A study of the damage tolerance of composite-metal hybrid joints reinforced by multiple and penetrative thin pins

Abstract

The application of adhesive bonding technology in aircraft structures can reduce the total wight greatly, but the bonded joints are very sensitive to the possible manufacturing defects and damages during service operations, which makes them difficult to meet the damage tolerance requirements of the current transport airplane structures. In this study, the damage tolerance of composite-metal hybrid joints reinforced by multiple and penetrative thin pins was studied. The damage tolerance performance of the composite-metal joint is supposed to be enhanced by multiple through-the-thickness penetrative thin reinforcements in the bonding region, and the thin reinforcements were bonded together with both the composite and metallic joint plates. Both experimental tests and finite element simulations were conducted to investigate the effects of the through-the-thickness reinforcements on the damage tolerance performance of the joints with and without pre-fabricated disbond defects. Through the comparative analyses, it was found that the penetrative thin pins in the bonding region significantly improved the static load carrying capacity, the failure strain, the fracture energy, and the fatigue lives of the composite-metal bonded joints. Moreover, the reinforcements decreased the sensitivity of the bonded joints to the disbond defects in the bonding region. The damage tolerance performance of the composite-metal adhesively bonded joints was significantly increased by the through-the-thickness penetrative reinforcements and the enhancement mechanism was revealed by the combined analysis of test results and simulation results.

Keywords

composite-metal bonded joint through-the-thickness reinforcement damage tolerance disbond fatigue life finite element

Introduction

The joints between carbon fiber reinforced composite materials and metals are commonly used in modern airplane structures, where a critical challenge is to effectively connect the materials with different physical and mechanical properties. The adhesively bonded joints have high load transfer efficiency and hence, can be used to reduce the structural weight. Nevertheless, it is difficult for adhesive joints to meet the current large transport airplane structural design philosophy of damage tolerance since they often fail abruptly and catastrophically and loose load capacities completely due to the brittle nature of the polymer adhesive and the high stress concentration at the ends of adhesive joints.¹ In addition to that, bonding is prone to a wider variety of manufacturing defects, such as porosities and voids, adhesive cracking, weak bonds, disbonds, and unbonds, even with a rigorous manufacturing quality control.² These manufacturing defects will decrease the load carrying capacity of the joint greatly and may improve the cracks initiation and propagation during the service life of a bonded joint. This makes the current adhesive bonding technology hard to meet the current regulations in damage tolerance of transportation category airplanes, such as the FAR 25.571³ and the AC 20 107 B.⁴

A considerable number of studies have been conducted to develop the strengthening and toughening technologies of the adhesively-bonded joints. For example, a European research project called Boltless Assembling of Primary Aerospace Composite Structures (BOPACS), executed by a consortium of European research partners, developed and evaluated different crack stopping concepts in order to increase damage tolerance in bonded composite joints for certification purposes.^5–7

Besides the efforts made by BOPACS, some scholars have recently examined and found the beneficial effects of the toughness of adhesive,^8–13 hybrid bondline,^14–16 chicken bolt,^17–20 and corrugation design.²¹ Beyond these, more focus is given on the advantages of the through-the-thickness reinforcements. For example, the Welding Institute (TWI) developed the Comeld™ joining method for composite-metal structures, which produced short pins by Cold Metal Transfer method on the joint area of a metallic structure and increased the load carrying capacity and energy absorption of the joint significantly.²² Many other scholars^23–33 used the modern manufacturing techniques to produce an array of pins onto a metallic substrate and investigated the structural performance of the metal-composite joint. However, a major problem with these methods is that they are just suitable for uncured composites because the pins cannot be embedded into the cured composite plates. Therefore, it results in a certain degree of limitations in practical applications, such as bonding repair of the damaged structures.

