Journal Information
Vol. 9. Issue 1.
Pages 455-464 (January - February 2020)
Share
Share
Download PDF
More article options
Visits
...
Vol. 9. Issue 1.
Pages 455-464 (January - February 2020)
Original Article
DOI: 10.1016/j.jmrt.2019.10.073
Open Access
Effect of silica nanoparticles on the curing kinetics and erosion wear of an epoxy powder coating
Visits
...
María Fernández-Álvarez
Corresponding author
mfalvare@ing.uc3m.es

Corresponding author.
, Francisco Velasco, Asunción Bautista, Juana Abenojar
Department of Materials Science and Engineering, IAAB, Universidad Carlos III de Madrid. Avda. Universidad 30, Leganés, Madrid, Spain
Article information
Abstract
Full Text
Bibliography
Download PDF
Statistics
Figures (10)
Show moreShow less
Tables (5)
Table 1. Characteristic bands of the epoxy powder [29,38].
Table 2. Ea calculated with the Kissinger equation for AR organic powder and all mixed organic powders. Peak temperatures are also shown.
Table 3. Ea calculated with the MFK for AR organic powder and all mixed organic powders at 50% of curing degree.
Table 4. Curing time for a 98% curing degree at 160 and 180 °C for AR and all mixed organic powders, calculated by the MFK.
Table 5. Tg (°C) values, stiffness, hardness Vickers (HV0.3) and rugosity for AR and all mixed organic powder coatings.
Show moreShow less
Abstract

In this study, the wear resistance of an epoxy powder coating was improved by SiO2 nanoparticles and their possible effect on curing kinetics of the coating was also evaluated. The epoxy powder coating was prepared with different percentages of nanoparticles (1-3 by wt.%) using a hot mixer, a method that can be more economic than other ones. The particle size distribution and Fourier-transform infrared spectroscopy (FT-IR) of epoxy powder were evaluated to examine the effect of mixing on the powder. The effect of SiO2 on the curing of epoxy powder was studied by differential scanning calorimetry (DSC). The Kissinger and model free kinetics (MFK) methods were used to calculate the activation energy (Ea) of the curing process of powders. The coating spraying process was carried out in an industrial installation on carbon steel substrates. The glass transition temperature (Tg) of the coatings was also studied using DSC. The morphology of the cured organic coatings was observed by scanning electron microscopy (SEM). Stiffness and hardness Vickers (HV) were evaluated. A test based on ASTM D969 was developed to perform erosion measurements. The results obtained by both the Kissinger and MKF methods showed that nanoparticles do not influence significantly the Ea of curing of the coatings. The addition of 1% SiO2 improves the erosion wear at 45 and 60°, due to the increase in stiffness and hardness provided by the nanoparticles, though, when particles collide at 60° with the samples, the lowest thickness loss was found for the epoxy with 3% nanoreinforcements.

Keywords:
Epoxy powder coating
Nanoparticles
Silica
Differential scanning calorimetry
Kinetic models
Erosion wear
Full Text
1Introduction

Powder coatings are progressively replacing conventional liquid coatings [1,2]. Nowadays, the use of powder organic coatings is widespread in several applications (automotive, street furniture, domestic appliances). The great advantage of powder coatings over conventional liquid coatings is the absence of solvents, which makes the powder coatings eco-friendly. Moreover, powder coatings have other advantages such as superior mechanical and aesthetic properties, less waste generated, reduction in coating material loss and fast curing. The main disadvantages of the coating process are the difficulties to achieve thin layers, the high startup costs or the baking step required [3,4].

Thermosetting resins are widely used due to their high tensile strength and modulus, good thermal and dimensional stability, excellent fatigue properties and good creep resistance. For outdoor applications, polyester based coatings are usually employed as they have high resistance to ultraviolet radiation (UV) [5,6]. However, epoxy coatings present better mechanical and thermal properties, excellent adhesion to most substrates and high corrosion resistance [7–10]. There are few studies of epoxy powder coatings at present [7,11,12], since most of the studies on organic powder coatings have been focused on polyester [11,13–16].

