\documentclass[10pt]{article}
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
\usepackage{hyperref}
\hypersetup{colorlinks=true, linkcolor=blue, filecolor=magenta, urlcolor=cyan,}
\urlstyle{same}
\usepackage{amsmath}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage[version=4]{mhchem}
\usepackage{stmaryrd}
\usepackage{graphicx}
\usepackage[export]{adjustbox}
\graphicspath{ {./images/} }
\usepackage{multirow}

\title{Controlled Crystallographic Texture Orientation in Structural Materials Using the Laser Powder Bed Fusion Process-A Review }


\author{Prince Valentine Cobbinah,* Sae Matsunaga, and Yoko Yamabe-Mitarai*}
\date{}


%New command to display footnote whose markers will always be hidden
\let\svthefootnote\thefootnote
\newcommand\blfootnotetext[1]{%
  \let\thefootnote\relax\footnote{#1}%
  \addtocounter{footnote}{-1}%
  \let\thefootnote\svthefootnote%
}

%Overriding the \footnotetext command to hide the marker if its value is `0`
\let\svfootnotetext\footnotetext
\renewcommand\footnotetext[2][?]{%
  \if\relax#1\relax%
    \ifnum\value{footnote}=0\blfootnotetext{#2}\else\svfootnotetext{#2}\fi%
  \else%
    \if?#1\ifnum\value{footnote}=0\blfootnotetext{#2}\else\svfootnotetext{#2}\fi%
    \else\svfootnotetext[#1]{#2}\fi%
  \fi
}

\DeclareUnicodeCharacter{0131}{$\imath$}

\begin{document}
\maketitle


\begin{abstract}
Advantages of the laser powder bed fusion (LPBF) process over conventional processing methods include microstructural control capability and complex geometry parts fabrication. For most alloys, the microstructure-property relationship typically guides the LPBF process toward achieving application-specific properties. Nonetheless, the LPBF process produces a high degree of anisotropy primarily from the epitaxial preferred grain growth mechanisms during solidification and the consequent highly textured microstructures produced. The crystallographic texture-controlled build concept through the LPBF process is a material design and processing approach gaining a lot of interest because of the promising unique and superior properties offered by the anisotropy accompanying the LPBF process. The numerous attempts at intentional texture-controlled building or parts fabrication from structural materials using the LPBF process warrant an up-to-date review. Thus, this review presents solidification mechanisms that influence texture evolution during the LPBF process such as the melting mode, melt pool shape, influence of thermal gradient, and solidification rate. Furthermore, this review delves into the influence of LPBF process parameters on texture, accompanying microstructural evolution, and the intentional control of texture effects on properties such as mechanical, corrosion, and oxidation resistance.
\end{abstract}

\section*{1. Introduction}
Fundamentally, composition and processing route significantly influence the microstructures of alloys. The microstructure, in turn, determines the final properties and alloy behavior. In most studies, phase transformation, phase distribution, grain size, and morphology are the key features in microstructures usually altered or optimized to improve the properties of alloys such
\footnotetext{P. V. Cobbinah, S. Matsunaga, Y. Yamabe-Mitarai

Graduate School of Frontier Sciences

Advanced Materials Science Department

The University of Tokyo

5-7-5 Kashiwanoha, Kashiwa, Chiba 277-8561, Japan

E-mail: \href{mailto:p.cobbinah22s@ams.k.u-tokyo.ac.jp}{p.cobbinah22s@ams.k.u-tokyo.ac.jp}

\href{mailto:mitarai.yoko@edu.k.u-tokyo.ac.jp}{mitarai.yoko@edu.k.u-tokyo.ac.jp}

The ORCID identification number(s) for the author(s) of this article can be found under \href{https://doi.org/10.1002/adem.202300819}{https://doi.org/10.1002/adem.202300819}.

(c) 2023 The Authors. Advanced Engineering Materials published by WileyVCH GmbH . This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
}

DOI: 10.1002/adem.202300819 as strength, ductility, toughness, hardness, wear resistance, corrosion, and oxidation resistance. ${ }^{[1]}$ To date, the microstructureproperty relationship serves as the primary guideline for understanding and developing various structural materials. The crystallographic texture is another intrinsic microstructural characteristic that influences property variations in alloys, which is mostly overlooked. Texture involves the study of crystal orientation distribution. The mechanical properties of alloys such as fatigue resistance, Young's modulus, yield strength and ductility, and creep resistance accentuate the potential and magnitude of texture effects. ${ }^{[2]}$ Furthermore, crystallographic texture provides detailed information about the kind or amount of processing that the material has undergone. ${ }^{[3]}$

Conventional processing methods such as the deformation processes (rolling, forging, and drawing) are well known to significantly influence crystallographic texture. ${ }^{[4]}$ The synergistic effects of deformation and recrystallization on microstructure, however complex, enable texture manipulation over a wide range of possibilities. During deformation, the grains undergo elongation in the deformation direction, followed by dynamic recovery and dynamic recrystallization leading to the nucleation or growth of new grains. ${ }^{[3]}$ Either new grains of certain crystallographic orientation grow in a preferred direction or nucleation rather preferentially "orient" thus strongly influencing the final texture that forms. ${ }^{[3]}$ A practical example is the drawing of face-centered cubic (FCC) metals into wires. ${ }^{[5]}$ The deformed grains form a fiber texture by aligning in the $<111>$ direction along the drawing direction. Recrystallization then alters the formed $<111>$ fiber texture into $<100>$ fiber texture. ${ }^{[5]}$ The example shows that texture can be manipulated through the drawing and annealing processes to control the relative intensities of the two fibers. Generally, exploiting the potential of deliberate crystallographic texture control in many alloys through the conventional deformation processing route can be extremely challenging. Additionally, the conventional deformation processes are cost intensive and strenuous; mostly require extra machining for dimensional accuracy resulting in material waste; and are restricted to products with simple shapes and sizes.

Conversely, the laser powder bed fusion (LPBF) technology facilitates the realization of complex-shaped products. LPBF\\
involves a layer-wise sequence build-up of materials resulting from the controlled interaction of a laser beam energy source with a metallic powder bed to fabricate functional parts. Exceptional to LPBF is the near fully dense ( $>99 \%$ density) metal parts capabilities and the better surface quality associated with the process. The process presents the opportunity to obtain refined grains, substructures within grains, and overall microstructural control. Additionally, the LPBF process affords the feasibility of wide texture control by enabling grain growth toward preferred grain orientation, thereby allowing the intentional control of texture. The concept of simultaneous control of product shape and material properties through texture control, owing to additive manufacturing (AM) technologies such as LPBF can help tailor-specific properties to specific parts of products to meet varying service conditions. ${ }^{[6]}$ The ability to tailor partspecific properties certainly expands the prospects and utilization of high-performance multifunctional products that cannot be realized using conventional manufacturing methods. However, the desire and success of this concept hinge on understanding and leveraging the knowledge of several materials whose texture can be controlled during the AM process.

There are several reviews focused on the LPBF process, parameters, and the influence of the LPBF process parameters on the starting alloy powders, defects formation mechanisms, residual stress, microstructural evolution, post-processing treatments, and mechanical properties. ${ }^{[7-9]}$ However, reviews dedicated to the influence of LPBF on texture evolution and the potential of texture control are few. Some examples include the comprehensive review of the influence of scan strategy on texture evolution in FCC, body-centered cubic (BCC), and non-cubic materials by Hagihara and Nakano. ${ }^{[6]}$ Suwas and Kumar ${ }^{[10]}$ reviewed the microstructure-texture relationship and accompanying mechanical properties in various alloy systems processed by LPBF. Also, the fundamentals of microstructure and texture evolution in some common structural materials such as steels, nickel-based superalloys (NBSAs), and Ti alloys processed by LPBF have been reviewed by Suwas and Vikram. ${ }^{[11]}$ The aforementioned reviews show that knowledge of the influence of other process parameters especially parameters that affect heat input and thermal history during the LPBF process in the effort to control texture is still lacking and yet to be presented. To address this gap and further add knowledge to the LPBF control of crystallographic texture, this review will take the following form. First, the melting modes and melt flow (Section 2) that occur during the LPBF process are addressed. This is because of their integral role in melt pool shapes, grain growth, epitaxy, and consequent solidification structures and texture. Second, although there are many reviews on microstructural evolution, we deemed it worthwhile to add microstructural evolution to this review (Section 3). This is in an attempt to make this review a good starting point for literature searches concerning LPBF texture control and the consequent generated microstructures. Third, this review delves into the influence of the various LPBF process parameters on crystallographic texture evolution, and the intentional control of texture (Section 4). Section 5 addressed the fundamentals of the solid-state phase transformation in titanium alloys during the LPBF process, while Section 6 presents the influence of texture control in refractory alloys. Last, the effects of tailored texture on properties such as mechanical, corrosion, and oxidation resistance are considered (Section 7). Section 8 summarizes all the sections and provides current challenges and prospects of controlled crystallographic texture.

\section*{2. LPBF Melting Modes and Melt Flow}
The melting mode and melt flow within the formed melt pools are important to both microstructural and texture evolution during the LPBF process. This is because the melting mode and melt flow determine the melt pool shape. The formation of stable and consistent melt pool shapes for each layer of the building process improves fusion and affects the final microstructure and texture evolution. The melting mode and melt flow are strongly controlled by the laser beam spot size ( $\sigma$ ), laser power intensity $(P)$, laser scan speed $(v)$, hatch distance $(d)$, powder layer thickness ( $t$ ), and intrinsic properties of the metal (or alloy) such as thermal diffusivity, conductivity, and laser absorptivity depending on composition. The LPBF process is strongly driven by heat and characterized by localized melting and rapid solidification. In the process of stacking up powder layers, a molten pool develops as the high-power laser beam melts a row of metal powder along a given direction. Depending on the amount of energy density absorbed by the powder, two main melting modes (conduction or keyhole) occur (Figure 1). ${ }^{[12]}$

\subsection*{2.1. Melting Mode}
Depending on the LPBF processing conditions, when the energy density is below a certain limit value, the conduction melting mode occurs. In the conduction mode, the dominant heat transfer mechanism is heat conduction. Usually, not more than $20 \%$ of the laser beam energy is efficiently used in melting, which consequently accounts for the depth of the melt pool that forms. ${ }^{[13]}$ Thus, melt pools formed through conduction melting

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-02}
\end{center}

(b)

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-02(1)}
\end{center}

Figure 1. Schematic diagrams illustrate the forming mechanism of a) the conduction mode and b) the keyhole mode. Reproduced with permission. ${ }^{[12]}$ Copyright 2017, Elsevier Ltd.\\
mode are wide and shallow. When viewed from the cross-section, the melt pool formed from the conduction melting mode has been described as semi-circular, ${ }^{[12]} \mathrm{U}$-shaped, ${ }^{[14]}$ hemisphereshaped, ${ }^{[15]}$ and elliptical-shaped, or half-ellipse. ${ }^{[16]}$ From a quantitative perspective (melt pool aspect ratio), the melt pool depth in the conduction melting mode is less than or equal to its halfwidth. ${ }^{[14,17]}$

On the other hand, when the energy density is higher than a certain threshold value, the keyhole melting mode occurs. In the keyhole mode, the absorbed laser energy for melting is about $70-90 \%,{ }^{[13]}$ sufficient to cause metal vaporization and plasma formation. The metal vaporization induces large recoil pressure on the molten pool, which exceeds the surface tension and hydrostatic pressure of the molten pool. ${ }^{[18]}$ The large recoil pressure creates a vapor-filled cavity (keyhole) from the surface down to the depths of the pool, allowing the laser beam to penetrate deeper. ${ }^{[18]}$ The melt pool, therefore, appears narrow and deep from the keyhole effect. The dominant heat transfer mechanism in this mode is the convective heat transfer attributed to thermocapillary convection (Bénard-Marangoni convection). Examples of melt pool shapes ascribed to the keyhole melting mode from a cross-sectional view include goblet, ${ }^{[12]}$ V-shaped, ${ }^{[14]}$ bowlshaped, ${ }^{[19]}$ nail-head-shaped, ${ }^{[20]}$ hourglass, ${ }^{[21]}$ and others. Compared to the conduction melting mode, the melt pool depth formed owing to the keyhole mode is greater than the half-width of the melt pool. ${ }^{[14,17]}$ Other quantitative means used to distinguish between the keyhole and conduction mode melting are laser power intensity, ${ }^{[22]}$ and normalized enthalpy. ${ }^{[23]}$

Furthermore, depending on the process conditions and material, the possibility of a threshold is speculated where the melting mode exhibits characteristics of both the conduction and keyhole melting modes. ${ }^{[24]}$ This threshold or region is referred to as the transition region. ${ }^{[25,26]}$ At this threshold, the induced recoil pressure is inadequate to overcome the surface tension on the molten pool, leading to insufficient penetration. Most of the applied laser energy in the transition mode or region, therefore, goes into increasing vaporization rates till a deep penetration (keyhole) forms. ${ }^{[26}$

In summary, the driving force for conduction, transition, and keyhole melting mode is the heat input (laser power intensity). Generally, in the LPBF, vaporization of the processed material is inevitable. The magnitude (depth) of vaporization, dependent on the heat input, accounts for the thresholds between the three melting modes. The keyhole mode develops when the laser is absorbed through several reflections. ${ }^{[21]}$ When the laser is reflected outside the developed vaporized keyhole after a few reflections, the conduction mode is dominant. In the case of the transition mode melting in the LPBF, fluctuations in laser absorptivity control the melt pool depths resulting in characteristics of the conduction and keyhole melting modes. ${ }^{[21]}$

\subsection*{2.2. Melt Flow}
Regarding melt flow in the melting modes, the recoil pressure and surface tension effects (Bénard-Marangoni convection) principally drive melt flow in the molten pool. ${ }^{[27]}$ These flows, which depict the process of energy transfer from the laser beam to the material, govern the heat and mass balance system in the LPBF process. Foremost, the material absorbs sufficient energy from the laser beam at the laser spot, which leads to melting and the formation of a melt pool, and then these flows transfer the energy to the other parts of the molten pool. ${ }^{[28]}$ The initiation of thermocapillary forces for fluid flow because of the temperature gradient in the melt pool (Marangoni flow) gives rise to a corresponding differential surface tension between the edge and center of the melt pool. The Marangoni effect, hence, drives the melt flow from the molten pool surface to the cold bottom. This increases the melting rate and the melt depth and recirculates the melt flow (leading to the cooling of the laser spot location). Wang and Zou ${ }^{[28]}$ (Figure 2) working with Ti-6Al-4V observed one circular flow in the longitudinal section of the molten pool in the conduction mode, whereas the keyhole mode showed two circular flows. The one circular flow in the conduction mode is reported to have a lesser fluid velocity magnitude and more uniformity resulting in smoother tracks after solidification. ${ }^{[29]}$ The two circular flows in the case of the keyhole mode include a backward flow generated when the recoil pressure forces the molten metal away from the center of the laser beam and the recirculating flow returning from the tail of the melt pool. ${ }^{[29]}$ Since the two flows are opposite, the flows collide at some point which sometimes appear as bulges in (or along) the track after solidification. The physics behind the different melt flows is quite complex and thus is mostly studied or rationalized using various computational models, ${ }^{[27]}$ which is outside the scope of the present review. In most of these studies validated by experiments, the conduction and keyhole melting modes have been found to influence solidification and final microstructure formation. Readers can refer to ref. [30] for a detailed discussion.