Lobel et al.³⁴ presented a strengthening method to composite bonded joints by implementing staples as load-bearing elements. Both conventional staples and purpose-made stainless steel wire staples were used as fastener elements, and the influence of the element diameter on the tensile strength was evaluated. Heimbs et al.³⁵ proposed a novel reinforcement technique in the through-thickness direction using metallic arrow-pins to increase failure resistance and damage tolerance. They also studied the failure behavior of composite T-joints under quasi-static and high-rate dynamic loading. The novel concept with the arrow-pin reinforcement showed a significantly increase on the post-damage load levels and energy absorption capability with the pins being pulled out of the laminate under large global deformations. Nogueira et al.³⁶ introduced an innovative technology denominated Redundant High Efficiency Assembly (RHEA) joints as a high-performance lightweight joint that combines damage tolerance with efficient load transfer. Through experimental analysis, it was found that the composite joint with RHEA insert represents a performance improvement in terms of load-bearing capability, in the endurance of high strain levels, in damage tolerance and in a longer fatigue life comparing with a co-bonded reference.³⁶ Bisagni et al.³⁷ have carried out experimental study to investigate the behavior of co-bonded carbon fiber reinforced plastics joints with a novel design incorporating a through the thickness local reinforcement. The mechanical performance of the specimens with local reinforcement, consisting of the insertion of spiked thin metal sheets between co-bonded laminates, were compared with those ones obtained from specimens with purely co-bonded joints. This novel design demonstrated by tests that damage progression under cycling load results significantly delayed by the reinforcements. Sarantinos et al.³⁸ also have researched a novel advanced joining technique, utilizing metal additive manufacturing, named μPinning. μPins are small pin-like structures manufactured on a metal substrate and used to penetrate and be consolidated inside a fiber-reinforced polymer (FRP) laminate as through-the-thickness reinforcement during curing. Numerical optimization, experimental testing was performed to validate the assumption of the importance of the μPin shape in the joint loading response.

These methods, however, are only suitable for co-curing composite parts and therefore, not applicable for the joint between composite and metallic structures. In the previous study,³⁹ the authors proposed a new joining method between metallic and composite structures to enhance the mechanical performance of metal-composite hybrid structures. The mechanism of the proposed joint is to arrest the crack in the bond line by some thin pins running through-the-thickness of the joint plates in the bonding area. The test results show that the pin reinforcements can improve the ultimate failure load by 25%, energy absorption capacity by 10.4 times, and decrease the suddenness of the failure of the joint significantly.

In this study, the damage tolerance of this new joint and the crack arrest effects of the pins are experimentally investigated. Cyclic fatigue tests are carried out on the specimens with and without prefabricated debonding and simulation work is also conducted. Thus, the effects of the reinforcing pins on the static load capacity and the failure lives of the joints with pre-inserted debonding are comparatively studied. Finite element simulations are conducted to reveal the enhancement mechanism of the reinforcing pins on the fatigue lives of the joints.

Preparation of specimens

The specimens with reinforcements and disbond features, as shown in Figure 1, were used to study the effects of the reinforcements on the structural performance of metal-composite adhesively bonded joints. The geometric features of the specimens are in accordance with the American Society for Testing and Materials Standards D 1002⁴⁰ and D 2093.⁴¹ Dimensions of the two constituent plates are the same, that is, 100 mm × 25 mm × 3 mm in length, width, and thickness, respectively. The length of the faying zone is 12 mm. There are 15 pins (3 rows and 5 columns) running through the joint plates in the faying region. The distance between rows, p, is 3 mm and the distance between columns, s, is 5 mm. There is a layer of adhesive on the surfaces of the pins and thus, the pins are bonded together with the metallic and composite joint plates after the adhesive is cured.

Figure 1.

Metal-composite adhesive bonded joint with through-thickness reinforcements and disbond (all dimensions in mm).

Four groups of specimens with different reinforcements and disbond features were manufactured, which are shown in Table 1. The pre-fabricated disbond was set to be on the metallic plate side. Six specimens were manufactured for each configuration and all the specimens were cured together in one time to ensure the cure process to be consistent. The manufacturing process is generally the same with that described in the reference article³⁹ and the only difference is the disbond was realized through put a double-layer plastic on the faying surface. The composite plates were made of carbon fiber/epoxy composite laminate with 0.188 mm nominal ply thickness and with a stacking sequence of [±45/0/90]_2s. The metallic plates were made of aluminum alloy 2A12. The metallic pins were made of 20# galvanized iron wire with the nominal diameter of 0.914 mm. J-116 adhesive film (Institute of Petrochemistry Heilongjiang Academy of Sciences) with a thickness of 0.1 mm was used to bond the composite and metallic joint plates together, and the liquid adhesive used to bond the metallic pin together with the two joint plates was HY-914 (Tianjin Haiyan). Specimens with and without reinforcements are shown in Figure 2.

