Accepted Manuscript

A novel metal-composite joint and its structural performance

Huaqing Tang, Longquan Liu

PII: S0263-8223(17)33385-8
DOI: https://doi.org/10.1016/j.compstruct.2018.07.111
Reference: COST 10029

To appear in: Composite Structures

Received Date: 24 October 2017

Revised Date: 7 June 2018
Accepted Date: 30 July 2018

Please cite this article as: Tang, H., Liu, L., A novel metal-composite joint and its structural performance, Composite
Structures (2018), doi: https://doi.org/10.1016/j.compstruct.2018.07.111

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
 A novel metal-composite joint and its structural performance

Huaqing Tanga, Longquan Liua*

a School of Aeronautics and Astronautics, Shanghai Jiao Tong University, 800 Dongchuan RD. Shanghai,
China, 200240.
Corresponding author:
Longquan Liu
School of Aeronautics and Astronautics, Shanghai Jiao Tong University
800 Dongchuan RD, Minhang District, Shanghai, 20040, PR China.
Telephone: 86-13901761419
Email: [email protected]

Abstract

A novel joining method between metallic and composite structures was developed to enhance the

mechanical performance of metal-composite hybrid structures. According to this joining method, the

metallic and composite components are adhesively bonded together and there are also some thin pins in

the overlap region running through the joint plates. The pins are bonded together with the joined

components as well. Comparatively tensile tests were conducted on the metal-composite joints so as to

investigate the advantages of the proposed joining method. The load-displacement relationships, fracture

modes and fatigue life were analyzed in accordance with the test results of the two different types of

joints. The results show that the proposed joining method can improve ultimate failure load, failure

displacement, energy absorption capacity and fatigue life significantly. Furthermore, the novel joining

method decreases the suddenness of the failure of the joint and provides some plastically like behavior

to the joint. Finite element models were developed to investigate the enhancement mechanism of the

proposed method. Through the analysis of the failure process, it can be concluded that there are bridging

force between the pins and the jointed components and the bridging force not only transfers load between

the components together with the adhesive layer, also inhibits the adhesive layer from peeling.

Keywords: Adhesive-multi pin joint; Composite-to-metal; Tensile test; Finite element analysis; Failure

1
process; Enhancement mechanism

1 Introduction

The application of carbon fiber reinforced polymers (CFRP) in the aerospace industries has become

increasingly popular for their desirable mechanical and physical properties. Composite materials were

accepted as the first choice of structural materials particularly in situations where weight saving is an

essential consideration. In many large-scale applications, a total composite solution may be unrealistic

and thus, composites must be combined with metals [1]. Reliable application of composite materials

depends on reliable joining of the structure, which means the strength of joints between composites and

metallic materials is the key issue. The correct choice of joining technique is essential to providing a safe

load-transfer at low weight. Hence, building a high-strength interface between metallic and FRP parts is

an important technology and has gained more and more interests from researchers [2].

Adhesive bonding and mechanical fastening are two major joining methods for composite materials.

Adhesive bonding is a non-disassembled joining technique for two dissimilar materials and is widely

used in the connection of composites and metals. However, the failure of adhesively bonded composites

usually results from the transverse failure of the adhesive layer. Such failure arises from the transverse

peel stress which exists in the ends of adhesive joints for any geometry [3]. Furthermore, the failure of

adhesive bonded joints often occurs unexpectedly and abruptly due to the high stress concentration at the

ends of the adhesive joints [4]. Therefore, the adhesive joints cannot meet the requirements of damage

tolerance in transport category airplanes up to now.

Mechanical fastening involves the process of drilling holes into the structures. The strength reduction

due to these holes for composite structures is much greater than that of metals because of the high notch

sensitivity of composite laminates [5]. In addition to that, the use of a large number of fasteners leads to
2
high cost and long period for research and manufacturing process, and brings an increase in structural

weight as well. This may weaken the reduction of structural weight which benefits from the use of

composite materials.

