Powder Bed Fusion

24
Srujana Rao Yarasi, Andrew R. Kitahara, Elizabeth A. Holm, and
Anthony D. Rollett

Contents
24.1 Introduction to PBF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 389 techniques. This includes a discussion on the powder
24.1.1 Laser Powder Bed Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391 production methods, available characterization techniques
24.1.2 Electron Beam Powder Bed Fusion . . . . . . . . . . . . . . . . . . . . . . 391 for both powder and part, as well as their interpretation as
24.2 Machine Parameters for PBF . . . . . . . . . . . . . . . . . . . . . . . . . . . 392 it relates to part performance. The section discusses rele-
24.2.1 Process Parameter Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392 vant examples of how this process is used in the aero-
24.2.2 Energy Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392 space, medical, and energy sectors to produce complex,
24.2.3 Scanning Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
24.2.4 Thermal Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394
high-value parts such as heat exchangers whose perfor-
mance can be increased with individualized design and
24.3 Powder Parameters for PBF . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394
seamless one-step manufacturing. A few challenges
24.3.1 Powder Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394
24.3.2 Powder Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395 unique to powder bed fusion processes are also discussed,
24.3.3 Powder Recycling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396 including limitations in powder recycling, relatively long
24.4 Post-processing Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397 production times, and the need for post-processing.
24.4.1 Support Structure Removal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 397
24.4.2 Improving Mechanical Properties . . . . . . . . . . . . . . . . . . . . . . . . 398 Keywords
24.5 Part Properties and Characterization . . . . . . . . . . . . . . . . . 398 Additive manufacturing · Powder bed fusion · LPBF · EB-
24.5.1 Mechanical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 398
24.5.2 Porosity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399
PBF · Metal powders · Flowability · Porosity

24.6 PBF in Practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399

24.6.1 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399
24.6.2 Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 400 24.1 Introduction to PBF
24.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401
Powder bed fusion (PBF) is a popular type of additive
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402 manufacturing (AM), wherein thin layers of powder are
spread over a build plate and selected areas of the powder
bed belonging to the part model are fused with an energy
Abstract source (laser or electron beam, for example). These steps are
This chapter covers the different types of metal powder repeated over numerous layers until the final three-
bed fusion processes, specifically expanding on electron dimensional part is formed [1]. In such a manner, a complex
beam powder bed fusion (EB-PBF) and laser powder bed and intricate CAD part model can result in a tangible AM
fusion (LPBF), and their advantages. The section also part. Once the part is finished and extracted, the remaining
covers the starting powder material characterization and powder may be recycled. The advantages of such a system
the resultant part characterization, which includes topics include its ability to produce high-resolution features and
such as flowability, porosity, and post-processing internal passages, and maintain dimensional control [2]. In
Fig. 24.1, the thin-walled hollow cylinders that are efficiently
S. R. Yarasi (*) · A. R. Kitahara · E. A. Holm · A. D. Rollett (*) packed in a hexagonal close packing arrangement would be
Department of Materials Science and Engineering, Carnegie Mellon incredibly difficult to create using conventional manufactur-
University, Pittsburgh, PA, USA
e-mail: [email protected]; [email protected]; ing techniques but have the ability to work well with PBF
[email protected]; [email protected] methods. The thin nozzles and the baseplate can be

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 389
E. Pei et al. (eds.), Springer Handbook of Additive Manufacturing, Springer Handbooks,
https://doi.org/10.1007/978-3-031-20752-5_24
390 S. R. Yarasi et al.

Fig. 24.1 An example of a CAD model of a part and the resulting finished AM part

manufactured in one step rather than having to require a large feedstock materials used in LPBF and EB-PBF; these con-
number of individual welds, for example. siderations are discussed in Sect. 24.3. Section 24.4 discusses
While PBF processes use a broad range of materials post-processing considerations for metal PBF, and to little
including ceramics, polymers, and composites, the focus in surprise, there are also some key differences between LPBF
this chapter will be on metal PBF. There are several technol- and EB-PBF there as well.
ogies that fall within the category of metal PBF such as direct One of the main advantages of PBF and AM in general is
metal laser melting (DMLM), direct metal laser sintering that parts with complex and customized shapes can be
(DMLS), selective laser melting (SLM), selective laser manufactured, such as parts with cavities and unusual geom-
sintering (SLS), and electron beam melting (EBM). These etries. Apart from the general advantages of AM like part
can each be categorized into one of two primary types of count reduction, part consolidation, and the availability of
metal PBF processes: laser powder bed fusion (LPBF), where complex part design, PBF kinematic joints, for example, may
a laser beam is used as the energy source, and electron beam be fabricated that are immediately operational once powder is
powder bed fusion (EB-PBF), where an electron beam is used removed from the joint regions [3]. Additionally, the rapid
as the energy source to fully melt the metal powders.1 LPBF solidification that occurs in PBF processes results in unique
therefore includes both melting and sintering processes; how- microstructures that can give rise to distinct, and sometimes
ever, this chapter will focus primarily on the LPBF processes desirable, properties as compared to cast or wrought compo-
that melt metal powder materials. As such, in both LPBF and nents. Conversely, there are also times where these micro-
EB-PBF, the energy source melts the required area of material structures are problematic. Compared to other forms of AM
in each layer to a depth exceeding the layer thickness. such as metal extrusion, directed energy deposition, or binder
Repeating this over subsequent layers causes a portion of jetting, PBF remains the dominant leader in AM technolo-
the previously solidified layers to remelt, thus ensuring a gies. In 2020, PBF accounted for 54% of the metal AM
well-bonded, high-density part. market [4]. Metal PBF is expensive compared to other
The obvious difference between LPBF and EB-PBF is the metal AM technologies, and it is often limited to the produc-
energy source used to melt the powder, and this is cause for a tion of relatively small parts, but its advantages in producing
list of process deviations. This one change affects every parts with complex geometries and with a variety of possible
downstream decision. The machine parameters discussed in materials make PBF the dominant leader in this technology
Sect. 24.2 are drastically different, which is a direct result of sector.
LPBF using optical absorption as the primary heat transfer There are several steps in the PBF process, apart from the
mode while EB-PBF relies on electrical conduction for general description given so far, that are important to con-
heating. This affects the required beam power, which in sider at this stage. Some of the initial steps taken, before the
turn affects the shape of the melt pool, which then affects machine door is closed, are to load the powder delivery
the range of scanning speeds that can be used, which also systems with enough powder to complete the build and
affects the shape of the melt pool. Beyond the machine place an appropriate build plate upon which the part will be
parameters, there are also requisite differences in the made. Then once the machine door is closed, the atmosphere
may be cleared to suit the material system being printed with
and the machine may be preheated to an appropriate temper-
1
We have used the abbreviations, LPBF and EB-PBF, for the sake of
ature. Preheating the build chamber may include elevating
familiarity of these terms with users. the temperature of the build plate and the area surrounding
24 Powder Bed Fusion 391

the part being built, the powder in the build platform, as well 5. The powder preheating uses infrared heaters in LPBF and
as the powder prior to spreading. This preheating prevents the electron beam in EB-PBF. EB-PBF typically operates
warping in the part due to nonuniform thermal gradients and at a higher temperature than LPBF, which can influence
minimizes energy requirements of the power source [5]. the resultant material properties.
These machine parameters, which include printing parame- 6. The surface finish is generally better in LPBF than in
ters such as those of power, velocity, hatch spacing, and EB-PBF processes, which influences the feature resolu-
thermal parameters such as preheat temperature, are an tion of the part. This is generally due to the use of powder
important set of decisions to be made, which are discussed with a higher average particle size in EB-PBF, though
in Sect. 24.2. Once the build is started, the spreading process many other factors such as processing parameters and
is vital to maintain a uniform powder layer on the build plate beam diameter do come into play. Under optimal condi-
to obtain a defect-free build. This is elaborated further in tions, the surface finish of a part can be coaxed to be the
Sect. 24.3. Post-processing steps that may be necessary best in either of the processes.
such as heat treating or optional surface finishing are also 24
discussed in Sect. 24.4. Section 24.5 deals with the properties
and porosity content that are quantified in the AM parts to 24.1.1 Laser Powder Bed Fusion
ensure they are within the limits of their applications. This
entire process is iterative as parameters that yield the best Laser beams are coherent light that can pass through a gas
properties are considered and selected for further builds. The unaffected as long as the gas is transparent at the laser
design of components for AM PBF processes is not discussed wavelength. Therefore, with the typical wavelength of
in this chapter; however, suggested resources are given at the ≈1 μm delivered by a fiber laser, LPBF processes can be
end. Some challenges and application of the powder bed practiced in an inert gas at atmospheric pressure within the
fusion process are also discussed at the end. build chamber. The laser beam is controlled using mirrors
attached to motors, which limits the speed at which the laser
Laser and Electron Beam PBF beam focal point can move. The powder particles absorb the
Some of the main differences between LPBF and EB-PBF photons in the laser beam and thus melt. Radiative and
are: resistive heaters are used in most LPBF systems to heat the
powder bed. This system is visualized in Fig. 24.2.
1. The energy source used to melt the powder is a laser beam
in LPBF and an electron beam in EB-PBF. Because of
this, the type of materials that can be used in LBPF can be 24.1.2 Electron Beam Powder Bed Fusion
anything that can absorb the laser beam photons, such as
metals, ceramics, and polymers. On the other hand, The electrons in an electron beam travel near the speed of
EB-PBF is restricted to materials that are conductive, light and interact with atoms in the atmosphere of the build
such as metals. chamber. For this reason, EB-PBF processes are conducted in
2. The atmosphere inside the machine is typically a protec- a vacuum environment with a small amount of inert gas
tive inert gas in LPBF but is required to be a vacuum in flowing through the chamber to remove spatter particles,
EB-PBF. An inert gas atmosphere is usually used to min-
imize contamination, prevent oxidation, and carry away
splatter or soot [6]. This can even be nitrogen. Vacuum is Laser
Scanner
necessary in EB-PBF; otherwise, the electron beam would source
be scattered and absorbed by any gas.
3. The energy absorption of the powder bed is absorptivity
limited in LPBF and conductivity limited in EB-PBF Build chamber
Laser beam
owing to the nature of the energy source being used.
4. The scan speeds are limited in LPBF, whereas they can be Recoater
very high in EB-PBF because of the type of mechanism
that drives the energy source: a galvanometer scanner for a Powder
Build
source plate
LPBF machine and magnetic coils for a EB-PBF machine.
For example, the EOS M290 has a maximum scan speed Build
of 7 m/s, whereas the Arcam EBM S12 machine can platform
easily exceed scan speeds of 1000 m/s. The typical scan
speeds that are used, however, depend on the material,
part design, and machine. Fig. 24.2 Laser powder bed fusion process overview
392 S. R. Yarasi et al.