Table 1.

Configurations of the specimens.

Configuration name	Pins condition	Disbond dimension (mm)
B-N	No pins	0
B-D	No pins	6
R-N	15 pins in the bonding area	0
R-D	15 pins in the bonding area	6

Figure 2.

Specimens of the adhesive bonded joints with and without reinforcements. (a) Typical specimen with reinforcements and (b) typical specimen without reinforcements.

Experimental studies

Both quasi-static and fatigue tests were conducted for all the groups of specimens. Tabs in dimension of 25 mm × 25 mm × 3 mm were attached on both sides at the ends of the specimen for grip. Quasi-static tests were carried out on a servo-hydraulic universal testing machine of type MTS E45.105 (MTS Systems Corp.) in accordance with ASTM D 5868 Standard.⁴² The rated load of the testing machine is 50 kN with a measuring error less than 0.1%. The tests were conducted in a displacement control mode with a loading rate of 0.2 mm/min. The load and displacement were recorded using the embedded transducers inside the testing machine with a sampling frequency of 5 Hz.

Fatigue tests were carried out on a servo-hydraulic universal test machine of type MTS 370.10 with a 50 kN (±0.1%) load range in accordance with ASTM D 3166 Standard.⁴³ The test setup is shown in Figure 3. The tests were in load control and the peak load was set to be 3 kN and stress ratio was 0.06 for all the tests. The cyclic loading was in sinusoidal waveform and the loading frequency of all fatigue tests was 10 Hz. The load and displacement were also measured using the embedded transducers. The strain was measured with an extensometer with a gauge length of 25 mm, which is expressed as the variation values measured by the extensometer divided by the gauge length. The load, displacement, and strain values were all recorded using a VTI EX1629 data acquisition system and the sampling frequencies were all 100 Hz.

Figure 3.

Setup of fatigue tests.

Test results and discussion

Static testing results

All the quasi-static test specimens were broken on the interface of the two joint plates. Figure 4 illustrates the failure modes of the four types of specimens listed in Table 1. From Figure 4(a) and (b) it can be seen that, as for the joints without through-the-thickness reinforcements, there are both composite fibers broken failure and debonding in the specimens without pre-fabricated disbond (Figure 4(a)), however, there is only disbond failure in the specimens with pre-fabricated disbond between the joint plates (Figure 4(b)). For all the four configurations, the dominate quasi-static failure mode of the adhesive layers between the two joint plates is adhesive failure but not cohesive failure, which indicates that the surface of substrates has not been properly treated due to manual manufacturing process. However, effect of the reinforcing pins can still be obtained through the comparisons of the mechanical performance of the joints with and without reinforcing pins. The pins were both shearing broken in the reinforced joints with and without pre-fabricated disbond.

Figure 4.

Failure modes of the joints with different reinforcement and disbond conditions after quasi-static tension tests. (a) B-N specimen, (b) B-D specimen, (c) R-N specimen, and (d) R-D specimen.

The load–displacement relationships of the specimens with the same configuration have a rather good consistency. Figure 5 shows the comparison of the typical results of all configurations. Through the comparisons of the load-displacement curves of the joints with and without through-the-thickness reinforcements, it can be concluded that the reinforcements increase the ultimate failure load and energy absorption of the joints significantly. From Figure 5, it can also be seen that the peak load values of the B-D, B-N, R-D, and R-N joints are about 3.78 kN, 4.64 kN, 4.91 kN, and 5.88 kN, respectively. This indicates that the load capacities are decreased about 18.5% and 16.5%, respectively, by the 6 mm-long pre-fabricated disbond for the joints without and with reinforcements, whereas the total energy absorptions are decreased by about 45.0% and 35.2%, respectively. The effects of the disbond on both the load carrying capacity and the fracture energy in the joints with reinforcements are lower than in those without reinforcements. This indicates that the penetrative reinforcements decrease the sensitivities of the static strength and fracture toughness to the disbond. However, from Figure 5, it can also be seen that the stiffnesses of the joints did not change much by the pre-fabricated disbond and reinforcements. This is because the total deformation of the joints mainly comes from the deformation of the sections of the joint plates outside the bonding area and out-of-plane rotation deformation of the joints due to the eccentric load of the single-lap joints, but not the shearing deformation in the bonding area.