The adhesive-bolts hybrid joining method overcomes the shortcomings of bonding and fastening

joining methods to some content. Kelly [6] investigated the strength and fatigue life of hybrid

(bonded/bolted) joints with carbon fiber reinforced plastic adherends. Hybrid joints presented greater

strength, stiffness and fatigue life in comparison with adhesive bonded joints. Zhao et al. [7] compared

bonded connection, bolted connection and hybrid connection. The results showed that hybrid joints have

higher stiffness, better ductility, and reasonable stress distribution. However, those additional fasteners

have to be capable to carry the limit load in case of a global failure of the bondline when the hybrid joint

is used as primary joints [8]. Design benefits that come along with adhesive bonding do not come into

effect, since fastening elements must be taken into account for part design and the use of fasteners also

increases the cost and weight of the structure.

In order to improve both the reliability and structural efficiency of the joints, scholars continued to

develop a variety of three-dimensional reinforcement technologies for composite materials. Application

of Z-pin technology is one of the most effective way for the improvement of connection performance.

Uncured prepreg that implanted a certain density of slender rods (called Z-pin) along the thickness

direction can significantly improve the damage tolerance, interlayer toughness, in-plane shear strength

and other properties of composite laminates [9-12]. Nevertheless, the Z-pin cannot be embedded in the

metallic plate, which makes this technique be not suitable to reinforce the joints between composite and

metallic structures. The invention of Surfi-Sculpt™ by The Welding Institute (TWI) provides the basis

for an integrative hybrid metal-composite joining method, Comeld™ joining method [13]. Some scholars

3
[14-18] investigated the structural performances of the joint structures, which connects fibers reinforced

composite materials and metals by pre-treating the metallic joint component using the Surfi-Sculpt™

surface treatment technology [19]. They also investigated the load carrying capacity of Comeld™

composite-metal joints by using experimental and numerical methods. It was found that the protrusion

of the surface reduced the stress concentration of the joint, and thus improve the carrying capacity of the

joint. Another promising manufacturing technology to produce these small pins is a welding technique

called Cold Metal Transfer (CMT) [20-21]. However, these two technologies are both just suitable for

uncured composites because of the pins cannot be embedded into the cured composite plate. Therefore,

it results in a certain degree of limitations in practical application.

In a word, conventional joining techniques have their major drawbacks. This paper will present a new

joining method between metallic and composite structures, which can improve the reliability and

structural efficiency of the metal-composite joints. Metal-composite joints made by traditional adhesive

bonded method and new joining method were both tensile tested for comparison to investigate the

advantages of the proposed joining method. The load-displacement relationships and failure modes of

the joints were analyzed in detail. A finite element modelling approach was developed and proposed to

analyze the failure process and strengthen mechanism. The good agreement between the numerical

results and experimental data testifies the validity of the proposed model. The results show that the

proposed joining method improves the ultimate failure load, failure strain, and energy absorption capacity

significantly. The novel joining method decreases the suddenness of the failure of the joint.

2 Adhesive-multi pin joining method

2.1 Working principle

The joint proposed in this paper(we name it as adhesive-multi pin joint) is integrated with adhesion
4
which bonds the metallic and composite plates, and an array of pins which run through the joined plates

and distribute in the overlap region as depicted in Fig. 1. The pins are bonded with the joined plates by

adhesion. The arrangement of pins can be decided in accordance with the specific joint and the level of

the eternal load.

2.2 Specimens and materials

A typical composite to metal adhesive-multi pin single lap joint that designed referring to ASTM D

1002-10 test standard [22] is shown in Fig. 2. Dimensions of the plates are all 100 mm × 25 mm × 3 mm

in length, width and thickness. In the Fig.2, p is the row distance of the pins and s is the column distance.

Three different arrangements of the pins will be investigated in this study, which named as Array-1,

Array-2 and Array-3 as shown in Fig. 2. As for Array-1, there are three rows of pins and five metallic

pins in each row. These pins were equally distributed with a p equaling to 3.0 mm and s equaling to 5

mm. The other dimensions are shown in Fig. 2(a). As for Array-2, there are two rows of pins and also

five metallic pins in each row. As for the Array-3, there are two rows of pins and three pins in each row.

The geometric features and dimensions of the joint with Array-2 and Array-3 are shown in Fig. 2(b) and

(c), respectively.

