International Journal of Mechanical Sciences 157–158 (2019) 471–484

Contents lists available at ScienceDirect

International Journal of Mechanical Sciences

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

Energy absorption of thin-walled tubes enhanced by lattice structures

Erhan Cetin, Cengiz Baykasoğlu∗
Hitit University, Faculty of Engineering, Department of Mechanical Engineering, Cevre Yolu Avenue, 19030 Corum, Turkey

a r t i c l e i n f o a b s t r a c t

Keywords: Thin-walled structures filled or covered by various materials have been proposed for energy absorption applica-
Thin-walled tubes tions in recent years. At this point, additive manufacturing technologies provide an unprecedented opportunity
Lattice structures to produce nontraditional low-density filler materials to further improve the energy absorption performance of
Hybrid structures
thin-walled structures. With a similar motivation, novel hybrid structures, in which thin-walled tubes filled with
Crashworthiness
periodic lattice materials, are proposed and the energy absorption behaviors of these structures are investigated
Finite element method
under axial impact loading conditions. Two different types of lattice structures (i.e. body-centered cubic structure
and body-centered cubic structure with vertical strut) are considered as filler materials, and the effects of number
of lattice unit cell, diameter of lattice member and tube thickness on energy absorption characteristics of hybrid
structures are examined using validated nonlinear finite element models. The results show that the tube and
lattice structures contribute the buckling and bending resistance of each other during progressive deformation
of hybrid structures, and a considerable enhancement in energy absorption performance could be achieved with
appropriate selection of tube and filler lattice structure parameters. Particularly, the result revealed that the hy-
brid structures can absorb up to 115% higher impact energy compared with the sum of individual parts of hybrid
structures. Besides, the hybrid structures also show promising performance in terms of crashworthiness parame-
ters such as specific energy absorption and crash force efficiency, and thus these structures are recommended as
potential candidates for crashworthiness applications.

1. Introduction 19]. On the other hand, researchers have also been focusing on filling the
thin-walled tubes by low-density materials [20–23]. These studies have
Thin-walled tubular structures or in other words thin-walled tubes also revealed that the energy absorption performances of thin-walled
have been mostly used as crash energy absorbers in many applications tubes could be significantly improved by using the filler materials since
due to their superior energy absorption capabilities [1–3]. In recent these structures improve the progressive deformation patterns of struc-
decades, comprehensive studies have been carried out for enhancing ture with little increase in total mass. A comprehensive review of studies
the crushing performance of thin-walled tubes. At this point, different on energy absorption performance of filled or covered thin-walled tubes
materials (e.g., aluminum, steel and composites), geometric patterns can be found in Refs. [24,25].
(e.g., origami, windowed, corrugated and functionally graded), cross- Additive manufacturing technology also known as 3D printing, al-
sectional shapes (e.g., circular, square, triangular and hexagonal) and lows for the production of complex structures due to its own layer by
semi-apical angles (e.g., straight, frusta and tapered) are proposed to layer nature. Thanks to this feature, the additive manufacturing tech-
improve the crushing characteristics of thin-walled tubes. A brief lit- nique enables the researchers to create and develop novel complex
erature review on thin-walled structures and their applications can be shaped structures. Among them, low-density lattice structures [26] are
found in Refs. [4–8]. the promising candidate as filler materials for the impact energy ab-
The impact energy absorbers are expected to have the maximum sorber applications [27,28]. In this study, the two most common types
energy absorption capabilities throughout crushing distance with min- of lattice structures (i.e. body-centered cubic (BCC) and body-centered
imum mass. In this regard, low-density materials such as composites cubic with vertical strut (BCC-Z) structures) are proposed as filler ma-
[9–11], metal/polymer foams [12,13] and honeycombs [14,15] provide terials. At this point, the mechanical properties, deformation behav-
an unprecedented opportunity to further enhance the energy-absorbing ior and energy absorption capacity of BCC and BCC-Z lattice struc-
capability of the thin-walled tubes. At this point, the compressive re- tures are deeply investigated in Refs. [29–38]. The related studies are
sponse of thin-walled tubes could be considerably enhanced when these summarized as follows. Gümrük and Mines [29] presented a compre-
structures are covered or wrapped with the composite materials [16– hensive study on the compressive behavior of BCC and BCC-Z lattice

∗
Corresponding author.
E-mail addresses: [email protected] (E. Cetin), [email protected] (C. Baykasoğlu).

https://doi.org/10.1016/j.ijmecsci.2019.04.049
Received 24 December 2018; Received in revised form 17 March 2019; Accepted 23 April 2019
Available online 23 April 2019
0020-7403/© 2019 Elsevier Ltd. All rights reserved.
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

Fig. 1. Schematic view of the structures.

structures having different lattice unit cell sizes. McKown et al. [30] in- tubes in literature, to the best of our knowledge, there is no comprehen-
vestigated the mechanical behavior of lattice structures based on BCC sive work on the energy absorption of the thin-walled tubes enhanced by
and BCC-Z unit cell topologies under quasi-static and blast loadings. In periodic lattice structures. To this motivation, two different types of lat-
another study, Gümrük et al. [31] compared the crushing characteristics tice structures (i.e. BCC and BCC-Z) are proposed as filler materials and
of lattice structures having BCC, BCC-Z, and F2 BCC unit cell topologies the effects of diameter of lattice member, number of lattice unit cells and
under different loading scenarios. Merkt et al. [32] investigated the in- tube thickness on energy absorption characteristics of hybrid structures
fluence of strain rate on the mechanical behavior of lattice structures. are investigated by using validated nonlinear FE models. The simulation
Besides, Smith et al. [33] used experimental measurements and finite results showed that the energy absorption capacities of the thin-walled
element (FE) methods to investigate the quasi-static response of lattice tubes could be significantly improved by enhancing the proposed lat-
structures having BCC and BCC-Z unit cell topologies. On the other hand, tice structures. It is observed that the hybrid structures can absorb more
BCC and BCC-Z structures are also used as core materials of some sand- than two times impact energy compared with the sum of the individual
wich structures under impact loading. Mines et al. [34] carried out an parts, which make up the hybrid structure. Besides, the crashworthiness
experimental investigation focusing on the mechanical behavior of sand- parameters such as the specific energy absorption and crash force effi-
wich panels with BCC micro-lattice cores, and the impact energy val- ciency of the hybrid tubes can be improved with the proper selection of
ues are compared to that of conventional aluminum honeycomb. Other tube and lattice configurations.
similar studies, Shen et al. [35] and Turner et al. [36] carried out ex- The rest of this paper is organized as follows. The energy absorption
perimental investigations on low-velocity impact performance of the performance parameters, the components of hybrid structures, the de-
BCC and BCC-Z lattice structures core based sandwich panels. On the sign variables, the finite element model, its components and validations
other hand, the lattice structures can be formed by changing geomet- are described in Section 2. The numerical results are presented and dis-
ric pattern in order to increase their energy absorption capacities. To cussed in Section 3. Finally, the main findings of the present study are
this aim, Maskery et al. [37,38] compared the energy absorption behav- summarized in Section 4.
ior of the uniform and functionally graded BCC lattice structures under
compression loading. A brief literature review on energy absorption per- 2. Materials and methods
formances of BCC, BCC-Z and other lattice structural topologies can be
found in Refs. [39,40]. 2.1. Crashworthiness parameters
The above mentioned experimental and numerical investigations
demonstrated the potential of these lattice structures for several applica- The energy absorption characteristics of proposed structures are ob-
tions. Although there are several studies on filled or covered thin-walled tained from their force-displacement responses. Total energy absorption

Table 1
Proposed structures and design parameters.

