Abstract
According to previous studies, the bond strength between the carbon fiber–reinforced plastics and metals mainly depends on the bond layer between the carbon fiber–reinforced plastic composites and the metal. This article presents the results of the experiments conducted on the tensile bond strength between the carbon fiber–reinforced plastic tubes and aluminum joints. The parameters examined in these experiments include adhesive type, surface roughness of the bonded area, and thickness and length of the bond layer. The results show that the 3M DP-460NS Off-White adhesive was suitable for our experiment. The bonded surfaces of the specimens were smoothened with a P600 sandpaper to achieve higher bond strengths with the adhesive. For the six bond thicknesses tested, a thickness of 0.2 mm was optimal for the tensile experiments. The ultimate tensile force was approximately directly proportional to the bond length.
Keywords
Introduction
The carbon fiber–reinforced polymer (CFRP) composites are used to repair the deteriorated structure of steel and concrete and to manufacture the body or frame for certain vehicles such as airplanes, automobiles, and boats.1–4 The suspension of many racing cars used in Formula SAE consists of a carbon fiber and aluminum alloy bonding structure5–7 as shown in Figure 1. Compared with the traditional repair method for the metal and concrete, such as welding, bolting, and attaching steel plates, the CFRP composite provides higher strength with less weight and higher resistance against the effects of high temperatures and humid environments.8,9 The connection between the CFRP and the substrate should be efficient for the repaired structures or for the body and frame. Because of the different properties of the CFRP and substrate, bonding with an adhesive is the most suitable means of achieving an efficient connection.
HRT14 carbon fiber suspension.
For the composite structures, the adhesive between the CFRP and the substrate is the determining factor for the connection strength. Currently, studies conducted on the bond strength between the CFRP and substrate are limited and mainly focus on the plane of the connected surface. Bond behavior between the CFRP and steel or concrete plates has been studied through the static tensile experiment,10–14 fatigue load experiment,15–17 large deformation cyclic loading experiment,18,19 environmental effect experiment,20–22 dynamic loading rate tensile experiment,23,24 and so on. Generally, they were all conducted between the CFRP and substrate plate and obtained a number of useful results and formulas for the bonding connection between the CFRP and substrate plate.
To study the bonding behavior between the CFRP tubes and metal joints, and evaluate the performance of the bond, it is necessary to design a reasonable experiment schedule. In the area where carbon fiber tubes are applied, the truss work bears the axial directional force and the drive shaft bears the cyclic torsion loading.24,25 To determine the bond strength under these conditions, static tensile and torsional experiments are conducted.26,27
The aim of this article is to present the complete experiments conducted on the bond strength between the CFRP tubes and aluminum joints. It will include material selection, experiment preparation, experiment process, and analysis of the experiment results.
Preparation of experiments
Material properties
Excluding the adhesive, three materials including the CFRP tubes and aluminum alloy joints used in the tensile experiments were used as specimens. Specifically, the CFRP tubes were made by T300-Epoxy-Prepreg-3k-Woven-Carbon Fibers. The monolayer thickness was approximately 0.2 mm with a nominal ultimate tensile strength of 800 MPa, tensile elasticity modulus of 65 GPa, and density of 1600 kg/m3.The weave style is twill. The weight of carbon fiber prepreg is 320 g/m2 and the content of the resin is 40%. Most of the carbon fibres aligned, and transverse is the direction of the axis. The manufacturer of carbon fiber tube is to lay fiber on the mandrel, then heat cured vacuum.
The aluminum alloy was 7075-T6. Table 1 shows the properties of the materials.
Material properties.
CFRP: The carbon fiber–reinforced polymer.
Experiment instrumentation
The CSS-44300 electronic universal testing machine was used in the tensile experiments. The CSS-44300 can be used for tensile experiments, compression experiments, bending experiments, and controlled stress and strain experiments for metal, metalloids, and composites. The experiment provided the data for the axial deformation and tensile force. In the experiments, the speed of tensile is 1 mm/min.
