Design, Preparation and Characterization of a High-Performance Epoxy Adhesive with Poly (Butylacrylate-block-styrene) Block Copolymer and Zirconia Nano Particles in Aluminum- Aluminum Bonded Joints

Epoxy adhesives are one of the polymers used as high-performance matrix in adhesives. However, the high brittleness and low toughness of epoxy adhesives are critical challenges during their service in structural applications due to their high-crosslinking degree. The lack of appropriate high-temperature thermal stability is another drawback of these valuable materials. This study addressed the effect of hybrid reinforcement comprising zirconium oxide nanoparticles (NPs), phenolic resin (resol type), and poly (butyl acrylate-block-styrene) copolymer (BCP) on mechanical, adhesion, thermal, and morphological properties of the epoxy adhesive. Mechanical properties, thermal stability, and microstructure of the epoxy adhesive was assessed using tensile test, TGA, and FESEM tests, respectively. The adhesion features of the formulated adhesive were evaluated in lap joint bonding of an aluminum to aluminum. A new approach was developed to design advanced adhesives with high mechanical, adhesion, and thermal properties by adding hybrid additives. Based on the tensile test results, adding 5 phr of zirconium oxide nanoparticles to the epoxy adhesive increased the tensile strength, modulus, and the toughness of the dumbbell-shaped samples by 69, 33 and 175% compared to the neat epoxy adhesive, respectively. Furthermore, the sample containing 10 phr phenolic resin, 5 phr zirconia NPs, and 2.5 phr block copolymer exhibited the highest improvement (420%) in the shear strength in the single lap joint increment compared to pure epoxy, reflecting the synergistic impact of these compounds at the mentioned percentage. The TGA results indicated the highest initial degradation temperature in the sample containing 5 phr zirconia NPs which was 54.4 °C higher than that of the pure epoxy. The images of the fracture surface of the optimal samples in the tensile test showed the cavitation, shear band formation, crack deviation, and crack tip blunting as major mechanisms in the toughness enhancement of the samples.


Introduction
Thanks to their high load-bearing properties, epoxy resins have found extensive applications in the production of composites, especially in the engineering fields.They are also widely utilized in coatings due to their excellent chemical resistance and electrical resistivity.As a high-performance polymeric matrix, epoxy resin has found unique position in the construction, automotive, and aerospace fields [1].The structure of the epoxy resin, however, suffers from poor resistance against the crack initiation and growth and low toughness due to its high crosslink degree which can limit its applications in the fields requiring high fracture strength.The tensile strength and toughness of the epoxy resin depend on various factors such as the type of hardener, diluent, and filler [2].
Numerous studies have recently addressed the enhancement of the fracture strength and toughness of the epoxy resin, among which, block copolymers can be mentioned.Block copolymers (BCP) refer to a class of polymers comprising epoxy-phillic and epoxy-phobic blocks within an epoxy matrix, giving rise to intertwined spherical and cylindrical micelles depending on the block ratios and the content.The fracture toughness can be remarkably enhanced by adding a small amount of these BCPs (less than 5 phr) while showing no significant negative impact on the T g and modulus [3].
Moreover, the incorporation of a small content of nanoparticles into the epoxy adhesive can elevate its toughness, glass transition temperature, and single lap shear strength at room temperature [4].
Studies have revealed that inorganic NPs such as SiO 2 [5,6], nano clay [7], Al 2 O 3 [8], Fe 2 O 3 [9], and GO [10] can improve the mechanical and thermal features of the epoxy resin.Among these NPs, zirconia nanoparticles (ZrO 2 ) have shown high strength, stiffness, corrosion resistance, hardness, and fracture toughness, introducing them as a proper candidate for enhancing the mechanical properties of the polymeric composites [11,12].However, a limited number of studies have addressed this material.
Brochier et al. assessed the effect of zirconia NPs on the single lap joint of the epoxy adhesives.They reported a 60% increment the single lap shear strength upon the incorporation of 1% zirconia NPs [13].Khosravi et al. reported 27,62, and 110% enhancement in the tensile strength, stiffness, and fracture toughness of the epoxy/glass fiber upon adding zirconia NPs, respectively [14].
Azizi and Eslami examined the influence of silane-modified zirconia NPs on epoxy/basalt composite.Their results indicated that the incorporation of 3% zirconia NPs managed to respectively improve the bending strength, bending modulus, bending fracture elongation, tensile strength, tensile modulus, and tensile fracture elongation by 90, 74, 84, 76, 85, and 14% compared to samples free of zirconia NPs [15].
Rosa Medina et al. evaluated the effect of zirconia NPs on the epoxy resin.They found that the addition of 10% zirconia NPs can elevate the modulus and fracture toughness by 37 and 100%, respectively.Moreover, the incorporation of 8% zirconia NPs enhanced the glass transition temperature by 8% when compared with the pure epoxy samples [16].
Xiaoqian Ma et al. examined the effect of zirconia NPs on the epoxy resin.They indicated that the best mechanical properties could be achieved at 3 wt% NPs content showing 44, 29, and 63% increment in the tensile strength, tensile modulus, and fracture toughness compared to the pure epoxy sample, respectively [17].
Gomez-del Río et al. explored the effect of styrene-butadiene methyl methacrylate (SBM) block copolymer and CNT on epoxy.The incorporation of 5 phr SBM increased the fracture toughness and fracture energy by 115 and 410%, respectively.Furthermore, adding 0.25 phr CNT increased the fracture toughness by 14.5%; while the hybrid of these two compounds exhibited no effect on fracture toughness [18].
Jojibabu Pantaa et al. studied the effect of the block copolymer and surface-modified carbon nanotubes on the lap shear strength.The results revealed that the addition of surface-modified CNT to pure epoxy increased its shear strength by 26%.A hybrid of surface-modified CNT and SBM tri-block copolymer (styrene-butadiene methyl methacrylate) elevated shear strength by 137% compared to pure epoxy, reflecting their synergistic effect at this percentage [19].
This research thus presents a comprehensive study on the effect of zirconia nanoparticles and block copolymer (butyl acrylate-block-styrene) on mechanical properties, adhesion strength, and thermal stability of epoxy-based nanocomposite adhesives by studying the microstructure and toughening mechanisms and also the effect of temperature on mechanical properties, adhesion strength epoxy-based nanocomposite adhesives.