Parameter Level 1 Level 2 Level 3 Level 4 Level 5

Structure BCC BCC-Z Empty Tube BCC hybrid BCC-Z hybrid

Number of lattice unit cells per structure 3 4 5 6 7
Diameter of lattice member 1 mm 2 mm 3 mm 4 mm 5 mm
Tube thickness
(For only hybrid structures with 7 cells) 0.5 mm 1 mm 1.5 mm – –

472
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

(EA), specific energy absorption (SEA), peak crush force (PCF), mean
crush force (MCF) and crush force efficiency (CFE) parameters are used
to investigate and compare the energy absorption performances of pro-
posed structures. The EA represents total absorbed energy during crush-
ing, and can be calculated as
𝛿
𝐸𝐴 = 𝐹 (𝛿)𝑑𝛿 (1)
∫0
where F(𝛿) is the instantaneous crushing force throughout the displace-
ment 𝛿. The PCF indicates the highest force in the force- displacement
curve. The MCF is defined as the ratio of energy to displacement and is
given as follows:
𝐸𝐴
𝑀𝐶𝐹 = (2)
𝛿
The SEA is the absorbed energy per unit mass, and is calculated as
𝐸𝐴
𝑆𝐸 𝐴 = (3)
𝑚 Fig. 2. 3D solid BCC lattice structure models with different lattice unit cell
Where m is the total mass of the structure. The higher SEA values repre- numbers a) three cells (L/a = 3), b) four cells (L/a = 4), c) five cells (L/a = 5),
sent a higher energy absorption performance of the structures [41,42]. d) six cells (L/a = 6) and e) seven cells (L/a = 7).
CFE is the ratio of mean crush force to the maximum crush force given
by
𝑀𝐶 𝐹
𝐶𝐹 𝐸 = (4)
𝑃 𝐶𝐹
The CE (crush efficiency) is described as the maximum crash length
per initial tube length (L), and is given as
𝛿
𝐶𝐸 = (5)
𝐿
It should be note that above mentioned crashworthiness parameters
of structures are calculated by considering the crush length immediately
before densification.

2.2. Description of geometric features

In this study, the thin-walled tubes are filled with the BCC and BCC-
Z types of lattice structures. At this point, five different lattice member
diameters and the lattice unit cell numbers are also considered in design. Fig. 3. 3D solid BCC-Z lattice structure models with different lattice unit cell
In addition, three different tube thicknesses are also taken into account numbers a) three cells (L/a = 3), b) four cells (L/a = 4), c) five cells (L/a = 5), d)
for hybrid structures having 7 lattice unit cells. Design configurations six cells (L/a = 6) and e) seven cells (L/a = 7).
and schematic view of the structures are given in Table 1 and Fig. 1,
respectively.
The length of all structures is chosen as design constant and taken
as 120 mm. The square thin-walled tubes are preferred in all designs
due to geometric compatibility with the proposed lattice structures. The
proposed lattice structure models are designed to fit with the inside of
square tubes. To this aim, the unit cells are established similar to the
Refs. [43–45], and then replicated depending on the number of cells
in the vertical direction. On the other hand, the square tubes are de-
signed by considering mid-surface finite element shell formulation in
order to prevent penetration between the lattice and tube structures.
Five different aspect ratios of structures are considered depending on
the number of unit lattice cells. At this point, the width of the lattice
structures with 3, 4, 5, 6 and 7 unit lattice cells are selected as 40, 30,
24, 20 and 17.14 mm, respectively. The BCC and BCC-Z lattice structure
models with different aspect ratios (i.e., L/a) are shown in Figs. 2 and 3,
where L and a are the initial length and width of the structure, respec- Fig. 4. The components of FE model for single lattice structures.
tively. In addition, the lattice structures are modeled using an idealized
structural geometry, in which the members are assumed to be perfectly
straight, and have a constant diameter. Fig. 4. The bottom rigid plate is fixed for translation and rotation in all
directions, while the top rigid plate having 250 kg impacting mass with
2.3. FE models and their validations 10 m/s initial velocity (𝜈 0 ) is permitted only to translate in a vertically
upward direction similar to the Refs. [46,47] All of the interactions be-
The FE simulations are carried out by using nonlinear explicit FE tween the rigid plates, the tubes and the lattice structures are defined
code, in Abaqus. In all analyzes, the structures are located between a by the general contact property. The contact behavior in the tangential
fixed rigid plate (bottom) and a moving rigid plate (top) as shown in direction is modeled by the penalty formulation in the Ref. [48]. Note

473
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

Fig. 5. Convergence test results for (a) lattice structures and

(b) empty tubes.

Fig. 6. The stress-strain curves of (a) AlSi10Mg and (b)

Al6063-T5.

that the penalty contact algorithm is very suitable for general contact gence test results for the lattice structures and empty tubes are shown in
modeling (i.e. contact between rigid bodies having multiple contacts Fig. 5.
per node). For all of the contacts, the friction coefficient is set as 0.25 The material of lattice structures is chosen as AlSi10Mg with Young’s
[46]. On the other hand, the contact behavior in the normal direction modulus of 69.3 GPa, yield strength of 160 MPa, density of 2670 kg/m3
is defined as ‘hard contact’. The moving rigid plate and the structures and Poisson’s ratio of 0.3 [51]. On the other hand, the material of
are connected through ‘tie’ constraints through a master-slave node re- the thin-walled tube is selected as Al6063-T5 with Young’s modulus of
lationship. This is to ensure that the crushing structures do not penetrate 68.2 GPa, yield strength of 187 MPa, density of 2700 kg/m3 and Pois-
the moving rigid plates. A reference point is assigned to the edge of the son’s ratio of 0.33 [52,53]. The true stress-strain curves of the AlSi10Mg
moving rigid plate to record the displacement, while another reference and Al6063-T5 are given in Fig. 6.
point is assigned to the edge of the fixed rigid plate to record the reaction The von Mises type yield criterion (J2-plasticity) in conjunction with
force. isotropic hardening is employed to model plasticity of the materials. The
The fixed and moving rigid plates are modeled using four-node lin- Cowper–Symonds overstress power law is used to capture strain-rate-
ear quadrilateral R3D4 elements to describe the 3D rigid bodies. On the sensitive behavior of Al6063-T5 in the simulation
other hand, the BCC and BCC-Z structures are modeled using four-node ( )1
linear tetrahedron C3D4 elements and the tube structures are modeled 𝑅 = 1 + 𝜀̇ 𝑝 ∕𝐷 𝑛 , 𝜎𝑑 ≥ 𝜎0 (6)
using four-node S4R shell elements with reduced integration. At least where 𝜀̇ 𝑝 is the von Mises equivalent plastic strain rate, R is the ratio of
three elements are used through the diameter of the BCC and BCC-Z lat- dynamic stress (𝜎 d ) at 𝜀̇ 𝑝 to the static flow stress (𝜎 0 ), D and n are ma-
tice structures to obtain reasonable results, which is similar in the Ref. terial constants which are selected to be 128,800 s−1 and 4 [46,52,54],
[49]. The lattice structures are also modeled using two-node linear B31 respectively. On the other hand, Johnson–Cook plasticity model is used
beam elements. The beam model is a simpler model which reduces the for AlSi10Mg to capture the rate dependency
computation time compared with the solid model. However, the beam
model has some disadvantages especially at the junction region due to 𝑅 = 1 + 𝐶𝑙𝑛(𝜀̇ 𝑝 ∕𝜀̇ 0 ), 𝜎𝑑 ≥ 𝜎0 (7)
a lack of stiffness in the areas. In this study, similar to the method used where, C and 𝜀̇ 0 are material parameter and the reference strain rate,
by Labeas et al. [50], the diameters of beam elements at the junction respectively. C and 𝜀̇ 0 values are chosen to be 0.02 and 0.001 s−1 , re-
areas are defined 40% more than that of the remainder of the beam el- spectively [51]. The energy absorption performance of proposed struc-
ements in order to simulate the higher material concentration in these tures may be significantly affected by material failure such as crack
regions. After the mesh convergence test, global element size of 2 mm formation and propagation. For this reason, damage models are em-
is found to be suitable for getting the results for the BCC and BCC-Z ployed along with the plasticity models, which is based on the value
lattice structures and this value is 1 mm for the tubes. The mesh conver- of von Mises equivalent plastic strain at element integration points. A