Figure 5.

Comparison of the load-displacement curves of the joints with different configurations.

Fatigue testing results

Under the same cyclic loading, the typical fatigue lives of the different joints are shown in Table 2. From Table 2, it can be seen the fatigue lives of the joints without and with disbond were increased about 35 times and 69 times by the through-the-thickness reinforcements. The effect of the reinforcing pins on fatigue lives of the joints is dramatic and much higher than that on static load capacities. The fatigue life is decreased by 55.6% by the disbond for the joints without reinforcement, however, the reduction is just 13.7% for the joints with through-the-thickness reinforcements. This indicates that the reinforcements decrease the sensitivity of the fatigue life to the pre-fabricated disbond defects. The maximum load values of the fatigue tests are about 79.3%, 64.7%, 61.1%, and 51.0% of static load capacities of the B-D, B-N, R-D, and R-N joints, respectively. The cyclic maximum load values of the joints without reinforcing pins are closer to the relevant static ultimate load values than those of the joints with reinforcing pins, and this partially explains the reason that fatigue lives of the former joints are very short and the latter ones are much longer.

Table 2.

Fatigue lives of the joints with different configurations.

Configuration	B-N	B-D	R-N	R-D
Number of cycles	4087	1813	148,006	127,712

Figure 6 shows the failure modes of the different kinds of specimens which are with/without reinforcements and with/without pre-fabricated disbond defects. Different from the failure modes after static loading, the main failure modes include both disbond and fiber breaking failure of the composite plate in the bonded joints without reinforcements. As for the bonded joints with reinforcements, some pins are broken by shearing and some others are pulled out from either plate for both the two configurations. In addition to that, there is a crack on the faying surface of the metallic plate (Fig. 6(a)). Obviously, the joints with through-the-thickness reinforcements have more complex failure modes comparing with those without reinforcements, and also the former have to consume more fatigue energy before the final failure.

Figure 6.

Failure modes of the joints with different reinforcement and disbond conditions after fatigue tests. (a) B-N specimen, (b) B-D specimen, (c) R-N specimen, and (d) R-D specimen.

Figure 7 shows the strain changing process from the beginning of the loading till the breakage of the specimen (the last cycle was not included since the strain values of the last cycle were very large). An examination of Figure 7(a) and (b) show that the wave shapes of the joints without reinforcements are kept to be in sinusoidal waveform till just several cycles before the complete failure of the joints and the maximum strain values are changed less than one thousand microns from beginning till the ultimate failure. Nevertheless, from Figure 7(d) and (e), it can be seen that the wave shapes of the joints with reinforcements are not in standard sinusoidal waveform any more before the complete failure and the maximum strain values are changed more than 20 thousand micro-strains from beginning till failure. This indicates that fatigue failure of the joints without reinforcements happens suddenly without any clear warning, however, that of the joints with reinforcements behaves as a progressive manner.

Figure 7.

Strain variation during fatigue tests. (a) B-N specimen, (b) B-D specimen, (c) R-N specimen, and (d) R-D specimen.

The joint stiffness, which is expressed as the stress variation on the bonding surface divided the strain variation measured by the extensometer and shown in Equation (1) below, was used to evaluate the load transfer properties of the joints. And the change of the joint stiffness was used to evaluate the degradation of the joint with fatigue test cycles

E = \frac{F_{m a x} - F_{m i n}}{S \cdot (ε_{m a x} - ε_{m i n})}

(1)

where

F_{m a x}

and

F_{m i n}

denote the maximum and minimum load values in each loading cycle;

ε_{m a x}

and

ε_{m i n}

denote the strain values related to the maximum and minimum load in each cycle; S denotes the area of the faying zone, which is 317.5 mm² (12.7 mm×25 mm) in all these four cases.

