Journal Pre-proofs

Compression behaviors of 3D printed pyramidal lattice truss composite struc-

tures

Gaoyuan Ye, Hongjie Bi, Yingcheng Hu

PII: S0263-8223(19)33402-6
DOI: https://doi.org/10.1016/j.compstruct.2019.111706
Reference: COST 111706

To appear in: Composite Structures

Received Date: 10 September 2019

Revised Date: 7 November 2019
Accepted Date: 18 November 2019

Please cite this article as: Ye, G., Bi, H., Hu, Y., Compression behaviors of 3D printed pyramidal lattice truss
composite structures, Composite Structures (2019), doi: https://doi.org/10.1016/j.compstruct.2019.111706

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will
undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing
this version to give early visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Published by Elsevier Ltd.

 1 Compression behaviors of 3D printed pyramidal lattice truss

2 composite structures

3 Gaoyuan Ye a, Hongjie Bi a, Yingcheng Hu a,*

4 a Key Laboratory of Bio-based Material Science and Technology of the Ministry of Education of

5 China, College of Material Science and Engineering, Northeast Forestry University, Harbin

6 150040, China

7 * Corresponding author. E-mail address: yingchenghu@nefu.edu.cn; Tel: +86-451-82190394

8 Abstract: The lattice truss structure is a porous lightweight periodic structure with high

9 specific stiffness and specific strength. In this research, pyramid lattice truss structures

10 were designed and manufactured by three-dimensional (3D) printing technology

11 through horizontal printing and vertical printing manners, respectively. Quasi-static

12 axial compression tests were conducted to study the mechanical properties and energy

13 absorption of the pyramidal lattice truss structures. These two types of printed lattice

14 truss materials had close strength and rigidity. However, the vertically printed lattice

15 truss structures had excellent ductility and their struts never broken off during

16 compression. Three post failure styles of the vertically printed lattice truss structures

17 were observed, including strain-hardening, stable deforming and softening and

18 depending on the slenderness ratio of the strut. Theoretic analysis and finite element

19 method (FEM) were performed to investigate the compression behaviors of the

20 vertically printed lattice truss materials. Appropriate printing method and relative

21 density could make the 3D printed pyramidal structures achieve excellent specific
22 energy absorption.

23 Keywords: Pyramidal lattice structure; Energy absorption; Specific strength; Specific

24 stiffness

25 1 Introduction

26 Lattice truss and anisogrid structures have excellent specific stiffness, specific

27 strength, and have been applied in transportation structures for their lightweight [1, 5]

28 and protective structures for their excellent energy absorbing ability [6, 7]. The

29 mechanical property and energy absorption of lattice structures prepared by different

30 methods were studied. Wang et al. [8-13] presented a fundamental study on the

31 mechanical behavior of composite sandwich structures with pyramidal truss core. The

32 prepared composite pyramid lattice sandwich structures have excellent mechanical

33 properties. Li et al. [14-16] designed and optimized the compressive behavior of bio-

34 based 2-D lattice structure. Results show that bio-based 2-D lattice structure has a

35 superior behavior compared to other biomaterials. Zhang et al. [17] used a combination

36 of numerical simulation and experiment to study the impact performance of the

37 pyramidal structure, and the result shows that the impact location would greatly

38 influence the impact damage.

39 Xiong et al. [18] tested sandwich panels with hollow Al-Si tube core, and reported

40 that the structure achieved superior specific strength. Li et al. [19] studied the

41 compression and shear properties of pyramid structures under vacuum thermal cycling.

42 Micro-lattice structures were manufactured via selective laser melting, and its
43 compression property was studied [20]. An octet truss lattice structure was designed

44 via selective laser melting, and its specific energy absorption was high [21]. Feng et al.

45 [22] designed an hourglass lattice structure, and presented an analysis of its in-plane

46 and out-of-plane compression performances.

47 In this paper, polylactic acid (PLA) was selected as three-dimensional (3D) printing

48 material. The method to print the structure was first studied. Then, the compression

49 performance and energy absorption characteristic of the lattice structures with the same

50 strut diameter but different strut lengths were studied.

51 2 Manufactures

52 2.1 Lattice truss structure

53 A pyramidal lattice structure was designed to study its compression properties. The

54 design was modified based on both strut diameter and length. The schematic diagram

55 of the utilized pyramidal lattice structure is shown in Fig. 1. The parameters 𝑡, 𝑑, 𝑙,

56 and 𝜔 represent the distance between two struts, strut diameter, strut length, and

57 inclination angle, respectively. The geometrical parameters are assumed as t = 15 mm,

58 ω = 60°, 𝑑 = 6 mm, and 𝑙 is a variable.

