Additive Manufacturing Letters 10 (2024) 100216

Contents lists available at ScienceDirect

Additive Manufacturing Letters

journal homepage: www.elsevier.com/locate/addlet

Short Communication

Additive manufacturing of NiTi architected metamaterials

C.A. Biffi a, C. Soyarslan b, J. Fiocchi a, C. Bregoli a, A. du Plessis c, A. Tuissi a, M. Mehrpouya b, *
a
CNR ICMATE, National Research Council, Institute of Condensed Matter Chemistry and Technologies for Energy, Unit of Lecco, CNR ICMATE; Via Previati 1/e, 23900
Lecco, Italy
b
Faculty of Engineering Technology, University of Twente, P.O. Box 217, 7500 AE Enschede, the Netherlands
c
Comet Technologies Canada Inc, 460 Sainte-Catherine W, Montreal (Québec) H3B 1A7, Canada

A R T I C L E I N F O A B S T R A C T

Keywords: Additive manufacturing has revolutionized the creation of complex and intrinsic structures, offering tailored
Shape memory alloys designs for enhanced product performance across various applications. Architected cellular or lattice structures
Functional performance exemplify this innovation, customizable for specific mechanical or functional requirements, boasting advantages
Architected metamaterials
such as reduced mass, heightened load-bearing capabilities, and superior energy absorption. Nonetheless, their
Additive manufacturing
Laser Powder Bed Fusion
single-use limitation arises from plastic deformation resulting from localized yield damage or plastic buckling.
Incorporating NiTi shape memory alloys (SMAs) presents a solution, enabling structures to recover their original
shape post-unloading. In this study, an NiTi architected metastructure, featuring auxetic behavior and a negative
Poisson’s ratio, was designed and fabricated via laser powder bed fusion (LPBF). The samples exhibit promising
superelastic performance with recoverable deformation strains at room temperature. Comprehensive charac­
terization processes evaluated the functional performance of the fabricated metastructures. The metastructure
geometry promoted microstructure formation primarily along the wall thickness. Cycling compression tests,
conducted at three applied force levels, demonstrated stable cyclic behavior with up to 3.8 % reversible
deformation strain, devoid of plastic buckling or yielding damage. Furthermore, the NiTi metastructures dis­
played robust energy absorption capacity and damping behavior, underscoring their potential for reusable en­
ergy dissipators in various industries including aerospace, automotive, construction, and etc.

1. Introduction fine-tune tailored functional performance [6–8]. These structures

possess a remarkable ability to dissipate or absorb kinetic energy from
Shape memory alloys (SMAs) belong to the category of smart ma­ external mechanical loads in a controlled manner [9–13]. Nonetheless, a
terials with a unique capability to recover their original shape after challenge encountered with architected metallic structures is the
deformation [1]. This thermally reversible shape recovery is a result of occurrence of unrecoverable deformation caused by inelastic buckling.
the phase transformation from low temperature (martensite) to high However, the integration of SMAs provides an attractive alternative as it
temperature (austenite), which is known as the shape memory effect allows for recoverable deformation and exhibits energy dissipation
(SME). Additionally, the superelasticity (SE) phenomenon is observed in characteristics through the SE effect. This, in turn, opens up exciting
the austenite phase (high-temperature) when mechanically stressed al­ possibilities for designing and fabricating a wide range of energy ab­
loys have the ability to restore their original shape [2,3]. The SE prop­ sorption systems and structures with recoverable performance [14].
erties hold promising potential for applications in energy and vibration There are only a limited number of studies that explore the func­
damping, as they occur during phase transformations in tional performance of SMA architecture structures. The difficulty lies in
loading-unloading cycles [4,5]. printing SMAs to achieve the desired functionality, as even slight
Additive manufacturing (AM) has brought about a revolution in the changes in alloy composition or process parameters can significantly
fabrication of complex-shaped structures, offering a unique solution for impact the transformation temperature and, subsequently, the thermo-
designers. Numerous studies have demonstrated the benefits of the laser mechanical performance of the printed structures [15,16]. In one
powder bed fusion (LPBF) method in producing architected cellular study, Khademzadeh tackled this challenge by utilizing a small spot laser
structures with precise and diverse geometries, enabling designers to with low energy density [17]. This approach aimed to enhance the

* Corresponding author.
E-mail address: [email protected] (M. Mehrpouya).

