Keywords

1 Introduction

Automobile lightweight can reduce the weight of the body and improve fuel efficiency, which is considered to be an important means to solve current energy and environmental issues [1,2,3]. For the ever-growing new energy vehicles (NEVs), lightweight automobile can improve the mileage of NEVs and solve the problem of “mileage anxiety” to a certain extent. Automobile lightweight has a positive effect on both traditional fuel vehicles and NEVs [4]. FMLs are a new type of lightweight engineering material that combines the respective advantages of metal and fiber-reinforced composite materials, which has been widely focused by the automotive industry for its excellent mechanical properties such as low density, corrosion resistance, and fatigue resistance [5,6,7].

However, FMLs is a hyper-hybrid composite material with multiple interfaces, which may have various failure behaviors such as fiber fracture and matrix damage during the stamping forming process, especially the delamination failure between heterogeneous interfaces would seriously affect the mechanical properties of the component. Vieille [8] found that the softening of the resin matrix at high temperature would lead to serious degradation of the fiber/resin bonding interface. Hirsch [9] confirmed that the interface roughness has a positive impact on the bonding performance of the FMLs heterogeneous interface. How to accurately quantify the mechanical properties of FMLs heterogeneous interfaces and accurately predict the interlayer damage failure is an urgent problem to be solved.

In this paper, the preparation of CFRTP/AL laminates and the stamping of S-beams were firstly carried out. Then DCB test and ENF test were executed according to ASTM test standard, which accurately quantified the mechanical properties of FMLs prepared by the hot stamping process and obtained the fracture toughness of type I and type II that characterize the interfacial bonding strength. Furthermore, the finite element simulation method coupled with CZM was introduced to simulate the experimental process of DCB and ENF to verify the correctness of the model. Finally, the developed numerical modeling strategy of CFRTP/AL was applied to the stamping simulation of S-beams successfully, which realized the accurate prediction of layered failure of heterogeneous interfaces in the process of large deformation of FMLs.

2 Fabrication of CFRTP/AL Laminates

FMLs is an interlayer hyper-hybrid composite material, which is obtained by stacking metal sheets and fiber-reinforced composite materials at a certain angle and curing under a certain temperature and pressure conditions [10]. In this study, FMLs were prepared by a HPFM-100D comprehensive compression molding test platform. AA5754 aluminum alloy produced by Southwest Aluminum is selected as the surface metal, and its material composition and performance parameters are shown in Tables 1 and 2. The fiber-reinforced composite material is the unidirectional prepreg of the thermoplastic PA6 carbon fiber produced by Zhongfu Shenying Carbon Fiber Co., Ltd. The relevant material parameters are shown in Table 3.

Table 1 Constituent elements and compositions of AA5754 aluminum alloy
Full size table
Table 2 Parameters of AA5754 aluminum alloy
Full size table
Table 3 Parameters of thermoplastic PA6 carbon fiber reinforced composites
Full size table

After the preparation of CFRTP/AL laminate, a 40 T hydraulic press was utilized to stamp the S-beam. The mold gap of the S-beam was 1.6 mm. Therefore, when preparing the laminate, two layers of carbon fiber prepreg were stacked between the upper and lower surface aluminum alloys. An adhesive film was added between the heterogeneous interfaces to further enhance the interlayer bonding ability. In order to obtain better formability, the CFRTP/AL laminate was heated to 150 ℃ and quickly transferred to the mold for stamping, and then cooled to room temperature with the furnace to obtain the S-beam formed part The layering method of CFRTP/AL and related hot molding process parameters are shown in Fig. 1c.

Fig. 1
figure 1

Preparation and forming of CFRTP/AL laminates a: HPFM-100D comprehensive molding test platform b: S-beam forming press c: The temperature–pressure curve of hot molding

Full size image

3 Interfacial Fracture Toughness Test

During the forming process of CFRTP/AL, the failure of delamination cracking between the heterogeneous interfaces of laminates is a key issue of concern. In order to accurately measure the interfacial fracture toughness of CFRTP/AL laminates prepared by hot molding process, DCB and ENF tests were carried out on the basis of ASTM D5528 and ASTM D7905 standards [11, 12]. The total thickness of the specimen is 4 mm. In order to ensure that the single component material of the specimen deforms harmoniously under the action of the test force, the carbon fiber prepregs are stacked in the orthogonal lay-up method of 0°/90° (fiber direction is specified as 0°) for 10 layers, so that the thickness of the composite material and the thickness of the aluminum alloy are both 2 mm. To prefabricate the crack, a layer of 0.05 mm polyimide release paper was added between the carbon fiber prepregs and the aluminum alloy. The release paper should not be too thick, as excessively thick release paper will lead to a lipid-rich area at the binding tip of the heterogeneous interface, resulting in inaccurate experimental results.