59
60 Fig. 1 Diagram of the pyramidal structure

61 The relative density of the pyramidal structure is defined as:

πd 2
62 ρ= (1)
sin ω ( 2 l cos ω  2 t ) 2

63 The model was drawn with the 3D software SolidWorks 2018, and the pyramidal

64 lattice structure and its strut were printed with the Raise 3D Pro2 Plus 3D printer,

65 purchased from Shanghai Fuzhi Information technology co., LTD in China. Four

66 models of each type were printed. The printing temperature was 205 °C, the printing

67 speed was 50 mm/min, and the layer thickness was 0.2 mm.

68 2.2 Printing methods

69 The lattice truss structures are printed using PLA (1.75 mm diameter). The density

70 of the PLA is 1.08 g/cm3. PLA is a biocompatible and biodegradable material, which

71 does not pollute the environment. It also has good mechanical properties.

72 Since the model is 3D printed, the printing method will affect the performance of the

73 model. In this research, two printing methods were firstly compared. The sandwich

74 panel is parallel to the platform of the 3D printer in printing is defined as horizontal

75 printing, while the structure panel is vertical to the platform of the 3D printer when

76 printing is defined as vertical printing (See Movie 1 and Movie 2). The printed model

77 is selected to have a strut diameter of 6 mm and a strut length of 30 mm.

78 Usually, the lattice truss structure is printed layer by layer along the thickness of the

79 sandwich panel. Along the length of strut, a number of horizontal stacks of interfaces

80 are observed, as shown in Fig. 2 (a). If the vertical printing method is applied, there are

81 many interfaces that are at an angle of 60 degrees to the structure panel, as shown in

82 Fig. 2 (b).

(a)
 (b)

83 Fig. 2 Printing model of (a) horizontal printing and (b) vertical printing

84 2.3 Failure styles relating to the two printing methods

85 To study the mechanical property of the 3D printed pyramidal lattice structure, a

86 compression test of out of plane was conducted by using a universal mechanical testing

87 machine according to the ASTM C365 standard. The loading speed was 2 mm/min. The

88 strut was compressed according to ASTM D695-2010. The photo of 3D printer and the

89 compression diagram are shown in Fig. 3.

 (a) (b)

90 Fig. 3 (a) 3D printer and (b) Compression diagram

91 The structural load displacement curves of different printing methods are shown in

92 Fig. 4. The compression curve of a horizontally printed structure has two phases: an

93 elastic phase and a phase during which the force decreases rapidly with increasing

94 displacement. For a vertically printed structure, the compression curve has three phases:

95 elastic phase, plateau phase, and densification phase. It can be deduced from Fig. 4 that

96 the failure mode of a horizontally printed lattice structure is that the strut breaks,

97 because the compressive curve drops sharply for displacements between 5-6 mm. The

98 actual destruction mode of the structure (shown in Fig. 5) also indicates that this

99 inference is correct. The failure mode of the horizontally printed structure is the

100 separation and crush between the strut and the panel. The failure mode of the vertically

101 printed structure is that the strut is bent, but does not break, as shown in Fig. 5 (b).

102 The horizontally and vertically printed structures are similar in experimental

103 phenomena during the elastic phase. When the strut begins to plastically deform, the
104 hinges are produced. Each layer of the strut of the horizontally printed structure is

105 parallel to the structure panel, and the bonding area of the layer between the strut, the

106 strut and the panel is small (see Movie 1), so that the strut can’t withstand shear

107 deformation and breaks on the core or between the core and the panel; The strut printing

108 layer of the vertically printing lattice structure is along the length direction of the strut,

109 and the bonding area between the layer and the layer is large (see Movie 2), so the strut

110 does not break in the shear deformation.

Exp.1
5 Exp.1
Exp.2 5
Exp.3 Exp.2
Exp.4 Exp.3
4 4 Exp.4
Force/kN

Force/kN

3 3

2 2

1 1

0 0
0 1 2 3 4 5 6 0 5 10 15 20
Displacement/mm Displacement/mm

(a) (b)

111 Fig. 4 Compressive curves of (a) horizontally printed and (b) vertical printed lattice

112 truss structures

(a) (b)

113 Fig. 5 Failure modes of (a) horizontally printed and (b) vertically printed lattice truss

114 composite structures.

115 Table 1 Mechanical properties of printed lattice structures.

Horizontally printed structure Vertically printed structure

Peak force Elastic modulus Peak force Elastic modulus
(N) (MPa) (N) (MPa)
1 4718 45.48 5051 46.2
2 4668 45.1 5107 46.63
3 4725 44.54 5023 46.25
4 4623 40.23 4513 41.28
Averag 4684.5 43.84 4923 45.17
e

116 The mechanical properties of printing structure are listed in Table 1. The peak force

117 and elastic modulus of the structure of both printing methods are very close. Since the

118 vertically printed structure does not suddenly break during the compression process and

119 has a densification stage (which it is very important for the structure), the vertically

120 printing method was selected for the following structure.