https://doi.org/10.1016/j.addlet.2024.100216
Received 11 February 2024; Received in revised form 20 April 2024; Accepted 22 April 2024
Available online 23 April 2024
2772-3690/© 2024 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
C.A. Biffi et al. Additive Manufacturing Letters 10 (2024) 100216

precision of the LPBF process when fabricating NiTi lattice structures,

reducing the deviation of the printed lattices from the designed model.
In another study, Gustmann et al. focused on investigating the impact of
scanning strategy on the quality and mechanical properties of NiTi lat­
tice structures [18]. The research demonstrated that the functional
performance of these lattices can be precisely tailored by adapting the
scanning vectors in the local area. Biffi et al. similarly highlighted the
impact of scanning strategy on the print quality and functional perfor­
mance of the lattices [19]. In their research, NiTi lattices showed
partially superelastic properties during the compression test at room
temperature, and the residual deformation was effectively recovered
after the heating process above the temperature of 80◦ C.
Such smart structures hold promising potential for biomedical ap­
plications, such as bone implants, due to their elastic modulus and
hysteretic behavior during stress cycling, which are similar to natural
bone [20]. Following this idea, Adnani et al. investigated the shape Fig. 1. The geometry of the CAD model for the 3D-printed structure. All
memory performance and mechanical response of NiTi lattice structures specified dimensions are in millimeters. While Lx and Ly both measure 15 mm,
fabricated by the LPBF technique [21]. They considered five levels of accounting for the introduced undulation, the wall thickness results in the
lattice porosity ranging from 32 % to 69 %. The study demonstrated that structure’s outer dimensions being 17 × 17 mm.
it is possible to decrease the elastic modulus of the lattices up to 86 % by
enhancing the porosity level while maintaining the shape memory
Table 1
performance of the NiTi lattices. In another study, Dadbakhsh et al. also
List of the adopted process parameters used for the printing
showed that the shape and geometry of the unit cells in NiTi lattice of the NiTi samples.
structures directly impact the shape memory response, and these factors
Parameters Value
can be controlled and manipulated during the design process [22].
Chen et al. employed a topology optimization strategy to design and Power 150 W
Exposure time 75 µs
fabricate gradient NiTi lattice structures [23,24]. Their findings
Scanning strategy Meander
demonstrated the suitability of these structures for energy absorption Atmosphere Argon
applications while retaining their shape memory performance. The lat­ Layer thickness 30 µm
tice structure offers the advantage of a larger displacement recovery Hatch distance 50 µm
interval and a smaller accumulative irrecoverable strain. Yan et al. also Point distance 120 µm
Laser spot size 65 µm
demonstrated the recoverable strain and energy dissipation capabilities
Platform temperature 30 ◦ C
of NiTi lattice structures [25]. They reported that the sample density and Platform material NiTi
cell architecture directly influence the superelasticity and effective Tilting angle 67◦
transformation stress of the lattices. Notably, their research revealed Oxygen level < 20 ppm
stable cyclic behavior with 2.8 % of reversible strain achieved through
cyclic compression tests conducted below the plastic deformation
lamina. This behavior mirrors the sinusoidal patterns outlined in Dolla
threshold.
model [33]. Zhang et al. further explored deformation-dependent
This study is dedicated to the investigation of the functional per­
Poisson’s ratios of slotted assemblies of these lattices, modeling the
formance of NiTi architected metastructures fabricated using the LPBF
waviness characterized as a sequence of circular arcs with consistent
process. These metastructures are inspired by wavy tetra-achiral lattices
radii and alternating orientations [34]. In scenarios devoid of undu­
as introduced by [26]. These lattice designs, along with their tetra-chiral
lations—where the arc radius approaches infinity—the lattice converges
counterparts, exhibit auxetic behavior [27,28]. The distinctive charac­
to a conventional square lattice design. Contrasting traditional methods
teristics of these structures make them particularly well-suited for
like slotting and stamping, 3D printing technology is employed for
structural applications, especially in energy absorption domains
fabricating these structures, emulating the absorption efficiency [29].
[29–31]. Employing a comprehensive range of microstructural and
This necessitates the creation of precise CAD models to accurately direct
mechanical characterization techniques, this research aims to compre­
material deposition in the additive manufacturing phase.
hensively explore the thermo-mechanical performance of these NiTi
Fig. 1 illustrates the design dimensions. Importantly, this model
metastructures
evolves from an undulated square lattice, preserving the foundational
lattice dimensions at lx = ly = 3 mm as the lattice spacing. Following the
2. Experimental and numerical methods methodology in [29], we adopted a 5 × 5 2D layout, resulting in an
edge-centered structural dimension of Lx = Ly = 15 mm. Considering
2.1. Metastructure design the introduced undulation and the wall thickness, the outer dimensions
of the structure are 17 × 17 mm. Undulations are achieved by trans­
In this study, the NiTi samples are configured as wavy tetra-achiral forming square edges into a three-point arc: the arc’s endpoints are
lattices, inspired by [26]. These lattices, alongside their tetra-chiral defined by the square lattice’s corners, while the third point is deter­
analogs, are associated with isogonal tilings IG62 and IG63, as delin­ mined by offsetting the edge center in the pertinent direction by 0.5 mm.
eated by Grünbaum and Shephard [32]. Owing to the intrinsic symmetry Consequently, given the parameters used, an arc radius of 2.5 mm is
of tetra-achiral lattices, they lack the axial-shear coupling effect, formed. This central undulated pattern is then thickened and extruded in
resulting in enhanced stability under uniaxial compression, in contrast the z-direction to finalize the design. The resulting watertight solid
to their tetra-chiral counterparts [26]. The shape of the tetra-achiral model is then transitioned into an STL model, readying it for 3D
lattice aligns with the third resonant mode of a square lattice when printing.
analyzed through eigenmode examination. The parameters established
by [27] were used as the basis for these structures, leading to their
auxetic behavior with in-plane Poisson’s ratios of − 0.3 along two pri­
mary structural directions, behaving similarly to a planar orthotropic

