Thin-Walled Structures 176 (2022) 109369

Contents lists available at ScienceDirect

Thin-Walled Structures
journal homepage: www.elsevier.com/locate/tws

Full length article

A hierarchical fabrication method and crushing response of metallic 3D

re-entrant honeycomb
Lian-Zheng Pei a , Chang Qi a,b , Shu Yang a,b ,∗, Yuan-Hang Ma a , Peng-Cheng Wu a
a
State Key Laboratory of Structural Analysis for Industrial Equipment, School of Automotive Engineering, Dalian University of Technology, Dalian 116024, China
b
Ningbo Institute of Dalian University of Technology, Ningbo 315016, China

ARTICLE INFO ABSTRACT

Keywords: Due to the negative Poisson’s ratio (NPR) effect, auxetic structures such as re-entrant honeycombs show
Auxetic honeycomb excellent energy absorption and force mitigation capabilities under crushing load. Compared with two-
Negative Poisson’s ratio dimensional (2D) auxetic structure, three-dimensional (3D) ones are expected to have better performance, but
Fabrication method
they are more difficult to fabricate. In this study, a hierarchical fabrication method for metallic 3D re-entrant
Crush loading
honeycomb (3D-RH) using traditional manufacturing techniques is proposed. A 3D-RH specimen was crafted
Blast loading
and tested under quasi-static crushing condition. The finite element (FE) model was established and verified by
the experimental results. The numerical simulation results showed that the friction condition between 3D-RH
specimen and loading surfaces has a great influence on the deformation mode of specimen. Moreover, the
energy absorption capability of 3D-RH was numerically compared with a similar structure, i.e., the warp and
woof periodic auxetic cellular (WWPAC) structure. The results showed that the specific energy absorption (SEA)
of 3D-RH is 1.59 times that of WWPAC and the 3D-RH outperformed the latter in lightweight performance.
Finally, the blast resistance of 3D-RH sandwich panel was numerically studied. Compared with 2D re-entrant
expanded honeycomb (2D-REH) core, 3D-RH core can reduce the maximum deflection of lower face-sheet by
14.6%, and increase SEA by 20.5%, showing better blast resistance.

1. Introduction Therefore, the performance of 2D auxetic honeycomb is orthotropic

and related to cell orientation. In order to obtain more isotropic struc-
Auxetic honeycomb structure has become one of the research tures, many researchers have explored the possibility of revealing NPR
hotspots in recent years because of its great energy absorption poten- effects in two or more directions, i.e., three-dimensional (3D) auxetic
tials, lightweight and excellent mechanical properties [1,2]. Different structures. Due to the lack of technical means to manufacture 3D NPR
from traditional honeycombs, the auxetic honeycomb will shrink lat- honeycomb, early research mainly focused on studying of material
erally towards the loading area when encountering crush or impact properties by theoretical analysis and numerical simulation. Imbalzano
loadings. This unique behavior is characterized by negative Poisson’s et al. [23,24] numerically studied the blast and impact resistance of
ratio (NPR) effect. Due to this effect, auxetic honeycomb can dissipate a 3D auxetic lattice. The results revealed the superiorities of such
energy at a relatively low and constant stress level under compression. 3D auxetic structure in energy absorption and back facet deforma-
This makes auxetic honeycomb very suitable for energy absorption and tion. Inspired by the rotational rigidity mechanism, Gao et al. [25]
force mitigation purposes, such as constructing protective devices for introduced a method to generate a series of novel 3D auxetic honey-
infrastructure and vehicle to resist blast shock waves. In recent years, combs, whose Poisson’s ratio can be adjusted in a wide range from
the mechanical performances of various auxetic honeycomb structures negative to positive. Li et al. [26,27] designed a graded 3D auxetic
represented by re-entrant honeycomb [3–11], double arrowhead hon- lattice and compared its impact resistance with non-graded one. The
eycomb [12–15], chiral honeycomb [16–18], and their derivatives [19– simulation results show that the graded honeycomb outperforms the
22], have been thoroughly studied by theoretical, experimental, and non-graded one in both the back facet deflection and contact force.
numerical methods. Gao et al. [28] theoretically studied the mechanical properties of 3D
Most of the auxetic honeycombs investigated in the above studies double-arrowhead lattice, including Poisson’s ratio, Young’s modulus
are two-dimensional (2D), and their NPR effect could only be observed and static collapse stress. The proposed new theoretical method is
from one direction that is perpendicular to the loading direction. more accurate when considering the influence of adjacent cells. Li

∗ Corresponding author at: State Key Laboratory of Structural Analysis for Industrial Equipment, School of Automotive Engineering, Dalian University of
Technology, Dalian 116024, China.
E-mail address: [email protected] (S. Yang).

https://doi.org/10.1016/j.tws.2022.109369
Received 29 December 2021; Received in revised form 28 March 2022; Accepted 25 April 2022
Available online 16 May 2022
0263-8231/© 2022 Elsevier Ltd. All rights reserved.
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

Fig. 1. Specimens of (a) 3D re-entrant auxetic [33] and (b) WWPAC [34].

Fig. 2. Geometric configurations of 3D-RH structure and its unit cell: (a) unit cell in iso-view, (b) unit cell in side view, (c) 3D-RH in iso-view, and (d) 3D-RH in side view.

