Materials and Design 55 (2014) 591–596

Contents lists available at ScienceDirect

Materials and Design

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

Technical Report

Mechanical behavior of carbon fiber reinforced polymer compositesandwich panels with 2-D lattice truss cores

0261-3069/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.matdes.2013.10.025

⇑ Corresponding author at: Center for Composite Materials, Harbin Institute ofTechnology, Harbin 150001, PR China. Tel./fax: +86 451 86402376.

E-mail address: wangbing86@hit.edu.cn (B. Wang).

Bing Wang a,c,⇑, Guoqi Zhang a, Qilin He b, Li Ma a, Linzhi Wu a, Jicai Feng c

a Center for Composite Materials, Harbin Institute of Technology, Harbin 150001, PR Chinab Beijing Institute of Astronautical Systems Engineering, Beijing 100076, PR Chinac School of Materials Science and Engineering, Harbin Institute of Technology, Harbin 150001, PR China

a r t i c l e i n f o

Article history:Received 11 October 2012Accepted 10 October 2013Available online 21 October 2013

a b s t r a c t

Composite sandwich structures with lattice truss cores are attracting more and more attention due totheir superior specific strength/stiffness and multi-functional applications. In the present study, the car-bon fiber reinforced polymer (CFRP) composite sandwich panels with 2-D lattice truss core are manufac-tured based on the hot-pressing method using unidirectional carbon/epoxy prepregs. The facesheets areinterconnected with lattice truss members by means of that both ends of the lattice truss members areembedded into the facesheets, without the bonding procedure commonly adopted by sandwich panels.The mechanical properties of the 2-D lattice truss sandwich panels are investigated under out-of-planecompression, shear and three-point bending tests. Delamination of the facesheets is observed in shearand bending tests while node failure mode does not occur. The tests demonstrate that delamination ofthe facesheet is the primary failure mode of this sandwich structure other than the debonding betweenthe facesheets and core for conventional sandwiches.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Sandwich structures consisting of low density core and solid facesheets are widely used in naval and aeronautical applications due totheir high stiffness/strength-to-weight ratio [1]. The conventionalcores are honeycombs [2,3] and foams [4,5]. These sandwich panelshave superior mechanical properties to their solid plate counterpart;however, these closed-cell configurations limit their multi-func-tional applications [6,7]. Recently, lattice truss cores have begun tobe explored as a candidate core material because of their superiorspecific strength/stiffness and the large interconnected void space.The stiffness and strength of lattice materials scale linearly withthe relative density q, which results that the lattice materials, atlow relative densities, can therefore be more than an order of mag-nitude stronger and stiffer than equivalent mass per unit volumefoams made from the same parent material. In order to explore themechanical properties of lattices, the corresponding techniques tomanufacture such sandwich structures have rapidly expanded dur-ing the past years. The investment casting, metal weaving and per-forated wrought mental sheet folding methods have emerged tofabricate metal lattice structures [8–10]. For the design calculations,the mechanism maps based on beam theory have been given by

Desphande and Fleck [11]. Wicks and Hutchinson have optimallydesigned such panels subject to prescribed combinations of bendingand transverse shear loads [12]. Besides, the quasi-static /dynamicmechanical response of the lattice sandwich structures has beeninvestigated by numerous scholars [13–15]. The measurementshave shown that the lattice sandwich structures have desirablemechanical properties such as specific strength/stiffness and impactenergy absorption. In addition, such materials are also expected tofind applications in lightweight, compact structural heat exchangers[16]. Their open-cell configuration allows heat exchange along thepanel core, making them as attractive candidates for developmentof multi-function systems. Moreover, the Kagome core sandwichstructures have been investigated as shape morphing structure bySymons et al. [17,18]. As we know, the behavior of sandwich struc-tures primarily depends on the topology and the parent material.The optimized lattice topology and superior parent material proper-ties can be combined to create new engineering materials, which ex-pand the material property space. In view of this, the CFRP which canprovide high uniaxial specific strength is used as the parent materialof the lattice core sandwich structures.

Fan et al. [19,20] have manufactured a carbon fiber-reinforcedthree-dimensional lattice core sandwich panel using an intertwin-ing method. Finnegan et al. [21] manufactured CFRP pyramidaltruss cores by slot-fitting method. In order to improve the shearperformance of pyramidal lattice core sandwich structure, YuguoSun and Liang Gao presented an improved pyramidal lattice trusscore sandwich structure by introducing a series of parallel

Fig. 2. Photograph of the molds.