\section*{3. Microstructure Development}
Several attempts to explain the microstructural development associated with LPBF have been based on two theories: constitutional supercooling ${ }^{[31]}$ (also referred to as solute undercooling $(\Delta T)^{[32]}$ ) and interfacial stability ${ }^{[33]}$ theories. These theories cover the thermodynamics and kinetics relations governing solidification mode and generated microstructural features. ${ }^{[34]}$ The alloy composition effect, the temperature gradient in the melt pool $(G)$, solidification (or growth) rate $(R)$, and the liquid phase undercooling $(\Delta T)$ are essential factors of the constitutional supercooling theory. ${ }^{[31,34]}$ Conversely, the influence of the processed alloy's interfacial kinetics and heat transfer on solidification mode form the interfacial stability theory. In the interfacial stability theory, the migration velocity and solute trapping (or push-out) at the solid-liquid interface determine the solidification behavior of the alloy. ${ }^{[33,34]}$

\subsection*{3.1. The Effect of Temperature Gradient and Growth Rate}
Since alloys constitute different elements with varying individual melting temperatures, the alloy composition effect involves alloys solidification occurring over a range rather than a melting point as seen for pure metals. Alloys solidifying over a range lead to constitutional supercooling, which affects redistributions between the solid and liquid interface. ${ }^{[35]}$ Also, high temperatures and high cooling rates during the LPBF process result in temperature differences $(G)$ in the melt pool. $G$ depending

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-04(1)}
\end{center}

Conduction mode

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-04}
\end{center}

Keyhole mode

Figure 2. Schematics of the circular flows in two typical modes. The red line represents the boundary of the molten pool, while the blue arrow represents the flow of melted metal. Metal in the molten pool flows along the direction of the arrow, keeping the balance of heat and mass. Reproduced with permission. ${ }^{[28]}$ Copyright 2019, Elsevier Ltd.

on time and location in the melt pool differs during the LPBF process. On the other hand, $R$ closely relates to scan speed and the angle between the heat flow and the growth directions of the solidifying material. ${ }^{[36]}$ As such, the constitutional supercooling theory uses the relationship between $G$ and $R$ in the forms $G / R$ and $G * R$ to predict the solidification front stability (solidification mode) and the fineness of the solidification microstructure respectively ${ }^{[37]}$ (Figure 3). With decreasing G/R, solidification microstructures sequentially evolve from planar to cellular, or dendritic. ${ }^{[38]}$ On the other hand, the higher the product $G * R$ (cooling rate), the finer the solidification microstructure. ${ }^{[37]}$

During the LPBF process, the melt pool experiences not only temperature gradients but also varying growth rates $(R)$. The center of the melt pool experiences the highest energy (fast growth rate) and the melt pool boundary, the least energy (slow growth rate). ${ }^{[39]}$ At the center of the melt pool, the heat source (laser beam) is parallel to the heat flow direction leading to a fast growth rate. The slow growth rate at the melt pool boundary is because

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-04(2)}
\end{center}

Figure 3. Effect of temperature gradient $(G)$ and growth rate $(R)$ on the morphology and size of solidification microstructure. Reproduced with permission. ${ }^{[31]}$ Copyright 2003, John Wiley and Sons. the heat flow direction becomes perpendicular to the moving laser direction. Furthermore, $G, R$, and cooling rates differ at the surface and bottom of the melt pool. G, G/R ratios, and cooling rates are highest at the melt pool bottom (or fusion line), while $R$ is extremely low. ${ }^{[40]}$ Conversely, at the melt pool surface, $R$ directly influenced by laser scan speed is the highest, whereas $G$, cooling rates, and $G / R$ ratios are lowest. ${ }^{[40]}$ The different $G$ and $R$ in the melt pool mean solidification mode and microstructure will always vary. ${ }^{[39]}$

\subsection*{3.2. Planar and Cellular Structure}
High $G$ and low $R$ (high $G / R$ ratio) favor the planar solidification mode. In the planar solidification mode, competitive growth governs the microstructure. The competitive growth phenomenon is when grains with "easy growth" directions closest to the maximum heat flow direction grow the fastest. ${ }^{[8]}$

As shown in Figure 3, the planar microstructure evolves into the cellular microstructure at a relatively lower $G / R$ ratio during solidification. The cellular microstructure is made up of cells seen as a recurring pattern of the solid-liquid interface with strongly connecting tips as shown in Figure 4. Prashanth and Eckert ${ }^{[4]]}$ investigated Al-12Si, AlSi10Mg, CoCrMo, and the stainless steel 316 L systems and asserted that, aside from the $G$ and $R$ effects, the cellular microstructure formation in some materials during the LPBF process occurs due to the Marangonidriven instability (surface tension effect) and the particle accumulated structure formation (mass transport effect) mechanisms.

The cellular microstructure will form provided the processed material meets the following criteria ${ }^{[34]}$ : 1) The processed material during solidification forms a binary alloy system. For a multicomponent system, the alloy should result in the formation of only two phases with one phase acting as solute and the other phase as the solvent. 2) The solute(s) should be immiscible with the solvent or the solubility of solute(s) in the solvent decreases with temperature and becomes negligible around the solidification temperature of the solvent phase. This means that the equilibrium partition coefficient between the solute(s) and the solvent should be less than unity. 3) The solvent phase\\
\includegraphics[max width=\textwidth, center]{2024_07_13_01fad27876b5c2169f6dg-05}

Figure 4. SEM images of SLM parts made of a) AlSi10Mg (Reproduced with permission. ${ }^{[39]}$ Copyright 2012, Elsevier Ltd.) b) CoCrMo (Reproduced with permission. ${ }^{100]}$ Copyright 2015, Elsevier Ltd.) c) SS316L (Reproduced with permission. ${ }^{[101]}$ Copyright 2016, Elsevier Ltd.) depicting cellular microstructures.

should have a lower melting temperature than the solute(s). 4) The difference in melting point between the solute(s) and the solvent should be greater than 673 K .

\subsection*{3.3. Dendritic Structure}
\subsection*{3.3.1. The Formation Kinetics}
The dendritic microstructure described as "tree-like" (Greek word: "Dendron") exhibits a branched morphology with primary, secondary, tertiary, and higher order branches. ${ }^{[41]}$ Dendrites form owing to undercooling $(\Delta T)$ and the composition $\left(\mathrm{C}_{\mathrm{o}}\right.$ ) of the processed alloy at the solidification front. ${ }^{[22]}$ In dendrites formation, the growing solid crystals adopt a nonplanar (i.e., dendritic) shape depending on the growth rate $(R)$ of the crystal and the rate at which "piled up" solute elements are pushed outside the solidifying front by diffusion. ${ }^{[42]}$ These pushed-out solutes (or solute segregation pattern) manifest as the various branches or interarm spacings of the dendrites and usually form secondary phases such as precipitates or pores at the interdendritic areas. The amount and solute types contained in the alloy significantly influence the dendritic morphology.

In addition, there is a close connection between the evolution and growth of dendrites to latent heat dissipation from the solidification interface. The solidification interface abounds with heat (latent heat) that requires dissipation. ${ }^{[41,43]}$ The manner in which latent heat dissipates dictates the growth condition of the dendrites. Also, the solidification interface serves as a solute source (or sink) during solute redistribution between the solid and liquid, inducing solute diffusion for dendrite growth. ${ }^{[41,43]}$

\subsection*{3.3.2. Equiaxed Dendritic Structure and Columnar Structure}
There are two growth conditions that support the evolution and growth of dendrites: 1) nucleated equiaxed crystals growth in an undercooled isothermal melt and 2) the columnar structure. ${ }^{[41]}$ The equiaxed dendritic and columnar structures are the most common microstructures in the LPBF process. In nucleated equiaxed crystals growth, the latent heat dissipates from the crystal through the undercooled melt where $G$ is nearly zero in the solid but negative in the liquid (Figure 5a,b). The crystal grows with heat flux dissipation occurring equally in all directions. The solidification behavior of the crystal results in the equiaxed dendritic (or eutectic) morphologies forming a polycrystalline solid with randomly oriented grains. ${ }^{[41]}$ Conversely, the columnar structure characterizes the second growth condition and forms if the heat flux orients unidirectionally. Also described as constrained growth or directional solidification, columnar dendritic growth involves latent heat dissipation from the superheated melt through the solid due to a positive $G$ in the liquid (Figure $5 c$ ).

Furthermore, the solidification front instability in equiaxed growth owing to the equiaxed morphology results in the solidification of a single phase or dendritic grains. The case of columnar growth, depending on the local growth conditions, may take a planar, cellular, or dendritic morphology. ${ }^{[44]}$ As exemplified in Figure 6, Marattukalam et al. ${ }^{[20]}$ observed columnar growths with planar and dendritic morphologies in the melt pools of the LPBFbuilt stainless steel 316 L. The LPBF parameters employed in the study include a 107 W laser power input, scan velocity of 800 mm $\mathrm{s}^{-1}$, hatch distance of $70 \mu \mathrm{m}$, and three different scan strategies (Z-scan, Y-scan, and a $67^{\circ}$ rotational-scan with bidirectional laser scanning vectors).

Columnar grain morphology and growth depend on heat flux direction, energy source velocity, melt pool boundary shape, and melt pool shape. According to Grong and Grong, ${ }^{[45]}$ a sphericalshaped melt pool formed from low scan speed, low laser power input, and high thermal diffusivity process parameter combination promotes curved and tapered columnar grains. These curved and tapered columnar grains evolve because the maximum thermal gradient direction shifts in the liquid from the melt pool boundary toward the liquid-free surface. However, a cometshaped melt pool from a high scan speed, high laser power input, and low thermal diffusivity supports straight and broad columnar grains since the maximum thermal gradient direction in the melt pool does not significantly change during the solidification process. ${ }^{[45]}$

Moreover, columnar grains along the melt pool boundary grow in the thermal gradient direction of the melt pool. The relationship $R$ has with laser scan speed $(v)$ and the angle between the heat flow and the solidifying material's growth directions $(\theta)$ can be summarized as $R_{N}=v \cdot \cos \theta \cdot{ }^{[38,40]}$ The steady-state nominal growth rate $\left(R_{N}\right)$ for metals and alloys that exhibit preferred growth directions (e.g., $<100>$ for cubic materials) is invariably lower than the local growth rate $\left(R_{\mathrm{L}}\right)$ of the crystal. Therefore, the relationship between $R_{N}$ and $R_{\mathrm{L}}$ can be presented as

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-06(1)}
\end{center}

Figure 5. a,b) Free growth and c) constrained growth conditions: compositions and temperature fields are given along the dendrite axis. Reproduced with permission. ${ }^{[4]]}$ Copyright 1994, Taylor \& Francis.

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-06}
\end{center}

Figure 6. SEM image showing planar and dendritic growth patterns along OA and OB , respectively, which is dependent on the temperature gradient $(G)$ within the melt pool. Reproduced with permission. ${ }^{[20]}$ Copyright 2020, Elsevier Ltd.