displaced powder particles, byproducts of outgassing, and 3. Thermal parameters such as powder bed temperatures
residual oxygen [7, 8]. The electron beam is focused and 4. Powder parameters such as particle size distribution,
deflected magnetically, which permits very high scan speeds shape, powder packing, layer thickness, and material
because the beam can be moved quasi-instantaneously to properties
another position. As the powder particles absorb the kinetic
energy of the incoming electrons, the increasing negative Our discussion combines the first three as machine param-
charge repels the incoming electrons, which diffuses the eters and treats the powder parameters separately. These
beam. It can also create a powder cloud (“smoke”) caused parameters are highly interdependent and must be chosen
by the expulsion of charged powder particles. Therefore, it is with care. Higher scan speeds may be chosen for faster builds
essential in EB-PBF to have sufficient conductivity in the but can result in lack-of-fusion porosity if melt pool overlap
powder bed to avoid charging, which limits the types of is insufficient [9, 10]. Choosing the right combination of
materials that can be used. Accordingly, the EB-PBF process processing parameters is critical in finding the appropriate
preheats each powder layer by defocusing the electron beam process window to eliminate porosity. This will be explored
and scanning it very rapidly over the powder bed to induce in the following section.
partial sintering of the particles and thus raise the electrical
conductivity. A side effect of defocusing the electron beam is
the increased area of the heat-affected zone, which results in a
24.2 Machine Parameters for PBF
larger minimum resolution and feature size. The typical
EB-PBF machine is visualized in Fig. 24.3.
24.2.1 Process Parameter Mapping
Using optimal process parameters is important for produc-
ing high-density and defect-free parts. The process parame-
In this process, selecting the best set of parameters is a tricky
ters involved in PBF can be divided into four categories:
but crucial step. As discussed earlier, there are several sets of
parameters that affect the build properties: energy beam,
1. Energy beam parameters such as power and spot size
scanning, powder, and thermal parameters, visualized in
2. Scan parameters such as scan speed and hatch spacing
Fig. 24.4. We will discuss the powder properties and methods
to characterize them in the next section. The processing
parameters: (i) energy beam, (ii) scanning, and (iii) thermal
parameters require tuning to find the right spot and directly
Electron
source affect the microstructure, defects, melt pool geometry, etc., of
the build [8, 11]. There is often a trade-off between these
parameters that needs to be balanced to ensure the best
Electromagnetic coils
possible performance such as minimal defects, good surface
roughness, lower build time, and optimal mechanical prop-
erties. For example, while a high-energy beam power with a
high powder bed temperature can produce high-density parts,
it also results in poor recyclability of the powder. Choosing a
E-beam
low-energy beam power with a low powder bed temperature
Powder Powder
can mitigate this but is disadvantageous to the density of the
hopper hopper part. Trying to find the middle ground with a high-energy
beam power and a low powder bed temperature can result in
warping due to nonuniform temperatures [5].

Rake

24.2.2 Energy Parameters

Build
plate Energy parameters include the power of the energy beam and
Build spot size, as seen in Fig. 24.4a. There are other laser param-
Build chamber
platform eters in LPBF that are influential as well, such as the wave-
length and operation mode (pulse or continuous), but power
and spot size are most often varied by users. Higher power
causes higher energy density that can vaporize the material.
Fig. 24.3 Electron beam powder bed fusion process overview This metal vapor is often trapped by the surrounding molten
24 Powder Bed Fusion 393

Fig. 24.4 Illustration of some

key processing parameters in
a)
powder bed fusion: (a) energy
parameters (side view), (b)
scanning parameters (top view),
and (c) powder parameters (side Beam power
view)
Beam spot
size

Preheat
temperature

24
b) Scan speed c)

Powder size and

shape

Scan patterns Powder packing

Hatch spacing

metal after solidification, leading to vapor pores. The energy the time spent melting a particular area in the scan pattern,
density also increases with a decrease in spot size which is the thus decreasing the energy density. This can lead to insuffi-
diameter of the beam on the build plate. Penetration depth, cient melting when the scan speed is quite low, and these
which is the depth to which 99% of the electron beam power unmelted powder volumes cause a lack-of-fusion porosity.
is absorbed by the material [12] or the transmission depth of Similarly, low scan speeds can increase the energy density,
the laser beam into the substrate [13], is also affected by spot causing keyholing. Hatch spacing refers to the distance
size, wherein a larger spot size implies a lower penetration between two successive scan paths, whether that is by the
depth. EB-PBF has an advantage of increased penetration laser or electron beam. Choosing a larger hatch spacing to
depths as electrons penetrate deeper into the metal whereas limit scanning time in a given layer can also lead to insuffi-
the laser beam tends to reflect. The energy density in EB-PBF cient melting between the scan paths. There must be enough
can additionally be changed by changing the speed of the overlap to ensure proper melting.
electron beam since this depends on the applied voltage. This The internal cross-section of the part, minus the contour,
cannot happen in LPBF, however, since the laser beam moves can also be subdivided into smaller, square sections, with
at the constant speed of light [14]. The formula for volumetric each section being rastered individually and random sections
energy density is as follows [15–17]: being processed one after the other. Randomness in the
scanning strategy often prevents any preferential direction
P for residual stresses. Another advantage of this strategy is
E¼
ðv  h  tÞ avoiding balling. By reducing the melt pool track length, the
energy beam returns to a particular spot in a square section
where P is power, v is the scan speed, h is the hatch spacing,
due to overlapping scan lines, which would not have solidi-
and t is the layer thickness.
fied completely. This has the presumed effect of a longer scan
speed and thus a longer dwell time, which prevents balling
24.2.3 Scanning Parameters from occurring. The dwell time here refers to the time
between each energy input across subsequent layers for a
Scan speed, hatch spacing, and scan patterns are some of the particular volume of the part. This includes recoating time
scanning parameters considered in this section (illustrated in as well [18]. Balling and other porosity mechanisms are
Fig. 24.4b). Generally, increasing the scan speed decreases explained in further detail in Sect. 24.5.2.
394 S. R. Yarasi et al.