Several studies have shown how, by adding small amounts of fillers or nanoparticles, certain properties can be improved [7–9,17]. Currently, it is widely known that wear is one of the most interesting characteristics to be enhanced for organic coatings [18]. Other properties such as corrosion resistance [7–11] or UV resistance [15] are looking to be improved. SiO2 nanoparticles are the additions most used to improve wear resistance, although other nanoparticles, such as alumosilicate [18] or calcium carbonate [19] are also employed. This improvement has especially been observed for conventional liquid organic coatings [8,20–22]. For example, Palraj et al. [20] have reported improvements in the abrasion resistance of organic coatings thanks to the use of nanoparticles instead of SiO2 microparticles. Kang et al. [23] also have managed to reduce the wear of an epoxy coating with the addition of hydrophilic SiO2 nanoparticles. Although good results have been demonstrated in liquid coatings, there is less literature on powder organic coatings, although several pioneering studies have already been published [13,18]. Alumosilicate nanofillers in amounts up to 5% are able to reduce wear of epoxy powder coatings against a metallic counterbody through affecting viscoelastic properties [18], while SiO2 nanoparticles can exert abrasive wear resistance on textured powder polyester coatings [13] although the complex structure of textured coatings affects the enhancement. For this reason, it has been considered interesting to deepen this study.

The mixing method is important to obtain an adequate distribution of the nanoparticles in the powder matrix. The improvement in the wear of the organic coatings could depend on the homogeneous distribution of the nanoparticles in the resin matrix. In previous published studies, organic powder and the nanoparticles are usually mixed in an extruder [1,7,11,20,24,25] There are also other different mixing methods that result in good final properties for the coating and which are more economical, for example, a mechanical stirrer [14] or ball milling [13].

For this research, a hot mixer was used. This mixing method is faster and more economical than the extruder method, working at temperatures higher than room temperature, but not as high as those of extruders. Scant previous studies have focused on its use to manufacture reinforced powders for organic coatings. Sharifi et al. used a mixer [7] with an epoxy powder coating and clay nanoparticles, obtaining a homogenous coating with percentages of clay up to 5%. In addition, the temperature seems to have a positive effect on the separation of aggregates and agglomerates, as shown for pyrogenic silica [26]. Hence, there are previous published results that suggest the feasibility of the approach [7,26].

The coating application process is also a critical step, especially after adding the nanoparticles. For powder coatings, the charge of the powders may be affected because of the nanoadditions. The nanoparticles could also influence the powder flow and the viscosity during melting [27].

Another important issue is understanding the effect of nanoadditions on the curing kinetics of the organic coatings. The mechanical properties (hardness, impact resistance) or the chemical resistance could be affected by the degree of the crosslinking. Few kinetic studies in powder coatings with nanoparticles are found in the literature. A study on the addition of nano-TiO2 to a polyester powder coating shows that the nanoaddition decreases the temperature at which the curing reaction starts and ends [1]. The glass transition temperature (Tg) of the cured coating also reveals important data on the interaction of the nanoparticles and the resin matrix [7].

Therefore, the main aim of this research is to achieve different epoxy organic coatings by adding different percentages of silica nanoparticles, trying not to change the curing kinetics and improving their erosion wear.

2Materials and Methods2.1Characterization of organic powders

The epoxy powder (RAL 5005) was supplied by Cubson International Consulting S.L. (Humanes de Madrid, Spain). The silica nanoparticles were hydrophilic fumed silica AEROSIL® 90 (Degussa, Germany) with a specific surface area of 75-105 m2 g-1 and an average size of 20 nm. The mixing of powder and nanoparticles was carried out in a Haake Rheocord 252p (Thermo Fisher Scientific, Massachusetts, USA) mixer at 72 ± 1 °C and 40 rpm for 15 min. All the mixes were carried out under dry conditions.

Five different organic powders were studied: the as received organic powder (labelled as AR), the AR passed through the mixer (labelled as 0%) and three different mixtures with different percentages of nanoparticles: 1%, 2% and 3% by wt. No mixtures were carried out with higher amount of nanoparticles because they could start to interfere the charging process of the powders in the gun and, so, affect the applicability of the coating.

