Materials Letters 164 (2016) 456–459

Contents lists available at ScienceDirect

Materials Letters
journal homepage: www.elsevier.com/locate/matlet

A unit cell structure with tunable Poisson's ratio from positive to

negative
Dong Li a,n, Liang Dong b, Roderic S. Lakes c
a
College of Sciences, Northeastern University, Shenyang 110819, PR China
b
Materials Science and Engineering, University of Virginia, Charlottesville, VA 22904, USA
c
Department of Engineering Physics, University of Wisconsin, Madison, WI 53706-1687, USA

art ic l e i nf o a b s t r a c t

Article history: A novel unit cell structure with re-entrant hollow skeleton was designed and its Poisson's ratio was
Received 23 September 2015 studied using the finite element method (FEM) as a function of the geometric variables and the parent
Received in revised form material Poisson's ratio. The simulation results showed that the Poisson's ratio of the unit cell structure is
27 October 2015
tunable over the full range of Poisson's ratio (
1 to þ0.5) by varying the geometric variables and parent
Accepted 7 November 2015
material Poisson's ratio. Three types of samples with identical geometric variables were fabricated via 3D
Available online 10 November 2015
printing using three different materials, and their Poisson's ratios were measured and compared with the
Keywords: ones from FE simulation, and excellent agreement was found between the experimental results and FE
Re-entrant cellular structure simulation.
Negative Poisson's ratio
& 2015 Elsevier B.V. All rights reserved.
Simulation and modeling
Elastic properties
Structural

1. Introduction measurements and simulation. This design concept can be used to

fabricate hybrid cellular structures with diverse physical and me-
Poisson's ratio (ν) is defined as the ratio of the transverse chanical properties at different parts of the structures to meet
contraction strain to the longitudinal extension strain in tension. special robust mechanical performance requirements.
Poisson's ratio can range from
1 to 0.5 for isotropic materials,
but most solids have a positive Poisson's ratio (0.25–0.33). Nega-
tive Poisson's ratio was initially studied in 2-D structures [1–3]. 2. Simulation
Negative Poisson's ratio in 3-D isotropic materials has been ob-
served in designed cellular materials with re-entrant structures The Poisson's ratio of a unit cell structure with a re-entrant
[4,5], polymer gels [6] and ferroelastic ceramic [7] near phase hollow skeleton has been studied using the commercial finite
transition temperatures and in In–Sn alloy [8,9] near a morpho- element software ANSYS (see Fig. 1a and b). Poisson's ratio of the
tropic phase boundary. Negative Poisson's ratio is also known in parent material used to make the structure was set to vary from
anisotropic materials such as orthorhombic alloy [10], and certain 0 to 0.5 with an incremental step of 0.1. Free meshes of 82,980
in-plane directions of 3D structures [11,12] and in more recent elements (Solid 186, 20 nodes) were used. h/l was set to vary from
0 to 0.45 with an incremental step of 0.025, and L/l was set to vary
study of 2-D structures [13–15].
from 1.1 to 1.4 with an incremental step of 0.1. The bottom surface
In this paper, a unit cell structure with a re-entrant hollow
of the unit cell model is fixed, with all degrees of freedom being
skeleton has been designed, analyzed, fabricated and tested. The
constrained, and the compressive strain was set as 0.1% on the top
unit cell Poisson's ratio has been studied using FEM to investigate
surface of the model in the vertical direction. It was found that the
the effects of geometric variables and parent material properties
simulation results of the unit cell Poisson's ratio are independent
upon the unit cell Poisson's ratio. Three samples with identical
of the Young's modulus of the parent material.
geometric variables were fabricated by 3D printing from three
different materials (Vero White Plus, Tango Black Plus and PLA),
and their Poisson's ratios were measured and compared with si-
3. Experiment
mulation results; excellent agreement was found between
Object Eden 260 3D Printer and MakerBot Replicator z18 3D
n
Corresponding author. Printer were used to fabricate the unit cell samples. Three types of
E-mail address: [email protected] (D. Li). unit cell samples with identical geometric variables were printed

http://dx.doi.org/10.1016/j.matlet.2015.11.037
0167-577X/& 2015 Elsevier B.V. All rights reserved.
 D. Li et al. / Materials Letters 164 (2016) 456–459 457

Fig. 1. (a) The ANSYS model of the unit cell structure with a re-entrant hollow skeleton; (b) schematic drawing of the unit cell structure viewed in the x–z plane; (c) a sample
fabricated by 3D printing from Vero White Plus.

