This study presents a comprehensive machine-learning (ML)-based surrogate modeling framework to predict multiple, coupled thermo-mechanical responses of laser powder bed fusion additively manufactured (L-PBF-AM) NiTi shape memory alloys (SMAs). Four supervised ML algorithms, Linear Regression (LR), Random Forest (RF), Artificial Neural Network (ANN), and Support Vector Machine (SVM), were systematically trained and benchmarked using curated experimental data. The input features included feedstock chemical composition, L-PBF processing parameters, and loading mode/test temperature, while the target outputs encompassed relative density, phase transformation temperatures, ultimate tensile/compressive strength, elongation strain, total superelastic/shape memory strain, and recovery ratio. Comparative evaluation revealed that RF achieved the best overall predictive accuracy, delivering high coefficients of determination (R²) and low mean absolute errors (MAE) across most property categories. However, ANN outperformed RF in predicting transformation temperatures, highlighting its enhanced sensitivity to nonlinear phase-transformation phenomena and suggesting the complementary potential of hybrid ML schemes. Model validation using independent datasets excluded from training confirmed strong generalization performance, with RF exhibiting less than 15

Melt pool geometry governs porosity formation, inter-layer bonding, microstructural evolution, and residual stress development in Laser Powder Bed Fusion (L-PBF), making its accurate prediction central to process qualification and process-window development. However, experimental qualification is resource-intensive, high-fidelity multi-physics simulations remain computationally expensive for large-scale parameter exploration, and reduced-order analytical models often lose accuracy near melting-regime transitions. Existing machine learning (ML) approaches provide deterministic predictions but do not explicitly address uncertainty or systematic prediction bias, limiting their reliability across different operating conditions and experimental datasets. This study presents an uncertainty-aware hybrid ML framework for melt pool geometry prediction and process qualification in L-PBF of Inconel 718 (IN718). Single-track and multi-track specimens were fabricated over a broad range of laser powers, scan speeds, hatch spacings, and layer counts and characterized using optical and scanning electron microscopy. The proposed framework combines k-fold ensemble regression with physics-informed directional classification, where the ensemble standard deviation serves as a physically interpretable uncertainty signal and is subsequently converted into a targeted bias-correction mechanism. The melt pool depth-to-width ratio (D/W) was adopted as a unifying descriptor linking melt pool morphology, thermal-gradient-driven grain growth direction, melting-regime transitions, and process-map development. The framework was trained using 68 experimentally measured melt pools and evaluated using an external dataset of 54 measurements compiled from published L-PBF IN718 studies, including validation and independent test datasets. Compared with established analytical models and polynomial curve-fitting baselines, the proposed approach improved predictive accuracy, enhanced cross-study generalization, and provided useful ensemble-based uncertainty estimates. Multi-track experiments further validated the maximum hatch spacing criterion for predicting lack-of-fusion onset from single-track measurements, achieving prediction errors below 6% and enabling the identification of defect-free processing windows. The results demonstrate that ensemble prediction uncertainty contains information related to regime-dependent model sensitivity and can be exploited to improve prediction reliability near conduction-transition-keyhole boundaries. Beyond the specific IN718 case, the proposed framework provides a potentially transferable strategy for uncertainty-aware process qualification and rapid process-window development and offers a computationally efficient foundation for digital twin development, in-situ monitoring, and future closed-loop control of metal additive manufacturing (AM).

Additive manufacturing (AM) is redefining the design space of NiTi shape memory alloys (SMAs), enabling architected geometries and spatially tailored functionality not accessible through conventional processing. However, the extreme thermal gradients and rapid solidification inherent to AM fundamentally reshape transformation thermodynamics, microstructure, and functional reliability. This review synthesizes current understanding of process-structure-property relationships across laser powder bed fusion, electron-beam powder bed fusion, directed energy deposition, and solid-state AM routes, framing them within a unified thermal-compositional landscape.NiTi performance in AM cannot be evaluated solely through densification metrics. Instead, subtle variations in local energy distribution, evaporation-driven nickel redistribution, oxygen uptake, and crystallographic texture govern transformation temperatures, anisotropy, and superelastic stability. A curated database compiled from over 300 studies is used to perform cross-study statistical mapping, revealing that energy-density metrics are non-unique descriptors, where identical nominal values can produce divergent phase states depending on melt-pool mode and time-temperature history. Texture control via build orientation and scan strategy has emerged as a powerful lever for engineering anisotropic functional response, while post-build heat treatments enable secondary-phase tuning but remain highly sensitive to as-built chemistry.Looking forward, the field is transitioning from empirical parameter optimization toward physics-informed and data-driven design frameworks. Standardized reporting of composition shifts, thermal history, and transformation metrics is needed to enable cross-platform comparability. Integration of in situ monitoring, predictive modeling, and curated databases offers a pathway toward closed-loop control of transformation behavior. Establishing thermodynamically grounded design principles will be essential for translating AM NiTi into reliable biomedical, aerospace, and adaptive structural applications. This framework provides a basis for rational process design and standardized reporting in AM NiTi systems.