An additional consideration for EB-PBF processes is that effect on their performance in the additive manufacturing
changing the scan speed or scan strategy may result in a domain [29–32].
change in power. When designing experiments to find the
optimal parameter set, this should be taken into account. This
does not become a problem in LPBF, as the power of the laser 24.3.1 Powder Production
beam is an independent quantity [14].
Metal powders that are used for powder bed fusion additive
manufacturing can be produced either by mechanical
24.2.4 Thermal Parameters methods or chemical methods. Some of the most commonly
used ones are gas or water atomization, plasma rotating
Preheating the build plate is often an excellent way to prevent electrode process (PREP), hydride-dehydride process,
solidification cracking by decreasing the temperature gradi- mechanical milling, and reduction of oxides. The atomization
ent and allowing backfilling to occur. It also prevents warping techniques are more expensive than mechanical milling but
and sometimes removes the need for supports. In EB-PBF generally result in highly spherical powder particles.
especially, preheating, by scanning the defocused electron
beam several times across the powder bed, can prevent Gas/Water Atomization
smoking (the expulsion of powder particles due to charging Atomization processes begin with melting the feedstock alloy
by the electron beam) and thus build failure [12, 19]. By in a furnace, typically vacuum induction melting (VIM)
preheating in EB-PBF processes, the powder particles sinter, furnaces to control interstitial elements. Once melted, the
which stabilizes the powder bed. The preheat temperatures liquid metal enters the atomization chamber below, where it
are generally higher in EB-PBF processes, which has the falls through freely. In the case of water atomization, the
effect of a reduced thermal gradient and longer cooling furnace is separate from the atomization chamber, and a
times, which can result in lower residual stresses as compared tundish is used to allow the liquid metal to flow into the
to those of typical LPBF processes [20]. However, preheating atomizer at a constant rate. High-pressure jets of inert
can increase the overall process time and also result in poor gas/water are used to atomize the liquid metal stream into
recyclability of the powder due to the sintering effects [21]. particles, which solidify as they continue to fall through. In
case of gas atomization, the metal droplets have an increased
solidification time, resulting in more spherical powder parti-
24.3 Powder Parameters for PBF cles due to the lower heat capacity of the gas. The powder is
then collected at the bottom of the atomization chamber,
Powder parameters not only influence powder spreading and following which it may be dried. Gas atomization (GA) is
powder bed packing density, they also have an effect on the typically used for Ti-6Al-4V alloys to minimize contamina-
laser absorption characteristics. For instance, smaller parti- tion, while water atomization is used for iron and steel alloys.
cles provide a greater surface area and can absorb laser Water-atomized powders tend to be more irregular in shape as
energy more efficiently than bigger particles. However, they compared to gas-atomized powders and have higher surface
are harder to spread due to higher cohesive forces [22]. oxygen contents [33, 34]. Currently, GA is the most popular
Typically, powder size ranges between 25–45 μm in LPBF method of producing powders for PBF AM purposes.
and 40–105 μm in EBPBF [23], which is smaller than pow-
ders used in direct energy deposition processes that range Plasma Rotating Electrode Process
between 50 and 150 μm [24]. There are different kinds of Plasma rotating electrode process (PREP) is a centrifugal
powders, with their morphology ranging from perfectly atomization process where a metal bar is melted at the end
spherical (made with highly expensive processes such as while being rotated about its longitudinal axis. This molten
PREP) to highly crunchy (with irregular surface morphol- metal that falls off the end of the bar gets atomized and
ogy), as seen in Figs. 24.4c and 24.6. The importance of solidifies in an inert gas atmosphere. This is an expensive
metal powder characterization in the powder-bed fusion process, however, that generates highly spherical powder
AM processes is becoming increasingly recognized. How particles with low porosity [35].
the powder flows and packs can have a significant impact
on powder bed formation and hence the development of melt Hydride De-Hydride Process
pools and microscopic homogeneity [25]. For instance, a The hydride de-hydride (HDH) process does not involve the
powder bed with an inhomogeneous packing bed density melting of the feedstock. Certain metals become brittle when
has an increased risk of defect formation [26–28]. These exposed to hydrogen; for instance, titanium forms brittle
differences in powder characteristics, apart from surface titanium hydrides in a hydride unit at high temperatures.
chemistry and environmental variables, have a significant These are then crushed, milled, and screened to form smaller
24 Powder Bed Fusion 395

powder particles in the desired size range. This powder is Hall/Carney Flow Test
then heated again to remove the excess hydrogen leaving The Hall/Carney flow test involves using a flowmeter funnel
behind the required powder [34, 36]. to measure the flow rate of a powder sample [38]. The time
required for 50 g of powder to flow through the orifice is
measured and repeated three times to give the result. While
this test is useful to know whether vertical flow of powder is
24.3.2 Powder Characterization
feasible, it does not translate well to determining the spread-
ability of powder systems. It is also highly sensitive to exter-
Understanding both powder flowability and spreadability as
nal conditions such as humidity [39].
well as its size and morphology distributions and the
interconnected relationships between these two types of
Hausner Ratio
properties is crucial for the choice of powder and in the
The Hausner ratio is the ratio between tapped and bulk
design of the powder handling systems. For instance, as
particle size decreases, interparticle friction and electrostatic
density which is a metric that can be used to compare powder 24
flowability between systems [40]. It has been used frequently
forces increase, resulting in loss of flowability. Smaller sized
in AM; however, it is not a direct indicator of powder
powder systems are also more dangerous to handle and can
flowability. The test is highly sensitive to the number of
wreak havoc on components inside the build chamber. How-
taps and conditioning of the powder and does not reflect the
ever, smaller powder particles enable better surface finish,
spreadability behavior in AM [41].
higher accuracy, and thinner layers [5]. These powder prop-
erties directly influence the packing behavior on the build
Rheometer
plate, which affects the final density of the AM part.
The Freeman FT4 rheometer is another device that has been
The distinction between flowability and spreadability is
used in a wide range of industries, from pharmaceutical to
important to make. A good flowing powder may not spread
AM. There are several types of tests that can be done with the
well and vice versa. Simply put, flowability is the ability
rheometer: namely for bulk properties, dynamic properties,
of the powder to flow in a desired manner in a specific
and shear properties [42]. For the purpose of AM, some
piece of equipment [37]. Spreadability is the specific appli-
dynamic metrics and some shear metrics measured by the
cation of flow as it relates to the spread of powder on a
rheometer are more relevant. For the dynamic tests, a cylinder
surface, such as over a build plate. Both flowability and
is packed with powder and a specially designed blade is
spreadability are hard to quantify in one single test and
rotated and moved through the packed cylinder vertically.
multiple test types are advisable.
The torque and force required to move through the powder is
measured to give metrics that describe the flowability of the
Flowability and Spreadability powder. For the shear tests, a specially designed shear cell is
A good powder delivery system and a good flowing powder packed with powder and a shear force is applied on the top
must ensure that the powder spread across the build plate to layer to measure the maximum shear stresses for a given
form a smooth, thin, repeatable layer of powder which does normal stress. The various metrics derived from rheometry
not disturb the previously processed layers [5]. In a powder experiments can give us different levels of understanding
bed fusion additive manufacturing process, powder moves in how powder will behave in AM. It is up to the interpretation
two modes: of the user; however, this too does not directly give us the
spreadability behavior of powder.
1. Powder is transferred from a powder reservoir to the build
plate in front of a recoater blade/roller (typically Revolving/Rotating Drum
vertical flow). The revolving or rotating drum method involves taking
2. Powder is spread over the build plate by the horizontal images of powder that flows in a cylinder with transparent
motion of the recoater blade/roller (horizontal flow). glass sides as it is rotated at various RPM. The powder air
interface is isolated from each image taken at regular inter-
Various flowability metrics used so far in AM studies aim vals and this gives various metrics that describe powder flow.
to mimic these processes to accurately determine a powder Two instruments, the GranuDrum [43] and the Mercury Rev-
feedstock performance. There are several instruments that are olution Powder Analyzer [44], utilize this concept. This
discussed in this section that aim to solve this problem; method may be a better way to understand the performance
however, no single technique can fully characterize the pow- of powders in AM as the tumbling of powder in the cylinder
der flowability properties. Additionally, while it is important is similar to the tumbling of powder ahead of the recoater
to measure the flowability of powders, it is also important to blade or roller [41, 45]. It must be noted that there have been
map it to its spreadability in the AM machine. no comprehensive studies done yet that can indicate cut off
396 S. R. Yarasi et al.

wavelength several times smaller than the diameter of the

Good flow Poor flow powder particle, which diffracts off of powder particles onto
an array of photodetectors. Smaller particles generate more
diffuse diffraction patterns, and larger particles generate high
Processed data

intensity near the laser center. The intensity and radial posi-
tion of the resulting diffraction pattern can be correlated to a
PSD if the refractive indices of both the particle and the
suspension medium are known. This method can be skewed
when irregular powder particles or agglomerated/satellite
powder particles are present, as the diameter of such particles
depends on where the laser hits [50]. Image analysis tech-
Fig. 24.5 Powder-air interface processed through the rotating drum niques involve using SEM images of powder particles. If the
method for good flowing and poor flowing powders pixel size to micrometer ratio is known, the powder particles
can be segmented from the image and an equivalent diameter
values for good flowability of powders; instead, this system found for each powder particle. This technique requires the
has been typically used to compare the flowability of differ- proper dispersion of particles onto the SEM stub to ensure no
ent powder systems, with lower flow angles and a smoother overlapping occurs between particles and enough number of
interface indicating better flowability. For example, in images to ensure the particles being imaged are representa-
Fig. 24.5, the better flowing powder has a lower flow angle tive of the powder feedstock.
and a smoother powder-air interface as compared to the poor When it comes to powder morphology, while ASTM
flowing powder. B243 serves as a baseline for qualitative descriptions of
morphology, there are few quantitative methods to character-
Spreadability Tests ize powder shape. Typically, studies [45, 51, 52] have qual-
There are several instruments being developed by companies, itatively examined powder morphology through SEM images
such as the SpreadStation Powder Analyzer from Mercury and discussed sphericity, relative surface smoothness, and the
Scientific and research labs [46, 47] that aim to mimic the number of satellites. While it may be relatively easier to
process of powder flow and spreading across the build that is describe spherical, elongated, and even satellite particles,
seen in PBF processes. These instruments typically consist of there is difficulty when dealing with the crunchy (with irreg-
a test bed upon which powder is spread with the help of a ular surface texture) particles as seen in Fig. 24.6. Having a
recoater blade/roller. A camera is placed in different loca- spherical powder morphology can result in better flowability
tions, depending on the metric of measurement, to take properties due to lower interparticle friction. Better
images of the powder bed. These images are analyzed to flowability properties often translate to a more uniformly
determine powder spreading quality that can include aspects packed powder bed, which is highly desirable for a good
like uniformity. This area of research is in the nascent stages build with low porosity. However, achieving perfect spherical
of development and needs further validation and morphology is an expensive process and a huge cost addition
standardization. to AM processes. Recent work on hydride-dehydride (HDH)
powders provides evidence that irregular powders are indeed
Morphology usable in powder bed machines, contrary to popularly held
There are several aspects of morphology that are pertinent to opinion [23, 53, 54]. Ultimately, there is no conclusive state-
PBF AM. Particle size distribution (PSD) is one of them and ment that can be made on how exactly morphology influ-
an oft-analyzed one. A number of techniques exist to measure ences the build process and which morphology/size
the PSD of powder systems: sieving, laser scattering, and distribution is the best.
image analysis are the most commonly used methods in AM
[48, 49]. d10, d50, and d90 are the most frequently used
metrics by powder manufacturers and researchers alike. 24.3.3 Powder Recycling
These refer to the tenth, fiftieth, and ninetieth percentiles of
the PSDs. The volume of powder in the PBF machine is much greater
Sieving involves passing powder through a series of than the volume of powder required for the build, which
sieves of varying mesh sizes, which allows powder particles makes recycling powder an attractive option. The quantity
to be separated into different size ranges. This method pro- of powder that can be recycled viably depends on the condi-
duces a rough size range as irregularly shaped particles and tion of the powder. The properties of powders, both physical
agglomerated particles can slip through smaller meshes [49]. and chemical, can change during the course of the build. The
Laser scattering involves a laser of a characteristic extent of this change determines the recyclability of the
24 Powder Bed Fusion 397