121 3 Mechanical behaviors of vertically printed lattice truss structures

122 3.1Three crushing modes

123 Table 2 Structural parameters and relative density

Type t ω 𝒅 𝒍 Theoretic Actual Core layer Slenderness

/mm /mm /mm 𝝆 𝝆 ρ g/cm3 ratio 𝒅/𝒍
A 10 9.5% 10.7% 0.12 0.6
B 20 6.7% 7.4% 0.08 0.3
C 15 60° 6 30 4.9% 4.7% 0.05 0.2
D 40 3.8% 3.6% 0.039 0.15
E 50 3.1% 3.1% 0.034 0.12

124 The vertical printing method was selected for the following printed structures. While

125 the strut diameter was fixed to 6 mm, the lengths of the strut were 10, 20, 30, 40, and

126 50 mm. The lattice structure of strut diameter 6 mm and strut length 10 mm is named

127 A, and the lattice structure of strut diameter 6 mm and strut length 20 mm is named B,
128 and so on. The parameters of structures are showed in Table 2.

129

8
A
B
C Densification
6 D
Stress (MPa)

E
Peak efficiency
Peak load
4 Densification
Strain-hardening
Peak load
2 Stable deforming

Peak load
Softening
0
0.0 0.2 0.4 0.6 0.8
130 Strain
131 Fig. 6 Compression curves of lattice truss structures with different strut length.

132 Typical compression curves are compared in Fig. 6. Three post-failure deformation

133 styles are revealed by the experiments, including strain hardening, stable deforming and

134 softening. When the strut is only 10 mm long, the strut is stout enough with slenderness

135 ratio of 0.6 and compression-shear coupling deformation mode is observed, as shown

136 in Fig. 7. After the initial peak load, the load still increases gradually and the post-

137 failure deformation is strain hardening induced by the contact between the struts and

138 the skins. When the strut length is enlarged to 20 mm or 30 mm, the slenderness ratio

139 is greater than 0.2. The deformation curves have long and stable plateau. The strut still

140 has compression-shear coupling effect. When the strut length is extended to 40 and 50
141 mm, it is slender with slenderness ratio smaller than 0.2. The flexural deformation lets

142 the deformation curves softening after the peak load. Densification disappears when the

143 slenderness ratio is smaller than 0.2.

144 The failure modes of the lattice truss structures in the post-failure stage are compared

145 in Fig. 7. When the strut is only 10 mm long, the strut has one compression-shear plastic

146 hinge. With the thickening of the strut, the cross section of the hinge is enlarged and

147 the struts easily contact with the skins and then the strength is improved. When the strut

148 length is enlarged to 30 mm, the strut has compression-shear plastic hinge at the end.

149 When the strut length is enlarged to 50 mm, the strut has three flexure plastic hinges.

(a) A lattice structure

(b) B lattice structure

(c) C lattice structure

 (d) D lattice structure

(e) E lattice structure

150 Fig. 7 Failure modes of different lattice truss structures.

151 Table 3 Compression properties of lattice truss structures.

Type Strength Elastic Specific Specific Load mass

(MPa) modulus strength stiffness/ ratio
(MPa) (103m2s-2) (103m2s-2) (N/g)
A 4.08 69.4 34 578.3 544
B 2.56 62.6 32 782.5 325
C 1.62 41.3 32.4 825.6 205
D 1.19 32.5 30.5 833.3 152
E 0.95 30.8 27.9 905 120

152 The compression performances of structure are listed in Table 3. As shown in Table

153 3, a larger strut slenderness ratio indicates greater strength and elastic modulus. The

154 strength and elastic modulus of the A lattice structure were largest, and the strength and

155 elastic modulus of the E lattice structure was smallest. The maximum strength was more

156 than four times of the minimum strength; the maximum elastic modulus was more than

157 twice of the minimum elastic modulus.

158 Specific strength and stiffness are also important indexes for measuring structural
159 compression performance. These are defined as the ratio of strength and elastic modulus

160 to the core layer density ρ (the relative density of the structure, multiplied by the density

161 of the material). The core layer density of lattice structure is shown in Table 2. Specific

162 strength and stiffness are listed in Table 3. Table 3 shows that the specific strength

163 increases almost with the increasing slenderness ratio, and the specific stiffness

164 decreases with increasing slenderness ratio. The reason why the specific stiffness

165 decreases as the core slender ratio increases is that the increasing trend of the core layer

166 density is stronger than the increase of the elastic modulus.