Figure 8 shows the change in joint stiffness from the beginning to the end of the loading cycle of the four types of joints. From Figure 8, it can be seen the joint stiffness of the bonded joints without reinforcements decreases in a steep rate till the fatigue failure. On the contrary, the joint stiffness of the bonded joints with reinforcements after the initial steep drop decreases in a much slower pace with significantly more cyclic numbers before the final steep drop at the end of the fatigue test. This also indicates that the joints with reinforcements have a much lower progressive degradation process in most of the fatigue lives and the joints with through-the-thickness reinforcements have higher stability as regards load transfer performance.

Figure 8.

Change in joint stiffness during the fatigue tests.

Finite element simulations

The finite element modelling

In order to reveal the enhancement mechanisms of the through-the-thickness reinforcing pins on the fatigue lives of the composite-metal adhesively bonded joints, finite element models of the joints with/without pins and with/without pre-inserted disbond were developed using commercial code, Abaqus,⁴⁴ which are shown in Figure 9. Longitudinal tensile load was applied on the joints. Eight-node linear brick elements with reduced integration and hourglass control, C3D8R, were used to simulate the plates. The steel pins were modeled as C3D8R elements as well and the cylindrical surfaces of the pin was connected to the interior cylindrical surface of the hole through a layer of cohesive elements COH3D8, which are shown in Figure 9(a) and (b). As shown in Figure 9(c), a layer of 0.1 mm thick 8-nodeed three-dimensional cohesive elements, COH3D8, were tied to the faying surfaces of the joint plates, which was used to simulate the bonding interface between the composite and metallic plate. A layer of 0 mm thick COH3D8 elements locate between the outer surface of the thin pins and the walls of holes on the joint plates, and the two faces of the cohesive elements were tied to the exterior cylindrical surface of the pin and the interior cylindrical surfaces of the hole correspondingly. The cohesive elements in the corresponding location were deleted to simulate the disbond between the composite and metallic joint plates, as shown in Figure 9(d).

Figure 9.

Finite element modelling method of the joints. (a) FE model of the joint with through-the-thickness reinforcements, (b) Local view of the FE model of the joint with through-the-thickness reinforcements, (c) Cohesive elements in the FE model, (d) FE model of the joint with through-the-thickness reinforcements, and (e) FE model of the joint without through-the-thickness reinforcements.

The material of the metallic joint plates is aluminum alloy and its elastic modulus is 70 GPa and Poisson’s ratio is 0.3. The steel pins’ elastic modulus is 210 GPa and Poisson’s ratio is also 0.3. The material parameters of the composite joint plates and two different kinds of adhesive are shown in Tables 3–5, respectively. The failure criterion of the cohesive elements under mixed-mode is governed by an interaction law of the energies which cause failure in the single (normal and two shear) mode, which is shown in Equation (2) as below

\frac{G_{I}}{G_{I C}} + \frac{G_{I I}}{G_{I I C}} + \frac{G_{I I I}}{G_{I I I C}} = 1

(2)

where G_IC, G_IIC, and G_IIIC refer to the critical fracture energies required to cause failure in the normal, the first and the second shear direction, respectively. The damage factor D (SDEG: Scalar stiffness degradation, 0 ≤ D ≤ 1) was introduced to characterize the degree of damage for the cohesive element. The detail modelling methods of the pins, plates, their interactive relationships and failure criteria of the cohesive elements can be referred to the article.³⁹ The modelling method of the bonded joints without through-the-thickness reinforcements is similar with that of the joints with reinforcements, except the former model has no pins and relevant cohesive elements, as shown in Figure 9(e).

Table 3.

Material properties of CFRP.³⁹

E₁₁,GPa	E₂₂, GPa	E₃₃, GPa	ν ₁₂	ν ₁₃	ν ₂₃	G₂₃, GPa	G₁₃, GPa	G₁₃, GPa
121	9.93	9.93	0.32	0.32	0.34	4.75	3.6	3.6

Table 4.

Material parameters of adhesive HY-914.³⁹

t_n, MPa	t_s, MPa	t_t, MPa	GIC, J/m²	GIIC, J/m²	GIIIC, J/m²
10	10	10	250	670	670

Table 5.

