Abstract
In this research, the interfacial interaction between nanosilica particles increased by using emulsifiers and treated by polyethylene oxide non‐ionic surfactant. Several resins were prepared via emulsion polymerisation and were converted into water based paints. The silica content and mole fraction of acrylic monomers were altered into nanocomposite emulsions to improve the adhesion and flammability of final resins. The dynamic light scattering, transmission electron microscopy and scanning electron microscopy were used to evaluate the size distribution and morphology of silica particles. The flammability was decreased by increasing silica content of nanocomposite resins. The optimum adhesion of paints on all surfaces was on 0·4 methyl methacrylate mole fraction. The adhesion behaviour of paints on glass, ceramic and steel surface was not similar to each other by changing the silica contents and stabling copolymers mole fractions. The highest adhesion strength was on a ceramic surface with 60 g silica content.
Introduction
The properties of nanocomposites are greatly influenced by both the dispersing of nanoparticles in the base polymers and the interfacial adhesion between the inorganic and organic components. Although nanocomposites can be prepared by simply blending the nanoparticles with base polymers by high shear stirring or ball milling, the dispersing degree of nanoparticles and the interfacial adhesion were insufficient for obtaining desirable properties.1 Recent nanocomposites are useful in a wide range of industrial applications such as cosmetics, plastics, rubbers, binders ink, paints, biomedicals, and so forth,2 but the effect of silica and copolymer mole fractions on latexes has not been reported in detail. No practical usage with industrial silica particles has been reported for nanocomposite paints. 2 , 3 One approach was the treatment of silica with silyl coupling agents4 which were applied by using a proper silyl coupling agent to improve the encapsulation efficiency of copolymers in water on the surface of silica particles. Nanosilica particles are flame retardants and burning characteristics should be considered associated with the gas barrier properties of nanolayers which impede gas diffusion and the retarding combustion nature of the silicate layers.5
In general, nanocomposites are formed via the polymerisation of monomers in the presence of nanoinorganic particles by one of the suspension,6 dispersion, 7 , 8 emulsion 3 , 9 and miniemulsion10 processes. The influence of the silica content and growth condition on the microstructure, adhesion strength and mechanical properties of Ti–Si–N nanocomposite films was studied in a sol–gel method by using organic dispersant.11 However, the preparation of organic–inorganic nanohybrids in the form of emulsion is more preferable because of its easy processability and environmental friendly purposes.
The use of ionic surfactants with the opposite electrical charge to the silica particles has been previously reported. The absorbed surfactant on the particles surface is believed to improve the affinity to the organic monomers. 2 , 12 In another study, UV curable epoxy/silica nanocomposites were prepared and characterised by proper adhesion strength;13 however, they were not applicable as nanocomposite coatings and thermal stability of nanocomposites were not enhanced with the addition of silica particles. In another article, the influence of matrix polarity and the presence of reactive functional groups on polymer chain on the dispersion of nanoparticles, the polymer/filler interfacial adhesion and the mechanical properties of the nanocomposites was investigated,14 but the adhesion and flammability role of silica particles and copolymers was not determined. The results by Zhu et al. revealed that filling grafted–silica hybrid nanoparticles into poly styrene butyl acrylate–acrylic acid latex resulted in significant improvement in interfacial adhesion properties, UV shielding, water resistance and mechanical properties.15 However, as it will be stated later, nanosilica was directly added to water to prepare stable dispersions. But the problem is that because of a large surface area, silica aggregates embedded in a polymer matrix were formed easily and led to no anticipative enhancement in mechanical properties.16 However, the properties of latexes were improved by using proper surfactants. Therefore, the surface modification of nanofillers is necessary for the preparation of nanocomposites. Nanoparticle/matrix interface adhesion, particle loading, the particle size and the particle content are important factors that affect adhesion strength and the mechanical properties. Therefore, the use of surfactant coupling agents that increase the nanoparticle–matrix adhesion leads to higher adhesion strength,17 so the adhesion of paints to substrates will improve properly.
