1. Introduction
Over the last few years, mechanical metamaterials have attracted significant attention due to their novel design, combining hierarchical architecture with material size effects at the micro/nanoscale [
1]. The research in this field is inspired by several natural materials, such as bone, nacre, enamel, and wood, which demonstrate high stiffness, high strength, and high toughness in combination with a very low density. By tuning a material’s architecture along with its microstructure, unique property combinations not found in conventional engineering materials (metals, ceramics, polymers, and composites) can be achieved [
1,
2]. Examples include exotic and mechanical metamaterials, which may present a negative Poisson’s ratio and negative compressibility [
2].
Lattices (such as pentamodes) are metamaterials with a highly ordered three-dimensional architecture, which are created by periodically repeating a unit cell lattice composed of beams or rods [
1]. Pentamodes are three-dimensional (3D) lattice structures composed of bi-cone beams, which were introduced as a subcategory of mechanical metamaterials by Christensen [
3] and Milton and Cherkaev [
4]. They are named ‘solid waters’, because they exhibit decoupling of bulk and shear moduli; that is, they can have high bulk modulus and low shear modulus [
3,
4].
The main characteristic of pentamodes is the low value of the ratio shear to bulk modulus [
5,
6,
7,
8], which makes them suitable for a wide range of applications from bioengineering applications (bone healing, muscle healing, etc.) [
9,
10,
11] to wave guiding, aeronautical applications (empennage, wings, fuselage design, etc.) [
12], and vibrations isolation (anti-seismic design, elastodynamic cloaking, soundproofing, etc.) [
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24].
Nowadays, the manufacturing of pentamodes is feasible due to the advancement of 3D printing technologies, which did not exist in the 1990s [
25]. Pentamodes have been manufactured since 2012 utilizing 3D printing technology [
26,
27,
28]. Furthermore, pentamodes have been manufactured for a wide range of applications including buckling and crack propagation [
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24,
25,
26,
27,
28]. Therefore, it is necessary to understand their behaviour under different loading conditions [
29,
30].
An established 3D printing technology is the selective laser sintering (SLS) technique [
30,
31,
32,
33]. According to this additive manufacturing (AM) technique, the material, which may be a powder polymer, is selectively sintered by a laser beam, layer by layer. The cross-section is derived from a computer-aided design (CAD) file or a 3D scan. This 3D printing technique has advantages, such as the printing of complex structures with complex geometry characteristics and the manufacturing of parts with high strength and stiffness. A great disadvantage of the method is the rough surface of the manufactured parts, but with post-processing techniques, this disadvantage is eliminated. In addition, the technique consumes a high amount of energy due to the laser scanning. For this reason, a laser pulse is preferred.
The experimental testing of new materials, such as pentamode metamaterials, is essential for understanding their behaviour. Pentamode metamaterials have high bulk strength. For this reason, the behaviour of pentamodes under compressive loading is a very interesting topic.
At first pentamodes were studied analytically by Christensen [
3], Milton and Cherkaev [
4], who calculated the Young’s and shear moduli. The advance of numerical methods and computer power resulted in the study of pentamodes using techniques such as the finite element method. The majority of studies are oriented around the experimental, computational, and analytical evaluation of pentamodes. The finite element analyses provided the opportunity to study a wide range of pentamode structures [
5,
6,
8,
9,
10,
11,
12,
13,
14,
15,
16,
17,
18,
19,
20,
21,
22,
23,
24,
25,
26,
27,
28]. In addition, with the introduction of 3D printing, complex structures such as pentamode lattice structures became feasible to manufacture. Thus, a wide range of experiments may be considered for pentamode lattice structures in order to validate the experimental results with those from analytical and numerical studies [
7,
26,
27,
28,
29,
30,
31,
32,
34,
35,
36,
37,
38].
Our study is divided into two parts. First, four specimens from two batches were manufactured using the SLS technique [
32,
33]. The specimens were compressed in a tensile-compression machine in the laboratory. A prescribed displacement is applied on the pentamode structure’s upper surface, whereas the bottom surface is fully constrained. The reaction force is measured until the pentamode structure fractures. Then, the pentamode structure is numerically confronted using the finite element method, and the experimental results are compared with the numerical ones. In conclusion, the behaviour of different pentamode specimens is numerically studied, and the results are compared with those obtained from experiments. The comparison between the experimental and numerical results is performed up to the failure of the specimens in contrast with previous studies, and very interesting results arise. Finally, the results were also compared with those from analytic solutions.
