An experimental study on the failure mechanism of local-strengthening glass fiber reinforced polymer T-joints under/after the low-velocity impact loading

Abstract

In this study, an experimental methodology, consisting of low-velocity impact (LVI) and quasi-static compressive test, is conducted to investigate the performance and failure mechanism of GFRP T-joints, fabricated by vacuum-assisted resin infusion process (VARI), when subjected to out-of-plane LVI loading at the corner. Further, the effect of LVI loading on the reduction of strength is analyzed by the tensile test after LVI. In order to improve the strength of the GFRP T-joints, one layer of stainless-steel wire net (SSWN) is inserted into the corner of the T-joints structure. The experimental results indicate that the addition of SSWN decreases the damaged area of delamination in the web, meanwhile, connecting the nearby GFRP layers to avoid the structure failure suddenly, which improves its performance of tensile test after LVI.

Keywords

Glass fiber reinforced polymer T-joints reinforcement by metal wire net low-velocity impact failure mechanism

Introduction

Due to its remarkable advantages, such as high stiffness-to-weight ratio and low cost, fiber-reinforced polymers (FRP) are widely employed in aircraft, machinery, and other sectors.^1–4 Meanwhile, FRP is used in the field of civil engineering for the replacement of rebar and enhancement. the T-shaped structure involved in this study is widely used in Bridges and other structures in civil engineering,^5,6 which can provide references for the promotion and application of materials and structures in civil engineering. As manufacture-process constraints, the generated structures require connecting pieces. Then for a composite structure, there are two widely used methods, bolted and adhesive bonding joint, to joint composite together. Traditional bolted joints have severe difficulty with increased stress concentration around the fastener hole, which weakens the bearing capacity of composite constructions.^7,8 According to reference, 9^9,10 the fastening joint may lead to a total reduction by a ratio from 60% to 85% in the bearing capacity of the failure strength of composite laminate structures.

Because of its low weight and fewer components, the adhesive bonding technique has become one of the most critical technologies for composite structures.^11–13 Though the composite structures have a variety of advantages such as light weight, high corrosion, and fatigue resistance et al, they are sensitive to low-velocity impact loading, which could cause various types of damage and threaten their mechanical performance further. After the LVI event, the damages on the composite structures usually consist of fiber breakage, matrix crack, and interface delamination.^14,15 In our previous study,^16–18 it is indicated that minor incident energy of drop weight could cause some barely visible damage to the composite structures, resulting in a significant reduction in the bearing capacity of the structure. As a result, fiber-metal laminates (FML), a combination of sheets of fiber-reinforced polymer and metals, are designed and used in various aerospace applications nowadays.¹⁹ The most common combination of glass fiber reinforced plastics with aluminum sheets is referred to as GLARE (glass laminate aluminum reinforced epoxy). FMLs show excellent mechanical properties and outstanding fatigue behavior.^20,21 Further, due to the addition of the metal layer, the FMLs show better impact resistance properties than the FRP; on the other hand,²² weaker interfacial strength especially near the metal layers, which is one of the most important key properties under impact loading.^23,24

In order to improve the interfacial strength of FMLs, the main method is a surface treatment which can be classified into three categories: (i) mechanical treatments such as sandblasting, sanding, and shock peening,²⁵ (ii) chemical treatments such as anodizing, etching,²⁶ and (iii) the addition of interfacial layers such as coupling agent, sol/gel,²⁷ multi-walled carbon nanotube,^13,28–30 and electro grafted coating.³¹ Moreover, our previous study^16,17 showed that the addition of metal wire nets in the FRP could enhance the energy absorb ability as well as the compression-after-impact strength significantly without any surface treatment.

As our previous study¹⁵ shows, the delamination in the bottom and web is the main failure mode of a T-joints structure under tensile load. Therefore, its delamination behavior drew considerable attention from other researchers^31–35 as well.