Polymer layered inorganic nanocomposites have been extensively studied because of their excellent properties. These improved properties may be due to the synergistic effects of the nanometer dimensional dispersion of polymers and inorganic components.18 Layered silicates and graphite oxide systems18 are typically two‐dimensional solids in the bulk form composed of layered nanosheets possessing hydroxyl group to reduce flammability of nanocomposites. Flammability characteristics were affected by the silica layers which generated flame retardation. Mizutani et al.2 have developed the preparation of emulsion water based paints, and improved the flammability resistance of paint by using 50% silica dispersion, but they had limitation in preparation and extending the range of copolymers ratio. They reported that some cracks were formed in most mole fractions and it happened due to the high silica content.
In the present study, the flammability resistance was improved by using less silica in emulsions, and prepared a wide range of monomer ratios with no cracks by reducing the silica content, changing surfactants and preparation methods. The preparation of emulsions was investigated and used to make the latexes and studied the adhesion strength to different surfaces and flammability by changing monomer ratios and silica contents. Finally, the improved optimal and stable conditions were declared. The application area of Tg for decorative paints was determined between 10 and 40°C (Ref. 19) and it was considered to adjust the latexes in this range.
Material and methods
Materials
The nanosilica was Aerosil 200 whose particle diameter was reported 12–16 nm by Degussa. Triton X‐114 (70 wt‐%) was used as non‐ionic surfactant which was a clear oily liquid with an average molecular weight of 558 g (Across Co.) as a poly ethylene oxide surfactant); its cloud point was reported to be 43°C. Poly ethylene oxide 200 (PEO‐Merck) was used as another reactive non‐ionic surfactant. The anionic surfactant was decyl‐benzene sulphonic acid sodium salt (DBS‐Across). The monomers supplied were selected from industrial grade of methyl methacrylate (MMA), n‐butyl acrylate (BA) and methacrylic acid (MAA) and ammonium peroxy disulphate (APS, Merck) was the initiator of the reactions.
Preparation of emulsions, paints
Ten different emulsions, which had been subdivided into two groups, were prepared; the first group consisted of P1, P2, P3, P4 and P5 emulsions (the same compositions of these resin emulsions are shown in Table 1). The different monomer mole fractions and modified compositions of these latexes are shown in Table 2. The second group of emulsions consisted of P6, P7, P8, P9 and P10 which had the same monomer mole fractions (Table 3) and different nanosilica contents. The detailed composition of paints is shown in Table 4 Tables 4 and 5 and the silica content and total weight of emulsions and latexes are shown in Table 6.
Composition of resin emulsions in P1, P2, P3, P4 and P5 latexes
Components of resin emulsions of P1, P2, P3, P4 and P5 latexes by changing copolymers mole fractions
Compositions of resin emulsions of P6, P7, P8, P9 and P10 samples
Content of assistant solvent, coalescing agent, thickener and antibacterial agents in all latexes
Composition of pigment dispersion of all latexes
Silica contents, emulsion resin contents and total weight of emulsion latexes
The preparation of P3 emulsion was as follows: a monomer mixture consisting of 38·6 g MMA, 58 g n‐BA and 0·8 g MAA prepared by slow mixing, a 1 L flask equipped with a condenser, a dropping funnel and a mechanical stirrer. It was mixed nanosilica by adding 60 g aerosil 200 into 345 g water and continued mixing it mechanically until a stable colloid aqueous was carefully prepared. To dissolve nanosilica in water without flocculation, 1·5 g poly ethylene oxide 200 (PEO200) was added to water to improve the dissolution. Then, the nanosilica was added slowly to water and stirred by a mechanical stirrer. The suspension was vibrated carefully by an ultrasonic vibrator at 55°C for 3 h. Experience proved that an ultrasonic vibration time between 2·5 and 3 h was suitable for this nanosilica particles dispersion. Then, the suspension was stirred mechanically for 10–20 min to improve the tixotropic properties (stable nanosilica dispersion). The suspension was charged into a flask reactor, to start copolymerisation. A solution of 0·4 g APS and 3·9 g Triton X‐114 in 100 g water was added dropwise to nanosilica sol which was in flask with vigorous stirring at 55±1°C in a controllable temperature water bath. Then, to solve silica sols by non‐ionic surfactant, the temperature of the solution was increased to 67±1°C (higher than the cloud point of non‐ionic surfactant) and 8 g of monomer mixture was added and the solution was mixed under a nitrogen atmosphere for 1 h. The solution's colour turned milky, and the viscosity of the reaction liquid increased quickly. Then, 0·4 g APS and 5·8 g DBS were solved in 45 g water and added to the solution. After 5 min, the remaining monomer mixture was added to start post‐polymerisation in one additional hour. The mixture viscosity reduced quickly and a homogenised liquid was formed. The emulsion was cooled down to room temperature and 1·9 g ammonia solution was added to stable the pH.