2
C.A. Biffi et al. Additive Manufacturing Letters 10 (2024) 100216

Fig. 2. The optical microscopy images of the NiTi lattices show the quality of the sample fabrication by the LPBF technique on the node (a) and on the edge (b).

2.2. Material and fabrication and a total of 3000 images were acquired for the examination. The ob­
tained tomography datasets were reconstructed using Dragonfly soft­
NiTi samples were produced utilizing an LPBF system (Renishaw, ware (Comet Technologies Canada Inc), and the 3D distribution of
AM400, UK) employing Ni56Ti46 (wt.%) spherical powder (Nanoval, deformation or damage in the samples was determined using Dragon­
Germany), having a mesh distribution from 16 μm to 63 μm. The LPBF fly’s segmentation toolkit.
process parameters outlined in Table 1 were chosen for the fabrication of
the NiTi architected structures. The adopted LPBF system was equipped
with a pulsed wave laser, and the pulse duration, rather than scanning 2.4. Microstructural characterization
speed, was utilized to define the appropriate process conditions for
achieving the desired NiTi properties [19,35]. The samples were The martensitic transformation temperatures were detected through
extracted from the platform using a wire electrical discharge cutting differential scanning calorimetry (DSC, model Q25 from TA Instrument,
machine. To enhance the functional characteristics of the printed USA), carried out with a heating/cooling scan rate of 10 ◦ C/min, within
specimens, we applied a specific low-temperature heat treatment pro­ a temperature range from − 100 ◦ C up to 100 ◦ C. Initial NiTi powder, as
cess. This process primarily aims to release the residual stresses induced built and heat treated samples were analyzed through DSC.
by the LPBF process. As a result, the as-built samples underwent Diffraction data (CuKα, λ=1.5418 Å) were collected on a θ:θ vertical
annealing at 500 ◦ C for 5 min, followed by water quenching. scan P analytical X’Pert PRO diffractometer, equipped with parallel
(Soller) slits – 0.04 rad – and a real-time multiple strip detector. The
generator was operated at 40 kV and 30 mA; slits were used with a
2.3. 2D and 3D visual characterization divergence of 0.5◦ The scans were performed in the 10–120◦ 2θ range at
25 ◦ C. Phase identification was carried out on the heat treated NiTi
Optical microscopy images were captured using a Digital Microscope sample.
(Keyence, VHX-7000 N, Belgium) for surface investigation of the NiTi Scanning electron microscopy (SEM) was applied for additional ex­
lattice samples. A selected sample was subjected to X-ray micro­ amination, employing a Thermo Fisher (Model Phenom XL) and JEOL
–Computed Tomography (µ-CT) for detailed qualification. The µ-CT was (Model JSM-7200F) equipped with EDS and electron backscatter
performed using GE equipment (model V, TOMEX), featuring a screen diffraction (EBSD) detectors to thoroughly analyze and characterize the
DXR-250 with a pixel size of 2000 × 2000 | 200 μm. The acquisition microstructure. Image acquisition was conducted using a back-scattered
parameters were set as follows: power=13 W, spatial resolution= 14 µm, electron (BSE) detector to offer a comprehensive depiction of various

Fig. 3. Micro-CT analysis of the built NiTi architecture sample: xy view (a) and 3D visualization (b).