The particle size distribution of organic powders was measured with a laser analyzer Mastersizer 2000 (Malvern Instruments, Worcestershire, UK) to obtain the particle size distribution. An absorbance of 0.1 and a refractive index of 1.5 were used, typical values of an opaque black pigment. Three measurements of each sample were carried out. The D10, D50 and D90 values were calculated, being the diameters at which 10%, 50% and 90% of the mass of the sample have a smaller diameter than this value.

To evaluate if the use of the mixer can promote any change on the functional groups of the AR organic powders, they were analyzed by Fourier-transform infrared spectroscopy (FT-IR), using a spectrometer Spectrum GX Instrument (PerkinElmer, Massachusetts, USA). Tested pellets were made of KBr and the organic powder. FT-IR spectra of the organic powders were recorded in the range of 4000-400 cm-1 at a resolution of 4 cm-1 and 10 scans. To verify that the curing reaction of the organic coating after the mixer (0%) had not yet started, the possible curing degree between AR and 0% was calculated with equation 1 (at time of curing 0). The ratio of the areas of the bands 910 cm-1 (Abs910, characteristic of the oxirane group in the epoxy ring) and 1040 cm-1 (characteristic of C-H stretching in the aromatic ring of the bisphenol) were selected. The curing reaction does not affect the aromatic ring of the bisphenol at 1040 cm-1, so it was used as a reference [28,29].

Thermal measurements were performed using differential scanning calorimeter (DSC) equipment, model 822 (Mettler Toledo GmbH, Greifensee, Switzerland). Each sample weighed approximately 10 mg and went into aluminum crucibles of 40 µl. The samples were heated in the temperature range of 25-250 °C and nitrogen (35 ml·min-1 flow) was used as purge gas. Three different heating rates (5, 10 and 20 °·min-1) were used to calculate kinetic parameters of the different powders.

The Kissinger method [30] and the model free kinetic (MFK) method [31,32] were used to calculate the activation energy of the process and to evaluate possible effects of SiO2 on the curing behavior. The Kissinger method uses equation 2, where Ea is the activation energy (J·mol-1), β is the heating rate (°·min-1), Tp is the peak of curing curve (K), R is the gas constant (J·mol-1 K-1) and C is a constant. In this method, the Ea is supposed to be constant during the curing process:

The STARe software (Mettler Toledo GmbH, Greifensee, Switzerland) was used to evaluate the kinetics by the MFK. The MFK is based on at least three temperature scans at different rates. In this model, the Ea is not constant; it is calculated as a function of conversion degree of curing. Moreover, isothermal curves at different temperatures were simulated with the MFK method.

The epoxy powder was applied on carbon steel substrates (152 × 76 x 0.8 mm), using the 5 different powders. An electrostatic spray gun (Pulverizadora Manual Easyselect) with a control unit OPTITRONIC (ITW GEMA, San Galo, Switzerland) was used at 100 kV DC voltage source. Carbon steel sheets were degreased with acetone before the coating process. The curing of the applied powders was carried out in an oven at 180 °C for 15 min. The final thickness of the organic coatings -measured with a thickness gauge (Elcometer 456, Manchester, UK) ranged between 60-80 µm.

2.2Characterization of organic coatings

Scanning electron microscopy (SEM) images were taken with a Teneo SEM (Thermo Fisher Scientific Inc., Waltham, USA). A 10 kV electron beam was used and also an energy-dispersive spectroscopy (EDS) was used to perform semi-quantitative analyses.

The glass transition temperature (Tg) of the cured organic coatings was also calculated by DSC. A heating rate of 20 ᵒ·min-1 (in the range of 25-125 °C) was used. Approximately 10 mg of each sample were analyzed.

Stiffness and Vickers hardness (HV) of the organic coatings were studied. Stiffness measurements were carried out with a universal durometer ZHU 2.5 (Zwick Roell, Ulm, Germany). The applied load was 5 N, the speed load application was 1 mm·min-1 and the speed load removal was 10 mm·min-1. Stiffness was calculated during the unloading [33]. HV measurements were performed with a Zwick CMT (Zwick Roell, Ulm, Germany) durometer. A load of 300 gf and a loading time of 10 s were used. In both cases, 9 measurements were carried out.