Materials
In this research, epoxy Diglycidyl Ether Bisphenol A (DGEBA; with the brand name of EPONATE 500, viscosity of 14,000-11,000 cp, and room-temperature density of 1.16 g/cm 3 ) and cycloaliphatic hardener (with the brand name of EPIKURE F205; Bitex composite; Turkey) were prepared at the resin/hardener ratio of 100:55 and density of 0.98 g/cm 3 .

Preparation Method
To prepare the samples, epoxy resin (20 g) was first poured into a suitable container followed by adding 2 ml of xylene.The mixture was then mechanically stirred for 3 min at 200 rpm.Zirconia nanoparticles were then added to the mixture in specified percentages (before addition, the zirconia nanoparticles were weighed and placed in the oven at 110 °C for 2 h for dehumidification).To better mix the nanoparticles in the epoxy matrix, they were also stirred by a mechanical stirrer at a certain speed.
In the following, liquid phenolic resin (2, 4, and 6 g) was added and mechanically stirred for 15 min at 200 rpm.Afterward, butyl-acrylate-block-styrene (0.25, 0.5, and 0.75 g) was added as a toughening agent and stirred for 20 min at 200 rpm.The mixture was then placed in an oven to remove bubbles that may enter the mixture during the stirring process.Subsequently, the curing agent (10 g) was added and mechanically stirred for 10 min at 200 rpm.The prepared adhesive was then poured into dumbbell-shaped molds to prepare specimens for mechanical evaluation tests.Some parts of the adhesive were also used for metal-metal lap joint, butt joint, and T-peel tests.The samples were removed from the molds after 24 h and placed at ambient temperature for 7 days for final curing.

Formulation
The following formulations were designed to assess the effect of phenolic resin, BCP, and zirconia NPs contents on the mechanical properties of the epoxy adhesive.The formulation of epoxy adhesives with different components are presented in Table 1.

Effect of Temperature on the Mechanical Properties of the Optimal Pure Epoxy Sample
In this test, the optimal composition comprising 10 phr phenolic resin, 2.5 phr toughening agent, and 5 phr nanoparticles was added to the epoxy.Then, dumbbellshaped samples and Al-Al single lap joint were prepared at room as well as low temperature (− 60 °C).Al-Al butt joint samples were also fabricated at room, low (− 60 °C), and high (+ 120 and 180 °C) temperatures.T-peel samples were prepared at room temperature to evaluate the mechanical properties.Table 2 lists the factors and their corresponding levels in the samples.

Surface Treatment
Anodizing treatment was carried out before applying adhesive on the Al surface: Al foil (as the anode) was connected to the positive pole of the source while the auxiliary electrode (Pt) was connected to the negative pole to serve as a cathode.They were placed in sulfuric acid (15%) solution by a clamp to conduct electrochemical anodizing at 9 V for 5 min.

Tensile Test on Dumbbell-Shaped Specimens
The tensile test was conducted on the dumbbell-shaped samples according to ISO 527-1,2,3 to determine the tensile properties of the polymers.The length, width, and thickness of the dumbbell-shaped samples were 75, 25, and 5 mm, respectively.Each sample was tested in 5 replicates.According to the mentioned standard, the loading rate of the sample for the tensile test was 2 mm/min.This test was performed by a SANTAM STM-20 tensile device in the central lab of Isfahan University.

Lap Shear Strength Test
This test measures the strength of metal-metal joints based on the ASTM D1002 standard.Al pieces were cut into 100*25*2 mm 2 and the lap domain was 12.5 mm.Each sample was tested in 5 replicates.According to the mentioned standard, the thickness of the adhesive was 0.8 mm while the tensile rate was set to 1.3 mm/min.

Butt Joint Strength Test
Al-Al joint specimens were placed in the tensile test apparatus according to the ASTMD2094-00 standard to characterize adhesives in metal-metal joints.A definite length of the samples (12.7 mm) was placed in the pins while the loading rate was set to 0.5 mm/min.According to the mentioned standard, the thickness of the applied adhesive was 0.1 mm while the lap domain was 3.24 mm 2 .This test was carried out at room temperature as well as − 60, 120 and 180 °C.The two latter temperatures were prepared using dry ice and a furnace, respectively.The results of this test were presented as the findings of butt joint strength test, as depicted in Fig. 1a-d.

T-peel Strength Test
This test is indeed a groove test; a T-shape is the major type of peeling according to the ASTM D1876.Two Al layers with similar thickness are attached by an adhesive; then the ends of the test sample are detached by a tensile test device at the rate of 127 mm/min.Figure 2 shows the T-Peel test device and the dimensions of the used sample.