167 Considering the weight of the load and the weight of the structure that can be

168 described by the load mass ratio, the load mass ratio can be defined as:

Fmax
169 X (2)
m
170 Where X represents the load mass ratio, 𝐹𝑚𝑎𝑥 represents the peak force of the lattice

171 structure, and 𝑚 represents the mass of the lattice structure. The load mass ratios of

172 lattice structures are listed in Table 3. Table 3 shows that the load mass ratio decreases

173 gradually with decreasing slenderness of the strut ratio. This is the case since a structure

174 with high slenderness ratio of the strut is lighter than a structure with small slenderness

175 ratio of the strut. Consequently, when the peak forces structures can withstand are

176 basically identical, the larger slenderness ratio of the strut indicates the larger load mass

177 ratio of the structure. Among the lattice structures, the structure with the largest load

178 mass ratio was the structure with a strut length of 10 mm, and its load mass ratio was

179 544 N/g. The maximum load mass ratio was more than four times the minimum load
180 mass ratio, and the minimum load mass ratio was 120 N/g for a strut length of 50 mm.

181 The load mass ratio of the pineapple leaf fiber lattice cylinder structure prepared by

182 Hao et al. [23] was 79.323 N/g, and the load mass ratios of the glass fiber and carbon

183 fiber lattice cylinder structure prepared by Zhang et al. [24] were 47.23 N/g and 69.35

184 N/g, respectively. The highest load mass ratio of 3D printed structure is about seven

185 times that of those structures. Therefore, 3D printed pyramidal lattice structures have

186 an excellent load mass ratio.

187 The strain value corresponding to the maximum value of the energy absorption

188 efficiency is the strain value of the densification point. The energy absorption efficiency

189 is defined as:

ε1

190 Ee 
 0 σ (ε)dε (3)
σ (ε1 )

191 The densification point is at a point where the increase in per unit volume energy

192 absorption of the structure is less than the corresponding increase in stress. The peak

193 efficiency is considered to be below 1 [25], because due to its structure, the stress does

194 not decrease during the plateau stage. This does not apply to the structure that was

195 investigated in this paper; therefore, this paper can only qualitatively analyze the peak

196 efficiency and cannot offer quantitative considerations. The peak efficiency points are

197 shown in Fig. 6, which shows that a larger the slenderness ratio of the strut indicates a

198 greater stress at the peak efficiency and a smaller strain. The stress at the peak efficiency

199 point of A lattice structure was largest, and even exceeded the peak stress of A lattice

200 structure. The structure has a large stress during the platform stage, and the energy
201 absorption was also large. The stress increase can exceed the energy increase under a

202 very fast growth trend; therefore, the stress at the peak efficiency point is large. For

203 structures with less stress at the platform stage and that would continue to fall or remain

204 unchanged, the peak efficiency point appears at the point where the stress shows a

205 relative growth trend; therefore, its stress is small.

206 The relationship of the total energy absorption, the slenderness ratio of the strut, and

207 stress is shown in Fig. 8. The total energy absorption and stress correspond to the

208 slenderness ratio of the strut. The total energy absorption is defined as:

209 EAT   Fds (4)

210 The total energy absorption refers to the energy absorption to the densification point.

211 As shown in Fig. 8, the total energy absorption decreases with increasing slenderness

212 ratio. The total energy absorption decreases with increasing stress and the stress

213 increases with increasing slenderness ratio. Although both the slenderness ratio and the

214 structural strength are large, the displacement of the structure when reaching the

215 densification point is small; therefore, the total energy absorption of the structure is

216 small (the force displacement can be seen in Fig. S1 in the supplementary materials).

217 By plotting the energy absorption stress curve for various structures, the most suitable

218 application can be selected. Given a specific energy absorption value, the optimal

219 slenderness ratio is the ratio that yields the smallest stress.
220

221 Fig. 8 Relationship between energy absorption, slenderness, and stress.

222 3.2 Theoretical analysis

223 A single cell has four struts, which experience the same force; therefore, the force of

224 a single strut was analyzed. Fig. 9 shows a force diagram of one strut. As shown in Fig.

225 9, assuming that the displacement of the core in the Z direction under the action of force

226 𝐹 is δ𝑍, the axial force FA and normal force of a single strut are:

1 δ sin ω
227 FA  πd 2 z Ec (5)
4 l
12 Ec Iδz cos ω
228 Fs  (6)
l3

229 where Ec represents the modulus of elasticity of the strut and I represents the moment

230 of the section inertia.

231
232 Fig. 9 Free body diagram of strut of lattice truss structure.