et al. [29] proposed a 3D foam derived from Voronoi structure. The test indicated that the manufacturing quality and internal structure
crushing simulation shows that the foam possesses negative Poisson’s interaction may affect the accuracy of the analytical model.
ratio and isotropic mechanical properties. In addition, the 3D foam Although additive manufacturing technology is very convenient to
exhibits better blast resistance than the re-entrant honeycomb. During fabricate 3D auxetic structures, metal 3D printing technology, for exam-
the same period, only a few studies have managed to fabricate and test ple, SEBM (selective electron beam melting) [39,46–48] or SLM (selec-
3D auxetic structures. For example, by using silicon injection moulding tive laser melting) [49–52], is still too expensive and time-consuming
technique, Smardzewski et al. [30–32] prepared auxetic springs that for mass production at this stage. In addition, some studies have
can be installed in the seat cushion, and investigated their crushing reported that additive manufacturing methods may introduce defects
behaviors via experimental, numerical and theoretical means. such as rough surface [53], irregular cross section [53] and stair
With the development of additive manufacturing technology in re- stepping effect [54]. To avoid the above shortcomings, Wang et al. [33,
cent years, the fabrication of 3D auxetic honeycomb becomes feasible, 55] introduced the interlocking assembly method and constructed a
so the mechanical properties of 3D auxetic honeycomb can be tested
3D re-entrant lattice, as shown in Fig. 1a. However, this method is
and verified through experimental methods. Gao et al. [25,35] tested
more suitable for constructing truss-type structures. Inspired by the
the crushing behavior of 3D double-arrowhead lattice under quasi-
interlocking concept, Safikhani et al. [34] fabricated the warp and woof
static crushing and further studied its crashworthiness. Li et al. [36,37]
periodic auxetic cellular (WWPAC) structure specimens, as shown in
designed a novel 3D re-entrant unit cell structure and a hierarchical
Fig. 1b, which is similar to the 3D re-entrant hexagonal honeycomb
octet-truss structure with re-entrant hexagon as higher order structure.
(3D-RH) studied in the presented research, as shown in Fig. 2. The
With the help of 3D printing, the crushing behaviors of the two struc-
tures were experimentally examined. Novak et al. [38–40] observed the WWPAC specimens are made by riveting pre-bended metal strips layer
inertial effect from the experimental tests when the metallic 3D chiral by layer. This fabrication method causes two main disadvantages of
lattice was crushed under quasi-static, low- and high-speed impact the specimen. First, the thickness of the horizontal cell walls is twice
loads. Li et al. [41–43] tested the vibration response of a novel 3D that of the inclined cell walls. The results discussed in Section 6.3 show
re-entrant lattice. Base on validated FE models, the effects of thermal that the inclined cell walls absorb most of the energy. Therefore, the
conditions, boundary conditions and structural gradient design were energy absorption per unit mass is reduced. Secondly, the WWPAC
studied. Shen et al. [44] designed and printed a new type of 3D specimen contains many overlapping regions of the metal strips. These
re-entrant lattice by combining the 2D re-entrant components. The overlapping regions and the rivets left in the specimen have very high
effective modulus and yield strength of the material were derived by stiffness and are not easy to deform during crushing. Therefore, the
using the beam theory and verified by the compression test results. overlapping regions and rivets not only weaken the energy absorption
Yang et al. [45] established an analytical model of 3D re-entrant capacity, but also add redundant mass to the WWPAC. The energy
lattice based on beam model. The results of quasi-static compression absorption per unit mass is further decreased. It is not difficult to find

2
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

Fig. 3. Main gist of hierarchical fabrication method.

that the lightweight characteristics of WWPAC are deeply affected by

its manufacturing method.
In this paper, we propose a hierarchical fabrication method to
construct 3D-RH structure, as shown in Fig. 2. This method only
adopts conventional manufacturing processes such as cutting, bending
and wielding, which has high efficiency and low cost. Moreover, this
method can fully arrange the substrate material and will not introduce
excess mass into the specimen, so it will not damage the lightweight
performance of the specimen. A 3D-RH specimen was fabricated and
quasi-static crushing test was carried out. The finite element (FE) model
was established and validated by the test results. Good qualitative and
quantitative consistencies were obtained. On this basis, the influence of
friction coefficient on the crushing response was numerically studied.
Then, the energy absorption capability of the 3D-RH was compared
with the WWPAC. Finally, the blast resistance performance of 3D-RH
and 2D-RH structures was compared.

2. Physical model and fabrication method

2.1. Model description Fig. 4. Shapes and dimensions of basic parts: (a) waveform part, (b) half-cross part,
and (c) long and short bands.

The three-dimensional re-entrant honeycomb (3D-RH) and its unit

cell are shown in Fig. 2. The unit cell can be considered as two
traditional re-entrant hexagons merged together perpendicularly, so its shape and are easy to fabricate using conventional manufacturing tech-
top and bottom are cross shaped. As depicted in Fig. 2a and b, the niques such as cutting, bending, and stamping etc. Taking the waveform
geometric parameters of 3D-RH cell are cross length 𝑙1 , inclined wall part as example, it can be fabricated by firstly cutting a long strip from
length 𝑙2 , horizontal wall length 𝑙3 , cell wall width b, re-entrant angle the metal plate and then bending it into waveform.
𝜃, and cell wall thickness t. In order to prevent improper interference The process of constructing a sub-structure using the basic parts is
of cell wall, considering the contraction behavior of the re-entrant schematically described in Fig. 5. First, the basic parts are assembled
hexagon, the geometric parameters should meet the criterion of 2𝑙2 + into an ‘‘L’’ shape using traditional manufacturing procedures, such
b < 𝑙1 . By repeating the unit cell along the X, Y and Z directions, the as welding. Then, four ‘‘L’’ shapes are connected head-to-tail to form
3D-RH can be constructed, as shown in Fig. 2c and d. In this study, the one of the sub-structures in Fig. 3, i.e., the O-block. In a similar way,
baseline 3D-RH model was configured as 𝑙1 = 2 mm, 𝑙3 = 50 mm, 𝑙2 = the other two sub-structures, half O-block and quarter O-block, can be
16 mm, 𝑡 = 1 mm, 𝑏 = 10 mm, 𝜃 = 50◦ , and it has 3 units in X, Y and Z constructed, as shown in Fig. 6.
directions, respectively. Like LEGO toys, the sub-structures are the bricks to assemble the
3D-RH specimen. It takes two O-blocks, four half O-blocks, and two
2.2. Hierarchical fabrication method quarter O-blocks to assemble one baseline 3D-RH specimen, as shown
in Fig. 2c. For this purpose, the eight sub-structures shall be welded
The main gist of the hierarchical fabrication method is described in together in the following order: ⃝ 1 to ⃝, 4 as shown in Fig. 7. For
Fig. 3. From left to right, the 3D-RH specimen can be disassembled into better visualization, the O-blocks, half O-blocks and quarter O-blocks
three types of sub-structures, and further into four types of basic parts. are colored in blue, green and red, respectively, in Fig. 7.
From right to left, the basic parts are assembled into the sub-structures The proposed hierarchical fabrication method has several advan-
and further into the specimen. From the perspective of hierarchy, it tages. Firstly, the assembly procedure used in this method is simple
can be seen that the 3D-RH specimen represents the macroscopic scale, and low-cost. Traditional manufacturing techniques, such as cutting,
the sub-structure represents the mesoscopic scale, and the basic part bending and welding, are sufficient to fabricate 3D-RH specimens.
represents the micro scale. The relationship between these hierarchies Secondly, this method can better arrange the substrate material. The
is very similar to that between fibers, yarns and fabrics. cell wall thickness of WWPAC is uneven, and the thickness of horizontal
Fig. 4 shows the shape and dimensions of the four basic parts, cell walls is twice that of inclined cell walls. The 3D-RH fabricated
named as waveform part (Fig. 4a), half-cross part (Fig. 4b), and long herein has uniform cell wall thickness. The results discussed in Sec-
and short bands (Fig. 4c), respectively. Considering 𝑙1 = 2𝑙3 in the tion 6.3 show that the inclined cell walls absorb much more energy
configuration of baseline 3D-RH model, the length of long band is twice than the horizontal ones. Therefore, by arranging more substrate ma-
that of short band. It can be seen that all the basic parts are in simple terial in the high-energy absorption region, i.e., inclined cell walls,