Thermogravimetric Analysis (TGA)
TGA was carried out to check the thermal stability of the samples on the tensile test samples with the best response.The samples were heated from the ambient temperature to 500 °C under the nitrogen atmosphere at a rate of 10 °C/min.The sample weight was 10 g which was cut into small plates and placed on Al planes to be heated up to 500 °C.The initial degradation temperature (T IDT ), maximum degradation temperature (T max ), and residual char or non-volatile fraction were determined up to the temperature of 500 °C.A TGA device model DSC1 METTLER TOLEDO, Switzerland was employed for this purpose.

Field-Emission Scanning Electron Microscopy (FESEM)
For FESEM imaging, the samples were first coated with a thin layer of gold to achieve clear imaging.After the preparation process, the samples were placed in the imaging device.The toughening mechanism and distribution of zirconia NPs in the epoxy matrix were explored by the FESEM technique (MIRA 3, TESCAN).The imaging was carried out on the dumbbell-shaped samples of the tensile test with the best response.

Fourier-Transforms Infrared (FTIR) Spectroscopy
To investigate the reaction and the bonds formed upon curing the adhesive, the dog-bone specimens with the best tensile performance were ground into fine particles for FTIR analysis.The IR spectrum of the sample was recorded in the range of 4000-400 cm −1 with spectral resolution of 4 cm −1 using a spectrophotometer (Model 6300 Japan).

Differential Scanning Calorimetry (DSC)
Cure characterization of the samples were examined by DSC analysis under nitrogen gas and heating rate of 10 °C/min in the temperature range of 30-150 °C.Three samples (pure epoxy, epoxy containing 5 phr zirconia NPs, and epoxy containing 2.5 phr BCP along with the curing agent) were prepared.The tests were conducted in the central lab of Isfahan University using a DSC device (DSC 131, Setaram, France).

Results and Discussions
Figure 3 depicts the findings of the tensile test conducted on the dumbbell-shaped samples.Accordingly, tensile strength and modulus declined upon adding phenolic resin to the epoxy which can be assigned to the lower strength and modulus of the phenolic resin compared to the epoxy.Moreover, phenolic resin showed high shrinkage during the curing process, giving rise to pores and cracks in the adhesive structure; which can consequently decrement the strength and modulus compared to the pure epoxy [8,29].The highest decline in the modulus and tensile strength (1672 and 7.3 MPa) was observed in the sample containing 30 phr liquid phenolic resin.The toughness of the system also showed a decrement at all phenolic resin levels.
As suggested in Fig. 3, the modulus continuously decremented upon adding BCP to the epoxy at all levels.Such a decrement can be attributed to the presence of soft blocks of butyl acrylate in the copolymer with a lower modulus relative to the epoxy.Tensile strength also increased up to the BCP level of 2.5 phr followed by a decrease which can be due to the hydrogen bonding between the carbonyl groups of the butyl acrylate and the hydroxyl group of the epoxy [30].The highest tensile strength (39.4 MPa) was related to the BCP level of 2.5 phr, showing a 40% increment compared to pure epoxy.Furthermore, the highest toughness was recorded at this BCP level, exhibiting a 78% enhancement compared to the pure epoxy due to the proper interaction of the BCP with the epoxy and the formation of nanostructured spherical micelles.The increase in toughness can be also assigned to the enhancement of the elastic energy stored in butyl acrylate rubber particles during the sample elongation [31].At BCP levels beyond 2.5 phr, the modulus and tensile strength decreased to 2009.25 and 35.5 MPa, respectively; due to the enlargement of the BCP structure in the epoxy matrix, reducing its interface with epoxy chains and declining the mechanical properties [32].
According to Fig. 3, the modulus and strength rose by adding zirconia NPs to epoxy such that the highest modulus and strength reached 2867 and 47.61 MPa at the NPs level of 5%, showing 33 and 69% enhancement compared to the pure epoxy, respectively.The reason can be the higher modulus and strength of nanoparticles compared to the epoxy.Moreover, the addition of zirconia NPs restricted the movement of epoxy chains [12,15].Zirconia nanoparticles prevent the opening of a growing crack due to their spherical geometry.Therefore, more energy is required for crack growth and as a result, the energy absorption increases.On the other hand, the hydroxyl groups on zirconia NPs facilitate the curing reaction, leading to higher ring-opening and improved mechanical properties.The highest toughness was also observed in 5 wt%, exhibiting 175% enhancement compared to pure epoxy [15][16][17].
Figure 4 depicts the tensile test results of ternary and quaternary dumbbell-shaped samples.Accordingly, the  and strength increased by elevating the content of zirconia NPs in the epoxy/phenolic matrix, such that the highest modulus, strength, and toughness were recorded in the 10P-5Zr sample.A rise in NPs content compensated for the structural defect due to the presence of resin and increased the mechanical properties.Overall, the mechanical properties decremented relative to pure epoxy at these NPs levels, mainly due to the presence of phenolic resin in the structure of the adhesive, which releases formaldehyde during the curing process, causing cavitation and defects in the adhesive structure and reducing the modulus, strength, and toughness [7,8,26].
According to Fig. 4, the incorporation of zirconia NPs to the epoxy/BCP matrix increased the modulus, strength, and toughness; such that the greatest improvement in mechanical properties was observed in the 2.5C-5Zr sample whose modulus and strength were 1566 and 31 MPa, respectively while its toughness reached 1239 kJ/m 3 .The toughness increased by 117% compared to pure epoxy.Such an enhancement can be due to the good adhesion and strong interface of the nanoparticles with the matrix [27,28].
Based on Fig. 4, the tensile strength and toughness of the adhesive in the 10P-C2.5 dumbbell-shaped sample increased by the incorporation of BCP.The tensile strength and toughness reached 32.9 MPa and 672 kJ/m 3 showing 17 and 18% enhancement compared to pure epoxy, respectively.
Concerning the optimal 2.5C-10P-5Zr sample, the modulus and strength decreased by 69 and 54% compared to pure epoxy, respectively reflecting a negative interaction of materials at this level.However, the toughness increased by 51%.Instead of reacting with epoxy groups, the hardener was surrounded by acrylate groups of BCP, hydroxyl groups of phenolic, and nanoparticles, thus failing to open the epoxy oxirane rings.This reduced the crosslinking degree and enhanced the toughness while decreasing the modulus and strength.At this level of the materials, the viscosity of the adhesive increased, hindering the exit of the bubbles, thus, reducing the mechanical properties [8,37,38].

