- Email: [email protected]
Mechanical and fracture properties of epoxy adhesives modified with graphene nanoplatelets and rubber particles
Author’s Accepted Manuscript Mechanical and fracture properties of epoxy adhesives modified with graphene nanoplatelets and rubber particles Dong Quan, Declan Carolan, Clemence Rouge, Neal Murphy, Alojz Ivankovic www.elsevier.com/locate/ijadhadh
PII: DOI: Reference:
S0143-7496(17)30162-8 https://doi.org/10.1016/j.ijadhadh.2017.09.003 JAAD2054
To appear in: International Journal of Adhesion and Adhesives Received date: 13 March 2017 Accepted date: 2 September 2017 Cite this article as: Dong Quan, Declan Carolan, Clemence Rouge, Neal Murphy and Alojz Ivankovic, Mechanical and fracture properties of epoxy adhesives modified with graphene nanoplatelets and rubber particles, International Journal of Adhesion and Adhesives, https://doi.org/10.1016/j.ijadhadh.2017.09.003 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 galley proof before it is published in its final citable 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.
Mechanical and fracture properties of epoxy adhesives modified with graphene nanoplatelets and rubber particles Dong Quan a, Declan Carolan b,c, Clemence Rouge a, Neal Murphy a, Alojz Ivankovic a a b
School of Mechanical and Materials Engineering, University College Dublin, Dublin, Ireland Department of Mechanical Engineering, Imperial College London, SW7 2AZ, UK
cFAC
Technology, London, UK
_________________________________________________________________________________________________ Abstract Graphene nanoplatelets (GNP) were introduced into a rubber-modified epoxy adhesive in order to simultaneously improve the bulk mechanical properties, fracture toughness and single joint lap shear strength of the adhesive. The Young’s modulus was observed to increase marginally from 2.46 GPa to 2.56 GPa due to the addition of 0.1 wt.% GNPs. No further increase in modulus was observed for GNP loading above 0.1 wt.%. A negligible effect on the measured tensile strength was observed. The fracture energy of the bulk adhesive increased by 21 % due to the addition of 0.1 wt.% GNPs. No further increase in measured fracture energy was observed as the GNP content was further increased to 0.5 wt.%. A systematic decrease in the lap shear strength was observed due to the addition of GNPs, i.e. the lap shear strength decreased from 21.7 MPa of the control adhesive gradually to 17.2 MPa of the adhesive modified by 0.5 wt.% GNPs. Imaging analysis of the failed adhesive joints reveal that the reduction in lap shear strength was attributed to the preferential alignment of the GNPs in the direction parallel to the adhesive bonding surface. This was further confirmed by comparing the electrical behaviour of the lap shear joints with that of the bulk adhesive samples.
Keywords: Rubber-modified epoxy adhesive, Graphene nanoplatelets, fracture toughness, lap shear strength. _________________________________________________________________________________________________ 1. Introduction Epoxy polymers possess many favourable engineering properties due to their highly cross-linked structure. However, this type of structure also has short-comings, such as inherently low fracture toughness and poor resistance to fracture [1, 2]. The addition of modifiers forming a second distributed phase was found to be a promising method to improve the fracture toughness of epoxy polymers. Different types of modifiers, such as silica spheres [3–5], graphene nanoplatelets [6–8], carbon nanotubes [9, 10] and rubber particles [11–13] have been proven to improve the fracture toughness of epoxies considerably. Among different modifiers of epoxies, rubber particles demonstrate superior performance in toughness enhancement. It has been widely reported [13–17] that the addition of rubber particles can increase fracture toughness significantly. However, the addition of rubber particles results in a considerable reduction in mechanical properties. This limited the application of rubber-modified epoxies, especially for the systems with high rubber content. A possible solution to this is to create a so called hybrid epoxy, containing a blend of both soft rubber particles and rigid particles dispersed on the nano-scale. The rubber particles provide an effective toughening mechanism, while the rigid particles contribute to offset the loss in mechanical properties. In some cases, e.g. silica, the rigid particles also contribute to toughening and can provide a synergistic effect [5, 18]. The effects of graphene on the mechanical properties, electrical conductivity and fracture toughness of epoxy have been widely investigated. Tang et al. [8] studied the mechanical properties and fracture toughness of graphene modified epoxy
composites. They report that the addition of 0.2 wt.% graphene oxide into the neat epoxy increased the Young’s modulus from 2.93 GPa to approximately 3.1 GPa with the fracture toughness improved from 0.49 MPam1/2 to 0.75MPam1/2. The debonding and crack bridging of the graphene oxide platelets are identified as the main toughening mechanisms. Rafiee et al. [7] investigated the fracture and fatigue behaviour of graphene modified epoxy nanocomposites. It was observed that adding 0.125 wt.% functionalized graphene sheets increased the fracture toughness of the neat epoxy by 65 % and the fracture energy by 115 %. Moreover, incorporation of 0.125 wt.% functionalized graphene sheets was found to dramatically reduce the rate of crack propagation under fatigue conditions. The superior performance of graphene to resist fracture and fatigue was attributed to the two-dimensional structure of graphene, which enabled it to deflect cracks effectively. AhmadiMoghadam et al. [19] demonstrated that blending small amount of graphene nanoplatelets increased the fracture toughness of an epoxy by 82 %, while a concurrent improvement in Young’s modulus and tensile strength was also observed. Crack bridging was proposed as the main toughening mechanism. Wajid et al. [20] presented that, even at small graphene filler content, the epoxy composites exhibit a 10-40 % increase in the mechanical properties, and the electrical conductivity was enhanced by seven orders of magnitude. Lim et al. [21] and Chandrasekaran et al. [22] studied the thermo-mechanical, electrical and fracture properties of rubber-graphene-epoxy nanocomposites and presented that both the electrical conductivity and fracture toughness increased significantly. Based on the literature review, it is clear that blending small amount of graphene into epoxies can increase the mechanical properties and fracture toughness of the polymer considerably. However, limited work has been performed to study the effect of graphene on the fracture behaviour of epoxy adhesive joints. Moriche et al. [23] studied the influence of the incorporation of GNPs on the lap shear strength of an epoxy adhesive. They reported that the lap shear strength remained constant for
neat adhesives and adhesives modified by GNPs. This was attributed to the weak adhesion between the substrates and the adhesives, which resulted in interfacial failure of the adhesive joints. In this case, the lap shear test is measuring the adhesion between the adhesives and the substrates instead of the lap shear strength of the adhesives. Guadagno et al. [24] used epoxy adhesive to bond epoxy adherents and analysed the effect of adding GNPs in the adhesive and adherent on the tensile strength of the joints. It was found that the incorporation of GNPs into the adhesives and the substrates significantly increased the tensile strength of the joints. In previous work by the current authors [25], the effect of multi-walled carbon nanotubes (MWCNTs) on the mechanical properties, electric conductivity, fracture toughness and lap-shear strength of a rubber-modified epoxy adhesive was studied. It was found that blending small amount of MWCNTs moderately increased the Young’s modulus and fracture toughness, and significantly increased the electric conductivity and lap-shear strength of the rubber-modified adhesives. In the current work, the effects of graphene nanoplatelets (GNP) on the mechanical properties, fracture toughness and lap-shear strength of a rubber-modified epoxy adhesive are studied. 2. Experimental 2.1. Materials The rubber-modified adhesive is an experimental grade structural epoxy adhesive supplied by Henkel. In the rest of this paper, this adhesive is referred to as the ‘Control’. It contains 10 vol.% core-shell rubber (CSR) nanoparticles as a toughening agent as standard. The basic resin of this adhesive is a standard diglycidylether of bis-phenol A (DGEBA) epoxy (Epon828 from HEXION) with an epoxy equivalent molecular weight between 185-192 g/eq. The curing agent is dicyandiamide using fenuron as accelerator. Further additives are included in the resin to improve its performance. The exact nature and quantities of these additives remains the
proprietary information of Henkel. The CSR (Kane Ace MX 153 from Kaneka) is butadiene-acrylic copolymer with a mean diameter of approx. 74 nm [13]. The graphene nanoplatelets (GNP) were obtained in powder form from Graphene Supermarket, USA. These GNPs have an average flake thickness of 5-30 nm and average particle lateral size of 5-25 μm. 2.2. Nanocomposite preparation The GNPs and the rubber-modified adhesive were firstly pre-mixed using a mechanical mixer (IKA RW20 digital mixer) at 2000 rpm for two hours. The temperature of the mixture was elevated to 50 C using water bath in order to reduce the viscosity sufficiently to allow for easier processing. After that, the mixture was shear mixed using a high shear mixer (Silverson L4RT) at 3000 rpm initially at 50 C with the temperature slowly reduced to 0 C over two hours. The mixture was subsequently processed using the high shear mixer for a further three hours at approximately 0
C. The increased viscosity of the mixture at low
temperature allows the shear mixer to generate sufficiently high shear forces to effectively break up GNP clusters. A stoichiometric quantity of hardener was then added and mixed evenly throughout the liquid resin. Following the mixing processes, the mixture was degassed under a vacuum of -1 bar for an hour at 60 C in a vacuum oven. The samples for characterising the mechanical behaviour of bulk adhesives were then cured in aluminium moulds in a fan-assisted oven at 180 C for 90 minutes. It should be noted that the curing took place immediately after the mixing and casting process in order to limit the formation of excessive GNP agglomerates in the liquid resin. Adhesives with the following GNP content were manufactured: 0 wt.%, 0.1 wt.%, 0.2 wt.%, 0.3 wt.% and 0.5 wt.%, and referred to as Control, G0.1, G0.2, G0.3 and G0.5 throughout the remainder of this paper. 2.3. Experimental methods
Tensile mechanical properties of the composites, i.e. Young’s modulus and tensile strength, are measured using the uniaxial tensile test according to BS ISO 527 Standard [26]. The dumbbell specimens with 25 mm gauge length and 5 mm × 4 mm cross-section area were machined from a 4 mm thick cured plate using a sharp cutter in a CNC machine. A 1200P grade sand paper was then used to remove the visible defects, such as micro cracking and voids, along the sample surfaces. The tests were conducted at a loading rate of 0.5 mm/min at room temperature (nominally 20±1 C) on a Zwick/Roell material testing machine. The change of gauge length was measured using a non-contact video extensometer. At least five replicate tests were conducted for each material. Dynamic mechanical thermal analysis (DMTA, DMTA 242E from Netzsch, Germany) was performed in tension mode at 1 HZ. The dimension of the specimens was 10 mm × 2.5 mm × 0.6 mm. The specimens were heated up from 22 C to 170 C at a constant heating rate of 3 C /min. The maximum stationary point of the tan δ versus temperature curve was taken as the glass transition temperature. The Mode I fracture toughness, KIC and fracture energy, GIC, of the bulk adhesives were measured using the single edge notch three-point bending (SENB) test according to ASTM D5045-99 standard [27]. A sharp pre-crack was introduced by tapping a chilled razor blade into the sample. The tests were conducted at room temperature with a constant displacement rate of 1 mm/min. Nine specimens were tested for each material. The fracture toughness, KIC, values were calculated using ( ) where P is the load at failure, B is the sample thickness, W is the sample width, a is the length of the pre-crack, and f(a/W) is a non-dimensional shape factor. The fracture energy, GIC, was calculated be the equation below [27].
