A Comprehensive Study on the Effect of ZnO and ZnAl2O4 Nanoparticles on the Mechanical Properties of Epoxy Coating: Tensile and Hardness

ZnO and ZnAl2O4 nanoparticles (NPs) were successfully prepared by the co-precipitation method and characterized by x-ray powder diffraction, transmission electron microscopy, and Fourier transform infrared spectroscopy. The prepared NPs were incorporated into epoxy (EP) coating with mass ratios 200 − 800 mg/kg of ZnO NPs/EP and ZnAl2O4 NPs /EP. The prepared coatings were characterized by scanning electron microscopy and Fourier transform infrared spectroscopy, and their mechanical properties were investigated, at room temperature, after 5, 10, 15, and 20 days of preparation. Tensile tests showed an improvement in the tensile properties, with the best improvement in ultimate tensile strength (93.2%) for 800 mg/kg ZnAl2O4 NPs/EP coating after 15 days of preparation. The ZnO NPs/EP and ZnAl2O4 NPs/EP coatings exhibited noticeable sensitivity to the stretching rate. Vickers microhardness (Hv) investigations showed normal indentation size effect behavior for all the samples. The best improvement in Hv was attained after 5 days of preparation, for all coatings, with the best improvement (9.15%) for 700 mg/kg ZnO NPs/EP.


Introduction
Industrial interest in epoxy (EP) originates from its vast properties, including high chemical resistance, simple processability, good adhesion to heterogeneous materials, and excellent electrical insulation (Ma et al., 2018;Samad et al., 2018). Such remarkable properties have made EP a material of choice in the marine, automotive, construction, and aircraft industries (Adesina et al., 2020;Yazman & Samancı, 2019). The curing of EP resin can be induced by heat or accelerated by adding curing agents (hardeners) (Baghdadi et al., 2020). However, when cured, EP resin tends to become very brittle with low impact resistance and toughness. This reduction in the mechanical properties of EP is due to the high cross-link density of the formed EP polymer (Baghdadi et al., 2020;M. Wu et al., 2020). Epoxies (EPs) are commonly chosen as organic coatings for protecting metals, mainly against corrosion (Samad et al., 2018;Wang et al., 2019;Yu et al., 2020). Nonetheless, the low mechanical properties of EPs make them fail on extended exposure to the environment (Samad et al., 2018).
Recently, several researchers have studied the effect of the incorporation of inorganic nanoparticles into EP resin in enhancing their tensile properties (e.g., Young's modulus, tensile strength, elongation, yield strength, yield point, and elastic limit (Rahman & Zhafer Firdaus Syed Putra, 2019)) and hardness. SiO 2 nanoparticles, used by Xiao et al. (Xiao et al., 2018) as fillers in EP resin, were effective in improving the tensile strength, flexural strength, and elastic modulus of the EP resin. Bazrgari et al. (Bazrgari et al., 2018) found that introducing Al 2 O 3 nanoparticles into EP resin has increased the bending strength and impact strength of EP. Moreover, Kumar et al. (Kumar et al., 2016) showed that TiO 2 nanoparticles' dispersion in EP resin has enhanced its tensile strength and failure energy. Furthermore, the introduction of AgO nanoparticles into EP could improve their mechanical Additionally, several research papers have discussed the effect of ZnO NPs on the mechanical properties of different coatings. ZnO NPs added into acrylic coating (Vu et al., 2021), polyurethane acrylate coating (Ariffin et al., 2020), and alkyd coating (Sturdy et al., 2020) showed an improvement in the mechanical properties of the coatings. To our knowledge, the effect of ZnAl 2 O 4 NPs on the mechanical properties of epoxy coating has not been previously studied. Therefore, this article will aim at studying the effect of ZnAl 2 O 4 NPs on the mechanical properties of EP coating. The high surface area of the ZnAl 2 O 4 NPs, used as fillers in the EP matrix, enhances EP mechanical properties through a more efficient load transfer mechanism. Moreover, a comparative study between ZnO and ZnAl 2 O 4 nanoparticles' addition and time effect on the mechanical properties of EP coating is conducted to find out the nanoparticles that provide the best mechanical properties.