Predictive Modeling of Mechanical Properties of Nanocomposites
The properties of the polymer and the reinforcement, their volume fractions, the distance between the reinforcement particles, and their distribution and interaction with the matrix are among the influential factors in the mechanical properties of polymer nanocomposites.However, the nanocomposites are mainly developed through experimental approaches with difficulties in the control of properties.In this regard, modeling the mechanical properties of nanocomposites is highly important.Among the modeling methods of mechanical properties, continuum media mechanical methods have gained increasing popularity due to their feasibility and the ability to predict properties regardless of the type of microstructure in atomic dimensions [68].For this purpose, theoretical predictive models of the mechanical properties of nanocomposites such as Halpin-Tsai, series, parallel, and Kerner models were used and In the above equations, E c , E m , and E f respectively denote the modulus of the composite, matrix, and reinforcement.Moreover, V f and V m show the volume fraction of the reinforcement and matrix, respectively; while ν m in the Kerner equation (Eq.5) represents the Poisson coefficient of the epoxy resin which is equal to 0.33.In the Halpin-Tsai equation (Eq.1), ξ = 2 for spherical particles.
The modulus of the pure epoxy was calculated as 2.149 GPa while the modulus and density of zirconia NPs were 200 GPa and 5.89 g/cm 3 , respectively.The volume fraction of zirconia NPs was determined 0.1, 0.2, and 0.5 [35,36].
The general rule of mixtures is one of the basic theories of composites with extensive applications.This theory has upper (parallel) and lower (series) bounds.The upper (1) bound can be obtained by assuming equal strains for filler and matrix; the series model can be achieved by assuming equal stress for the filler and matrix.Halpin-Tsai equation is also a semi empirical equation capable of evaluating the mechanical properties of nanocomposites based on the matrix and geometry of filler particles.According to Fig. 3c, experimental results indicated a 33% increment in the modulus at 5 vol% of zirconia NPs.As mentioned before, the higher modulus of nanoparticles, their proper dispersion, and interaction with the epoxy matrix are due to their hydroxyl groups.The Young moduli predicted by Kerner, series, and Halpin-Tsai models showed a proper consistency with the empirical results at NP content of 0.1 vol%.However, predictive models showed no significant enhancement in the modulus with increasing the volume fraction of NPs.
In the parallel model, the modulus showed a 10% increase compared to the experimental results at NPs volume fractions of 0.5 and 0.2%.It can be said that theoretical models do not have good accuracy in predicting the elastic modulus as they consider an ideal bond between nanoparticles while ignoring defects such as bubbles and aggregation as well as the defects of the nanoparticles themselves [33][34][35].

Results of Single Lap Shear Strength Test for Metal-Metal Joints
As shown in Fig. 6a, the shear strength continuously increased with enhancing phenolic resin levels.Due to the increase in hydroxyl groups in phenolic resin, a proper interaction is formed with the substrate surface, incrementing the adhesion of the adhesive to the Al substrate.Furthermore, phenolic hydroxyl groups increased the ring-opening and crosslinking degree, thus, augmenting the adhesion strength of the adhesive to the substrate.The highest shear strength (1.48 MPa) was detected in the sample containing 30 phr phenolic resin, exhibiting a 41% increment compared to pure epoxy [6,7,40].According to Fig. 6b, the incorporation of BCP continuously enhanced the shear strength; such that the highest shear strength (1.72 MP) was observed at the BCP level of 3.75 phr level, showing a 63% increment compared to the pure epoxy.Such an enhancement can be attributed to the hydrogen bonds between the hydroxyl groups of epoxy and the ester groups of butyl acrylate in the BCP.The presence of spherical BCP micelles in the adhesive increased energy dissipation, improving the stress transfer from the epoxy matrix in the presence of BCP, thus, increasing the shear strength [41,42].
As presented in Fig. 6c, the shear strength increased with raising NPs content as the highest shear strength (3.2 MPa) was observed in the 5Zr sample, exhibiting a 212% increment compared to the pure epoxy.Such an enhancement can be ascribed to the chemical interactions between zirconia  NPs and epoxy chains as the hydroxyl groups on the surface of zirconia NPs can also form hydrogen bonds with epoxy [20,27,28].Higher percentages of zirconia NPs enhanced the surface roughness, hence, improving their adhesion to the Al substrate.Moreover, the increased adhesion between zirconia nanoparticles and epoxy can augment the adhesion of nanocomposite to the metal substrate as ZrO 2 NPs bridge between the epoxy adhesive and the substrate, further incrementing the shear strength [43][44][45]65].
Based on Fig. 6d, the shear strength increased with elevating the content of zirconia NPs in the epoxy/phenolic matrix.The highest shear strength (4.01 MPa) was observed in the 30P-5Zr sample, showing a 382% enhancement compared to the pure epoxy.The presence of hydroxyl groups on zirconia NPs and phenolic resin led to strong bonding between the adhesive and the surface [40,43,65].
According to Fig. 6e, the shear strength reached 4.34 MPa in the optimal sample of 2.5C-5Zr, showing a 313% increment compared to pure epoxy.In the case of the optimal 2.5C-10P-5Zr sample, the shear strength further grew to 5.47 MPa, offering a 420% rise relative to the pure epoxy.The stress transfer from the matrix was improved due to the proper interaction and hydrogen bonding between the hydroxyl groups of phenolic resin, carbonyl groups of BCP, and hydroxyl groups on the surface of zirconia NPs with epoxy rings and the anodized aluminum substrate, which significantly increased the shear strength in the adhesive formulation.At these levels, materials can be said to have a synergistic effect [42][43][44].
Figure 7 depicts the mechanism of the interaction between the materials in the adhesive with the anodized aluminum substrate which indicates the hydrogen bonding between the adhesive formulation components as well as the anodized substrate with the hydroxyl groups of the phenolic resin, BCP, and zirconia NPs.