233 The resultant force 𝐹 in the Z direction is:

Ec πd 2δz  2 
2
3d 
234 F =FA sin ω  Fs cos ω  sin ω    cos 2 ω  (7)
4l  4 l  

235 The equivalent stress and strain of the pyramid structure are:

 2  
4F Ec πd 2δz 3d  2
236 σz    sin ω    cos ω  (8)
( 2l cos ω  2t ) 2 l ( 2l cos ω  2t ) 2  4 l  
δ
237 εz = z (9)
l sin ω
238 According to the definition formula of the elastic modulus and substituting Eq. (1) into

239 Eq. (8), the elastic modulus of the pyramid structure is:

2
3 d 
240 Ez =Ec ρ sin 4 ω+ Ec ρ   cos 2 ω sin 2 ω (10)
4 l

241 Because the pyramid structure is a stretch-oriented structure, the elastic modulus of the

242 structure is:

243 Ez =Ec ρ sin 4 ω (11)

244 The strength of the pyramid structure is:

245 σ cr  σ s ρ sin 2 ω (12)

246 where 𝜎𝑠 represents the strength of strut. When the slenderness of the strut is relatively

247 large, buckling damage is likely to occur. The buckling strength of the strut is:

π 2 Em d 2
248 σb  (13)
4l 2

249 where 𝐸𝑚 represents the elastic modulus of the material. When the strut buckles, the

250 structural strength is:

π 2 d 2 Em
251 σ cb  ρ (14)
4l 2

252 Since the relative density of the lattice structure is large, the strut of the lattice

253 structure is short, and the diameter is large. Consequently, the critical load of buckling

254 increases, and the strut tends to be crushed before it buckles. Therefore, Eq. (11) and

255 (12) can be used to calculate the properties of the lattice structure.

256 Table 4 Theoretic and experimental compression performance

Theoretic Experimental Error Theoretic Experimental Error

strength strength elastic modulus elastic modulus
(MPa) (MPa) (MPa) (MPa)
A 4.23 4.08 3.7% 85.3 69.4 22.8%
B 2.81 2.56 9.8% 74 62.6 18.2%
C 1.76 1.62 8.6% 49.9 41.3 20.8%
D 1.26 1.19 5.9% 39.8 32.5 22.5%
E 1.05 0.95 10.5% 34.2 30.8 11%

257 As shown in Table 4, the theoretic compression performance and experimental

258 performance are very close. The maximum difference between the theoretic

259 compressive stress and the experimental compressive stress is 10.5%. The maximum

260 difference between the theoretic elastic modulus and the experimental elastic modulus
261 is 22.8%. These errors remain within acceptable limits. The reason for the error is

262 mainly that the test strut is printed separately and the struts in the structure are not

263 printed in exactly the same way.

264 3.3 Crush strain and strength coefficient

265 The crush strain of the structure refers to the strain, corresponding to the first peak

266 load of the structure on the stress strain curve. In the force analysis of the structural in

267 Fig. 8, the axial force of the strut is:

Fz
268 FA = sin ω (15)
4
269 The axial displacement of the strut is:

FAl Fz l sin ω
270 δA =  (16)
Em A πd 2 Em

271 The vertical displacement of the structure is:

δA Fl
272 δz =  2z (17)
sin ω πd Em

273 Table 5 Crush displacement of structures

Type Experimental displacement Theoretic displacement

(mm) (mm)
A 0.91 0.44
B 1.27 0.8
C 1.72 1.2
D 2.07 1.6
E 1.72 1.78

274 Experimental displacement and theoretic displacement of structures are shown in

275 Table 5. The theoretic displacement and the actual displacement are close, and the error

276 that occurs remains within an acceptable range.

277 The critical slenderness ratio of the strut can be obtained when the critical buckling
278 stress of the strut being is equal to the yield stress of the material. The critical

279 slenderness ratio is:

d 2 σy
280  (18)
l π Em

281 𝜎𝑦 and 𝐸𝑚 are determined by the PLA material. The yield stress 𝜎𝑦 of the material is

282 30.1 MPa, and its elastic modulus 𝐸𝑚 is 1001 MPa. The critical slenderness ratio of

283 the structure is 0.11. Because the diameter of the strut is 6 mm in this study, the critical

284 strut length is 54.5 mm; therefore, when the length of the strut is greater than or equal

285 to 54.5, the structural failure mechanism is the buckling failure of the strut.

286 The non-dimensional peak strength coefficient is defined as:

σ pk
287  (19)
σs ρ

288 The non-dimensional elastic modulus coefficient is defined as:

E
289  e (20)
Ec ρ

290 The theoretic strength coefficient is 0.75, and the theoretic elastic modulus coefficient

291 is 0.56. The strength coefficient and elastic modulus coefficient of structures are shown

292 in Fig. 10. The experimental strength coefficient value is approximately ∑ = 0.75, and

293 the experimental elastic modulus coefficient value is approximately ∑𝑒 = 0.56. The

294 coefficients are both independent of the slenderness ratio of the strut. The theoretically

295 predicted value is in good agreement with the experimentally obtained value.
 1.0 0.8

0.8 =0.75 (theory)

e=0.56 (theory)