The main objective of this work is to reveal the damage mechanism and the reduction in the mechanical strength of the GFRP T-joints under low-velocity impact loadings. Further, one layer of steel wire net is inserted in the GFRP layers to enhance its impact resistance property. A series of effective experimental methods are conducted which can be considered as an appropriate reference to study the low-velocity impact behavior of the FRP T-joints structures.

Experimental schema

Fabrication of specimens

The fabrication process, the lay-up configuration of GFRP, and the size of specimens are depicted in Figure 1, seven layers of woven glass fiber (Damin Tents Co., Ltd, Shanghai, China), with a density of 300 g/m² and a thickness of 0.25 mm per one layer, are laid on the outer surfaces of two angle-iron molds with the same direction along the X axis. Then the molds are connected by several bolts, after that, the third part of seven layers of woven glass fiber is applied on the plane of two-angle irons. The conventional vacuum-assisted resin infusion process (VARI), which is described in reference, 1^16,17 is used to completely inject vinyl ester resin (Dongguan Yuan Bon Plastic Industry, Guangzhou, China) into the vacuum space, which is created by the sealants between two molds and the vacuum bag on the layers. A proper mixing ratio of the hardening agent, methyl ethyl ketone peroxide (MEKP), and the accelerating agent, dimethylaniline, would help us to control the curing process at room temperature accurately. In order to improve the impact resistance of GFRP T-joints, as shown in Figure 1, a 40 mm length of metal wire net, 10 mesh/inch and 0.3 mm diameter of single wire, is inserted into the interface at the corner point of T-joints which leads an increase of 0.1 mm in laminate thickness. Besides, the average volume fraction of fiber in the composites is 55.23%. Finally, with the help of a diamond cutting machine, all T-joint specimens are cut into the same size, a length of 200 mm × height of 100 mm × width of 36 mm.

Low-velocity impact test

The INSTRON 9340 LVI testing machine, meeting the standard test method ASTM D 7136 for the low-velocity impact of GFRP, is utilized to investigate the performance of GFRP T-joints during the LVI events. The hammer has a total mass of 4.5 kg and a semicircular top of 16 mm in diameter, the range of impact energy level was changed by controlling the height of the drop weight. In the experiments, as shown in Figure 2(a), we designed the fixture with a circle hole of 76 mm diameter, consisting of three rectangle steel plates which are fixed on the platform with a rectangle hole of 125 mm * 75 mm, meeting ASTM D 7136 M for LVI of FRP laminate. Three kinds of specimens such as rectangle plates (RP), T-joints (TS0), and enhanced T-joints (TS40) are tested in the LVI behavior under various incident energy levels of 15 J, 20 J, 25 J, 30 J, and 35 J. The response history, including contracted force, displacement, and absorbed energy, is recorded by the device. (RP: comparative model; 15 J–2.37 m/s, 20 J–2.73 m/s, 25 J–3.06 m/s, 30 J–3.34 m/s, 35 J–3.61 m/s)

Static test as low-velocity impact and tensile test after low-velocity impact

In the static testing, as shown in Figure 2(c), with the exception of loading at a speed of 1.25 mm/min, the loading and boundary conditions are set up the same as low-velocity impact testing. Furthermore, both the results in the contact force and the evolution of delamination degree are obtained. After the low-velocity impact testing, as pictured in Figure 2(b), the specimens are clamped to take a tensile test at a loading speed of 1 mm/min. In this test, all fixed bolts are applied at the same moment of 4 Nm to keep the same boundary condition. Each kind of test is repeated at least three times to ensure the reliability of the experimental data.