$R_{\mathrm{L}}=R_{N} / \cos \alpha$, where $\alpha$ is the angle between the melt pool boundary normal and the preferred growth direction, such as $<100>{ }^{[38]}$ An increase in crystal misalignment owing to the maximum thermal gradient direction in the melt pool increases $R_{\mathrm{L}}$. ${ }^{[38]}$

The transition from columnar to equiaxed growth (CET) occurs when the constitutionally undercooled liquid consists of sufficiently many nucleated equiaxed grains ahead of the columnar dendritic front. ${ }^{[46]}$ Once nucleated, a certain volume fraction of equiaxed grains will form depending on solidification factors such as the temperature gradient in the liquid $(G)$, the solidification rate of the columnar front, the nuclei density of the alloy, etc. This eventually results in the CET. ${ }^{[46]}$

\section*{4. Texture Evolution}
\subsection*{4.1. Texture Evolution Mechanisms}
The melting mode, melt flow, and melt pool solidification determine also the crystallographic texture of LPBF-built parts. The numerous attempts at explaining texture evolution mechanisms in most LPBF research, center around 1) the influence of melt pool shape and solidification behavior ( $G$ and $R$ ) in the melt pool and 2) texture evolution from already solidified deposited layers (epitaxial crystal growth).

Sun et al. ${ }^{[47]}$ reported that the geometry of the melt pool strongly influences crystallographic texture formation. The melt pool shape reflects the melting mode and affects the solid-liquid interface migration direction $(R)$, which closely relates to the heat flow direction and $G$ within the melt pool. For example, during solidification in the conduction mode melting, the bottom of the melt pool curvature decreases, and the solid-liquid interface migration is usually toward the build direction. ${ }^{[47]}$ On the other hand, in the keyhole mode melting, the curvature of the melt pool increases, and lateral migration of the solid-liquid interface dominates at the melt pool bottom. ${ }^{[47]}$

Epitaxy, on the other hand, describes the process involving the deposition of a material possessing one crystal type on a substrate\\
with identical or almost identical lattice dimensions. ${ }^{[48]}$ Thus, epitaxial solidification involves a solid grain or nucleus nucleation process from a liquid in contact with a substrate. ${ }^{[38]}$ The partial remelting or the complete melt-back of the substrate (i.e., previously deposited underlying or adjacent tracks and layers) facilitates epitaxial solidification. Epitaxial growth plays an essential role in microstructural and crystallographic texture formation and continuity. So, as solidification occurs, the newly melted tracks or layers inherit the crystal orientation of the previously solidified layer over multiple layers with columnar grain growth. ${ }^{[38]}$ Additionally, based on the premise that the melt pool is a contiguous semi-circle, epitaxial solidification (with specific grain orientation) follows the maximum heat flow direction occurring in the connecting centers of the circle and at the overlapping melt pool edges along the build direction as exemplified in studies such as refs. $[49,50]$.

\subsection*{4.2. Influence of Solidification Parameters (such as $G, R, \Delta T$ ) on Texture Development}
\subsection*{4.2.1. Influence of Laser Heat Input Parameters ( $P, v, d$, and $t$ )}
LPBF process parameters such as laser power $(P)$, scan speed $(v)$, scan strategy, hatch distance (d), and powder layer thickness ( $t$ ) influence the effects that melt pool shape, thermal gradient $(G)$, and migration velocity of the solid-liquid interface $(R)$ have on texture development. There are different explanations that have been given in various studies on the influence of these process parameters in the formation of various crystallographic textures. In this section, we consider some studies and highlight the attributed LPBF process parameter(s) in the formation of the different crystallographic textures. Additionally, Table 1 summarizes studies targeted at texture control using the LPBF process.

Energy input usually quantified as volumetric energy density (VED) or energy density involves the combined effects of $P, v, d$, and $t$, which accounts for the heat-flow direction, the balance between $G$ and $R$ over the melt pool, and the melt pool shape. VED, which is proportional to the laser power input, has been linked to the $<001>$ family texture development in the LPBF process. ${ }^{[51-54]}$ High VED from a high-power input and decreased scan speed favor the $<001>$ family crystallographic texture formation mainly by affecting the melt pool shape. ${ }^{[5]}$ When deep melt pools form with a solidification front almost horizontal at the melt pool centerlines, two easy-growth directions perpendicular to each other from the melt pool sides and center occur which culminate in the formation of a $<001>$ cubic texture. ${ }^{[47]}$ Similar influence of energy input has been reported by Garibaldi et al. ${ }^{[56]}$ in the LPBF processing of high silicon steel parts. As the authors observed, by increasing energy input, the $<001>$ fiber texture changed into the $<001>$ cube texture in the build direction. This change was also attributed to the formation of deeper melt pools. However, compared to ref. [47], the solidification front at the melt pool sides shifted directions $\left(90^{\circ}\right)$ toward the build direction leading to the $<001>$ cube texture in the build direction. Contradictory to the above-mentioned studies, Sun et al. ${ }^{[57]}$ by using the kinetic Monte Carlo mesoscale simulation approach ascribed the formation of the $<001>$ texture to shallow melt pools while deep melt pools promoted the development of the $<011>$ texture.

Another school of thought asserts that a high $G / R$ ratio (Figure 7) supports the formation of the $<001>$ texture owing to the competitive growth of grains. For certain materials such as the FCC and BCC materials, grain growth innately aligns in the three $<100\rangle$ easy growth directions [100], [010], and [001] parallel to the building direction. This is because the preferred $<001>$ family easy growth direction in cubic materials stems from the anisotropy of the crystals together with how atoms are packed at the crystallographic planes. ${ }^{[9]}$ The axis with the densest direction grows the slowest during solidification, whereas the axis with the least atoms $(<001>$ ) grows the fastest. ${ }^{[9]}$ Additionally, it has been asserted that the $<001>$ texture in the BCC structure requires less energy for growth, thus accounting for its quickest evolution and growth during solidification. ${ }^{[51]}$

The influence of laser scan speed (v) on crystallographic texture formation has also been reported. From Figure 7, a high scan speed is likely to result in the $<001>$ texture. Higashi and Ozaki ${ }^{[58]}$ observed that higher VEDs used in the processing of pure molybdenum (Mo) produced strong $\langle 110\rangle$ fiber texture. However, by increasing speed (low VED), the formed $<110>$ fiber texture completely changed into a strong $<001>$ fiber texture. Further investigations showed the independent influence of scan speed on the crystallographic texture. The reason for the impact of scan speed on texture was again attributed to the formed melt pool morphology. Teardrop-like melt pools with high solidification angles relative to the surface plane formed at higher scanning speeds supporting the formation of the $<001>$ fiber texture. Additionally, a decreased VED (relatively high laser power and low scan speed) compared to that used in obtaining the $<110\rangle$ fiber texture resulted in a weak $<111>$ crystallographic texture. ${ }^{[58]}$

The $<110\rangle$ texture can be tailored as well through the careful manipulation of VED. One means is by employing a relatively low powder layer thickness. ${ }^{[59]}$ A low powder layer thickness means a low laser power (low VED) is required for melting, which is found to affect the melt pool size that forms and increases the tendency of the $<110\rangle$ texture formation. ${ }^{[59]}$ Also, Sofras et al. ${ }^{[60]}$ showed that decreased laser power input and lower scan speed combination (low VED) can produce strong $<110>$ texture in SS304L. Another shrewd means is taking advantage of the inclined growth of grains when the laser beam moves away from the molten pool. Recent studies show that while the $<100>$ direction of the FCC lattice aligns with the inclined heat flow, the $<110\rangle$ crystallographic texture consequently aligns along the build direction. This results in a strong $<110>$ texture resembling a single crystal-like parallel to the build direction. ${ }^{[60,61]}$

The influence of hatch distance on texture is mainly by affecting the temperatures of the build process. For instance, short hatch distance increases the peak temperatures for scanning tracks. ${ }^{[2]}$ Therefore, consecutive scans with smaller hatch distances would lead to increased scan track temperatures and a larger temperature gradient along the building direction. This increases preferential growth along the build direction. ${ }^{[62]}$ Nadammal et al. ${ }^{[63]}$ observed that shorter hatch distance yields stronger texture, particularly in the $<001>$ direction parallel to the overall

Table 1. Crystallographic texture evolution, processed material, attributed LPBF parameter(s), and influence on mechanical, corrosion, and oxidation resistance properties.

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-08}
\end{center}

Table 1. Continued.

\begin{center}
\begin{tabular}{|c|c|c|c|c|}
\hline
Reference & Resulting texture & Material & Attributed LPBF Parameter(s) & Comments \\
\hline
\multirow[t]{2}{*}{[52]\{\}} & \begin{tabular}{l}
$<100>$ fiber texture along the \\
scanning direction \\
\end{tabular} & AlSi10Mg & Laser power & \begin{tabular}{l}
Both columnar and equiaxed grains were \\
observed. \\
\end{tabular} \\
\hline
 &  &  &  & \begin{tabular}{l}
Fine zone, coarse zone, and heat-affected \\
zone characterize the microstructure \\
across the melt pool. \\
\end{tabular} \\
\hline
\multirow[t]{2}{*}{[53]} & \begin{tabular}{l}
$<100>$ along the building direction (with \\
$\qquad P=1000 \mathrm{~W}$ ) \\
\end{tabular} & SS316L & Laser power & \begin{tabular}{c}
Samples fabricated with higher power \\
$(1000 \mathrm{~W})$ showed improved mechanical \\
properties compared. \\
\end{tabular} \\
\hline
 & Near random (with $P=400 \mathrm{~W}$ ) &  &  & \begin{tabular}{l}
Elongation of LPBFed samples better than \\
cast and wrought materials. \\
\end{tabular} \\
\hline
$[55]$ & $<100>$ parallel to build direction & IN718 & Laser power and power profile shape & \begin{tabular}{l}
Tensile properties of LPBFed samples \\
were superior to cast samples but inferior \\
to the properties of wrought sample. \\
\end{tabular} \\
\hline
\multirow[t]{3}{*}{[47]} & \begin{tabular}{c}
SCM with $<110>$ along the building \\
direction \\
\end{tabular} & IN718 & Bidirectional laser scanning strategy & \begin{tabular}{l}
CLM exhibits a balance in strength and \\
ductility properties. \\
\end{tabular} \\
\hline
 & \begin{tabular}{c}
CLM with $a<110>$ oriented main layer \\
and $<100>$-oriented sublayer in the \\
building direction \\
\end{tabular} &  & Energy density &  \\
\hline
 & Polycrystalline with a weak orientation &  &  &  \\
\hline
$[56]$ & \begin{tabular}{l}
$<100>$ fiber-texture along the build \\
direction \\
\end{tabular} & \begin{tabular}{l}
High silicon \\
steel parts \\
\end{tabular} & \begin{tabular}{l}
Bidirectional scan strategy (XY-scan) \\
including variation in scan speed \\
\end{tabular} & \begin{tabular}{l}
Formation of the $<001>$ crystallographic \\
texture promotes easy magnetization of \\
the alloy. \\
\end{tabular} \\
\hline
[57] & $<011>$ parallel to the building direction & SS316L & X-scan strategy & \begin{tabular}{l}
Simultaneous improvement in tensile \\
strength $(\approx 16 \%$ ) and ductility ( $\approx 40 \%$ ) \\
\end{tabular} \\
\hline
\multirow[t]{2}{*}{[67]} & SCM & SS316L & X-scan strategy &  \\
\hline
 & \begin{tabular}{c}
CLM $<011>+<001>$ in building \\
direction \\
\end{tabular} &  & laser scan speed &  \\
\hline
\multirow[t]{15}{*}{[103]} & $<001>$ along the transverse direction & SS316L & Laser hatch style (double stagger melt) & Yield strength (MPa) \\
\hline
 & (TD) &  & \begin{tabular}{l}
and part placement strategy \\
(at $0^{\circ}, 45^{\circ}, 60^{\circ}$, and $90^{\circ}$ ) \\
\end{tabular} & $0^{\circ}-494.8 \pm 17.6$ \\
\hline
 &  &  &  & $45^{\circ}-498.6 \pm 13.5$ \\
\hline
 &  &  &  & $60^{\circ}-536.6 \pm 18.9$ \\
\hline
 &  &  &  & $90^{\circ}-489.9 \pm 17.2$ \\
\hline
 &  &  &  & Tensile strength, (MPa) \\
\hline
 &  &  &  & $0^{\circ}-640.3 \pm 23.6$ \\
\hline
 &  &  &  & $45^{\circ}-606.4 \pm 19.2$ \\
\hline
 &  &  &  & $60^{\circ}-601.7 \pm 21.4$ \\
\hline
 &  &  &  & $90^{\circ}-548.3 \pm 18.8$ \\
\hline
 &  &  &  & Elongation (\%) \\
\hline
 &  &  &  & $0^{\circ}-56.7 \pm 1.4$ \\
\hline
 &  &  &  & $45^{\circ}-59.9 \pm 1.7$ \\
\hline
 &  &  &  & $60^{\circ}-62.6 \pm 2.1$ \\
\hline
 &  &  &  & $90^{\circ}-43.7 \pm 1.5$ \\
\hline
\multirow[t]{2}{*}{[60]\{\}} & \begin{tabular}{l}
$\langle 110\rangle,\langle 100\rangle$, and $\langle 111\rangle$ along \\
loading direction of tensile test samples ${ }^{\text {a) }}$ \\
\end{tabular} & SS304L & \begin{tabular}{l}
Chess-board scanning pattern \\
$\left(4 \times 4 \mathrm{~mm}^{2}\right.$ squares) with $90^{\circ}$ rotation \\
and 1 mm shifts in $x$ and $y$ with respect to \\
the previous layer. \\
\end{tabular} & \begin{tabular}{c}
$\sigma_{Y(100\rangle}, \mathrm{UTS}_{\langle 100\rangle}<$ \\
$\sigma_{\gamma(110\rangle}, \mathrm{UTS}_{\langle 110\rangle}$ \\
$<\sigma_{\gamma(111\rangle}, \mathrm{UTS}_{\langle 111\rangle}$ \\
\end{tabular} \\
\hline
 &  &  & Power and scan speed variation &  \\
\hline
[61] & $<100>$ in the scanning direction & SS316L & \begin{tabular}{l}
Bidirectional "serpentine" scan strategy \\
involving laser scan angle variation for \\
each layer. \\
\end{tabular} &  \\
\hline
\end{tabular}
\end{center}

Table 1. Continued.