In this work, we study the nanoparticles' addition and time effect on the mechanical properties of EP coating. ZnO NPs and ZnAl 2 O 4 NPs were synthesized via the co-precipitation method. The synthesized nanoparticles were characterized by X-ray powder diffraction (XRD), transmission electron microscopy (TEM), and Fourier transform infrared spectroscopy (FTIR). The prepared nanoparticles were incorporated into EP coating, with different mass ratios of 200, 300, 700, and 800 mg/kg of nanoparticles/coating. Moreover, the prepared coating samples were characterized by scanning electron microscopy (SEM) and FTIR. The effect of ZnO and ZnAl 2 O 4 nanoparticles on the tensile properties and the microhardness of EP coatings were investigated after 5, 10, 15, and 20 days of preparation. This work is organized in the present paper as follows: Sect. 2 presents the details of the preparation technique. Then, Sect. 3 illustrates the obtained results and their accompanying discussions. Finally, Sect. 4 summarizes the conclusions.

Synthesis and Characterization of Zinc Oxide and Zinc Aluminate Nanoparticles
ZnO and ZnAl 2 O 4 nanoparticles were prepared via the co-precipitation method. The detailed synthesis and the characterization of the ZnO NPs were reported in our previous work (Shmait et al., n.d.). As for the synthesis of the ZnAl 2 O 4 NPs, zinc chloride (ZnCl 2 , ≥98%, Sigma Aldrich), aluminum chloride anhydrous (AlCl 3 , ≥98%, Sigma Aldrich), and sodium hydroxide (NaOH, ≥98%, Alpha Chemika) were used as precursors. Firstly, aqueous solutions of 1 M ZnCl 2 and 1 M AlCl 3 were prepared, then mixed and stirred to obtain a uniform solution. Afterward, 1 M NaOH solution was added dropwise, while stirring, until the pH of the mixture became 12. The resulting mixture was then heated at 80°C for 6 hours with continuous stirring. After cooling, filtration was performed and the solid was washed with distilled water until the pH of the filtrate reached 7. Thereafter, the wet solid was dried in a box furnace at 100 °C for 18 hours. Eventually, the dried solid was ground and calcined at 550°C for 4 hours. The XRD pattern of the synthesized ZnAl 2 O 4 NPs was obtained using Bruker D8 Focus powder diffractometer with Cu-K α radiation (λ=1.54056 Å), in the range of 20° ≤ 2θ ≤ 80°, to confirm the phase formation and the purity of the prepared ZnAl 2 O 4 NPs. TEM was implemented to study morphology and the particle size of the ZnAl 2 O 4 NPs using JEM 100CX operated at 80 kV. FTIR analysis was carried out to identify the chemical bonds and the composition of the tested samples by Nicolet iS5 FTIR spectrometer in the spectral region between 4000 and 400 cm -1 , at room temperature.

Preparation of Epoxy Coatings for Tensile and Microhardness Tests
The EP resin (Epichlorohydrin + Bisphenol A) and its hardener (Cycloaliphatic polyamine), shown in Figures 1 (a) and (b) (Nikolic et al., 2010;Wetzel et al., 2003), respectively, were provided by Tinol Industries -Lebanon. For all EP coatings preparations, the mass ratio of the EP resin to its hardener was 2:1. The neat EP coating was prepared by stirring EP resin with its hardener for 30 minutes. To prepare EP coatings containing ZnO NPs (ZnO NPs/EP), the required mass of the ZnO NPs was stirred with the EP resin for 10 minutes. This was followed by adding the hardener and all components were stirred for 20 minutes. Different ZnO NPs/EP coatings were prepared with concentrations of nanoparticles/(EP+hardener) of 200, 300, 700, and 800 mg/kg. The same procedure was followed in preparing EP coatings containing ZnAl 2 O 4 NPs (ZnAl 2 O 4 NPs /EP) with the same concentrations. The prepared EP coatings, with and without nanoparticles, were poured into molds, and left for 5, 10, 15, and 20 days, to form films to perform the tensile and microhardness tests. The mean thickness of each of the dry films was 500 ± 50 µm. The surface morphology of the coating samples was characterized by a JEOL scanning electron microscope (JSM-5300). FTIR analysis of the coatings was carried out by Nicolet iS5 FTIR spectrometer in the spectral region between 4 000 and 400 cm -1 , at room temperature. Tensile tests were performed, using ESM303 Tensile Tester (Mark-10), for films cut into dog bone specimens having 0.9 cm width, 8 cm gauge length, and 14 cm overall length. Microhardness measurements were performed using a digital Vickers microhardness tester (SIOMM model, MHVD-1000IS). Moreover, the absence of extra peaks in the ZnO NPs pattern ensures the high purity of the prepared nanoparticles, which is confirmed by implementing a refinement to the XRD pattern, performed on the MAUD program and shown in Figure 2 (c). On the other hand, the formation of face-centered cubic spinel structure of ZnAl 2 O 4 is confirmed by the presence of the diffraction peaks located at 31. 