3
C.A. Biffi et al. Additive Manufacturing Letters 10 (2024) 100216

Fig. 4. The trend of the wall thickness of the sample with its statistical analysis.

phases present in the microstructure. To enhance image quality and tests at three levels of force (4–8–12 kN), and each test was repeated
mitigate charging issues during EBSD imaging, specific samples were three times to ensure accuracy and reliability.
embedded in conductive epoxy.
3. Results and discussion
2.5. Mechanical characterization
3.1. Visual investigation of NiTi structures
The mechanical behavior of the NiTi architected structures was
The NiTi structures were successfully printed following the experi­
investigated through a cyclic compression test, conducted with 10 cycles
mental plan outlined in the previous section. As depicted in Fig. 2, the
of loading and unloading to ensure stable mechanical responses for each
printed NiTi lattice exhibits a defect-free surface with no discernible
sample. The tests were performed at room temperature, 25 ◦ C, (using a
cracks. The scanning strategy employed during printing, comprising
tensile test machine (Zwick, model Z100, Germany) equipped with a
meander patterns consisting of contour and hatch paths, is visible on the
100 kN load cell. A constant speed of 0.5 mm/min was maintained for all

Fig. 5. DSC scans of the initial powder, as built and heat-treated NiTi samples.

4
C.A. Biffi et al. Additive Manufacturing Letters 10 (2024) 100216

Table 2
Transformation Temperatures (TT) of the initial powder, as built and heat-treated NiTi samples. All temperature units are ◦ C.
TT As Ap Af Rs Rp Rf Ms Mp Mf

Powder − 32.5 14 25 3 – – – − 30.8 − 58

As-built − 25.7 − 1.5 17 1 – – – − 40.7 − 58.5
HT500 2.5 15.5 28 10.5 0.5 − 10.7 − 37 − 50 − 60

upper surfaces. The choice of the meander strategy, splitted in hatch and
contour, depends on the significant thickness of the architected struc­
ture. Previous works were carried out on the printing of NiTi thinner
lattice samples, requiring the adjustment of the principal process pa­
rameters for implementing a different scanning strategy, namely lattice
[19]. Additionally, linear laser scans were used in the central part of the
samples, while the border was created using concentric lines.
To evaluate the inner quality of the fabricated structures, a µ-CT scan
was performed. The 3D reconstruction in Fig. 3 illustrates the defects’
size through a color scale representation. Notably, the nodes of the
lattice structure exhibit relatively more defects, with an average size of
approximately 100 µm. Also visible are cracks due to the compression
testing. Despite the presence of some defects, the volume fraction was
measured as 0.07% and this is well within the normal range for an
additively manufactured material.
From the µ-CT scans, a wall thickness evaluation could be made.
Fig. 4 shows that the wall thickness ranged mainly around 1 mm (shown
in yellow) with some thicker regions at the junctions (shown in red). The
statistical thickness distribution is included. Fig. 6. XRD diffractogram of the heat-treated NiTi sample.