The composite plate is made of carbon fiber/epoxy composite laminate with 0.188 mm nominal ply

thickness and the laminate’s stacking sequence is [±45/0/90]2s. The metallic joint plate are made of LY12

aluminum alloy (LY12 is a Chinese aluminum alloy serial number and the material properties are

equivalent to 2024-T3). The metallic pins are made of galvanized iron wire with serial number of 20 (the

nominal diameter of 0.914 mm). The mechanical properties of the unidirectional lamina are listed in

Table 1, where Eij (i= 1, 2, 3) denote the elastic modulus in the three principle directions, respectively.

Gij (i =1, 2, 3) denote the shear modulus in the three principle planes, and νij (i= 1, 2, 3) denote the

5
Poisson’s ratios, respectively. The material properties of the aluminum alloy used in this study can be

referred to the Mil-handbook 5J and the elastic modulus of the iron wire is 68 GPa, yield strength is 260

MPa, ultimate strength is 400 MPa and elongation is 20%. The adhesive used to bond the metallic pin

with the joint plates is HY-914 and that between the two adherends is J-116. Their mechanical properties

are listed in Table 2 and Table 3, respectively. Parameters of adhesive are listed in the section of numerical

simulation.

2.3 Manufacturing of test specimens

The adhesive-multi pin joints were manufactured in accordance with the following procedures. Firstly,

surface treatment was implemented in the overlap region of metallic and composite plates with desired

size in accordance with ASTM D 1002 and D 2093 standards [22, 23], which includes acetone wiping,

sanding and wiping with a clean dry cloth. The after-treated bonding surfaces were wiped used three

successive portions of a clean acetone-dampened cloth. The adherends were kept standing in a clean,

dust-free area with bonding surfaces upwards for 20 minutes to allow evaporation of solvents. Then the

bonding surfaces of the adherends were ground with a fine-grit sandpaper till no evidence of surface

gloss is visible. The particles were removed by wiping with a clean dry cloth.

After surface treatment finished, one piece of J-116 adhesive film with the same size of the overlap

region was pasted on the treated faying surfaces of the two joint plates and then, the joint plates were

clamped together using two foldback clips so as to keep the relative position of the joint plates. Then an

array of through holes (φ=1 mm) were drilled in the overlap region of the joined plates and HY-914

liquid adhesive was filled into the small holes. After that, fifteen metallic pins dipped into HY-914 liquid

adhesive with the adhesive spreading on the outer surfaces and then, the pins were inserted into the holes.

Finally, the joint assemblies were placed into an oven to keep the temperature being 120℃ for one hour

6
to cure the adhesive film and the liquid adhesive. The adhesive-multi pin joints will be successfully

manufactured as shown in Fig. 3

For comparison, adhesive bonded joint specimens with the same geometry as the newly introduced

joint specimen were manufactured using the similar procedures and were chosen as a reference. The

traditional adhesive joint specimens and the adhesive-multi pin joint specimens are shown in Fig. 3,

which are named as J 1-n and N 1-n (n=1, 2, 3, 4, 5) here and after in this article, respectively.

3 Experimental tests

Tensile tests were conducted on a universal material testing machine of type MTS E45.105 in

accordance with the test standard of ASTM D 1002-10 [22]. The load range of the testing machine is

from -100 kN to 100 kN and the relative tolerance of the measured load is ±1%. Specimens were clamped

vertically by the grips of the testing machine as shown in the Fig. 4. The clamping length is 25 mm at

the both ends. Pads with the same thickness of the joined plates were attached on both ends of the

specimens so as to ensure the longitude axis of the test specimen to be coincided with the direction of

applied load. The load was displacement-controlled with a free crosshead velocity of 0.2 mm/min to

eliminate the dynamic effect of loading. The load and displacement were recorded using the embedded

transducers with a sampling frequency of 5 Hz.

In addition to tensile tests, fatigue tests were also carried out to compare the mechanical properties

between the traditional adhesive structure and the novel structure under alternative loading. Fatigue tests

were conducted on a universal fatigue testing machine of type MTS-370.05, whose load capacity is a 50

kN. The load was applied repeatedly by using the load control method with a frequency of 10 Hz, a

maximum load of 3 kN and the stress ratio of 0.06.