Elastic mdulus cofficient

0.6
Strength coefficient

0.6 Plastic yield

0.4

0.4
Buckling
0.2
0.2

0.0 0.0
0.0 0.2 0.4 0.6 0.8 0.1 0.2 0.3 0.4 0.5 0.6
Slenderness ratio Slenderness ratio of strut

(a) (b)

296 Fig. 10 Coefficient of (a) strength and (b) elastic modulus

297 3.4 Post-failure analysis

298 The post failure stress,  P , and strain,  P , of the lattice truss material for

299 compression-shear failure is

300  P   y cos  / cos P (21)

301 and

302  P  1  sin P / sin  (22)

303 where  y is the yield strength and  P is the inclination of the strut in plastic

304 deformation. The post-failure stress-strain relationship is given by

P cos 
305  (23)
y 1  1   P  sin 2 
2

306 This model is applied to predict the post-failure deformation of d6-l20 and d6-l30,

307 which have stable post-failure deformation plateau. As shown in Fig. 11, the model

308 suggests consistent post-failure deformation curves and mean crushing forces (MCFs).
 4 2.5
Test Test
Prediction Prediction
MCF 2.0 MCF
3 Predicted MCF Predicted MCF
Stress (MPa)

Stress (MPa)
1.5
2
1.0

1
0.5

0 0.0
0.00 0.14 0.28 0.42 0.56 0.70 0.00 0.16 0.32 0.48 0.64 0.80
Strain Strain

(a) (b)

309 Fig. 11 Post-failure deformation prediction for (a) B and (b) C lattice structures.

310 3.5 FEM Simulation

311 Finite element method (FEM) was conducted based on the ABAQUS/Standard to

312 predict the deformation process of lattice structures under quasi-static compression.

313 The PLA material can be considered as a homogeneous isotropic material. The elastic

314 modulus, Poisson’s Ratio, and yield stress of the tested material are 1001 MPa, 0.3, and

315 30.1 MPa, respectively. Solid Element C3D4 was selected to stimulate the structure.

316 Tie and general contact were selected between the plates and the model. Applying the

317 displacement load to the model yields the deformation map of the model.

318 Table 6 Actual and simulated compression performances

Simulated Actual Error Simulated Actual Error

peak force peak force displacement displacement
(N) (N) (mm) (mm)
A 5730.4 5001 14.6% 1.04 0.91 14.3%
B 4810.5 4522 6.4% 1.32 1.27 3.9%
C 4639.3 4513.4 2.8% 1.8 1.72 4.7%
D 4352.7 4301.7 1.2% 1.8 2.07 -13%
E 4110.3 4022 2.2% 2.04 1.72 18.6%

319 Actual and simulated compression performances are shown in Table 6. For the

320 peak force, the experimental and simulation results differ by a maximum of 14.6%. For
321 displacement when the peak force is reached, the experiment and simulation results

322 differ by a maximum of 18.6%. The errors in the experiment and the method of printing

323 impact the results.

(a) A lattice structure

(b) B lattice structure

(c) C lattice structure

(d) D lattice structure

(e) E lattice structure

324 Fig. 12 Deformation map of simulated structures.

325 The simulated structural deformation diagram is shown in Fig. 12. The simulated
326 structural failure mode is very close to the actual failure mode. Plastic hinges appear in

327 the simulating failure modes of the structure, and the position of the plastic hinge on

328 the strut is also very similar. However, there is no strut damage in the failure mode of

329 the simulated structures D and E, which is because the damage model is not built well.

330 4 Discussions
Hierarchical thermopl-

Energy absorption per unit mass (J/g)

astichoneycomb [33]
10
Specific energy absorption (J/cm3)

40 HolyPan panel [33]

Hierarchical lattice [32]
Woven textile panel [32]
Lattice structure [34]
30 Woven kagome truss[35]
Polymer lattice truss [36]
Aluminum trusss [25]

20 Hollow lattice structure [27]

1 Aluminum lattice structure [26]
Hollow pyramidal truss [37]
Hollow lattice structure [27]
3d printed structure
Hierarchical lattice structure [28]
Steel square honeycomb [29]
10
3d printed pyramidal structure
Wood-based lattice structure [30]
CFRP lattice structure [31]
Hierarchical lattice [32] 0
0.1 0.01 0.1 1
0.01 0.1 1
Density (g/cm3) Density (g/cm3)

(a) (b)

331 Fig. 13 Energy absorption of different structures of (a) per unit volume and (b) per

332 unit mass

333 The specific energy absorption (SEA) of different structures is shown in Fig. 13. The

334 SEAV of the 3D printed pyramidal structures varies from 0.49 J/cm3 to 2.13 J/cm3, and

335 SEAM varies from 9.67 J/g to 12 J/g. The SEAV of the 3D printed structure is

336 comparable to that of an aluminum lattice structure [26] under the same density, which

337 exceeds that of a carbon fiber structure [31] of the same density. The SEAM of the 3D

338 printed structure is almost identical to that of the aluminum structure with the same

339 density [26], which is higher than the SAEM of many structures.