0 24.6 µm
100 µm 1009 x

Elongated Satellited

24
Spherical
Crunchy

Fig. 24.6 Examples of different powder morphologies: spherical, elongated, and with satellites in SEM images

powder systems. The elevated temperatures in the build

chamber, especially that surrounding the part, can cause 24.4 Post-processing Techniques
powder particles to fuse, for instance, which changes the
size and morphology distribution of the powders, ultimately 24.4.1 Support Structure Removal
affecting the spreadability. The atmosphere can also change
the surface composition of the powder particles, which, when Support removal is an important post-processing step. For
fused to form the part, affect the resulting mechanical prop- metal PBF, complex builds often require support structures to
erties. The oxygen content in the powder can increase, which prevent warping during the build process. They not only
may affect the mechanical properties of the build, especially physically support the parts but also act as conductors to
for titanium alloys [55]. Other reactive agents such as nitro- draw heat away during the process. This reduces the resid-
gen or carbon may also form oxide layers on the particles ual stresses in the part which in turn positively affects the
[56]. The most common approach to recycling used powders resultant mechanical properties. Small and internal features
in PBF processes is to mix them with a certain ratio of unused do not always need support structures, but external features
powder. Overflow powder and part-bed powder are typically such as large overhangs, such as that seen in Fig. 24.7,
handled separately, as they experience different temperature cannot be sufficiently supported by the powder bed alone.
profiles during the build. Thorough mixing of recycled and Support structures are built with the part but must be
virgin powder is necessary to maintain homogeneity of the removed following the build process. Depending on the
powder properties. Sieving is typically done to prevent large part geometry, the support structures may be removed by
agglomerates or very small particles from changing the grinding, milling, or cutting. These are often manual pro-
behavior of the powder. However, there will come a point cesses, which increase the production cost for the part and
when repeated recycling renders some powder in becoming also affect the surface finish. There are different types of
unusable. Studies need to be done to evaluate how many support structures that exist such as grid, lattice, and block
recycling cycles can be done while keeping part properties supports. Optimization is done to minimize the resources
at their desired levels [23]. Since the temperatures in EB-PBF spent in building and removing supports while maintaining
are typically higher and sintering is encouraged to some adequate support to the part [58]. In order to minimize post-
extent to prevent smoking, the unused powder often forms processing time, the contact area between the supports and
agglomerates in a process known as caking, which affects the the part is kept to a minimum. This is additionally important
recyclability of powders specifically in these processes [57]. to ensure good surface quality of the part. In some cases, the
398 S. R. Yarasi et al.

part geometries, the surface finish can be modified by grind-

ing and polishing. However, some parts may benefit from
higher surface roughness, so those parts may be used with
minimal post-processing.
Part Support
structure
24.5 Part Properties and Characterization

24.5.1 Mechanical Properties

Support Build plate
structure Typically, the mechanical properties of AM parts that are
important in their applications are metrics like yield, ultimate
tensile strength, ductility, creep, and fatigue. These metrics
are often compared with the cast or wrought material prop-
Fig. 24.7 An example of support structures for overhang erties. The first step in part characterization is to build a set of
standard designs at various machine parameter combinations
geometry and spacing of the supports can lead to porosity in with which mechanical properties and porosity information
the part [59–61]. can be obtained through a series of tests. This allows the user
to determine which parameter combinations result in the best
part with minimal porosity and optimal performance.
24.4.2 Improving Mechanical Properties Depending on the application, some mechanical properties
may be of higher priority than others and the design of
Most post-processing techniques are designed to enhance experiments should revolve around those properties. Once
mechanical properties of AM parts, such as machining, heat parameters have been selected, prototypes of the actual part
treating, among others. Heat treating is most commonly used may be built and tested once more to determine porosity,
to address issues of residual stress and porosity, which microstructure, and part performance. There may be a need
improve mechanical properties. Often, an optimal heat to extrapolate data from available literature and experiments
treating process has to balance between two conflicting con- to determine how the part will behave at the service condi-
ditions. For instance, heat treatment can reduce residual tions. This is an iterative process, to narrow down the optimal
stresses but impact the mechanical strength negatively. A process window, which involves redesigning parts and
typical heat treatment method for LPBF processes is to heat balancing the trade-off between parameters. Once optimal
the parts in an inert atmosphere for a period of time to remove parameters have been selected, there may be a need for
residual stress. Other heat treatment strategies can improve further redesign of the end product to improve efficiency.
properties like fracture toughness and reduce the overall The advantages of additive manufacturing come into play
porosity [62]. Post-processing such as hot isostatic pressing here in this rapid prototyping and optimization stage. It
(HIPing) can reduce pores but can age the material. There are should be noted that microstructures seen in AM built parts
also different finishing techniques to improve the surface are fundamentally different from those of cast or wrought
finish of AM parts. In most cases, a smooth surface finish is parts, even when the exact same design is built. This results in
desired to prevent stress concentrations from occurring, different mechanical properties of AM parts, which may be
which detrimentally affect the fatigue life of a part. When better or worse than those of conventionally made parts [8].
investigating surface roughness, upskin, downskin, and ver- The impact of factors like heat treating, HIP, orientation on
tical surfaces are most encountered. These are the upward tensile, fracture toughness, and other mechanical properties
facing, downward facing, and vertical surfaces of a part being has been published in a comprehensive review [65]. Some
built on a build plate. While upskin surface roughness significant differences in microstructure and properties
depends on the melt pool size, the downskin surface tends between traditionally manufactured materials and PBF mate-
to be influenced by the depth of the melt pool. Using a rials have been reviewed for aluminum alloys [66], titanium
contour scanning strategy, in which the outline of the build alloys, stainless steels [20, 67], and nickel-based superalloys
is scanned with the energy beam, can improve the surface [68]. This occurs due to the repeated layer-by-layer melting
finish of the build [63, 64]. The rest of the build in this case is process which causes grains to solidify in an epitaxial manner
filled in with a rastering strategy. Powder selection is an that branch out in columnar grains. However, part micro-
important factor, especially with the contour scanning strat- structure does have the potential to be controlled locally
egy; smaller powder particles tend to result in higher quality through experimentation with power, scan speed, and scan
surface finishes and improve the build resolution. For simple strategy [69, 70].
24 Powder Bed Fusion 399

24.5.2 Porosity sectioned AM part and look at the cross section under a
microscope. This can be done at various heights to get a
Porosity refers to the absence of material in a solid and can be better understanding of the distribution of porosity in the
present both in the powder (due to entrapped gas during part, as well as size and shape information. Another way is
production) and the AM part [71]. The powder porosity can the Archimedes density method, which is largely accurate in
be transferred into the build if there is not enough time for the determining the percentage of porosity in the part [76]. A
pores to escape from the melt pool [72, 73]. There are two nondestructive method to characterize powder porosity and
mechanisms in which porosity is induced during the build in part porosity is micro-X-ray computed tomography (μXCT).
a PBF machine, due to keyholing or insufficient melting [74]. This technique uses X-ray to illuminate either the part or a
Keyhole porosity typically occurs when high-energy concen- sample of powder from different directions and then recon-
tration in the melt pool causes evaporation of metal from the structs the object in the form of 3D image slices [39]. These
surface and a deep, unstable keyhole leading to trapped pores image slices show the 2D cross-sections of the part or powder
that are spherical in shape. In contrast, lack-of-fusion poros- sample, complete with inner porosity data. 24
ity is caused due to insufficient melting between the layers
and is irregular in shape. Characterizing and quantifying the
amount of porosity in AM parts is an important area of 24.6 PBF in Practice
research. It is critical that parts used in high stress or fatigue
applications are fully dense, so as to prevent part failure 24.6.1 Applications
during service. In contrast, some amount of porosity, espe-
cially on the surface, is intentionally built into the parts such The advantages of AM lend itself to complex, critical, and
as biomedical implants where pores promote high-value applications. The ability to create complex parts
osseointegration with the bones [8, 75]. with high resolution gives it an edge over other AM technol-
We have mentioned keyholing and lack of fusion porosity ogies like direct energy deposition (DED). The aerospace and
as two of the regions surrounding the optimal P-V area. medical industries are areas where PBF AM has made a mark
Another constraint surrounding this region is the phenome- in producing innovative parts with consistent and predictable
non of balling. This may occur when high scan speeds are properties. Commonly, the aerospace industry uses alumi-
combined with high-energy beam power, which is a critical num, titanium, stainless steel, nickel-based superalloys,
limitation in maximizing production rate. The surface tension among other metal alloys in AM applications. This is due to
forces of the molten metal on the powder bed cause it to reasons such as high strength-to-weight ratio, lightweight of
minimize its surface area by balling up. These three phenom- aluminum alloys, corrosion resistance of titanium alloys, etc.
enon are visualized in the P-V space in Fig. 24.8, where the [77]. The medical industry has typically favored titanium
scan speed is equivalent to the velocity. alloys due to the biocompatibility and non-allergenic nature
There are various ways to characterize porosity in powder of titanium [78]. PBF has been used to make turbine blades
and AM parts. A simple, inexpensive way is to mount the with IN718 [79], fuel nozzles in airplanes [80], heat
exchangers with AlSi10Mg alloy, for example [81], dental
implants [82], and orthopedic implants with titanium alloys
[75]. These industries are ripe with potential for high-value
part deployment, which is driving significant interest in PBF
Keyholing porosity AM technologies. The design flexibility with PBF AM is a
Ba

