The Bending Responses of Sandwich Panels with Aluminium Honeycomb Core and CFRP Skins Used in Electric Vehicle Body

-e aim of this paper was to investigate bending responses of sandwich panels with aluminium honeycomb core and carbon fibrereinforced plastic (CFRP) skins used in electric vehicle body subjected to quasistatic bending. -e typical load-displacement curves, failure modes, and energy absorption are studied. -e effects of fibre direction, stacking sequence, layer thickness, and loading velocity on the crashworthiness characteristics are discussed. -e finite element analysis (FEA) results are compared with experimental measurements. It is observed that there are good agreements between the FEA and experimental results. Numerical simulations and experiment predict that the honeycomb sandwich panels with ±30° and ±45° fibre direction, asymmetrical stacking sequence (45°/−45°/45°/−45°), thicker panels (0.2mm∼0.4mm), and smaller loading velocity (5mm/min∼30mm/min) have better crashworthiness performance. -e FEA prediction is also helpful in understanding the initiation and propagation of cracks within the honeycomb sandwich panels.

e honeycomb sandwich structures are widely used in several engineering applications in the transport industry.
ere have been extensive studies regarding the impact and bending behaviour of aluminium honeycomb panels with and without external skins.Liu et al. [2] explored the crashworthiness of CFRP square tubes filled with aluminium honeycomb subjected to quasistatic axial crushing.By comparison, the peak load and absorbed energy of the filled tubes increased by more than 10% as compared with those of the bare CFRP tubes, ranging approximately from 12.41% to 27.22% and from 10.49% to 21.83%, respectively.
For three-point bending (TPB), energy absorption (EA) and specific energy absorption (SEA) were found by Sun et al. [17] to be largely influenced by the structural parameters in the honeycomb core, but not much by the skin thickness.
ey also compared the crashworthiness of empty circular CFRP with CFRP/aluminium/steel tubes filled with aluminium foam or aluminium honeycomb under axial quasistatic crushing.With the increase in R of CFRP tubes, both the energy absorption and loading capacities increase, with specific energy absorption (SEA) increasing from 48.60 J/g to 60.37 J/g. e SEAs of CFRP tubes filled with honeycomb were slightly lower than the empty counterparts but far better than those of all metal specimens [18].Hazizan et al. [19] investigated the low-velocity impact response of two glass fibre/epoxy aluminium honeycomb sandwich structures.Crupi et al. [20] investigated the mechanical behaviour under bending and impact loading of AHS panel reinforced by GFRP outer skins and a comparison with the AHS panels (without GFRP skins) was done.e experimental tests demonstrated that the amount of energy absorption of the honeycomb sandwiches was highly improved, reinforcing them by means of GFRP outer skins.Hussein et al. [21] studied the axial crushing behaviour of aluminium honeycomb-filled square carbon fibre-reinforced plastic (CFRP) tubes.e results showed that the energy absorption (EA) of aluminium honeycomb-filled CFRP tubes increased from 20% to 36% more than the energy absorption of hollow CFRP tubes at different crushing velocities.Liu and Wu [22] investigated the lateral planar crushing and bending responses of CFRP square tubes filled with aluminium honeycomb.
e results of lateral threepoint bending tests showed that the peak load, EA, and SEA of honeycomb-filled CFRP tubes increased by 17%, 32%, and 0.9%, respectively, compared with the CFRP hollow tubes.
Although the honeycomb sandwich panels have been widely studied, there have been limited studies on the honeycomb sandwich panels with CFRP skins.Furthermore, it is a new attempt that the lightweight sandwich panels with aluminium honeycomb core and CFRP skins are applied to the body panel in our developed lightweight CFRP electric vehicle, which would be subject to lateral bending load.erefore, their application requires a better understanding of the bending response of such lightweight composites, and most of the studies do not consider the material design of CFRP to improve bending performance.

e CAD Model of Sandwich Panels with Aluminium Honeycomb Core and CFRP Skins.
e CAD model of sandwich panels with aluminium honeycomb core and CFRP skins is shown in Figure 1.It is made of two parts, known as upper and lower CFRP skins and middle aluminium honeycomb core.
e CFRP skins are bonded to aluminium honeycomb core by DG-4 epoxy adhesive.e adhesive can be cured at room temperature and withstand temperatures from −60 °C to +120 °C.Moreover, the bonding process is simple, convenient to use and fast curing.e length and width of CFRP skins are 200 mm and 30 mm, respectively, and the thickness of CFRP skin is determined by layer thickness.
e aluminium honeycomb core is made from the aluminium alloy with the cell thickness of 0.07 mm and side length of 4 mm.e height of aluminium honeycomb is 8.4 mm.