474
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

cumulative failure model is used in simulation in which the failure of Table 2

an element is controlled by damage parameter, w, which is defined as Process parameters used to build AlSi10Mg samples.
∑
𝜀0 + Δ𝜀𝑝 Laser power (W) Scan speed (mm/s) Hatch spacing (mm) Layer thickness (mm)
𝑤= (8)
𝜀𝑓 370 1300 0.19 0.03

where 𝜀0 is initial value of the von Mises equivalent plastic strain, Δ𝜀p
denotes an increment of the equivalent plastic strain and 𝜀f represents
the equivalent failure strain. The damage parameter is set to zero when
damage initiates and decreases the yield strength according to the rela-
tion (1-w) 𝜎 0 . Failure is assumed to occur when w exceeds unity. It is
assumed that the damage initiates when the equivalent plastic strain ex-
ceeds a critical value, 𝜀c , then evolves based on the following equation

𝐿𝜀̇ 𝑝 𝜎𝑦0
𝑤̇ = (9)
2𝐺𝑓

where L is the characteristic length of the element, 𝜎 y0 is the yield

strength value when the failure creation is reached and Gf denotes the
fracture energy per unit area. Gf is given as
𝜀𝑓
𝐺𝑓 = 𝐿 𝜎0 𝑑 𝜀 𝑝 (10)
∫𝜀𝑐

The stress-strain relations at different strain rates, and corresponding

Johnson-Cook and ductile damage model parameters of Al6063-T5 and
AlSi10Mg in Refs. [46,51–55] are considered in finite element models.
It should be noted that, the effect of strain ratio and temperature are not
considered in failure models.
The initial geometric imperfections have generally significant effects
on the crash behavior of tubes, especially under axial loadings. On the
other hand, it is very difficult to measure the distribution of imperfec-
tions in the real sample. At this point, eigenvalue imperfections are gen-
erally introduced in the geometry using eigenvalue buckling simulations
before crash simulations in order to take the geometric imperfections
into account [52,56–59]. The magnitudes for the first buckling mode
is adjusted to 0.02 of the thickness of the tubes using trial and error
method, and the buckling mode is defined in numerical models by us-
ing ‘imperfection’ option in Abaqus.
The BCC and BCC-Z lattice structures are manufactured from
AlSi10Mg with particle size of 30 μm using EOS M 290 direct metal Fig. 8. Comparison of the experimental and FE results (a) empty tube, (b) BCC
laser sintering (DMLS) based additive manufacturing process and then, hybrid structure, (c) BCC-Z hybrid structure.
assembled in the aluminum square tubes to construct hybrid structures
to use in FE validation tests (Fig. 7). The process parameters applied
to the samples are given in Table 2. The mechanical, physical, chem-
ical and thermal properties of AlSi10Mg parts produced by DMLS can Experimental crushing tests are performed under quasi-static axial
be found in Refs. [60,61]. The lattice samples have 4 unit cells, cross- loading and the test results are compared with the proposed FE model
section dimensions of 16.5 × 16.5 mm2 , a lattice diameter of 3 mm and results. Axial compression tests are performed using a Shimadzu Auto-
a length of 66 mm. On the other hand, the square tube samples have graph AG-IS universal testing machine having capacity of 100 kN. The
external dimensions of 20 × 20 mm2 , a thickness of 1.5 mm and a length structures are placed between the two flat platens and compressed at
of 66 mm. a rate of 2 mm/min. The load and displacement data are recorded at
a sampling rate of 20 points/sec. The experimental and FE results are
presented in Fig. 8. It is shown that the beam models of the BCC and
BCC-Z hybrid structures have relatively poor results, especially at the
densification regions. Since the self-contact mechanism cannot be de-
fined in the beam model, deviation occurs in the progressive deforma-
tions, therefore the densification region of the stress-strain curve is not
compatible with the experimental results. For these reasons, the beam
model formulation is not considered in this study. On the other hand, the
solid model of the hybrid structures predicts the elastic, elastic–plastic,
plateau, and densification regions considerably compatible with the ex-
perimental results. Similarly, FE results of the thin-walled tube are also
in a similar trend with the experimental results. In terms of the PCF and
EA values, the maximum differences between the experimental and FE
simulation results for empty tube, BCC and BCC-Z hybrid structures are
found lower than 5.2%, and 5.5%, respectively. Consequently, it is con-
Fig. 7. Al6063-T5 tubes and manufactured AlSi10Mg BCC and BCCZ lattice cluded that the FE models provide adequate confidence to examine the
structures. energy absorption performance of both the tubes and hybrid structures.

475
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

Fig. 9. The force-displacement and energy-displacement

graphs of single and hybrid BCC structures having different
cell numbers: (a) 3, (b) 4, (c) 5, (d) 6 and (e) 7 cells.

Fig. 10. The force-displacement and energy-displacement

graphs of single and hybrid BCC-Z structures having different
cell numbers: (a) 3, (b) 4, (c) 5, (d) 6 and (e) 7 cells.

476
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

Fig. 11. The SEA vs lattice member diameter of the single and
hybrid structures for different cell numbers (a) 3, (b) 4, (c) 5,
(d) 6 and (e) 7 cells.

3. Results and discussion Fig. 10 shows the energy absorption characteristics of empty tube,
single BCC-Z lattice structure, sum of single components and BCC-Z hy-
The force-displacement and energy-displacement curves of the brid tube for lattice member diameter of 3 mm and tube thickness of
empty tubes, single lattice structures, hybrid structures, sum of empty 0.5 mm. As clear from Figs. 9 and 10, BCC-Z structures have better en-
tube and single lattice structure are plotted in Figs. 9 and 10 for lat- ergy absorption capacities than that of the BCC structures due to the
tice member diameter of 3 mm and tube thickness of 0.5 mm. As clear presence of the vertical struts in the structure. Similarly, the BCC-Z hy-
from Fig. 9, the BCC structures generally have uniform force curves, and brid structures also have better energy absorption capacities than the
the curves generally do not contain peaks due to the absence of vertical BCC hybrid structures for all lattice unit cell numbers. For example,
struts in the structures. With this feature, the crushing characteristics the EA values of BCC-Z hybrid structures with 3, 4, 5, 6, and 7 lattice
of the BCC structures are similar to that of the foam structures [26]. unit cells for lattice member diameter of 3 mm are respectively 36.8%,
Since the BCC structures have uniform force curves, the force curves of 40.2%, 57.7%, 121.5%, and 24.5% higher than that of the BCC hybrid
sum of single lattice structures and empty tubes exhibit similar trend structures. Besides, Figs. 9 and 10 also demonstrated that the PCF of the
with that of empty tubes. When a unit cell is crushed, the load is trans- BCC and BCC-Z hybrid structures are considerably lower than the sum
ferred to the next unit cell, and the force value decreases during the force of that of the empty tube and single lattice structures, separately.
transfer between the cells. The decrease in the force value depends on Increment in the diameter of the lattice member of structure can
the number of unit cells in the structure. The force values in the struc- increase the energy absorption capacity of structure; however, this
ture having large BCC unit cells decrease sharply, while the force values increment also causes an increase in the mass of the structure (see
of the structure with small unit cells show relatively uniform crushing. Appendix A and B). For this reason, the SEA is also a quite critical pa-
Hence, as the number of lattice unit cells in the structure increases, more rameter like the EA. The SEA values of single and hybrid structures hav-
uniform crushing occurs for BCC hybrid structures. On the other hand, ing different lattice member diameters and lattice unit cell numbers for
from Fig. 9 it is seen that the BCC hybrid structures have considerably the tube thickness of 0.5 mm are shown in Fig. 11. It should be men-
better energy absorption capacities compared with the empty tubes and tioned that the CE of the hybrid structures are naturally lower than
single BCC structures. Moreover, significant differences are observed in that of simple BCC, BCC-Z and tube structures, and thus the EA and
the force and absorbed energy values between the BCC hybrid structures corresponding SEA values presented in Fig. 11 are calculated by con-
and the sum of their individual components since both progressive de- sidering the crushing distance immediately before densification starts.
formation behavior and global buckling resistance of single components As seen in Fig. 11, as the aspect ratio of the empty tubes increase, the
enhance with the use of hybrid structure. SEA values of the tubes increases since the tubes have similar energy