7
4 Experimental results and discussion

4.1 Load-displacement curve

The load–displacement relationships of the two different types of joints shown in Fig. 5. From Fig. 5,

it can be seen that the results of the both types of specimens have good consistency. The comparison of

each typical experimental result is shown in Fig. 5(c). From Fig. 5(c), it can be seen that the adhesive-

multiple pin joints have a significantly greater load carrying capability than the adhesive joints.

Comparing with the traditional joint, the stiffness of the proposed joint is almost unchanged, and the

ultimate load of novel joint is increased by 25%. The area under the load–displacement curves represents

the energy absorbed before the finial failure of the specimens. For both types of joints, the energy

absorbed by adhesive-multi pin joint (10.71 J) is more than 10 times than that absorbed by the adhesive

joint (0.94 J). Moreover, the adhesive joint fails at the bond line and these failures are catastrophic, as

indicated by the deep decrease in load of the corresponding curves.

The failure of the adhesive-multi pin joint is much more progressive than that of the traditional

adhesive joint. The load-displacement curves of adhesive-multi pin joint exhibit greater changes in slope

than that of the adhesive joints. In addition, the load drop-off of the adhesive-multi pin joint that

represents the final failure appears much later than that of the traditional joint.

The hybrid joints of Array-2 and Array-3 as shown in Fig. 2 were also tensile tested at the same test

setup. The load-displacement curves of the novel joints with different configurations and the traditional

adhesive joints are shown in Fig. 6. It can be seen that the ultimate load capacity of the joints are not

obvious changed with the number of metallic pins. However, the final failure displacement and the energy

absorption capacity is obviously decreased with the number of pins. The decrease of the number of

metallic pins leads to the failure of interface adhesive layer, and the metallic pins does not have enough

8
load capacity, resulting in a large drop in load and a significant decrease in energy absorption capacity.

Test results prove that metallic pins in the overlapping region play a major role in carrying capacity.

Different structural design can be chosen according to different conditions

4.2 Fracture mode of joints

The fracture modes of the two types of joints are different, as shown in Fig. 7(a) and (b). From Fig. 7,

it is clear that adhesive joint was broken under the shear force. The adhesion between the metallic and

the composite plates was subjected to cohesive destruction. There was very little damage to the metallic

plates as shown in Fig. 7(a). The surface of the composite plate is destroyed by the shearing force. For

adhesive-multi pin joint, adhesion do not fail at lower loads condition, and the damage to composite

material occurs when loads reach the failure strength. From Fig. 7(b), it can be found that the composite

material was destroyed at the ply angle of 45°. Moreover, some metallic pins deformed significantly and

some others broken completely. The adhesive-multi pin joint eventually failed due to fracture failure of

metallic pins and shear failure of the composite, which contributed to different mechanical behaviors.

4.2 Fatigue tests result

Under the same fatigue load, the number of cycles to failure of different specimens is shown in Table

4. The test results show that, the cycles to failure of the composite-metal hybrid structure is increased by

more than 74 times than that of the traditional adhesive joint structure. It can be seen that the metallic

pins improve the fatigue life of joining structures.

5 Numerical simulation

5.1 Meshes and boundary conditions

The simulations were carried out using the finite element code, ABAQUS of version 6.13. A finite

element model of the adhesive-multi pin joint was built to analyze the structural performance, as
9
schematically illustrated in the Fig. 8. Both the metallic and composite plates and the pins were modelled

as C3D8R elements. The adhesive layer on the faying surfaces of the plates and around the pins are

modeled as cohesive elements, COH3D8. The composite laminate is modelled with four layers of

elements in the thickness direction, and each layer of elements represent four plies of lamina. The

mechanical properties of each ply were calculated in accordance with the local fiber orientations related

to the main coordinate axis. The adhesive on the surface of the pins was modeled as a layer of 8-node

cohesive elements (COH3D8) with a thickness of zero. The nodes on each side of the zero-thick cohesive

elements were tied together with the pins and hole walls, respectively. The nodes of plain adhesive

element are tied with two adherends of the joint, of which the thickness is 0.1 mm.

The boundary conditions are shown in Fig. 8 (a). The left end of steel plate is fixed in all three

translational directions (Ux, Uy and Uz). The right end of the laminate plate is declared as a rigid body

and has tie relationship with a reference node. Therefore, the motion of the right end surface is governed

by the motion of the reference node, which is fixed in two translational directions (Uy and Uz), while a

pull load is applied along Ux direction.