T-peel Test Analysis
T-peel strength depends on the angle of peeling, test speed, nature of the adhesive, physical and mechanical properties of the substrate, temperature, and humidity of the environment.
Figure 8 shows the results of the T-peel test for pure epoxy and the optimal sample, which indicates the highest records in the lap shear strength test.In the case of pure epoxy, the T-peel strength was 27.8 N/mm which rose to 83.4 N/mm in the 10p-2.5C-5Zrquaternary sample, showing a three-fold enhancement.It can be said that the presence of zirconia nanoparticles and BCP transferred the stress to the substrate.Moreover, the energy dissipation during the crack growth in the T-peel test enhanced the T-peel strength in the quaternary specimen [66,67].

Analysis of the Dumbbell-Shaped Samples at − 60 and 120 °C
Temperature is an effective factor in the strength of the adhesive joints playing a key role in the design of these joints.The applicability of the adhesive in a wide range of temperatures is highly desired.Temperature also affects the structure of the adhesive, especially at temperatures near the glass transition temperature (T g ).At temperatures above T g , the properties of the adhesive such as elasticity, fracture strength, and elastic modulus are declined.At temperatures far below T g , however, these features may show reverse behavior [25].
Fig. 7 Proposed mechanism describing the interaction of materials in a single lap metal-metal joint Fig. 8 T-peel test results for pure epoxy and optimal quaternary sample High temperatures decrement the strength of the adhesive; whereas low temperatures may result in its brittleness.The strength of the adhesive generally declines with temperature elevation; while its ductility shows an increment.High-temperature adhesives are often highly brittle at lower temperatures; while those designed for low-temperature applications usually exhibit poor strength at high temperatures [23,25].
Bai et al. studied the effect of temperature on the single lap metal joint of carbon-epoxy adhesive composite and showed an 80% reduction in the single lap strength and stiffness at temperatures beyond the glass transition temperature [21].
Lu Ke et al. addressed the effect of temperature on the single lap joint of carbon-epoxy composite and revealed a 50% reduction in the adhesion strength at 70 °C compared with room temperature [22].
Banea and da Silva studied the influence of temperature on the epoxy single lap strength at − 40, room temperature, and 80 °C.They observed a 30 and 10% reduction in the shear strength of the composite at 80 °C compared to the room temperature and − 40 °C, respectively [23].
Banea and colleagues also investigated the effect of temperature on single lap metal-to-metal joints at 100, 125, 175, and 200 °C.At temperatures below T g , the shear strength increased by approximately 9 and 30% at 100 and 125 °C, respectively.However, at temperatures above the glass transition temperature, a noticeable decrease can be seen in shear strength.Single lap shear strength decreased by 54 and 75% at 175 and 200 °C, respectively [25].
Figure 9 presents the tensile test results for pure epoxy and optimal quaternary dumbbell-shaped specimens at room temperature, − 60, and 120 °C.
The modulus of the pure epoxy and 10P-2.5C-5Zrshowed a slight increase at − 60 °C compared to the room temperature which can be assigned to the harder movement of the polymeric chains and a reduction in the free space between the chains.In the case of pure epoxy and optimal quaternary samples, toughness showed a 32 and 12% decline due to the decrease in the toughening mechanisms and easier initiation of the crack growth.The toughening mechanisms used by zirconia NPs and BCP to increase plastic deformation and energy loss got more difficult at low temperatures, lowering the toughening effect of zirconia NPs and BCP.At − 60 °C, the tensile strength of pure epoxy and the optimal quaternary samples showed 46 and 23% decrement, respectively.
By temperature decline, the adhesive fracture occurs earlier, i.e., before the adhesive undergoes massive deformation, which reduces the strength and toughness [46,50].
At 120 °C, the tensile strength and modulus of optimal quaternary samples showed 41 and 5% decrement, respectively.
According to Fig. 9, as the temperature increases, the interaction between the materials in the adhesive decreases and reduces the cohesive strength [58].