where E is the tensile Young’s modulus and ν is the Poisson’s ratio of the epoxy taken as 0.36 [13]. The single lap shear (SLS) test was conducted to study the shear strength and inservice performance of the adhesively bonded joints. The A109 steel substrates were supplied by Q-Lab Corporation (Q-Panel RS-14 [28]) with a dimension of 102 mm×25 mm × 1.6 mm. The length of the overlap of the joints was 12.5 mm. The bonding surfaces of the substrates were treated carefully prior to bonding to obtain good adhesion. The surface treatment consisted of acetone degreasing with subsequent grit blasting. The grit blasting used compressed air at approximately 100 psi to fire 180 mesh Aluminium Oxide grit at the surface for approx. 10 seconds. Following the grit blasting, the substrates were rinsed clear with running hot water, dried, and wiped with acetone. The substrates were then placed into an aircirculating oven at 50 C for 30 minutes to remove the moisture. The substrates were bonded immediately after being taken out of the oven to eliminate any further contamination. A constant adhesive layer thickness of 0.4 mm was maintained by two shims at either end of the lap shear joint. The manufacturing procedure to produce good-quality SLS adhesive joints follows best practice in industry. Firstly, slightly more adhesive than required amount was placed between two substrates, then a pressure was applied on the joint during the curing process to remove the extra adhesive and to obtain the required (0.4 mm) adhesive thickness. The tests were conducted at room temperature with a constant displacement rate of 0.5 mm/min. At least five replicate tests were conducted for each adhesive. A scanning electron microscope equipped with a field emission gun (SEM, FEI Quanta 3D) was used to inspect the fracture surfaces of SENB and SLS specimens. The acceleration voltage was 5 kV. The samples were gold sputter coated at a current of 30 mA for 15 seconds resulting in a coating thickness of 5 nm The
fracture subsurface of SENB samples were explored using transmission optical microscope (TOM, Nikon E80i (Orina)), and the samples were taken from the centre-plane of the specimens that is perpendicular to the fracture surfaces. The samples were ground and fine polished to thin sections with a thickness of approximately 60 μm according to the technique described in [29]. Electrical impedance spectroscopy was performed using a Reference 600 plus Potentiostat from Gamry Instruments, UK. Bulk samples were prepared by coating a sample of dimension 15 mm × 20 mm × 2.5 mm with a conductive silver paint to reduce the contact resistance. The electrical impedance of the adhesive in the lap shear joint was calculated by considering only the dimensions of the adhesive in the joint, i.e. a volume of 12.5 mm × 25 mm × 0.4 mm. The electrical conductivity of the steel substrates will be several orders of magnitude higher than that of the adhesive and the intimate molecular contact necessary to achieve a good adhesive bond would render any contact resistance between the steel substrates and the adhesives negligible. The electrical characteristics were analysed as a function of frequency using an RMS AC voltage of between 100 and 500 mV over a frequency range of 100 mHz-1 MHz. 3. Results and Discussion 3.1. GNP dispersion Figure 1 presents TOM mages of epoxy polymers modified with GNPs after curing. It can be clearly observed that the high shear mixing process resulted in the GNPs being well dispersed throughout the epoxy adhesive, as previously reported in the literature [8, 30]. At each GNP fraction loading some small GNP clusters can be observed. These are more prevalent at higher GNP contents. 3.2. Thermal and mechanical properties The measured glass transition temperatures, Tg, for all the adhesives are
summarised in Table 1. Tg of the control adhesive was measured to be 127.2 C. The addition of GNPs had negligible effect on the Tg. This agrees with the observation of Lim et al. [21] and Chong and Taylor [31], but contradicts the finding of Galpaya et al. [32] who report a small reduction in Tg with the addition of GNPs to epoxies. It should be noted that the materials used in [31, 32] were epoxy-graphene systems without any rubber phase. Wajid et al. [33] and Naebe et al. [34] report that the addition of small amount of graphene increases Tg considerably. This is correlated with a strong adhesion between the graphene sheets and the matrix. It is noteworthy that there is a distinction in the literature between the effect of either graphene or GNPs on the measured Tg of a thermoset matrix. Indeed, Allaoui and El Bounnia [35] indicate that agglomeration of carbon nanotubes is associated with a reduction in Tg rather than an increase as is the case with well-dispersed nanotubes. The mechanism suggested the agglomerations of nanotubes physically inhibited the formation of a fully cured thermoset network and some uncured or partially uncured resin remained trapped by the closely packed agglomerations. A similar argument can be applied to graphene and agglomerations of graphene. In the strictest sense, graphene nanoplatelets can be considered as small agglomerations of graphene. Thus, it can be expected that for the large graphene nanoplatelets, a reduction in Tg could be observed while for nanoplatelets which consist of only a few layers of graphene, an increase would be observed. The Young’s modulus, E, and tensile strength, uts, of all the adhesive formulations are summarised in Table 1. It is found that the incorporation of 0.1 wt.% GNPs increased the Young’s modulus moderately from 2.46 GPa of control to 2.56 GPa of G0.1. This indicates that there exists effective load transfer between GNPs and epoxy matrix at the elastic deformation stage under loading. No further improvement in the Young’s modulus was observed as the GNP content increased up to 0.5 wt.%. This is due to the increasing amount of GNP agglomerates at higher GNP contents. From Table 1, one can see that the tensile strength has no significant change due to
the addition of GNPs. This demonstrates that the interfacial adhesion between GNPs and epoxy matrix is not strong enough to maintain load transfer at the failure stage in the tensile test. 3.3. Mode I fracture behaviour 3.3.1. Fracture toughness Table 1 presents the mode-I fracture toughness, KIC, and fracture energy, GIC from the SENB tests. KIC of control was measured to be 2.17 MPam 1/2. This increased to 2.40 MPam1/2 with the addition of 0.1 wt.% GNPs. No further increase in KIC was observed at a higher loading of GNP, up to a particle loading of 0.5 wt.%. Indeed, a slight reduction is observed for G0.3 and G0.5. This is likely attributed to the increased prevalence of GNP agglomerates at higher GNP loading, see Figure 1. The GNP agglomerates appeared as defects in the fracture process and prevented further enhancement in the fracture energy for high GNP content. The fracture energy exhibited a similar trend as the fracture toughness. GIC increased significantly from 2136 J/m2 for control to 2590 J/m2 for G0.1, and then plateaued for higher GNP loading. 3.3.2. Toughening mechanisms Typical TOM micrographs taken under bright-field light (top) and cross-polarised light (bottom) at the fracture subsurface of SENB specimens are presented in Figures 2. The bottom half of each figure represents the subsurface damage. In each image crack propagation is from left to right. A plastic zone with dark colour is observed at the fracture subsurface of all the formulations on the bright-field micrographs. A bright region was observed on the cross-polarised micrographs. This typically indicates the presence of shear band yielding in the bright region on the cross-polarised micrographs and bulk plasticity in the ‘far-field’ plastic damage region on the bright-field micrographs for rubber modified epoxies [13, 25, 36]. The
depth of the plastic zone in the direction perpendicular to the fracture surface is summarised in Table 2. It is clear that the depth of the plastic zone decreased from 117 μm for the control to 73 μm for G0.1 due to the addition of 0.1 % GNPs. This indicates a decrease in the fracture toughness by the dissipation of plastic energy (induced by the CSR nanoparticles [13, 25, 36]) for G0.1 formulation. From Table 1, it is found that the total fracture energy of G0.1 is much higher than that for the control. Hence, additional fracture mechanisms, i.e. crack pinning and crack deflection [21], were introduced by the GNPs to dissipate more fracture energy in the fracture process. These fracture mechanisms do not result in an increase in the plastic zone size. Moreover, the presence of the GNPs impedes the classical CSR toughening mechanisms by restricting the availability of matrix to plastically deform, thus restricting the depth of the plastic zone in the hybrid adhesives. Figure 3 presents typical SEM images of the fracture surfaces of the SENB samples. The images have been taken from just ahead of the pre-crack tip in each sample. A relatively smooth fracture surface with many river-lines was observed for the control adhesive, see Figure 3(a). A much rougher appearance of the fracture surface with was observed due to the addition of GNPs, see Figures 3 (b) and (c). Figures 3 (d)-(f) present images of the same fracture surfaces at higher magnification. The higher magnification reveals an extensive network of voids on the fracture surfaces of the control adhesive, see Figure 3 (d). These voids were generated by the toughening mechanisms of the CSR nanoparticles, i.e. CSR cavitation and plastic void growth in the epoxy matrix [1, 13, 15]. Figures 3 (e) and (f) present typical fracture surfaces of the adhesives containing both CSR nanoparticles and GNPs. Voids generated by cavitation and subsequent plastic void growth can also be observed, similarly to the control adhesive. Furthermore, many pulled out and debonded GNP flakes were also observed on the fracture surfaces. There is significant evidence of gross plastic deformation of the epoxy-CSR matrix in the regions immediately surrounding the GNP stacks. Therefore, it is likely that the