e Material Property of Sandwich Panels with Aluminium Honeycomb
Core and CFRP Skins.Upper and lower CFRP skins are full carbon fibre structures, and their material is T700/FAW100.Table 1 lists the material properties of CFRP T700/FAW100.Nine material constants in Table 1 will be used in the finite element analysis.
e middle aluminium honeycomb core has an isotropic material property, and its material is 3003 aluminium alloy.
e material properties are shown in Table 2.

Lay-Up Schemes of Sandwich Panels with Aluminium
Honeycomb Core and CFRP Skins.In order to analyze the effect of stacking sequence, fibre direction, and layer thickness on the crashworthiness of sandwich panels, the specimens in this study are divided into three groups (Group A/Group B/Group C) as listed in Table 3.A total of eight different lay-up schemes (A/B/C/D/E/F/G) are considered.e fibre directions with 15 °, 30 °, 45 °, 60 °, and 75 °are mainly considered.
e purpose of Group A is to compare and analyze the effect of fibre direction on the crashworthiness of sandwich panels.e purpose of Group B is to compare and analyze the effect of stacking sequence on the crashworthiness of sandwich panels.e purpose of Group C is to compare and analyze the effect of layer thickness on the crashworthiness of sandwich panels.

Testing Methods.
Different from the bending mechanical properties of conventional solid metal materials, the quasistatic three-point bending test is based on the GB/T1449-2005 Testing Standard "Test method for bending properties of carbon fiber reinforced plastics."As shown in Figure 2, the sandwich panel with aluminium honeycomb core and CFRP skins is placed parallel to the center position of the support seat of the universal testing machine.e loading roller moves downward at a constant rate of V until the specimen is broken.e load-displacement data are recorded during loading, and the experimental results of different cases are compared and analyzed.

Methods of Analyses
3.1.Orthotropic Material Property.Under the Cartesian coordinate 1-2-3, the constitutive equation of the orthotropic material such as carbon fibre-reinforced plastic is [22] as follows: Aluminium honeycomb core CFRP panel where σ { }, ε { }, and [C] are the stress, strain, and stiffness matrix, respectively.e compliance matrix [C] is the inverse of [C] as follows: where E i , G ij , and ] ij are Young's modulus, shear modulus, and Poisson's ratio, respectively.e symmetrical matrix [S] in (3) has nine independent material constants for the orthotropic material.
In the carbon/epoxy composite laminate, each ply has the orthotropic material property.e fibre directions of each ply can be different for practical applications.e bicycle frame is a full carbon fibre structure, and its material is T700/FAW100.Table 1 lists the material properties of CFRP T700/FAW100.Nine material constants in Table 1 will be used in the finite element analysis.

Finite Element Analysis (FEA).
In this study, finite element analysis (FEA) is performed using ABAQUS software including model and mesh, load, boundary condition, solving, and postprocessing.Figure 3 shows the finite element model of sandwich panels with aluminium honeycomb core and CFRP skins.It is supported by a rigid support seat at the bottom.e rigid loading roller moving at a constant velocity was built to represent the loading for three-point bending.
e sandwich panels necessitate a progressive failure model for solid elements using a modified Chang-Chang failure criterion, which is capable of predicting tensile and compressive fibre failure, as well as tensile and compressive matrix failure.
e aluminium honeycomb is modelled by solid elements with isotropic material property.
ere are three types of contact defined between the loading roller, CFRP skins, aluminium honeycomb core, and support seat, namely, automatic single surface, tied surface to surface, and automatic nodes to surface (as illustrated in Figure 3).e loading speed and boundary conditions were prescribed consistently with the experimental setup.
e initial peak force (F max ) can be obtained directly from the load-displacement curve which separates the    Advances in Materials Science and Engineering loading process into the initial elastic bending stage and bending collapse stage.e value of average crushing force (F avg ) is defined mathematically as where d is the collapse distance and F(δ) is the instantaneous crush force.Crush force efficiency (CFE), defined as the ratio of the mean crushing force (MCF) to the initial peak force (F max ), is used to measure the uniformity of crushing force as e higher the value of CFE, the better the crashworthiness performance.
e energy absorption (EA) is obtained by integrating the load-displacement curve during the loading process as e higher the energy absorption (EA), the better the crashworthiness.To account for the effect of mass (weight), specific energy absorption (SEA), defined as is frequently used as one of the most critical crashworthiness criteria.