3.2. Microstructural analysis of NiTi structures was carried out during the LPBF process.
Fig. 6 shows the XRD spectrum that was carried out on the heat-
The microstructure of the NiTi samples was analyzed through DSC, treated sample at room temperature. Austenite (cubic structure, B2)
XRD, and EBSD tests to obtain a complete overview in terms of trans­ was predominantly detected in the NiTi sample, consistent with the
formation temperatures of the martensitic transformation, phase iden­ Austenite Finish (Af) temperature measured from the DSC scans. The
tification, and distribution inside the architectural samples. sample exhibited a preferential crystallographic orientation, influenced
Initially, the transformation temperatures (TTs) of the initial pow­ by the rapid laser motion on the powder bed and the subsequent rapid
der, as-built, and heat-treated samples were measured via DSC. Due to solidification. The most intense diffraction reflection was associated
the significant sensitivity of the performances in NiTi SMAs with respect with the (110) peak. Epitaxial growth and columnar grain growth
to the chemical composition, it was decided to proceed with a controlled the solidification, attributed to the very high cooling rates
comparative analysis of the TTs from the feedstock material through the characteristic of LPBF.
LPBF process to the post-treated sample. Further microstructural Additionally, minor peaks associated with the B19′ phase
investigation would focus only on the heat-treated samples. (martensite) were observed. These minor peaks were expected, given
Fig. 5 presents the DSC test results for the initial NiTi powder, as- that the patterns were collected at room temperature, very close to Af.
built, and heat-treated architected structures. The NiTi powder exhibi­ For a more in-depth understanding of the microstructure, EBSD analysis
ted a multistage martensitic transformation (MT) upon heating and was conducted.
cooling due to compositional inhomogeneities and residual stresses The EBSD analysis provides a comprehensive insight into the grain
arising from the gas atomization process [36]. In contrast, the as-built formation of the 3D printed samples across various sections of the
structure displayed a single-stage MT upon heating and a double-stage structure. Fig. 7 illustrates the outcomes of the EBSD test conducted on
MT upon cooling, with the two peaks largely overlapping each other, the XZ section of the sample. Indexing is carried out based on the body-
resulting in TT values shifting to lower values. The heat treatment centered cubic austenite. Consequently, the orientation image micro­
conducted at 500 ◦ C for 5 min (indicated as HT 500) promoted a sharper graphs (OIM) are computed for the normal direction of the section,
peak upon heating, indicating the martensite-to-austenite trans­ along with the inverse pole figures (IPF) in the Z, Y, and X directions.
formation. Additionally, two well-defined peaks were observed during As depicted, there are non-indexed areas represented in black, sug­
cooling. The first peak corresponded to the transformation from gesting a potential association with localized martensite regions, as re­
austenite to R-phase, distinguished by limited thermal hysteresis be­ ported in other work in the literature [17]. However, most of the
tween the peaks upon heating and cooling. The second peak was related analyzed surfaces exhibited the presence of austenite, which is in good
to the phase transformation from R-phase to martensite. agreement with the XRD (Fig. 6) and DSC (Fig. 5) analyses. Similar re­
Table 2 presents the values of TTs, obtained from the DSC scans sults were achieved in the literature [37,38]. The results further reveal a
shown in Fig. 5. This includes the evaluation of austenite start (As), peak lack of uniform texture, with the latter manifesting as a more reddish
(Ap), and finish (Af), R-phase start (Rs), peak (Rp), and finish (Rf), as center and displaying random colors in the OIM. The specimen exhibited
well as martensite start (Ms), peak (Mp), and finish (Mf) temperatures, elongated grains in the building direction (see Fig. 7), while the shape of
respectively. the typical liquid pools is not detectable. On the contrary, columnar
In conclusion, the NiTi powder shows wide and partially overlapped grains can be observed in Fig. 7, growing orthogonally to the platform,
peaks of the MT; the corresponding values of TTs confirmed a complete as confirmed by other works in the literature [35].
phase transformation in austenite at room temperature, promoting a Fig. 8 presents the results of the EBSD analysis conducted on the XY
superelastic behavior. The TTs of the as-built structure, ranged in a section of the NiTi samples. In these images, epitaxial grains with di­
similar temperature interval of the power, meaning no significant Ni loss mensions significantly smaller than those in the XZ view (refer to Fig. 7)

5
C.A. Biffi et al. Additive Manufacturing Letters 10 (2024) 100216

Fig. 7. EBSD analysis of the XZ section in the NiTi structure, depicted using orientation image micrographs (OIM). The visuals correspond to (a) Band Contrast (BC)
and (b-d) inverse pole figures (IPF) in the Z, Y, and X directions, respectively.

Fig. 8. EBSD findings on the XY section of the NiTi structure, illustrated through orientation image micrographs (OIM). The images correspond to (a) Band Contrast
(BC) and (b-d) inverse pole figures (IPF) in the Z, Y, and X directions, respectively.

6
C.A. Biffi et al. Additive Manufacturing Letters 10 (2024) 100216

Fig. 9. The result of fracture compression test for a NiTi re-entrant

auxetic structure.

were observed. This observation suggests the presence of a notable

anisotropic microstructure.
Notably, it highlights a junction on the cellular face where four struts
converge at a single point. Similar to the previous picture, the non-
indexed regions align with martensite areas, potentially indicative of
the presence of stress-induced martensite (SIM). This occurrence may be
attributed to stress concentration regions between the struts and nodes.
Similar results were also achieved in other works in literature [17]. It
can be also noticed that finer microstructure can be seen in the center of
the node, while the border of the lattices exhibited slightly larger grains.
This may be due to the adopted scanning strategy (meander), in which
the contour and the hatch were carried out with different process
parameters.