Table 1 Table 2 Tables 1–3 show the recipes for the preparation of all polymer silica nanocomposite emulsion resins. White outdoor paints with acceptable properties were prepared from emulsions. The detailed composition of paints is shown in Table 4 Table 5 Tables 4–6. The wetting and dispersing agents, defoamer and TiO2 white pigment were added to water and mixed gently. This suspension was vibrated carefully in an ultrasonic vibrator for 10 min to complete the wetting and dispersing of TiO2 pigments. The resin emulsions were added to the pigment dispersions and mixed for 10 min slowly.
Characterisation techniques
The morphology of silica particles and silica–acrylic nanocomposites were observed by a SEM PHILIPS XL30. The samples were coated with gold using the sputtering technique. The transmission electron microscopy (TEM) images were carried out by a Carl Zeiss CEM902A with an in‐column filter at 120 kV. The measurements of particle size and size distribution of silica sol and nanocomposites were performed by dynamic light scattering (DLS, Malvern Zetasizer ZS, Malvern, UK) on a particle size analyser. The adhesion of paint films on different surfaces were carried out by cross‐cut tests by a CC 1000 device (TQC Co., Zevenhuizen, The Netherlands) according to ASTM D3359‐02 standard (with six grids). This test method is the most applicative method in paint producing plants to test on surfaces. The weakest adhesion level was 0B and the highest level was 5B and other levels were changing between these two levels. If none of the paint samples were removed by the test tape, the adhesion level will be 5B, less than 5% is 4B level, between 5 and 15% is 3B, between 15 and 35% is 2B, between 35 and 65% is 1B, and less than 65% removal shows 0B as the weakest level of adhesion. If a result was observed between two levels, it was as a result of two test results by repeating one test. The flame resistance of paint films was tested by burning the paint surfaces on steel by a handy burner with the flame temperature of 1250±50°C on the edge of flame from 30 cm distance for 3 min with a contact angle of 90° between steel plates and flame direction.
Theory/calculation
All surfactant molecules contain two different parts, one insoluble in specific fluid (the lypophilic part) and another soluble (the lypophobic) part.20 When the fluid is water, the hydrophilic part is the head and the hydrophobic part is the tail. The head of anionic surfactant (DBS) is anionic sulphonic acid sodium salt substituted on aromatic ring, and the tail is an aliphatic carbon chain.
Surfactant adsorption at hydrophilic solid surfaces is often pictured as leading to mono and bilayers, but it has become increasingly clear that this process is also best regarded as a process of surfactant self‐assembly. In such cases, the picture of continuous surfactant layers must be as illustrated in Fig. 1.20
Self‐assembly mechanism of surfactant into discrete micelles (at surface of hydrophilic nanosilica) by bilayer array
To select monomer ratios in copolymerisation and adjust the Tg of resulted copolymer to the range of decorative resins, the following formula was used21
,
are Tg of MMA and n‐BA monomers, and W1 and W2 are weight fractions of these two monomers. The formula can be changed as follows to calculate the Tg(Co) of the final copolymer by considering the monomer mole fractions
These emulsions had different Tg as different monomer ratios and silica contents. After that, the results of adhesion, Tg and flammability of samples were compared when they were converted to the paints according to recipes in Table 1 Table 2 Table 3 Table 4 Table 5 Tables 1–6.