This study presents a comparative experimental investigation of the elastocaloric (EC) behavior of a quaternary NiTiHfPd shape memory alloy (SMA) and a conventional binary NiTi under uniaxial compressive loading. Nearadiabatic temperature changes (Delta T) arising from stress-induced martensitic transformations were measured using high-speed, non-contact infrared thermography across a systematic range of applied strain levels (up to 7 %) and strain rates (0.0007-0.30 s- 1). Binary NiTi exhibited a more uniform temperature distribution and achieved a slightly higher average Delta T, while NiTiHfPd demonstrated superior peak cooling performance, attaining a maximum Delta T of -16.80 degrees C at 7 % strain and a strain rate of 0.30 s-1, with lower energy dissipation (Delta W). The influence of applied strain and strain rate on Delta T, hysteresis loss (Delta W), and the coefficient of performance (COP) was systematically evaluated and quantitatively compared between the two alloys. Results indicate that NiTiHfPd's reduced hysteresis and strong peak cooling performance offer distinct advantages for applications requiring high stress tolerance and rapid, localized cooling.

This paper explores the additive manufacturing of nickel-titanium (Ni-Ti) shape memory alloys (SMAs) using binder jetting and subsequent solid-state sintering under an Ar atmosphere. The consolidated 3D-printed Ni-Ti parts exhibit a correlation between pore morphology, pore fraction, and phase formation at different solidstate sintering temperatures. At a sintering temperature of 1175 degrees C, NiTi grains with TiC precipitates along grain boundaries are observed, accompanied by irregular, interconnected pores constituting similar to 15 % of the volume. Increasing the sintering temperature to 1185 degrees C leads to the formation of a minor phase of Ni4Ti3 within NiTi grains, along with grain boundary TiC precipitates, and isolated pores with a volume fraction of similar to 5 %. The higher sintering temperature corresponds to a higher average nanohardness value (8.1 GPa or 763 Hv compared to 6.9 GPa or 654 Hv), indicating the presence of a greater abundance of secondary phases and higher densification at the elevated sintering temperature. While the transformation temperatures (TTs) were undetectable in the bulk-printed samples, a different structure featuring designed channels and thin struts, sintered under similar conditions, exhibited detectable TTs, with a martensite start temperature of 34 degrees C. Biocompatibility tests demonstrated cell spreading and attachment on both sintered Ni-Ti samples. These findings offer valuable insights into the potential use of binder jetted Ni-Ti for medical implants and tissue engineering applications.

Thin features are integral components of most lightweight cellular lattice structures; however, limited studies are carried out to understand their property/quality aspects. This experimental study investigates the influence of powder feedstock size, feature geometry, and process parameters on the property/quality of thin lightweight features fabricated using a laser powder bed fusion additive manufacturing (L-PBF-AM) system. Three different Ti6Al4V powder feedstocks (fine, medium, and coarse) were utilized, and the dimensions of the features were varied in the range of 0.1 to 0.5 mm. The experimental data were analyzed to gain insights into the powder feedstock-geometry-process-property/quality (PGPP/PGPQ) characteristics, which had not been previously explored but are crucial for designing lightweight structures using L-PBF-AM. The results indicated that both powder feedstock size and feature dimension significantly influenced the properties and quality of the fabricated thin features. Additionally, feature type, volumetric energy density, and their interactions exhibited varying effects on geometrical accuracy, porosity, grain size, and flexural properties. Struts showed lower success rates, grain sizes, and dimensional errors but higher mechanical properties compared to walls. However, both features exhibited similar porosity characteristics. Regarding powder feedstock size, smaller powder sizes were found to be advantageous for fabricating lower-dimensional features and improving their mechanical properties. The feature geometry type also significantly influences the final material properties. A notable observation was that the 0.1 mm wall features exhibited the lowest mechanical properties, particularly in terms of yield strength, while the 0.5 mm strut features exhibited lower mechanical properties among the struts. The findings of the study underscored the importance of understanding the compound relationships between powder feedstock, feature geometry, process parameters, and the resulting properties/quality for lightweight features in L-PBF-AM. Further research is necessary to establish the knowledge and understanding of L-PBF-AM thin features and elucidate the PGPP/PGPQ characteristics in greater detail.