3
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

Fig. 5. Process to construct O-block.

Fig. 6. Trimetric and top views of (a) O-block, (b) half O-block and (c) quarter O-block.

Fig. 7. Order of connecting sub-structures to construct 3D-RH specimen. (For interpretation of the references to color in this figure legend, the reader is referred to the web
version of this article.)

4
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

Fig. 8. Examples of 3D-RH specimen with in-plane shape of (a) triangular and (b) ‘‘T’’ shape.

Fig. 9. Fabrication of 3D-RH physical samples: (a) basic parts, (b) handheld laser welding machine, (c) welded sub-structures, and (d) finished 3D-RH specimen.

the proposed hierarchical fabrication method endows the 3D-RH with Fig. 9b. The maximum power of the laser welding machine is 500 kW,
better lightweight performance than the WWPAC. Thirdly, the method which can easily weld 1 mm thick steel plate without solder. According
can expand the in-plane (X–Y plane) shape of 3D-RH specimens. The to the above assembly method, one 3D-RH specimen was manually
commonly studied 3D-RH specimen has, say three cells in the X and constructed. The assembled sub-structures and the 3D-RH specimen are
Y directions to form a rectangular in-plane shape. However, using shown in Fig. 9c and d, respectively.
the three types of sub-structures, as shown in Fig. 8, 3D-RH speci-
mens with varied in-plane shapes can be constructed to meet different 3. Experimental setup
requirements.
In this study, 3D-RH specimens were made of 201 stainless steel Crushing tests have been widely used to study the deformation
for lower cost and better solderability. Four types of basic parts man- behavior and load-bearing capacity of cellular structures, such as
ufactured by laser cutting and bending are shown in Fig. 9a. One wood [56], metallic foam [57], and honeycomb [58,59]. In this study,
3D-RH specimen consists of 36 waveform parts, 72 half-cross parts, quasi-static crushing test was carried out on the 3D-RH specimen to
36 long bands and 36 short bands. The basic parts were assembled check its integrity and load-bearing capability. The test was performed
by laser welding, and the handheld laser welding machine is shown in with a universal testing machine, and its setting is shown in Fig. 10a.

5
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

Fig. 10. Experimental setup for quasi-static crushing of 3D-RH specimen.

Fig. 11. FE modeling method: (a) setup for quasi-static crushing simulation; (b) mesh size.

The 3D-RH specimen was placed on the lower platen, and the upper with 5 integration points. The common-node method was adopted to
platen crushed the specimen from the top at a constant speed of connect adjacent elements. By balancing the computational effort and
2 mm/min. The crushing force was recorded by the load cell installed accuracy, the average element size of 3D-RH specimen was set to 2 mm.
on the upper pressing platen. In order to observe the NPR effect of Assumed as rigid, the platens were modeled by 8-node solid elements,
the specimen, two 4 mm thick steel plates were placed between the and the average element size was 10 mm. The meshed model is shown
specimen and the upper and lower platens, respectively. Lubricant was in Fig. 11b.
greased on the surfaces of two steel plates facing the specimen to reduce
the friction between them. 4.2. Material model and parameters

4. Finite element model

The material of 3D-RH specimen is 201 stainless steel, which is de-
scribed by *MAT_ PIECEWISE_LINEAR_PLASTICITY constitutive model
4.1. Modeling method
in the FE model. The material parameter values are as follows: mass
density (RO) is 7.85 g/cm3 , Young’s modulus (E) is 200 GPa, Poisson’s
To numerically study the crushing behavior of 3D-RH, a finite
ratio (PR) is 0.33, yield stress (SIGY ) is 397 MPa, and failure strain
element (FE) model was developed and solved by the nonlinear explicit
(FAIL) is 0.52 [60]. Fig. 12 shows the true stress-effective plastic strain
program LS-DYNA 971. The FE model consists of the 3D-RH specimen,
the lower and upper platens, as shown in Fig. 11. Since the steel plates curve of 201 stainless steel. The strain rate effect was accounted for
showed no deformation during the crushing test, they were amalga- using the Cowper–Symonds (C-S) model with related parameters 𝐶 =
mated with the platens in the FE model to reduce computational cost. 40.4 s−1 and 𝑃 = 5 [5]. The platens were treated as rigid bodies
The lower platen was fixed with 6 degrees of freedom, while the upper and modeled using *MAT_RIGID. The input deck in the numerical
platen was only allowed to move at a constant speed of 1 m/s in the Z- simulations is tabulated in Table 1.
direction [34]. An AUTOMATIC_SURFACE_TO_SURFACE contact algo-
rithm was used to deal with the contact between the specimen and the 5. Experimental results and FE model validation
two platens. Among the cell walls, the AUTOMATIC_SINGLE_SURFACE
algorithm was defined to treat the self-contact. The friction coefficient Final deformation of the 3D-RH specimen after quasi-static crush-
of the contact algorithms was set to 0.1. ing test is shown in Fig. 13a. It can be seen that the specimen has
The 3D-RH specimen was modeled using fully integrated 4-node been crushed to the densification stage, and the cell walls are stacked
shell elements. The thickness of these elements was assigned as 1 mm together. Under such severe deformation, except for some weld seam

6
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

Table 1
Input deck of material models and C-S model parameters in numerical simulations (unit: mm, kg, ms).
Material Part LS-DYNA material type, material input deck and C-S model parameters
201 stainless steel 3D-RH *MAT_PIECEWISE_LINEAR_PLASTICITY
RO E PR SIGY FAIL C P LCSS
7.85e−6 200 0.33 0.397 0.52 40.4 5 Fig. 12
Rigid Platens *MAT_RIGID
RO E PR
7.85e−6 210 0.33

results turned out matching the test results perfectly. The influence of
the friction coefficient will be discussed in Section 6.2.
Fig. 15 compares the quasi-static crushing force–strain curves of
the 3D-RH specimen obtained from test and simulation. Very good
consistency has been achieved in terms of both force level and den-
sification strain. Combined with the above deformation results, it can
be shown that the proposed FE modeling method is capable to simulate
the behavior of 3D-RH under quasi-static load.