340 PLA material is a 3D printing material with good plasticity. The PLA pyramid lattice

341 structure prepared by 3D printing produces plastic hinges in the plastic deformation
342 stage. Plastic hinges keep the structural stress at a high level during the plastic

343 deformation phase. The density of the PLA material is small, and the density of the

344 prepared structural core layer is also small. Therefore, the structure prepared by PLA

345 material has better energy absorption than most other lattice materials.

346 5. Conclusions

347 A pyramidal structure was fabricated via 3D printing technology, and its quasi-static

348 axial compression was tested. The following conclusions were obtained:

349 (1) A horizontally printed structure had similar mechanical properties than a

350 vertically printed structure, but the failure mechanism of the horizontally printed

351 structure was strut breakage, while that of the vertically printed structure was that the

352 strut produces plastic hinges.

353 (2) Both the strength and elastic modulus of the structure increased and the specific

354 stiffness decreased with increasing slenderness ratio.

355 (3) The maximum load to mass ratio of the structure was 544 N/g, which exceeded

356 the load mass ratio of carbon fiber and pineapple leaf fiber structures.

357 (4) A higher strut slenderness ratio indicated a stronger stress when the structure

358 reached the densification point.

359 (5) The compression performance obtained via FEM and theoretic analysis were very

360 close to that of experimentally obtained values. This 3D printed pyramidal structure

361 had a good energy absorption capacity.

362
363 Acknowledgement

364 This project was supported by the Fundamental Research Funds for Central

365 Universities (2572016EBJ1), and the National Natural Science Foundation of China

366 (31470581).

367

368 Reference

369 [1] Fan H, Liu F, Long H. Locomotion of vehicles on hinged road mats. Journal of

370 Terramechanics. 2011; 48:47-55.

371 [2] Meng F, Li W, Fan H, Zhou Y. A nonlinear theory for CFRP strengthened aluminum

372 beam. Composite Structures. 2015; 131:574-7.

373 [3] Meng F, Zhang B, Zhao Z, Xu Y, Fan H, Jin F. A novel all-composite blast-resistant

374 door structure with hierarchical stiffeners. Composite Structures. 2016; 148: 113-126.

375 [4] Fan H, Ouyang J, Sun F, Yu P, Kuang N, Hu Y. Lightweight design of CRH wind

376 deflector panels based on woven textile sandwich composites: Criterion and

377 experimental research. Acta Mechanica Solida Sinica. 2016; 29(2): 208-220.

378 [5] Zhang H, Kuang N, Sun F, Fan H, Wang X. Ultra-light CRH wind deflector

379 fabricated by woven lattice sandwich composites. Composites Science and Technology.

380 2014;102:145-51.

381 [6] Xia Z, Wang X, Fan H, Li Y, Jin F. Blast resistance of metallic tube-core sandwich

382 panels. International Journal of Impact Engineering. 2016; 97: 10-28.

383 [7] Fan H, Hong W, Sun F, Xu Y, Jin F. Lateral compression behaviors of thin-walled

384 equilateral triangular tubes. International Journal of Steel Structures. 2015;15(2):433-

385 443.
386 [8] Wang B, Wu L, Jin X, Du S, Sun Y, Ma L. Experimental investigation of 3D

387 sandwich structure with core reinforced by composite columns. Materials and Design.

388 2010; 31(1):158-165.

389 [9] Wang B, Wu L, Sun Y, Du S. Mechanical behavior of the sandwich structures with

390 carbon fiber-reinforced pyramidal lattice truss core. Materials and Design. 2010; 31(5):

391 2659-2663.

392 [10] Wang B, Wu L, Ma L, Feng J. Low-velocity impact characteristics and residual

393 tensile strength of carbon fiber composite lattice core sandwich structures. Composites

394 Part B: Engineering. 2011; 42(4):891-897.

395 [11] Wang B, Zhang G, He Q, Ma L, Wu L, Feng J. Mechanical behavior of carbon

396 fiber reinforced polymer composite sandwich panels with 2-D lattice truss cores.

397 Materials and Design. 2014; 55:591-596.

398 [12] Wang B, Xiong J, Wang X, Ma L, Zhang G, Wu L, Feng J. Energy absorption

399 efficiency in carbon fiber composite laminates under high velocity impact. Materials

400 and Design. 2013; 50:140-148.

401 [13] Zhang G, Wang B, Ma L, Wu L, Pan S, Yang J. Energy absorption and low velocity

402 impact response of polyurethane foam filled pyramidal lattice core sandwich panels.

403 Composite Structures. 2014; 108:304-310.

404 [14] Li S, Qin J, Li C, Feng Y, Zhao X, Hu Y. Optimization and compressive behavior

405 of composite 2-D lattice structure. Mechanics of Advanced Materials and Structures.