This study presents the shape memory behavior of Ni-rich NiTi shape memory alloy fabricated by Laser Powder Bed Fusion Additive Manufacturing (L-PBF-AM) before and after post-processing heat treatment. The microstructural features and thermo-mechanical responses were systematically investigated to understand the effects of processing on the behavior of the specimens. It was shown that the L-PBF-AM process improves the functionality of NiTi components by illustrating perfect superelastic behavior at higher-temperature windows compared to the casted ingot. In addition, it was revealed that shape memory responses were tailored by altering hatch distance, which significantly controls the texture formation along the building direction. After post-processing treatments, transformation temperatures were increased, hysteresis was decreased, and the strength of the samples was significantly improved. The aged L-PBF-AM sample with a smaller hatch distance (80 µm) and intense [001] texture illustrated perfect superelastic behavior with a recoverable strain of 7

In recent years shape memory alloys (SMAs) have gained significant attention as potential damping device materials. This article presents an extensive review of the damping characteristics of SMAs, as well as experimental methods used to characterize their damping properties. The shape memory response and associated damping quality are discussed for three popular families of SMAs; Fe-based, Cu-based, and NiTi-based alloys including their behaviors and limitations. This review article also summarizes the most important parameters that impact the damping behavior of SMAs which are necessary to be investigated by researchers and manufacturers to address the current design challenges.

Shape memory alloys are a unique class of materials that are capable of large reversible deformations under external stimuli such as stress or temperature. The present study examines the phase transformations and mechanical responses of NiTi and NiTiHf shape memory alloys under the loading of a spherical indenter by using a finite element model. It is found that the indentation unloading curves exhibit distinct changes in slopes due to the reversible phase transformations in the SMAs. The normalized contact stiffness (F/S2) of the SMAs varies with the indentation load (depth) as opposed to being constant for conventional single-phase materials. The load-induced phase transformation that occurred under the spherical indenter was simulated numerically. It is observed that the phase transformation phenomenon in the SMA induced by an indentation load is distinctly different from that induced by a uniaxial load. A pointed indenter produces a localized deformation, resulting in a stress (load) gradient in the specimen. As a result, the transformation of phases in SMAs induced by an indenter can only be partially completed. The overall modulus of the SMAs varies continuously with the indentation load (depth) as the average volumetric fraction of the martensite phase varies. For NiTi (Ea > Em), the modulus decreases with the depth, while for NiTiHf (Ea < Em), the modulus increases with the depth. The predicted young modules during indentation modeling agree well with experimental results. Finally, the phase transformation of the SMAs under the indenter is not affected by the post-yield behavior of the materials.

This study systematically evaluates the effects of laser powder bed fusion additive manufacturing (L-PBF-AM) parameters (hatch spacing and laser power) on the thermomechanical behavior and microstructure of Ni50.8Ti49.2 shape memory alloy. The samples were fabricated with hatch spacings from 40 to 240 mu m and laser powers of 50 and 100 W at a constant scanning speed of 125 mm/s, resulting in parts with volumetric energy density levels from 55 to 666 J/mm(3) and two sets of linear energy densities of 0.4 and 0.8 J/mm. The results showed a reduced melt pool size and discontinuity of scan tracks with decreased laser power. Additionally, the porosity level was increased with larger hatch spacing and lower laser power. More notably, the transformation temperatures increased, and the critical stress, recoverable strain, and functional stability of samples improved with lower hatch spacing, where the recovery ratio of up to 90% was observed, regardless of the employed laser power. This study also discussed the relationship between the fabrication process and texture formation in the L-PBF-AM process. The advantage of L-PBF-AM was revealed in tailoring the microstructure from highly textured samples in [1 1 1] or [001] direction when hatch spacing lower than laser beam focused was employed, to the appearance of equiaxed solidification front with island grains and random orientations.