Analysis of Single Lap Joint Test at − 60 and 120 °C
Figure 10 shows the results of the single lap tensile strength test for pure epoxy and the optimal quaternary samples at room temperature as well as − 60 and 120 °C.
In the case of pure epoxy, the shear strength decreased by 26% at − 60 °C compared to room temperature.The shear strength of the optimal 10P-2.5C-5Zrreached 4.32 MPa at − 60 °C, showing a 78% decrease relative to the optimal sample at room temperature.At low temperatures, the movement of the epoxy chains on top of each other is hindered, which increases external stresses and facilitates crack growth.At low temperatures, the adhesive loses its ductility and energy absorption hence, fails to withstand external stresses, which reduces the shear strength.In the optimal sample, temperature elevation to 120 °C enhanced the shear strength to 1.2 MPa, i.e., approximately 4.5 times lower than that of the room temperature due to the reduction of intermolecular forces and the interactions between the components of the nanocomposite adhesive with each other and the substrate [54,55].

Analysis of Butt Joint Tensile Strength at Different Temperatures
The butt joint tensile strength of the 10P-2.5C-5Zrsample is depicted in Fig. 11 at room temperatures as well as 120, 180, and − 60 °C.The butt joint tensile strength of the optimal sample was 0.31 MPa at room temperature which declined to 0.22 MPa at 120 °C, showing a 29% decrement compared to the room temperature.At 180 °C, the butt joint tensile strength reached 0.21 MPa, exhibiting a 32% reduction compared to the room temperature.
It can be said that at temperatures higher than T g , the flowability of the adhesive increased and the adhesive became softer and separated from the substrate.At temperatures above T g , all the mechanical properties of the adhesive reduced, which is consistent with previous reports [50,51].
According to Fig. 11, at − 60 °C, the butt tensile strength reached 0.25 MPa, showing a 19% decrease compared to the room temperature.Similar to the trend of the single lap joint, it can be said that the toughness mechanisms decreased at this temperature, which reduced the tensile strength.Moreover, the glass transition temperature of epoxy adhesive is close to 90 °C below which, the adhesive has a brittle state with low fracture energy [52,53,55].

Analysis of FTIR Spectra of Pure Epoxy, and Optimal Binary, Ternary, and Quaternary Samples
Figure 12a depicts the FTIR spectra of the dumbbell-shaped pure epoxy, binary, ternary, and quaternary samples.
The peak emerging at 3431 cm −1 can be assigned to the stretching vibration of the hydroxyl groups of epoxy.The stretching vibration of C-H in the aromatic ring of epoxy also appeared at 3037 cm −1 .The peaks at 2973 and 2933 cm −1 can be attributed to the stretching vibrations of CH 3 and CH 2 groups; while those at 1572 and 1510 cm −1 are related to C-C stretching vibrations in the aromatic ring of epoxy.An absorption peak at 922 cm −1 is attributable to the epoxy ring.The emergence of some peaks at 649 and 559 cm −1 indicate the bending vibration of N-H and C-H.A weak peak at 1731 cm −1 is indicative of the carbonyl groups in BCP.While the one appearing at 765 cm −1 is related to the O-Zr-O bond in zirconia NPs [56,57,59].The intensity of hydroxyl peaks showed an increase in 5zr, 2.5C, and 10P indicating a rise in the ring-opening process and an enhancement in the hydroxyl groups in the mentioned formulation of epoxy resin.FTIR spectrum of pure zirconia NPs was also recorded as presented in Fig. 12b.A broad band at 3500 cm −1 represents the hydroxyl groups in the structure of nanoparticles which can enhance the interactions and H-bonding with the epoxy chain, hence, improving the mechanical and thermal properties as confirmed by the results of mechanical properties [48,62,66].