presence of the GNPs contributes an additional dissipative toughening mechanism. 3.4. Single lap shear (SLS) test Figure 4 presents the lap shear strength of the different adhesives tested. The lap shear strength was measured as 21.7 MPa for the control adhesive. The incorporation of GNPs resulted in a steady decrease of lap shear strength, i.e. from 21.7 MPa of the control to 17.2 MPa of G0.5. Figure 5 presents the optical micrographs of the homologous fracture surfaces of the SLS adhesive joints. The blue rectangle demarcates the adhesive bonding area. It should be noted that an extremely thin residual adhesive layer was observed on the left half of the joint for all the adhesives. This is difficult to discern for the control samples as there is a poor contrast in colour between the adhesive and the substrate. Hence, the failure of all the adhesives was cohesive, i.e. the crack was observed to grow inside the adhesive layer, although very close to a bonded interface. Typical SEM images of the SLS fracture surfaces for the control, G0.2 and G0.5 adhesives are presented in Figure 6. A rough fracture surface with many small platelets of polymer was observed for the control adhesive, see Figure 6 (a). The SEM image of the control adhesive with high magnification (Figure 6(b)) shows that the fracture surface is covered with a “sea” of voids, which were generated by rubber cavitation and following plastic void growth [5, 13, 15]. This is similar in appearance to the corresponding fracture surface of the bulk adhesive, see Figure 3 (d). The addition of the GNPs into the control adhesive changed the appearance of the fracture surface significantly, i.e. the fracture surfaces of G0.2 and G0.5 adhesives are characterised by the presence of alternating smooth regions (indicated by red arrows in Figures 6 (c) and (d)), approximately 20-50 μm in size, and rough regions (indicated by yellow arrows in Figures 6 (c) and (d)). A high
magnification image of the roughened regions of the fracture surface, Figure 6 (e), reveal that these rough regions are characterised by the presence of both GNP flakes and voids initiated by cavitation of the rubber particles. It is interesting that very limited number of GNP flakes, such as the indicated one on Figure 6 (e), are observed on the fracture surfaces of adhesives modified with GNPs. The smooth regions identifiable in Figures 6 (c) and (d) are the surfaces of GNP flakes aligned in the direction parallel to the fracture surface. This is demonstrated schematically in Figure 7 (a). The cross-section of the adhesive layer of a fractured G0.5 SLS specimen was imaged using SEM to closer inspect the orientation of the GNPs. Figure 7(b) demonstrates the process used to prepare the SEM sample. A fine-tooth saw was firstly used to cut almost completely through the substrates. The samples were then fractured dynamically into two pieces to reveal a surface of the cross-section. The SEM images of the cross-section are shown in Figure 8. Note that the gap between the adhesive layer and the substrate arose from unavoidable debonding introduced in the sample preparation process. The presence of GNP clusters (indicated by red solid arrows) are observed on the adhesive layer side, and the size of these clusters are approximately the same as the size of the smooth region observed on the fracture surface of the SLS specimens, see Figure 6 (d). This further demonstrates that the smooth region on the fracture surface of SLS specimens are introduced by the GNP clusters, which are aligned in the direction parallel to the fracture surface. The preferential alignment of the GNP clusters in the direction parallel to the fracture surface can be explained by considering the individual steps in the manufacturing procedure of the SLS adhesive joints. In the first step of the manufacturing procedure, an excess of adhesive was placed between two substrates to be joined. A pressure was then applied to the joint during the curing process to obtain the required (0.4 mm) adhesive thickness and remove the excess adhesive. The pressure applied to the joints spread the adhesive evenly in the adhesive joints and
pushed the adhesive flow to the edge of the bonding area. This caused the twodimensional GNP clusters oriented in the adhesive flowing direction, i.e. the direction parallel to the bonding surface. Figure 9 presents a comparison of the measured electric conductivity between the bulk samples and the SLS adhesive joint for both the control, i.e. no GNPs, and G0.5 adhesive formulations. Firstly, it can be observed that in the absence of any conducting carbon nano-filler, the measured electrical conductivity follows a power law behaviour, with an increase in conductivity measured with increasing electrical frequency. Moreover, with the addition of GNPs, it can be seen that in the case of the bulk material, a frequency independent value of electrical conductivity for the composite emerges at lower frequency. This is indicative of the formation of a percolation network formation and can be attributed to the reasonably good dispersion and random orientation of the GNPs and GNP clusters in the nanocomposite. This is in stark contrast to the behaviour of the SLS joint with 0.5 % GNPs. It is found that the electric conductivity of the SLS joint is much lower than the electric conductivity of the bulk specimens due to the identical alignment of the GNP clusters in the SLS adhesive joints. This is further evidence that the orientation of the GNPs in the SLS specimen is no longer random as it is not forming a percolation network. 4. Conclusions This work studied the effect of GNPs on the mechanical properties, fracture toughness and lap shear strength of a rubber-modified epoxy adhesive. The addition of GNPs had no effect on the glass transition temperature. The tensile modulus increased due to the addition of 0.1 % GNPs with no subsequent increase observed for higher GNP loading. The tensile strength was unaffected by the addition of small amount GNPs, up to 0.5 wt.%. Blending small amount of GNPs, i.e. 0.1 %, into the control adhesive increased the fracture energy moderately from 2136 J/m 2 to 2590
J/m2. The addition of further quantities of GNPs did not result in a further increase in the fracture energy. This is attributed to increased agglomeration of the GNPs in the curing process. The addition of GNPs decreased the lap shear strength of the adhesive significantly, i.e. from 21.7 MPa of the control adhesive by approx. 20 % gradually to 17.2 MPa of the adhesive modified by 0.5 % GNPs. This was caused by the preferential alignment of the GNP clusters in the adhesive joints. The alignment of the GNP clusters, introduced by the manufacturing process of the joints, also caused a significant drop in the electric conductivity of the adhesive joints. This, combined with the poor mechanical performance, means that GNP modified adhesives cannot be considered as suitable candidate materials for structural adhesive joints requiring in-situ structural monitoring. Acknowledgements Financial support from the Irish Centre for Composite Research, China Scholarship Council and Henkel (Ireland) is gratefully acknowledged. References [1]
A. J. Kinloch, S. J. Shaw, D. A. Tod, D. L. Hunston, Deformation and fracture behaviour of a rubber-toughened epoxy: 1. Microstructure and fracture studies, Polymer 24 (10) (1983) 1341-1354.
[2]
A. J. Kinloch, Toughening Epoxy Adhesives to Meet Today’s Challenges, MRS Bulletin 28 (2003) 445-448.
[3]
A. I. Dmitriev, I. Husler, W. sterle, B. Wetzel, G. Zhang, Modeling of the stress strain behavior of an epoxy-based nanocomposite filled with silica nanoparticles, Materials & Design 89 (2016) 950-956.
[4]
M. Imanaka, Y. Takeuchi, Y. Nakamura, A. Nishimura, T. Iida, Fracture toughness of spherical silica-filled epoxy adhesives, International Journal of Adhesion and Adhesives 21 (5) (2001) 389-396.
[5]
D. Carolan, A. Ivankovic, A. Kinloch, S. Sprenger, A. Taylor, Toughening of epoxy-based hybrid nanocomposites, Polymer 97 (2016) 179-190.
[6]
G. Marami, S. A. Nazari, S. A. Faghidian, F. Vakili-Tahami, S. Etemadi, Improving the mechanical behavior of the adhesively bonded joints using RGO additive, International Journal of Adhesion and Adhesives 70 (2016) 277-286.
[7]
M. A. Rafiee, J. Rafiee, I. Srivastava, Z. Wang, H. Song, Z.-Z. Yu, N. Koratkar, Fracture and fatigue in graphene nanocomposites, Small 6 (2) (2010) 179183.
[8]
L.-C. Tang, Y.-J. Wan, D. Yan, Y.-B. Pei, L. Zhao, Y.-B. Li, L.-B. Wu, J.-X. Jiang, G.-Q. Lai, The effect of graphene dispersion on the mechanical properties of graphene/epoxy composites, Carbon 60 (2013) 16-27.
[9]
F. Gojny, M. Wichmann, U. Kpke, B. Fiedler, K. Schulte, Carbon nanotubereinforced epoxy-composites: enhanced stiffness and fracture toughness at low nanotube content, Composites Science and Technology 64 (15) (2004) 2363–2371.