3.3. Mechanical performance of NiTi structures

The compression test was also performed to assess the mechanical

strength of the fabricated NiTi lattices at room temperature and deter­
mine the point of complete fracture. As shown in Fig. 9, the initial crack
initiation occurred at a force of 16.1 kN; however, the complete fracture
of the structure was observed at a higher compression force of 19.1 kN,
with a corresponding deformation of approximately 7.6 %. The fracture
is attributed to stress concentration in the corner of the sample with
sharp angles, leading to the failure of the entire sample.
The results from the cyclic compression test demonstrated that the
NiTi architected structures displayed stable superelastic responses
across all three loading forces (4 kN, 8 kN, and 12 kN). As shown in
Fig. 10, except for the first few cycles, the samples maintained their
mechanical integrity and superelastic behavior throughout multiple
cycles of loading and unloading.
Fig. 11 illustrates the total displacement and recoverable de­
formations of the NiTi lattice samples subjected to compression tests at 4
Fig. 10. The result of the cyclic compression test with a maximum force of (a) 4
kN, 8 kN, and 12 kN. As anticipated, an increase in the applied force,
kN, (b) 8 kN, and (c) 12 kN after 10 cycles.
ranging from 4 kN to 12 kN, results in a notable growth in the total
displacement by the end of the final loading cycle, measuring 0.4507
mm, 0.5004 mm, and 0.7658 mm respectively. This trend is also kN, and 12 kN compression tests in Fig. 12. To ascertain the specific
observed in the recoverable deformation values, calculated from the dissipation energy, this total value is normalized by the mass of the
unloading displacement of the 10th cycle, revealing values of 0.2572 corresponding lattice.
∫ ∫
mm, 0.4081 mm, and 0.6499 mm for the 4 kN, 8 kN, and 12 kN
compression tests, respectively. These findings emphasize the struc­ EDisspiated = Fdu − Fdu (1)
loading unloading
tures’ capacity to endure repetitive stress and effectively revert to their
initial shape, thus making them well-suited for applications demanding The respective lattice masses for compression tests conducted at 4
reliable and recoverable mechanical responses. kN, 8 kN, and 12 kN are 9.327 gs, 9.308 gs, and 9.203 gs. Notably, the
Eq. (1) gives the total dissipated energy for a cyclic force- initial observations indicate that the obtained results remain stable
displacement curve. To this end, the energy associated with the beyond the initial cycles, persisting throughout the subsequent repeti­
unloading phase, i.e., energy release, is subtracted from the energy tions up to the 10th cycle. Furthermore, the specific energy dissipation
associated with the loading phase, i.e., energy input. The total energy rate can be significantly augmented through an increment in compres­
dissipation for each successive loading cycle is depicted for the 4 kN, 8 sion force. This augmentation is evidenced by the obtained saturated

7
C.A. Biffi et al. Additive Manufacturing Letters 10 (2024) 100216

Fig. 11. Summary of compression testing results, showing the total displacement and the recoverable deformation ratio of the NiTi architected structures when
compressed to 4 kN, 8 kN, and 12 kN.

Fig. 12. Dissipated energy and specific dissipation energy values for the NiTi lattice samples were compressed to 4 kN, 8 kN, and 12 kN.