Results and discussion

Static test under the same boundary condition as the low-velocity impact

Figure 3(a) shows the force-displacement curves of TS0, TS40, and RP under the static testing as depicted in Figure 3(b). The maximum value of contact force is 2383 N, 3256 N, and 4086 N, respectively. When inserting a layer of steel wire net into the corner, the strength of T-joints under static test as the low-velocity impact could be increased by up to 36.6%. In the tests for TS0/TS40 panels, the contact force increases sharply till some permanent damages are found in the contacted region, as shown in Figure 4. Then, as the damage spreads around the contact region, the contact force increases slowly and reaches its peak value when the top semicircle begins to penetrate the panel.

According to the failure evolution of TS0 and TS40 in the static test shown in Figure 4, four stages can be summarized. Firstly, the diagrams start from a straight line with a slope similar to RP, at the same time, the deformation is distributed along the entire skin part. Then, a small initial failure appears at the corner of the vertical interface under a similar load force of 688.5 N and displacement of 2.36 mm, resulting in a significant decrease in the slope of the contact force-displacement curve. Meanwhile, it is considered as the beginning of the second stage. As the increase of the displacement, the contact force and vertical cracks of TS0 and TS40 propagate along the intermediate interface in the web. Nevertheless, at the end of the second stage, at position C, a new crack appeared at the horizontal interface. In the third stage, as the initiation of crack at the skin laminate, and the increase of vertical crack as well as the contact force, the slope of the force-displacement declines again. In the fourth stage, as shown in Figure 4, the extended cracks in both directions and concentrated tensile failure in the skin at the corner are obtained, which leads the contact force to decline sharply from the maximum value. As a result, the T-joint structures totally fail.

Low-velocity impact testing

According to our previous study,¹⁵ the low strength of the web interface could reduce the static mechanical properties of a T-joint structure. Similarly, as shown in the force-time and displacement-time curves of Figures 5 and 6, the T-joints structures present relatively poor performance in impact resistance. In general, TS40 cases show a higher ductility than TS0 cases, which may cause a lower contact force. Moreover, compared with TS0, TS40 has much higher ductility, but a small lower ultimate contact force. As a result, the higher ductility of materials results in higher impact resistance due to absorbing more impact energy.

Figure 7 shows in detail the comparisons of energy-time curves of RP, TS0, and TS40 specimens. It is found clearly that the absorbed energy value of RP and TS0 specimens under each incident energy can reach the incident energy. However, because the insertion of wire mesh improves the ductility and plastic deformation of specimens, a portion of the energy is not responded to by the TS40 specimens undergoing high deformation at high incident energies (30 J and 35 J) to the impactor.

The comparison of delamination length for each case under 5 incident energies is shown in Figure 8 and is summarized in Figure 9. The primary failure mode of RP cases is interface delamination. With the increase of incident energy, the delamination area gradually expands from the impact point to the fixed boundary. In terms of the T-joints cases, the crack length of both TS0 and TS40 cases increases with the increase of incident energy. In addition, the vertical lengths of all cracks are greater than the horizontal lengths. As summarized in Figure 9, under various incident energies, the TS40 cases are about 2.5 mm shorter in vertical length and much shorter than TS0 in horizontal length. For instance, under the incident energy of 35 J, the TS0 has cracks, 33.5 mm, and 21.2 mm, in the horizontal and vertical interface, while TS40 cases have a shorter value of 31.5 mm and 11.3 mm, which means reductions at a ratio of 6.0% and 46.7%. Hence, the addition of a metal wire net effectively prevents the interface crack propagation in LVI events, especially along the horizontal direction.

Tensile test after low-velocity impact event of T-joints structure

As shown in Figure 10(a), both force-displacement curves of TS0 and TS40 without drop-weigh loading present a sharp and slow linear increase in succession. Figure 11 shows the comparisons of crack evolution of TS0 and TS40 in the tensile test after low-velocity impact under incident energy of 0/30 J. In order to reveal the damage mechanism, three key points, such as the turning point of TS40-0 J curve, peak value point of TS0-30 J and TS40-30 J, marked as dotted-line in Figure 10(a), are chosen as the critical moment during the tensile test.