3. Results
Figure 7 presents the pentamodes of the first batch of specimens (
Table 1) and the results from the finite element analyses under different compressive displacements applied at the top surface nodes
, considering the bottom surface nodes
fully constrained.
The plots in
Figure 7 show that the numerical and experimental results for the first batch of specimens almost coincide until failure in the case of large displacement and bilinear material analysis [
41]. The results from the second batch of specimens are presented in
Figure 8.
Similarly, for the second batch of specimens, according to the plots in
Figure 8, the numerical and experimental results almost coincide in the case of large displacement and bilinear material analysis. However, in
Figure 8, small deviations between the numerical and experimental results are observed up to 1.5 mm, which may be due to the prestressing effect, which does not appear for the first batch of specimens (
Figure 7). The prestress is the consideration of a preload of 1/100 of the maximum load defined in the machine in order to have a better grab on the plates. The deviation in the second batch may be due to the existence of some defects in the specimens. From the experiments, stress relaxation issues are not observed.
In the sequel, the pentamode behaviour was studied numerically (
Figure 9) for constant
and different small diameters
in the case of large displacements and bilinear material analyses. In addition, the case
is examined correspondingly to the second batch of specimens.
According to the plots in
Figure 9, the small diameter d of the bi-cone beams appears to make a significant contribution to the pentamode’s strength. It is also observed that the results for the different
are between the two experimental results for the two batches of specimens. It should be noted that the pentamodes with the same small diameter
and different diameters
appear to have approximately the same strength, whereas the force–displacement curve appears to have discrepancies.
Furthermore, the effective Young’s moduli
, for the different diameter d (
Figure 9), calculated by the following Equation (1) [
5,
21], are presented in
Table 3.
where F is the total reaction force at the nodes of the bottom surface of the pentamodes,
is the pentamode height,
is the applied displacement, and
is the area of the pentamode where the displacement is applied. In our study,
represents the ratio of the average stresses to the average strains in the body when subjected to pure compression.
In order to study the stress field under compressive loading, the stresses according to the Von Mises criterion are also plotted, and the results for the first batch of specimens (
Table 1) and the different applied displacements
are presented in
Figure 10.
The maximum stresses according to Von Mises criterion from
Figure 10 are presented in
Table 4. According to the results, the maximum stresses are close to the material’s ultimate tensile strength (
Table 2) for the applied displacements
. It should be mentioned that the maximum stress for the applied displacement of 2 mm has a higher value than the maximum stress for the applied displacements of 3 mm and 4 mm. This may be explained by the fact that for the cases of 3 mm and 4 mm, the specimen has already failed. The maximum stresses were observed near the beam connections (
Figure 10c).
The stress field according to the Von Mises criterion was also studied for the second batch of specimens (
Table 1), and the results for the different compressive displacements
are plotted in
Figure 11.
From
Figure 11, the maximum stresses according to the Von Mises criterion are presented in
Table 5. The maximum stresses are observed to be close to the material’s ultimate tensile strength (
Table 2) for the applied compressive displacements
. The maximum Von Mises stress values also appeared at the beam’s connections (
Figure 11c).
In the sequel, to study the behaviour of pentamode specimens for different heights (
), the effective Young’s modulus is calculated and plotted in
Figure 12 and
Figure 13 for the first and second batches of specimens, respectively, and for the applied compressive displacement of 1 mm (
Table 1).
It is observed that for the first batch of specimens (
Figure 12), the pentamode with
appeared to have the highest effective Young’s modulus. As the pentamode height
increases, the effective Young’s modulus decreases in value. In the case of the second batch of specimens, the results for
are presented in
Figure 13.
It is observed (
Figure 13) that the results for the second batch of specimens show a different behaviour than those of the first batch (
Figure 12). There is an abrupt reduction of the value of
between
and
, compared with the reduction of the value of
in the first batch of specimens (
Figure 12). Higher values of
also appeared in the second batch than in the first batch of specimens.