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-10}
\end{center}

Table 1. Continued.

\begin{center}
\begin{tabular}{|c|c|c|c|c|}
\hline
Reference & Resulting texture & Material & Attributed LPBF Parameter(s) & Comments \\
\hline
 & \begin{tabular}{l}
Fiber texture (strategy C), with a strong \\
alignment of $<001>$ parallel to the build- \\
\end{tabular} &  & \begin{tabular}{l}
A. Scan vectors rotate $90^{\circ}$ (parallel/ \\
perpendicular to the reference direction) \\
\end{tabular} & \begin{tabular}{l}
Scanning strategy, A-148 GPa (at $23^{\circ} \mathrm{C}$ ) \\
and 108 GPa (at $850^{\circ} \mathrm{C}$ ) \\
\end{tabular} \\
\hline
 & \begin{tabular}{l}
up direction and a rather uniform \\
distribution of $<100>$ and $<110>$ in the \\
\end{tabular} &  & \begin{tabular}{l}
B. Scan vectors rotate $90^{\circ}\left(+45^{\circ} /-45^{\circ}\right.$ to \\
the reference direction) \\
\end{tabular} & \begin{tabular}{l}
Scanning strategy, B-225 GPa (at $23^{\circ} \mathrm{C}$ ) \\
and 164 GPa (at $850^{\circ} \mathrm{C}$ ) \\
\end{tabular} \\
\hline
 &  &  & \begin{tabular}{l}
C. Scan vectors rotate $67^{\circ}$ \\
(pseudo-uniform around the \\
building direction) \\
\end{tabular} & \begin{tabular}{l}
Scanning strategy, $\mathrm{C}-183 \mathrm{GPa}$ (at $23^{\circ} \mathrm{C}$ ) \\
and 134 GPa (at $850^{\circ} \mathrm{C}$ ) \\
\end{tabular} \\
\hline
\multirow[t]{4}{*}{[107]\{\}} & \begin{tabular}{l}
$<100>\{001\}$ scan strategy A parallel to \\
the scanning direction \\
\end{tabular} & Hastelloy X & Scan strategy & \begin{tabular}{l}
Young's modulus (tensile tests \\
perpendicular to the build direction) \\
\end{tabular} \\
\hline
 & \begin{tabular}{c}
$<110>\{001\}$ scan strategy $B$ parallel to \\
the scanning direction \\
\end{tabular} &  & \begin{tabular}{l}
A. Scan vectors rotate $90^{\circ}$ (parallel/ \\
perpendicular to the reference direction) \\
\end{tabular} & \begin{tabular}{l}
Scanning strategy, A-146 GPa (at $23^{\circ} \mathrm{C}$ ) \\
and 99 GPa (at $850^{\circ} \mathrm{C}$ ) \\
\end{tabular} \\
\hline
 & \begin{tabular}{l}
Fiber texture with $<001>$ texture parallel \\
to the building direction (scan strategy C ) \\
\end{tabular} &  & \begin{tabular}{l}
B. Scan vectors rotate $90^{\circ}\left(+45^{\circ} /-45^{\circ}\right.$ to \\
the reference direction) \\
\end{tabular} & \begin{tabular}{l}
Scanning strategy, B-190 GPa (at $23^{\circ} \mathrm{C}$ ) \\
and 129 GPa (at $850^{\circ} \mathrm{C}$ ) \\
\end{tabular} \\
\hline
 &  &  & \begin{tabular}{l}
C. Scan vectors rotate $67^{\circ}$ (pseudo- \\
uniform around the building direction) \\
\end{tabular} & \begin{tabular}{l}
Scanning strategy, $\mathrm{C}-171 \mathrm{GPa}$ (at $23^{\circ} \mathrm{C}$ ) \\
and 118 GPa (at $850^{\circ} \mathrm{C}$ ) \\
\end{tabular} \\
\hline
$[63]$ & \begin{tabular}{l}
Strong Goss texture $\{110\}<001>$ in the \\
hatch direction - scanning direction \\
(X-Y plane) \\
\end{tabular} & IN718 & Hatch length variation & \begin{tabular}{l}
Increase in hatch length tends to decrease \\
texture intensity. \\
\end{tabular} \\
\hline
\multirow[t]{17}{*}{[108]\{\}} & $<110>$ parallel to the build direction & SS316L & Alternating meander stripe with $90^{\circ}$ from & Tensile properties \\
\hline
 &  &  & layer to layer. & Elongation (\%) \\
\hline
 &  &  &  & Sample $0^{\circ}-54-56.5$ \\
\hline
 &  &  &  & Sample $45^{\circ}-53.0-56.5$ \\
\hline
 &  &  &  & Sample $90^{\circ}-53.5-60$ \\
\hline
 &  &  &  & Young modulus (GPa) \\
\hline
 &  &  &  & Sample $0^{\circ}-215 \pm 3$ \\
\hline
 &  &  &  & Sample $45^{\circ}-202 \pm 8$ \\
\hline
 &  &  &  & Sample $90^{\circ}-192 \pm 7$ \\
\hline
 &  &  &  & Yield strength (MPa) \\
\hline
 &  &  &  & Sample $0^{\circ}-581$ \\
\hline
 &  &  &  & Sample $45^{\circ}-563-564$ \\
\hline
 &  &  &  & Sample $90^{\circ}-506-514$ \\
\hline
 &  &  &  & UTS (MPa) \\
\hline
 &  &  &  & Sample $0^{\circ}-689-691$ \\
\hline
 &  &  &  & Sample $45^{\circ}-670-671$ \\
\hline
 &  &  &  & Sample $90^{\circ}-611-620$ \\
\hline
\multirow[t]{13}{*}{[59]} & $\{110\}<001>$ Goss texture on the $X Y$ & SS316L & VED & Tensile properties \\
\hline
 & plane of the cube (SCM) &  &  & Elongation (\%) \\
\hline
 &  &  &  & $<100>$ sample- $36.2 \pm 0.9$ \\
\hline
 &  &  &  & $<110>$ sample- $96.3 \pm 3.0$ \\
\hline
 &  &  &  & $<111>$ sample- $58.5 .4 \pm 2.4$ \\
\hline
 &  &  &  & Yield strength (MPa) \\
\hline
 &  &  &  & $<100>$ sample- $546.1 \pm 10.0$ \\
\hline
 &  &  &  & $<110>$ sample $-495.4 \pm 15.1$ \\
\hline
 &  &  &  & $<111>$ sample- $710.0 \pm 11.6$ \\
\hline
 &  &  &  & UTS (MPa) \\
\hline
 &  &  &  & $<100>$ sample-645.4 $\pm 1.1$ \\
\hline
 &  &  &  & $<110>$ sample- $607.2 \pm 11.3$ \\
\hline
 &  &  &  & $<111>$ sample- $840.8 \pm 10.1$ \\
\hline
\end{tabular}
\end{center}

Table 1. Continued.

\begin{center}
\begin{tabular}{|c|c|c|c|c|}
\hline
Reference & Resulting texture & Material & Attributed LPBF Parameter(s) & Comments \\
\hline
\multirow[t]{3}{*}{[94]\{\}} & $<100>$ along the scanning direction & \begin{tabular}{l}
Ti-45 wt\% \\
Nb (30 at\%) \\
\end{tabular} & \multirow[t]{3}{*}{}\begin{tabular}{l}
Bidirectional scanning without \\
rotation between layers \\
\end{tabular} & \multirow{3}{*}{}\begin{tabular}{l}
Average elastic modulus in all three \\
directions range from 56 to 59 GPa , hence \\
suitable for biomedical applications. \\
\end{tabular} \\
\hline
 & $<100>$ along the build direction & $(\beta$-Titanium &  &  \\
\hline
 & $<100>$ in the perpendicular direction & alloy) &  &  \\
\hline
\end{tabular}
\end{center}

a) $\mathrm{T} 1<110>, P=150 \mathrm{~W}, \nu=450 \mathrm{~mm} \mathrm{~s}^{-1}, d=111 \mu \mathrm{m}, t=30 \mu \mathrm{m} ; \mathrm{T} 2<100>, P=175 \mathrm{~W}, \nu=600 \mathrm{~mm} \mathrm{~s}^{-1}, d=83 \mu \mathrm{m}, t=30 \mu \mathrm{m} ; \mathrm{T} 3<111>, P=100 \mathrm{~W}, \nu=300 \mathrm{~mm} \mathrm{~s}$ $d=111 \mu \mathrm{m}, t=30 \mu \mathrm{m}$.

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-12}
\end{center}

Temperature Gradient, G

Figure 7. Variation of solidification texture of pure aluminum as a function of $G / R$ ratio on the solidification map. Reproduced under terms of the Creative Commons CC BY 4.0 license. ${ }^{[102]}$ Copyright, The Authors. Published by Journal of Japan Institute of Light Metals.

maximum heat flow direction. A combination of high laser power, low scan speed, and a short hatch distance produce deeper melt pools, which are generally ascribed to support the $<100>$ epitaxial growth. ${ }^{[64]}$ Similarly, when low laser power, a high scan speed, and longer hatch lengths are applied, broad and shallow melt pools form which also support the $<100>$ texture evolution. ${ }^{[64]}$ Nonetheless, these larger melt pools exhibit less depth and hence are characterized by finer grains between and within the columnar grains. The formation of the finer grains interrupts the $<100>$ texture epitaxial growth leading to texture reduction.

The aforementioned studies all show that the laser power input and scan speed influence texture by mainly affecting the melt pool size. Although high VED promotes the $<001>$ family texture development, the high laser power input involved has the tendency to increase the overall temperature of the LPBF process. The increased temperature of the LPBF process means rates of the keyhole melting mode increase, forming deeper melt pools and also causing instabilities in the melt pools. ${ }^{[65]}$ The increased instabilities in the melt pool cause periodic collapse of keyholes leading to many pore formations. Moreover, high VED in various studies has been found to be the main parameter that affects cracking. ${ }^{[66]}$ On the other hand, increasing scan speed lengthens and decreases the melt pool temperature. ${ }^{[65]}$ This enhances solidification rates but forms smaller grains which reduces texture evolution. Optimizing powder layer thickness is also essential in texture evolution because it influences the intensity of the required laser power input. Insufficient laser power input used in melting thicker powder layers means more powder particles will remain unmelted, which will impede epitaxial texture growth and also form pores. Insufficient hatch distance and scan strategies with more partitions affect texture in the overlapping zones of melt pools through deflection. This can be attributed to their influence on residual stresses during the LPBF process.

In summary, it is clear from this section that the evolution and control of crystallographic texture heavily depend on the synergistic effects of the laser power, laser scan speed, hatch distance, and powder layer thickness. However, the existence of some contradicting findings and the difficulty in attaining appropriate process parameters suggest that substantial research gaps still exist in optimizing the LPBF process parameters and understanding their influence on underlying mechanisms of texture evolution.

\subsection*{4.2.2. Effect of Scan Strategy}
Scan strategy is another parameter that strongly influences texture formation by controlling the local heat flow direction and competitive grain growth. The unidirectional ${ }^{[39]}$ or bidirectional ${ }^{[67]}$ scanning strategy without scan direction rotation in subsequent layers (X-scan) is likely to lead to the $<001>$ family crystallographic texture formation. The $<001>$ crystallographic texture evolves because the laser trajectories remain the same in each layer. For instance, in the unidirectional scanning, the principal heat flow directions at the solid-liquid interface occur at an angle of about $60^{\circ}$ relative to the substrate. ${ }^{[68]}$ As such, the resulting thermal flux becomes homogenous for every layer, thereby promoting the epitaxial growth of the $<001>$ crystal orientation. ${ }^{[99]}$ However, in bidirectional scanning, for successful epitaxial growth of the $<100>$ texture, grains orientation deviates at a $15^{\circ}$ angle to the maximum local heat flow direction. A fiber texture evolves when the grains perfectly align with the heat flow direction, while misalignment results in a cube texture. ${ }^{[68]}$ Also, a bidirectional scan strategy with $90^{\circ}$ rotation in layers (XY-scan) has been shown to also result in strong $<001>$ textures in the $x$-, $y$-, and $z$-directions. ${ }^{[49]}$ It is argued that the XY-scan strategy supports strong $<100>$ cube texture development, rather than crystal nucleation because of the small angular mismatch owing to the rotations. ${ }^{[70]}$ The common explanation that can be drawn from the above-mentioned studies is that preferential\\
grain growth, particularly in cubic materials, essentially controls the $<001>$ texture evolution.

The $<110>$ family crystallographic texture may also form from unidirectional scanning with no layer rotations. ${ }^{[64,71]}$ Likewise, the bidirectional scan strategy with $67^{\circ}$ and $180^{\circ}$ rotation (XY-scan) can yield a $<011>$ family crystallographic texture. ${ }^{[61,72]}$ In both the X- and XY-scan strategies mentioned here, the preferential orientation of the $<011>$ texture along the build direction is a result of the $\pm 45^{\circ}$ growth in the $<100>$ equilibrium cells in the building plane. As previously stated, the $\pm 45^{\circ}$ growth direction of the elongated grains occurs to facilitate epitaxial growth which consequently yields the $<011>$ texture.