Fig. 2. The relationships between the Poisson's ratio of the designed cellular structure and the h/l ratio from 0 to 0.45. Plot (a), L/l¼1.1; (b), 1.2; (c), 1.3; (d), 1.4. Symbols (■),
(●), (▲), (▼), (◆) and (◄) represent different parent materials with Poisson's ratio of 0, 0.1, 0.2, 0.3, 0.4, and 0.45, respectively, used to make the designed cellular structures.

from three different materials: Vero White Plus, Tango Black Plus, materials have three independent elastic constants [16]. The unit
and PLA. Vero White Plus has the highest stiffness among the cell samples were compressively loaded in the Z direction using a
three materials while the Tango Black Plus is most compliant. A universal testing machine (Zwick010, Germany) with a testing
printed sample is shown in Fig. 1c. One face of the cube was re- speed of 2 mm/min (corresponding to a strain rate of 10
3/s). The
moved to show in the inner structure. The geometric variables of transverse displacements in both the X and Y directions in the
the printed samples are h ¼12 mm, l¼ 30 mm, L ¼33 mm. middle plane of the unit cell sample were recorded by a digital
Cubic materials have equal moduli and Poisson's ratios in the outside micrometer (Guilin Measuring and Cutting Tool Co. Ltd.
principal directions but can differ in oblique directions. Cubic China) at the unit cell 0.4% normal strain for Vero White Plus and
458 D. Li et al. / Materials Letters 164 (2016) 456–459

2% normal strain for Tango Black Plus and PLA. The strains were
selected because they are lower than the unit cell elastic limit but
sufficiently high that the end effect (due to frictions between the
test sample and the surfaces of the platens) can be neglected. The
high machining precision of 3D printing route ensures a geome-
trically accurate final product. If the symmetry is orthotropic, there
are three independent Poisson's ratios. To test symmetry, load was
applied in each principal direction and Poisson's ratio measured.
Negligible difference (less than 5%) was found between νxy , νyz , and
νxz , a fact that suggests good cubic responses of the fabricated unit
cell structure. In this study, the arithmetic mean of the X and Y
direction displacements was recorded as the transverse strain. The
arithmetic mean of the transverse strains recorded from 8 separate
measurements was used to determine the unit cell Poisson's ratio
v.