distinct advantage over conventional manufacturing. The

l
lin
g

layer-by-layer build process affords opportunities to build

ph
en
om

lighter weight components with lattice internal structures

Power (W)

en
on

[83], integrated systems in single parts that remove assembly

Process window process requirements and traditional joining methods [77],
on-demand manufacturing that does not require retooling of
conventional manufacturing sites [84], and generally reduces
Lack-of-fusion porosity the human, time, and material resources to produce a
part [85].
The aerospace and medical device industries both require
high-value parts that meet strict performance criteria. In many
Velocity (mm/s)
cases, these parts have severe consequences in the event of part
failure. These consequences could be considered in measures
Fig. 24.8 Process window in PBF and its surrounding porosity zones of financial loss or endangerment to livelihood. In any case,
400 S. R. Yarasi et al.

certification is an essential process in PBF manufacturing, where a low volume of parts are created and tested. In these
which is further discussed in Sect. 24.6.2. Assuming the cases, printing a small number of parts layer by layer is faster
certification process can be managed well, these industries than retooling an entire production line every time a slight
show great promise for the integration of PBF. For instance, design change is desired. However, in production where
aviation flight systems such as airplanes often include mil- many parts need to be manufactured, the volume build rate
lions of parts. These parts are assembled into complex sub- limitations of the layer-by-layer process are detrimental. This
systems where each plays a vital role in the success of the is especially apparent in industries that rely on serial produc-
aircraft. PBF has potential impact as a means of designing tion schemes such as in automotive manufacturing. One
higher complexity parts – what might have once been many strategy to overcome such a challenge is to optimize the
parts assembled into a subsystem can now be produced all build parameters discussed in Sect. 24.2 to maximize the
at once in a PBF system component. This is appealing for build rate while remaining within the favorable process win-
the reasons previously discussed: lighter weight systems, dow. Even with process optimization, volume build rates for
avoidance of traditional part joinery, reduction of human most commercial systems are presently limited to around
labor time and cost, and perhaps most notably, the oppor- 20–30 cm3/h for LPBF [87] and up to 80 cm3/h for
tunity for on-demand manufacturing. Because PBF produc- EB-PBF [88]. Some newer LPBF systems have innovations
tion only requires feedstock material, the manufacturing that expand build rate capabilities to 40–60 cm3/h for
hardware (e.g., the PBF machine), and computer software, medium-size machines, and for exceptionally large
parts can be produced only when they are required. There machines, this can be further extended to 100–180 cm3/h
may be a need for higher volumes of parts or system com- [86]. The major innovation that enables these comparatively
ponents at the start of a production run, but over time, the high build rates is the use of multiple lasers for heating the
need to produce individual replacement parts may be lim- metal powder feedstock. However, this added complexity to
ited and temporally distributed. This is a major advantage of the PBF process introduces challenges of its own such as
PBF; replacement parts do not need to be stored in high laser interactions with nearby plumes caused by other beams
volumes in warehouses and can be more easily or excess heating induced by adjacent melt pools [89].
manufactured close to local demand. The slow speed of manufacturing with PBF compounds
The medical industry is familiar with the types of chal- with other factors such as powder production that further
lenges faced by the aerospace industry, and PBF similarly increase the cost of PBF AM. As was discussed in Sect.
fills a critical gap to address these problems. Similar to flight 24.3, GA is currently the dominant method for producing
system hardware, structural biomedical implants must also be metal powder feedstock materials for PBF. However, the
lightweight and strong. Further, these materials must also be yield for GA powders is relatively low; it is typical for around
biocompatible, meaning that the material will not be harmful 10–20% of powder from a GA batch to be suitable for PBF
to existing body tissue. Generally, this results in the choice of applications [90]. The labor required to separate the low yield
titanium alloys, which tend to be biocompatible while of useful powder from the entire volume of atomized powder
maintaining the strength and weight constraints for patient further increases the cost of PBF. A current strategy to reduce
application [78]. From a practical perspective, biomedical the feedstock material cost is to recycle and reuse unmelted
components with human applications are also well-aligned powder; however, there are some morphological transforma-
with the current size limitations of PBF manufacturing; mod- tions that occur to heat-affected powder, and it is not conclu-
erately sized LPBF and EB-PBF machines tend to have sizes sively understood how recycled powders interact with other
around 200 mm  200 mm  200 mm to 400 mm  build parameters that affect the final part properties. Another
400 mm  400 mm [86]. strategy to reduce the powder cost is to use less expensive
feedstocks such as water-atomized powders or HDH powders
as discussed in Sect. 24.3.1. Further, PBF inherently requires
24.6.2 Challenges the handling of metal powder materials, which necessitates
the labor of a more skilled and trained workforce. Finer
Although the design freedom of PBF AM is quite attractive, powders are preferred for better build resolution, but these
there are a number of challenges to overcome first. The layer- are challenging to handle and present new safety concerns
by-layer build process involved in PBF is inherently slow. such as explosions from rapid oxidation. The powder itself is
Most commercial PBF systems feature a single beam that also problematic after the build process because the solidified
selectively melts the powder feedstock material. Multiple part will still be immersed in the powder bed. The part must
parts can be built simultaneously so long as they all fit within be removed from the build plate and cleared of excess pow-
the build chamber, but the build rate still suffers compared to der. Although there are some efforts to build robotic systems
traditional metal production methods such as casting. This is to automate this process, it is currently most common for this
less of a concern in engineering design and development to be performed by a technician, which presents a severe dust
24 Powder Bed Fusion 401

exposure risk. This challenge must be addressed with a properties. That is, the same design for the same part built
thorough safety protocol. on the same machine using the same powder and same set of
Though there are many materials suitable for PBF process parameters do not always result in the same proper-
machines, challenges arise from the process requirements ties or performance. The reason this happens is not well
and further, the certification of the processes for a given understood, though there are some hypotheses that query
material. To start with, materials used in these machines the inherent uncertainty within the process. It is known that
should be able to be converted into powder form, melt with- changing the process parameters can affect the porosity of a
out crack formation concerns, and endure the repeated melt- part as discussed in Sect. 24.5. Aside from porosity, one
ing process. The performance of the PBF process is enhanced might also consider the uniquely different microstructures
when we use materials that are tailored to the specific PBF parts have compared to their conventionally
demands of the AM process. This tailoring process is com- manufactured counterparts. There are hundreds of parameters
putationally expensive, as even small linear changes to the to control in PBF including everything from feedstock selec-
composition of elements in an alloy can result in drastic tion to processing and post-processing methods and even 24
nonlinear jumps in behavior. Furthermore, there are specific environmental variables. There are certainly many ongoing
concerns with different material types: for instance, alumi- efforts to better understand, control, and optimize PBF, but
num alloys pose solidification cracking concerns, titanium for now, part-to-part variation remains a significant challenge
alloys exhibit columnar growth that can be detrimental to inhibiting wide-scale adoption of this technology. Neverthe-
mechanical properties, and unstable phases that limit perfor- less, there are some cases where these processes are well
mance arise in nickel alloys [91]. The challenges inherent to studied and commercially available PBF parts succeed.
the processing parameters are present in all levels of PBF use,
from laboratory scale basic scientific research to applications
in industrial manufacturing. The questions that arise from the 24.7 Summary
metallurgical challenges in PBF present a major barrier for
PBF certification; regulatory bodies such as the Federal Avi- PBF AM as a topic was introduced along with the two
ation Administration in the United States will need to suffi- primary modes: LPBF and EB-PBF. Machine parameters
ciently understand PBF processes such that parts intended for used in PBF such as beam power and spot size, scan speed,
any application can be inspected to reveal potential safety hatch spacing, and thermal parameters were discussed. These
hazards. Traditional manufacturing methodologies benefit parameters were introduced, but the optimization of these is
from a long history of other standards for certification, but presently a required task for particular parts and designs.
PBF is comparatively young and does not have a long record Powder selection is described in greater detail because it
of certification. Further complicating the certification process has such a significant influence on the rest of the build
is the rapid pace of innovations being made to the PBF process and has post-processing implications. Characteriza-
process space; it is not well understood how every new tion of the PBF AM process scope is discussed including
feature might balance with the multitude of other process powder characterization and porosity; however, there are
variables. certainly more details to consider such as the microstructure
Capital investment for current AM machines is on the of as-built parts, in-situ build monitoring for defect detection,
order of several hundreds of thousands of dollars, and oper- and application-specific performance testing for the intended
ation and maintenance expenses are significant and recurring. parts. Finally, this chapter concludes with a brief discussion
Many PBF AM machines have a suite of features including of current applications for PBF AM as well as the challenges
preheating build plates, inert gas flow, and vacuum pumps. that must be overcome to adopt the technology.
Each feature has an associated cost and energy requirement in PBF is a complex manufacturing process with scores of
addition to the requisite cost of operating a high power laser processing variables, many of which are interdependent for
or electron beam. For instance, in an effort to increase AM optimization. Design considerations begin at the geometric
productivity, a user might set the laser power to 300 W with a part design. PBF affords the user a high degree of flexibility
preheat of 200  C. For small quantities of parts, this may be to incorporate functional complexity into parts; systems of
insignificant, but the energy requirements for large-scale assembled parts made with traditional manufacturing prac-
production can be disadvantageous compared to other tices can sometimes be consolidated into a single PBF com-
manufacturing methods. ponent. This offers significant advantages for weight
Aside from slow rates of production and high investment reduction and custom-specific purpose solutions. However,
cost, there still exists part-to-part variation in the build pro- for the many opportunities PBF offers, there are perhaps
cesses. Section 24.4 discusses post-processing methods that equally many challenges to overcome. Careful consideration
improve mechanical properties of PBF parts, but the concern must be made in choosing the specific type of PBF; the metal
should be focused on the high variation of part-to-part feedstock material; the PBF machine; process parameters
402 S. R. Yarasi et al.