Experimental Procedures
4.1.Specimen Preparation.For electric vehicles, the most effective way to increase the extension mileage is to reduce weight.e traditional materials of vehicle body are highstrength steel or aluminium alloys, which are heavy and do not meet the growing extensive mileage requirements of electric vehicles.e sandwich panels with aluminium honeycomb core and carbon fibre-reinforced plastic (CFRP) skins can be reasonably designed to make electric vehicle body with the advantages of lightweight and better energy absorption performance.As shown in Figure 4, the filled structures are applied to the typical structural elements such as the body panels in our developed lightweight CFRP electric vehicle, which would be subject to lateral bending load. is work aimed at investigating the lateral bending capability of sandwich panels with aluminium honeycomb core and CFRP skins.Figure 5 shows the specimens of sandwich panels with eight different lay-up schemes.In order to analyze the effect of stacking sequence, fibre direction, and layer thickness on the crashworthiness of sandwich panels, a total of eight different lay-up schemes (A/B/C/D/E/F/G/H) are considered.To ensure the accuracy of the experimental results, each case is repeated three times   under the same testing conditions.In addition, in order to analyze the effect of loading velocity on the crashworthiness of sandwich panels, the loading experiments are carried out at different loading rates for Case E4, Case E5, and Case E6, respectively, and the specimens of Case E4, Case E5, and Case E6 are shown in Figure 6.

Quasistatic ree-Point Bending
Test. e quasistatic three-point bending tests are carried out to study the bending behaviours of these different sandwich panels.All the tests are performed at room temperature in the electronic universal testing machine DNS-100 with a loading capacity of 100 kN.As shown in Figure 7, the speed of the loading roller is from 5 mm/min to 30 mm/min.When the specimens of three groups are tested, the loading velocity of the loading roller is set as 5 mm/min.
In order to analyze the effect of loading velocity on the crashworthiness of sandwich panels, the quasistatic loading test of the specimens of Case E4, Case E5, and Case E6 is carried out, and the loading rates are 5 mm/min, 15 mm/min, and 30 mm/min, respectively.e final bending displacement is set as 25 mm to ensure complete damage for each specimen.
e bending load and its corresponding displacement are recorded by a data acquisition system, where the load-displacement curves can be plotted by the system.e deformation behaviours of the specimens are photographed during the quasistatic bending process.e specimens after quasistatic bending failure are shown in Figure 8.

Load-Displacement Curves.
e load-displacement curves of the sandwich panels with aluminium honeycomb core and CFRP skins under the three-point bending test are shown in Figure 9. e curves can be divided into two stages, namely, the initial elastic bending stage and bending collapse stage [23].Taking the specimens of Case D (D1, D2, and D3) as an example, in the initial elastic bending stage, the bending load kept increasing until it reaches the first peak (the average value is 435 KN).Following the first peak, extensive microfracture was observed at the corner of the upper panel in contact with the loading roller; the fracture spread quickly in a direction perpendicular to the CFRP skin, causing several drops in the bending load curve within the bending collapse stage as shown in Figure 9; and the numerically predicted peak load matched the experimental measurement with as small error given except for the specimen of Case D1.It is seen that the duration of the elastic deformation stage is very short and the collapse stage is the main energy absorption stage during the bending.

Failure Modes.
e failure progress of the sandwich panels with aluminium honeycomb core and CFRP skins subjected to three-point bending is shown in Figure 10.It shows a complex failure mode, including the plastic hinges, buckling, indentation, core failure, and shear interaction.It is evident that the microfracture initiated at the corner of the top wall in contact with the loading roller  Advances in Materials Science and Engineering because of the stress concentration.e cracks spread from the top wall of the filled CFRP panels to the aluminium honeycomb.In the bending process, the hexagonal cell layers of aluminium honeycomb are squeezed each other on the upper surface (compression surface) and the cell layers stretched on the bottom surface (tensile surface), leading to a fan shape.But the bending deformation was not enough to induce the cells of the aluminium honeycomb debonding from the adjacent adhesive layers.erefore, the damage of degumming between aluminium honeycomb cell layers was not found in these three-point bending tests [23].