Coating
Soft steel, building ceramic tiles and glass sheet were the surfaces which were used for coating by emulsion paints. All test surfaces were washed with the mixture of water and detergents before coating. Paints were coated on the panels by using a Baker industrial film applicator (TQC Co.) to achieve at the same thickness for all samples (120 μm) according to ASTM D 82353‐FMTS standard method. Finally, the paint films were dried at 25°C for one week under 55±5% of relative humidity (to simulate the dry condition of outdoor paints).
Results and discussion
The ultrafine colloid silica used in this study has a size of 12–16 nm. These particles were highly hydrophilic because the partial ionisation of terminal hydroxysilyl groups to silicate anions creates these properties. These particles were aggregated by each other to form bigger particles with the size of 152 nm. The homogeneous and stable dispersion of such tiny inorganic particles in an organic matrix is always a challenging task to the material scientists. Because of the large surface area and attractive interaction between the nanoparticle layers, strong adhesion at the polymer/filler interface is necessary to obtain a homogenous dispersion of the nanoparticles in the polymer and improvement in mechanical properties with a small filler concentration. The differences in the dispersion of the nanoparticles and the interfacial adhesion directly influence the mechanical properties of the nanocomposites.14 At the prepolymerisation step, hydroxysilyl groups covered nanosilica particles, by the acrylic copolymer mixture. Therefore, the radical polymerisation was directed on the surface of nanoparticles by adding the non‐ionic surfactant (with a low cloud point) with a bilayer self‐assembly array around nanosilica particles. After modifying the surface of colloid silica, the turbidity of the mixture slightly increased. The prepolymerisation step was performed by adding a small portion of monomer mixtures to form a thin acrylic layer around nanosilica particles. The dissolution of nanocomposites occurred by adding anionic surfactant to form a new anionic thin layer. Finally, the post‐polymerisation of acrylic monomers was followed by adding the remaining of monomer mixture to form a complete acrylic layer around nanoparticles and make the final resins. Compared with other researches,2 we got to resins by decreasing the silica content and applying bigger nanosilica particles and well dispersion of these particles in water to the well adsorption of copolymers on nanosilica surfaces to study the adhesion and flame resistance of samples. The monomer mole ratios were extended similar to prepared emulsions, but in wider ranges, it was not possible by this preparation technique because the emulsions flocculated and resulted paints were destroyed. The extended monomer ratios were achieved to 25 wt‐% MMA by reducing the silica content and dispersion of silica particles was improved in water and a new mixture of non‐ionic surfactants was used.
Size distribution of silica, and acrylic–silica nanocomposites
The DLS charts of silica sols are shown in Fig. 2. The average diameter of all nanosilica particles (100% intensity) was 152 nm (Fig. 2a). In Fig. 2b, the DLS chart of P3 emulsion particles after post‐polymerisation is shown. The average diameter of final hybrid particles was 250 nm (87·5% intensity of this size of particles). This increase in the particle size confirms the formation of final hybrid particles and the reduction of intensity shows that 12·5% of particles did not completely convert to the final composite particles. The morphology and particle size studies indicated that silica particles were dispersed homogenously through the matrix by using proper surfactants and improving the dispersion of silica in water.