6. Discussion

6.1. Poisson’s ratio

As a fundamental feature of the structure, the Poisson’s ratio of 3D-

RH will be discussed in this section. It is assumed that the deformation
of 3D-RH unit cell in the elastic deformation stage is sufficiently small,
whereupon the change of unit cell geometry only depends on the re-
Fig. 12. True stress-effective plastic strain curve of 201 stainless steel [60]. entrant angle 𝜃. Therefore, the horizontal and vertical strains can be
given as
𝑑𝑊 1 𝑑𝑊 1 𝑑(2𝑙1 − 2𝑙2 cos 𝜃) 𝑙2 sin 𝜃
failures caused by relatively rough manual welding process (indicated 𝜀h = = 𝑑𝜃 = 𝑑𝜃 = 𝑑𝜃 (1)
𝑊 𝑊 𝑑𝜃 𝑊 𝑑𝜃 𝑙1 − 𝑙2 cos 𝜃
by red arrows in Fig. 13a), the specimen basically maintains its in-
tegrity. If NC welding process and special fixture were used in mass and
production, such failure modes can be avoided. The effectiveness of the 𝑑𝐻 1 𝑑𝐻 1 𝑑(2𝑙2 sin 𝜃) cos 𝜃
𝜀v = = 𝑑𝜃 = 𝑑𝜃 = 𝑑𝜃 (2)
proposed fabrication method has been verified. Fig. 13b shows the final 𝐻 𝐻 𝑑𝜃 𝐻 𝑑𝜃 sin 𝜃
deformation of the specimen simulated by FE model, which is in good where W and H are the width and height of the 3D-RH unit cell,
agreement with the experimental results. By embedding some defects in respectively, as in Fig. 2. Then, the Poisson’s ratio during the elastic
the FE model, the weld seam failure in the test can even be accurately deformation stage can be written as
simulated.
𝜀h 𝑙2 sin2 𝜃
Several snapshots were captured from the test and simulation to re- 𝜈elastic = − =− (3)
store the quasi-static crushing process of the 3D-RH specimen, as shown
𝜀v 𝑙1 cos 𝜃 − 𝑙2 cos2 𝜃
in Fig. 14. In the figure, four green dashed lines are added to denote Given the condition that 𝑙1 > 2𝑙2 + b and 𝜃 ∈ (0, 𝜋∕2), the value of 𝜈elastic
the original width of the specimen. With their help, the decrease of should always be less than 0. Considering the unit cell configuration,
specimen width can be easily observed, indicating a negative Poisson’s 𝑙1 = 2, 𝑙3 = 50 mm, 𝑙2 = 16 mm, and 𝜃 = 50◦ , the Poisson’s ratio of 3D-
ratio of the 3D-RH. Fig. 14 also reflects the behavior of 3D-RH under RH specimen during the elastic deformation stage is obtained as −0.368
quasi-static crushing. After the crushing began, the 3D-RH specimen according to Eq. (3).
started to deform, and the NPR effect appeared simultaneously, and The Poisson’s ratio of 3D-RH specimen during the whole crushing
the cell walls showed the tendency to contract towards the center of process was traced by the FE model, as shown in Fig. 16. The figure
the specimen. However, despite the presence of lubricant, the friction shows that the Poisson’s ratio increases (absolute value decreases) as
between the specimen and the two steel plates was not uniform, and the crushing strain increases, but it is always lower than 0. The min-
only some cells could slide along the steel plate surfaces. As a result, the imum Poisson’s ratio is −0.371, which is obtained near the beginning
waveform parts in column I and III were tilted to the left of the view. of crushing, which is very close to the Poisson’s ratio in the elastic
One typical moment of this phenomenon is demonstrated in Fig. 14 deformation stage calculated above.
when the crushing strain of the specimen is 0.25, i.e., crushed by 25%
of its original height. When the strain was 0.51, the bottom end of 6.2. Effect of the friction coefficient
column I and the top end of column III reached the maximum lateral
displacement, and the other ends of these two columns began to slide As can be seen from Fig. 14, the lateral contraction of 3D-RH
until the end of the crushing process. structure, i.e., NPR effect, is largely affected by boundary conditions,
The snap shots clearly show that the simulation vividly imitated especially the friction condition between the specimen and the two
the whole crushing process. It is noted that, in the simulation, the pressing platens (i.e., the steel plates in the test). In this section, the
friction coefficient between the specimen and the two platens was effect of the friction coefficient between the 3D-RH specimen and
slightly tuned. Since the bottom end of cell column I and the top end the platens on the crushing response of the specimen is discussed
of column III were found to moving prior to the other column ends, see numerically. Here, the friction coefficient solely refers to that between
Fig. 14, the friction coefficient between these two ends and the platens the 3D-RH specimen and the platens in the FE model, which is referred
was maintained at 0.1, while the friction coefficient between the other to as ‘‘𝜇f ’’ for brevity. Besides, to emphasize the efficacy of 𝜇f , the
column ends and the platens was increased to 0.4. And the simulation material failure is removed from the FE model.

7
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

Fig. 13. Final deformation of 3D-RH specimen under quasi-static crushing: (a) experiment; (b) simulation.

Fig. 14. 3D-RH specimen deformation at typical strains of quasi-static crushing test: (a) experiment; (b) simulation. (Note: dashed lines denote original width of the specimen.).

8
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

Fig. 15. Tested vs. simulated quasi-static crushing force–strain curve of 3D-RH specimen.

Fig. 16. Poisson’s ratio vs. strain of 3D-RH specimen under quasi-static crushing.