477
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

Fig. 12. The MCF vs lattice member diameter of the single

and hybrid structures for different cell numbers (a) 3, (b) 4,
(c) 5, (d) 6 and (e) 7 cells.

absorption values although the mass of the tubes decreases. The figure of the tube could significantly affect the energy absorption performance
shows that the SEA values of both the single BCC and BCC-Z structures of the hybrid structures and this issue is discussed in the end of this
generally increase with the increasing lattice member diameter, and the section.
single BCC-Z structures have higher SEA values than the BCC ones, as The MCF values of the single and hybrid structures with different
expected. For example, the SEA values of BCC-Z structures having 3, lattice member diameter and lattice unit cell numbers for the tube thick-
4, 5, 6, and 7 lattice unit cells are respectively 67.6%, 66.9%, 69.3%, ness of 0.5 mm are illustrated in Fig. 12. The figure demonstrates that
105.5%, and 244.9% higher than that of the BCC structures for lattice the increment of lattice member diameter significantly enhance the MCF
member diameter of 2 mm. As shown in Fig 11, the energy absorption values of both the single and hybrid structures except for certain con-
capacities of empty tubes are significantly higher than that of the single figurations. For instance, the BCC and BCC-Z hybrid structures having
lattice structures for the diameter range of 1–3 mm, thus the crushing be- 4 unit cells with diameter of 4 mm have 98.5% and 96.4% higher MCF
havior of hybrid structures are mainly controlled by the tubes, and the values than the corresponding BCC and BCC-Z hybrid structures having
lattice structures slightly affect the energy absorption performance of diameter of 3 mm, respectively. On the other hand, the global buckling
proposed hybrid structures in this range. On the other hand, the lattice adversely affects the energy absorption performances of hybrid struc-
structures have significant effects on the crushing behavior of hybrid tures especially for larger aspect ratio values, and this phenomenon pre-
structures for higher lattice member diameter values, and a consider- vents the further enhancement of energy absorption capacity of the hy-
able enhancement is observed in the SEA values of hybrid structures brid structures. Similar to the SEA values, the MCF values of the empty
as shown in Fig 11. As clear from Fig. 11, the SEA values of hybrid tubes are considerably higher than that of single lattice structures for
tubes generally increase with the increasing of lattice unit cell num- the member diameter range of 1–3 mm, therefore the crushing behav-
bers in the hybrid structures. However, the desired energy absorption ior of the hybrid structures are mainly controlled by the tubes in this
performance could not be achieved especially for the hybrid structures range. On the other hand, the lattice structures have dominant effect
having higher lattice unit cell numbers and larger member diameters on the MCF values of the hybrid structures for larger lattice member
since these structures tend to buckle globally after a certain progressive diameters. Note that, as the member diameter and aspect ratio of lattice
deformation (i.e., see Fig 14). Particularly, a limited enhancement in structures increase, the tendency of these structures to global buckling
SEA values is observed for the hybrid structures having 7 unit cells due also increases, and this leads to the buckling of the hybrid structure,
to their global buckling tendency. Hence, it’s clear that the appropriate globally (e.g. see Fig. 14).
selection of the structural parameters is quite critical to optimize the en- The CFE values of the empty tubes and hybrid structures with dif-
ergy absorption efficiency of hybrid structures. Note that the thickness ferent lattice member diameters and lattice unit cell numbers for the

478
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

Fig. 13. The CFE vs lattice member diameter of the empty

tube and hybrid lattice structures for different cell numbers
(a) 3, (b) 4, (c) 5, (d) 6 and (e) 7 cells.

tube thickness of 0.5 mm are demonstrated in Fig. 13. The CFE values ratio increases, the tendency to global buckling of the hybrid structures
of the empty tubes increase with the increasing aspect ratios, since the increases especially for thicker lattice member diameter.
PCF values of the empty tubes decrease while the tubes have similar The tube thicknesses of hybrid structures are also studied in order
MCF values in this range. When a tube is filled with any material, it is to investigate the effect of the tube thickness on the energy absorption
expected that the PCF value of the tube increases as well, and this in- performance of the hybrid structures. At this point, three different tube
crease may cause reduction in the CFE value. However, a considerable thicknesses (i.e. 0.5, 1 and 1.5 mm) are considered for the BCC and BCC-
improvement is observed in the CFE values of hybrid structures in most Z hybrid structures having 7 lattice unit cells. Fig. 15 shows the SEA
of the cases due to the high MCF values of the hybrid structures. On the values of the hybrid structures having different lattice member diame-
other hand, the hybrid structures with high aspect ratio have generally ters and different tube thicknesses. The figure demonstrates that the SEA
better CFE values due to their relatively low PCF and high MCF. Note values of both the BCC and BCC-Z hybrid structures increase with the
that the CE value of the hybrid structures decreases with the increase the increasing tube thickness for all cases. In addition, it is clearly observed
lattice member diameter and aspect ratio of the structures, and the CFE that the increasing tube thickness contributes to the global buckling re-
values are calculated by considering the effective crush distance of the sistance of the hybrid structures, and this allows the structure to absorb
structures. In addition, after progressive buckling deformation modes, more impact energy. In particular, the hybrid structures with thinner
the global buckling deformations occur in some cases (e.g., Fig 14), and tube thicknesses have relatively lower energy absorption capacities due
these deformations significantly reduce the energy absorption capaci- to global buckling, and the energy absorption performance can be im-
ties, thereby MCF and CFE values of hybrid structures. proved by increasing the tube thickness of the structures. For instance,
The deformation characteristics of the hybrid structures having dif- the 7-cell BCC-Z hybrid structures having tube thickness of 1.5 mm and
ferent lattice member diameters and cell numbers are demonstrated in lattice member diameter of 1 mm are absorbed 8.2 times more energy
Fig. 14 for the tube thickness of 0.5 mm. It is observed that the hybrid than that of hybrid structures having tube thickness of 0.5 mm. More-
structures having 3 and 4 lattice unit cells generally show fully pro- over, the hybrid structures have significantly better energy absorption
gressive buckling behavior, while the hybrid structures having 5, 6 and capacities than the sum of the individual components separately for
7 lattice unit cells show both progressive and global buckling modes. higher lattice member diameter values. For instance, the EA values of
In addition, 7-cell BCC hybrid structures having lattice member diam- the 7-cell BCC and BBC-Z hybrid structures having lattice member di-
eters of 1 and 2 mm exhibit only progressive characteristics whereas ameter of 5 mm are respectively 25.9%, and 10.5% higher than sum of
progressive and global buckling modes are observed in 7-cell BCC-Z hy- EA values of individual parts of hybrid structures having 1.5 mm tube
brid structures. On the other hand, as clear from Fig. 14, as the aspect thickness.

479
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

Fig. 14. The deformation characteristics of the

hybrid structures having different lattice mem-
ber diameters and number of cell numbers.