26°, 36.97°, 44.97°, 49.2°, 55.92°, 59.43°, 65.37°, 74.2°, and 77.5°, which correspond to the planes of (220), (311), (400), (331), (422), (511), (440), (620), and (533), respectively, as shown in Figure 2 (b) (Xing et al., 2020). However, the appearance of additional peaks in the ZnAl 2 O 4 XRD pattern indicates the presence of an impurity phase, mainly attributed to the formation of zinc oxide (Xing et al., 2020). This is verified by the refinement to the XRD pattern, as seen in Figure 2 (d), where 88.2 and 11.8% of ZnAl 2 O 4 and ZnO phases, respectively, were obtained. The weighted profile R-factor (R wp ), expected R-factor (R exp ), and the goodness of the fit (χ 2 ) are shown in Figures 2 (c) and (d). The lattice parameters, obtained from the Rietveld refinement (Table 1), are found agreeable with that reported in the previous literature (a = 3.25 and c = 5.21 Å for ZnO NPs, a = 8.09 Å for ZnAl 2 O 4 NPs (Tangcharoen et al., 2019;Zak et al., 2011)). Moreover, the lattice parameters of the prepared ZnO nanoparticles are found larger than that of their bulk counterpart (a = 3.12 and c = 5.05 Å (Gallegos et al., 2019)), demonstrating that the lattice parameters have increased in the nanoscale (Fukuhara, 2003), subsequently, inducing microstrain in the crystalline lattice. Microstrains, which are very common in nanoparticles, are mainly caused by the high surface energy of the nanoparticles. A large fraction of the surface atoms in materials at the nanoscale are under-coordinated and present a bond order deficiency, as compared to their bulk equivalent form (Alsaad et al., 2020;Andrade et al., 2017). The obtained microstrain (ε) values for ZnO and ZnAl 2 O 4 nanoparticles are listed in Table 1. The microstrain value of the synthesized ZnO NPs is quite similar to that reported by Almeida et al. for ZnO NPs, prepared by the sol-gel method (Lopes de Almeida et al., 2021). However, a lower microstrain value for ZnAl 2 O 4 NPs is obtained in this work, as compared to that reported by El-Fadl et al. (El-Fadl et al., 2019), for ZnAl 2 O 4 NPs prepared via microwave combustion method. A lower microstrain reflects a reduction in lattice defect along the grain boundaries (Shaban et al., 2020).  where 0.9 is Scherrer's constant, λ is the X-ray wavelength, β is the full width at half maximum height (in radians), and 2θ is the diffraction angle. The crystallite size of the ZnO and ZnAl 2 O 4 nanoparticles was found to be 23.84 nm and 6.61 nm, respectively.

Nanoparticles Characterization
The x-ray density ( ), the dislocation density (δ), and the specific surface area (S) were calculated using the following equations (El-Fadl et al., 2019;Kamareddine et al., 2020): where n, M, N A , and V refer to the number of atoms per unit cell (4 for ZnO (Galsin, 2019) and 8 for ZnAl 2 O 4 (El-Fadl et al., 2019)), molecular weight, Avogadro's number, and the unit cell volume. The results, listed in Table 1, show that the specific surface area of ZnAl 2 O 4 NPs is greater than that of ZnO NPs. This helps in many applications as improving the mechanical properties of EP coatings upon the incorporation of the ZnAl 2 O 4 NPs.  Figure 3 (a), shows slightly agglomerated, homogeneous particles with distorted hexagonal morphology. On the other hand, the TEM image of ZnAl 2 O 4 NPs shows agglomerated nanoparticles exhibiting homogeneous, nearly spherical morphology, as seen in Figure 3 (c). Moreover, the histograms displayed in Figures 3 (b) and (d), obtained using the ImageJ software, show that the maximum number of particles have a particle size of 49.09 and 7.73 nm for ZnO NPs and ZnAl 2 O 4 NPs, respectively. The high surface area of the nanoparticles and the strong attractive forces between the nanoparticles lead to their aggregation (Ashraf et al., 2018). The final particle of ZnO has reached a larger size as compared to the crystallite size obtained from XRD, which confirms that the ZnO particles were formed from the combination of crystallites (Jensen et al., 2006;Kibasomba et al., 2018). The particle size of ZnO NPs larger than crystallite size was reported by Vijayakumar et al. (Vijayakumar et al., 2020) for ZnO NPs prepared by the co-precipitation method. Nonetheless, the particle size of ZnAl 2 O 4 is in good agreement with the crystallite size obtained from the XRD analysis. Likewise, the particle size agrees with the crystallite size for ZnAl 2 O 4 prepared by Jagadeeshwaran et al. (Jagadeeshwaran & Murugaraj, 2019) by the co-precipitation method.  