406 2018; 1-10.

407 [15] Li, S, Qin, J, Wang, B, Zheng T, Hu, Y. Design and Compressive Behavior of a

408 Photosensitive Resin-Based 2-D Lattice Structure with Variable Cross-Section Core.

409 Polymers. 2019; 11(1):186.

410 [16] Li S, Zheng T, Li, Q, Hu Y, Wang B. Flexural and energy absorption properties of

411 natural-fiber reinforced composites with a novel fabrication technique. Composites

412 Communications. 2019; 16:124-131.

413 [17] Zhang G, Wang B, Ma L, Xiong J, Yang J, Wu L. The residual compressive

414 strength of impact-damaged sandwich structures with pyramidal truss cores. Composite

415 Structures. 2013;105:188-98.

416 [18] Xiong J, Ma L, Wu L, Li M, Vaziri A. Mechanical behavior of sandwich panels

417 with hollow Al–Si tubes core construction. Materials & Design. 2011;32(2):592-7.

418 [19] Li X, Xiong J, Ma L, Wu L, Yan X. Effect of vacuum thermal cycling on the

419 compression and shear performance of composite sandwich structures containing

420 pyramidal truss cores. Composites Science and Technology. 2018;158:67-78.

421 [20] Gümrük R, Mines RAW. Compressive behaviour of stainless steel micro-lattice

422 structures. International Journal of Mechanical Sciences. 2013;68:125-39.

423 [21] Tancogne-Dejean T, Spierings AB, Mohr D. Additively-manufactured metallic

424 micro-lattice materials for high specific energy absorption under static and dynamic

425 loading. Acta Materialia. 2016;116:14-28.

426 [22] Feng L, Wu L, Yu G. An Hourglass truss lattice structure and its mechanical

427 performances. Materials & Design. 2016;99:581-91.

428 [23] Hao M, Hu Y, Wang B, Liu S. Mechanical behavior of natural fiber-based isogrid

429 lattice cylinder. Compos Struct 2017;176:117-23.

430 [24] Zhang C. Manufacturing and Properties of Two-dimension Lattice Composite

431 Structures. Master: National University of Defense Technology, 2008.

432 [25] Bates S, Farrow I, Trask R. 3D printed polyurethane honeycombs for repeated

433 tailored energy absorption. Materials & Design. 2016; 112:172-83.

434 [26] Kooistra G. Compressive behavior of age hardenable tetrahedral lattice truss

435 structures made from aluminium. Acta Materialia. 2004;52(14):4229-37.

436 [27] Queheillalt DT, Wadley HNG. Cellular metal lattices with hollow trusses. Acta

437 Materialia. 2005;53(2):303-13.

438 [28] Sun F, Lai C, Fan H, Fang D. Crushing mechanism of hierarchical lattice structure.

439 Mechanics of Materials. 2016;97:164-83.

440 [29] Côté F, Deshpande VS, Fleck NA, Evans AG. The out-of-plane compressive

441 behavior of metallic honeycombs. Materials Science and Engineering: A. 2004;380(1-

442 2):272-80.

443 [30] Jin M. Mechanical Behaviors of Wood-based Lattice Truss Core Sandwich

444 Structure. Master: Northeast Forestry University, 2015.

445 [31] Xiong J. Design and Mechanical Behavior of Lightweight Composite Innovative

446 Lattice Truss Structures. Doctor: Harbin Institute of Technology, 2013.

447 [32] Zheng J, Zhao L, Fan H. Energy absorption mechanisms of hierarchical woven

448 lattice composites. Composites Part B: Engineering. 2012;43(3):1516-22.

449 [33] Liu H, Chen L, Du B, et al. Flatwise compression property of hierarchical

450 thermoplastic composite square lattice. Compos Struct. 2019;210:118-33

451 [34] McKown S, Shen Y, Brookes WK, Sutcliffe CJ, Cantwell WJ, Langdon GS, et al.

452 The quasi-static and blast loading response of lattice structures. International Journal of

453 Impact Engineering. 2008;35(8):795-810.

454 [35] Lee BK, Kang K-J. A parametric study on compressive characteristics of Wire-

455 woven bulk Kagome truss cores. Composite Structures. 2010;92(2):445-53.

456 [36] Jacobsen AJ, Barvosa-Carter W, Nutt S. Compression behavior of micro-scale

457 truss structures formed from self-propagating polymer waveguides. Acta Materialia.

458 2007;55(20):6724-33.

459 [37] Queheillalt DT, Wadley HNG. Pyramidal lattice truss structures with hollow

460 trusses. Materials Science and Engineering: A. 2005;397(1-2):132-7.

461 All authors have no conflicts of interest to this work.

462