Fig. 15. The SEA values for different tube

thicknesses.

480
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

Consequently, it is seen that the tube thickness is also found to be structures significantly affect the energy absorption performance of the
a quite important parameter on the energy absorption behavior of the structures. Namely, the hybrid structures with low aspect ratios gen-
hybrid structures, and the energy absorption performance of the thin- erally show fully progressive buckling deformations, whereas the struc-
walled tubes could be further enhanced with the appropriate selection tures with high aspect ratios generally show both progressive and global
of the tube thickness of the hybrid structures. buckling modes. At this point, the global buckling of the hybrid structure
can be prevented by selecting the optimum tube and lattice structure
4. Conclusion parameters, and thereby the energy absorption capacities of the hybrid
structures can be enhanced, significantly. The crushing performance of
In this study, the enhancement of the energy absorption perfor- the hybrid structures are also investigated in terms of crashworthiness
mances of thin-walled tubes by the lattice structures are examined under parameters such as the specific energy absorption and the crush force
axial impact loading. Two different lattice structure types are consid- efficiency, and a considerable enhancement is also observed in most of
ered as filler materials, and effects of unit cell numbers, lattice member the designs. The thin-walled structures filled with lattice structures are
diameters, and tube thicknesses are taken into account as design pa- thus recommended as potential candidates in various crashworthiness
rameters. The crashworthiness parameters and deformation modes are applications. In future research, functionally graded lattice filled thin-
also examined to compare the energy absorption performance of the walled tubular structures will be investigated to optimize the energy
structures. The results showed that the tube and lattice structures con- absorption performance of the hybrid structures. In addition, further
tribute to the buckling and bending behaviors of each other during the studies will focus on multiobjective optimization considering the effects
progressive deformation of the hybrid structures. It is observed that the of various design parameters.
energy absorption performance of thin-walled tubes could be consid-
erably improved by filling them with the proposed lattice structures, Acknowledgment
and the hybrid structures can absorb more than two times impact en-
ergy compared with the sum of the individual parts, which make up This work was supported by the Scientific Research Projects Govern-
the hybrid structure. The results revealed that the aspect ratios of the ing Unit of Hitit University, project No: MUH19004.18.001

Appendix A. The crashworthiness parameters of BCC, BBC-Z and empty tube structures

Structure type Aspect ratio Diameter of lattice member (mm) Tube thickness (mm) EA (mJ) Mass (gr) SEA (kJ/kg) MCF (kN) PCF (kN) CFE

BCC 3 1 – 3251.32 1.72 1.89 0.03 0.30 0.10

BCC 3 2 – 23,286.1 6.73 3.46 0.24 1.46 0.16
BCC 3 3 – 83,055.8 14.80 5.61 0.89 4.04 0.22
BCC 3 4 – 254,906 25.70 9.92 2.74 16.23 0.17
BCC 3 5 – 428,641 39.10 10.96 4.98 16.76 0.30
BCC 4 1 – 3864.23 1.71 2.26 0.04 0.25 0.14
BCC 4 2 – 32,122.1 6.63 4.84 0.33 1.64 0.20
BCC 4 3 – 127,297 14.40 8.84 1.38 6.10 0.23
BCC 4 4 – 254,141 24.80 10.25 3.10 6.90 0.45
BCC 4 5 – 607,072 37.50 16.19 7.59 31.27 0.24
BCC 5 1 – 4564.88 1.70 2.69 0.04 0.31 0.14
BCC 5 2 – 41,438.2 6.52 6.36 0.44 1.61 0.27
BCC 5 3 – 146,929 14.10 10.42 1.55 3.07 0.50
BCC 5 4 – 262,244 24 10.93 3.75 8.49 0.44
BCC 5 5 – 436,162 35.80 12.18 7.27 9.19 0.79
BCC 6 1 – 1801.61 1.68 1.07 0.03 0.10 0.31
BCC 6 2 – 17,490.1 6.42 2.72 0.29 0.45 0.65
BCC 6 3 – 60,182.9 13.70 4.39 1.20 1.73 0.70
BCC 6 4 – 157,809 23.10 6.83 3.16 4.41 0.72
BCC 6 5 – 408,509 34.20 11.94 8.17 10.46 0.78
BCC 7 1 – 1859.33 1.67 1.11 0.03 0.09 0.34
BCC 7 2 – 17,304.4 6.31 2.74 0.35 0.55 0.63
BCC 7 3 – 72,921.8 13.40 5.44 1.46 2.22 0.66
BCC 7 4 – 175,562 22.30 7.87 3.51 5.35 0.66
BCC 7 5 – 364,346 32.50 11.21 9.11 13.75 0.66
BCC-Z 3 1 – 4777.69 1.97 2.43 0.04 0.30 0.15
BCC-Z 3 2 – 44,459.3 7.67 5.80 0.43 1.47 0.29
BCC-Z 3 3 – 169,498 16.80 10.09 1.69 5.05 0.34
BCC-Z 3 4 – 334,063 29.10 11.48 3.75 8.40 0.45
BCC-Z 3 5 – 769,772 44.20 17.42 8.75 17.99 0.49
BCC-Z 4 1 – 6766.57 1.95 3.47 0.06 0.34 0.18
BCC-Z 4 2 – 60,914.3 7.54 8.08 0.60 1.96 0.31
BCC-Z 4 3 – 225,318 16.40 13.74 2.45 5.67 0.43
BCC-Z 4 4 – 436,644 28 15.59 4.85 10.50 0.46
BCC-Z 4 5 – 1,050,960 42.10 24.96 12.22 18.46 0.66
BCC-Z 5 1 – 7431.78 1.93 3.85 0.07 0.24 0.30
BCC-Z 5 2 – 79,795 7.41 10.77 0.81 2.16 0.38
BCC-Z 5 3 – 271,639 16.90 17.08 2.89 5.89 0.49
BCC-Z 5 4 – 480,747 27 17.81 6.87 8.19 0.84
BCC-Z 5 5 – 814,612 40.10 20.31 14.81 18.38 0.81
BCC-Z 6 1 – 9135.67 1.92 4.76 0.09 0.32 0.29
BCC-Z 6 2 – 40,673.2 7.27 5.59 0.68 1.32 0.51
BCC-Z 6 3 – 103,519.7 15.50 6.68 1.88 2.18 0.87
BCC-Z 6 4 – 414,516 25.90 16.00 9.21 10.54 0.87
(continued on next page)

481
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

(continued)

Structure type Aspect ratio Diameter of lattice member (mm) Tube thickness (mm) EA (mJ) Mass (gr) SEA (kJ/kg) MCF (kN) PCF (kN) CFE

BCC-Z 6 5 – 597,231 38 15.72 13.27 20.58 0.64

BCC-Z 7 1 – 13,107 1.90 6.90 0.12 0.77 0.15
BCC-Z 7 2 – 67,476.1 7.14 9.45 1.04 1.61 0.64
BCC-Z 7 3 – 196,786 15.00 13.12 3.28 5.05 0.65
BCC-Z 7 4 – 451,453 24.80 18.20 8.21 11.29 0.73
BCC-Z 7 5 – 626,953 35.90 17.46 12.54 22.12 0.57
Empty Tube 3 – 0.5 261,343 26.20 9.97 2.69 17.78 0.15
Empty Tube 4 – 0.5 243,307 19.80 12.29 2.48 13.27 0.19
Empty Tube 5 – 0.5 235,789 15.90 14.83 2.43 11.29 0.22
Empty Tube 6 – 0.5 208,788 13.30 15.70 2.20 8.96 0.25
Empty Tube 7 – 0.5 201,858 11.40 17.71 2.12 7.71 0.28
Empty Tube 7 – 1 732,856 23.50 31.19 7.63 19.49 0.39
Empty Tube 7 – 1.5 1,548,860 36.20 42.79 16.13 28.26 0.57