Material parameters of adhesive Hysol EA9696.³⁹

t_n, MPa	t_s, MPa	t_t, MPa	GIC, J/m²	GIIC, J/m²	GIIIC, J/m²
35	35	35	440	850	850

Nonlinear static general algorithm in Abaqus was used to calculate the quasi-statical tensile failure process. All simulations were run till the load drop since the failure process of the cohesive elements was in a progressive way and calculation of the cohesive element’s failure was very time consuming. This indicates the pinned models were not run after the full failure of the cohesive interface between the composite plate and the metallic plate.

Simulation results and analyses

The comparisons of the relationships of load versus displacements from testing and modelling are shown in Figure 10, from which it can be seen that the simulation results correlate the experimental results reasonably well. Basic features of the joints, such as the initial stiffness, onset of damage, and the peak load have been captured by the finite element modelling and are quite close the test results.

Figure 10.

Load-displacement curves of the test results and simulation results.

The failure indexes (scalar stiffness degradation values) of the cohesive elements corresponding to the external load of 3 kN are shown in Figure 11. From the figure, it can be seen the peak failure index values of cohesive elements in the models with the same through-the-thickness pins but different disbond conditions are almost the same, which gives the reason that the joints with and without disbond have very similar fatigue lives. It can also be seen that the peak failure index values of the joints with through-the-thickness reinforcements are much lower than those without through-the-thickness pins, moreover, the number of elements with high failure index values is much less than the joints without through-the-thickness reinforcements. This indicates, under the same outer load, there are less area of adhesive in high failure risk status in the joints with reinforcing pins than those without reinforcing pins, besides, the risk levels are much lower. These further explain the reason that the fatigue lives are increased dramatically by the reinforcing pins.

Figure 11.

Failure indexes of the cohesive elements in different joint models. (a) B-N joint (b) B-D joint, (c) R-N joint, and (d) R-D joint.

Conclusions

The effects of the penetrative thin pins on the damage tolerance performance of the composite-metal adhesively bonded joints have been investigated through the quasi-static and cyclic tensile tests as well as the finite element modelling. Based on the outputs from both experimental work and numerical simulations presented, the following conclusions can be drawn:

1) The through-the-thickness reinforcing method of the composite-metal adhesively bonded joints can not only improve the static mechanical performance such as the ultimate failure load, failure strain, and energy absorption but also increases the fatigue lives of the joints significantly. Moreover, the benefits of the reinforcing pins on fatigue lives are much higher than that on static load capacities.

2) The through-the-thickness reinforcements decrease the sensitivity of the joints on the disbond defect in the joint region, and can withstand further loading after the initial failure at the joint interface. This indicates the usage of reinforcing pins in bonding region increase the damage tolerance performance of the adhesively bonded joints and leave the airplane inspectors more chances to detect the damage and therefore, increase the safety level of the composite-metal bonded joints.

3) The degradation rate of the load transfer properties of the joint with cyclic cycles was decreased by the reinforcing pins and thus, the joints with through-the-thickness reinforcements have higher stability in stiffness.

4) Comparing with the joints without reinforcements under the same cyclic load, both the peak failure index values and the area of adhesive in high failure risk in the joints with through-the-thickness reinforcements are much lower than those without reinforcements, and this explains the enhancement mechanism of the through-the-thickness reinforcements on the fatigue lives of the composite-metal adhesively bonded joints.

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 No. 52075326).

ORCID iD

Longquan Liu

References

Budhe

Banea

de Barros

, et al. An updated review of adhesively bonded joints in composite materials. Int J Adhes Adhesives 2017; 72: 30–42.

Adams

. Adhesive Bonding: Science, Technology and Applications. Oxford: CRC Press, Woodhead Pub, 2005.

U.S. Department of Transportation – Federal Aviation Administration (FAA) . FAR 25.571 Damage—Tolerance and Fatigue Evaluation of Structure, 2018, www.FAA.gov

U.S. Department of Transportation – Federal Aviation Administration (FAA) . Advisory Circular 20-107B Subject: Composite Aircraft Structure, 2018, www.FAA.gov

Boltless assembling Of Primary Composite Structures (BOPACS). Description of Work. BOPACS, 2012. EU FP7 Research Project: Boltless assembling Of Primary Composite Structures (BOPACS). https://cordis.europa.eu/project/id/314180 (Accessed 2022).