Moreover, a bidirectional scan strategy with either $55^{\circ}, 67^{\circ}$, $90^{\circ}$, or $120^{\circ}$ scan direction rotation enables the evolution of the $<111>$ crystallographic texture. ${ }^{[19,61,69]}$ The rotation of scan direction between the layers suggests thermal flux directions also rotate, causing fewer remelted grains to have an optimal crystal orientation. In addition, the scan direction rotation in these angles promotes grain selection by competitive growth mechanism. This can be attributed to the increase in thermal gradient between the layers which weakens texture and perhaps favors the development of the $<111>$ crystallographic texture (Figure 7).

Crystallographic texture evolution can also be controlled using the island scan strategy. Compared to the unidirectional (X-) and bidirectional (XY-) scan strategies, the island scan strategy tends to reduce the overall residual stresses because the heat input is uniformly distributed across the surface. ${ }^{[73]}$ The use of smaller island sizes favors the epitaxial growth of the $<001>$ texture. ${ }^{[74]}$ The short distance traveled by the laser for each scan means the smaller islands will have more heat continuously applied (or accumulated) over a longer period. This causes grains to elongate in the maximum heat flow direction parallel to the building direction signaling the epitaxial growth of the $<001>$ texture. An advantage of using the island scan strategy to control texture compared to heat input parameters, particularly hatch distance is the effect on part geometry. Part geometry usually influences hatch distance. For example, ${ }^{[74]}$ when the rastering scan strategy is used for large areas, the hatch distance becomes longer. Whereas if the island scan is used, the island size could be varied circumventing the large area problem presented by the geometry of the part. Rather, heat input is uniformly distributed over large areas and texture can be duly controlled.

\subsection*{4.2.3. Crystallographic Lamellar-Like Microstructure (CLM), Single Crystal-Like Microstructure (SCM), and Polycrystalline-Like Crystallographic Microstructure (PCM) Texture}
This section presents some uniquely developed crystallographic textures generated through scan strategy and careful control of other process parameters such as $P, v, d$, and $t$. As shown in Figure 8, they include the single crystal-like microstructure (SCM), crystallographic lamellar-like microstructure (CLM), and polycrystalline-like microstructure (PCM). ${ }^{[47,49,50,67,75]}$

In the SCM texture, cellular growths in the melt pools are identical, hence a single-crystal orientation develops throughout the build, whereas for CLM, multiple textures evolve in a characteristic lamellar pattern. Ishimoto et al. ${ }^{[49]}$ produced the SCM in Ti-15Mo-5Zr-3Al (wt\%) alloy by utilizing the X- and XY-scan strategies as shown in Figure 9. The X-scan strategy under appropriate processing conditions resulted in SCM with strong $<001>$ and $<011\rangle$ crystallographic textures preferentially oriented in the scanning $(x)$ and building $(z)$ directions, respectively (Figure 9a). In the case of the XY-scan strategy, the SCM formed in the $x$-, $y$-, and $z$-directions as shown in Figure 9b. For the\\
(a)

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-14(1)}
\end{center}

(c)

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-14(2)}
\end{center}

(b)

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-14}
\end{center}

(d)

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-14(3)}
\end{center}

SD : scanning directio

\begin{center}
\begin{tabular}{|c|c|}
\hline
$\uparrow \underset{\text { oradient directiona }}{\text { Maximum therma }}$ & - Growth direction of \\
\hline
\end{tabular}
\end{center}

Figure 9. Schematic of $\mathrm{a}, \mathrm{b}$ ) scan strategy and corresponding texture and $\mathrm{c}, \mathrm{d}$ ) their evolution mechanisms during L-PBF of bcc-Ti-15Mo-5Zr-3Al alloy using the (a,c) X-scan or (b,d) XY-scan method. Reproduced under terms of the Creative Commons CC BY 4.0 license. ${ }^{[6]}$ Copyright 2021, The Authors. Published by Springer Nature.

XY-scan strategy, both the initial X- and subsequent Y-scans showed similar cellular growths around the melt pool bottom and upper walls (Figure 9d), and the $<001>$ elongated cells evolved and grew parallel to the building direction. Thus, in the X- and XY-scan strategies, the $<001>$ direction became fixed owing to the two orthogonal directional growths of the elongated cells in the plane perpendicular to the scanning direction (Figure 9c,d). The observed growth behavior in both scan strategies resulted in the SCM crystallographic texture. Sun et al. ${ }^{[50]}$ as presented in Figure 10a, observed the simultaneous evolution and growth of the $<001>$ and $<011>$ textures alternatively stacked like the lamellar pattern in the sample produced with low energy density. In the sample fabricated, the $<100>$ crystallographic orientation was the major orientation in the $x$-axis, and $<1 \overline{1} 0>$ was minor; in the $y$-axis, $<1 \overline{1} 0>$ was the major and $<110>$ the minor; and in the $z$-axis, $<011>$ was the major and parallel to the $<001>$ minor orientation (Figure 10d). The periodicity of the lamellar structure was observed to match with the laser hatch distance which is unique to the LPBF process amongst other AM processes (Figure 10c). Additionally, for the PCM condition, a randomized crystallographic orientation is observed while texture is maintained to a certain extent. The PCM texture usually evolves if the melt pool depth is not deep enough to remelt already solidified layers (at the fusion lines) to promote epitaxial growth. The lack of epitaxial growth results in the solidification of the melt pools forming grains with randomized crystal orientations.

\subsection*{4.2.4. Influence of Laser Remelting on Crystallographic Texture Development}
Laser remelting is when the fusion line of already deposited layers is partially remelted during subsequent layer deposition or the sides of previous melt pools owing to the hatch distance are partially remelted resulting in an overlap. In some instances, laser remelting involves the intentional application of lower laser power to partially remelt the deposited layers before the actual subsequent layer deposition. Figure 11 elaborates on the influence of overlap and laser remelting on texture growth. ${ }^{[76]}$

Herein, we use the study by Shao et al. ${ }^{[76]}$ to elucidate the influence of laser remelting on crystallographic texture development. On the influence of the melt pool width on texture evolution, the study demonstrated that with hatch distance of 0.11 mm leading to a $19 \%$ overlapping rate and with or without laser remelting yielded the CLM texture. The major layer observed was the $<011>$ texture, and the $<001>$ was the minor texture oriented parallel to the build direction. As Figure 11a demonstrates, the $<001>$ texture growth occurred at the center of the melt pool parallel to the build direction while the $<011>$ grains oriented $\pm 45^{\circ}$ from the melt pool sides of the overlapping layer. When the hatch distance decreased to $0.07 \mathrm{~mm}(49 \%$ overlapping rate), the $<011>$ crystal orientation deflected to $<112>$. Further decrease in hatch distance $(0.05 \mathrm{~mm}, 59 \%$ overlapping rate) almost completely deflected the $<011>$ texture to the $<111>$ texture to form a CLM texture in the build direction.

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-15}
\end{center}

Figure 10. Crystal orientation maps in the specimens fabricated in X-scan strategy with a) low energy density and b) high energy density, observed along the $x$-, $y$-, and $z$-directions, respectively, taken by SEM-EBSD. c) Misorientation angle variation along line AB in (a). d) Schematic illustration showing the crystallographic orientation relationship between major and minor layers in the CLM formed under low energy density. Reproduced under terms of the Creative Commons CC BY 4.0 license. ${ }^{[50]}$ Copyright 2018, The Authors. Published by Elsevier Ltd on behalf of Acta Materialia Inc.

As generally known and observed in the above study, the LPBF energy input increased with decreasing hatch distance. This consequently generates a high thermal gradient in the solidification front. Additionally, the width of the melt pool which is altered by hatch distance influences the thermal gradient as well. Hence increasing the overlapping rate means more heat input, likely altering the grain orientation in the overlapping zone, and influencing the thermal gradient of the adjacent melt pool. ${ }^{[62,74]}$ The epitaxial growth of grains at the centerline in the $<001>$ direction remains unaffected if less than half of the adjacent melt pool is remelted or overlapped. When the overlapping is more than half, the $<001>$ growth is interrupted, and the $<011>$ texture at the overlapping zone is deflected to the scanning direction because of the high temperature of the previous track. This results in the nucleation and grain growth along the maximum thermal gradient forming the $<111>$ texture.

Furthermore, laser remelting before the deposition of subsequent layers removes heterogeneous nucleation sites and solidification barriers at the fusion line of already solidified layers. The laser remelting process and accompanying lower nucleation rate supported epitaxial growth. In the $19 \%$ overlapping rate, the laser remelting process favored the epitaxial growth of the $<001>$ texture at the centerline of the melt pools parallel to the build direction. Simultaneously, the elliptical shape of the formed melt pool had grain growth oriented in the direction of the thermal gradient $(<011\rangle$ ) from the melt pool sidewalls. In the layers with $50 \%$ melt pool overlapping rate and more, the laser remelting process formed broad and shallow melt pools governed by the epitaxial growth of grains with small angle between the crystal orientation and thermal gradient. At the front of the solidification interface, the thermal gradient in the direction of the $<001>$ prevailed. Thus, grains aligned near the $<001>$ direction parallel to the build direction grew faster. It was noticed that after several layers deposited and repeated interlayer laser remelting, competitive growth caused more grains to align and epitaxially grow in the $<001>$ direction. In all, the laser remelting process is considered to influence crystallographic texture development by selectively promoting the preferential orientation of grains and epitaxial growth.

\section*{5. Texture Control via Phase Transformation}
Controlling texture in alloys that undergo solid-state phase transformation is currently one of the challenges encountered by the LPBF process. A primary example is the $\alpha+\beta$ titanium (Ti) alloys (Ti-6Al-4V (wt\%), Ti-6Al-2Sn-4Zr-6Mo (wt\%)). Many

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-16}
\end{center}

(b)\\
\includegraphics[max width=\textwidth, center]{2024_07_13_01fad27876b5c2169f6dg-16(1)}

Figure 11. Schematic illustration of the effect of overlap and laser remelting on the growth texture with a,g) S1, b) S2, c, h) S3, d, i) R1, e) R2, and f,j) R3. The arrows indicate the trend direction of dendrites growth, the blue arrows, green arrows, and red arrows represent the direction of $<111\rangle,<011>$, and $\langle 001>$, respectively, and the pink ellipses represent the surface nucleation. LPBF process parameters include: alloy $=\mathrm{IN} 718, P=285 \mathrm{~W}, v=960$ $\mathrm{mm} \mathrm{s}^{-1}, t=40 \mu \mathrm{m}, \sigma=90 \mu \mathrm{m}, \mathrm{S} 1(d=0.11 \mathrm{~mm}), \mathrm{S} 2(d=0.07 \mathrm{~mm}), \mathrm{S} 3(d=0.05 \mathrm{~mm}), \mathrm{R} 1(d=0.11 \mathrm{~mm} ; 150 \mathrm{~W}$ laser remelting), R2 $(d=0.07 \mathrm{~mm}$; 150 W laser remelting), R3 ( $d=0.05 \mathrm{~mm}$; 150 W laser remelting). Reproduced with permission. ${ }^{[76]}$ Copyright 2022, Elsevier Ltd.

studies have shown that the first layer of these $\alpha+\beta \mathrm{Ti}$ alloys during the LPBF process solidifies as $\mathrm{L} \rightarrow \beta \rightarrow \alpha^{\prime}+$ retained $\beta .{ }^{[77,78]}$ The rapid cooling rates and multiple thermal cycles suggest that solid-state transformation occurs in two forms during the LPBF process. First, the rapid cooling rates of the LPBF process favor a displacive phase transformation mechanism resulting in martensitic microstructure formation. ${ }^{[79]}$ The martensitic microstructure, depending on composition, is made up of laths or acicular (or needle) precipitate phases, with high density of lattice defects such as stacking faults, dislocations and nano-twins, and can be highly crystallographic. ${ }^{[7]}$ These solid-state $\beta \rightarrow \alpha^{\prime}$ transformations are governed by the Burger's orientation relationship (OR) wherein the $\alpha$ laths or precipitates exhibit a specific crystallographic orientation relationship with the BCC $\beta$ matrix phase. The second group of solid-state phase transformations is diffusioncontrolled wherein the high-temperature $\beta$ phase is metastable and transforms with slow cooling. ${ }^{[77]}$ This solid-state $\beta \rightarrow \alpha^{\prime}$ (or $\alpha$ ) transformation occurs from the multiple thermal cycles or reheating during subsequent layers deposition which can cause decomposition and randomization of crystallographic texture. It is argued that despite the repeated reheating during the LPBF process, most martensitic products retain their crystallography because the reheating do not cause transformations to again start

Table 2. Twelve possible variant selections by the $\beta \rightarrow \alpha$ martensitic phase transformation through the Burgers orientation relationship (or). Reproduced with permission. ${ }^{[1]}$ copyright 2003, Elsevier Ltd.