Appendix B. The crashworthiness parameters of BCC and BBC-Z hybrid structures

Structure Type Aspect ratio Diameter of lattice member (mm) Tube thickness (mm) EA (mJ) Mass (gr) SEA (kJ/kg) MCF (kN) PCF (kN) CFE

BCC Hybrid 3 1 0.5 306374 27.92 10.97 2.92 16.29 0.18

BCC Hybrid 3 2 0.5 332227 32.93 10.09 3.57 17.65 0.20
BCC Hybrid 3 3 0.5 440008 41.00 10.73 5.09 17.37 0.29
BCC Hybrid 3 4 0.5 566120 51.90 10.91 6.54 17.54 0.37
BCC Hybrid 3 5 0.5 759705 65.30 11.63 9.04 27.04 0.33
BCC Hybrid 4 1 0.5 218315 21.51 10.15 2.25 13.01 0.17
BCC Hybrid 4 2 0.5 356228 26.43 13.48 3.75 12.91 0.29
BCC Hybrid 4 3 0.5 539551 34.20 15.78 5.83 31.46 0.19
BCC Hybrid 4 4 0.5 1040930 44.60 23.34 11.57 73.95 0.16
BCC Hybrid 4 5 0.5 1706180 57.30 29.78 20.07 98.42 0.20
BCC Hybrid 5 1 0.5 217973 17.60 12.38 2.25 10.14 0.22
BCC Hybrid 5 2 0.5 300629 22.42 13.41 3.34 10.18 0.33
BCC Hybrid 5 3 0.5 466088 30.00 15.54 5.48 20.99 0.26
BCC Hybrid 5 4 0.5 899397 39.90 22.54 10.58 32.86 0.32
BCC Hybrid 5 5 0.5 1360330 51.70 26.31 18.14 64.09 0.28
BCC Hybrid 6 1 0.5 200007 28.11 7.12 2.11 7.22 0.29
BCC Hybrid 6 2 0.5 275210 19.72 13.96 3.28 9.05 0.36
BCC Hybrid 6 3 0.5 302318 27.00 11.20 4.65 8.73 0.53
BCC Hybrid 6 4 0.5 483849 36.40 13.29 8.06 10.20 0.79
BCC Hybrid 6 5 0.5 812026 47.50 17.10 13.53 17.05 0.79
BCC Hybrid 7 1 0.5 181136 13.07 13.86 1.97 7.86 0.25
BCC Hybrid 7 2 0.5 261405 17.71 14.76 3.27 7.11 0.46
BCC Hybrid 7 3 0.5 305658 24.80 12.32 4.37 7.72 0.57
BCC Hybrid 7 4 0.5 526486 33.70 15.62 7.52 10.21 0.74
BCC Hybrid 7 5 0.5 808969 43.90 18.43 13.48 18.43 0.73
BCC Hybrid 7 1 1 599354 25.17 23.81 6.51 14.24 0.46
BCC Hybrid 7 2 1 699631 29.81 23.47 8.23 11.69 0.70
BCC Hybrid 7 3 1 1046150 36.90 28.35 13.08 17.72 0.74
BCC Hybrid 7 4 1 1155160 45.80 25.22 17.24 24.48 0.70
BCC Hybrid 7 5 1 1217350 56.00 21.74 20.29 25.77 0.79
BCC Hybrid 7 1 1.5 1236140 37.87 32.64 13.44 33.34 0.40
BCC Hybrid 7 2 1.5 1322330 42.51 31.11 15.03 26.51 0.57
BCC Hybrid 7 3 1.5 1619060 49.60 32.64 19.05 28.39 0.67
BCC Hybrid 7 4 1.5 2004840 58.50 34.35 28.71 31.49 0.91
BCC Hybrid 7 5 1.5 2408910 68.70 35.06 34.4100 40.34 0.85
BCC-Z Hybrid 3 1 0.5 346615 28.17 12.30 3.18 19.53 0.16
BCC-Z Hybrid 3 2 0.5 423557 33.87 12.51 4.01 23.86 0.17
BCC-Z Hybrid 3 3 0.5 601921 43.00 14.00 5.90 25.89 0.23
BCC-Z Hybrid 3 4 0.5 1156870 55.30 20.92 11.45 74.77 0.15
BCC-Z Hybrid 3 5 0.5 1824980 70.40 25.92 18.07 119.26 0.15
BCC-Z Hybrid 4 1 0.5 192211 21.75 8.84 1.92 9.37 0.21
BCC-Z Hybrid 4 2 0.5 355276 27.34 12.99 3.59 12.11 0.30
BCC-Z Hybrid 4 3 0.5 756545 36.20 20.90 7.57 49.10 0.15
BCC-Z Hybrid 4 4 0.5 1413010 47.80 29.56 14.87 77.46 0.19
BCC-Z Hybrid 4 5 0.5 2060190 61.90 33.28 22.39 82.75 0.27
BCC-Z Hybrid 5 1 0.5 241500 17.83 13.54 2.28 12.41 0.18
BCC-Z Hybrid 5 2 0.5 317451 23.31 13.62 3.34 9.48 0.35
BCC-Z Hybrid 5 3 0.5 735266 31.80 23.12 7.82 34.03 0.23
BCC-Z Hybrid 5 4 0.5 860069 42.90 20.05 9.15 30.75 0.30
(continued on next page)

482
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

(continued)

Structure Type Aspect ratio Diameter of lattice member (mm) Tube thickness (mm) EA (mJ) Mass (gr) SEA (kJ/kg) MCF (kN) PCF (kN) CFE

BCC-Z Hybrid 5 5 0.5 1260310 56.00 22.51 19.39 51.10 0.38

BCC-Z Hybrid 6 1 0.5 270220 15.22 17.75 2.53 13.63 0.19
BCC-Z Hybrid 6 2 0.5 411455 20.57 20.00 3.96 18.41 0.21
BCC-Z Hybrid 6 3 0.5 669639 28.80 23.25 8.59 18.76 0.46
BCC-Z Hybrid 6 4 0.5 960824 39.20 24.51 12.81 14.15 0.91
BCC-Z Hybrid 6 5 0.5 1399960 51.30 27.29 22.22 25.65 0.87
BCC-Z Hybrid 7 1 0.5 126961 13.30 9.55 1.59 5.16 0.31
BCC-Z Hybrid 7 2 0.5 274417 18.54 14.80 3.43 7.12 0.48
BCC-Z Hybrid 7 3 0.5 389679 26.40 14.76 5.57 8.75 0.64
BCC-Z Hybrid 7 4 0.5 654862 36.20 18.09 10.91 15.86 0.69
BCC-Z Hybrid 7 5 0.5 1149330 47.30 24.30 19.16 27.28 0.70
BCC-Z Hybrid 7 1 1 554446 25.40 21.83 5.66 19.44 0.29
BCC-Z Hybrid 7 2 1 675963 30.64 22.06 7.86 14.23 0.55
BCC-Z Hybrid 7 3 1 835810 38.50 21.71 11.14 17.10 0.65
BCC-Z Hybrid 7 4 1 1056113 48.30 21.87 15.09 21.17 0.71
BCC-Z Hybrid 7 5 1 1622780 59.40 27.32 24.97 34.59 0.72
BCC-Z Hybrid 7 1 1.5 1046800 38.10 27.48 11.02 31.43 0.35
BCC-Z Hybrid 7 2 1.5 1259010 43.34 29.05 13.99 28.35 0.49
BCC-Z Hybrid 7 3 1.5 1667970 51.20 32.58 20.85 31.32 0.67
BCC-Z Hybrid 7 4 1.5 2045310 61.00 33.53 27.27 34.25 0.80
BCC-Z Hybrid 7 5 1.5 2405130 72.10 33.36 37.00 45.11 0.82