[10] M. R. Gude, S. G. Prolongo, A. Urea, Toughening effect of carbon nanotubes and carbon nanofibres in epoxy adhesives for joining carbon fibre laminates, International Journal of Adhesion and Adhesives 62 (2015) 139-145. [11]
M. Imanaka, S. Motohashi, K. Nishi, Y. Nakamura, M. Kimoto, Crack growth behavior of epoxy adhesives modified with liquid rubber and cross-linked rubber particles under mode I loading, International Journal of Adhesion and Adhesives 29 (1) (2009) 45-55.
[12] A. Rezaifard, K. Hodd, D. Tod, J. Barton, Toughening epoxy resins with poly (methyl methacrylate)-grafter-natural rubber, and its use in adhesive formulations, International Journal of Adhesion and Adhesives 14 (2) (1994) 153-159. [13]
D. Quan, A. Ivankovic, Effect of core-shell rubber (CSR) nano-particles on mechanical properties and fracture toughness of an epoxy polymer, Polymer 66 (2015) 16-28.
[14]
G. Giannakopoulos, K. Masania, A. C. Taylor, Toughening of epoxy using core– shell particles, Journal of Materials Science 46 (2) (2011) 327-338.
[15]
J. Chen, A. J. Kinloch, S. Sprenger, A. C. Taylor, The mechanical properties and toughening mechanisms of an epoxy polymer modified with polysiloxanebased core-shell particles, Polymer 54 (16) (2013) 4276-4289.
[16]
B. T. Marouf, R. A. Pearson, R. Bagheri, Anomalous fracture behavior in an epoxy-based hybrid composite, Materials Science and Engineering: A 515 (1– 2) (2009) 49-58.
[17]
S. A. Ngah, A. C. Taylor, Toughening performance of glass fibre composites with core-shell rubber and silica nanoparticle modified matrices, Composites Part A: Applied Science and Manufacturing 80 (2016) 292-303.
[18]
Y. L. Liang, R. A. Pearson, The toughening mechanism in hybrid epoxy-silicarubber nanocomposites (HESRNs), Polymer 51 (21) (2010) 4880-4890.
[19]
B. Ahmadi-Moghadam, M. Sharafimasooleh, S. Shadlou, F. Taheri, Effect of functionalization of graphene nanoplatelets on the mechanical response of graphene/epoxy composites, Materials & Design (1980-2015) 66, Part A (2015) 142-149.
[20] A. S. Wajid, H. S. T. Ahmed, S. Das, F. Irin, A. F. Jankowski, M. J. Green, Highperformance pristine graphene/epoxy composites with enhanced mechanical and electrical properties, Macromolecular Materials and Engineering 298 (3) (2013) 339-347. [21]
Y. J. Lim, D. Carolan, A. C. Taylor, Simultaneously tough and conductive rubber–graphene–epoxy nanocomposites, Journal of Materials Science 51 (18) (2016) 8631-8644.
[22] S. Chandrasekaran, C. Seidel, K. Schulte, Preparation and characterization of graphite nano-platelet (GNP)/epoxy nano-composite: Mechanical, electrical and thermal properties, European Polymer Journal 49 (12) (2013) 38783888. [23]
R. Moriche, S. Prolongo, M. Snchez, A. Jimnez-Surez, F. Chamizo, A. Urea, Thermal conductivity and lap shear strength of GNP/epoxy nanocomposites adhesives, International Journal of Adhesion and Adhesives 68 (2016) 407410.
[24]
L. Guadagno, M. Sarno, U. Vietri, M. Raimondo, C. Cirillo, P. Ciambelli, Graphene-based structural adhesive to enhance adhesion performance, RSC Advances 5 (35) (2015) 27874-27886.
[25] D. Quan, D. Carolan, C. Rouge, N. Murphy, A. Ivankovic, Carbon nanotubes and core-shell rubber nanoparticles modified structural epoxy adhesives, Journal of Materials Science (2016) 1-16.
[26]
BS ISO 527-2:1996, Plastics determination of tensile properties-part 2: test conditions for moulding and extrusion plastics, British Standard Institute (1996).
[27]
ASTM D5045-99, Standard Test Method for Plane-Strain Fracture Toughness and Strain Energy Release Rate of Plastic Materials (1999).
[28]
Q-Lab Corporation, http://www.q-lab.com/q-panel-selector/, last Accessed 03th January 2017.
[29]
A. S. Holik, R. P. Kambour, D. G. Fink, S. Y. Hobbs, Grinding and polishing techniques for thin sectioning of polymeric materials for transmission light microscopy, Microstructure Science 7 (1979) 357-367.
[30] W. Zhao, M. Fang, F. Wu, H. Wu, L. Wang, G. Chen, Preparation of graphene by exfoliation of graphite using wet ball milling, J. Mater. Chem. 20 (2010) 58175819. [31]
H. M. Chong, S. J. Hinder, A. C. Taylor, Graphene nanoplatelet-modified epoxy: effect of aspect ratio and surface functionality on mechanical properties and toughening mechanisms, Journal of Materials Science 51 (19) (2016) 8764– 8790.
[32]
D. Galpaya, M. Wang, C. Yan, M. Liu, N. Motta, E. R. Waclawik, Fabrication and mechanical and thermal behaviour of graphene oxide/epoxy nanocomposites, Journal of Multifunctional Composites 1 (2) (2013) 91-98.
[33] A. S. Wajid, H. S. T. Ahmed, S. Das, F. Irin, A. F. Jankowski, M. J. Green, Highperformance pristine graphene/epoxy composites with enhanced mechanical and electrical properties, Macromolecular Materials and Engineering 298 (3) (2013) 339-347. [34]
M. Naebe, J. Wang, A. Amini, H. Khayyam, N. Hameed, L. H. Li, Y. Chen, B. Fox, Mechanical property and structure of covalent functionalised graphene/epoxy nanocomposites, Scientific Reports 4 (2014) 4375.
[35]
A. Allaoui, N.-E. El Bounia, How carbon nanotubes affect the cure kinetics and glass transition temperature of their epoxy composites? – A review, Express Polymer Letters 3 (9) (2009) 588-594.
[36]
Y. Liang, R. Pearson, Toughening mechanisms in epoxy-silica nanocomposites (ESNs), Polymer 50 (20) (2009) 4895-4905.
Table 1: Thermal and mechanical properties and fracture toughness of different adhesive formulations. Formulation
E (GPa)
uts
wt.% Graphene
Tg (°C)
Control
0
127.2
2.46±0.08 68.1±3.3
2.17±0.14
2136±121
G0.1
0.1
126.9
2.56±0.02 69.5±0.8
2.40±0.17
2590±178
G0.2
0.2
125.9
2.59±0.06 66.8±1.1
2.40±0.09
2604±191
G0.3
0.3
126.7
2.61±0.05 63.4±2.9
2.35±0.11
2580±215
G0.5
0.5
127.5
2.60±0.06 65.8±0.7
2.32±0.12
2499±130
(MPa)
KIC GIC (J/m2) (MPam1/2)
Table 2: Measurement of the fracture subsurface damage. Formulation Control Plastic zone depth (µm) 117
G0.1 73
G0.2 79
G0.3 74
G0.5 71
200 µm (a) G0.1
200 µm (b) G0.2
200 µm (c) G0.3
200 µm (d) G0.5 Figure 1: Dispersion of the GNPs in different adhesive formulations.
G0.3
Control
0.5mm
0.5mm
Figure 2: Typical TOM micrographs of the fracture subsurface of SENB specimens under bright field light (top) and cross polarised light (bottom). Crack growth direction is from left right. The black boxes indicate the plastic deformation zone.
(a) Control-Low magnification
(b) G0.2-Low magnification
(c) G0.5-Low magnification
(d) Control- High magnification
Debonded GNPs Pull-out GNPs
(e) G0.2- High magnification
Pull-out GNPs
Debonded GNPs
(f) G0.5- High magnification Figure 3: SEM images of the fracture surface of SENB samples.
Lapshear strength (MPa)
25
22
19
16
13 0
0.1
0.2
0.3
0.4
0.5
GNP content (wt.%) Figure 4: Lap shear strength of different adhesive formulations.
0.6
Figure 5: Optical micrographs of the fracture surface of the SLS adhesive joints. The blue box indicates the bonding surface.
(a) Control-Low magnification
(b) Control-High magnification
(c) G0.2-Low magnification
(d) G0.5-Low magnification
Rough Region
GNP Flake
Smooth Region
(e) G0.5-High magnification Figure 6: Typical SEM images of the fracture surface of SLS specimens. The red arrows indicate the smooth regions and the yellow arrows indicate the rough regions in (c) and (d).
Adhesive layer GNP clusters
Substrates (a)
Adhesive layer
Substrates (b) Figure 7: Schematics to show (a) the alignment of the GNP clusters in the SLS specimens and (b) the sample preparation for SEM analysis to confirm the GNP alignment. The dashed line in (b) indicates the broken plane, and SEM images were taken from the top.
Adhesive layer
Substrate
Figure 8: SEM images of the cross-section of the adhesive layer of a fractured G0.5 SLS specimen. The arrows indicate GNP agglomerates.
1.E+00 1.E-01 Bulk-Control
Electric Conductivity (S/m)
1.E-02 1.E-03 1.E-04
Bulk-G0.5 SLS Joint-Control SLS Joint-G0.5
1.E-05 1.E-06 1.E-07 1.E-08 1.E-09 1.E-10 1.E-11 1.E-12 1.E-01
1.E+00
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
1.E+06
Frequency (Hz) Figure 9: The electric conductivity of the bulk specimens and the SLS adhesive joints for the Control and G0.5 adhesives.