According to the evolution of damage morphology shown in Figure 11, regarding TS0-0J and TS40-0J, as the tensile loading increases up to about 400 N, an initial delamination at the middle interface in the corner of web is found, then both force-displacement curves tend to increase slowly. It is because they have similar web in the layer configuration and material which causes the same interfacial strength. As the increase of tensile displacement, the crack length at the web and the tensile force grows up to the maximum value. However, as shown in Figure 11, when an initial crack is observed at point C in the horizontal area of the corner, the force first begins to decrease. After that, the horizontal crack length of TS0-0J and TS40-0J increased dramatically, besides, the force of the TS0 case suddenly decreased rather than being as gentle as that of TS40.

The force-displacement curves of the tensile test after low-velocity impact events with various incident energies for TS0 and TS40 are pictured in Figure 10 as well and all curves’ maximum values are listed in Table 1. In general, the low-velocity impact loading could reduce the performance of T-joint structures on the subsequent tensile test. In addition, after low-velocity impact loading at the same incident energy, the tensile strength of TS40 is more than 15% higher than TS0. At the same time, it is noted that, after low-velocity impact events, all force-displacement curves linearly increase with a similar slope until reaching the peak. It is because there is no big difference in the length of the vertical crack, as listed in Figure 9. During the process of tensile test, the crack propagation is concentrated at the intermediate interface along the horizontal direction, marked in Figure 11, where compare the crack-propagation evolution of TS0 and TS40 in the tensile test after low-velocity impact with an incident energy of 30 J at the key points. When the force of TS0-30 J reaches the maximum value and drops dramatically, however, the curve of TS40-30 J reaches its first peak force, with a greater value, and presents a longer crack length in both directions. After a slight decrease, even a slight growth of force to maximum value can be continuously observed. Meanwhile, the delamination extends along both sides of the metal wire net. As a result, though greater delamination is presented, the metal wire net could build up a connection as a bridge at the middle interface of the skin panel so as to enhance the strength of the T-joints structure.

Table 1.

The peak force in the tension after the low-velocity impact of different incident energies.

Incident energy(J)	TS0		TS40
Incident energy(J)	Average peak force/N	Decline ratio to TS0-0 J/%	Average peak force/N	Decline ratio to TS40-0 J/%	Growth ratio to TS0/%
0	863.11 ± 30.82	0	1067.06 ± 11.63	0	23.63
20	704.26 ± 26.35	18.40	875.64 ± 18.42	17.94	24.33
25	589.78 ± 13.54	31.67	688.71 ± 09.64	35.46	16.77
30	510.22 ± 18.68	40.88	596.36 ± 21.39	44.11	16.88
35	434.82 ± 23.61	49.21	512.71 ± 24.14	51.95	17.91

Conclusion

In this paper, the mechanical properties of GFRP T-joints are studied experimentally. The experiment consists of static load, low-velocity impact, and tensile test after it. In the dynamic testing, the contact force history, displacement history, and damage morphology are recorded. In the static test, both force-displacement curves and damage evolution are obtained. Combining the experimental results of the static and low-velocity impact test with the same boundary condition, the damage mechanism of T-joints structures under the low-velocity impact loading is revealed deeply. Further, the strength of T-joint structures is enhanced effectively by inserting a steel wire net at the corner, as well as the strengthening mechanism is illustrated. Based on these investigations, the main conclusions can be summarized as follows:

1. Under a static concentrate press loading at the corner point, the failure of the T-joints, caused by the interface delamination, contains four stages, which are distinguished by the initiation of the interface delamination. The addition of one layer of steel wire net in the middle interface could prevent the extension of the interface delamination so as to improve the maximum contact force of the GFRP T-joints structure by a ratio of 36.6%.

2. In the low-velocity testing, the TS40 cases presented higher ductility to TS0 cases so as to absorb the impact energy with a higher displacement, lower contact force, and less damage degree, which means that the TS40 exhibits higher performance under the loading of low-velocity impact.