\begin{center}
\begin{tabular}{lccc}
\hline
Variant & Plane parallel & Direction parallel & Rotation angle/axis from V1 \\
\hline
V1 & $(1 \overline{1} 0)_{\beta} \|(0001)_{\alpha}$ & $[111]_{\beta} \|[11 \overline{2} 0]_{\alpha}$ & - \\
V2 & $(10 \overline{1})_{\beta} \|(0001)_{\alpha}$ & $[111]_{\beta} \|[11 \overline{2} 0]_{\alpha}$ & $60^{\circ} /[11 \overline{2} 0]$ \\
V3 & $(01 \overline{1})_{\beta} \|(0001)_{\alpha}$ & $[111]_{\beta} \|[11 \overline{2} 0]_{\alpha}$ & $60^{\circ} /[11 \overline{2} 0]$ \\
V4 & $(110)_{\beta} \|(0001)_{\alpha}$ & $[\overline{1} 11]_{\beta} \|[11 \overline{2} 0]_{\alpha}$ & $90^{\circ} /[1 \overline{2.38} 1.380]$ \\
V5 & $(101)_{\beta} \|(0001)_{\alpha}$ & $[\overline{1} 11]_{\beta} \|[11 \overline{2} 0]_{\alpha}$ & $63.26^{\circ} /[\overline{10} 55 \overline{3}]$ \\
V6 & $(01 \overline{1})_{\beta} \|(0001)_{\alpha}$ & $[\overline{1} 11]_{\beta} \|[11 \overline{2} 0]_{\alpha}$ & $60.83^{\circ} /[\overline{1.377} \overline{1} 2.3770 .359]$ \\
V7 & $(110)_{\beta} \|(0001)_{\alpha}$ & $[1 \overline{1} 1]_{\beta} \|[11 \overline{2} 0]_{\alpha}$ & $90^{\circ} /[1 \overline{2.38} 1.380]$ \\
V8 & $(10 \overline{1})_{\beta} \|(0001)_{\alpha}$ & $[1 \overline{1} 1]_{\beta} \|[11 \overline{2} 0]_{\alpha}$ & $60.83^{\circ} /[\overline{1.377} \overline{1} 2.3770 .359]$ \\
V9 & $(011)_{\beta} \|(0001)_{\alpha}$ & $[1 \overline{1} 1]_{\beta} \|[11 \overline{2} 0]_{\alpha}$ & $63.26^{\circ} /[\overline{10} 55 \overline{5}]$ \\
V10 & $(1 \overline{1} 0)_{\beta} \|(0001)_{\alpha}$ & $[11 \overline{1}]_{\beta} \|[11 \overline{2} 0]_{\alpha}$ & $10.53^{\circ} /[0001]$ \\
V11 & $(101)_{\beta} \|(0001)_{\alpha}$ & $[11 \overline{1}]_{\beta} \|[11 \overline{2} 0]_{\alpha}$ & $60.83^{\circ} /[\overline{1.377} \overline{1} 2.3770 .359]$ \\
V12 & $(011)_{\beta} \|(0001)_{\alpha}$ & $[11 \overline{1}]_{\beta} \|[11 \overline{2} 0]_{\alpha}$ & $60.83^{\circ} /[1.377 \overline{1} 2.3770 .359]$ \\
\hline
\end{tabular}
\end{center}

from the $\beta$ phase field. ${ }^{[80]}$ In the crystallography of the $\beta \rightarrow \alpha$ martensitic transformation, each $\beta$ grain can transform to 12 possible $\alpha$ variants ${ }^{[81]}$ as shown in Table 2, reiterating the difficulty of controlling texture in these alloys. So far, four mechanisms have been used to explain the variant selection or texture development in these allotropic alloys. They include ${ }^{[79]} 1$ ) the stress retained in the high-temperature $\beta$ phase, owing to the substantial volume and shape change on heating and/or externally applied deformation, favors specific transformation product characteristics on cooling, ${ }^{[82]}$ 2) the presence of metastable $\alpha$-phases at high temperatures, which act as nuclei for growth on cooling, ${ }^{[83]} 3$ ) the formation of specific variant cluster arrangements to accommodate strains produced by the $\beta \rightarrow \alpha$ transformation, and ${ }^{[77,81]}$ 4) the presence of specific grain boundary characteristics in the high-temperature $\beta$ phase, enhancing the nucleation of certain orientations (or variants) during phase transformation. ${ }^{[84]}$

Since controlling texture using the LPBF has proven difficult thus far in these allotropic alloys, some studies have resorted to the use of post processing treatments such as annealing and solution heat treatments to control the extent of variant selection or texture development. ${ }^{[85,86]}$ The application of these post processing treatments gives the diffusion process enough time and usually results in a randomized crystallographic texture. ${ }^{[85,86]}$ This is because additively manufactured components exhibit a different recovery and recrystallization behavior compared to the conventionally processed counterparts. ${ }^{[86]}$ For instance, the recrystallization kinetics in LPBFed-AISI 316 L is sluggish and reported to only commence at higher temperature ( $>1100^{\circ} \mathrm{C}$ ). For further insights into the recrystallization kinetics and mechanisms in LPBFed parts, readers may refer to ref. [86].

\section*{6. Texture Control in Refractory Metals and Alloys}
Refractory materials are well known for their high hardness, high-temperature strength and ductility, and oxidation resistance capabilities. ${ }^{[87]}$ Nonetheless, refractory metals and alloys at room temperature are brittle and possess extremely high melting temperatures. ${ }^{[87]}$ The low ductility at room temperature and high melting temperatures make processing refractory alloys by conventional means such as casting and forging challenging. The LPBF process has successfully been used to fabricate refractory alloy parts. ${ }^{[19,88]}$ Despite the high melting temperatures of refractory metals and alloys, crystallographic texture control has been found to be feasible in these alloys. For instance, Thijs et al. ${ }^{[69]}$ produced a $<111>$ texture along the build direction of tantalum (Ta) samples with high energy density inputs using the LPBF process. Sidambe et al. ${ }^{[19]}$ achieved a $\left.<111\right\rangle$ texture in the build direction of LPBFed-tungsten (W). Gokcekaya et al. ${ }^{[51]}$ used high energy density to form strong $<100>$ texture along the build direction in pure chromium (Cr) samples. Furthermore, Higashi and Ozaki ${ }^{[58]}$ showed that in LPBF processing of pure molybdenum (Mo), an increase in scan speed from 400 to $800 \mathrm{~mm} \mathrm{~s}^{-1}$ at 350 W changes strong $<110>$ fiber texture to $<001>$ along the build direction. Todo et al. ${ }^{[15]}$ showed that eliminating high-angle grain boundaries (HAGB) through careful control of crystallographic texture in forming the $<011>$ single crystalline-like texture improves W densification (99.1\%). HAGB prevention according to the authors suppresses crack formation, many of which were observed in previous work such as Sidambe et al. ${ }^{[19]}$ on LPBF of W with a $<111>$ crystallographic texture development along the build direction.

\section*{7. Controlled Texture Effect on LPBF-Built Material Properties}
Many material properties are anisotropic. This anisotropy stems from the directional properties of single crystals and also depends on the degree of preferred crystal orientations or texture of the polycrystalline material. ${ }^{[89]}$ Hence, the deliberate control of texture in a particular direction can be exploited to enable enhanced material properties. This section (and Table 1) presents the effects of texture control on material properties such as mechanical properties and corrosion and oxidation resistance.

\subsection*{7.1. Mechanical, Corrosion, and Oxidation Resistance Properties}
Young's modulus is well known to strongly depend on crystallographic orientation. Generally, the magnitude of Young's modulus hinges on atomic arrangement along a specific crystallographic direction. This is because of the direct relationship of Young's modulus to interatomic forces density and strength. ${ }^{[90]}$ The elastic modulus for an FCC system and single crystal with respect to interatomic distance thus follows the order $\mathrm{E}_{[111]}>\mathrm{E}_{[110]}>\mathrm{E}_{[100]}{ }^{[91]}$ By employing different scan strategies in LPBF processing of SS316L, Marattukalam et al. ${ }^{[20]}$ observed a similar Young's modulus order, where bidirectional scanning with $67^{\circ}$ rotation of layers leading to a $<111>$ texture produced an elastic modulus of $193 \pm 16 \mathrm{GPa}$. The bidirectional scanning in the y-direction perpendicular to the raking direction resulted in a $<110>$ texture with an elastic modulus of $167 \pm 3 \mathrm{GPa}$ while bidirectional scanning in the $z$-direction, i.e., along the rake direction of the powder bed produced $129 \pm 3 \mathrm{GPa}$. The yield\\
strength showed a similar trend by increasing from $554 \pm 5 \mathrm{MPa}$ (Z-scan with (100) texture) to $603 \pm 2 \mathrm{MPa}$ (67-rot-scan with fiber texture).

The significance of tailored crystallographic texture on deformation behavior is clearly noticed when a $<011>$ growth orientation is achieved in the build direction of SS316L instead of the preferential $<001>$ texture expected for an FCC alloy. Sun et al. ${ }^{[57]}$ observed a drastic improvement in tensile strength (by about $16 \%$ ) and ductility ( $40 \%$ ) in the $<011>$ aligned SS316L compared to the industrial grade. ${ }^{[57]}$ The improved strength was ascribed to the activation of more nano-twins which allowed higher strains during the plastic deformation in the $<011>$ direction. ${ }^{[57]}$ It is also worth noting that in stainless steels such as the duplex stainless steels, crystallographic texture can exist between the FCC austenite phase usually oriented in the $<001>$ and(or) $<111>$ directions and the BCC ferrite phase which preferentially orient in the $<011>$ direction. These texture orientations result from phase transformation during cooling and reheating during deposition of subsequent layers. ${ }^{[22]}$ The existence of these orientations facilitates the propagation of dislocation slip from one phase to the other and can reduce the distribution of residual stresses. ${ }^{[92]}$ Again, in BCC alloys such as tungsten (W), the [110] direction is reported to exhibit the highest tensile strength compared to the [111] and [001] directions. The observed strength in the [110] direction has been attributed to the closely packed crystalline plane of the (110) compared to the (100) and (111) planes which lead to stronger interactions of atoms. ${ }^{[93}$

Also, the formation of the CLM texture has been reported to demonstrate higher yield stress (about 12\%) compared to the SCM texture. ${ }^{[50]}$ The higher yield stress exhibited by the CLM texture is surmised to be a result of the alternating layer's boundaries impeding dislocation movements. ${ }^{[50]}$

Furthermore, the low Young's modulus of the SCM texture, as shown in the previous paragraphs, could be utilized for biomed ical applications. The $\beta$-Ti alloys such as the $\mathrm{Ti}-15 \mathrm{Mo}-5 \mathrm{Zr}-3 \mathrm{Al}$ $(\mathrm{wt} \%)^{[49]}$ and $\mathrm{Ti}-45 \mathrm{Nb}$ (wt\%) ${ }^{[94,95]}$ are well known for their biocompatibility. Hence, with low Young's modulus of 44.4 GPa and $56-59 \mathrm{GPa}$, respectively, close to the elastic modulus of the human cortical bone ( $\approx 30 \mathrm{GPa}$ ), the LPBFed Ti-15Mo$5 \mathrm{Zr}-3 \mathrm{Al}$ and $\mathrm{Ti}-45 \mathrm{Nb}$ alloys can serve as medical implants and suitably replace the conventional Ti64 (110-120 GPa) alloy usually used.

LPBF has been reported to improve the corrosion resistance of alloys such as SS316L. ${ }^{96]}$ Even though the occurrence of grain boundaries in CLM textured material should negatively affect corrosion resistance because grain boundaries usually serve as initiation sites for corrosion such as pitting corrosion, CLM textured LPBFed-SS316L, however, demonstrated enhanced corrosion resistance in $0.9 \mathrm{wt} \% \mathrm{NaCl}$ solution compared to the industrially processed SS316L reference samples. ${ }^{[50]}$ The observed enhanced corrosion resistance characterized by breakdown potential was more than twice $(1.2 \mathrm{~V}$ ) than that of the industrially processed reference SS316L ( 5 V ). The controlled crystallographic build using LPBF, perhaps, led to the improvement of corrosion resistance of the SS316L alloy, although the fundamental mechanisms involved are yet to be elucidated. ${ }^{[50]}$

Additionally, Tsutsumi et al. ${ }^{[75]}$ observed enhanced corrosion resistance in LPBFed SS316L under physiological conditions regardless of the plane exposed. Although the exposed planes exhibited either a CLM, SCM, or PCM texture with random orientation, the rapid cooling rate associated with the LPBF process accounted for the suppressed inclusion coarsening and resulted in improved corrosion resistance in the fabricated SS316L samples.

Also, high-temperature oxidation resistance is another material property that can be influenced by crystallographic texture. Preferential crystal-oriented builds affect occurring diffusion processes and the formation and growth of surface oxide layers. For instance, the $<100>$ crystal orientation in an LPBFed-Cr with a single crystalline-like texture has been shown to hinder intergranular Cr diffusion as well as slow intergranular oxidation at $800^{\circ} \mathrm{C} .{ }^{[51]}$

\section*{8. Summary, Challenges, and Prospects}
Texture is an additional independent parameter that deserves inclusion in the processing-microstructure-properties relationship governing materials design. Crystallographic texture control using the LPBF process hinges on careful manipulation of process parameters such as scan strategy, laser power, laser scan speed, powder layer thickness, and hatch spacing. These LPBF process parameters ultimately control the heat input and thermal history which consequently influence the melting modes, melt pool geometry, and solidification parameters (thermal gradient, cooling rate, and solidification rate) resulting in specific texture evolution. Additionally, directional solidification owing to epitaxial grain growth when already solidified tracks and layers are partially remelted during subsequent scans of the LPBF process leads to strong texture development in materials.