An opto-digital microscope Olympus DSX500 (Olympus Corporation, Tokyo, Japan) was used to calculate the roughness values (Sa) of the epoxy based coatings. At least three measurements of Sa were made in different zones of each organic coating sample.

Erosion wear was evaluated using a test based on ASTM D968-17 (“Standard Test Methods for Abrasion Resistance of Organic Coatings by Falling Abrasive”). Twenty liters of silica coarse sand (ø = 1-2 mm) were dropped from a height of 930 mm through a 19 mm diameter. Fig. 1 shows an erosive wear scheme. Samples were tilted 45° and 60° (α) above the horizontal. The impact angle of the particles (ẞ), complementary to α, was also marked in Fig. 1. Three tests were performed on each material, measuring the variation of thickness with a thickness gauge (Elcometer 456, Manchester, UK).

Fig. 1.

Scheme of erosive wear test (α = 45° and 60°).

(0.07MB).
3Results and Discussion3.1Characterization of organic powders

The hot mixing process of the SiO2 nanoparticles with the organic powder was carried out to allow a good homogeneity without approaching a temperature close to that the curing process takes place (at temperatures above 110 °C). It is important to know the particle size distribution because it affects the powder flow, the uniformity of charging and the agglomeration [34]. The size of the powder also influences its electrostatic charge during the gun application [35].

As it can be seen in Fig. 2, the powder distribution of the AR and the organic powder with the highest percentage of nanoparticles (3%) are similar. No dramatic effect of mixing process on particle size distribution can be appreciated. The presence of big agglomerates has not been found in the distributions.

Fig. 2.

Particle size distribution of AR organic powder and with 3% SiO2 mixed.

(0.11MB).

Fig. 3 shows the most typical parameters used to describe a powder distribution (D10, D50 and D90). These parameters are almost similar for the different organic powder samples. Even so, two different behaviors can be observed. Regarding the 0% sample, a small increase of the three parameters is seen compared to the AR. This could be related to a light bonding among epoxy particles due to the temperature applied during the hot mixing method. Furthermore, there is a slight decrease in the three parameters (D10, D50 and D90) for the samples that contain nanoparticles with respect to the 0%. This can be understood taking into account that nanoparticles impede the bonding between the epoxy particles. In addition, nanoparticles could affect the spraying process and the smoothness of the organic coatings. Homogenizing through hot mixing (72 ± 1 °C) does not change the particle size distribution so the mixing process would not be a problem for the spraying process.

Fig. 3.

D10, D50 and D90 parameters from particle size distribution of AR organic powder and all mixed organic powders.

(0.15MB).

In addition, fine particles can affect the surface finish of the coating. Heated organic powders melt at different rates and coarse particles simply take longer to melt than fine particles. Moreover, large particles can form hills in the film, influencing final coating smoothness (this is called ‘’orange peel’’ [36,37]). As hot mixing has not promoted fine formation, a good finishing could be expected for final coatings.

Fig. 4 shows FT-IR spectra of the AR and 0% organic powders. It can be seen that the two spectra are similar. The amount of organic powder (that is opaque) has to be very small to be able to distinguish the vibrations of the functional groups. The most important groups of the two components of the epoxy have also been indicated in Table 1. The functional groups appear to be the same in both cases, hence heating in the mixer has not influenced them. The areas of the 1040 cm-1 and 915 cm-1 peaks are also checked for AR and 0%. These areas are practically the same for both cases, so using equation 1, the curing degree is 0.

Fig. 4.

FI-IR spectroscopy of AR and 0% organic powders.

(0.12MB).
Table 1.

Characteristic bands of the epoxy powder [29,38].