5.2 Failure criteria of the cohesive elements

The parameters used for the adhesive are shown in Table 2 and Table 3. Cohesive elements were used

to model the debonding crack along the bondline. A bilinear traction-separation law was used to define

the constitutive response of cohesive elements, as shown in Fig. 9.

Here, t n0 represents the peak value of the nominal stress when the deformation is purely normal to

the contact surface, t s0 represents the peak value of the nominal stress when the deformation is purely

in the first shear direction, and t t0 represents the peak value of the nominal stress when the deformation

is purely in the second shear direction. Likewise, δn , δ s and δ t represent the values of the nominal

10
strain when the deformation is purely normal, purely in the first and the second directions to the contact

surface, respectively. In ABAQUS, the damage factor D [24] (SDEG: Scalar stiffness degradation, 0 ≤ D

≤ 1) is introduced to characterize the degree of damage for the cohesive element. The stiffness coefficient

(K) in damage evolution is expressed by

K =(1−D) K0 (1)

where K0 is the stiffness of complete material. When D equals to 0, it means that the material does not

yield or has just begun to yield; when D equals to 1, it means the material has been damaged and the load

carrying capacity is lost.

5.3 Model validation

The comparison of the simulated and experimental load-displacement curve of the joint without pins

was shown in Fig. 10(a). It is observed from Fig. 10 that simulated results of FE model show good

agreement with the experimental data, which validates the cohesive finite model for the adhesive joint.

The FE model provides an accurate prediction for the joint strength (error < 2.5%). The load-

displacement curve remains linear elastic until the yield stress of the metallic plate is reached. The load

of adhesive joint increases linearly with the displacement at the beginning, then goes nonlinear with the

displacement until reaches its peak.

The numerical and experimental results of load-displacement of the adhesive-multi pin joint are shown

in Fig. 10(b). The load increases linearly in the beginning and when it accesses its peak, the adhesive

layer begins to break. From this point of view, the load is completely carried by the pin bridging force.

The numerical results are in good agreement with the experimental data in whole process. With the

increase of displacement, the metallic pin is gradually cut off or pulled out and the carrying capacity of

the joint is gradually reduced to zero.

The pin-model is able to predict the debonding of contact surface and the damage of metallic pins,

demonstrating the suitability of the modelling method in the period of whole tensile process. This finite

11
element model can be used to evaluate the effect of design parameters of pin reinforced composite single

lap joint.

5.4 Damage mechanisms

Fig. 11 shows the failure process of adhesive layer in conventional adhesive structure. In Fig. 11, the

s and F represent the displacement and force, respectively. The adhesive layer is failed when the value

of SDEG reaches to 1. The cracks of adhesive layer initiated from the two ends and spread to the middle.

Finally and quickly, the two cracks meted each other and the adhesive was completely failed. Thus the

load carrying capacity of the joint will rapidly and catastrophically reduced to zero along with the failure

of adhesive layer.

The failure process of the adhesive in adhesive-multi pin joints is shown in Fig. 12. It can be seen that

the adhesive layer on the faying surface of the two adherends is more susceptible to fail than the adhesive

layer on the metallic pin. The cracks of adhesive layer initiated from two ends and spread to the middle

slower than that of adhesive joint. Moreover, after the adhesive layer on the faying surface was broken,

the metallic pin and its adhesive layer can still carry some load. In addition, the structure does not fall

rapidly, which ensures the stability of the connecting structure and greatly improve the failure

displacement of the structure.

Therefore, it can be concluded that the improvement of mechanical behaviors between the composite

part and the metallic part mainly because pin not only inhibits the peeling of the adhesive layer, but also

transfer some load between the connected parts together with the adhesive layer.

Conclusions

In this study, an adhesive-multi pin joining method is proposed to enhance the mechanical performance

of hybrid metal-composite joints. Adhesive joints and adhesive-multi pin joints were both tested under
12
tensile loading for comparison. A finite element model was established for numerical investigation.

Based on experimental and numerical results, conclusions can be drawn as follows:

1) A new joining method between the metallic and composite materials was presented and compared

with adhesive joint, the joint reinforced by metallic pin improves the ultimate failure load, and moreover,

it increases the failure strain and energy absorption capacity much more. Furthermore, the novel joining

method decreases the suddenness of the failure of the joint.