From the literature presented, high VED from a high laser power input, slow scan speed, short hatch spacing, and scan strategy such as the island scan strategy increase heat input resulting in the keyhole melting mode characterized by two circular flows. The keyhole melting mode with its characteristic severely agitated melt flow produces deep melt pools which usually promote the $<001>$ family texture formation parallel to the build direction and culminate in the columnar microstructure. Conversely, fast scan speed, long hatch spacing, low powder layer thickness, and some scan strategies involving rotations in layers lead to the conduction melting mode with wide and shallow melt pools. The shallow melt pools stymie epitaxial solidification due to inadequate remelting, the existence of inclusions, or fastgrowing nucleation sites at the fusion lines. This leads to the randomization of texture and forms the polycrystalline or equiaxed microstructure. Moreover, unique textures such as the CLM and SCM (with a $<011>$ texture along the build direction) can be achieved by altering particular the scan strategies (X- and XY-scans). In all, including the usual heat input parameters considered in this review, the LPBF is suggested to have about $130^{[97]}$ or $157^{[98]}$ process parameters. Most of these parameters have synergistic effects on each other, which directly or indirectly influence texture evolution. Currently, only the conventional five (5) process parameters are constantly varied and studied with respect to texture and microstructure evolution. This, perhaps, also accounts for the differing explanations given in the numerous studies on the underlying mechanisms involved in texture evolution when using the LPBF process, signaling\\
the present need for further investigations. Additionally, material-related properties (e.g., thermal conductivity, diffusivity, and laser reflectivity) are other factors that require substantial attention to understand their contribution to the control of texture when using the LPBF.

Furthermore, the main challenge in controlling texture using the LPBF process involves optimizing the process conditions over a wide range of possibilities. Optimizing the LPBF process parameters is essential especially since the same material can behave differently under varying conditions. However, currently, the trial-and-error optimization process is tedious and timeconsuming. Therefore, the use of computational models and machine learning programs to optimize and predict texture development and changes during the LPBF process would be of great benefit in the quest to control texture. Additionally, the LPBF process produces refined grains that offer property improvement benefits but sometimes reduce texture development. Analyzing the texture of these extremely refined grains is currently extremely difficult in some alloys owing to the resolution limits of equipment used such as the electron backscatter diffractometer (EBSD). The development of higher-resolution equipment or software capable of nanoscale texture analysis would change the current outlook of texture control and come with numerous application possibilities.

Also, it is worth emphasizing again that crystallographic texture control in alloys that are allotropic or undergo solid-state transformation, particularly during cooling, is still a challenge. Aside from the extensively reported $\mathrm{Ti}-6 \mathrm{Al}-4 \mathrm{~V}$ (wt\%) alloy, the near- $\beta \mathrm{Ti}$ alloys (e.g., Ti6246) is another alloy worth investigating. The Ti6246 is considered $\beta$ stabilized although an $\alpha+\beta$ alloy and is known to contain more equilibrium (or retained) $\beta$ phases than the Ti- $6 \mathrm{Al}-4 \mathrm{~V}$ (wt\%). Thus, achieving a strong $\beta$ texture over the transformed $\alpha$ or $\alpha^{\prime}$ martensitic phases would immensely improve the alloy's low-cycle fatigue and fatigue crack growth resistance comparable or even surpassing the conventionally used $\beta$-forged Ti6246.

Along similar lines, controlling texture in titanium aluminides (TiAl) using the LPBF process is yet to be investigated. TiAl alloys are lightweight intermetallic alloys with high-temperature capabilities and compose of the $\alpha_{2}-\mathrm{Ti}_{3} \mathrm{Al}$ phase (an ordered hexagonal crystal structure, $\mathrm{DO}_{19}$ ) and the $\gamma$-TiAl phase (an ordered facecentered tetragonal structure, $\mathrm{L1}_{0}$ ). However, TiAl alloys are usually susceptible to cracking when processed by the LPBF process. Leveraging knowledge from the use of texture control to improve the LPBF processing of refractory alloys could be applied to LPBF TiAls in an attempt to control cracking and simultaneously attain exceptional directional properties.

Also, to the best of the authors' knowledge, there are no attempts yet on the deliberate control of texture in any of the numerous high entropy alloy or superalloy (HEA or HESA) systems. HESAs are currently the most researched alloy systems owing to their countless application possibilities in comparison to conventional alloys. Although most systems are yet to be understood, attempts of texture control in the few that have so far exhibited exceptional properties ${ }^{[99]}$ would offer valuable insights on alloy design and development and serve as a reasonable guide for future research directions.

In all, intentional crystallographic texture control, either sitespecific or directional, offers a means to obtain unique set of properties or optimize existing properties to meet various engineering demands. One application envisaged is in compressor blades. For instance, to withstand creep at elevated temperatures, parts of the compressor blade subject to centrifugal force caused by rotation could be tailored to exhibit an anisotropic texture and microstructure. ${ }^{[6]}$ While its fixed lower section is also tailored to show isotropic texture and equiaxial grain microstructure to ensure fatigue resistance and a low-temperature toughness. ${ }^{[6]}$ In that sense, understanding texture evolution in more alloys is paramount to exploiting their use for the many existing and advanced engineering applications.

\section*{Acknowledgements}
P.V.C., S.M., and Y.Y.-M. contributed equally to this work. This work was supported by Grants-in-Aid for Transformative Research Area A, 21 H05198, and The Light Metal Educational Foundation (Japan).

\section*{Conflict of Interest}
The authors declare no conflict of interest.

\section*{Keywords}
crystallographic texture, laser powder bed fusion, mechanical properties melt pool shape, microstructure evolution, solidification rate, thermal gradient

Received: May 30, 2023

Revised: October 8, 2023

Published online: October 31, 2023

[1] a) R. B. Figueiredo, T. G. Langdon, J. Mater. Sci. 2022, 57, 5210; b) B. N. Mordyuk, G. I. Prokopenko, Y. V. Milman, M. O. Iefimov, K. E. Grinkevych, A. V. Sameljuk, I. V. Tkachenko, Wear 2014, 319, 84; c) S. D. Nath, H. Irrinki, G. Gupta, M. Kearns, O. Gulsoy, S. Atre, Powder Technol. 2019, 343, 738; d) J. Dai, J. Zhu, C. Chen, F. Weng, J. Alloys Compd. 2016, 685, 784; e) D. Q. Martins, W. R. Osório, M. E. P. Souza, R. Caram, A. Garcia, Electrochim Acta 2008, 53, 2809; f) A. Ghosh, S. Sivaprasad, A. Bhattacharjee, S. K. Kar, Mater. Sci. Eng., A 2013, 568, 61.

[2] a) M. R. Bache, W. J. Evans, Mater. Sci. Eng., A 2001, 319-321, 409 b) S. Birosca, J. Y. Buffiere, F. A. Garcia-Pastor, M. Karadge, L. Babout, M. Preuss, Acta Mater. 2009, 57, 5834; c) J. S. Weaver, I. Rosenthal Advanced Manufacturing Series (NIST AMS), U.S. Department of Commerce 2021, p. 100; d) J. J. Lewandowski, M. Seifi, Annu. Rev. Mater. Res. 2016, 46, 151; e) T. P. Gabb, J. Gayda, R. V. Miner, Metall. Trans. A 1986, 17, 497

[3] H. R. Wenk, P. V. Houtte, Rep. Prog. Phys. 2004, 67, 1367.

[4] I. L. Dillamore, W. T. Roberts, Metall. Rev. 1965, 10, 271.

[5] U. F. Kocks, C. N. Tomé, H. R. Wenk, Texture and Anisotropy: Preferred Orientations in Polycrystals and Their Effect on Materials Properties, Cambridge University Press, UK 2000.

[6] K. Hagihara, T. Nakano, JOM 2021, 74, 1760.

[7] a) B. Parvaresh, H. Aliyari, R. Miresmaeili, M. Dehghan, M. Mohammadi, Met. Mater. Int. 2023, 29, 3103; b) M. H. Sehhat, A. Mahdianikhotbesara, Granular Matter 2021, 23, 89; c) S.-G. Chen, H.-J. Gao, Q. Wu, Z.-H. Gao, X. Zhou, J. Mater. Res. Technol. 2022, 17, 2950; d) D. Herzog, V. Seyda,

E. Wycisk, C. Emmelmann, Acta Mater. 2016, 117, 371; e) A. Mostafaei, C. Zhao, Y. He, S. Reza Ghiaasiaan, B. Shi, S. Shao, N. Shamsaei, Z. Wu, N. Kouraytem, T. Sun, J. Pauza J. V. Gordon, B. Webler, N. D. Parab, M. Asherloo, Q. Guo, L. Chen, A. D. Rollett, Curr. Opin. Solid State Mater. Sci. 2022, 26 100974; f) W. Abd-Elaziem, S. Elkatatny, A.-E. Abd-Elaziem M. Khedr, M. A. Abd El-baky, M. A. Hassan, M. Abu-Okail, M. Mohammed, A. Järvenpää, T. Allam, A. Hamada, J. Mater. Res. Technol. 2022, 20, 681; g) X. Zhang, C. J. Yocom, B. Mao, Y. Liao, J. Laser Appl. 2019, 31, 031201; h) S. Chowdhury, N. Yadaiah C. Prakash, S. Ramakrishna, S. Dixit, L. R. Gupta, D. Buddhi, J. Mater. Res. Technol. 2022, 20, 2109; i) P. V. Cobbinah R. A. Nzeukou, O. T. Onawale, W. R. Matizamhuka, Metals 2020 11, 58 .

[8] T. DebRoy, H. L. Wei, J. S. Zuback, T. Mukherjee, J. W. Elmer, J. O. Milewski, A. M. Beese, A. Wilson-Heid, A. De, W. Zhang Prog. Mater. Sci. 2018, 92, 112.

[9] O. Andreau, I. Koutiri, P. Peyre, J.-D. Penot, N. Saintier, E. Pessard, T. De Terris, C. Dupuy, T. Baudin, J. Mater. Process. Technol. 2019, 264, 21 .

[10] S. Suwas, D. Kumar, Trans. Indian Natl. Acad. Eng. 2020, 5, 1.

[11] S. Suwas, R. J. Vikram, Trans. Indian Natl. Acad. Eng. 2021, 6, 991.

[12] T. Qi, H. Zhu, H. Zhang, J. Yin, L. Ke, X. Zeng, Mater. Des. 2017, 135, 257.

[13] M. J. Torkamany, F. Malek Ghaini, R. Poursalehi, A. F. H. Kaplan, Opt. Lasers Eng. 2016, 79, 9.

[14] J. Yang, J. Han, H. Yu, J. Yin, M. Gao, Z. Wang, X. Zeng, Mater. Des. 2016, 110, 558 .

[15] T. Todo, T. Ishimoto, O. Gokcekaya, J. Oh, T. Nakano, Scr. Mater. 2022, 206, 114252.

[16] K. Hagihara, T. Nakano, M. Suzuki, T. Ishimoto, . Suyalatu, S.-H. Sun, J. Alloys Compd. 2017, 696, 67

[17] M. Cloots, P. J. Uggowitzer, K. Wegener, Mater. Des. 2016, 89, 770

[18] T. Heeling, M. Cloots, K. Wegener, Addit. Manuf. 2017, 14, 116.

[19] A. T. Sidambe, Y. Tian, P. B. Prangnell, P. Fox, Int. J. Refract. Met. Hard Mater. 2019, 78, 254

[20] J. J. Marattukalam, D. Karlsson, V. Pacheco, P. Beran, U. Wiklund U. Jansson, B. Hjörvarsson, M. Sahlberg, Mater. Des. 2020, 193 108852.

[21] S. Patel, M. Vlasea, Materialia 2020, 9, 100591.

[22] T. Sibillano, A. Ancona, V. Berardi, E. Schingaro, G. Basile, P. M. Lugarà, J. Mater. Process. Technol. 2007, 191, 364.

[23] a) J. Metelkova, Y. Kinds, K. Kempen, C. de Formanoir, A. Witvrouw, B. Van Hooreweder, Addit. Manuf. 2018, 23, 161; b) E. Soylemez, Addit. Manuf. 2020, 31, 100984.

[24] W. E. King, H. D. Barth, V. M. Castillo, G. F. Gallegos, J. W. Gibbs, D. E. Hahn, C. Kamath, A. M. Rubenchik, J. Mater. Process. Technol. 2014, 214, 2915.

[25] a) G. Buvanashekaran, N. S. Shanmugam, K. Sankaranarayanasamy, R. Sabarikanth, Proc. Inst. Mech. Eng., Part C 2009, 223, 1141 b) P. Colegrove, P.-E. Simiand, A. Varughese, S. Williams D. Yapp, ASM Proc. of the 8th Int. Conf. 2008 2009, pp. 303-312.

[26] E. Assuncao, S. Williams, D. Yapp, Opt. Lasers Eng. 2012, 50, 823.

[27] S. A. Khairallah, A. T. Anderson, A. Rubenchik, W. E. King, Acta Mater. 2016, 108, 36

[28] H. Wang, Y. Zou, Int. J. Heat Mass Transfer 2019, 142, 118473.

[29] A. Chouhan, A. Aggarwal, A. Kumar, Int. J. Heat Mass Transfer 2021, 178,121602 .

[30] M. Asta, C. Beckermann, A. Karma, W. Kurz, R. Napolitano, M. Plapp, G. Purdy, M. Rappaz, R. Trivedi, Acta Mater. 2009, 57, 941.

[31] S. Kou, Welding Metallurgy, Willey, New Jersey 2003, p. 461.

[32] T. Debroy, S. A. David, Rev. Mod. Phys. 1995, 67, 85.

[33] R. F. Sekerka, J. Phys. Chem. Solids 1967, 28, 983.\\
[34] K. G. Prashanth, J. Eckert, J. Alloys Compd. 2017, 707, 27.