PII: DOI: Reference:
S0143-7496(17)30162-8 https://doi.org/10.1016/j.ijadhadh.2017.09.003 JAAD2054
To appear in: International Journal of Adhesion and Adhesives Received date: 13 March 2017 Accepted date: 2 September 2017 Cite this article as: Dong Quan, Declan Carolan, Clemence Rouge, Neal Murphy and Alojz Ivankovic, Mechanical and fracture properties of epoxy adhesives modified with graphene nanoplatelets and rubber particles, International Journal of Adhesion and Adhesives, https://doi.org/10.1016/j.ijadhadh.2017.09.003 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 galley proof before it is published in its final citable 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.
Mechanical and fracture properties of epoxy adhesives modified with graphene nanoplatelets and rubber particles Dong Quan a, Declan Carolan b,c, Clemence Rouge a, Neal Murphy a, Alojz Ivankovic a a b
School of Mechanical and Materials Engineering, University College Dublin, Dublin, Ireland Department of Mechanical Engineering, Imperial College London, SW7 2AZ, UK
cFAC
Technology, London, UK
_________________________________________________________________________________________________ Abstract Graphene nanoplatelets (GNP) were introduced into a rubber-modified epoxy adhesive in order to simultaneously improve the bulk mechanical properties, fracture toughness and single joint lap shear strength of the adhesive. The Young’s modulus was observed to increase marginally from 2.46 GPa to 2.56 GPa due to the addition of 0.1 wt.% GNPs. No further increase in modulus was observed for GNP loading above 0.1 wt.%. A negligible effect on the measured tensile strength was observed. The fracture energy of the bulk adhesive increased by 21 % due to the addition of 0.1 wt.% GNPs. No further increase in measured fracture energy was observed as the GNP content was further increased to 0.5 wt.%. A systematic decrease in the lap shear strength was observed due to the addition of GNPs, i.e. the lap shear strength decreased from 21.7 MPa of the control adhesive gradually to 17.2 MPa of the adhesive modified by 0.5 wt.% GNPs. Imaging analysis of the failed adhesive joints reveal that the reduction in lap shear strength was attributed to the preferential alignment of the GNPs in the direction parallel to the adhesive bonding surface. This was further confirmed by comparing the electrical behaviour of the lap shear joints with that of the bulk adhesive samples.
Keywords: Rubber-modified epoxy adhesive, Graphene nanoplatelets, fracture toughness, lap shear strength. _________________________________________________________________________________________________ 1. Introduction Epoxy polymers possess many favourable engineering properties due to their highly cross-linked structure. However, this type of structure also has short-comings, such as inherently low fracture toughness and poor resistance to fracture [1, 2]. The addition of modifiers forming a second distributed phase was found to be a promising method to improve the fracture toughness of epoxy polymers. Different types of modifiers, such as silica spheres [3–5], graphene nanoplatelets [6–8], carbon nanotubes [9, 10] and rubber particles [11–13] have been proven to improve the fracture toughness of epoxies considerably. Among different modifiers of epoxies, rubber particles demonstrate superior performance in toughness enhancement. It has been widely reported [13–17] that the addition of rubber particles can increase fracture toughness significantly. However, the addition of rubber particles results in a considerable reduction in mechanical properties. This limited the application of rubber-modified epoxies, especially for the systems with high rubber content. A possible solution to this is to create a so called hybrid epoxy, containing a blend of both soft rubber particles and rigid particles dispersed on the nano-scale. The rubber particles provide an effective toughening mechanism, while the rigid particles contribute to offset the loss in mechanical properties. In some cases, e.g. silica, the rigid particles also contribute to toughening and can provide a synergistic effect [5, 18]. The effects of graphene on the mechanical properties, electrical conductivity and fracture toughness of epoxy have been widely investigated. Tang et al. [8] studied the mechanical properties and fracture toughness of graphene modified epoxy
composites. They report that the addition of 0.2 wt.% graphene oxide into the neat epoxy increased the Young’s modulus from 2.93 GPa to approximately 3.1 GPa with the fracture toughness improved from 0.49 MPam1/2 to 0.75MPam1/2. The debonding and crack bridging of the graphene oxide platelets are identified as the main toughening mechanisms. Rafiee et al. [7] investigated the fracture and fatigue behaviour of graphene modified epoxy nanocomposites. It was observed that adding 0.125 wt.% functionalized graphene sheets increased the fracture toughness of the neat epoxy by 65 % and the fracture energy by 115 %. Moreover, incorporation of 0.125 wt.% functionalized graphene sheets was found to dramatically reduce the rate of crack propagation under fatigue conditions. The superior performance of graphene to resist fracture and fatigue was attributed to the two-dimensional structure of graphene, which enabled it to deflect cracks effectively. AhmadiMoghadam et al. [19] demonstrated that blending small amount of graphene nanoplatelets increased the fracture toughness of an epoxy by 82 %, while a concurrent improvement in Young’s modulus and tensile strength was also observed. Crack bridging was proposed as the main toughening mechanism. Wajid et al. [20] presented that, even at small graphene filler content, the epoxy composites exhibit a 10-40 % increase in the mechanical properties, and the electrical conductivity was enhanced by seven orders of magnitude. Lim et al. [21] and Chandrasekaran et al. [22] studied the thermo-mechanical, electrical and fracture properties of rubber-graphene-epoxy nanocomposites and presented that both the electrical conductivity and fracture toughness increased significantly. Based on the literature review, it is clear that blending small amount of graphene into epoxies can increase the mechanical properties and fracture toughness of the polymer considerably. However, limited work has been performed to study the effect of graphene on the fracture behaviour of epoxy adhesive joints. Moriche et al. [23] studied the influence of the incorporation of GNPs on the lap shear strength of an epoxy adhesive. They reported that the lap shear strength remained constant for
neat adhesives and adhesives modified by GNPs. This was attributed to the weak adhesion between the substrates and the adhesives, which resulted in interfacial failure of the adhesive joints. In this case, the lap shear test is measuring the adhesion between the adhesives and the substrates instead of the lap shear strength of the adhesives. Guadagno et al. [24] used epoxy adhesive to bond epoxy adherents and analysed the effect of adding GNPs in the adhesive and adherent on the tensile strength of the joints. It was found that the incorporation of GNPs into the adhesives and the substrates significantly increased the tensile strength of the joints. In previous work by the current authors [25], the effect of multi-walled carbon nanotubes (MWCNTs) on the mechanical properties, electric conductivity, fracture toughness and lap-shear strength of a rubber-modified epoxy adhesive was studied. It was found that blending small amount of MWCNTs moderately increased the Young’s modulus and fracture toughness, and significantly increased the electric conductivity and lap-shear strength of the rubber-modified adhesives. In the current work, the effects of graphene nanoplatelets (GNP) on the mechanical properties, fracture toughness and lap-shear strength of a rubber-modified epoxy adhesive are studied. 2. Experimental 2.1. Materials The rubber-modified adhesive is an experimental grade structural epoxy adhesive supplied by Henkel. In the rest of this paper, this adhesive is referred to as the ‘Control’. It contains 10 vol.% core-shell rubber (CSR) nanoparticles as a toughening agent as standard. The basic resin of this adhesive is a standard diglycidylether of bis-phenol A (DGEBA) epoxy (Epon828 from HEXION) with an epoxy equivalent molecular weight between 185-192 g/eq. The curing agent is dicyandiamide using fenuron as accelerator. Further additives are included in the resin to improve its performance. The exact nature and quantities of these additives remains the
proprietary information of Henkel. The CSR (Kane Ace MX 153 from Kaneka) is butadiene-acrylic copolymer with a mean diameter of approx. 74 nm [13]. The graphene nanoplatelets (GNP) were obtained in powder form from Graphene Supermarket, USA. These GNPs have an average flake thickness of 5-30 nm and average particle lateral size of 5-25 μm. 2.2. Nanocomposite preparation The GNPs and the rubber-modified adhesive were firstly pre-mixed using a mechanical mixer (IKA RW20 digital mixer) at 2000 rpm for two hours. The temperature of the mixture was elevated to 50 C using water bath in order to reduce the viscosity sufficiently to allow for easier processing. After that, the mixture was shear mixed using a high shear mixer (Silverson L4RT) at 3000 rpm initially at 50 C with the temperature slowly reduced to 0 C over two hours. The mixture was subsequently processed using the high shear mixer for a further three hours at approximately 0
C. The increased viscosity of the mixture at low
temperature allows the shear mixer to generate sufficiently high shear forces to effectively break up GNP clusters. A stoichiometric quantity of hardener was then added and mixed evenly throughout the liquid resin. Following the mixing processes, the mixture was degassed under a vacuum of -1 bar for an hour at 60 C in a vacuum oven. The samples for characterising the mechanical behaviour of bulk adhesives were then cured in aluminium moulds in a fan-assisted oven at 180 C for 90 minutes. It should be noted that the curing took place immediately after the mixing and casting process in order to limit the formation of excessive GNP agglomerates in the liquid resin. Adhesives with the following GNP content were manufactured: 0 wt.%, 0.1 wt.%, 0.2 wt.%, 0.3 wt.% and 0.5 wt.%, and referred to as Control, G0.1, G0.2, G0.3 and G0.5 throughout the remainder of this paper. 2.3. Experimental methods
Tensile mechanical properties of the composites, i.e. Young’s modulus and tensile strength, are measured using the uniaxial tensile test according to BS ISO 527 Standard [26]. The dumbbell specimens with 25 mm gauge length and 5 mm × 4 mm cross-section area were machined from a 4 mm thick cured plate using a sharp cutter in a CNC machine. A 1200P grade sand paper was then used to remove the visible defects, such as micro cracking and voids, along the sample surfaces. The tests were conducted at a loading rate of 0.5 mm/min at room temperature (nominally 20±1 C) on a Zwick/Roell material testing machine. The change of gauge length was measured using a non-contact video extensometer. At least five replicate tests were conducted for each material. Dynamic mechanical thermal analysis (DMTA, DMTA 242E from Netzsch, Germany) was performed in tension mode at 1 HZ. The dimension of the specimens was 10 mm × 2.5 mm × 0.6 mm. The specimens were heated up from 22 C to 170 C at a constant heating rate of 3 C /min. The maximum stationary point of the tan δ versus temperature curve was taken as the glass transition temperature. The Mode I fracture toughness, KIC and fracture energy, GIC, of the bulk adhesives were measured using the single edge notch three-point bending (SENB) test according to ASTM D5045-99 standard [27]. A sharp pre-crack was introduced by tapping a chilled razor blade into the sample. The tests were conducted at room temperature with a constant displacement rate of 1 mm/min. Nine specimens were tested for each material. The fracture toughness, KIC, values were calculated using ( ) where P is the load at failure, B is the sample thickness, W is the sample width, a is the length of the pre-crack, and f(a/W) is a non-dimensional shape factor. The fracture energy, GIC, was calculated be the equation below [27].