592 B. Wang et al. / Materials and Design 55 (2014) 591–596

distributed cross-bars to core members [22]. Compared with 3-Dlattice sandwich structures, the preparation process of 2-D latticesandwich structures is easier [23]. In contrast, the longitudinalshear performance of 2-D lattice sandwich structures is better,which may be used under shear loading in one direction. Thus,the aim of the present work is to develop a new method to manu-facture the composite 2-D lattice sandwich structures. In addition,the mechanical properties of 2-D lattice truss core sandwich panelsare tested under the out-of-plane compression, shear and three-point bending loadings. Finally, the failure mechanism andmechanical properties of the new structure are also discussed.

2. Fabrication

The unit cell of 2-D lattice truss core is shown in Fig. 1. The rel-ative density of the 2-D lattice truss cores depends on the diameterof the truss member d, inclination angle x, length l of truss mem-ber, width b of the unit cell, and the distance t between the adja-cent truss members, which is as following

�q ¼ pd2

4b sin xðl cos xþ tÞ ð1Þ

The geometrical parameters are taken as d = 3 mm, l = 17 mm,b = 20 mm, t = 3 mm, and x = 45� in the present study, so the rela-tive density of the truss core, q, is 3.33%.

The composite sandwich panels with 2-D lattice truss cores aremanufactured based on the hot-pressing method. In order to con-figure the lattice core topology, many cuboid flat steel plates withsemicircular grooves on the side face are used as molds, as shownin Fig. 2. In the manufacturing process, the molds are assembledafter the surface is cleaned with acetone and coated with the moldrelease agent. Carbon fiber prepregs are cut to the required size,and then rolled into rods as truss members with fibers along thedirection of truss members, whose radius coincides with that ofthe semicircular grooves on the side face of molds. Then, the com-posite rods are inserted into the holes of the assembled mold withabout 0.5 cm outside the molds, as depicted in Fig. 3. After that, theunidirectional carbon/epoxy (T700/TDE85) prepregs are layered onthe top surface of molds to configure the facesheets, with thestacking sequence [00/450/�450/00]s, and both ends of truss mem-bers are dispersed and gradually embedded into the eight layers offacesheets. The preforms are then cured at 170 �C under pressureof 0.7 MPa for 3 h on the hot press. Finally, removing the moldsand cutting out the panels to the required dimensions, we getthe corresponding samples, as shown in Fig. 4.

3. Experiments

3.1. Out-of-plane compression of 2-D truss core sandwich panels

Compression tests are performed with panels of73 � 60 � 15 mm to determine their compressive stiffness and

Fig. 1. Schematic of 2-D lattice truss core unit cell.

strength in the out-of-plane direction in accordance with ASTM:C365/C365 M-11a A screw driven testing machine (Instron 5569)is used to test samples at a rate of 0.5 mm/min. At least three testsare conducted to confirm the repeatability of the measurements.

3.2. Shear behavior of 2-D truss core sandwich panels

As we know, sandwich structures are usually used in situationswhere they are subjected to significant bending loads. The appliedbending moment is balanced by the bending stress in the face-sheets. On the other hand, the traverse shear load is supportedmainly by the core. Typically, cores are the weakest part of sand-wich structures. Based on the work done by Deshpande and Fleck[11], we know that this 2-D lattice is anisotropic in bending andshearing loads. Along the orthogonal direction, the shear/bendingstrength must be much smaller. Thus, the core should be optimallydesigned to select the appropriate orientation to achieve a highload carrying capacity, and the case that the core along the orthog-onal direction bears the load is not desirable. Therefore, it isinstructive to explore the properties of this core along the wavedirection. In this case, the shear or bending tests for this sandwichstructure is also along this direction.

Shear tests are performed in accordance with ASTM: C273/C273 M-11. The test fixture used to obtain shear stress/strain ofthe core is shown in Fig. 5. The specimens are rigidly bonded tosteel plates by adhesive, and the dimensions of the specimensare 150 � 60 � 15 mm.

3.3. Bending behavior of 2-D truss core sandwich panels

Three-point bending tests are conducted using the Instron 5569universal testing machine according to ASTM C393/C393 M-11e1The specimens with the dimension of 200 � 60 � 15 mm, and thedistance between the supports is 180 mm, as shown in Fig. 6.The column radius of the support and loading pins is 10 mm.