[35] W. A. Tiller, K. A. Jackson, J. W. Rutter, B. Chalmers, Acta Metall. 1953, $1,428$.

[36] J. Liu, A. C. To, Addit. Manuf. 2017, 16, 58.

[37] D. Bourell, J. P. Kruth, M. Leu, G. Levy, D. Rosen, A. M. Beese, A. Clare, CIRP Ann. 2017, 66, 659.

[38] A. Basak, S. Das, Annu. Rev. Mater. Res. 2016, 46, 125.

[39] L. Thijs, K. Kempen, J.-P. Kruth, J. Van Humbeeck, Acta Mater. 2013, 61, 1809

[40] P. Köhnen, M. Létang, M. Voshage, J. H. Schleifenbaum, C. Haase, Addit. Manuf. 2019, 30, 100914.

[41] R. Trivedi, W. Kurz, Int. Mater. Rev. 1994, 39, 49

[42] G. Sigworth, Mod. Cast. 2014, 104, 39

[43] M. Gäumann, R. Trivedi, W. Kurz, Mater. Sci. Eng., A 1997, 226, 763.

[44] W. Kurz, C. Bezençon, M. Gäumann, Sci. Technol. Adv. Mater. 2001, 2, 185.

[45] $\varnothing$. Grong, D. Grong, Metallurgical Modelling of Welding, Vol. 2, Institute of Materials, London, UK 1997

[46] M. Gäumann, C. Bezençon, P. Canalis, W. Kurz, Acta Mater. 2001, 49, 1051.

[47] O. Gokcekaya, T. Ishimoto, S. Hibino, J. Yasutomi, T. Narushima, T. Nakano, Acta Mater. 2021, 212, 116876.

[48] T. Kuech, Handbook of Crystal Growth: Thin Films and Epitaxy, North Holland [Imprint], Oxford 2014.

[49] T. Ishimoto, K. Hagihara, K. Hisamoto, S.-H. Sun, T. Nakano, Scr. Mater. 2017, 132, 34.

[50] S.-H. Sun, T. Ishimoto, K. Hagihara, Y. Tsutsumi, T. Hanawa, T. Nakano, Scr. Mater. 2019, 159, 89.

[51] O. Gokcekaya, N. Hayashi, T. Ishimoto, K. Ueda, T. Narushima, T. Nakano, Addit. Manuf. 2020, 36, 101624.

[52] X. Liu, C. Zhao, X. Zhou, Z. Shen, W. Liu, Mater. Des. 2019, 168, 107677.

[53] T. Niendorf, S. Leuders, A. Riemer, H. A. Richard, T. Tröster, D. Schwarze, Metall. Mater. Trans. B 2013, 44, 794

[54] M. Ghayoor, K. Lee, Y. He, C.-H. Chang, B. K. Paul, S. Pasebani, Addit. Manuf. 2020, 32, 101011.

[55] V. A. Popovich, E. V. Borisov, A. A. Popovich, V. S. Sufiarov, D. V. Masaylo, L. Alzina, Mater. Des. 2017, 114, 441.

[56] M. Garibaldi, I. Ashcroft, M. Simonelli, R. Hague, Acta Mater. 2016, $110,207$.

[57] Z. Sun, X. Tan, S. B. Tor, C. K. Chua, NPG Asia Mater. 2018, 10, 127.

[58] M. Higashi, T. Ozaki, Mater. Des. 2020, 191, 108588

[59] X. Wang, J. A. Muñiz-Lerma, M. Attarian Shandiz, O. Sanchez-Mata, M. Brochu, Mater. Sci. Eng., A 2019, 766, 138395.

[60] C. Sofras, J. Čapek, A. Arabi-Hashemi, C. Leinenbach, M. Frost, K. An, R. E. Logé, M. Strobl, E. Polatidis, Mater. Des. 2022, 219, 110789.

[61] K. A. Sofinowski, S. Raman, X. Wang, B. Gaskey, M. Seita, Addit. Manuf. 2021, 38, 101809 .

[62] Z. Dong, Y. Liu, W. Wen, J. Ge, J. Liang, Materials 2018, 12, 50

[63] N. Nadammal, S. Cabeza, T. Mishurova, T. Thiede, A. Kromm, C. Seyfert, L. Farahbod, C. Haberland, J. A. Schneider P. D. Portella, Mater. Des. 2017, 134, 139.

[64] S. D. Jadhav, S. Dadbakhsh, L. Goossens, J. P. Kruth, J. Van Humbeeck, K. Vanmeensel, J. Mater. Process. Technol. 2019, $270,47$.

[65] H. Shipley, D. McDonnell, M. Culleton, R. Coull, R. Lupoi, G. O'Donnell, D. Trimble, Int. J. Mach. Tools Manuf. 2018, 128, 1.

[66] S. Sanchez, P. Smith, Z. Xu, G. Gaspard, C. J. Hyde, W. W. Wits, I. A. Ashcroft, H. Chen, A. T. Clare, Int. J. Mach. Tools Manuf. 2021, 165, 103729 .

[67] T. Ishimoto, S. Wu, Y. Ito, S.-H. Sun, H. Amano, T. Nakano, ISIJ Int. 2020, 60, 1758 .

[68] H. L. Wei, J. Mazumder, T. DebRoy, Sci. Rep. 2015, 5, 16446.

[69] L. Thijs, M. L. Montero Sistiaga, R. Wauthle, Q. Xie, J.-P. Kruth, J. Van Humbeeck, Acta Mater. 2013, 61, 4657.

[70] H. Y. Wan, Z. J. Zhou, C. P. Li, G. F. Chen, G. P. Zhang, J. Mater. Sci. Technol. 2018, 34, 1799.

[71] Z. Sun, X. P. Tan, M. Descoins, D. Mangelinck, S. B. Tor, C. S. Lim, Scr. Mater. 2019, 168, 129.

[72] S.-H. Sun, K. Hagihara, T. Nakano, Mater. Des. 2018, 140, 307.

[73] J.-P. Kruth, J. Deckers, E. Yasa, R. Wauthlé, Proc. Inst. Mech. Eng., Part B 2012, 226, 980.

[74] B. Attard, S. Cruchley, C. Beetz, M. Megahed, Y. L. Chiu, M. M. Attallah, Addit. Manuf. 2020, 36, 101432.

[75] Y. Tsutsumi, T. Ishimoto, T. Oishi, T. Manaka, P. Chen, M. Ashida, K. Doi, H. Katayama, T. Hanawa, T. Nakano, Addit. Manuf. 2021, 45, 102066 .

[76] W. Shao, B. He, C. Qiu, Z. Li, Opt. Laser Technol. 2022, 156, 108609 .

[77] N. Sridharan, Y. Chen, P. Nandwana, R. M. Ulfig, D. J. Larson, S. S. Babu, J. Mater. Sci. 2020, 55, 1715.

[78] J. Chen, D. Fabijanic, T. Zhang, E. W. Lui, M. Brandt, W. Xu, Addit. Manuf. 2022, 58, 103041.

[79] H. Beladi, Q. Chao, G. S. Rohrer, Acta Mater. 2014, 80, 478

[80] S. S. Babu, S. M. Kelly, E. D. Specht, T. A. Palmer, J. W. Elmer, In Int. Conf. on Solid-Solid Phase Transformations in Inorganic Materials 2005, Phoenix, AZ 2005.

[81] S. C. Wang, M. Aindow, M. J. Starink, Acta Mater. 2003, 51, 2485

[82] W. B. Hutchinson, L. Ryde, P. S. Bate, In Materials Science Forum, Vol. 495, Trans Tech Publications Ltd, Switzerland 2005, pp. 1141-1150.

[83] S. C. Vogel, D. Bhattacharyya, G. B. Viswanathan, D. J. Williams, H. L. Fraser, In Materials Science Forum, Vol. 495, Trans Tech Publications Ltd, Switzerland 2005, pp. 681-686.

[84] D. Bhattacharyya, G. B. Viswanathan, R. Denkenberger, D. Furrer, H. L. Fraser, Acta Mater. 2003, 51, 4679.

[85] a) M. N. Doğu, Z. Esen, K. Davut, E. Tan, B. Gümüş, A. F. Dericioglu, Mater. Charact. 2020, 168, 110549; b) M. N. Doğu, K. Davut, M. A. Obeidi, M. A. Yalçın, H. Gu, T. S. E. Low, J. Ginn, D. Brabazon, J. Mater. Res. Technol. 2022, 19, 4242.

[86] L. S. Aota, P. Bajaj, K. D. Zilnyk, E. A. Jägle, D. Ponge, H. R. Z. Sandim, D. Raabe, Materialia 2021, 20, 101236.

[87] E. M. Savitskii, Physical Metallurgy of Refractory Metals and Alloys, Springer Science \& Business Media, New York, NY 2012.

[88] A. Iveković, N. Omidvari, B. Vrancken, K. Lietaert, L. Thijs, K. Vanmeensel, J. Vleugels, J.-P. Kruth, Int. J. Refract. Met. Hard Mater. 2018, 72, 27.\\
[89] F. Larson, A. Zarkades, Properties of Textured Titanium Alloys, Metals and Ceramics Information Center, Columbus, OH 1974, p. 0085.

[90] a) H. W. Hayden, W. G. Moffatt, J. Wulff, The Structure and Properties of Materials. V. 3. Mechanical Behavior, John Wiley and Sons, New York, NY 1965, p. 248; b) M. Ashby, H. Shercliff, D. Cebon, Materials: Engineering, Science, Processing and Design/Michael Ashby, Hugh Shercliff and David Cebon, Butterworth-Heinemann, Amsterdam, Boston 2007.

[91] R. Komanduri, N. Chandrasekaran, L. M. Raff, Wear 2000, 242, 60

[92] R. Badji, T. Chauveau, B. Bacroix, Mater. Sci. Eng., A 2013, 575, 94

[93] Y.-L. Liu, H.-B. Zhou, Y. Zhang, S. Jin, G.-H. Lu, Nucl. Instrum Methods Phys. Res., Sect. B 2009, 267, 3282.

[94] S. Tekumalla, J. E. Chew, S. W. Tan, M. Krishnan, M. Seita, Addit. Manuf. 2022, 59, 103111

[95] S. Pilz, T. Gustmann, F. Günther, M. Zimmermann, U. Kühn, A. Gebert, Mater. Des. 2022, 216, 110516.

[96] G. Sander, S. Thomas, V. Cruz, M. Jurg, N. Birbilis, X. Gao, M. Brameld, C. R. Hutchinson, J. Electrochem. Soc. 2017, 164, C250.

[97] I. Yadroitsev, Selective Laser Melting: Direct Manufacturing of 3DObjects By Selective Laser Melting of Metal Powders, LAP Lambert Academic Publishing, Saarbüken 2009

[98] O. Rehme, Cellular Design for Laser Freeform Fabrication, Vol. 4, Cuvillier Verlag, Germany 2010.

[99] a) X. J. Fan, R. T. Qu, Z. F. Zhang, J. Mater. Sci. Technol. 2022, 123, 70; b) D. Li, Y. Zhang, Intermetallics 2016, 70, 24; c) H. Zhang, Y. He, Y. Pan, Scr. Mater. 2013, 69, 342.

[100] B. Qian, K. Saeidi, L. Kvetková, F. Lofaj, C. Xiao, Z. Shen, Dent Mater. 2015, 31, 1435.

[101] D. Wang, C. Song, Y. Yang, Y. Bai, Mater. Des. 2016, 100, 291.

[102] Y. Watanabe, H. Sugano, T. Chiba, H. Sato, N. Sato, S. Nakano, J. Jpn. Inst. Light Met. 2022, 72, 198

[103] D. Gu, H. Chen, Mater. Sci. Eng., A 2018, 725, 419.

[104] T. Nagase, T. Hori, M. Todai, S.-H. Sun, T. Nakano, Mater. Des. 2019, 173, 107771.

[105] T. D. McLouth, D. B. Witkin, G. E. Bean, S. D. Sitzman, P. M. Adams, J. R. Lohser, J.-M. Yang, R. J. Zaldivar, Mater. Sci. Eng., A 2020, 780, 139184.

[106] F. Geiger, K. Kunze, T. Etter, Mater. Sci. Eng., A 2016, 661, 240

[107] T. Etter, K. Kunze, F. Geiger, H. Meidani, IOP Conf. Ser.: Mater. Sci. Eng. 2015, 82, 12097

[108] A. Charmi, R. Falkenberg, L. Ávila, G. Mohr, K. Sommer, A. Ulbricht, M. Sprengel, R. Saliwan Neumann, B. Skrotzki, A. Evans, Mater. Sci. Eng., A 2021, 799, 140154.

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-22}
\end{center}

Sae Matsunaga is an assistant professor affiliated with the Mitarai laboratory, Advanced Materials Science Department, Graduate School of Frontier Sciences of The University of Tokyo, Japan. Her research interest includes the study of alloy design and deformation mechanisms of nickel-based superalloys and $\mathrm{Nb}-\mathrm{Si}$ alloys.

\begin{center}
\includegraphics[max width=\textwidth]{2024_07_13_01fad27876b5c2169f6dg-22(1)}
\end{center}

Yoko Yamabe-Mitarai is a professor and current head of the Advanced Materials Science Department at the Graduate School of Frontier Sciences of The University of Tokyo, Japan. She is well known for her valuable research and contributions to alloy design and development, processing, microstructure, and properties of various materials such as PGM-based superalloys, shape memory alloys, high entropy superalloys, and Ti alloys.


\end{document}