References [20] Sun G, Li S, Liu Q, Li G, Li Q. Experimental study on crashworthi-

ness of empty/aluminum foam/honeycomb-filled CFRP tubes. Compos Struct
2016;152:969–93. doi:10.1016/j.compstruct.2016.06.019.
[1] Marzbanrad J, Ebrahimi MR. Multi-objective optimization of aluminum hollow tubes
[21] Balaji G, Annamalai K. Crushing response of square aluminium column filled with
for vehicle crash energy absorption using a genetic algorithm and neural networks.
carbon fibre tubes and aluminium honeycomb. Thin-Walled Struct 2018;132:667–
Thin-Walled Struct 2011;49(12):1605–15. doi:10.1016/j.tws.2011.08.009.
81. doi:10.1016/j.tws.2018.07.037.
[2] Meran A.P., Baykasoglu C., Mugan A. Development of a design for a crash en-
[22] Li Z, Chen R, Lu F. Comparative analysis of crashworthiness of empty
ergy management system for use in a railway passenger car 2016;230:206–19.
and foam-filled thin-walled tubes. Thin-Walled Struct 2018;124:343–9.
doi:10.1177/0954409714533321.
doi:10.1016/j.tws.2017.12.017.
[3] Marsolek J, Reimerdes HG. Energy absorption of metallic cylindrical shells with
[23] Hussein RD, Ruan D, Lu G, Guillow S, Yoon JW. Crushing response of square alu-
induced non-axisymmetric folding patterns. Int J Impact Eng 2004;30(8–9):1209–
minium tubes filled with polyurethane foam and aluminium honeycomb. Thin-
23. doi:10.1016/j.ijimpeng.2004.06.006.
Walled Struct 2017;110:140–54. doi:10.1016/j.tws.2016.10.023.
[4] Alghamdi AAA. Collapsible impact energy absorbers: an overview. Thin-Walled
[24] Supian ABM, Sapuan SM, Zuhri MYM, Zainudin ES, Ya HH. Hybrid reinforced ther-
Struct 2001;39(2):189–213. doi:10.1016/S0263-8231(00)00048-3.
moset polymer composite in energy absorption tube application: a review. Def Tech-
[5] Abramowicz W. Thin-walled structures as impact energy absorbers. Thin-Walled
nol 2018;14(4):291–305. doi:10.1016/j.dt.2018.04.004.
Struct 2003;41(2–3):91–107. doi:10.1016/S0263-8231(02)00082-4.
[25] Baroutaji A, Sajjia M, Olabi AG. On the crashworthiness performance of thin-walled
[6] Olabi AG, Morris E, Hashmi MSJ. Metallic tube type energy absorbers: a synopsis.
energy absorbers: recent advances and future developments. Thin-Walled Struct
Thin-Walled Struct 2007;45(7–8):706–26. doi:10.1016/j.tws.2007.05.003.
2017;118:137–63. doi:10.1016/j.tws.2017.05.018.
[7] Yuen SCK, Nurick GN. The energy-absorbing characteristics of tubular struc-
[26] Gibson LJ, Ashby MF. Cellular Solids: structure and properties. Cambridge: Cam-
tures with geometric and material modifications: an overview. Appl Mech Rev
bridge University Press; 1997.
2008;61(2):020802. doi:10.1115/1.2885138.
[27] Banhart J. Manufacture, characterisation and application of cellular metals and
[8] Mat F, Ismail KA, Yaacob S, Inayatullah O. Impact response of thin-
metal foams. Prog Mater Sci 2001;46(6):559–632. doi:10.1016/S0079-6425(00)
walled tubes: a prospective review. Appl Mech Mater 2012;165:130–4.
00002-5.
doi:10.4028/www.scientific.net/AMM.165.130.
[28] Sun Y, Li QM. Dynamic compressive behaviour of cellular materials: a review
[9] Zhu G, Sun G, Yu H, Li S, Li Q. Energy absorption of metal, composite and
of phenomenon, mechanism and modelling. Int J Impact Eng 2018;112:74–115.
metal/composite hybrid structures under oblique crushing loading. Int J Mech Sci
doi:10.1016/j.ijimpeng.2017.10.006.
2018;135:458–83. doi:10.1016/j.ijmecsci.2017.11.017.
[29] Gümrük R, Mines RAW. Compressive behaviour of stainless steel micro-lattice struc-
[10] Zhu G, Sun G, Liu Q, Li G, Li Q. On crushing characteristics of different con-
tures. Int J Mech Sci 2013;68:125–39. doi:10.1016/j.ijmecsci.2013.01.006.
figurations of metal-composites hybrid tubes. Compos Struct 2017;175:58–69.
[30] McKown S, Shen Y, Brookes WK, Sutcliffe CJ, Cantwell WJ, Langdon GS, et al.
doi:10.1016/j.compstruct.2017.04.072.
The quasi-static and blast loading response of lattice structures. Int J Impact Eng
[11] Babbage JM, Mallick PK. Static axial crush performance of unfilled and foam-
2008;35(8):795–810. doi:10.1016/j.ijimpeng.2007.10.005.
filled aluminum-composite hybrid tubes. Compos Struct 2005;70(2):177–84.
[31] Gümrük R, Mines RAW, Karadeniz S. Static mechanical behaviours of stainless
doi:10.1016/j.compstruct.2004.08.021.
steel micro-lattice structures under different loading conditions. Mater Sci Eng A
[12] Hanssen AG, Langseth M, Hopperstad OS. Static and dynamic crushing of circular
2013;586:392–406. doi:10.1016/j.msea.2013.07.070.
aluminium extrusions with aluminium foam filler. Int J Impact Eng 2000;24:475–
[32] Merkt S, Hinke C, Bültmann J, Brandt M, Xie YM. Mechanical response of TiAl6V4
507. doi:10.1016/S0734-743X(99)00170-0.
lattice structures manufactured by selective laser melting in quasistatic and dynamic
[13] Zarei HR, Kröger M. Optimization of the foam-filled aluminum tubes
compression tests. J Laser Appl 2015;27(S1):S17006. doi:10.2351/1.4898835.
for crush box application. Thin-Walled Struct 2008;46(2):214–21.
[33] Smith M, Guan Z, Cantwell WJ. Finite element modelling of the compressive re-
doi:10.1016/j.tws.2007.07.016.
sponse of lattice structures manufactured using the selective laser melting technique.
[14] Paz J, Díaz J, Romera L, Costas M. Crushing analysis and multi-objective crashwor-
Int J Mech Sci 2013;67:28–41. doi:10.1016/j.ijmecsci.2012.12.004.
thiness optimization of GFRP honeycomb-filled energy absorption devices. Finite
[34] Mines RAW, Tsopanos S, Shen Y, Hasan R, McKown ST. Drop weight impact be-
Elem Anal Des 2014;91:30–9. doi:10.1016/j.finel.2014.07.006.
haviour of sandwich panels with metallic micro lattice cores. Int J Impact Eng
[15] Liu Q, Mo Z, Wu Y, Ma J, Pong Tsui GC, Hui D. Crush response of CFRP
2013;60:120–32. doi:10.1016/j.ijimpeng.2013.04.007.
square tube filled with aluminum honeycomb. Compos Part B Eng 2016;98:406–14.
[35] Shen Y, Cantwell W, Mines R, Li Y. Low-velocity impact performance of lat-
doi:10.1016/j.compositesb.2016.05.048.
tice structure core based sandwich panels. J Compos Mater 2014;48(25):3153–67.
[16] Sun G, Wang Z, Hong J, Song K, Li Q. Experimental investigation of the quasi-static
doi:10.1177/0021998313507616.
axial crushing behavior of filament-wound CFRP and aluminum/CFRP hybrid tubes.
[36] Turner AJ, Al Rifaie M, Mian A, Srinivasan R. Low-velocity impact behavior of sand-
Compos Struct 2018;194:208–25. doi:10.1016/j.compstruct.2018.02.005.
wich structures with additively manufactured polymer lattice cores. J Mater Eng
[17] Reuter C, Tröster T. Crashworthiness and numerical simulation of hybrid
Perform 2018;27(5):2505–12. doi:10.1007/s11665-018-3322-x.
aluminium-CFRP tubes under axial impact. Thin-Walled Struct 2017;117:1–9.
[37] Maskery I, Aboulkhair NT, Aremu AO, Tuck CJ, Ashcroft IA, Wildman RD, et al.
doi:10.1016/j.tws.2017.03.034.
A mechanical property evaluation of graded density Al-Si10-Mg lattice struc-
[18] Mirzaei M, Shakeri M, Sadighi M, Akbarshahi H. Experimental and analytical as-
tures manufactured by selective laser melting. Mater Sci Eng A 2016;670:264–74.
sessment of axial crushing of circular hybrid tubes under quasi-static load. Compos
doi:10.1016/j.msea.2016.06.013.
Struct 2012;94(6):1959–66. doi:10.1016/j.compstruct.2012.01.003.
[38] Maskery I, Hussey A, Panesar A, Aremu A, Tuck C, Ashcroft I, et al. An inves-
[19] Ma Y, Sugahara T, Yang Y, Hamada H. A study on the energy absorption prop-
tigation into reinforced and functionally graded lattice structures. J Cell Plast
erties of carbon/aramid fiber filament winding composite tube. Compos Struct
2017;53(2):109–33. doi:10.1177/ToBeAssigned.
2015;123:301–11. doi:10.1016/j.compstruct.2014.12.067.