2) The metallic pins are very thin and are bonded with the connected plates, which has little side effect

on the plates’ strength, thus overcoming the stress concentration problem caused by the conventional

mechanical connection.

3) The proposed finite element model can predict the debonding and the damage process of the joints.

It has instructive significance in the design parameters of metallic pin reinforced composite single lap

joint and reveals the enhancement mechanism of the hybrid joint as well.

4) There is bridging force between the pins passing through the plates so that the pins not only inhibit

the adhesive layer on the faying surface from peeling, but also transfer load together with the adhesive

layer on the faying surface. Therefore, the presence of the pins increase the load capacity and energy

absorbing capacity of the joints.

Acknowledgements

The authors disclosed receipt of the following financial support for the research, authorship, and/or

publication of this article: The research described in this paper was financially supported by the CALT

University Joint Foundation and Aeronautical Science Foundation of China (2016ZE57009).

References

13
1. Tu W, Wen PH, Hogg PJ, et al. Optimization of the protrusion geometry in Comeld™ joints.

Compos Sci Technol 2011; 71: 868–876.

2. Ucsnik S, Scheerer M, Zaremba S, et al. Experimental investigation of a novel hybrid metal–

composite joining technology. Compos Part A: Appl Sci Manuf 2010; 41:369-347.

3. Adams RD, Harris JA. Influence of local geometry on the strength of adhesive joints. Int J Adhes

Adhes 1987; 7(2):69–80.

4. Matthews FL, Kilty F, Godwin EW. A review of the strength of joints in fibre-reinforced plastics.

Part 2. Adhesively bonded joints. Compos 1982; 13(1): 29-37.

5. Sen F, Pakdil M, Sayman O, Benli S. Experimental failure analysis of mechanically fastened joints

with clearance in composite laminates under preload. Mater Des 2008; 29:1159–1169.

6. Kelly G. Quasi-static strength and fatigue life of hybrid (bonded/bolted) composite single-lap joints.

Compos Struct 2006; 72: 119-129.

7. Zhao XY, Huang SN, Feng P, et al. Experimental research on hybrid connecting method for FRP

constructional elements. Eng Mech 2015; 6(32): 314-321.

8. Kruse T. Bonding of CFRP primary aerospace structures: overview in the technology status in the

context of the certification boundary conditions addressing needs for development. In: Proceedings

of the 19th international conference on composite materials, Montreal, Canada, 28 July-2 August

2013, p. 5635-5643.

9. Byrda LB, Birman V. Effectiveness of z-pins in preventing delamination of co-cured composite

joints on the example of a double cantilever test. Compos Part B Eng 2006; 37: 365-378.

10. Yasaee M, Lander JK, Allegri G, et al. Hallett. Experimental characterisation of mixed mode
traction-displacement relationships for a single carbon composite Z-pin. Compos Sci Technol 2014;

94: 123-131.

11. Cui H, Li Y, Koussios S, et al. Mixed mode cohesive law for Z-pinned composite analyses. Compos
Sci Technol 2013; 75: 60-68.

14
12. Mouritz AP, Koh T M. Re-evaluation of mode I bridging traction modelling for z-pinned laminates
based on experimental analysis. Compos Part B Eng 2014; 56:797-807.

13. Dance BGI, Buxton AL. An introduction to Surfi-Sculpt technology – new opportunities, New
Challenges. In: Conference proceedings of 7th international conference on beam technology, Halle,

Germany, 17-19 April 2007, p. 75-84.

14. Buxton AL, Dance BGI. Surfi-Sculpt – revolutionary surface processing with an electron beam. In:

Proceedings of the 4th international surface engineering conference, Minnesota, USA, 1–3 August

2005, p. 107-110.

15. Freeman R, Smith F. Comeld: a novel method of composite to metal joining. Adv Mater Process

2005; 163(4): 33–40.

16. Smith F, Wylde G. Comeld – an innovation in composite to metal joining. Aust Weld J 2004; 49:

26–7.

17. Dance BGI. Rapid materials processing and surface sculpting using electron beam and laser

processes. In: TMS 2009 138th annual meeting & exhibition on proceedings, San Francisco, USA,

15–19 February 2009, Vol. 3, p. 167-174.