Table  2. The strong band appearing at 431.13 cm −1 in Figure 4 (a) corresponds to Zn-O stretching vibration mode confirming the successful preparation of ZnO NPs. As seen in Figure 4 (c), a broad absorption band appears at 3 446.80 cm −1 , representing the O-H stretching vibrations. Also, a weak band appears at 1 635.82 cm -1 , corresponding to H-O-H bending vibrations (Naik et al., 2020). These two bands indicate the small water content in the ZnAl 2 O 4 NPs samples (Hassan et al., 2020;Subhasree et al., 2020). There is no peak in the range 1 028 -1 157 cm -1 , which confirms the absence of Al-OH bending vibrations (Bayal & Jeevanandam, 2012). Moreover, the three peaks observed at 677.90, 554.39, and 501.46 cm -1 indicate the formation of the spinel-type structure (Guo et al., 2017). Similar results were obtained with Belyaev et al. (Belyaev et al., 2019). The spinel structures exhibit stretching bands in the 500-900 cm -1 range, which are associated with metal-oxygen, aluminum-oxygen, and metal-oxygen-aluminum stretching vibrations (Sameera et al., 2019;Yuvasravana et al., 2017).    Figure 5 (a-e). Figure 5 (a) shows that neat EP coating presents a smooth surface. However, a rough surface is observed in EP coatings incorporated with ZnO or ZnAl 2 O 4 nanoparticles, as seen in Figures 5 (b) and (c) and Figures 5 (d) and (e), respectively. Surface roughness increased with the increase in the concentration of ZnO and ZnAl 2 O 4 nanoparticles in EP. Moreover, the micrographs depicted in Figure 5 (b-e), show a nearly uniform distribution of ZnO and ZnAl 2 O 4 nanoparticles in the EP matrix. Nonetheless, some agglomeration appears in the 700 mg/kg ZnO NPs/EP coating. Additionally, some nanoparticles-free spaces are observed in the ZnO NPs/EP and ZnAl 2 O 4 NPs/EP samples, as shown in Figure 5 (b-e).

Tensile Tests
Studying the mechanical properties of materials is necessary since they help us understand the materials' behavior. As such, the tensile properties of EP coatings, with and without ZnO or ZnAl 2 O 4 nanoparticles, were investigated at room temperature, under stretching rates of 1 and 3 mm/min, after 10, 15, and 20 days of preparation.  Figure 7 (a) inset, an increase in the UTS of EP coating upon ZnO NPs incorporation was 20.7% with 200 mg/kg ZnO NPs/EP coating, as compared to neat EP. After that, an increase in the concentration of ZnO NPs in EP to 300 and 700 mg/kg has increased the UTS by 45.9 and 48.5%, respectively, as compared to neat EP. This increase in UTS may be attributed to the following mechanism: the fine particles of ZnO were evenly distributed in the EP resin, which makes the load transfer between the nanoparticles and the EP resin equal (Bazrgari et al., 2018;Kameswara Reddy et al., 2020). Additionally, the strong interfacial bonds between the ZnO NPs and the EP resin enhance the mechanical properties by providing an effective load transfer to the nanoparticles, which are more rigid in the EP resin (Ma et al., 2018;Yazman & Samancı, 2019 (Frigione & Lettieri, 2020). The results, presented in Figure 7 (a), are in good agreement with the SEM micrograph analysis for EP coatings, in which some agglomeration was noticed in 700 mg/kg ZnO NPs/EP coating. On the other hand, the UTS of neat EP coating was 28.4 MPa, on day 15, as seen in Figure 7 ( This might be attributed to the smaller particle size of the ZnAl 2 O 4 NPs, as compared to ZnO NPs, which enhances its ability to penetrate ultra-small holes, indentations, and capillary area in the EP matrix (Nguyen et al., 2016). ZnAl 2 O 4 NPs have a higher total surface area, as confirmed in the XRD analysis, which causes a more efficient transfer of stress between the nanoparticles and the EP polymer, hence, giving better reinforcement (Fu et al., 2008;Ma et al., 2018). Additionally, an enhancement in Young's modulus and the elongation-at-break with ZnO NPs/EP and ZnAl 2 O 4 NPs/EP coating, as compared to neat EP, was