4. Results

The typical compressive stress–strain response is shown inFig. 7. Following an initial linear response, the compressive stressapproaches the peak value. After that, the zigzag fluctuation ap-pears due to the splitting occurring at the ends of truss members,and then the load suddenly decreases because of the buckling ofthe fibers, as shown in Fig. 8. Based on the compressive stress–strain curve, the compressive strength and the Young’s modulus,2.64 and 210.2 MPa, respectively, are obtained for such sandwichstructure. Considering the composite density of 1560 kg/m3, thespecific strength (compressive strength/ density) of this 2-D latticetruss core is 51.28 � 103 N m/kg, which is compared well withother lightweight cores [24].

The shear stress–strain response is shown in Fig. 9. During theinitial stage of loading, the response is fairly linear. Thus, basedon this linear stage, we can obtain the shear modulus of94.8 MPa. Thereafter, a small fluctuation emerges when the shearstress approaches 1.1 MPa while the load continues to increaseas the shear stress arrives at the maximum value, 1.38 MPa. Inthe tests, delamination of facesheets is observed which is due tothe defects such as fiber tangle or voids generated when the trussmembers are embedded into the facesheets, as shown in Fig. 10.

Fig. 3. Fabrication of the CFRP 2-D lattice truss core sandwich structures. (a) The unidirectional carbon/epoxy prepregs; (b) the truss member rolled by the prepregs; (c) theassembled molds.

Fig. 4. The sandwich panel with carbon fiber-reinforced 2-D lattice truss core.

Fig. 5. Schematic of the tension shear test fixture.

Fig. 6. The three-point bending test.

Fig. 7. The stress–strain curve of out-of-plane compression test.

B. Wang et al. / Materials and Design 55 (2014) 591–596 593

However, the debonding failure mode between the core and face-sheets which is prone to occur for sandwich panels under sheartests is not observed [22]. This delamination failure in tests implies

that the future work will focus on improving the fabrication pro-cess to increase the interlayer property.

The load–displacement curve for the three-bending tests de-picted in Fig. 11 display two representative regions, referred toas the elastic region A and the delamination region B, C and D,

Fig. 8. The failure mode of the 2-D lattice truss core sandwich panels during out-of-plane compression test.

Fig. 9. The shear stress–strain curve for 2-D truss core sandwich panels.

(a) (b)

Fig. 10. Photographs of the experimental shear test. (a) Initial loading; (b)delamination of facesheets.

Fig. 11. The load–displacement curve of 2-D lattice truss core sandwich panelduring the three-point bending test.

594 B. Wang et al. / Materials and Design 55 (2014) 591–596

respectively. The calculated bending rigidity is 6.32 kN/m2 basedon the elastic region while the applied maximum load is 1887 Nin the test. As we know, the applied bending moment is supportedmainly by the facesheets and the top facesheet mainly carries thecompressive load while the bottom facesheet mainly bears the ten-sile load. On the other hand, the traverse shear load is sustainedmainly by the core. With the applied load increase, the top face-sheet tends to buckle under compressive load which also inducesdelamination. The delamination of the top facesheet leads to a sud-den drop of the load, as shown in Fig. 11, while the facesheets andthe truss cores are still connected together. Node failure or truss

member failure is not observed during the bending test. The threepoints in the load–displacement curve indicate that the interfacialdelamination occurs for three times, as shown in Fig. 12(b–d).

5. Discussion

Analytical expressions for the compressive modulus E of the2-D lattice truss cores are given by first analyzing the elastic defor-mations of a single strut of the 2-D lattice truss cores and thenextending the results to evaluate the effective properties of thewhole core. Considering a strut of length l and circle cross-sectionof diameter d as shown in Fig. 13, symmetry conditions indicatethat the top end of the strut is only free to move along the x3-direc-tion. For an imposed displacement d in the x3-direction, the axialand shear forces in the strut are given by elementary beam theoryas

FA ¼14pd2Es

d sinxl

ð2Þ

Fs ¼3EsId cos x

l3 ð3Þ

where I ¼ pd4

64 is the second moment of area of the strut cross-sec-tion. The total applied force F in the x3-direction then follows as

F ¼ FA sinxþ FS cos x ¼ Espd2d4l

sin2 xþ 316

dl

� �2

cos2 x

" #ð4Þ

Then, the through-thickness stress r and strain e applied to the 2-Dcores are related to the force F and displacement d via:

r ¼ 2Fb � ð2l cos xþ 2tÞ

¼ Espd2d4blðl cos xþ tÞ sin2 xþ 3

16dl

� �2

cos2 x

" #ð5Þ

and

e ¼ dl sin x

ð6Þ

respectively. Combining Eqs. (4)–(6) then gives the effectiveYoung’s modulus of the 2-D cores:

EC

ES¼ pd2 sin x

4bðl cos xþ tÞ2sin2 xþ 3

16dl

� �2

cos2 x

" #ð7Þ

Ec ¼ ES �q sin4 xþ 316

ES �qdl

� �2

sin2 x cos2 x ð8Þ

Fig. 13. (a) Sketch of the deformation of a single strut of 2-D lattice truss core underout-of-plane compression. (b) The free-body diagram of a strut loaded in acombination of compression and shear.