Specimen details
This study requires specimens to conduct the experiments. The inner and outer diameters of the CFRP tubes were 17 and 20 mm, respectively. Figure 2 shows a detailed schematic of the specimens. The strength obtained with the tensile test corresponds to just with the adhesive, so it does not consider the effect of the mechanical fit.
Schematic plot of the specimens.
The variable parameters included the length and thickness of the bond layer. The length depended on length l 1, thickness depended on the gap between the carbon fiber tubes and metal joints, and the convex ring on the aluminum joints was designed to guarantee the adhesive layer thickness and the concentricity of the CFRP tube and aluminum joint. The specimens for the tensile experiments are shown in Figure 3.
Specimens for tensile experiments.
Experiments for tensile strength
Experiment program
In the tensile experiments, the parameters that affected the bond strength between the CFRP tube and aluminum joint were adhesive type, surface roughness, thickness, and length of the bond layer between the CFRP tube and aluminum joint. Four series of tensile tests were undertaken for the tensile experiments.
The selection of the adhesive is critical because it significantly affects the bond strength. The adhesive must have the following characteristics: high compatibility, high mechanical performance, high bonding strength, high humidity resistance, and excellent technical properties. On the basis of the performance of three different 3M adhesives shown in Table 2 (the data in Table 2 come from the manufacturers’ data sheet), the most suitable one (DP-460NS Off-White) was selected for the composite material and metal, and it was used for the remaining experiments (Adhesive-x). The chemical nature of the adhesives is epoxy. Group 1 experiments compared the performance of the adhesives through a simple tensile experiment.
Performance of three different 3M adhesives.
Group 2 experiments compared substrate surface treatment to obtain an appropriate surface roughness. In this study, there were three different surface treatment methods to obtain three different surface roughness values. The three methods were precision lathing, P600 sandpaper rubbing, and surface knurling. In the first experiment, the surface treatment method was precision lathing to achieve a smooth surface. The most efficient surface treatment method among the three was Treatment-y, which was used in the second experiment group, and it was used in the third and fourth experiment groups.
Group 3 experiments compared bond-line thicknesses to obtain the relationship between the ultimate tensile strength and the bond layer thickness. Six different thicknesses were tested in these experiments: 0.05, 0.10, 0.15, 0.20, 0.25, and 0.30 mm. Initially, the thickness for the first and second experiment groups was 0.10 mm. For the fourth experiment group, Thickness-z, which was the appropriate bond thickness in the third experiment group, was used.
Group 4 experiments compared the bond lengths to obtain the relationship between the ultimate tensile strength and the bond layer length. Five different bond lengths were tested in this experiment: 18.7, 28.0, 37.5, 46.8, and 57.2 mm with bonding areas of 10, 15, 20, 25, and 30 cm2, respectively. For the first, second, and third experiment groups, a bond length of 37.5 mm, which is the average value of the five lengths, was selected.
To establish repeatability, each experiment was conducted six times. Table 3 shows details of the tensile experiment program.
Tensile experiment program.
Results and discussion
Table 4 lists the results of the tensile experiments. The first group’s data showed that the specimens using the DP-460NS Off-White adhesive achieved higher ultimate tensile strengths than those using other adhesives. Therefore, adhesive-x was DP-460NS Off-White, which was used for the remaining experiments.
Tensile experiment results.
The second group provided data on surface roughness. Since DP-460NS Off-White was adhesive-x, the bonding condition of number 3 in the first experiment group and number 1 in the second experiment group was identical. Therefore, number 1 in the second experiment group was ignored. The data of the three groups showed that the specimens rubbed with sandpaper achieved the highest ultimate tensile forces than the specimens subjected to precision lathing and surface knurling. The results showed that the adhesive had higher adhesive force on slightly rough surfaces than on smooth surfaces. However, for the knurling surface, the surface bulge pierced the bond layer and declined the shear strength of the bond layer, so the bond layer could not provide higher ultimate tensile forces. Therefore, it was reasonable to rub the bonded surface with P600 sandpaper in a circular direction for the bars bearing the tensile and compression forces. This rubbing method was used in the third and fourth experiment groups.