Kruse

Körwien

Heckner

, et al. Bonding of CFRP primary aerospace structures-crackstopping in composite bonded joints under fatigue. In: 20th International Conference on Composite Materials. Copenhagen, 19–24 July 2015.

Holzhüter

Löbel

Hühne

, et al. The Adhesive Zone Mix Disbond Arrest Feature Results (EU-FP7 Project BOPACS). Meeting on Use of Bonded Joints in Military Applications, Turin, Italy, 15-16 April, 2018.

Fenner

Daniel

. Hybrid nanoreinforced carbon/epoxy composites for enhanced damage tolerance and fatigue life. Composites A: Appl Sci Manufacturing 2014; 65(10): 47–56.

Becu

Maazouz

Sautereau

, et al. Fracture behavior of epoxy polymers modified with core‐shell rubber particles. J Appl Polym Sci 2015; 65(12): 2419–2431.

10.

Huang

Rapoff

Haftka

. Arresting cracks in bone-like composites. In: Aiaa/asme/asce/ahs/asc Structures, Structural Dynamics, and Materials Conference, Norfolk, Virginia, 07–10 April 2003.

11.

Khashaba

Aljinaidi

Hamed

. Fatigue and reliability analysis of nano-modified scarf adhesive joints in carbon fiber composites. Composites B Eng 2017; 120: 103–117.

12.

Sun

Shi

, et al. Short-aramid-fiber toughening of epoxy adhesive joint between carbon fiber composites and metal substrates with different surface morphology. Composites B: Eng 2015; 77: 38–45.

13.

Khoramishad

Razavi

SMJ

. Metallic fiber-reinforced adhesively bonded joints. Int J Adhes Adhesives 2014; 55: 114–122.

14.

Löbel

Holzhüter

Sinapius

, et al. A hybrid bondline concept for bonded composite joints. Int J Adhes Adhesives 2016; 68: 229–238.

15.

Yousefsani

Tahani

. Relief of edge effects in bi-adhesive composite joints. Composites Part B: Eng 2017; 108: 153–163.

16.

Khan

Kumar

. Migrating edge effects in adhesively bonded composite tubular lap joints under axial, pressure and torsional loading via stiffness-tailoring. In: 21st International Conference on Composite Materials, Xi’an, 20–25 August 2017.

17.

Chi

HEC

Gray

Lin

. Design and Optimization of an Axial Mode II Crack Arrest Specimen. In: Aiaa/asme/asce/ahs/asc Structures, Structural Dynamics and Materials Conference, Honolulu, Hawaii, 23–26 April 2012.

18.

Sachse

Pickett

Essig

, et al. Experimental and numerical investigation of the influence of rivetless nut plate joints on fatigue crack growth in adhesively bonded composite joints. Int J Fatigue 2017; 105: 262–275.

19.

Action

Engelstad

. Crack arrestment of pi-joined bonded composites. In: 55th AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference, National Harbor, Maryland, 13–17 January 2014.

20.

Kumar

Babu

. Investigation analysis of crack growth arresting with fasteners in hybrid laminated skin-stiffener joint. Appl Compos Mater 2017; 25: 1–14.

21.

Son

H-G

Park

Y-B

Kweon

J-H

, et al. Fatigue behaviour of metal pin-reinforced composite single-lap joints in a hygrothermal environment. Compos Structures 2014; 108: 151–160.

22.

Tserpes

Peikert

Floros

. Crack stopping in composite adhesively bonded joints through corrugation. Theor Appl Fracture Mech 2016; 83: 152–157.

23.

Guild

Hogg

. Comeld^TM joints: a novel technique for bonding composites and metal. Rare Metal Mat Eng 2009; 38(3): 134–141.

24.

Zhang

Wen

Cui

. Study on the strength prediction model of Comeld composites joints. Composites Part B: Eng 2012; 43(8): 3310–3317.

25.