3. In the tensile testing after the low-velocity impact loading, the failure evolution is presented. Moreover, it is found that, except for reducing the damage degree in the low-velocity impact examination, the steel wire net in the interface could connect the nearby GFRP layers to avoid a structure failure suddenly. For example, the peak tensile force of TS40-30 J is higher than the value of the TS0-30 J by a ratio of 16.88%.

Footnotes

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 work was supported by the National Natural Science Foundation of China (grant numbers 12002126, 12172130), Natural Science Foundation of Jiangxi Province (grant number 20224BAB201016, 20202ACB211002, 20202BAB201007), Primary Research and Development Plan of Jiangxi Province of China (grant number 20212BBE53016), Foundation of Jiangxi Province of China Educational Committee (grant number GJJ2200649, GJJ201907, GJJ211908, GJJ2201503), and The Innovative Projects of NIT(YC2023-S997).

ORCID iD

Yun Wan

Data availability statement

Data will be made available on request.

References

Al-Lami

Hilmer

Sinapius

. Eco-efficiency assessment of manufacturing carbon fiber reinforced polymers (CFRP) in aerospace industry. Aero Sci Technol 2018, 79: 669–678.

, et al. Experimental study of CFRP laser surface modification and bonding characteristics of CFRP/Al6061 heterogeneous joints. Compos Struct 2022; 283: 115030.

Quan

Wang

Zhao

, et al. On the fracture behaviour of aerospace-grade polyether-ether-ketone composite-to-aluminium adhesive joints. Compos Commun 2022; 30: 101098.

Zhang

Shen

, et al. Low-speed impact of glass/Kevlar hybrid composites: Experimental results and numerical verification. J Compos Mater. 2023; 57(19): 3079–3099.

Chen

Young

. Experimental investigation on cold-formed steel elliptical hollow section T-joints. Eng Struct 2023; 283: 115593.

Musa

Mashiri

. Parametric study and equations of the maximum SCF for concrete filled steel tubular T-joints under in-plane and out-of-plane bending. Thin-Walled Struct 2019; 135: 245–268.

Wang

Qin

Mekala

, et al. Three-dimensional progressive damage and failure analysis of double-lap composite bolted joints under quasi-static tensile loading. Compos Struct 2022: 285: 115227.

Kaybal

Ulus

Eskizeybek

, et al. An experimental study on low velocity impact performance of bolted composite joints-part 2: Influence of long-term seawater aging. Compos Struct 2021; 272: 113571.

Cao

Cardew-Hall

. Interference-fit riveting technique in fiber composite laminates. Aero Sci Technol 2006; 10(4): 327–330.

10.

Yang

Smith

, et al. Strengthening single-bolt timber joints with externally bonded CFRP composites. Structures 2020; 28: 2671–2685.

11.

Klein

Ihmels

Schürmann

, et al. Manufacturing influences of the adhesive layer on the load-bearing capacity of anchor bolts for thick FRP laminates. Int J Adhesion Adhes 2021; 106: 102817.

12.

Mai

Xie

, et al. Durability of components of FRP-concrete bonded reinforcement systems exposed to chloride environments. Compos Struct 2022; 279: 114697.

13.

Tang

Tan

, et al. Study on tensile performance of hybrid adhesive-riveted joint of carbon fibre sandwich structure based on energy method. J Compos Mater. 2023; 57(18): 2873–2885.

14.

Wan

. The low-velocity impact and compression after impact (CAI) behavior of foam core sandwich panels with shape memory alloy hybrid face-sheets. Sci Eng Compos Mater 2019; 26(1): 517–530.

15.

Wan

Yang

Yuan

, et al. Low-velocity impact damage localization of GF/epoxy laminates by the embedded MWCNT@ GF sensor network. J Mater Res Technol 2020; 9(4): 9253–9261.

16.