It should be noted that the reduction of the effective Young’s modulus
as
increases (
Figure 12 and
Figure 13) also appeared in the studies of Amendola et al. [
5,
6].
In order to study the behaviour of pentamodes under different applied vertical displacements and heights, the effective Young’s modulus is also calculated under different vertical displacements (
Figure 14 and
Figure 15) for the first and the second batch of specimens.
For the second batch of specimens, higher values of
appeared relative to the first batch of specimens. From
Figure 14 and
Figure 15, in pentamodes with
,
discrepancies appeared for the values of
when different compressive displacements were applied to them. For the cases of
,
,
appeared to approach a more constant value as the displacement increased. It is also observed that as the values of the applied compressive displacements increased, the values of the
generally decreased. But for the cases of
, after the applied compressive displacement of 3 mm, small discrepancies are observed.
Finally, the pentamode bulk modulus will be compared for the results from the analytical equations presented in the studies of Fabbrocino et al. (2016) [
6] and Lymperopoulos and Theotokoglou (2022) [
22], the effective Young’s modulus from the finite element analyses, and the experimental results. According to Fabbrocino et al. [
6], the bulk modulus is given by the following Equation (2),
whereas according to Lymperopoulos and Theotokoglou [
22], the bulk modulus is given by Equation (3),
where
is the Young’s modulus of the material,
is the small diameter,
is the wide diameter and
is the length of the bi-cone beam (
Figure 1,
Table 2). In the following table (
Table 6), the results are presented.
It is observed that for the second batch, the results from the analytical Equation (3), the finite element analyses, and the experimental results almost coincide.
4. Discussion
In our study, two batches of specimens were considered. The first batch of specimens
appeared to have ductile fracture behaviour according to the numerical and experimental analyses, while the second batch of specimens
appeared to have brittle fractures. Fracture is a multifactorial phenomenon that can be influenced by many factors, including environmental factors like temperature and humidity, materials composition, etc. The results of the reaction forces at the bottom nodes of the bi-cone specimens for the different diameters increased as the values of the diameter d increased. It was also observed that the dimension of the d diameter was a significant factor in the final strength of the structure as had also been observed in previous studies [
5,
6,
22]. The prescribed behaviour of the pentamode structures under different specimen dimensions showed great validity. The failure of the pentamode structure from the experimental procedure was observed to take place at the beam connections, which was also verified by the numerical simulations according to the maximum stresses from the Von Mises criterion observed at these points. It is worth noting that other studies [
5,
23] also found increased values of the specimens at the beam connections. However, these studies were performed only computationally or experimentally but without fracturing the specimens in a laboratory. In our study, analyses of pentamode lattice structures were performed computationally and experimentally for the first time up to the fracture of the specimens. Furthermore, some analytical results were also presented and were compared with numerical and experimental results.
From our experimental analyses, it was observed that the pentamode specimens start to rotate under compression loading at the beginning of fracturing. This behaviour may be explained by their lattice configuration.
The maximum stresses, according to the Von Mises criterion, were observed at the beam connections. In addition, it was observed that the maximum stresses appeared in the second batch of specimens were lower with respect to thoses in the first batch of specimens. This behaviour may be explained by the fact that the second batch of pentamode specimens has a wider beam cross section than that of the first batch, which resulted in reduced Von Misses stresses; consequently, the final strength of the second batch of specimens was higher than that of the first batch. Pentamodes of the first and second batches of specimens for the applied compressive displacements higher than 2 mm, fail according to von Mises criterion.
It was also observed that the value of the effective Young’s modulus
decrease as the heights
of the pentamode increase. The prescribed behaviour was similar to that proposed by Amentola et al. [
5,
6]. It was also observed that the values of
for the pentamode specimens with heights higher than
converge to a rather constant value for both batches of specimens.
It should also be noted that the small diameter
is a significant factor in pentamodes’ overall ultimate strength, rather than
. This was also referred to in the studies of Amendola et al. [
5], Fabbrocino et al. [
6,
7], and Lymperopoulos and Theotokoglou [
22].
Furthermore, the effective Young’s modulus [
5,
6] is a first approach to understanding the pentamode’s lattice structures’ behaviour. The bulk modulus [
6,
7] represents the elastic stiffness of the lattice structures taking into consideration their periodic nature.