Effect of Fibre Direction on Crashworthiness.
e FEA and experimental results of about five kinds of cases (Case A, Case B, Case C, Case D, and Case E) with different fibre directions are listed in Table 4.It indicates that Case A with ±45 °fibre direction has the lowest F max (the average value is 294.2N).On the contrary, Case C with ±60 °fibre direction has the highest F max (the average value is 511.5 N). e average value of F max of Case B with ±30 °fibre direction is 385.5 N. In addition, the value of F avg of Case B is the highest in these cases.e comparison of CFE of these five different cases is shown in Table 4, and both the CFEs of Case A and Case B are the highest in these cases, indicating that they have the smoothest loading-displacement process.e CFE of these cases are in a range from 0.21 to 0.49 as listed in Table 4. e SEA and EA of these cases are also listed in Table 4.It can be observed that both the SEA and EA of Case B are much higher than those of the other cases, indicating better crashworthiness performance.Except for Case B, both the SEA and EA of Case A are close to those of other cases.
From the above study, the crashworthiness of sandwich panels with aluminium honeycomb core and CFRP skins can be affected by fibre direction.ese FEA and experimental results revealed that Case A with ±45 °fibre direction has the   Summarizing the respective advantages of fibre direction of ±30 °and ±45 °, both of these fibre directions should be considered in the ply design.e deviation values of the experimental results for different fibre directions are shown in Table 5.It is found that the deviation values are less than 10%.erefore, the results of the experimental results for different fibre directions are reliable.It is noted that because the values of CFE are very small, it will lead to a large deviation in calculations.

Effect of Stacking Sequence on Crashworthiness.
Case A and Case F are considered to analyze the effect of stacking sequence on crashworthiness performance.e comparison of F max and F avg of the two cases is shown in Figure 11.It indicates that the F max of Case A is lower than that of Case F. e F avg of Case A is close to that of Case F. As shown in Figure 12, the CFE of Case A is higher than that of Case F, indicating that Case A has the smoother loadingdisplacement process.e SEA and EA of these two cases are shown in Figure 13.It can be observed that the SEA and EA of Case A are higher than those of Case F, indicating better crashworthiness performance.
It can be concluded that the crashworthiness of sandwich panels with aluminium honeycomb core and CFRP skins can be affected by the stacking sequence.By comparing both Case A and Case F crashworthiness criteria, the results reveal that Case A with an asymmetrical stacking sequence design has better crashworthiness performance.

Effect of Layer ickness on Crashworthiness.
As shown in Table 6, there are three kinds of cases (Case A, Case G, and Case H) with different layer thicknesses.For Case A, the layer thickness is 0.8 mm; for Case G, the layer thickness is 0.6 mm; and for Case H, the layer thickness is 0.4 mm.With the increase in the layer thickness, both the F avg and CFE increase.Interestingly, the F max is approximately the same.With the increase in the layers thickness, both the EA and SEA increase.Both the EA and SEA of Case A are the highest of these three kinds of cases.erefore, in general, with increase in the layer thickness, the sandwich panels with aluminium honeycomb core and CFRP skins would have higher energy absorption capacities.e deviation values of the experimental results for different layer thicknesses are shown in Table 7.It was found that the deviation values are less than 10%.erefore, the experimental results for different fibre directions are reliable.

Effect of Loading Velocity on Crashworthiness.
To investigate the effect of loading velocity on the crashworthiness of sandwich panels with aluminium honeycomb core and CFRP skins, the load-displacement curves of specimens (Case E4, Case E5, and Case E6) with different velocities are plotted in Figure 14.With the increase in the loading velocity, the F max of specimens of Case E increases.When the loading velocity is 30 mm/min, the F max of the specimen of Case E6 is the highest.
e comparison of CFE of three specimens (Case E4, Case E5, and Case E6) with different loading velocities is graphed in Figure 15, and the F max and F avg are also graphed in Figure 15.
e CFE of the specimen of Case E5 with minimum loading velocity is the highest.Advances in Materials Science and Engineering e EA and SEA of three specimens (Case E4, Case E5, and Case E6) are graphed in Figure 16.It can be observed that the EA and SEA of the specimen of Case E5 with minimum loading velocity are much higher than those of other specimens.
It can be concluded the crashworthiness of sandwich panels with aluminium honeycomb core and CFRP skins can be affected by loading velocity.By comparing the crashworthiness criteria, the results reveal that the specimen of    Advances in Materials Science and Engineering Case E5 with minimum loading velocity has the best crashworthiness performance.