The DCB/ENF test was conducted on the WDW-100 universal tensile testing machine, and the test process is shown in Fig. 2. In the DCB test, the carbon fiber bundles were pulled out of the resin matrix, showing the “bridging” phenomenon in Fig. 2a. At this point, the bearing capacity had been greatly reduced. The final calculated mode I fracture toughness data GIC is 753.91 J/m2. In the ENF specimen, the heterogeneous interface was mainly subjected to shear stress. Due to the large longitudinal stiffness of CFRTP, it cannot coordinate with the aluminum alloy layer to deform when subjected to a large bending displacement load. The glue layer on the left side of the punch mainly bore this incongruous deformation and suffered shear stress, resulting in shear failure. At the same time, the carbon fiber layer gradually broke and failed from the outside to the inside, and finally the carbon fiber layer and the aluminum alloy layer were completely cracked, as shown in Fig. 2b. The final calculated mode II fracture toughness data GIIC is 3.59 kJ/m2.

Fig. 2
figure 2

Interface mechanical properties test a: DCB b: ENF

Full size image

4 Numerical Simulation of Forming Process for CFRTP/AL

4.1 Cohesive Zone Model

CZM is widely used in the field of continuum fracture simulation and crack propagation, and has been derived into bilinear, trapezoidal, exponential, and other forms. Figure 3 shows the damage propagation process of the heterogeneous interface adhesive layer and the exponential CZM. It can be seen that the exponential traction–separation relationship assumes that the adhesive layer is initially in the linear elastic stage. In ABAQUS, both cohesive element and cohesive contact can be used to simulate the bonding behavior of heterogeneous interfaces. When the cohesive contact method is used for simulation, the slope of the elastic stage is the stiffness of adhesive layer. When the traction force reaches the initial damage, the damage evolves exponentially. The area enclosed by the traction force-separation curve and the coordinate axis represents the fracture energy \(G_{C}\). The damage initiation criteria commonly used in CZM include the maximum nominal stress criterion and the quadratic nominal stress criterion. In order to accurately describe the delamination failure between heterogeneous interfaces, the quadratic nominal stress criterion is introduced, which means that when the sum of the squares of the nominal stress and the maximum nominal stress ratio in the three directions is 1, it is determined that the adhesive layer fails, as shown in Eq. 1.

$$\left( {\frac{{\left\langle {t_{n} } \right\rangle }}{{t_{n}^{\max } }}} \right)^{2} + \left( {\frac{{t_{s} }}{{t_{s}^{\max } }}} \right)^{2} + \left( {\frac{{t_{t} }}{{t_{t}^{\max } }}} \right)^{2} = 1$$
(1)
Fig. 3
figure 3

Crack growth process and exponential CZM of heterogeneous interface

Full size image

where tn is the normal traction stress component, ts and tt are the tangential traction stress components, and “ <  > ” indicates that the failure of the interface would not be caused by the compressive deformation or stress in the normal direction.

When the adhesive layer reaches the damage initiation, the damage evolution is carried out. Considering that the heterogeneous interface of the CFRTP/AL laminate is a mixed failure model of type I and type II fracture during the forming process, it is necessary to consider the combined action of normal traction and tangential traction at the same time. In this paper, the damage propagation of the adhesive layer is simulated based on the second-order power law, as shown in Eq. 2.

$$\left( {\frac{{G_{I} }}{{G_{IC} }}} \right)^{2} + \left( {\frac{{G_{II} }}{{G_{IIC} }}} \right)^{2} = 1$$
(2)

where GIC is the energy released by the normal traction force, and GIIC is the energy released by the tangential traction force, namely, the type I fracture energy and the type II fracture energy.

4.2 DCB/ENF FE Simulation

A 3D finite element model is established in the commercial software ABAQUS, and the mechanical behavior of the DCB specimen under type I loading conditions is simulated using cohesive element. The model can be divided into three parts: the aluminum alloy layer, the CFRTP layer, and the middle adhesive interface. In order to accurately simulate the actual stress of the DCB specimen during the test, two reference points are established 5 mm above the end of the model, and the motion constraints are imposed on them respectively to simulate the role of the hinge in the test. Boundary conditions are shown in Fig. 4a. In ABAQUS, the simulation result of scalar stiffness degradation (SDEG) is expressed as the damage of the adhesive layer. Larger SDEG value indicates more serious damage of the adhesive layer. When its value is 1, it indicates that the adhesive layer has completely failed and the corresponding elements are deleted. The DCB simulation results are shown in Fig. 4b. It can be seen from the figure that with the increase of the tip displacement load, the heterogeneous interface experiences the continuous increase of traction stress, damage initiation, damage evolution, and complete failure elements deletion in four stages, which is consistent with the test results.