(b)

(e)

(d)

(a)

(c)

Fig. 12. Deformation history of composite sandwich structure with inclined lattice truss cores in the three-point bending test. (a) Elastic deformation; (b) splitting ofcomposite columns; (c–e) delamination of facesheet.

B. Wang et al. / Materials and Design 55 (2014) 591–596 595

where �q ¼ pd2

4b sinxðl cos xþtÞ is the relative density.

The first and second terms in Eq. (8) represent the contributionsto the stiffness of the 2-D cores due to the stretching and bendingof the struts, respectively. Compared with the first term, the sec-ond term is a small quantity and can be neglected.

Ec ¼ ES �q sin4 x ð9Þ

where ES = 33.2 GPa is the compressive Young’s modulus of unidi-rectional carbon fiber composite truss member measured by meansof a single truss member under compressive load [22]. However,due to the intricate failure mode as mentioned earlier, analyticalexpressions for the prediction of compressive strength can not bederived.

According to Eq. (9), the effective compressive Young’s modulusof the 2-D truss cores sandwich panel is 276.39 MPa, which isabout 31.2% larger than the measured Young’s modulus. It is be-cause that there may be many micro-defects in the truss memberscaused by the less pressure along the radial direction of fiber rodsin the molding of composite structures.

The shear stiffness of lattice truss cores is given by Deshpande[13] as

Gs ¼18

Es � sin2 2x � q ð10Þ

where Es = 33.2 GPa is the compressive Young’s modulus of the car-bon fiber composite truss members measured on the single trussmember under compressive test. According to Eq. (10), the effectiveshear modulus of 2-D lattice truss cores is 138.2 MPa, about 47.87%larger than the tests. This discrepancy may be attributed to thesample misalignment as well as the small gaps in the fixture thatmay introduce excessive relative displacement during the tests.The large discrepancy can also be found by Deshpande and Fleck[13]. Although debonding between the cores and facesheet whichalways occurred for sandwiches under shear load did not emergeduring the parent investigation, the shear strength was not desir-able compared with the work done by Tochukwu George et al.[25]. It is that the lattice truss members were embedded in the cru-ciform shaped slots around the facesheets in their investigation.

596 B. Wang et al. / Materials and Design 55 (2014) 591–596

During the tests, only buckling and delamination of truss membersoccurred, which can efficiently take advantage of the material prop-erty of composites. For the shear tests of metal lattice truss mem-bers, the fracture, yielding and buckling of truss members werethe main failure modes [8,13]. Therefore, the shear performanceof this composite lattice truss cores was not fully presented owingto the delamination of facesheets. The shear performance of suchsandwich structures can be better supposing the delamination offacesheets was restrained.

For the bending tests, the bending rigidity is desirable consider-ing the mass of such sandwich structures. However, also thedelamination failure mode of facesheets governed the subsequentperformance of such sandwich structures. Therefore, the futurework will focus on improving the interface property of facesheetsto prevent the early occurring of delamination.

6. Conclusions

In the present study, CFRP 2-D truss core sandwich panels aremanufactured based on the hot-pressing method. The 2-D trusscores and the facesheets are interconnected together by means ofboth ends of truss members embedded into the facesheets duringthe prepreg laying process. Out-of-plane compression, shear andthree-point bending tests are carried out to study the mechanicalbehaviors of the 2-D truss core sandwich panels.

According to the experiments, the mechanical behaviors of theCFRP 2-D truss core sandwich panels can be concluded as follows:Delamination of facesheets is the main weakness this composite2-D truss core sandwich panel; The shear strength of the 2-Dlattice truss core is higher than the interface strength of facesheets;The delamination always occurs around to the area where the trussmembers embedded into the facesheets. Therefore, the futurework will focus on improving the fabrication process to increasethe interface property.

Acknowledgements

The present work is supported by National Science Foundationof China under Grant No. 11202059, Natural Science Foundation ofHeilongjiang Province (No. A201204), the Major State Basic Re-search Development Program of China (973 Program) under GrantNo. 2011CB600303, Key Laboratory Opening Funding of AdvancedComposites in Special Environment(2011) and the fundamentalresearch funds for the central Universities (Grant No.HIT.NSRIF.2010069).