4. Results and discussion

Fig. 3. Poisson's ratio of the cellular structure as a function of that of the parent
4.1. Simulation results material used to fabricate the structure. The experimental data of samples made
from PLA, Vero White Plus and Tango Black Plus are represented by (□), (○), and (△),
respectively. (-■-) represents the simulation results, and (─) is the fitted curve for
Poisson's ratio is defined as the ratio of transverse contraction the simulation results.
strain to longitudinal extension strain in the direction of stretching
force as:
(see Fig. 3). The measured unit cell Poisson's ratios were
ν=−
ϵ′
0.455 70.033 (standard deviation),
0.604 70.035 and
ϵ (1)
0.66070.024 for samples made from Tango Black Plus, Vero
White Plus, and PLA, respectively.
where, ν is the Poisson's ratio, ϵ′ and ε are the transverse
The least square method was used to curve fit the simulation
contraction strain and longitudinal extension strain, respectively.
results:
Fig. 2 shows the relationships between the simulated Poisson's
ratio of the designed unit cell and its h/l ratio from 0 to 0.45 when ν = 0.47ν02 + 0.48ν0 − 0.80 (2)
the L/l ratio is varied from 1.1 to 1.4. The unit cell Poisson's ratio
decreases with a decreasing value of the Poisson's ratio ( ν0 ) of the In Eq. (2), ν refers to the unit cell Poisson's ratio and ν0 refers to
parent material used to make the structure. The unit cell Poisson's the parent material Poisson's ratio. It can be seen that the mea-
ratio exhibited a similar behavior for different parent material surements are in excellent agreement with the simulation results.
Poisson's ratios ( ν0 ): for any specific Poisson's ratio of the parent
material, the unit cell Poisson's ratio decreases rapidly with an
increasing h/l ratio from 0 to 0.2; the unit cell Poisson's ratio de- 5. Conclusions
creases at a much slower rate beyond h/l ¼ 0.2 in all cases except L/
l ¼1.1, where a slight increase is observed after h/l ¼0.2. The si- A novel unit cell structure with a re-entrant hollow skeleton
mulation results have shown that the unit cell Poisson's ratio is
has been proposed and studied via FEM. The unit cell Poisson's
tunable from positive to negative with an increasing h/l and an
ratio is found to be tunable between
1 and 0.5 by varying the
appropriate combination of L/l and the parent material Poisson's
geometric variables and the parent material Poisson's ratio. Three
ratio. A minimum unit cell Poisson's ratio,
0.84, is achieved
types of unit cell samples with identical geometric variables were
when h/l ¼0.2, L/l ¼1.1 and ν0 ¼0. The simulation results show the
fabricated via 3D printing from three different materials, and the
full range of the Poisson's ratio as a function of geometric vari-
unit cell Poisson's ratios were measured during uniaxial com-
ables; the wide tunable range of the Poisson's ratio is of interest
pression testing. The measurements have showed excellent
for practical designs. Classical elasticity has no length scale, so
there are no size effects in classical elastic moduli. The single cell agreement with the simulation results. The design concept pro-
will be equivalent to a lattice made of such cells provided the links posed in the present study can be used to fabricate hybrid cellular
between cells replicate the boundary conditions assumed and lattice structures with diverse physical and mechanical properties
provided applied stress is uniform. In future lattices, cells can be at different parts of the structures. As structures with negative
made smaller and lattices can be made with many cells. A cellular Poisson's ratio have higher indentation resistance and higher en-
material constructed by stacking identical unit cells is expected to ergy absorption capacity than conventional materials, the hybrid
exhibit homogeneous constitutive parameters; however, the cellular materials constructed by specially designed 3D packing of
properties of a material constructed by stacking hybrid unit cells the proposed unit cell structures with diverse geometric and
requires further study. material variables could have potential applications in such as
sports facilities, aerospace fillers and bio-implants.
4.2. Experimental results

The Poisson's ratios of three different parent materials, Vero Acknowledgments

White Plus, Tango Black Plus, and PLA, are 0.33, 0.49, and 0.23
respectively. Two samples were made for each type of unit cell This work is supported by “The National Natural Science
designed in the present study. The measured Poisson's ratios of Foundation of China (11304033) and the Specialized Research
the total six unit cell samples were plotted against their parent Fund for the Doctoral Program of Higher Education of China
material Poisson's ratios and compared with the simulation results (20120042120031)”.
 D. Li et al. / Materials Letters 164 (2016) 456–459 459

References [8] D. Li, T.M. Jaglinski, D.S. Stone, R.S. Lakes, Appl. Phys. Lett. 101 (2012) 251903.
[9] D. Li, L. Dong, R.S. Lakes, Mater. Lett. 143 (2015) 31–34.
[10] M. Rovati, Scr. Mater. 48 (2003) 235–240.
[1] L.J. Gibson, M.F. Ashby, G.S. Schajer, C.I. Robertson, Proc. R. Soc. Lond. A 382 [11] H. Hu, V. Silberschmidt, J. Mater. Sci. 48 (2013) 8493–8500.
(1982) 25–42. [12] X.N. Hou, H. Hu, V. Silberschmidt, Composites: Part B 58 (2014) 152–159.
[2] A.G. Kolpakov, Prikl. Mat. Mekh. 49 (1985) 969–977. [13] Y. Prawoto, Comput. Mater. Sci. 58 (2012) 140–153.
[3] K.W. Wojciechowski, Phys. Lett. A 137 (1989) 60–64. [14] N.T. Kaminakis, G.E. Stavroulakis, Composites: Part B 43 (2012) 2655–2668.
[4] R.S. Lakes, Science 238 (1987) 551. [15] A.A. Pozniak, J. Smardzewski, K.W. Wojciechowski, Smart Mater. Struct. 22
[5] T. Buckmann, R. Schittny, M. Thiel, M. Kadic, G.W. Milton, M. Wegener, New J. (2013) 1–11.
Phys. 16 (2014) 033032. [16] J.F. Nye, Physical Properties of Crystals, Clarendon, Oxford, 1976.
[6] S. Hirotsu, J. Chem. Phys. 94 (1991) 3949.
[7] L. Dong, D.S. Stone, R.S. Lakes, Philos. Mag. Lett. 90 (2010) 23–33.