such as power, velocity, hatch spacing, scan rotations, and sciencedirect.com/science/article/pii/S2214860420309246. ISSN:
preheat temperature, among others; post-process procedures 2214-8604
11. Beuth, J., et al.: Process mapping for qualification across multiple
including part, support and powder removal, heat treatments, direct metal additive manufacturing processes. In: 24th International
and surface finishing. Additionally, all characterization meth- SFF Symposium – An Additive Manufacturing Conference, SFF 2013,
odologies must be carefully constructed to assess the quality pp. 655–665. https://www.scopus.com/inward/record.uri?eid¼2-s2.0-
of the parts produced. Some characterization methods are 8 4 8 9 8 4 7 0 8 5 1 & p a r t n e r I D ¼4 0 & m d 5 ¼f 0 0 6 d 5 d f 1 f a =
420ac078b7619a54b7f5f (2013)
destructive and are inherently not useful for qualification 12. Zäh, M.F., et al.: Determination of process parameters for electron
and certification of parts intended for use. So nondestructive beam sintering (EBS). In: COMSOL Conference 2008,
evaluation techniques must also be investigated to correlate pp. 12–18 (2008)
to quantities of interest that are generally destructively eval- 13. Kim, J., et al.: Calibration of laser penetration depth and absorptivity
in finite element method based modeling of powder bed fusion melt
uated such as tensile or compressive strength, internal poros- pools. Metals and Materials International. 26(6), 891–902 (2020).
ity, fatigue life performance, among others. It is clear that https://doi.org/10.1007/s12540-019-00599-3. ISSN: 20054149
there are countless industrial applications in a wide breadth of 14. Kumar, S.: Additive Manufacturing Processes, pp. 1–205 (2020).
industries, including aerospace, automotive, biomedical, and https://doi.org/10.1007/978-3-030-45089-2. ISBN: 9783030450892
15. Caiazzo, F., Alfieri, V., Casalino, G.: On the relevance of volumetric
others, where low volumes of parts with specialized designs energy density in the investigation of inconel 718 laser powder bed
might be of significant value. Additionally, given all of the fusion. Materials. 13(3) (2020). https://doi.org/10.3390/
challenges ahead of the emerging PBF industry, it is apparent ma13030538. ISSN: 19961944
that there will continue to be research and development 16. Brown, C.U., et al.: The effects of laser powder bed fusion process
parameters on material hardness and density for nickel alloy 625.
interest across both academic and industrial institutions. NIST Advanced Manufacturing Series, pp. 100–119. https://doi.org/
10.6028/NIST.AMS.100-19 (2018)
17. Kahlert, M., et al.: Influence of microstructure and defects on
References mechanical properties of AISI H13 manufactured by electron beam
powder bed fusion. J. Mater. Eng. Perform. 30(9),
6895–6904 (2021). https://doi.org/10.1007/s11665-021-06059-7.
1. King, W.E., et al.: Laser powder bed fusion additive manufacturing
ISSN: 15441024
of metals; physics, computational, and materials challenges. Appl.
18. Mohr, G., Altenburg, S.J., Hilgenberg, K.: Effects of inter layer time
Phys. Rev. 2(4), 041304 (2015). https://doi.org/10.1063/1.4937809.
and build height on resulting properties of 316L stainless steel
http://aip.scitation.org/doi/10.1063/1.4937809. ISSN: 1931-9401
processed by laser powder bed fusion. Addit. Manuf. 32(July
2. Frazier, W.E.: Metal additive manufacturing: a review. J. Mater. Eng.
2019), 101080 (2020). https://doi.org/10.1016/j.addma.2020.
Perform. 23(6), 1917–1928 (2014). https://doi.org/10.1007/s11665-
101080. ISSN: 22148604
014-0958-z. ISSN: 1544-1024
19. Zachary C. Cordero, et al.: Powder bed charging during electron-
3. Mavroidis, C., et al.: Fabrication of non-assembly mechanisms and
beam additive manufacturing. Acta Mater. 124, 437–445 (2017).
robotic systems using rapid prototyping. J. Mech. Des. (1990).
https://doi.org/10.1016/j.actamat.2016.11.012. ISSN:13596454
123(4), 516–524 (2001) ISSN: 1050-0472
20. Wysocki, B., et al.: Laser and electron beam additive manufacturing
4. Vafadar, A., et al.: Advances in metal additive manufacturing: a
methods of fabricating titanium bone implants. Appl. Sci. (Switzer-
review of common processes, industrial applications, and current
land). 7(7), 1–20 (2017). https://doi.org/10.3390/app7070657.
challenges. Appl. Sci. 11(3), 1213 (2021)
ISSN: 20763417
5. Gibson, I., Rosen, D.W., Stucker, B.: Powder bed fusion processes.
21. Sigl, M., Lutzmann, S., Zaeh, M.F.: Transient physical effects in
In: Additive Manufacturing Technologies: Rapid Prototyping to
electron beam sintering. In: 17th Solid Freeform Fabrication Sym-
Direct Digital Manufacturing, pp. 120–159. Springer US, Boston
posium, SFF 2006, pp. 464–477 (2006)
(2010). https://doi.org/10.1007/978-1-4419-1120-9_5. ISBN:
22. German, R.M.: Powder Metallurgy Science. Metal Powder Industry
978-1-4419-1120-9
(1994) ISBN: 978-1878954428
6. Wirth, F., et al.: Influence of the Inert Gas Flow on the Laser Powder
23. Vock, S., et al.: Powders for Powder Bed Fusion: A Review. Pro-
Bed Fusion (LPBF) Process, pp. 192–204. Springer International
gress Addit. Manuf. 4, 383–397 (2019). https://doi.org/10.1007/
Publishing (2021). https://doi.org/10.1007/978-3-030-54334-1.
s40964-019-00078-6
ISBN: 9783030543341
24. Iams, A.D., et al.: Influence of particle size on powder rheology and
7. Körner, C.: Additive manufacturing of metallic components by
effects on mass flow during directed energy deposition additive
selective electron beam melting – a review. Int. Mater. Rev. 61(5),
manufacturing. Powder Technol. 396, 316–326 (2022). https://doi.
361–377 (2016). https://doi.org/10.1080/09506608.2016.1176289.
org/10.1016/j.powtec.2021.10.059
ISSN: 17432804
25. Qian, M.: Metal powder for additive manufacturing. JOM. 67(3),
8. Gong, X., Anderson, T., Chou, K.: Review on powder-based electron
536–537 (2015). https://doi.org/10.1007/s11837-015-1321-z. http://
beam additive manufacturing technology. Manuf. Rev. 1 (2014).
link.springer.com/10.1007/s11837-015-1321-z, ISSN: 1047-4838
https://doi.org/10.1051/mfreview/2014001. ISSN: 22654224
26. Simchi, A.: The role of particle size on the laser sintering of iron
9. Ming, T., Chris Pistorius, P., Beuth, J.L.: Prediction of lack-of-fusion
powder. Metall. Mater. Trans. B Process Metall. Mater. Process. Sci.
porosity for powder bed fusion. Addit. Manuf. 14, 39–48 (2017).
35(5), 937–948 (2004). https://doi.org/10.1007/s11663-004-0088-3.
https://doi.org/10.1016/j.addma.2016.12.001. https://www.
ISSN: 10735615
sciencedirect.com/science/article/pii/S2214860416300471. ISSN:
27. Parteli, E.J.R., Pöschel, T.: Particle-based simulation of powder
2214-8604
application in additive manufacturing. Powder Technol. 288,
10. Gordon, J.V., et al.: Defect structure process maps for laser powder
96–102 (2016). https://doi.org/10.1016/j.powtec.2015.10.035.
bed fusion additive manufacturing. Addit. Manuf. 36, 101552 (2020).
ISSN: 1873328X
https://doi.org/10.1016/j.addma.2020.101552. http://www.
24 Powder Bed Fusion 403