The data of the third group showed the relationship between the thickness of the bond layer and the ultimate tensile strength. Since the surfaces were rubbed with the P600 sandpaper, the bonding condition of number 2 in the second experiment group and number 2 in the third experiment group was identical. Therefore, number 2 in the third experiment group was ignored.
The results showed that the ultimate tensile force increased with an increase in the bond layer thickness up to a thickness of less than 0.2 mm, as shown Figure 4. When the thickness became greater than 0.2 mm, the ultimate tensile force decreased with increases in the bond thickness. A thickness of 0.2 mm was the most suitable for connecting the CFRP and 7075 T6 joint using the DP-460NS Off-White adhesive, among the six tested bond layer thicknesses. This thickness (thickness-z) was used in the fourth experiment group.
Relationship between the ultimate tensile force and bond layer thickness.
Figure 5 shows the relationship between the bond layer length and the ultimate tensile strength. Since thickness-z was 0.2 mm, the bonding condition of number 4 in the third experiment group and number 3 in the fourth experiment group was identical. Therefore, number 3 in the fourth experiment group was ignored. The data in Table 4 show different ultimate tensile forces of the specimens with the same bond length. Evidently, the ultimate tensile forces differed, but the value was relatively centralized. The average value was selected to be the ultimate tensile force for each bond length. Figure 5 shows that the ultimate tensile force has an approximately linear relationship with the bond length. Therefore, for the selected CFRP tube, a higher ultimate tensile force could be achieved by increasing the bond length within the range of conditions tested here.
Relationship between the ultimate tensile force and bond layer length.
Figure 6 shows the tensile process of different bond layer lengths. The curve could have been caused by the tensile test machine, which had not clamped the test sample. When the deformation reached a value that caused the machine to clamp the test samples, the relationship between the tensile force and the deformation was linear. Once the tensile force reached the ultimate tensile force, the tensile force decreased rapidly.
Tensile force versus deformation.
Throughout the tensile experiments of the four groups, a reasonable bonding process for the truss work was developed. The adhesive 3M DP-460NS Off-White was the most suitable for bonding the CFRP tube and aluminum joint. In the bonding process, higher adhesive and shear forces could be achieved by rubbing the surfaces of the specimens with P600 sandpaper, and a bond layer thickness of 0.2 mm was the most suitable. The ultimate tensile strength had an approximately linear relationship with the bond length; therefore, the bond length could be determined according to the tensile load.
Conclusion
This article presents an experimental study on the development of a proper bonding process for connecting CFRP tubes and aluminum joints by conducting a series of tensile experiments on the bonded specimens. The parameters examined include adhesive type, surface roughness, and thickness and length of the bond layer. From the test results and discussions presented in this article, the following conclusions can be drawn:
The 3M DP-460NS Off-White adhesive was suitable for bonding the CFRP tube and metal joint.
Regarding the bonding process, high adhesive forces could be achieved by rubbing the surfaces of the specimens with P600 sandpaper.
High adhesive strengths could be achieved using a bond layer thickness of 0.2 mm among the six tested thicknesses.
The ultimate tensile strength had an approximately linear relationship with the bond length.
The bonding method for connecting the CFRP and aluminum joints was effective.
This study only conducted the experiments in a static state; however, the working condition of automobiles, airplanes, and other machineries is accompanied with irregular vibrations. Therefore, additional studies on the bond strength under dynamic loading rates should be conducted to improve this study.
Footnotes
Acknowledgements
The authors are grateful to the Weihai Guangwei Group who sponsored the CFRP material and CFRP tubes. The experimental instruments were provided freely by the Material Laboratory at Harbin Institute of Technology and the authors appreciate their support. The authors are also grateful to the HRT Formula Student Team who applied the bonding process in the CFRP suspension of their racing cars to show its practicability.
Academic Editor: Adib Becker
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported, in part, by the Weihai Science and Technology Development Project (grant no. 2014DXGJ16).