Thermal Stability
Table 3 and Fig. 13 present the thermal stability of pure epoxy and optimal samples.Bisphenol A diglycidyl ether type epoxy resin has good thermal stability with a degradation onset temperature of 175.8 °C, maximum degradation temperature of 375 °C, and residual char of 9.63% at 500 °C due to the formation of networks with high crosslink density.The thermal degradation of epoxy resin involves three stages.The first stage of degradation occurs at 100-175 °C due to the loss of surface Fig. 12 a FTIR spectra of dumbbell-shaped pure epoxy and optimal binary, ternary, and quaternary specimens.b FTIR spectrum of pure zirconia NPs moisture and the degradation of the volatile compound with low molecular weight in the epoxy adhesive.A weight loss from 150 to 350 °C is due to the degradation of oxygenated groups such as the hydroxyl groups of epoxy resin.The main weight loss of the second stage at 175-375 °C can be assigned to the destruction of bisphenol A aromatic groups, the breakdown of the bonds formed with the curing agents, and the release of flammable gases, amines, and gaseous aromatic compounds.Complete degradation of the epoxy network occurred at 375-500 °C [58,59] According to Table 3, the increase in the amount of residual char can be due to the high thermal stability of the phenolic resin, high breakdown energy of the bonds formed with the phenol rings, and also the increase of the aromatic rings.Residual char acts as an insulating layer against the diffusion of oxygen and heat, preventing further degradation of the sample.The residual char continuously rose by enhancing the phenolic resin content such that the highest residual char (23%) was observed in the sample containing 30 wt% phenolic resin.Furthermore, the initial degradation temperature reached 201.7 °C, i.e., 25.8 °C higher than that of pure epoxy.The maximum degradation temperature also reached 385 °C, which is 10 °C higher than pure epoxy [59][60][61].
In 1.25C, 2.5C, and 3.75C samples, the initial and maximum degradation temperatures showed an enhancement.The mobility of the epoxy chains showed a decline due to the proper interactions between the acrylates of the BCP and epoxy rings and their H-bonding.The presence of benzene rings in the styrene structure of the BCP increased the thermal stability of the system which incremented the energy required for the release of volatiles.Additionally, the residual char continuously increased with elevating the BCP content due to the increment of the benzene rings.The highest residual char was observed in the 3.75C sample, showing a 2.33% increment compared to pure epoxy [8,57].
According to Table 3, a rise in zirconia content enhanced the initial degradation temperature as well as the maximum degradation temperature compared to the pure epoxy.The enhancement in the initial degradation temperature can be assigned to the proper dispersion of zirconia NPs in the epoxy matrix and their appropriate interaction and adhesion with the epoxy chains due to the presence of hydroxyl groups on the surface of NPs which declines the mobility and slipping of the chains on top of each other, hence, higher temperatures and energy will be required to overcome the forces between the epoxy chains and NPs.Due to their inorganic nature, zirconia NPs enjoy higher thermal stability, which can elevate the thermal stability of the entire system [53][54][55].The results of mechanical properties and FESEM images confirm the proper dispersion of nanoparticles in the epoxy matrix.
The initial degradation temperature and maximum degradation temperature of the 10P-5Zr sample showed an increment compared to the pure epoxy.Moreover, the residual char was enhanced due to the presence of phenolic resin in the composition of the adhesive which can augment the thermal stability of the system due to its high bonding energy and proper interaction with epoxy as a consequence of their benzene rings [61,65].
In the optimal 2.5C-5Zr sample, the initial degradation temperature and maximum degradation temperature also exhibited an increment compared to the pure epoxy.The presence of well-distributed zirconia NPs in the epoxy matrix hinders the mobility of the polymer chains.The presence of BCP also promotes interactions of H-bonding with Fig. 13 TGA diagram of the pure epoxy and optimal samples epoxy chains, hindering the release of volatile gases and compounds from the adhesive system [6,8,60].
According to Table 3, in the case of optimal 10P-2.5C-5Zr, the initial degradation and maximum degradation temperatures rose compared to the pure epoxy.The residual char was enhanced due to the presence of phenolic resin and BCP and a rise in the content of benzene rings, which incremented the thermal stability.Moreover, it can be said that the mobility and slipping of the epoxy chains decreased due to the proper interactions between zirconia NPs, acrylate blocks, and phenolic in the formulation of the adhesive.Such a decline in mobility is the main obstacle to the release of volatile compounds [8,37,57].4 present the results of studying the cure kinetics in pure epoxy as well as 5Zr and 2.5C samples.

DSC Analysis
The DSC diagram shows a peak for all three samples.This form of peaks is related to one-step reactions [24].Considering the shape of the peaks, it can be said the curing reaction of epoxy adhesives occurred in a single step and the type of reaction did not change with the addition of zirconia NPs and BCP.According to Table 4, the initial curing temperature of the optimal epoxy/nano-zirconia and epoxy/copolymer samples showed a decrease compared to the pure epoxy.The incorporation of zirconia NPs and BCP facilitated the reaction of the curing agent with epoxy, thus increasing the crosslinking degree [48].Furthermore, compared to the pure epoxy, the enthalpy of reaction was higher in the samples containing BCP and NPs, indicating an increase in the released heat, which enhanced the ringopening and crosslinking degree, thus, augmenting the tensile strength compared to pure epoxy.These results are consistent with the results of dumbbell-shaped samples [62][63][64].

Analysis of FESEM Images
The fracture cross-section of pure epoxy is shown in Fig. 15 which reveals a perfectly smooth and glossy surface with no plastic deformation, indicating brittle fracture in pure epoxy.
The fracture cross-section of epoxy and phenolic resin (at levels of 10, 20, and 30 phr) is depicted in Fig. 15.As seen, an increment in the phenolic resin content increased parallel lines in cross-section, reflecting brittle fracture.The more the number and the deeper the depth of these parallel lines, the higher the brittleness of the facture as also proved by the mechanical features of the dumbbell-shaped samples.A rise in the phenolic resin content reduced all mechanical properties including toughness due to formation of bubbles in the adhesive matrix [7,8].
Toughening mechanisms such as crack deviation (red circles), crack tip blunting (orange circles), and shear band formation (blue circles) are marked in FESEM images.
Figure 16 shows epoxy samples with different contents of block copolymer (1.25, 2.5 and 3.75 phr).This figure shows cavitation and formation of spherical micelles within the BCP-modified epoxy sample.Also, the fracture crosssection changed from a smooth and glossy surface to a rough one, indicating the resistance of the matrix against deformation and increased toughness and plastic deformation of the epoxy matrix.The rougher fracture cross-section also suggests a rise in the fracture toughness energy which can be mainly assigned to the crack deviation due to the presence of spherical micelles in the epoxy matrix.The distribution of spherical micelles within the epoxy matrix led to a plastic deformation zone and a shear band.The BCP-induced cavities are also observable in the images as cavitation is one of the toughness mechanisms in the studied adhesive.BCP particles can form a shear band in the epoxy matrix, altering the stress rate.The 1.25C sample contains low levels of BCP, leading to less numbers of micelles.The higher BCP content in 2.5C sample increased the number of micelles.By further increase in the BCP content in the 3.75C sample, the size of the micelle structure also rose, declining the contact area of epoxy chains with the copolymer, thus, decreasing the mechanical properties [30][31][32].Figure 17 shows epoxy and zirconia nanoparticles at levels of 1, 2 and 5 phr.
Based on Fig. 17a-c, dispersion of nanoparticles can be observed in the epoxy matrix at the NP level of 1, 2, and 5 phr.The surface also got rougher, suggesting the resistance of the epoxy matrix against fracture and hence, the rise of toughness.This image also shows crack deviation, crack tip blunting, and formation of shear bands as the main toughening mechanisms [16][17][18].
Figure 18a depicts the fracture cross-section of the 2.5C-5Zr sample which indicates the toughness of epoxy at all additive levels.Toughening mechanisms such as crack deviation and the formation of shear bands can be observed in this image which can be assigned to the tough fracture as an indicative of higher fracture energy [27,28,65].The fracture cross-section of the 10P-2.5C-5Zrsample can be seen in Fig. 18c.Mechanisms of toughness at this level of the additives include shear band formation, crack deviation, and crack tip blunting.Surface roughness also rose compared to the pure epoxy, reflecting plastic deformation and an increase of the fracture energy and toughness [6,8,28].