Wan

Yang

Jin

, et al. Impact and compression-after-impact behavior of sandwich panel comprised of foam core and wire nets/glass fiber reinforced epoxy hybrid facesheets. Jnl of Sandwich Structures Materials 2021; 23(6): 2614–2637.

17.

Wan

Diao

Yang

, et al. GF/epoxy laminates embedded with wire nets: a way to improve the low-velocity impact resistance and energy absorption ability. Compos Struct 2018; 202: 818–835.

18.

Wan

Yao

, et al. Experimental studies of low-velocity impact behavior on hybrid metal wire net/woven carbon-fiber reinforced composite laminates. Compos Commun 2022; 32: 101185.

19.

Sinmazçelik

Avcu

Bora

MÖ

, et al. A review: Fibre metal laminates, background, bonding types and applied test methods. Mater Des 2011; 32(7): 3671–3685.

20.

Wang

Dong

, et al. Metal surface nanopatterning for enhanced interfacial adhesion in fiber metal laminates. Compos Sci Technol 2021; 205: 108651.

21.

Yamada

Kötter

Nishikawa

, et al. Mechanical properties and failure mode of thin-ply fiber metal laminates under out-of-plane loading. Compos Appl Sci Manuf 2021; 143: 106267.

22.

Wang

Liu

, et al. On impact behavior of fiber metal laminate (FML) structures: a state-of-the-art review. Thin-Walled Struct 2021; 167: 108026.

23.

Yao

Wang

, et al. Numerical and experimental investigation on the oblique successive impact behavior and accumulated damage characteristics of fiber metal laminates. Thin-Walled Struct 2021; 166: 108033.

24.

Xiao

, et al. Development of an integrated model for prediction of impact and vibration response of hybrid fiber metal laminates with a viscoelastic layer. Int J Mech Sci 2021; 197: 106298.

25.

Kim

Surface modification of Ti dental implants by grit-blasting and micro-arc oxidation. Mater Manuf Process 2010; 25(5): 307–310.

26.

Chen

Yang

. Surface modifications of Ti alloy with tunable hierarchical structures and chemistry for improved metal–polymer interface used in deepwater composite riser. Appl Surf Sci 2015; 328: 614–622.

27.

Liang

, et al. Effect of modified polypropylene on the interfacial bonding of polymer–aluminum laminated films. Mater Des 2015; 81: 141–148.

28.

Wang

Tao

Jin

. The effect of MWCNTs with different diameters on the interface properties of Ti/CFRP fiber metal laminates. Compos Struct 2021; 266: 113818.

29.

Tayeb

Zidour

Bensattalah

, et al. Mechanical buckling of FG-CNTs reinforced composite plate with parabolic distribution using Hamilton’s energy principle. Advances in nano research 2020; 8(2): 135–148.

30.

Guo

, et al. Enhanced interfacial adhesion of CF/PEEK-titanium hybrid laminates via introducing micro-nano layers with multi-walled carbon nanotube networks. Compos Sci Technol 2022; 223: 109418.

31.

Pan

Duan

Zheng

, et al. Surface characteristics and adhesive strength to polyetheretherketone of titanium electrografted with aryl diazonium salt. Mater Des 2016; 95: 555–562.

32.

Guo

. Numerical analysis and experiment of sandwich T-joint structure reinforced by composite fasteners. Compos B Eng 2020; 199: 108288.

33.

Bai

Dong

Xiong

, et al. Progressive damage behaviour of RTM-made composite T-joint under tensile loading. Compos B Eng 2019; 160: 488–497.

34.

Philibert

Soutis

Gresil

, et al. Damage detection in a composite T-joint using guided lamb waves. Aerospace 2018; 5(2): 40.

35.

Lesani

Sadat Hosseini

Bahaari

. Load bearing capacity of GFRP-strengthened tubular T-joints: experimental and numerical study. Structures 2022; 38: 1151–1164.