Fig. 4
figure 4

DCB finite element simulation a: boundary conditions and motion constraints b: simulation results

Full size image

ENF finite element model and boundary conditions are shown in Fig. 5a. The radius of the punch and the supporting roller is 5 mm, which is set as the rigid body through reference point. Combined with the ENF experimental process, the degrees of freedom of the two rollers are set as complete fixed, and a downward displacement load is applied to the punch. The simulation result CSQUADSCRT represents the quadratic traction stress damage initiation criterion. When using cohesive contact calculation method, this parameter can be used to judge whether the heterogeneous interface is damaged. When its value is 1, it means that the damage has been reached. The ENF simulation results are shown in Fig. 5b. From the simulation results, it can be seen that in the early stage of simulation, the adhesive layer generates shear traction at the tip and first appeared damage failure. As the punch continues to move down, the failure gradually expands and finally gathers at the position of the punch. The adhesive layer failure at both ends of the punch is earlier than other positions. It can be concluded from the DCB/ENF simulation results that the damage simulation of the failure behavior of the CFRTP/AL heterogeneous interface adhesive layer can be realized based on both the cohesive element and the cohesive contact.

Fig. 5
figure 5

ENF finite element simulation a: boundary conditions and motion constraints b: simulation results

Full size image

4.3 Simulation of S-beam CFRTP/AL Stamping Forming

Furthermore, CZM is introduced into the stamping forming simulation of the S-beam for automobile parts. The finite element model is shown in Fig. 6. Among them, punch and die are meshed with R3D4 shell elements, which are constrained as discrete rigid bodies by reference points. The sheet material is deformable CFRTP/AL composite material, and the mesh type of surface aluminum alloy is C3D8R solid element, and the core composite material mesh type is SC8R continuous shell element. Considering that the central part of the sheet material is the main deformation area, this section is remeshed by transition mesh division method, which gradually densifies from both sides to the middle. Thickness direction is divided into three layers to prevent warping in the forming process. The adhesive ability between the heterogeneous interfaces of FMLs is the focus of attention in the stamping process. During the setting of contact properties, the contact type between the upper and lower surface metals and the core composite is specially designated as cohesive contact. Based on the second-order power law, the delamination failure between the heterogeneous interfaces is effectively predicted by using the type I and type II fracture toughness. The rest of the parts adopt the general contact properties and the contact law is penalty function method. Tangential friction coefficient is set as 0.125. Moreover, the cohesive element has high requirements on the quality of the mesh. The cohesive element is meshed at 0.5 mm to prevent the simulation from failing to converge.

Fig. 6
figure 6

Finite element model of stamping forming of CFRTP/AL S-beam laminate

Full size image

The results of finite element simulation are shown in Fig. 7. It can be seen from the results that S-beam has good formability and there is no delamination cracking between the heterogeneous interfaces. The S-beam part obtained by stamping with the 40T press is shown in Fig. 7b. The experimental results are consistent with the simulation results, which verifies that the modeling strategy proposed in this paper can effectively predict the ply directional failure of CFRTP/AL laminates during the forming process.

Fig. 7
figure 7

Finite element simulation results of CFRTP/AL laminated S-beam a: Coupled CZM finite element simulation results b: Experimental results

Full size image

The simulation result CSDMG is the stiffness degradation coefficient of cohesive surface, which can be used to describe the combination of heterogeneous interfaces of CFRTP/AL, and its value of 1 indicates that the interface is completely invalid. In order to observe the change history of the ply directional adhesive properties of heterogeneous interface during the forming process of S-beam, the bottom aluminum alloy is taken as an example. Its upper surface is in viscous contact with the surface of composite material. The simulation results are shown in Fig. 8. It can be seen that the initial damage between CFRTP/AL heterogeneous interface occurs at the contact area between the sheet metal and the die fillet. As the punch continues to move down, the damage gradually expands to both sides, and the maximum stiffness degradation coefficient is 0.829, which does not completely fail. However, due to the large damage, the interface delamination may occur under the influence of alternating stress during the service. The figure also shows the fiber tensile failure simulation results of the core composite material. From the simulation results, some fibers have broken, and the position is mainly concentrated in the punch corner.

Fig. 8
figure 8

Damage history of adhesive layer during CFRTP/AL forming process

Full size image

5 Conclusions

In this paper, CFRTP/AL laminates were firstly fabricated based on the hot molding process, and the stamping test of S-beams was completed with the 40T press. The formability of the parts was good, and no delamination crack occurred. According to ASTM D5528 and ASTM D7905 test standards, DCB and ENF tests were performed. The fracture toughness of type I was calculated as GIC = 753.91 J/m2, and the fracture toughness of type II was calculated as GIIC = 3.59 kJ/m2. FE simulation of DCB/ENF test process was carried out based on the damage constitutive model of coupled CZM, which confirmed that CZM can be used to describe the bonding mode between heterogeneous interfaces and accurately predict the damage and failure. By using the equivalent modeling strategy of discontinuous micro-shear, the simulation model constructed in this paper is introduced into the stamping process of S-beam of typical automobile parts. The results show that damage of heterogeneous interface mainly occurs in the rounded part of the die, and expands to both sides with the downward movement of the punch. The maximum stiffness degradation coefficient CSDMG of the adhesive layer is 0.829, and no stratification phenomenon occurs.