483
E. Cetin and C. Baykasoğlu International Journal of Mechanical Sciences 157–158 (2019) 471–484

[39] Rashed MG, Ashraf M, Mines RAW, Hazell PJ. Metallic microlattice materials: a [52] Karagiozova D, Nurick GN, Chung Kim Yuen S. Energy absorption of aluminium
current state of the art on manufacturing, mechanical properties and applications. alloy circular and square tubes under an axial explosive load. Thin-Walled Struct
Mater Des 2016;95:518–33. doi:10.1016/j.matdes.2016.01.146. 2005;43(6):956–82. doi:10.1016/j.tws.2004.11.002.
[40] Xiong J, Mines R, Ghosh R, Vaziri A, Ma L, Ohrndorf A, et al. Advanced micro-lattice [53] Baykasoğlu C, Baykasoğlu A, Tunay Çetin M. A comparative study on
materials. Adv Eng Mater 2015;17(9):1253–64. doi:10.1002/adem.201400471. crashworthiness of thin-walled tubes with functionally graded thick-
[41] Li G, Xu F, Sun G, Li Q. A comparative study on thin-walled structures with function- ness under oblique impact loadings. Int J Crashworthiness 2019;0:1–17.
ally graded thickness (FGT) and tapered tubes withstanding oblique impact loading. doi:10.1080/13588265.2018.1478775.
Int J Impact Eng 2015;77:68–83. doi:10.1016/j.ijimpeng.2014.11.003. [54] Karagiozova D, Alves M. Dynamic elastic-plastic buckling of structural elements : a
[42] Gao Q, Wang L, Wang Y, Wang C. Crushing analysis and multiobjective crashwor- review. Appl Mech Rev 2008;61(4):1–26. doi:10.1115/1.2939481.
thiness optimization of foam-filled ellipse tubes under oblique impact loading. Thin- [55] Varas D, Zaera R, López-Puente J. Numerical modelling of the hydrodynamic
Walled Struct 2016;100:105–12. doi:10.1016/j.tws.2015.11.020. ram phenomenon. Int J Impact Eng 2009;36(3):363–74. doi:10.1016/j.ijimpeng.
[43] Li P, Wang Z, Petrinic N, Siviour CR. Deformation behaviour of stainless steel mi- 2008.07.020.
crolattice structures by selective laser melting. Mater Sci Eng A 2014;614:116–21. [56] Nagel GM, Thambiratnam DP. Dynamic simulation and energy absorption of
doi:10.1016/j.msea.2014.07.015. tapered thin-walled tubes under oblique impact loading. Int J Impact Eng
[44] Al-Saedi DSJ, Masood SH, Faizan-Ur-Rab M, Alomarah A, Ponnusamy P. Mechanical 2006;32(10):1595–620. doi:10.1016/j.ijimpeng.2005.01.002.
properties and energy absorption capability of functionally graded F2BCC lattice [57] Shakeri M, Mirzaeifar R, Salehghaffari S. New insights into the collapsing of cylin-
fabricated by SLM. Mater Des 2018;144:32–44. doi:10.1016/j.matdes.2018.01.059. drical thin-walled tubes under axial impact load. Proc Inst Mech Eng Part C J Mech
[45] Contuzzi N., Campanelli S.L., Casavola C., Lamberti L., Meccanica D., Management Eng Sci 2007;221(8):869–85. doi:10.1243/09544062JMES562.
M., et al. Manufacturing and characterization of 18Ni Marage 300 lattice compo- [58] Song J, Guo F. A comparative study on the windowed and multi-cell square
nents by selective laser melting 2013:3451–68. doi:10.3390/ma6083451. tubes under axial and oblique loading. Thin-Walled Struct 2013;66:9–14.
[46] Baykasoglu C, Cetin MT. Energy absorption of circular aluminium tubes with doi:10.1016/j.tws.2013.02.002.
functionally graded thickness under axial impact loading. Int J Crashworthiness [59] Song J, Zhou Y, Guo F. A relationship between progressive collapse and initial buck-
2015;20(1):95–106. doi:10.1080/13588265.2014.982269. ling for tubular structures under axial loading. Int J Mech Sci 2013;75:200–11.
[47] Hou S, Han X, Sun G, Long S, Li W, Yang X, et al. Multiobjective op- doi:10.1016/j.ijmecsci.2013.06.016.
timization for tapered circular tubes. Thin-Walled Struct 2011;49(7):855–63. [60] Alsi Eosa, Alsi Eosa. Material data sheet EOS Aluminium AlSi10Mg Material data
doi:10.1016/j.tws.2011.02.010. sheet Technical data 2011;49:1–5. https://cdn0.scrvt.com/eos/f3f4bf485d78f2dd/
[48] Dassault Systems Inc. Abaqus User’s Manuel Ver. 6.13. Lowell, MA: 2014. b010c05abe97/AlSi10Mg-090-M400_Flexline_Material_data_sheet_09-15_en.pdf
[49] Tripathy L, Lu WF. Evaluation of axially-crushed cellular truss structures for crash- (Accessed 2 March 2019).
worthiness. Int J Crashworthiness 2017;8265(6):1–17. doi:10.1080/13588265. [61] Hadadzadeh A, Baxter C, Amirkhiz BS, Mohammadi M. Strengthening mechanisms
2017.1389630. in direct metal laser sintered AlSi10Mg: comparison between virgin and recycled
[50] Labeas GN, Sunaric MM. Investigation on the static response and failure pro- powders. Addit Manuf 2018;23:108–20. doi:10.1016/j.addma.2018.07.014.
cess of metallic open lattice cellular structures. Strain 2010;46(2):195–204.
doi:10.1111/j.1475-1305.2008.00498.x.
[51] Zhang Y, Liu T, Ren H, Maskery I, Ashcroft I. Dynamic compressive response of ad-
ditively manufactured AlSi10Mg alloy hierarchical honeycomb structures. Compos
Struct 2018;195:45–59. doi:10.1016/j.compstruct.2018.04.021.

484