Wavenumbers (cm-1Assignation 
3432  -OH st 
3055  ArC-H st, C-H st oxirane 
2964-2925-2871  -CH3, -CH2- st 
2066  ArC-C armonics 
1700  C = O st 
1611-1509  ArC-C 
1459  -CH3 δ as, -CH2- δ 
1299  C-O-C st as, oxirane 
1259  ArC-O-C-al st as 
1180  ArC-H δ ip 
1102  C-OH st 
1040  ArC-O-C-al st s 
910  C-O-C st s, oxirane 
826  ArC-H δ oop 
730  -CH2 δ, for C-(CH2)n-C n<4 

St: stretching, Ar: aromatic, δ: bending, s: symmetric, as: asymmetric, al: aliphatic, ip: in plane, oop: out of plane bending, γ: skeleton vibrations, n: number of CH2 groups.

The DSC analysis enables to determine adequate curing conditions for the powders [39]. DSC dynamic experiments were performed at three different heating rates in order to allow the Kissinger and MFK methods to be applied. The values of 5, 10 and 20 °·min-1 were chosen because they are heating rates already reported in other similar kinetic studies [38]. Fig. 5 shows the DSC spectra of all organic powders at 20 °·min-1. An endothermic transition from 65 °C to 85 °C turns up in the case of epoxy AR. This endothermic peak is higher than in the rest of powders. This is due to the enthalpy of relaxation [40,41]. The other organic powders do not present this enthalpy of relaxation because they have been mixed at high temperature and therefore this previous heating relieves possible stresses. The second peak that appears in Fig. 5 is an exothermic peak and corresponds to the curing process of the organic coatings (crosslinking). The temperature range of the curing reaction is between 110 °C to 240 °C approximately, for a heating rate of 20 °·min-1. The five non-isothermal curves have very similar curing ranges, so at first, curing temperature of organic coatings do not seem to be significantly affected by the nanoadditions.

Fig. 5.

Non-isothermal DSC curves of AR and all mixed organic powders (20 °·min-1 heating rate).

(0.11MB).

In order to study the curing process of the epoxy coating, the Ea was calculated with the Kissinger equation (Eq. 2). Table 2 shows the temperature corresponding to the minimum of the curing curve (Tp), according to the heating rate used in each case. It can be seen that there is a small increase when the percentage of nanoparticles increases, especially in the case of 5 °·min-1 and 10 °·min-1, although this increment is within the measurement error.

Table 2.

Ea calculated with the Kissinger equation for AR organic powder and all mixed organic powders. Peak temperatures are also shown.

  β (°·min-1AR  0%  1%  2%  3% 
Tp (°C)5  150  150  150  151  151 
10  165  164  165  165  166 
20  180  180  180  180  180 
Ea (kJ·mol-1  65  66  68  68  70 

The Ea value of the curing reaction is obtained by plotting –ln(β/T2p) vs. 1000/Tp (Fig. 6). The value of the slope of the line multiplied by the gas constant (8.314 J·mol-1 K-1) gives the Ea of each sample. The results obtained for Ea (Table 2) are similar to other works with epoxy resins [42]. When the amount of silica increases, Ea also does, so the curing reaction is slightly slower. This could probably be related to the inhibiting effect on the crosslinking caused by SiO2 nanoparticles in the epoxy. In any case, the differences compared to AR organic coating are small and the curing process does not seem to be affected.

Fig. 6.

The Kissinger plot for AR, 2%, and 3% mixed organic powders.

(0.1MB).

The Kissinger method [30] assumes that the curing reaction is of order 1. In a more complex cure system, however, a large number of parameters is required and the activation energy changes according to the extent of conversion without the assumption of a particular form of the reaction model [43]. In this situation, the model free kinetics (MFK) method can be an alternative approach because it is more accurate [44]. The MFK assumes that the activation energy varies during the curing reaction. To confirm that higher Ea is required for the curing of organic coatings with higher nanoparticle content, the MFK method was performed. To calculate the variation of Ea with the curing degree, first it is necessary to know the curing degree vs. temperature relationship. Fig. 7 shows the conversion curves of the 2% mixed powder coating at the three different heating rates.

Fig. 7.

Curing degree as a function of heating rate for 2% mixed organic powder.

(0.09MB).