References

[1] Gibson LJ, Ashby MF. Cellular solids: structure and properties. CambridgeUniversity Press; 1997.

[2] Papka SD, Kyriakides S. In-plane compressive response and crushing ofhoneycomb. J Mech Phys Solids 1994;42:1499–532.

[3] Klinworth JW, St ronge WJ. Plane punch indentation of a ductile honeycomb.Int J Mech Sci 1989;3:359–78.

[4] Banhart J, Stanzick H, Helfen L, Baumbach T. Metal foam evolution studied bysynchrotron radioscopy. Appl Phys Lett 2001;178:1152–4.

[5] Evans AG, Hutchinson JW, Fleck NA, et al. The topological multifunctionalcellular metals. Prog Mater Sci 2001;46:309–27.

[6] Evans AG, Hutchinson JW, Ashby MF. Multifunctionality of cellular metalsystems. Prog Mater Sci 1999;43:171–221.

[7] Wadley HNG, Fleck NA, Evans AG. Fabrication and structural performance ofperiodic cellular metal. Composi Sci Technol 2003;63:2331–43.

[8] Yuki Sugimura. Mechanical response of single-layer tetrahedral trusses undershear loading. Mech Mater 2004;36:715–21.

[9] Lim Ji-Hyun, Kang Ki-Ju. Mechanical behavior of sandwich panels withtetrahedral and Kagome truss cores fabrication from wires. Int J Solids Struct2006;43:5228–46.

[10] Kooistra Gregory W, Deshpande Vikram S, Wadley Haydn NG. Compressivebehavior of age hardenable tetrahedral lattice truss structures made fromaluminum. Acta Mater 2004;52:4229–37.

[11] Deshpande VS, Fleck NA. Collapse of truss core sandwich beams in 3-pointbending. Int J Solids Struct 2001;38:6275–305.

[12] Nathan Wicks, Hutchinson John W. Optimal truss plates. Int J Solids Struct2001;38:5165–83.

[13] Deshpande VS, Fleck NA, Ashby MF. Effective properties of the octet-trusslattice material. J Mech Phys Solids 2001;49:1747–69.

[14] Yungwirth Christian J, Radford Darren D, Aronson Mark, Wadley Haydn NG.Experiment assessment of the ballistic response of composite pyramidallattice truss structures. Composite: Part B 2008;39:556–69.

[15] Cote F, Fleck NA, Deshpande VS. Fatigue performance of sandwich beams witha pyramidal core. Int J Fatigue 2007;29:1402–12.

[16] Lu TJ, Valdevit L, Evans AG. Active cooling by metallic sandwich structureswith periodic cores. Prog Mater Sci 2005;50:789–815.

[17] Symons DD, Hutchinson RG, Fleck NA. Actuation of the Kagome Double-LayerGrid. Part 1: Prediction of performance of the perfect structure. J Mech PhysSolids 2005;53:1855–74.

[18] Symons DD, Shieh J, Fleck NA. Actuation of the Kagome Double-Layer Grid.Part 2: Effect of imperfections on the measured and predicted actuationstiffness. J Mech Phys Solids 2005;53:1875–91.

[19] Fan HL, Yang W, Yan Y, Fu Q, Fang DN, Zhuang Z. Design and manufacturing ofa composite lattice structure reinforced by continuous carbon fibers. StinghuaSci Technol 2006;11:515–22.

[20] Fan HL, Meng FH, Yang W. Sandwich panels with kagome lattice coresreinforced by carbon fibers. Compos Struct 2007;81:533–9.

[21] Finnegan K, Koositra G, Wadley HNG, Dshpande VS. The compressive responseof carbon fiber composite pyramidal truss sandwich cores. Int J Mat Res2007;98:1264–72.

[22] Wang Bing, Linzhi Wu, Jin Xin, Shanyi Du, Sun Yuguo, Ma Li. Experimentalinvestigation of 3D sandwich structure with core reinforced by compositestruts. Mater Des 2010;31:158–65.

[23] Sun Yuguo, Gao Liang. Structural responses of all-composite improved-pyramidal truss sandwich cores. Mater Des 2013;43:50–8.

[24] Jingjing Zheng, Long Zhao, Hualin Fan. Energy absorption mechanisms ofhierarchical woven lattice composites. Composite: Part B 2012;43:1516–22.

[25] Tochukwu George, Deshpande SVikram, Wadley Haydn NG. Mechanicalresponse of carbon fiber composite sandwich panels with pyramidal trusscores. Composite: Part A 2013;47:31–40.