specific energy dissipation values of 0.0821 kN.mm/gr, 0.2802 kN.mm/ an applied force of up to 12 kN. This underscores the capability of
gr, and 0.7747 kN.mm/gr for the respective loads of 4 kN, 8 kN, and 12 this design to enhance the functional performance and deformability
kN. of the structure, making it highly suitable for versatile and adaptive
applications.
4. Conclusion
The combination of superelasticity, high deformation capacity, and
This study investigated the mechanical performance of 3D printed efficient energy dissipation showcased in this study holds significant
NiTi architected metamaterials. The fabricated samples showcased a implications for various engineering fields, such as aerospace, robotics,
remarkable superelastic response emphasizing their potential for diverse and medical devices. These findings provide a solid foundation for
and adaptable applications. Moreover, these samples displayed an further advancements in the design and optimization of 3D printed NiTi
excellent capacity for energy dissipation and damping behavior, architected metamaterials, opening new avenues for materials engi­
underscoring their efficiency in dissipating mechanical energy. The neering and enhanced performance in engineered systems.
outcomes of this research are summarized as follows:
CRediT authorship contribution statement
• The designed auxetic meta-structures were successfully fabricated
using an LPBF printing setup and tested using a wide range of C.A. Biffi: Writing – review & editing, Resources, Investigation,
characterization methods. The use of µ-CT scanning allowed us to Formal analysis. C. Soyarslan: Writing – review & editing, Software,
evaluate the relative density of the lattice structures, which was Formal analysis, Conceptualization. J. Fiocchi: Investigation, Formal
found to be as high as 99.85 %, signifying excellent overall structural analysis. C. Bregoli: Investigation, Formal analysis. A. du Plessis:
integrity. Writing – review & editing, Investigation, Formal analysis. A. Tuissi:
• Low-temperature heat treatment, carried out at 500 ◦ C for 5 min, was Visualization, Investigation. M. Mehrpouya: Writing – original draft,
successfully performed to promote the superelastic properties of the Supervision, Methodology, Investigation, Formal analysis,
3D printed NiTi samples. The results showed that the martensitic Conceptualization.
transformation was set in a temperature range suitable to obtain the
austenitic phase.
Declaration of competing interest
• The fabricated NiTi metamaterials demonstrated significant recov­
erable deformation strains ranging from 1.5 to 3.8 %, achieved under
The authors declare the following financial interests/personal