This study is the first study on the compressive and tensile stress-strain revealing the deformation anisotropy among laser powder bed fusion NiTi parts fabricated with the same process conditions. We investigated the effects of building orientation on the microstructure and the resulting shape memory properties. To this end, three orientations were selected, namely 0, 45, and 90-degree, measured from the build plate and fabricated with the same process parameters. A strong (001) texture was formed along the building direction for all of the samples; a different texture could however be observed along the loading direction (LD). Samples fabricated with 45-degree showed a texture of (110) along the LD, as confirmed through X-ray diffraction and backscattered diffraction while 0 and 90 samples still had the (001) texture along with the LD. These texture variations created anisotropic compression-tension behaviors with deformation patterns consistent with single crystals. The (001) -textured parts showed higher strength and lower transformation strain (2.87% @ 200 MPa in tension for 0 degrees) while the (110) samples showed higher transformation strain at lower stresses (5.31% @ 150 MPa in tension for 45 degrees). (c) 2021 Elsevier B.V. All rights reserved.

This study investigates the high-temperature shape memory behavior of NiTiHf alloys fabricated via selective laser melting process. Specifically, the effects of laser power (100 W and 250 W) on their transformation temperatures, strain, and microstructure were investigated and compared to the ingot. The transformation temperatures of SLM fabricated alloys increased from 150 degrees C to 350 degrees C with elevated laser power due to Ni evaporation. The sample fabricated with 100 W showed sharp transformation peaks, good shape memory behavior with recoverable strain of 1.67% and superelasticity. The sample fabricated with 250 W had broad transformation peaks with low recoverable strain of 0.7% during thermal cycling. (C) 2019 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

The transformation temperatures, magnetization behavior, shape memory behavior, and mechanical properties of polycrystalline Ni 45 Mn 40 Co 5 Sb 10−x B x (at.%) ( x = 0, 1, 2, 3, 4, 5) alloys were systematically investigated. It was revealed that substituting Sb with B drastically increases the transformation temperatures, while it decreases the saturation magnetization due to the alteration of electron concentration of the matrix and formation of Co-rich second phases. With the substitution of Sb with 5% B, martensite start temperature and activation energy were increased from 50 to 316.8 °C, and 185 to 722.6 kJ mol −1 , respectively. The thermal cycling under stress, superelasticity, and failure experiments showed that shape memory properties and strength were improved by the substitution of Sb with B. The shape memory effect with maximum recoverable strain of 1.6% was observed with in Ni 45 Mn 40 Co 5 Sb 9 B 1 , and perfect superelasticity was exhibited at 220 °C in Ni 45 Mn 40 Co 5 Sb 8 B 2 . It was concluded that NiMnCoSb alloys can be used as high-temperature magnetic shape memory alloys as they exhibit transformation temperatures above 100 °C and show promising shape memory and superelasticity behavior, and there is a magnetization difference between their transforming phases.

In this work, the effects of process parameters on the fabrication of NiTiHf alloys using selective laser melting are studied. Specimens were printed using bidirectional scanning pattern and with various sets of process parameters of laser power (100–250 W), hatch spacing (60–140 µm), and scanning speed (200–1000 mm/s). Cracking and delamination formation, dimensional accuracy, density, and transformation temperatures were examined. Despite the brittle nature of the alloy, fully dense parts have been produced. Laser scanning speed and volumetric energy density were found to be the most influential process parameters on fabricating defect-free samples. It was shown that transformation temperatures are highly dependent on the process parameters. By proper choice of parameters, it is possible to tailor the austenite finish temperature from 100 to 400 °C. The most influential factors on transformation behavior were found to be the laser power and energy density. It is worth noting that these two parameters at higher levels resulted in high process temperatures and therefore a larger level of Ni evaporation. Among the four parameters that constitute the energy density, the hatch spacing does not significantly affect the transformation temperatures. These findings serve as the foundation of developing HTSMA devices with desired geometrical and functional properties.