Size distribution of a silica dispersion and b nanocomposite emulsion
Morphology of silica particles and acrylic–silica nanocomposites
Figure 3a shows the scanning electron microscopy (SEM) image of nanosilica particles before the dissolution in water and vibration. Many of particles were aggregated due to the fact that silica particles are hydrophilic. Figure 3b shows the SEM image of silica–acrylic nanocomposite with lots of holes in the surface. Figure 4a shows the typical TEM image of colloid silica particles in water after 2·5 h of ultrasonic vibration of P3 emulsion. In Fig. 4b, the TEM image of silica–acrylic particles was displayed in which the size of particles increased to 250 nm. The core–shell type particles aggregated during the drying time for the SEM analyses. Therefore, with comparing the TEM and the SEM images, the particles built a nanocomposite structural array because of the encapsulation of acrylic polymers around silica cores. It is an important and principal rule for the preparation of nanocomposite solutions.
Image (SEM) of a silica solid particles before dissolution in water and b P3 nanocomposite surface after drying
Image (TEM) of a silica dispersed in water and b silica–acrylic nanocomposite particles after radical polymerisation
Evaluation of prepared nanocomposite emulsion latexes
Adhesion and Tg results
The adhesion result of paint films were influenced by Tg, copolymers, physical and chemical properties of substrates and silica content of paints. The application area of Tg for decorative paints is between 10 and 40°C.19 In Fig. 5, the theoretical and practical Tg of the prepared samples (by changing n‐BA mole fractions) is shown and most of measured practical Tg values were in this range. The authors used different substrates to determine the adhesion behaviour of latexes. In these samples, the silica contents were stable (60 g), but monomer ratios were altered to detect the effect of copolymers on the Tg of latexes. The reason for higher Tg of measured practical values in comparison with the theoretical values was as a result of the presence of nanosilica in samples. Inorganic fillers usually increase the Tg of paint films. The increase in the silica content in paints with the same chemical content reduces the chemical adhesion strength to substrates because of the reduction of resin content. The increase in the MMA ratio in copolymers caused higher Tg values of paints. As the MMA mole fraction increased, Tg of copolymers and adhesion strength to polar and metal substrates (such as polyvinyl chloride substrate) increases, too. The increase in BA in copolymers reduces the Tg and increases the adhesion strength to non‐polar and plastic substrates. These factors affected the results of adhesion strength and flammability resistance. The adhesion of paint samples to glass, ceramic and steel plates was measured by changing MMA monomer ratios ( Figure 6 Figure 7 Figs. 6a, 7a and 8a) and the silica content ( Figure 6 Figure 7 Figs. 6b, 7b and 8b) and now discuss these results. The adhesion changes on glass, ceramic and steel (by changing silica contents and monomer mole fractions) showed that the adhesion behaviours of paints on different substrates were not similar to each other, because the physical properties of the substrates were different.
Effect of n‐BA mole fraction on modelled and measured Tg of paint films
Effect of MMA a mole fraction and b silica content on adhesion to glass sheets
Effect of a MMA mole fraction and b silica content on adhesion to ceramic tiles
Effect of a MMA mole fraction and b silica content on adhesion to steel plates
Adhesion to glass
The results of adhesion to glass plates were higher than those of steel plates but weaker than those of ceramic tiles. The highest adhesion was in 0·4 mole fraction of MMA monomers as can be seen in Fig. 6a. The results of adhesion to glass by changing silica contents are shown in Fig. 6b. The increase in silica reduced the chemical adhesion because of the reduction in the resin content of paint. But, adding 30 g silica did not have a considerable effect on adhesion. The highest adhesion to glass was in 0–30 g silica content by 0·4 mole fraction of MMA. The presence of silica particles improved the adhesion to glass more than that of steel plates and this was because of the chemical tendency of silica to glass, but the smoothness of glass prevents the physical penetration of paints and reduces the adhesion strength.