8
C.A. Biffi et al. Additive Manufacturing Letters 10 (2024) 100216

relationships which may be considered as potential competing interests: [13] A. du Plessis, et al., Properties and applications of additively manufactured
metallic cellular materials: a review, Prog Mater Sci (2021) 100918.
One of the authors of this article is part of the Editorial Board of the
[14] F. Yazdandoost, et al., Energy dissipation of shock-generated stress waves through
journal. To avoid potential conflicts of interest, the responsibility for the phase transformation and plastic deformation in NiTi alloys, Mech. Mater. 137
editorial and peer-review process of this article lies with the journal’s (2019) 103090.
other editors. Furthermore, the authors of this article were removed [15] D. Kim, et al., 3D and 4D Printing of Complex Structures of Fe Mn Si-Based Shape
Memory Alloy Using Laser Powder Bed Fusion, Adv. Mater Interfaces 9 (13) (2022)
from the peer review process and had no and will not have any access to 2200171.
confidential information related to the editorial process of this article. [16] R. Xi, et al., Effect of Fe addition on the microstructure, transformation behaviour
and superelasticity of NiTi alloys fabricated by laser powder bed fusion, Virtual
Phys. Prototyp. 18 (1) (2023) e2126376.
Data availability [17] S. Khademzadeh, Precision additive manufacturing of NiTi shape memory parts
using micro-laser powder bed fusion, Progr. Addit. Manuf. (2021) 1–14.
Data will be made available on request. [18] T. Gustmann, et al., Properties of a superelastic NiTi shape memory alloy using
laser powder bed fusion and adaptive scanning strategies, Progre. Addit. Manuf. 5
(2020) 11–18.
[19] C.A. Biffi, et al., Microstructural and mechanical response of NiTi lattice 3D
Acknowledgements structure produced by selective laser melting, Metals (Basel) 10 (6) (2020) 814.
[20] S. Bernard, et al., Compression fatigue behavior of laser processed porous NiTi
alloy, J. Mech. Behav. Biomed. Mater. 13 (2012) 62–68.
The authors express their gratitude to Nicola Bennato for his support [21] M.T. Andani, et al., Mechanical and shape memory properties of porous Ni50.
in the 3D printing process. Also, we would like to appreciate the support 1Ti49. 9 alloys manufactured by selective laser melting, J. Mech. Behav. Biomed.
Mater. 68 (2017) 224–231.
of Mr. Nick Helthuis and Mr. Bert Vos for their valuable assistance in
[22] S. Dadbakhsh, et al., Laser additive manufacturing of bulk and porous shape-
setting up the microstructure and mechanical testing procedures. memory NiTi alloys: from processes to potential biomedical applications, MRS
Bulletin 41 (10) (2016) 765–774.
References [23] W. Chen, et al., Compressive mechanical properties and shape memory effect of
NiTi gradient lattice structures fabricated by laser powder bed fusion, Int. J.
Extreme Manuf. 4 (4) (2022) 045002.
[1] M. Mehrpouya, H.C. Bidsorkhi, MEMS Applications of NiTi Based Shape Memory [24] J. Sun, et al., Laser powder bed fusion of diatom frustule inspired bionic NiTi
Alloys: a Review, Micro Nanosyst 8 (2) (2016) 79–91. lattice structures: compressive behavior and shape memory effect, Smart Mater.
[2] Elahinia, M., Shape Memory Alloy Actuators: Design, Fabrication and Experimental Struct. 31 (7) (2022) 074003.
Evaluation. 2015, John Wiley & Sons. [25] Z. Yan, et al., Superelastic response and damping behavior of additively
[3] Mehrpouya, M., et al., Laser Welding of Nickel-Titanium (NiTi) Shape Memory manufactured Nitinol architectured materials, Addit. Manuf. 68 (2023) 103505.
alloys, in Advanced Welding and Deforming. 2021, Elsevier. p. 203–230. [26] Z. Yuan, Z. Cui, J. Ju, Micropolar homogenization of wavy tetra-chiral and tetra-
[4] M. Köhl, et al., Characterization of porous, net-shaped NiTi alloy regarding its achiral lattices to identify axial–shear coupling and directional negative Poisson’s
damping and energy-absorbing capacity, Mater. Sci. Eng.: A 528 (6) (2011) ratio, Mater Des. 201 (2021) 109483.
2454–2462. [27] C. Körner, Y. Liebold-Ribeiro, A systematic approach to identify cellular auxetic
[5] S. Saedi, et al., Energy Damping in Shape Memory Alloys: a Review, J. Alloys materials, Smart Mater. Struct. 24 (2) (2014) 025013.
Compd. (2023) 170286. [28] C.A. Biffi, et al., Functional performance of NiTi shape memory architected
[6] M. Benedetti, et al., Architected cellular materials: a review on their mechanical structures produced by laser powder bed fusion (LPBF), Trans. Addit. Manuf. Meets
properties towards fatigue-tolerant design and fabrication, Mater. Sci. Eng.: R: Rep. Med. 5 (S1) (2023) 821.
144 (2021) 100606. [29] A. Mauko, et al., Dynamic deformation behaviour of chiral auxetic lattices at low
[7] D. Gu, et al., Additively manufacturing-enabled hierarchical NiTi-based shape and high strain-rates, Metals (Basel) 11 (1) (2020) 52.
memory alloys with high strength and toughness, Virtual Phys. Prototyp. 16 (sup1) [30] X.-c. Zhang, et al., Dynamic crushing responses of bio-inspired re-entrant auxetic
(2021) S19–S38. honeycombs under in-plane impact loading, Mater. Today Commun. 23 (2020)
[8] L. Riva, P.S. Ginestra, E. Ceretti, Mechanical characterization and properties of 100918.
laser-based powder bed–fused lattice structures: a review, Int.. J. Adv. Manuf. [31] Y. Zhang, et al., Static and dynamic properties of a perforated metallic auxetic
Technol. 113 (2021) 649–671. metamaterial with tunable stiffness and energy absorption, Int. J. Impact Eng. 164
[9] M.K. Ravari, et al., On the effects of geometry, defects, and material asymmetry on (2022) 104193.
the mechanical response of shape memory alloy cellular lattice structures, Smart [32] B. Grünbaum, G.C. Shephard, Tilings and Patterns, Courier Dover Publications,
Mater. Struct. 25 (2) (2016) 025008. 1987.
[10] S. Afkhami, et al., Mechanical performance and design optimisation of metal [33] Dolla, W.J.S., B.A. Fricke, and B.R. Becker, Structural and drug diffusion models of
honeycombs fabricated by laser powder bed fusion, Thin-Walled Struct. 180 (2022) conventional and auxetic drug-eluting stents. 2007.
109864. [34] Y. Zhang, et al., In-plane compressive properties of assembled auxetic chiral
[11] C.W. Isaac, F. Duddeck, Recent progress in 4D printed energy-absorbing honeycomb composed of slotted wave plate, Mater. Des. 221 (2022) 110956.
metamaterials and structures, Virtual Phys. Prototyp. 18 (1) (2023) e2197436. [35] C.A. Biffi, et al., Selective laser melting of NiTi shape memory alloy: processability,
[12] A. Jafarabadi, et al., 4D printing of recoverable buckling-induced architected iron- microstructure, and superelasticity, Shape Memory Superelastic. 6 (2020)
based shape memory alloys, Mater. Des. 233 (2023) 112216. 342–353.