 97% <d rel < 99%
- : very susceptible - : d rel ≤ 97 %
n.s : not said n.s : not said
1XXX Al + n.s [47]
AlCu2 - n.s [29-31,
AlCu5 0 n.s 54-57,61]

T
AlCu6.8 + n.s
AlCu11.8 + n.s

IP
AlCu5Mg (2022) - +
2XXX
AlCu4Mg1 (2024) 0 +

R
AlCu6Mn (2219) 0 +
AlCuMg 0 +

SC
AlCu2Mg1.5Ni + n.s
(2618)
AlSi7Mg0.3 + + [6-8,
(A356) 11-14, 20,24,
AlSi7Mg0.6
(A357)
+
U + 35-53, 58,
63,64]
N
AlSi20Fe5Cu3Mg1 +
4XXX AlSi20 + +
A
AlSi12 + +
AlSi50 + n.s
M

AlSi10Mg + +
AlSi12Mg + +
AlSiNi + +
ED

AlMg5.7 + n.s [25-28, 47,62]

AlMg6 + n.s
5XXX
AlMgScZr + +
(ScalmalloyⓇ)
PT

6XXX AlMg1SiCu (6061) 0 - [36]

AlZn5 0 n.s [29, 59,60, 65]
E

AlZn2 0 n.s
AlZn10 0 n.s
7XXX
CC

AlZnMgCu n.s n.s

AlZn5.5MgCu - 0
(7075)
AlFe8.5V1.3Si1.7 + + [33, 34, 52]
A

8XXX (FVS0812 ou
AA8009)
Al(Fe2O3)5 n.s n.s [32, 66]
NC
AlNdNiCo + +

30