28. Boley, C.D., Khairallah, S.A., Rubenchik, A.M.: Calculation of laser 43. Granudrum. Dynamic Angle of Repose | Understanding and Improv-
absorption by metal powders in additive manufacturing. In: Additive ing Powders Spreadability. Visited on 08/08/2019
Manufacturing Handbook: Product Development for the Defense 44. Zocca, A., et al.: Powder-bed stabilization for powder-based additive
Industry 54(9), pp. 507–517. https://doi.org/10.1201/ manufacturing. Adv. Mech. Eng. 2014, 491581 (2014). https://doi.
9781315119106 (2017) org/10.1155/2014/491581. ISSN:16878140
29. Strondl, A., et al.: Characterization and control of powder properties 45. Muñiz-Lerma, J.A., et al.: A comprehensive approach to powder
for additive manufacturing. JOM. 67(3), 549–554 (2015). https:// feedstock characterization for powder bed fusion additive manufactur-
doi.org/10.1007/s11837-015-1304-0. ISSN: 15431851 ing: a case study on AlSi7Mg. Materials. 11(12) (2018). https://doi.
30. Slotwinski, J.A., Garboczi, E.J.: Metrology needs for metal additive org/10.3390/ma11122386. ISSN: 19961944
manufacturing powders. JOM. 67(3), 538–543 (2015). https://doi. 46. Lee, Y., Simunovic, S., Gurnon, K.A.. Quantification of Powder
org/10.1007/s11837-014-1290-7. ISSN: 1543-1851 Spreading Process for Metal Additive. 2019. ISBN: 1800553684
31. Kappes, B., et al.: Machine learning to optimize additive 47. Oropeza, D., Roberts, R., Hart, A.J.: A modular testbed for mecha-
manufacturing parameters for laser powder bed fusion of inconel nized spreading of powder layers for additive manufacturing.
718. In: Proceedings of the 9th International Symposium on Super- Review of Scientific Instruments. 92(1) (2021). https://doi.org/10.
alloy 718 & Derivatives: Energy, Aerospace, and Industrial Appli- 1063/5.0031191. ISSN: 10897623
cations. Springer International Publishing (2018). https://doi.org/10. 48. Snow, Z.: Metal powder production and powder size and shape 24
1007/978-3-319-89480-5. http://link.springer.com/10.1007/978-3- distribution. Addit. Manuf. Processes. 24, 167–171 (2020). https://
319-89480-5. ISBN: 978-3-319-89480-5 doi.org/10.31399/asm.hb.v24.a0006567
32. Slotwinski, J.A., et al.: Characterization of metal powders used for 49. Schade, C.: Introduction to Metal Powder Production and Charac-
additive manufacturing. J. Res. Natl. Inst. Stand. Technol. 119, terization[1]. Samal, P., Newkirk J. (eds). https://doi.org/10.31399/
460–493 (2014). https://doi.org/10.6028/jres.119.018. https:// asm.hb.v07.a0006086 (2015)
pubmed.ncbi.nlm.nih.gov/26601040%20, https://www.ncbi.nlm. 50. Iacocca, R.G.: Particle Size and Size Distribution. Samal, P.,
nih.gov/pmc/articles/PMC4487284/. ISSN: 1044-677X Newkirk J. (eds.). https://doi.org/10.31399/asm.hb.v07.a0006096
33. Schade, C., Hoeganaes, G.K.N., Dunkley, J.J.: Atomization[1]. (2015)
Powder Metall. 7, 58–71 (2018). https://doi.org/10.31399/asm.hb. 51. Susan, D.: Stereological analysis of spherical particles: experimental
v07.a0006084 assessment and comparison to laser diffraction. Metall. Mater.
34. Dawes, J., Bowerman, R., Trepleton, R.: Introduction to the additive Trans. A Phys. Metall. Mater. Sci. 36(9), 2481–2492 (2005).
manufacturing powder metallurgy supply chain. Johnson Matthey https://doi.org/10.1007/s11661-005-0122-3. ISSN: 10735623
Technol. Rev. 59(3), 243–256 (2015). https://doi.org/10.1595/ 52. Bo, H.: Particle image analysis. In: Samal, P., Newkirk, J. (eds.)
205651315X688686. ISSN: 20565135 Powder Metallurgy ASM Handbook, vol. 7, pp. 154–155 (2015).
35. Neikov, O.D., Gopienko, V.G.: Chapter 18 – Production of Titanium https://doi.org/10.31399/asm.hb.v07.a0006102
and Titanium Alloy Powders. In: Neikov, O.D., Naboychenko, S.S., 53. Wu, Z., et al.: Study of Printability and Porosity Formation in Laser
Nikolay, A.B.T. (eds.) Handbook of Non-Ferrous Metal Powders Powder Bed Fusion Built Hydride-Dehydride (HDH) Ti-6Al-4V.
(Second Edition) Yefimov, pp. 549–570. Elsevier, Oxford (2019). Addit. Manuf.. in press. 47, 102323 (2021)
https://doi.org/10.1016/B978-0-08-100543-9.00018-X. https://www. 54. Narra, S.P., et al.: Use of non-spherical hydride-dehydride (HDH)
sciencedirect.com/science/article/pii/B978008100543900018X. powder in powder bed fusion additive manufacturing. Addit. Manuf.
ISBN: 978-0-08-100543-9 34, 101188 (2020). https://doi.org/10.1016/j.addma.2020.101188.
36. Barbis, D.P., et al.: Titanium powders from the hydride-dehydride h t t p s : / / w w w. s c i e n c e d i r e c t . c o m / s c i e n c e / a r t i c l e / p i i /
process. Titanium Powder Metall. Sci. Technol. Appl. 2015, S2214860420305601
101–116 (2015). https://doi.org/10.1016/B978-0-12-800054-0. 55. Tang, H.P., et al.: Effect of powder reuse times on additive
00007-1 manufacturing of Ti-6Al-4V by selective electron beam melting.
37. Prescott, J.K., Barnum, R.A.: On Powder Flowability. October JOM. 67(3), 555–563 (2015). https://doi.org/10.1007/s11837-015-
(2000) 1300-4. ISSN: 15431851
38. B964. Standard Test Methods for Flow Rate of Metal Powders 56. Powell, D., et al.: Understanding powder degradation in metal
Using the Carney Funnel. ASTM Standard, pp. 4–6. https://doi. additive manufacturing to allow the upcycling of recycled powders.
org/10.1520/B0964-16.2 (2016) J. Clean. Prod. 268, 122077 (2020). https://doi.org/10.1016/j.
39. Wu, Z., et al.: Powder Characterization for Metal Additive jclepro.2020.122077. ISSN: 09596526
Manufacturing. David L. Bourell et al. (eds.) https://doi.org/10. 57. Gruber, H., et al.: Effect of powder recycling in electron beam
31399/asm.hb.v24.a0006568 (2020) melting on the surface chemistry of alloy 718 powder. Metall.
40. ASTM International F3049. Standard guide for characterizing prop- Mater. Trans. A Phys. Metall. Mater. Sci. 50(9),
erties of metal powders used for additive manufacturing processes. 4410–4422 (2019). https://doi.org/10.1007/s11661-019-05333-7
In: F3049–14, pp. 1–3. https://doi.org/10.1520/F3049-14. file:///D:/ 58. Jiang, J., Xu, X., Stringer, J.: Support structures for additive
Sorted%20Papers/ASTM%20-%202014%20-%20Standard% manufacturing: a review. J. Manuf. Mater. Process. 2(4), 64 (2018)
20Guide%20for%20Characterizing%20Properties%20of%20Metal 59. Kurzynowski, T., et al.: Effect of scanning and support strategies on
%20Powders%20Used%20for%20Additive%20Manufacturing% relative density of SLM-ed H13 steel in relation to specimen size.
20Processes.pdf (2014) Materials. 12(2) (2019). https://doi.org/10.3390/ma12020239.
41. Spierings, A.B., et al.: Powder flowability characterisation method- ISSN: 19961944
ology for powder-bed-based metal additive manufacturing. Progress 60. Morgan, D., Agba, E., Hill, C.: Support structure development and
Addit. Manuf. 1(1–2), 9–20 (2016). https://doi.org/10.1007/s40964- initial results for metal powder bed fusion additive manufacturing.
015-0001-4. ISSN: 2363-9512 Procedia Manuf. 10, 819–830 (2017). https://doi.org/10.1016/j.
42. Freeman, R.: Measuring the flow properties of consolidated, condi- promfg.2017.07.083. ISSN: 23519789
tioned and aerated powders – a comparative study using a powder 61. Järvinen, J.P., et al.: Characterization of effect of support structures
rheometer and a rotational shear cell. Powder Technol. 174(1–2), in laser additive manufacturing of stainless steel. Phys. Procedia.
25–33 (2007). https://doi.org/10.1016/j.powtec.2006.10.016. ISSN: 56(C), 72–81 (2014). https://doi.org/10.1016/j.phpro.2014.08.099.
00325910 ISSN: 18753892
404 S. R. Yarasi et al.