Stelzer

Ucsnik

Pinter

. Fatigue behaviour of composite-composite joints reinforced with cold metal transfer welded pins. Int J Fatigue 2015; 81: 37–47.

26.

Stelzer

Ucsnik

Pinter

. Strength and damage tolerance of composite-composite joints with steel and titanium through the thickness reinforcements. Composites Part A: Appl Sci Manufact 2016; 88: 39–47.

27.

Stelzer

Ucsnik

Pinter

. Composite-composite joining with through the thickness reinforcements for enhanced damage tolerance. Mater Sci Forum 2015; 825-826: 883–890.

28.

Xiong

Wang

Dear

, et al. The effect of protrusion density on composite-metal joints with surfi-sculpt reinforcement. Compos Structures 2017; 180: 457–466.

29.

Mountasir

Hoffmann

Cherif

, et al. Mechanical characterization of hybrid yarn thermoplastic composites from multi-layer woven fabrics with function integration. J Thermoplastic Compos Mater 2012; 25(6): 729–746.

30.

Wang

Ahn

Kaboglu

, et al. Characterisation of composite-titanium alloy hybrid joints using digital image correlation. Compos Structures 2016; 140: 702–711.

31.

Parkes

Butler

Meyer

, et al. Static strength of metal-composite joints with penetrative reinforcement. Compos Structures 2014; 118(1): 250–256.

32.

Nguyen

ATT

Amarasinghe

Brandt

, et al. Loading, support and geometry effects for pin-reinforced hybrid metal-composite joints. Composites Part A: Appl Sci Manufacturing 2017; 98: 192–206.

33.

Eberl

Avila Gray

Zaremba

, et al. The effect of fiber undulation on the strain field for pinned composite/titanium joints under tension. Composites Part A: Appl Sci Manufacturing 2017; 103: 148–160.

34.

Löbel

Kolesnikov

Scheffler

, et al. Enhanced tensile strength of composite joints by using staple-like pins: working principles and experimental validation. Compos Structures 2013; 106(2013): 453–460.

35.

Heimbs

Nogueira

Hombergsmeier

, et al. Failure behaviour of composite T-joints with novel metallic arrow-pin reinforcement. Compos Structures 2014; 110(4): 16–28.

36.

Nogueira

A C

Drechsler

Hombergsmeier

. Analysis of the static and fatigue strength of a damage tolerant 3D-reinforced joining technology on composite single lap joints. In: Aiaa/asme/asce/ahs/asc Structures, Structural Dynamics and Materials Conference, Honolulu, Hawaii, 23–26 April 2012.

37.

Bisagni

Furfari

Pacchione

. Experimental investigation of reinforced bonded joints for composite laminates. J Compos Mater 2018; 52(4): 431–447.

38.

Sarantinos

Kostopoulos

Di Vita

, et al. Micro-pins: the next step in composite-composite and metal-composite joining. CEAS Space J 2019; 11: 351–358.

39.

Tang

Liu

. A novel metal-composite joint and its structural performance. Compos Struct 2018; 206(15): 33–41.

40.

ASTM International . ASTM D 1002-10: Standard Test Method for Apparent Shear Strength of Single-Lap-Joint Adhesively Bonded Metal Specimens by Tension Loading. West Conshohocken, PA, USA: ASTM International, 2010.

41.

ASTM International . Designation: D2093—03 (Reapproved 2017):Standard Practice for Preparation of Surfaces of Plastics Prior to Adhesive Bonding. West Conshohocken, PA, USA: ASTM International, 2017.

42.

ASTM International . ASTM D5868-01(2014) Standard Test Method for Lap Shear Adhesion for Fiber Reinforced Plastic (FRP) Bonding. West Conshohocken, PA, USA: ASTM International, 2014.

43.

ASTM International . ASTM D3166-99(2012) Standard Test Method for Fatigue Properties of Adhesives in Shear by Tension Loading (Metal/Metal). West Conshohocken, PA, USA: ASTM International, 2012.

44.

Hibbitt

Karlsson

Sorensen

. SIMULA. ABAQUS Version 6.6. Documentation. SIMULA, 2014. http://130.149.89.49:2080/v6.14/ (Accessed 2022).