18. Zhang HJ, Wen W, Cui HT. Study on the strength prediction model of Comeld composites joints.

Compos Part B Eng 2012; 43: 3310-3317.

19. Mouring SE, Janowski ME, Louca LA, et al. Structural performance of Comeld hybrid metal-to-

composite joints. In: Proceedings of the Twenty-second Internatinal Offshore and Polar Engineering

Conference, Rhodes, Greece, 17–22 June 2012; p. 117-122.

20. Schierl A. The CMT process a revolution in welding technology. Weld World 2005; 49(9): 38.

21. Stelzer S, Ucsnik S, Pinter G. Fatigue behavior of composite-composite joints reinforced with cold

metal transfer welded pins. Int J Fatigue 2015; 81: 37-47.

15
22. ASTM International. ASTM D1002-10: Standard test method for apparent shear strength of single-

lap-joint adhesively bonded metal specimens by tension loading. West Conshohocken,

Pennsylvania, United States, 2010.

23. ASTM International. ASTM D2093-03(2017): Standard practice for preparation of surfaces of

plastics prior to adhesive bonding. West Conshohocken, Pennsylvania, United States, 2017.

24. Yang L, Wu ZJ, Cao Y, et al. Micromechanical modelling and simulation of unidirectional fiber-

reinforced composite under shear loading, J REINF PLAST COMP 2015; 34(1): 74-83.

16
Table 1 The mechanical properties of the unidirectional lamina

E11,GPa E22,GPa E33,GPa ν12 ν13 ν23 G23,GPa G12,GPa G13,GPa

121 9.93 9.93 0.32 0.32 0.34 4.75 3.6 3.6

Table 2 The mechanical properties of the adhesive HY-914

tn, MPa ts, MPa tt, MPa GIC, J/m2 GIIC, J/m2 GIIIC, J/m2
10 10 10 250 670 670

Table 3 The mechanical properties of the adhesive J-116

tn, MPa ts, MPa tt, MPa GIC, J/m2 GIIC, J/m2 GIIIC, J/m2
20 20 20 440 850 850

Table 4 Cycles to failure of two joints

Types of joint cycles

Novel joint of Array-1 148,312

Traditional adhesive joint 998
Fig. 1 Concept of the adhesive-multi pin joint

（a）

（b）
 （c）
Fig. 2 Geometric configurations of the specimens (all dimensions in mm）: (a) Array-1, (b) Array-2, (c)
Array-3

(a)

(b)
Fig. 3 Single-lap joint specimens: (a) traditional adhesive joints and (b) adhesive-multi pin joints

Fig. 4 Setup of the tensile test

(a)
 (b)

(c)
Fig. 5 Load-displacement curves of single lap joints: (a) load-displacement curves of adhesive joint
specimens, (b) load-displacement curves of adhesive-multi pin joint specimens, (c) comparison of the
typical load-displacement curves of the two different joints
Fig. 6 Comparison between novel joint of different array and experimental load and displacement
curves

(a)

(b)
Fig. 7 Fracture modes of the joints: (a) traditional adhesive bonded joints, (b) novel hybrid joints
 (a)

(b)
Fig. 8 Finite element model: (a) meshes and boundary conditions, (b) detail view A

Fig. 9 Constitution model of cohesive element

 (a) (b)
Fig. 10 Comparison between numerical and experimental load and displacement curves: adhesive joint
(a) and adhesive-multi pin joint (b)

(1) s=0 mm, F=0 kN (2) s=0.21 mm, F=1.8 kN

(3) s=0.35 mm, F=3.3 kN (4) s=0.39 mm, F=3.8 kN

 (5) s=0.42 mm, F=4.1 kN (6) s=0.45 mm, F=0 kN

Fig. 11 Failure process of adhesive layer of adhesive joint

(1) s=0 mm, F=0 kN (2) s=0.16 mm, F=1.5 kN

(3) s=0.42 mm, F=4.0 kN (4) s=0.59 mm, F=5.1 kN

(5) s=1.51 mm, F=4.4 kN (6) s=5 mm, F=0 kN

Fig. 12 Failure process of adhesive layer of adhesive-multi pin joint