In the MFK method, the Ea varies according to the extent of conversion, as shown in Fig. 8. The main difference observed is that samples that do not contain nanoparticles need a higher energy at the beginning of the reaction than those that have a percentage of SiO2. The AR needs about 130 kJ·mol-1 to start curing, while the 0% needs 85 kJ·mol-1. The higher Ea implies a slower process, at least during the first stages of the curing. This can be understood because the 0% has been previously subjected to temperature in the mixer, so extra energy has been added to the organic coating. In both cases, the Ea takes a constant value of 68 ± 1 kJ·mol-1 from a degree of conversion of 30%.

Fig. 8.

Ea vs. curing degree obtained by the MFK for AR and all mixed organic powders.

(0.1MB).

Samples with nanoparticles begin the curing reaction with a lower Ea than AR and 0%, 1% and 2% mixed powder coatings have a similar kinetics and their Ea is more constant during the curing. Their initial Ea are 75 and 77 kJ·mol-1 respectively, then they increase slightly and finally they decrease until a stabilization of 69 and 71 kJ·mol-1 respectively. Furthermore, for the case of 3%, the difference is higher, as it requires an initial Ea of 61 kJ·mol-1 and then stabilizes to 73 kJ·mol-1. These results show that the silica nanoparticles are catalyzing the curing reaction [45] and, therefore, the Ea is lower at the beginning. At the beginning, a greater quantity of nanoparticles favors the catalytic mechanism of curing. However, when the oxiranes begin to open (autocatalytic mechanism), nanoparticles interfere since a steric effect is occurring, and consequently the Ea increases. The addition of nanoparticles increases the final Ea needed to carry out the curing process when it is stabilized. There are positions occupied by the silica, which hinders the end of the reaction. The results of the Ea when the curing process is already stabilized (at 50% of curing degree) can also be observed in Table 3. Therefore, although numerically the results are not equal, they have the same tendency of the results obtained with the Kissinger method.

Table 3.

Ea calculated with the MFK for AR organic powder and all mixed organic powders at 50% of curing degree.

  AR  0%  1%  2%  3% 
Ea (kJ·mol-165.3  65.9  67.6  67.6  70.3 

Thanks to the MFK method, the time required for each sample for curing could also be simulated at two different temperatures. The results are shown in Table 4. At the lowest curing temperature, longer times are logically required to cure the organic coating. The differences are very small among materials, although it can be assumed that the percentage of nanoparticles increases the curing time. Silica nanoparticles affect it slightly. The industrial conditions for curing recommend 15 min at 180 °C. The difference with the results obtained is related to the amount of mass that is used for the DSC test and a real specimen. Due to practically no changes with the AR, it is still considered that industrial conditions remain optimal for powder coatings with the studied nanoparticles.

Table 4.

Curing time for a 98% curing degree at 160 and 180 °C for AR and all mixed organic powders, calculated by the MFK.

  AR  0%  1%  2%  3% 
160 °C - t (min)  11.3  11.7  11.7  13.0  12.7 
180 °C - t (min)  5.0  5.1  5.2  5.3  5.3 
3.2Characterization of organic coatings

All organic coatings were examined by SEM. No relevant differences were observed among the 1, 2 and 3% coatings. Fig. 9 shows representative SEM images corresponding to the AR organic coating (Fig. 9a) and an organic coating with silica nanoparticles (Fig. 9b). The main particles found are also indicated in the images. The EDS study has also allowed to analyze different components of the organic coatings. It can be seen that the epoxy powder coatings contain fillers as talc (the big white areas) and TiO2 (the small white areas). Comparing the images corresponding to the AR coating (as Fig. 9 left and with those corresponding to coating with nanoparticles (as Fig. 9 right), it can be checked that the reinforcements form no detectable agglomerates of nanoparticles. The absence of these agglomerates for all the studied coatings can be granted after the careful SEM study carried out. Furthermore, the different results carried out about thermal and mechanical properties (Table 5) neither suggest the presence of detrimental agglomerates.

Fig. 9.

SEM analysis of AR and 2% organic coatings.

(0.23MB).
Table 5.

Tg (°C) values, stiffness, hardness Vickers (HV0.3) and rugosity for AR and all mixed organic powder coatings.