62. Ladani, L., Sadeghilaridjani, M.: Review of powder bed fusion 80. Kellner, T.: World’s first plant to print jet engine nozzles in mass
additive manufacturing for metals. Metals. 11(9) (2021). https:// production. In: GE Reports. https://www.ge.com/news/reports/
doi.org/10.3390/met11091391. ISSN: 20754701 worlds-first-plant-to-print-jet-engine-nozzles-in (2014)
63. Snyder, J.C., Thole, K.A.: Understanding laser powder bed fusion 81. Saltzman, D., et al.: Design and evaluation of an additively
surface roughness. J. Manuf. Sci. E. T. ASME. 142(7) (2020). manufactured aircraft heat exchanger. Appl. Therm. Eng.
https://doi.org/10.1115/1.4046504. ISSN: 15288935 138(April), 254–263 (2018). https://doi.org/10.1016/j.
64. Obilanade, D., Dordlofva, C., Törlind, P.: Surface roughness con- applthermaleng.2018.04.032. ISSN: 13594311
siderations in design for additive manufacturing – a literature review. 82. Yadroitsava, I., du Plessis, A., Yadroitsev, I.: Chapter 12 – Bone
Proc. Des. Soc. 1(August), 2841–2850 (2021). https://doi.org/10. regeneration on implants of titanium alloys produced by laser pow-
1017/pds.2021.545. ISSN: 2732527X der bed fusion: A review. In: Froes, F., Qian, M., Mitsuo, B.T. (eds.)
65. Lewandowski, J.J., Seifi, M.: Metal additive manufacturing: a Titanium for Consumer Applications Niinomi, pp. 197–233.
review of mechanical properties. Annu. Rev. Mater. Res. 46, Elsevier (2019). https://doi.org/10.1016/B978-0-12-815820-3.
151–186 (2016). https://doi.org/10.1146/annurev-matsci-070115- 00016-2. https://www.sciencedirect.com/science/article/pii/
032024. ISSN: 15317331 B9780128158203000162. ISBN: 978-0-12-815820-3
66. Kotadia, H.R., et al.: A review of laser powder bed fusion additive 83. Echeta, I., et al.: Review of defects in lattice structures manufactured
manufacturing of aluminium alloys: microstructure and properties. by powder bed fusion. Int. J. Adv. Manuf. Technol. 106(5),
Addit. Manuf. 46, 102155 (2021) 2649–2668 (2020)
67. Dryepondt, S., et al.: Microstructure and high temperature tensile 84. Khajavi, S.H., Partanen, J., Holmström, J.: Additive manufacturing
properties of 316L fabricated by laser powder-bed fusion. Addit. in the spare parts supply chain. Comput. Ind. 65(1), 50–63 (2014)
Manuf. 37, 101723 (2021) 85. Niaki, M.K., Ali Torabi, S., FabioNonino: Why manufacturers adopt
68. Gorsse, S., et al.: Additive manufacturing of metals: a brief review of additive manufacturing technologies: the role of sustainability.
the characteristic microstructures and properties of steels, Ti-6Al-4V J. Clean. Prod. 222, 381–392 (2019)
and high-entropy alloys. Sci. Technol. Adv. Mater. 18(1), 86. Khorasani, A.M., et al.: A review of technological improvements in
584–610 (2017). https://doi.org/10.1080/14686996.2017.1361305. laser-based powder bed fusion of metal printers. Int. J. Adv. Manuf.
ISSN: 18785514 Technol. 108(1), 191–209 (2020)
69. Sames, W.J., et al.: The metallurgy and processing science of metal 87. Pauzon, C., et al.: Argon-helium mixtures as laser-powder bed
additive manufacturing. Int. Mater. Rev. 61(5), 315–360 (2016). fusion atmospheres: towards increased build rate of Ti-6Al-4V.
https://doi.org/10.1080/09506608.2015.1116649. ISSN: 17432804 J. Mater. Process. Technol. 279, 116555 (2020)
70. Chadwick, A.F., Voorhees, P.W.: The development of grain structure 88. Bhavar, V., et al.: A Review on Powder Bed Fusion Technology of
during additive manufacturing. Acta Mater. 211, 116862 (2021). Metal Additive Manufacturing. Additive Manufacturing Handbook,
https://doi.org/10.1016/j.actamat.2021.116862. ISSN: 13596454 In (2017) p. 255
71. Cunningham, R., et al.: Analyzing the effects of powder and post- 89. Yavari, R., et al.: Part-scale thermal simulation of laser powder bed
processing on porosity and properties of electron beam melted fusion using graph theory: effect of thermal history on porosity,
Ti-6Al-4V. Materials Research Letters. 5(7), 516–525 (2017). microstructure evolution, and recoater crash. Mater. Des. 204,
https://doi.org/10.1080/21663831.2017.1340911. ISSN: 21663831 109685 (2021)
72. Cunningham, R., et al.: Synchrotron-based X-ray microtomography 90. Anderson, I.E., White, E.M.H., Dehoff, R.: Feedstock powder pro-
characterization of the effect of processing variables on porosity cessing research needs for additive manufacturing development.
formation in laser power-bed additive manufacturing of Ti-6Al-4V. Curr. Opin. Solid State Mater. Sci. 22(1), 8–15 (2018)
JOM. 69(3), 479–484 (2017). https://doi.org/10.1007/s11837-016- 91. Pollock, T.M., Clarke, A.J., Babu, S.S.: Design and tailoring of
2234-1. ISSN: 15431851 alloys for additive manufacturing. Metall. Mater. Trans. A Phys.
73. Cunningham, R., et al.: Analyzing the effects of powder and post- Metall. Mater. Sci. 51(12), 6000–6019 (2020). https://doi.org/10.
processing on porosity and properties of electron beam melted 1007/s11661-020-06009-3. ISSN: 10735623
Ti-6Al-4V. Materials Research Letters. 5(7), 516–525 (2017).
https://doi.org/10.1080/21663831.2017.1340911. ISSN: null
74. Cunningham, R.W.: Defect formation mechanisms in powder-bed
metal additive manufacturing. 2018. https://doi.org/10.1184/R1/
6715691.v1. https://kilthub.cmu.edu/articles/thesis/Defect_Forma
tion_Mechanisms_in_Powder-Bed_Metal_Additive_Manufactur
ing/6715691/1
75. Dall’Ava, L., et al.: 3D printed acetabular cups for total hip
arthroplasty: a review article. Metals. 9(7), 729 (2019)
76. Spierings, A.B., Schneider, M., Eggenberger, R.: Comparison of
density measurement techniques for additive manufactured metallic
parts. Rapid Prototyp. J. 17(5), 380–386 (2011). https://doi.org/10.
1108/13552541111156504. ISSN: 1355-2546
77. Blakey-Milner, B., et al.: Metal additive manufacturing in aero-
space: a review. Mater. Des. 209, 110008 (2021). https://doi.org/ Srujana Rao Yarasi earned her M.S. in Materials Science at Carnegie
10.1016/j.matdes.2021.110008. ISSN: 18734197 Mellon University (2019) and her B.E. (Hons) in Mechanical Engineer-
78. Haleem, A., Javaid, M.: 3D printed medical parts with different ing at the Birla Institute of Technology & Science (2017). She is
materials using additive manufacturing. Clin. Epidemiol. Glob. currently a Ph.D. candidate at Carnegie Mellon University working on
Health. 8(1), 215–223 (2020). https://doi.org/10.1016/j.cegh.2019. understanding the relationships between metal powder characteristics
08.002. ISSN: 22133984 such as morphology with flow properties for powder bed fusion using
79. Caiazzo, F., et al.: Laser powder-bed fusion of Inconel 718 to man- computer vision and machine learning.
ufacture turbine blades. Int. J. Adv. Manuf. Technol. 93(9),
4023–4031 (2017). https://doi.org/10.1007/s00170-017-0839-3.
ISSN: 02683768 (English)
24 Powder Bed Fusion 405

Dr. Andrew R. Kitahara earned his Ph.D. in Materials Science and

Engineering from Carnegie Mellon University (2021) and his B.S. in Dr. Anthony D. Rollett is the US Steel Professor in Materials Science
Materials Engineering from Iowa State University (2016). He is cur- and Engineering at Carnegie Mellon University, where he has been since
rently a postdoctoral researcher at Carnegie Mellon University, investi- 1995. He was previously at the Los Alamos National Laboratory from
gating materials data science challenges for powder bed fusion additive 1979 to 1995. He is the recipient of multiple awards, including Fellow of 24
manufacturing of Ti-6Al-4V. Andrew’s research interests focus on TMS in 2011 and International Francqui Professor for 2020–2021
applied machine learning for materials science and automating data (Belgium). He is a codirector of the Next Manufacturing Center at
acquisition and analysis. CMU and a leader in research and education in metal additive
manufacturing and processing-microstructure-properties relationships.

Dr. Elizabeth A. Holm Ph.D., F.A.S.M., F.T.M.S., is a Professor of

Materials Science at Carnegie Mellon University. Prior to joining CMU,
she spent 20 years as a Distinguished Member of the Technical Staff at
Sandia National Laboratories. Her research areas include computational
materials science at the atomic and mesoscales and AI and machine
learning for materials science.