Conclusions
This study presents new formulations of epoxy adhesive with improved thermal, mechanical, and adhesion properties by incorporating block copolymer, phenolic resin, and zirconia nanoparticles into DGEBA-based epoxy resin.
The incorporation of phenolic resin in dumbbell-shaped samples reduced the toughness, modulus, and strength.It also increased the adhesion in the single lap metal-metal joint samples.The highest shear strength was observed in the 10P-2.5C-5Zrsample, showing a 420% increase compared to pure epoxy.
The addition of zirconia NPs up to 5 phr incremented the toughness, modulus, and tensile strength compared to pure epoxy.The highest toughness was recorded in the 5Zr sample, showing a 175% increment compared to the pure epoxy.
Compared to pure epoxy, the T-peel strength showed a three-fold increment in the optimal 10P-2.5C-5Zrsample.
Temperature enhancement and decrease altered the butt joint tensile strength in the optimal quaternary sample.Compared to the room temperature, the butt joint tensile strength rose by 32%, 29% at 180, 120 °C, respectively while exhibiting a 19% reduction at − 60 °C.
The highest initial degradation temperature (230.2 °C) was observed in the 5Zr sample; while 2.5C specimen exhibited the highest maximum degradation temperature (407.5 °C).The 30P sample also showed the highest residual char (23%).

Fig. 1 a
Fig. 1 a Butt joint sample; b butt joint tensile strength at room temperature; c butt joint tensile strength at − 60 °C supplied by dry ice, and d butt joint tensile strength at 120 and 180 °C using a furnace

Fig. 2 aFig. 3
Fig. 2 a T-peel test specimens and b T-peel test modulus

Fig. 4
Fig. 4 Results of ternary and quaternary dumbbell-shaped samples a Tensile strength b Young modulus c Toughness

Fig. 5
Fig. 5 Predictive model for Young modulus of epoxy/zirconia NPs

Fig. 6 a
Fig. 6 a Results of single lap strength of epoxy samples at phenolic resin levels of 10, 20, and 30 phr. b Results of single lap strength of pure epoxy and copolymer at levels of 1.25, 2.5, and 3.75 phr.c Results of single lap shear strength test in the presence of zirconia

Fig. 9
Fig. 9 Tensile test results of pure epoxy and optimal quaternary dumbbell-shaped specimens at room temperature, − 60 and 120 °C a Tensile strength b Young modulus c Toughness

Fig. 10
Fig. 10 Results of single lap tensile strength test for pure epoxy and optimal quaternary sample at room temperature as well as − 60 and 120 °C Fig. 11 But joint test results for optimal quaternary specimen at room temperature, − 60, 120, and 180 °C

Figure 14
Figure 14 and Table4present the results of studying the cure kinetics in pure epoxy as well as 5Zr and 2.5C samples.The DSC diagram shows a peak for all three samples.This form of peaks is related to one-step reactions[24].Considering the shape of the peaks, it can be said the curing reaction of epoxy adhesives occurred in a single step and the type of reaction did not change with the addition of zirconia NPs and BCP.According to Table4, the initial

Fig. 14
Fig. 14 Diagram of heat flow vs. temperature for pure and optimal epoxy samples

Figure
Figure18billustrates the fracture cross-section of the 10P-5Zr specimen.The toughening mechanisms involve crack deviation, crack tip blunting, and formation of the shear band, suggesting a tough fracture in this specimen which coincides with the mechanical properties obtained from the tensile test results.The fracture cross-section of the 10P-2.5C-5Zrsample can be seen in Fig.18c.Mechanisms of toughness at this level of the additives include shear band formation, crack deviation, and crack tip blunting.Surface roughness also rose compared to the pure epoxy, reflecting plastic deformation and an increase of the fracture energy and toughness[6,8,28].

Fig. 15
Fig. 15 Fracture cross-section images of epoxy and epoxy/phenolic samples at a 200-micron scale (a pure epoxy, b 10P, c 20P, and d 30P)

Fig. 18 a
Fig. 17Fracture cross-section of epoxy/zirconia NPs samples at the scale of 200-micron a 1Zr, b 2Zr and c 5Zr

Table 1
Composition of the epoxy adhesive formulations used in this research

Table 3
Thermal stability indices of the pure epoxy sample as well as optimal ones

Table 4
Cure kinetics indices of pure epoxy and optimal epoxy/ copolymer and epoxy/zirconium oxide NPs samples