where E is the tensile Young’s modulus and ν is the Poisson’s ratio of the epoxy taken as 0.36 [13]. The single lap shear (SLS) test was conducted to study the shear strength and inservice performance of the adhesively bonded joints. The A109 steel substrates were supplied by Q-Lab Corporation (Q-Panel RS-14 [28]) with a dimension of 102 mm×25 mm × 1.6 mm. The length of the overlap of the joints was 12.5 mm. The bonding surfaces of the substrates were treated carefully prior to bonding to obtain good adhesion. The surface treatment consisted of acetone degreasing with subsequent grit blasting. The grit blasting used compressed air at approximately 100 psi to fire 180 mesh Aluminium Oxide grit at the surface for approx. 10 seconds. Following the grit blasting, the substrates were rinsed clear with running hot water, dried, and wiped with acetone. The substrates were then placed into an aircirculating oven at 50 C for 30 minutes to remove the moisture. The substrates were bonded immediately after being taken out of the oven to eliminate any further contamination. A constant adhesive layer thickness of 0.4 mm was maintained by two shims at either end of the lap shear joint. The manufacturing procedure to produce good-quality SLS adhesive joints follows best practice in industry. Firstly, slightly more adhesive than required amount was placed between two substrates, then a pressure was applied on the joint during the curing process to remove the extra adhesive and to obtain the required (0.4 mm) adhesive thickness. The tests were conducted at room temperature with a constant displacement rate of 0.5 mm/min. At least five replicate tests were conducted for each adhesive. A scanning electron microscope equipped with a field emission gun (SEM, FEI Quanta 3D) was used to inspect the fracture surfaces of SENB and SLS specimens. The acceleration voltage was 5 kV. The samples were gold sputter coated at a current of 30 mA for 15 seconds resulting in a coating thickness of 5 nm The
fracture subsurface of SENB samples were explored using transmission optical microscope (TOM, Nikon E80i (Orina)), and the samples were taken from the centre-plane of the specimens that is perpendicular to the fracture surfaces. The samples were ground and fine polished to thin sections with a thickness of approximately 60 μm according to the technique described in [29]. Electrical impedance spectroscopy was performed using a Reference 600 plus Potentiostat from Gamry Instruments, UK. Bulk samples were prepared by coating a sample of dimension 15 mm × 20 mm × 2.5 mm with a conductive silver paint to reduce the contact resistance. The electrical impedance of the adhesive in the lap shear joint was calculated by considering only the dimensions of the adhesive in the joint, i.e. a volume of 12.5 mm × 25 mm × 0.4 mm. The electrical conductivity of the steel substrates will be several orders of magnitude higher than that of the adhesive and the intimate molecular contact necessary to achieve a good adhesive bond would render any contact resistance between the steel substrates and the adhesives negligible. The electrical characteristics were analysed as a function of frequency using an RMS AC voltage of between 100 and 500 mV over a frequency range of 100 mHz-1 MHz. 3. Results and Discussion 3.1. GNP dispersion Figure 1 presents TOM mages of epoxy polymers modified with GNPs after curing. It can be clearly observed that the high shear mixing process resulted in the GNPs being well dispersed throughout the epoxy adhesive, as previously reported in the literature [8, 30]. At each GNP fraction loading some small GNP clusters can be observed. These are more prevalent at higher GNP contents. 3.2. Thermal and mechanical properties The measured glass transition temperatures, Tg, for all the adhesives are
summarised in Table 1. Tg of the control adhesive was measured to be 127.2 C. The addition of GNPs had negligible effect on the Tg. This agrees with the observation of Lim et al. [21] and Chong and Taylor [31], but contradicts the finding of Galpaya et al. [32] who report a small reduction in Tg with the addition of GNPs to epoxies. It should be noted that the materials used in [31, 32] were epoxy-graphene systems without any rubber phase. Wajid et al. [33] and Naebe et al. [34] report that the addition of small amount of graphene increases Tg considerably. This is correlated with a strong adhesion between the graphene sheets and the matrix. It is noteworthy that there is a distinction in the literature between the effect of either graphene or GNPs on the measured Tg of a thermoset matrix. Indeed, Allaoui and El Bounnia [35] indicate that agglomeration of carbon nanotubes is associated with a reduction in Tg rather than an increase as is the case with well-dispersed nanotubes. The mechanism suggested the agglomerations of nanotubes physically inhibited the formation of a fully cured thermoset network and some uncured or partially uncured resin remained trapped by the closely packed agglomerations. A similar argument can be applied to graphene and agglomerations of graphene. In the strictest sense, graphene nanoplatelets can be considered as small agglomerations of graphene. Thus, it can be expected that for the large graphene nanoplatelets, a reduction in Tg could be observed while for nanoplatelets which consist of only a few layers of graphene, an increase would be observed. The Young’s modulus, E, and tensile strength, uts, of all the adhesive formulations are summarised in Table 1. It is found that the incorporation of 0.1 wt.% GNPs increased the Young’s modulus moderately from 2.46 GPa of control to 2.56 GPa of G0.1. This indicates that there exists effective load transfer between GNPs and epoxy matrix at the elastic deformation stage under loading. No further improvement in the Young’s modulus was observed as the GNP content increased up to 0.5 wt.%. This is due to the increasing amount of GNP agglomerates at higher GNP contents. From Table 1, one can see that the tensile strength has no significant change due to
the addition of GNPs. This demonstrates that the interfacial adhesion between GNPs and epoxy matrix is not strong enough to maintain load transfer at the failure stage in the tensile test. 3.3. Mode I fracture behaviour 3.3.1. Fracture toughness Table 1 presents the mode-I fracture toughness, KIC, and fracture energy, GIC from the SENB tests. KIC of control was measured to be 2.17 MPam 1/2. This increased to 2.40 MPam1/2 with the addition of 0.1 wt.% GNPs. No further increase in KIC was observed at a higher loading of GNP, up to a particle loading of 0.5 wt.%. Indeed, a slight reduction is observed for G0.3 and G0.5. This is likely attributed to the increased prevalence of GNP agglomerates at higher GNP loading, see Figure 1. The GNP agglomerates appeared as defects in the fracture process and prevented further enhancement in the fracture energy for high GNP content. The fracture energy exhibited a similar trend as the fracture toughness. GIC increased significantly from 2136 J/m2 for control to 2590 J/m2 for G0.1, and then plateaued for higher GNP loading. 3.3.2. Toughening mechanisms Typical TOM micrographs taken under bright-field light (top) and cross-polarised light (bottom) at the fracture subsurface of SENB specimens are presented in Figures 2. The bottom half of each figure represents the subsurface damage. In each image crack propagation is from left to right. A plastic zone with dark colour is observed at the fracture subsurface of all the formulations on the bright-field micrographs. A bright region was observed on the cross-polarised micrographs. This typically indicates the presence of shear band yielding in the bright region on the cross-polarised micrographs and bulk plasticity in the ‘far-field’ plastic damage region on the bright-field micrographs for rubber modified epoxies [13, 25, 36]. The
depth of the plastic zone in the direction perpendicular to the fracture surface is summarised in Table 2. It is clear that the depth of the plastic zone decreased from 117 μm for the control to 73 μm for G0.1 due to the addition of 0.1 % GNPs. This indicates a decrease in the fracture toughness by the dissipation of plastic energy (induced by the CSR nanoparticles [13, 25, 36]) for G0.1 formulation. From Table 1, it is found that the total fracture energy of G0.1 is much higher than that for the control. Hence, additional fracture mechanisms, i.e. crack pinning and crack deflection [21], were introduced by the GNPs to dissipate more fracture energy in the fracture process. These fracture mechanisms do not result in an increase in the plastic zone size. Moreover, the presence of the GNPs impedes the classical CSR toughening mechanisms by restricting the availability of matrix to plastically deform, thus restricting the depth of the plastic zone in the hybrid adhesives. Figure 3 presents typical SEM images of the fracture surfaces of the SENB samples. The images have been taken from just ahead of the pre-crack tip in each sample. A relatively smooth fracture surface with many river-lines was observed for the control adhesive, see Figure 3(a). A much rougher appearance of the fracture surface with was observed due to the addition of GNPs, see Figures 3 (b) and (c). Figures 3 (d)-(f) present images of the same fracture surfaces at higher magnification. The higher magnification reveals an extensive network of voids on the fracture surfaces of the control adhesive, see Figure 3 (d). These voids were generated by the toughening mechanisms of the CSR nanoparticles, i.e. CSR cavitation and plastic void growth in the epoxy matrix [1, 13, 15]. Figures 3 (e) and (f) present typical fracture surfaces of the adhesives containing both CSR nanoparticles and GNPs. Voids generated by cavitation and subsequent plastic void growth can also be observed, similarly to the control adhesive. Furthermore, many pulled out and debonded GNP flakes were also observed on the fracture surfaces. There is significant evidence of gross plastic deformation of the epoxy-CSR matrix in the regions immediately surrounding the GNP stacks. Therefore, it is likely that the