observed, as shown in Figures 8 (a) mas.ccsenet.org Vol. 15, No. 5; and (b). Young's modulus, determined from the slope of the linear segment in the stress-strain curve, is a measure of the stiffness of a material (Ünal et al., 2016), and the elongation-at-break represents the ability of a material to resist changes of shape without cracking (Djafari Petroudy, 2017). An increase in Young's modulus and the elongation-at-break, upon the incorporation of ZnO or ZnAl 2 O 4 nanoparticles into EP resin, has ensured an improvement in the stiffness and the toughness of the coating . Young's modulus and the elongation-at-break of neat EP coating, on day 15, were 1.26 GPa and 2.3%, respectively, which increased to 2.038 GPa and 2.9% with 800 mg/kg ZnAl 2 O 4 NPs/EP specimen. The 200 mg/kg ZnAl 2 O 4 NPs/EP specimen, at failure, is presented in Figure 9.   & Bandyopadhyay, 2011;Savvilotidou et al., 2017). Water absorption, even at low amounts, weakens the interface between the nanoparticles and the EP matrix, which obstructs stress transfer under mechanical loads (Frigione & Lettieri, 2020). Similar results were obtained with Pattanaik et al. (Pattanaik et al., 2020), in which the tensile strength of fly ash-filled EP composites has decreased under a natural humid atmospheric environment. The effect of the stretching rate on the UTS of neat EP, ZnO NPs/EP (300 and 700 mg/kg), and ZnAl 2 O 4 NPs/EP (200 and 300 mg/kg) specimens, was studied for day 15 samples. The results are shown in Figures 11 (a) and (b). As noticed, when the stretching rate was decreased from 3 to 1 mm/min, the UTS of all specimens decreased. Similar results were obtained with Gurusideswar et al. (Gurusideswar et al., 2017) Vol. 15, No. 5;2021 57 decrease in the stretching rate. An increase in the stretching rate decreases the molecular mobility of the EP polymer chains, which leads to an increase in the stiffness and brittleness of the material (Gurusideswar et al., 2017).

Vickers Microhardness
At room temperature, the microhardness of EP coating samples, with and without ZnO or ZnAl 2 O 4 nanoparticles, was measured by applying Vickers pyramidal indenter after 5, 10, 15, and 20 days of preparation. Different loads varying from 0.245 to 4.90 N, at a dwell loading time of 30 seconds, were applied. For each load, three different hits at different positions on the sample's surface were applied and Vickers microhardness value was determined using the given equation (Srour et al., 2017;Zalaoglu et al., 2017): where H v is Vickers microhardness in (GPa), F is the applied load in (N), and d is the average of the diagonals of the indentation marks of the three hits in (µm). The variation of H v with F, after 5, 10, 15, and 20 days of preparation, for neat EP, ZnO NPs/EP, and ZnAl 2 O 4 NPs/EP (200, 300, 700 and 800 mg/kg) coating samples, is displayed in Figure 12 (a-d) and Figure 13 (a-d).   Vol. 15, No. 5; et al., 2014;Sangwal, 2000). The other two cases, n > 2 and n = 2, refer to reverse indentation size effect (RISE) behavior and H v independent of F, respectively (Awad et al., 2014;Sangwal, 2000). Figure 12 (a-d) and Figure 13 ( Hardness is a parameter that expresses a material's resistance to deformation against a force concentrated on its surface (Bar, 2020). Thus, the addition of ZnO and ZnAl 2 O 4 nanoparticles was effective in improving the deformation resistance of EP coating. This may be attributed to the uniform dispersion of the ZnO and ZnAl 2 O 4 nanoparticles in the EP matrix, which limits the molecular chain mobility (Dass et al., 2017;Ozcan et al., 2019;Salahuddin et al., 2017). The high surface area of the nanoparticles reinforces a greater volume of the EP matrix and the load can be transferred to the fillers more efficiently due to the high interfacial area between the EP resin and the fillers. Hence, the material's resistance to deformation caused by the load increases (Dass et al., 2017;Ozcan et al., 2019;Salahuddin et al., 2017). A further increase in the concentration of ZnO NPs in EP to 700 mg/kg ZnO NPs/EP has further increased the H v of EP coating. Nonetheless, a higher filler content ratio of 800 mg/kg ZnO NPs/EP showed a decrease in the H v of the coating, as compared to neat EP. This might be due to the particles' agglomeration, which initiates cracks at the interfaces between the EP matrix and the nanofiller particles (Bisht et al., 2018;Dass et al., 2017 Vol. 15, No. 5;