Fig. 17 shows snapshots of the 3D-RH deformation with different 𝜇f applied to the ends of the columns could not exceed the friction force.
settings during crushing. Apparently, 𝜇f can change the deformation Therefore, the cell columns are crushed in the ‘‘bended’’ position until
mode of 3D-RH specimen. When 𝜇f = 0.1, the column ends are easy the densification stage.
to slide along the platen surfaces, and the cell columns remain almost The change of deformation mode caused by boundary condition
straight during the whole crushing process. Name this deformation is also reflected in the crushing force history. Fig. 18 shows that the
mode as mode A. Deformation mode B appears when 𝜇f increases to crushing force history of 3D-RH is closely related to its deformation
0.4. In the early stage of crushing, strain = 0.25 for example, the friction mode with a certain friction coefficient. The friction coefficient corre-
hinders the movement of the column ends, and the cell columns bend sponding to deformation mode A (refer to Fig. 17) is between 0.0 and
due to the NPR effect. However, as the crushing progresses, the lateral 0.3. In this case, a long plateau stage can be observed in the crushing
force applied to the column ends can overcome the friction, and the force history, and the crushing force increases slightly with the increase
column ends begin to ‘‘shrink’’. Finally, near the end of crushing, strain of friction coefficient. Once 𝜇f is greater than 0.5, deformation mode
= 0.78 for instance, the cell columns return to the ‘‘straight’’ position C is triggered. Compared with mode A, after the strain of 0.45, the
again. When 𝜇f increases to 0.5, the deformation changes to mode C, crushing force of mode C increases rapidly due to the early internal
and the early stage deformation is very similar to that when 𝜇f = 0.4. contact. As a result, the energy absorbed by the 3D-RH specimen when
However, no matter how far the specimen is crushed, the lateral force 𝜇f ≥ 0.5 is much higher than that when 𝜇f ≤ 0.3. Besides, when the

9
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

Fig. 17. Deformation mode comparison of 3D-RHs with different settings of 𝜇f .

Fig. 18. Crushing force histories of 3D-RH with different friction coefficients.

friction coefficient exceeds 0.7, it no longer shows noticeable influence wall thickness and cell arrangement. As can be seen from Fig. 19a, the
on the crushing force history. The crushing force history corresponding units of 3D-RH and WWPAC have the same configuration, except that
to deformation mode B, with 𝜇f = 0.4, is between mode A and mode the thickness of the horizontal cell wall of WWPAC is 2t. In addition,
C. In the engineering application of such structures, when a long Fig. 19b shows that WWPAC has two more cell columns than 3D-RH
plateau stage for load bearing is required, the low friction boundary is for half of the structure, that is, four more cell columns for the whole
preferred. On the contrary, when using 3D-RH to absorb more energy, structure. However, the FE model cannot capture some defects caused
high friction boundaries or even full constraints are necessary. by WWPAC fabrication method, such as rivets left in the structure.
The energy absorption capabilities of 3D-RH and WWPAC are com-
6.3. Energy absorption comparison with WWPAC pared in this section base on numerical results. Fig. 20 shows the FE
models of the two structures with the same substrate material of 201
The WWPAC [34] proposed by Safikhani et al. is similar to the stainless steel, the properties of which are listed in Table 1. Due to
3D-RH in this study. However, the two structures are different in cell the difference of horizontal cell wall thickness, the mass of 3D-RH and

10
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

Table 2
Blast resistance performance indexes of 3D-RH and 2D-REH sandwich structures.
Core Permanent Maximum 𝐸i of core (kJ) SEA of core
structure deformation of deflection of lower (kJ/kg)
lower face-sheet face-sheet (mm)
(mm)
3D-RH 11.40 13.43 1.73 2.35
2D-REH 13.59 15.36 1.44 1.95

conventional honeycomb owing to its NPR effect. In order to high-

light the advantages of 3D-RH structural topology, the blast resistance
performance of 3D-RH and 2D-REH are compared through numerical
simulation.
The FE models for blast impact simulation are shown in Fig. 24.
Each model contains two face-sheets and a 3D-RH or 2D-REH as
deformable core, featuring a sandwich panel structure. All the face-
sheets are 4 mm thick with in-plane size of 250 mm × 250 mm. The
configurations of 3D-RH and 2D-REH are the same as the baseline
model shown in Fig. 2, except for the cell wall thickness of 2D-REH.
In order to ensure that the two structures have identical mass, the cell
wall thickness of 2D-REH was reduced to 0.24 mm. All parts are made
of 201 stainless steel with a failure strain of 0.5. Only the lower face-
sheets were fully clamped. The AUTOMATIC_SURFACE_TO_SURFACE
algorithm was defined to deal with the contact between the specimen
and the face-sheets. The friction coefficient was set to 0.3 [5]. 0.2 kg
Fig. 19. Comparison of 3D-RH and WWPAC in (a) unit cell and (b) structural scales. TNT was assumed to be detonated 100 mm above the upper face-
sheet center, and the blast load was generated by using the CONWEP
algorithm [6].
WWPAC are 0.74 kg and 1.70 kg, respectively. The crushing speed is set Final deformations and plastic strain contours of the 3D-RH and 2D-
at 1 m/s, and the lateral movement of the column ends are restrained, REH cores of the sandwich panels under blast load are shown in Fig. 25.
which is equivalent to infinite 𝜇f . Both structures have been severely compressed and the cell walls were
Fig. 21 shows the final deformed profiles of the two structures under packed together. However, no material failure occurred in both models.
quasi-static crushing, while Fig. 22 compares the crushing force–global NPR effect can be observed on both cores. In the current configuration,
strain relationships. The crushing force level of the WWPAC is about the NPR effect is more obvious on 3D-RH core.
twice that of the 3D-RH. This is mainly because the WWPAC has more The blast protection performance of sandwich cores (i.e., 3D-RH and
columns of cells to resist the crushing load. On the other hand, due 2D-REH in current case) can be quantitatively evaluated and compared
to the double thickness of the horizontal cell wall, the WWPAC reaches in terms of structural deformation and energy absorption. Fig. 26
the densification stage earlier than the 3D-RH. The densification strains compares the permanent deflection of the lower face-sheet with 3D-
of the WWPAC and 3D-RH specimens are 0.60 and 0.78, respectively. RH and 2D-REH cores. Obviously, with 3D-RH as the sandwich core,
When the densification stage is reached, the internal energy 𝐸i of the lower face-sheet obtained smaller permanent deflection. In addition
the WWPAC specimen is 2.52 kJ and that of the 3D-RH specimen is to permanent deflection, the maximum deflection of the lower face-
1.74 kJ. Fig. 23 shows the partition of 𝐸i between the horizontal and sheets during loading, which is a critical index of explosion-proof, is
inclined cell walls of the two specimens. Obviously, the inclined cell also listed in Table 2. The maximum deflection of 3D-RH lower face-
walls contribute the most to the total energy absorption, and the energy sheet is 13.43 mm and that of 2D-REH is 15.36 mm. Compared with
absorbed is about 4 times that of the horizontal cell walls in both the latter, 3D-RH core reduces the maximum deflection of the lower
structures. face-sheet by 14.6%.
When these two structures are used in protective devices such as The energy absorption indexes of the two types of cores are also
vehicle armor, their lightweight characteristics should also be consid- listed in Table 2. The internal energy 𝐸i of 3D-RH core is 1.73 kJ, which
ered. Specific energy absorption (SEA) is usually adopted to evaluate is 20.1% greater than that of 2D-REH core. Moreover, the SEA of 3D-
such performance, which is defined as follows: RH core is 2.35 kJ/kg, compared to that of 2D-REH core which is 1.95
𝐸𝑖 kJ/kg, an increase of 20.5%. These results prove that 3D-RH structure
𝑆𝐸𝐴 = could better resist blast load than 2D-REH in the current research case.
𝑚
where m is the structural mass. When the densification stage is reached,
the SEA of WWPAC is 1.48 kJ/kg, and that of 3D-RH is 2.36 kJ/kg. 7. Conclusions
This means that the energy absorption capacity per unit mass of 3D-
RH is 1.59 times that of WWPAC. The 3D-RH shows better lightweight In this study, a hierarchical fabrication method of metallic 3D-RH
performance than WWPAC. has been proposed. By this method, 3D-RH specimens with various
in-plane shapes can be easily fabricated by using traditional manufac-
6.4. Blast resistance comparison with 2D re-entrant expanded honeycomb turing techniques such as laser cutting, bending and welding. Stainless
steel 3D-RH specimens have been fabricated and tested under quasi-
One major application of auxetic honeycomb structure is to protect static crushing. The FE model has been established and verified by
lives and properties from extreme loads such as blast impact. The the results of quasi-static crushing test. Based on FE simulations, the
explosion resistance of 2D re-entrant expanded honeycomb (2D-REH) influence of boundary conditions, specifically the friction coefficient
has been studied by Qi et al. [5] through field blast test combined between 3D-RH specimen and loading surfaces, on 3D-RH’s deforma-
with FE simulation. 2D-REH showed better anti-blast performance than tion mode has been investigated. Moreover, the energy absorption