presence of the GNPs contributes an additional dissipative toughening mechanism. 3.4. Single lap shear (SLS) test Figure 4 presents the lap shear strength of the different adhesives tested. The lap shear strength was measured as 21.7 MPa for the control adhesive. The incorporation of GNPs resulted in a steady decrease of lap shear strength, i.e. from 21.7 MPa of the control to 17.2 MPa of G0.5. Figure 5 presents the optical micrographs of the homologous fracture surfaces of the SLS adhesive joints. The blue rectangle demarcates the adhesive bonding area. It should be noted that an extremely thin residual adhesive layer was observed on the left half of the joint for all the adhesives. This is difficult to discern for the control samples as there is a poor contrast in colour between the adhesive and the substrate. Hence, the failure of all the adhesives was cohesive, i.e. the crack was observed to grow inside the adhesive layer, although very close to a bonded interface. Typical SEM images of the SLS fracture surfaces for the control, G0.2 and G0.5 adhesives are presented in Figure 6. A rough fracture surface with many small platelets of polymer was observed for the control adhesive, see Figure 6 (a). The SEM image of the control adhesive with high magnification (Figure 6(b)) shows that the fracture surface is covered with a “sea” of voids, which were generated by rubber cavitation and following plastic void growth [5, 13, 15]. This is similar in appearance to the corresponding fracture surface of the bulk adhesive, see Figure 3 (d). The addition of the GNPs into the control adhesive changed the appearance of the fracture surface significantly, i.e. the fracture surfaces of G0.2 and G0.5 adhesives are characterised by the presence of alternating smooth regions (indicated by red arrows in Figures 6 (c) and (d)), approximately 20-50 μm in size, and rough regions (indicated by yellow arrows in Figures 6 (c) and (d)). A high
magnification image of the roughened regions of the fracture surface, Figure 6 (e), reveal that these rough regions are characterised by the presence of both GNP flakes and voids initiated by cavitation of the rubber particles. It is interesting that very limited number of GNP flakes, such as the indicated one on Figure 6 (e), are observed on the fracture surfaces of adhesives modified with GNPs. The smooth regions identifiable in Figures 6 (c) and (d) are the surfaces of GNP flakes aligned in the direction parallel to the fracture surface. This is demonstrated schematically in Figure 7 (a). The cross-section of the adhesive layer of a fractured G0.5 SLS specimen was imaged using SEM to closer inspect the orientation of the GNPs. Figure 7(b) demonstrates the process used to prepare the SEM sample. A fine-tooth saw was firstly used to cut almost completely through the substrates. The samples were then fractured dynamically into two pieces to reveal a surface of the cross-section. The SEM images of the cross-section are shown in Figure 8. Note that the gap between the adhesive layer and the substrate arose from unavoidable debonding introduced in the sample preparation process. The presence of GNP clusters (indicated by red solid arrows) are observed on the adhesive layer side, and the size of these clusters are approximately the same as the size of the smooth region observed on the fracture surface of the SLS specimens, see Figure 6 (d). This further demonstrates that the smooth region on the fracture surface of SLS specimens are introduced by the GNP clusters, which are aligned in the direction parallel to the fracture surface. The preferential alignment of the GNP clusters in the direction parallel to the fracture surface can be explained by considering the individual steps in the manufacturing procedure of the SLS adhesive joints. In the first step of the manufacturing procedure, an excess of adhesive was placed between two substrates to be joined. A pressure was then applied to the joint during the curing process to obtain the required (0.4 mm) adhesive thickness and remove the excess adhesive. The pressure applied to the joints spread the adhesive evenly in the adhesive joints and
pushed the adhesive flow to the edge of the bonding area. This caused the twodimensional GNP clusters oriented in the adhesive flowing direction, i.e. the direction parallel to the bonding surface. Figure 9 presents a comparison of the measured electric conductivity between the bulk samples and the SLS adhesive joint for both the control, i.e. no GNPs, and G0.5 adhesive formulations. Firstly, it can be observed that in the absence of any conducting carbon nano-filler, the measured electrical conductivity follows a power law behaviour, with an increase in conductivity measured with increasing electrical frequency. Moreover, with the addition of GNPs, it can be seen that in the case of the bulk material, a frequency independent value of electrical conductivity for the composite emerges at lower frequency. This is indicative of the formation of a percolation network formation and can be attributed to the reasonably good dispersion and random orientation of the GNPs and GNP clusters in the nanocomposite. This is in stark contrast to the behaviour of the SLS joint with 0.5 % GNPs. It is found that the electric conductivity of the SLS joint is much lower than the electric conductivity of the bulk specimens due to the identical alignment of the GNP clusters in the SLS adhesive joints. This is further evidence that the orientation of the GNPs in the SLS specimen is no longer random as it is not forming a percolation network. 4. Conclusions This work studied the effect of GNPs on the mechanical properties, fracture toughness and lap shear strength of a rubber-modified epoxy adhesive. The addition of GNPs had no effect on the glass transition temperature. The tensile modulus increased due to the addition of 0.1 % GNPs with no subsequent increase observed for higher GNP loading. The tensile strength was unaffected by the addition of small amount GNPs, up to 0.5 wt.%. Blending small amount of GNPs, i.e. 0.1 %, into the control adhesive increased the fracture energy moderately from 2136 J/m 2 to 2590
J/m2. The addition of further quantities of GNPs did not result in a further increase in the fracture energy. This is attributed to increased agglomeration of the GNPs in the curing process. The addition of GNPs decreased the lap shear strength of the adhesive significantly, i.e. from 21.7 MPa of the control adhesive by approx. 20 % gradually to 17.2 MPa of the adhesive modified by 0.5 % GNPs. This was caused by the preferential alignment of the GNP clusters in the adhesive joints. The alignment of the GNP clusters, introduced by the manufacturing process of the joints, also caused a significant drop in the electric conductivity of the adhesive joints. This, combined with the poor mechanical performance, means that GNP modified adhesives cannot be considered as suitable candidate materials for structural adhesive joints requiring in-situ structural monitoring. Acknowledgements Financial support from the Irish Centre for Composite Research, China Scholarship Council and Henkel (Ireland) is gratefully acknowledged. References [1]
A. J. Kinloch, S. J. Shaw, D. A. Tod, D. L. Hunston, Deformation and fracture behaviour of a rubber-toughened epoxy: 1. Microstructure and fracture studies, Polymer 24 (10) (1983) 1341-1354.
[2]
A. J. Kinloch, Toughening Epoxy Adhesives to Meet Today’s Challenges, MRS Bulletin 28 (2003) 445-448.
[3]
A. I. Dmitriev, I. Husler, W. sterle, B. Wetzel, G. Zhang, Modeling of the stress strain behavior of an epoxy-based nanocomposite filled with silica nanoparticles, Materials & Design 89 (2016) 950-956.
[4]
M. Imanaka, Y. Takeuchi, Y. Nakamura, A. Nishimura, T. Iida, Fracture toughness of spherical silica-filled epoxy adhesives, International Journal of Adhesion and Adhesives 21 (5) (2001) 389-396.
[5]
D. Carolan, A. Ivankovic, A. Kinloch, S. Sprenger, A. Taylor, Toughening of epoxy-based hybrid nanocomposites, Polymer 97 (2016) 179-190.
[6]
G. Marami, S. A. Nazari, S. A. Faghidian, F. Vakili-Tahami, S. Etemadi, Improving the mechanical behavior of the adhesively bonded joints using RGO additive, International Journal of Adhesion and Adhesives 70 (2016) 277-286.
[7]
M. A. Rafiee, J. Rafiee, I. Srivastava, Z. Wang, H. Song, Z.-Z. Yu, N. Koratkar, Fracture and fatigue in graphene nanocomposites, Small 6 (2) (2010) 179183.
[8]
L.-C. Tang, Y.-J. Wan, D. Yan, Y.-B. Pei, L. Zhao, Y.-B. Li, L.-B. Wu, J.-X. Jiang, G.-Q. Lai, The effect of graphene dispersion on the mechanical properties of graphene/epoxy composites, Carbon 60 (2013) 16-27.
[9]
F. Gojny, M. Wichmann, U. Kpke, B. Fiedler, K. Schulte, Carbon nanotubereinforced epoxy-composites: enhanced stiffness and fracture toughness at low nanotube content, Composites Science and Technology 64 (15) (2004) 2363–2371.