Adhesion to ceramic
A comparison of the adhesion strength of paints with glass, steel and ceramic tiles showed that the adhesion strength of ceramic was in the highest level. The highest adhesion was in 0·4 mole fraction of MMA (Fig. 7a). By changing the silica content, the highest adhesions to ceramic tiles were in the silica free samples in which, the adhesion strength was high up to 90 g silica content (Fig. 7b). It was because of the porous surface of ceramic which was more than glass and steel sheets. It shows a physical mechanism for high adhesion which was improved by silica particles, but the smoothness of glass and steel plates physically prevented the penetration of liquid paints to these surfaces. The adhesion level of these paints could be improved to the higher levels after the use of an acryl urethane primer.2
Adhesion to steel plate
All the results of adhesion to steel plates by changing monomer ratios were weak and close to each other, but the adhesion strength with 0·4–0·45 MMA mole fractions was stronger than other samples (Fig. 8a). The highest adhesion strength to steel by changing silica content was reported for the sample by 90 g silica content (Fig. 8b). By considering all of these results, it is not suitable to use these paints for soft steel plates because of the weak adhesion. Although the adhesion results for steel were weak, decreasing the smoothness of steel plate by using a soft abrasive paper will result in a higher adhesion level to 3B.
Flammability of paints
Thermal studies, under non‐isothermal crystallisation, indicated the lack of influence of silica on the thermal behaviour. Flammability characteristics were however affected by the silica layers which overall generated flame retardation in the complex nanocomposites.5 Recent studies 22 , 23 show that the lower flammability of nanocomposites is due to the formation of a multilayered carbon aqueous silicate structure in the condensed phase. This highperformance carbon aqueous silicate char builds up on the surface of burning polymer, insulates the underlying polymeric substrate, slows down heat and mass transfer between the gaseous and condensed phases, and thus retards the thermo‐oxidative degradation of the polymer.5
The flame resistances of all samples were compared by a 1250°C handy burner from the distance of 30 cm for 3 min and only the surfaces of P4 and P5 did not burn successfully. It means that after finishing the burning time, these two surfaces did not turn to brown. A few dark points were observed on the surfaces of P3, P6, P7, P8, P9 and P10. The surfaces of P1 and P2 completely turned to brown after burning. Therefore, the weak flame resistance of P1 and P2 was because of their low silica contents and high organic contents. With burning the surfaces, a silica layer was formed on the paint film surfaces by the condensation between the silica particles and burning of organic content, which acts as the insulating material to prevent further thermal decomposition.2 The silica particles tend to accumulate and coagulate near the paint surface, forming loose, granular particles instead of a tight continuous silica network. Therefore, adding nanosilica particles to the acrylic resin during the gasification leads to forming silica network which can enhance the formation of cross‐links among the particles, provided that appropriate surface treatments of silica particles are carried out.16 The flammability was affected by the silica layers which generated flame retardation in the nanocomposite.24
Conclusions
In conclusion, the good dispersion of silica particles in matrix brings in improved interfacial adhesion between the inorganic filler and the polymeric matrix and thus results in the enhanced properties in mechanical properties, thermoresistance and water‐resistance. It was found that the increase in silica content increased the Tg of copolymers more than the values calculated by theoretical formulas. This article offers a proper solution to prepare nanocomposite emulsions directly from the industrial silica particles and suggests how to disperse them properly. These nanoparticles had a proper size distribution and reduced the cost of raw materials in emulsions. The adhesion results were affected by copolymer mole fractions, resulted Tg and silica contents. These paints had the proper adhesion to building ceramic tiles and these results were higher than adhesion strength to glass sheets. The adhesions to steel were the weakest results. The best monomer ratio for the highest adhesion strength was found in 0·4 MMA mole fractions which were confirmed on all substrates. The optimum silica content for best adhesion was between 0 and 30 g for glass, between 0 and 60 g for ceramic and between 0 and 90 g for steel surface. These latex samples with >60 g silica content had a proper flame resistance, but none of the monomer ratios had any considerable effect on the flammability of nanocomposites. These test results could be used for the development of industrial acrylic resins and paints which did not develop by other researches for these nanocomposites.
Footnotes
Acknowledgements
This work was supported by the contributions of the authors’ colleagues in Amirkabir University of Technology and Department of Chemistry of University of Tehran.