11
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

Fig. 20. FE model of (a) 3D-RH and (b) WWPAC with upper platen being hidden.

Fig. 21. Final deformations of (a) 3D-RH and (b) WWPAC under quasi-static crushing.

Fig. 22. Comparison of crushing force history of 3D-RH and WWPAC.

capacity and anti-blast performance of 3D-RH have been numerically • The main advantages of hierarchical fabrication method are: high
predicted and compared with WWPAC and 2D-REH, respectively. The efficiency and low cost; the lightweight property of 3D-RH can be
main findings can be summarized as follows: maintained; and, the in-plane shape of 3D-RH can be expanded.

12
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

Fig. 23. Energy partition between horizontal and inclined cell walls in 3D-RH and WWPAC.

Fig. 24. FE model setups of (a) 3D-RH and (b) 2D-REH sandwich panels under blast load.

Fig. 25. Final deformation and plastic strain of (a) 3D-RH and (b) 2D-REH cores under blast load.

Fig. 26. Permanent deflection of lower face-sheet under blast load with (a) 3D-RH and (b) 2D-REH as sandwich core.

13
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

• The minimum Poisson’s ratio of the investigated 3D-RH is −0.371, [10] C. Li, H.-S. Shen, H. Wang, Nonlinear vibration of sandwich beams with
obtained near the beginning of crushing. After that, Poisson’s functionally graded negative Poisson’s ratio honeycomb core, Int. J. Struct. Stab.
Dyn. 19 (03) (2019).
ratio increases as crushing proceeds, but always less than zero.
[11] C. Li, H.-S. Shen, H. Wang, Nonlinear dynamic response of sandwich beams with
• Under quasi-static crushing, 3D-RH specimen shows three typical functionally graded negative Poisson’s ratio honeycomb core, Eur. Phys. J. Plus
deformation modes corresponding to 𝜇f ≤ 0.3, 0.3 < 𝜇f < 0.5, and 134 (2) (2019).
𝜇f ≥ 0.5, where 𝜇f is the friction coefficient between the specimen [12] J. Qiao, C.Q. Chen, Analyses on the in-plane impact resistance of auxetic double
and loading surfaces. arrowhead honeycombs, J. Appl. Mech. 82 (5) (2015).
[13] J.X. Qiao, C.Q. Chen, Impact resistance of uniform and functionally graded
• Change of deformation mode affects 3D-RH’s crushing force his- auxetic double arrowhead honeycombs, Int. J. Impact Eng. 83 (2015) 47–58.
tory: with low friction boundary (𝜇f ≤ 0.3), the specimen main- [14] X. Zhao, Q. Gao, L. Wang, Q. Yu, Z.D. Ma, Dynamic crushing of double-arrowed
tains a long plateau stage; with high friction boundary (𝜇f ≥ 0.5), auxetic structure under impact loading, Mater. Des. 160 (2018) 527–537.
the specimen absorbs more energy. [15] G. Chen, Y. Cheng, P. Zhang, S. Cai, J. Liu, Blast resistance of metallic double
arrowhead honeycomb sandwich panels with different core configurations under
• 3D-RH shows better lightweight performance than WWPAC. Un-
the paper tube-guided air blast loading, Int. J. Mech. Sci. 201 (2021).
der the studied quasi-static crushing, WWPAC absorbs more en- [16] C. Qi, F. Jiang, C. Yu, S. Yang, In-plane crushing response of tetra-chiral
ergy while 3D-RH shows a higher SEA of 2.36 kJ/kg, 1.59 times honeycombs, Int. J. Impact Eng. 130 (2019) 247–265.
that of WWPAC. [17] X. Lu, V.B.C. Tan, T.E. Tay, Auxeticity of monoclinic tetrachiral honeycombs,
Compos. Struct. (2020).
• 3D-RH shows better anti-blast performance than 2D-REH. The
[18] A. Alderson, K.L. Alderson, D. Attard, K.E. Evans, R. Gatt, J.N. Grima, W. Miller,
maximum lower face-sheet deflection of 3D-RH is 14.6% smaller N. Ravirala, C.W. Smith, K. Zied, Elastic constants of 3-, 4- and 6-connected
and the SEA is 20.5% larger, which shows the advantages of chiral and anti-chiral honeycombs subject to uniaxial in-plane loading, Compos.
structural topology of 3D-RH. Sci. Technol. 70 (7) (2010) 1042–1048.
[19] L.L. Hu, Z.R. Luo, Q.Y. Yin, Negative Poisson’s ratio effect of re-entrant anti-
trichiral honeycombs under large deformation, Thin-Walled Struct. 141 (2019)
CRediT authorship contribution statement 283–292.
[20] C. Qi, F. Jiang, S. Yang, A. Remennikov, Multi-scale characterization of novel
Lian-Zheng Pei: Software, Methodology, Conceptualization, Writ- re-entrant circular auxetic honeycombs under quasi-static crushing, Thin-Walled
ing – original draft. Chang Qi: Methodology, Project administration, Struct. 169 (2021).
[21] C. Qi, F. Jiang, A. Remennikov, L.-Z. Pei, J. Liu, J.-S. Wang, X.-W. Liao, S. Yang,
Supervision, Writing – review & editing. Shu Yang: Resources, Funding
Quasi-static crushing behavior of novel re-entrant circular auxetic honeycombs,
acquisition. Yuan-Hang Ma: Investigation. Peng-Cheng Wu: Investiga- Composites B 197 (2020).
tion. [22] D. Li, J. Yin, L. Dong, R.S. Lakes, Strong re-entrant cellular structures with
negative Poisson’s ratio, J. Mater. Sci. 53 (5) (2017) 3493–3499.
[23] G. Imbalzano, P. Tran, T.D. Ngo, P.V.S. Lee, A numerical study of auxetic
Declaration of competing interest