[10] M. R. Gude, S. G. Prolongo, A. Urea, Toughening effect of carbon nanotubes and carbon nanofibres in epoxy adhesives for joining carbon fibre laminates, International Journal of Adhesion and Adhesives 62 (2015) 139-145. [11]
M. Imanaka, S. Motohashi, K. Nishi, Y. Nakamura, M. Kimoto, Crack growth behavior of epoxy adhesives modified with liquid rubber and cross-linked rubber particles under mode I loading, International Journal of Adhesion and Adhesives 29 (1) (2009) 45-55.
[12] A. Rezaifard, K. Hodd, D. Tod, J. Barton, Toughening epoxy resins with poly (methyl methacrylate)-grafter-natural rubber, and its use in adhesive formulations, International Journal of Adhesion and Adhesives 14 (2) (1994) 153-159. [13]
D. Quan, A. Ivankovic, Effect of core-shell rubber (CSR) nano-particles on mechanical properties and fracture toughness of an epoxy polymer, Polymer 66 (2015) 16-28.
[14]
G. Giannakopoulos, K. Masania, A. C. Taylor, Toughening of epoxy using core– shell particles, Journal of Materials Science 46 (2) (2011) 327-338.
[15]
J. Chen, A. J. Kinloch, S. Sprenger, A. C. Taylor, The mechanical properties and toughening mechanisms of an epoxy polymer modified with polysiloxanebased core-shell particles, Polymer 54 (16) (2013) 4276-4289.
[16]
B. T. Marouf, R. A. Pearson, R. Bagheri, Anomalous fracture behavior in an epoxy-based hybrid composite, Materials Science and Engineering: A 515 (1– 2) (2009) 49-58.
[17]
S. A. Ngah, A. C. Taylor, Toughening performance of glass fibre composites with core-shell rubber and silica nanoparticle modified matrices, Composites Part A: Applied Science and Manufacturing 80 (2016) 292-303.
[18]
Y. L. Liang, R. A. Pearson, The toughening mechanism in hybrid epoxy-silicarubber nanocomposites (HESRNs), Polymer 51 (21) (2010) 4880-4890.
[19]
B. Ahmadi-Moghadam, M. Sharafimasooleh, S. Shadlou, F. Taheri, Effect of functionalization of graphene nanoplatelets on the mechanical response of graphene/epoxy composites, Materials & Design (1980-2015) 66, Part A (2015) 142-149.
[20] A. S. Wajid, H. S. T. Ahmed, S. Das, F. Irin, A. F. Jankowski, M. J. Green, Highperformance pristine graphene/epoxy composites with enhanced mechanical and electrical properties, Macromolecular Materials and Engineering 298 (3) (2013) 339-347. [21]
Y. J. Lim, D. Carolan, A. C. Taylor, Simultaneously tough and conductive rubber–graphene–epoxy nanocomposites, Journal of Materials Science 51 (18) (2016) 8631-8644.
[22] S. Chandrasekaran, C. Seidel, K. Schulte, Preparation and characterization of graphite nano-platelet (GNP)/epoxy nano-composite: Mechanical, electrical and thermal properties, European Polymer Journal 49 (12) (2013) 38783888. [23]
R. Moriche, S. Prolongo, M. Snchez, A. Jimnez-Surez, F. Chamizo, A. Urea, Thermal conductivity and lap shear strength of GNP/epoxy nanocomposites adhesives, International Journal of Adhesion and Adhesives 68 (2016) 407410.
[24]
L. Guadagno, M. Sarno, U. Vietri, M. Raimondo, C. Cirillo, P. Ciambelli, Graphene-based structural adhesive to enhance adhesion performance, RSC Advances 5 (35) (2015) 27874-27886.
[25] D. Quan, D. Carolan, C. Rouge, N. Murphy, A. Ivankovic, Carbon nanotubes and core-shell rubber nanoparticles modified structural epoxy adhesives, Journal of Materials Science (2016) 1-16.
[26]
BS ISO 527-2:1996, Plastics determination of tensile properties-part 2: test conditions for moulding and extrusion plastics, British Standard Institute (1996).
[27]
ASTM D5045-99, Standard Test Method for Plane-Strain Fracture Toughness and Strain Energy Release Rate of Plastic Materials (1999).
[28]
Q-Lab Corporation, http://www.q-lab.com/q-panel-selector/, last Accessed 03th January 2017.
[29]
A. S. Holik, R. P. Kambour, D. G. Fink, S. Y. Hobbs, Grinding and polishing techniques for thin sectioning of polymeric materials for transmission light microscopy, Microstructure Science 7 (1979) 357-367.
[30] W. Zhao, M. Fang, F. Wu, H. Wu, L. Wang, G. Chen, Preparation of graphene by exfoliation of graphite using wet ball milling, J. Mater. Chem. 20 (2010) 58175819. [31]
H. M. Chong, S. J. Hinder, A. C. Taylor, Graphene nanoplatelet-modified epoxy: effect of aspect ratio and surface functionality on mechanical properties and toughening mechanisms, Journal of Materials Science 51 (19) (2016) 8764– 8790.
[32]
D. Galpaya, M. Wang, C. Yan, M. Liu, N. Motta, E. R. Waclawik, Fabrication and mechanical and thermal behaviour of graphene oxide/epoxy nanocomposites, Journal of Multifunctional Composites 1 (2) (2013) 91-98.
[33] A. S. Wajid, H. S. T. Ahmed, S. Das, F. Irin, A. F. Jankowski, M. J. Green, Highperformance pristine graphene/epoxy composites with enhanced mechanical and electrical properties, Macromolecular Materials and Engineering 298 (3) (2013) 339-347. [34]
M. Naebe, J. Wang, A. Amini, H. Khayyam, N. Hameed, L. H. Li, Y. Chen, B. Fox, Mechanical property and structure of covalent functionalised graphene/epoxy nanocomposites, Scientific Reports 4 (2014) 4375.
[35]
A. Allaoui, N.-E. El Bounia, How carbon nanotubes affect the cure kinetics and glass transition temperature of their epoxy composites? – A review, Express Polymer Letters 3 (9) (2009) 588-594.
[36]
Y. Liang, R. Pearson, Toughening mechanisms in epoxy-silica nanocomposites (ESNs), Polymer 50 (20) (2009) 4895-4905.
Table 1: Thermal and mechanical properties and fracture toughness of different adhesive formulations. Formulation
E (GPa)
uts
wt.% Graphene
Tg (°C)
Control
0
127.2
2.46±0.08 68.1±3.3
2.17±0.14
2136±121
G0.1
0.1
126.9
2.56±0.02 69.5±0.8
2.40±0.17
2590±178
G0.2
0.2
125.9
2.59±0.06 66.8±1.1
2.40±0.09
2604±191
G0.3
0.3
126.7
2.61±0.05 63.4±2.9
2.35±0.11
2580±215
G0.5
0.5
127.5
2.60±0.06 65.8±0.7
2.32±0.12
2499±130
(MPa)
KIC GIC (J/m2) (MPam1/2)
Table 2: Measurement of the fracture subsurface damage. Formulation Control Plastic zone depth (µm) 117
G0.1 73
G0.2 79
G0.3 74
G0.5 71
200 µm (a) G0.1
200 µm (b) G0.2
200 µm (c) G0.3
200 µm (d) G0.5 Figure 1: Dispersion of the GNPs in different adhesive formulations.
G0.3
Control
0.5mm
0.5mm
Figure 2: Typical TOM micrographs of the fracture subsurface of SENB specimens under bright field light (top) and cross polarised light (bottom). Crack growth direction is from left right. The black boxes indicate the plastic deformation zone.
(a) Control-Low magnification
(b) G0.2-Low magnification
(c) G0.5-Low magnification
(d) Control- High magnification
Debonded GNPs Pull-out GNPs
(e) G0.2- High magnification
Pull-out GNPs
Debonded GNPs
(f) G0.5- High magnification Figure 3: SEM images of the fracture surface of SENB samples.
Lapshear strength (MPa)
25
22
19
16
13 0
0.1
0.2
0.3
0.4
0.5
GNP content (wt.%) Figure 4: Lap shear strength of different adhesive formulations.
0.6
Figure 5: Optical micrographs of the fracture surface of the SLS adhesive joints. The blue box indicates the bonding surface.
(a) Control-Low magnification
(b) Control-High magnification
(c) G0.2-Low magnification
(d) G0.5-Low magnification
Rough Region
GNP Flake
Smooth Region
(e) G0.5-High magnification Figure 6: Typical SEM images of the fracture surface of SLS specimens. The red arrows indicate the smooth regions and the yellow arrows indicate the rough regions in (c) and (d).
Adhesive layer GNP clusters
Substrates (a)
Adhesive layer
Substrates (b) Figure 7: Schematics to show (a) the alignment of the GNP clusters in the SLS specimens and (b) the sample preparation for SEM analysis to confirm the GNP alignment. The dashed line in (b) indicates the broken plane, and SEM images were taken from the top.
Adhesive layer
Substrate
Figure 8: SEM images of the cross-section of the adhesive layer of a fractured G0.5 SLS specimen. The arrows indicate GNP agglomerates.
1.E+00 1.E-01 Bulk-Control
Electric Conductivity (S/m)
1.E-02 1.E-03 1.E-04
Bulk-G0.5 SLS Joint-Control SLS Joint-G0.5
1.E-05 1.E-06 1.E-07 1.E-08 1.E-09 1.E-10 1.E-11 1.E-12 1.E-01
1.E+00
1.E+01
1.E+02
1.E+03
1.E+04
1.E+05
1.E+06
Frequency (Hz) Figure 9: The electric conductivity of the bulk specimens and the SLS adhesive joints for the Control and G0.5 adhesives.