The transformation temperatures, magnetization behavior, shape memory behavior, and mechanical properties of polycrystalline Ni45Mn40Co5Sb10−xBx (at.%) (x = 0, 1, 2, 3, 4, 5) alloys were systematically investigated. It was revealed that substituting Sb with B drastically increases the transformation temperatures, while it decreases the saturation magnetization due to the alteration of electron concentration of the matrix and formation of Co-rich second phases. With the substitution of Sb with 5% B, martensite start temperature and activation energy were increased from 50 to 316.8 °C, and 185 to 722.6 kJ mol−1, respectively. The thermal cycling under stress, superelasticity, and failure experiments showed that shape memory properties and strength were improved by the substitution of Sb with B. The shape memory effect with maximum recoverable strain of 1.6% was observed with in Ni45Mn40Co5Sb9B1, and perfect superelasticity was exhibited at 220 °C in Ni45Mn40Co5Sb8B2. It was concluded that NiMnCoSb alloys can be used as high-temperature magnetic shape memory alloys as they exhibit transformation temperatures above 100 °C and show promising shape memory and superelasticity behavior, and there is a magnetization difference between their transforming phases.

Porous NiTi scaffolds display unique bone-like properties including low stiffness and superelastic behavior which makes them promising for biomedical applications. The present article focuses on the techniques to enhance superelasticity of porous NiTi structures. Selective Laser Melting (SLM) method was employed to fabricate the dense and porous (32–58%) NiTi parts. The fabricated samples were subsequently heat-treated (solution annealing + aging at 350 °C for 15 min) and their thermo-mechanical properties were determined as functions of temperature and stress. Additionally, the mechanical behaviors of the samples were simulated and compared to the experimental results. It is shown that SLM NiTi with up to 58% porosity can display shape memory effect with full recovery under 100 MPa nominal stress. Dense SLM NiTi could show almost perfect superelasticity with strain recovery of 5.65 after 6% deformation at body temperatures. The strain recoveries were 3.5, 3.6, and 2.7% for samples with porosity levels of 32%, 45%, and 58%, respectively. Furthermore, it was shown that Young’s modulus (i.e., stiffness) of NiTi parts can be tuned by adjusting the porosity levels to match the properties of the bones.

This study evaluates the anisotropic tensile properties of Ni-50.Ti-1(49.9) (in at%) components fabricated using an additive manufacturing (AM) process of selective laser melting (SLM). Dog-bone shaped tensile specimens were fabricated in three orthogonal building orientations (i.e., horizontal, edge, and vertical) with two different scanning strategies (i.e., alternating x/y and alternating in +/- 45(degrees) to the x-axis). Next, the samples were subjected to tensile testing until failure, shape memory effect tests and thermal cycling under constant tensile stresses up to 500 MPa. Their failure surfaces were analyzed for possible microstructural defects. It was revealed that the build orientation and scanning strategy affect the texture/microstructure, and hence the failure stress, ductility, shape memory effect, and functional stability. Samples fabricated in the horizontal orientation with alternating x/y scanning strategy had the highest ultimate tensile strength (606 MPa) and elongation (6.8%) with the strain recovery of 3.54% after 4 shape memory effect cycles. At stress levels less than or equal to 200 MPa, these samples had the actuation strain greater than 3.8% without accumulation of noticeable residual strain. It was observed that the scanning strategy of alternating in +/- 45 degrees result in degraded mechanical and shape memory response, particularly in horizontal and edge samples.

Microstructure of NiTiHf shape memory alloys can be engineered to have high strength and operate at high stress levels for a large temperature window. Nanoprecipitation is well-known method to improve the strength of materials but it can be employed to NiTiHf alloys to substantially alter their phase transformation characteristics (martensite morphology, transformation strain, hysteresis and stress). The martensitic transformation of Ni-rich Ni51.2Ti28.8Hf20 was severely suppressed in the solution treated condition (900 degrees C-3h/water quench) and after aging at low temperatures, while the transformation temperatures were greater than 100 degrees C after 650 degrees C-3h aging. Generation of nanosize precipitates (similar to 20 nm in size) after 3 h aging at 450 degrees C and 550 degrees C improved the strength of the material, resulting in a near perfect dimensional stability during isobaric thermal cycling at stress levels of greater than 1500 MPa, with work output of 20-30 J cm(-3). Superelastic behavior with 4% recoverable strain was demonstrated at low temperatures (-20 to 40 degrees C) after aging at 450 degrees C-3h and at elevated temperatures (120-160 degrees C) after aging at 550 degrees C-3h, with stresses reaching 2 GPa without the onset of plastic deformation. A clear relationship between thermal treatments, microstructure, mechanical and shape memory properties will be shown. (C) 2017 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.