composite panels under blast loadings, Compos. Struct. 135 (2016) 339–352.
[24] G. Imbalzano, P. Tran, T.D. Ngo, P.V. Lee, Three-dimensional modelling of
The authors declare that they have no known competing finan- auxetic sandwich panels for localised impact resistance, J. Sandw. Struct. Mater.
cial interests or personal relationships that could have appeared to 19 (3) (2017) 291–316.
influence the work reported in this paper. [25] Q. Gao, L. Wang, Z. Zhou, Z.D. Ma, C. Wang, Y. Wang, Theoretical, numerical
and experimental analysis of three-dimensional double-V honeycomb, Mater. Des.
139 (2018) 380–391.
Acknowledgments [26] C. Li, H.-S. Shen, H. Wang, Postbuckling behavior of sandwich plates with
functionally graded auxetic 3D lattice core, Compos. Struct. 237 (2020).
This work was supported by the National Key Research and Devel- [27] C. Li, H.-S. Shen, J. Yang, H. Wang, Low-velocity impact response of sandwich
plates with GRC face sheets and FG auxetic 3D lattice cores, Eng. Anal. Bound.
opment Program of China (no. 2021YFB3702004), Fundamental Re- Elem. 132 (2021) 335–344.
search Funds for the Central Universities (no. DUT20LAB132, [28] Y. Gao, X. Wei, X. Han, Z. Zhou, J. Xiong, Novel 3D auxetic lattice structures
DUT21GF203), and KGJ Basic Scientific Research Project (no. developed based on the rotating rigid mechanism, Int. J. Solids Struct. 233
ZX20210935). (2021).
[29] W. Lv, D. Li, L. Dong, Study on blast resistance of a composite sandwich panel
with isotropic foam core with negative Poisson’s ratio, Int. J. Mech. Sci. 191
References (2021).
[30] M. Janus-Michalska, D. Jasińska, J. Smardzewski, Comparison of contact stress
[1] J. Zhang, G. Lu, Z. You, Large deformation and energy absorption of additively distribution for foam seat and seat of auxetic spring skeleton, ijame 18 (1) (2013)
manufactured auxetic materials and structures: A review, Composites B 201 55–72.
(2020). [31] J. Smardzewski, Auxetic springs for seating, Turk. J. Agric. For. 37 (3) (2013)
[2] X. Ren, R. Das, P. Tran, T.D. Ngo, Y.M. Xie, Auxetic metamaterials and structures: 369–376.
a review, Smart Mater. Struct. 27 (2) (2018). [32] J. Smardzewski, D. Jasińska, M. Janus-Michalska, Structure and properties of
[3] G. Imbalzano, S. Linforth, T.D. Ngo, P.V.S. Lee, P. Tran, Blast resistance of composite seat with auxetic springs, Compos. Struct. 113 (2014) 354–361.
auxetic and honeycomb sandwich panels: Comparisons and parametric designs, [33] X.-T. Wang, X.-W. Li, L. Ma, Interlocking assembled 3D auxetic cellular
Compos. Struct. (2017). structures, Mater. Des. 99 (2016) 467–476.
[4] X. Jin, Z. Wang, J. Ning, G. Xiao, E. Liu, X. Shu, Dynamic response of sandwich [34] M. Safikhani Nasim, E. Etemadi, Three dimensional modeling of warp and woof
structures with graded auxetic honeycomb cores under blast loading, Composites periodic auxetic cellular structure, Int. J. Mech. Sci. 136 (2018) 475–481.
B 106 (2016) 206–217. [35] Q. Gao, C. Ge, W. Zhuang, L. Wang, Z. Ma, Crashworthiness analysis of double-
[5] C. Qi, A. Remennikov, L.-Z. Pei, S. Yang, Z.-H. Yu, T.D. Ngo, Impact and close-in arrowed auxetic structure under axial impact loading, Mater. Des. 161 (2019)
blast response of auxetic honeycomb-cored sandwich panels: Experimental tests 22–34.
and numerical simulations, Compos. Struct. 180 (2017) 161–178. [36] W. Lv, D. Li, L. Dong, Study on mechanical properties of a hierarchical octet-truss
[6] C. Qi, L.-Z. Pei, A. Remennikov, S. Yang, J. Liu, J.-S. Wang, X.-W. Liao, structure, Compos. Struct. 249 (2020).
Parametric study and optimization of the protect system containing a re-entrant [37] D. Li, R. Gao, L. Dong, W.-K. Lam, F. Zhang, A novel 3D re-entrant unit cell
hexagon cored sandwich panel under blast impact, Compos. Struct. 252 (2020). structure with negative Poisson’s ratio and tunable stiffness, Smart Mater. Struct.
[7] D. Kalubadanage, A. Remennikov, T. Ngo, C. Qi, Close-in blast resistance of large- 29 (4) (2020).
scale auxetic re-entrant honeycomb sandwich panels, J. Sandw. Struct. Mater. 23 [38] N. Novak, L. Starčevič, M. Vesenjak, Z. Ren, Blast response study of the sandwich
(8) (2021) 4016–4053. composite panels with 3D chiral auxetic core, Compos. Struct. 210 (2019)
[8] C. Li, H.-S. Shen, H. Wang, Nonlinear bending of sandwich beams with 167–178.
functionally graded negative Poisson’s ratio honeycomb core, Compos. Struct. [39] N. Novak, M. Vesenjak, S. Tanaka, K. Hokamoto, Z. Ren, Compressive behaviour
212 (2019) 317–325. of chiral auxetic cellular structures at different strain rates, Int. J. Impact Eng.
[9] C. Li, H.-S. Shen, H. Wang, Thermal post-buckling of sandwich beams with 141 (2020).
functionally graded negative Poisson’s ratio honeycomb core, Int. J. Mech. Sci. [40] N. Novak, L. Krstulović-Opara, Z. Ren, M. Vesenjak, Compression and shear
152 (2019) 289–297. behaviour of graded chiral auxetic structures, Mech. Mater. 148 (2020).