Conclusions
is study presented a numerical and experimental study on bending responses of sandwich panels with aluminium honeycomb core and CFRP skins under quasistatic bending load.e typical load-displacement curves, failure modes, and energy absorption are explored.Within the limitation of the study, the following conclusions can be drawn: (1) e load-displacement curves of the sandwich panels with aluminium honeycomb core and CFRP skins under the three-point bending test can be divided into the initial elastic bending stage and bending collapse stage.e sandwich panels with aluminium honeycomb core and CFRP skins absorbed most of the energy during bending collapse stage.(2) e FEA prediction and experiment are used in understanding the initiation and propagation of cracks within the sandwich panels with aluminium honeycomb core and CFRP skins.e microfracture initiates at the corner of the top wall in contact with the loading roller.e cracks spread from the top wall of the filled CFRP skins to the aluminium honeycomb.In the bending process, the hexagonal cell layers of Advances in Materials Science and Engineering aluminium honeycomb are squeezed each other, leading to a fan shape.But the bending deformation was not enough to induce the cells of the aluminium honeycomb debonding from the adjacent adhesive layers.(3) e crashworthiness of sandwich panels with aluminium honeycomb core and CFRP skins can be affected by fibre direction.Case A with ±45 °fibre direction has the lowest F max , but not the highest EA and SEA.In addition, Case B with ±30 °fibre direction has the highest EA and SEA, but not the lowest F max .erefore, both of these fibre directions should be considered in the ply design.(4) e crashworthiness of sandwich panels with aluminium honeycomb core and CFRP skins can be affected by the stacking sequence.By comparing both Case A and Case F crashworthiness criteria, the results reveal that Case A with symmetrical ply design has a better crashworthiness performance.(5) In general, with the increase in the layer thickness, the sandwich panels with aluminium honeycomb core and CFRP skins would have higher energy absorption capacities.e crashworthiness of sandwich panels can be affected by loading velocity.e experimental results reveal that the specimen of Case E5 with minimum loading velocity (5 mm/min) has the best crashworthiness performance.
In summary, this study demonstrates the potential of sandwich panels with aluminium honeycomb core and CFRP skins to be an energy absorber used in electric vehicle body; and the experimental results can also be used for validating the numerical simulation for design optimization of the sandwich panels with aluminium honeycomb core and CFRP skins in the future.

Figure 1 :
Figure 1: CAD model of sandwich panels with aluminium honeycomb core and CFRP skins.

2
Advances in Materials Science and Engineering e abovementioned equation can be written in a simple form:

Figure 4 :
Figure 4: Schematic of sandwich structures with aluminium honeycomb core and CFRP skins in vehicle body.

Figure 5 :Figure 3 :
Figure 5: Specimens before quasistatic three-point bending test with different lay-up schemes.

Figure 6 :
Figure 6: Specimens before quasistatic three-point bending test with different loading speeds.

Figure 7 :
Figure 7: Specimens tested in the electronic universal testing machine.

Figure 9 :
Figure 9: Typical load-displacement curves of specimens in threepoint bending tests.

Figure 10 :
Figure 10: e failure process of sandwich panels in three-point bending tests.

Figure 11 :
Figure 11: Comparison of different stacking sequence for specimens: F max and F avg .

Figure 14 :
Figure 14: Load-displacement curves of specimens with different velocities on crashworthiness.

Figure 15 :Figure 16 :
Figure 15: Comparison of different loading velocities for specimens: F max , F avg , and CFE.

TABLE 3 :
Lay-up schemes of sandwich panels.

Table 4 :
Different fibre directions for FEA and experimental results.Advances in Materials Science and Engineering lowest F max , but not the highest EA and SEA.In addition, Case B with ±30 °fibre direction has the highest EA and SEA, but not the lowest F max .e lowest F max , highest EA, and highest SEA indicate better crashworthiness performance.

TABLE 5 :
e deviation values of the experimental results for different fibre directions.

Table 6 :
Different layer thicknesses for FEA and experimental results.

Table 7 :
e deviation values of the experimental results for layer thicknesses.