14
L.-Z. Pei, C. Qi, S. Yang et al. Thin-Walled Structures 176 (2022) 109369

[41] C. Li, H.-S. Shen, H. Wang, Z. Yu, Large amplitude vibration of sandwich plates [51] A. Arjunan, M. Singh, A. Baroutaji, C. Wang, Additively manufactured AlSi10Mg
with functionally graded auxetic 3D lattice core, Int. J. Mech. Sci. 174 (2020). inherently stable thin and thick-walled lattice with negative Poisson’s ratio,
[42] C. Li, H.-S. Shen, H. Wang, Full-scale finite element modeling and nonlinear Compos. Struct. 247 (2020).
bending analysis of sandwich plates with functionally graded auxetic 3D lattice [52] H. Lei, C. Li, J. Meng, H. Zhou, Y. Liu, X. Zhang, P. Wang, D. Fang, Evalu-
core, J. Sandw. Struct. Mater. 23 (7) (2021) 3113–3138. ation of compressive properties of SLM-fabricated multi-layer lattice structures
[43] C. Li, H.-S. Shen, H. Wang, Nonlinear dynamic response of sandwich plates with by experimental test and 𝜇-CT-based finite element analysis, Mater. Des. 169
functionally graded auxetic 3D lattice core, Nonlinear Dynam. 100 (4) (2020) (2019).
3235–3252. [53] L. Yang, Structural Design, Optimization and Application of 3D Re-Entrant
[44] J. Shen, K. Liu, Q. Zeng, J. Ge, Z. Dong, J. Liang, Design and mechanical Auxetic Structures, North Carolina State University, 2012.
property studies of 3D re-entrant lattice auxetic structure, Aerosp. Sci. Technol. [54] J. Schwerdtfeger, F. Schury, M. Stingl, F. Wein, R.F. Singer, C. Körner, Mechan-
118 (2021). ical characterisation of a periodic auxetic structure produced by SEBM, Phys.
[45] L. Yang, O. Harrysson, H. West, D. Cormier, Mechanical properties of 3D re- Status Solidi (B) 249 (7) (2012) 1347–1352.
entrant honeycomb auxetic structures realized via additive manufacturing, Int. [55] X.-T. Wang, B. Wang, X.-W. Li, L. Ma, Mechanical properties of 3D re-entrant
J. Solids Struct. 69-70 (2015) 475–490. auxetic cellular structures, Int. J. Mech. Sci. 131–132 (2017) 396–407.
[46] J. Schwerdtfeger, P. Heinl, R.F. Singer, C. Körner, Auxetic cellular structures [56] S.R. Reid, C. Peng, Dynamic uniaxial crushing of wood, Int. J. Impact Eng. 19
through selective electron-beam melting, Phys. Status Solidi (B) 247 (2) (2010) (5) (1997) 531–570.
269–272. [57] L. Jing, X. Su, F. Yang, H. Ma, L. Zhao, Compressive strain rate dependence
[47] I. Eldesouky, O. Harrysson, H. West, H. Elhofy, Electron beam melted scaffolds and constitutive modeling of closed-cell aluminum foams with various relative
for orthopedic applications, Addit. Manuf. 17 (2017) 169–175. densities, J. Mater. Sci. 53 (20) (2018) 14739–14757.
[48] J.A. Tamayo, M. Riascos, C.A. Vargas, L.M. Baena, Additive manufacturing of [58] G. Sun, H. Jiang, J. Fang, G. Li, Q. Li, Crashworthiness of vertex based
Ti6Al4V alloy via electron beam melting for the development of implants for hierarchical honeycombs in out-of-plane impact, Mater. Des. 110 (2016)
the biomedical industry, Heliyon 7 (5) (2021) e06892. 705–719.
[49] V.A. Lvov, F.S. Senatov, A.A. Stepashkin, A.A. Veveris, M.D. Pavlov, A.A. [59] S. Li, Z. Liu, V.P.W. Shim, Y. Guo, Z. Sun, X. Li, Z. Wang, In-plane compression
Komissarov, Low-cycle fatigue behavior of 3D-printed metallic auxetic structure, of 3D-printed self-similar hierarchical honeycombs – Static and dynamic analysis,
Mater. Today: Proc. 33 (2020) 1979–1983. Thin-Walled Struct. 157 (2020).
[50] P. Platek, J. Sienkiewicz, J. Janiszewski, F. Jiang, Investigations on mechanical [60] A.S. Hamada, L.P. Karjalainen, R.D.K. Misra, J. Talonen, Contribution of de-
properties of lattice structures with different values of relative density made from formation mechanisms to strength and ductility in two Cr–Mn grade austenitic
316L by selective laser melting (SLM), Materials (Basel) 13 (9) (2020). stainless steels, Mater. Sci. Eng. A 559 (2013) 336–344.

15