  AR  0%  1%  2%  3% 
Tg (°C)  99.4  99.4  99.0  98.8  98.6 
Stiffness (kN/mm29.5 ± 0.9  9.9 ± 0.8  11.1 ± 0.7  14.4 ± 1.3  20.3 ± 1.3 
HV0.3  18 ± 1  20 ± 1  20 ± 1  21 ± 1  22 ± 1 
Sa (µm)  5.8 ± 0.2  5.7 ± 0.5  5.9 ± 0.3  6.0 ± 0.1  7.3 ± 0.8 

However, in addition to the fillers and pigments that the organic coatings have, the presence of nanoparticles can be observed in the 2% since there is a white contrast (heavy particles such as SiO2) with the black matrix (light particles such as epoxy). This contrast is not noticeable in the AR, where the matrix has a homogenous color, since there are no nanoparticles.

After checking that the curing process of the different organic studied powders is similar, analogous thermal properties in the cured organic coatings are expected. Table 5 shows the small variation that exists in the Tg of all organic coatings. The main important factors that can affect the Tg are the curing process and the particle dispersion [46]. Generally, when the Tg of the organic coating increases with the presence of nanoreinforcements, it is due to the restricted mobility of the polymer chains, caused by the strong interactions between nanoparticles and the resin matrix [47–49]. In this case, the Tg is not affected by the silica nanoparticles, since there is no significant change among the different samples and this also gives an idea of the homogeneity of the mixture. If the Tg had decreased with a higher content of SiO2, this would have meant that there are agglomerates that do not allow a good crosslinking of the organic coating. Hence, the organic coating would have been softer.

As can be seen also in Table 5, stiffness and hardness Vickers augment with the increasing nanosilica content. Other studies of epoxy coatings with SiO2 nanoparticles also show the increase of stiffness due to the nanoreinforcements [17,50]. Nanoparticles are introduced between the chains of the epoxy [8], making the organic coating more rigid and harder than the unreinforced organic coatings. Furthermore, the results obtained from the Tg (Table 5), can be related to the values of hardness (Table 5). In this case, it is also possible to confirm the good crosslinking of the chains since the hardness does not decrease, but increases with the addition of the nanoparticles [51,52].

Roughness values (Sa) are also included in Table 5. It can be seen that AR y 0% coatings exhibit similar roughnesses, therefore, it can be concluded that the method of mixing does not affect the roughness of the organic coatings. However, Sa increases as the nanosilica content of the coating does. Other authors also obtain greater values of roughness when they add silica nanoparticles to their epoxy coatings [23].

Fig. 10 shows the two results for erosive wear. Epoxy is brittle [49] and erosion takes place then by the formation and intersection of subsurface cracks. The erosion resistance of the organic coatings with nanoparticles may decrease or increase depending on the degree of dispersion of the nanoparticles, the particle content, the particle size distribution and the test performed [20]. ASTM D968 standard indicates 45º tilting angle for testing. In this work, the tests were performed with the samples tilted at two different angles (α = 45° and α = 60°). The use of two different tilting angles allows obtaining a more detailed wear study, as the wear mechanism of a given material is usually affected by the titling angle. These angles were chosen because they had already been used in similar studies [53,54] that show the influence of different damaging behaviors depending on the tilting angle. It is generally assumed that erosion of materials takes place under two main mechanisms: cutting and deformation [53]. Cutting is usually associated with the impact of particles with enough energy to gouge fragments of material. This mechanism easily causes a loss of material when α is close to 90°. On the other hand, deformation relates to particles that impact perpendicular to the surface and that are able to generate stresses higher than yield strength [54] causing this type of damage on the coating. Actually, when the samples are titled at intermediate angles, the mechanism of erosion by solid particles and the total erosion caused in the coating can be often expressed as the sum of effects of the deformation and the effects of the cutting erosion [53,54]. Initially, deformation should be the main effect of the impacts, increasing the roughness of the surface. In rough, deformed surfaces, the loss of thickness of the coating by cutting would start to be the dominant wear mechanism.

Fig. 10.

Thickness loss during erosion wear test of AR organic powder and all mixed organic powder, at two different angles (α = 45° and α = 60°).

(0.2MB).

The obtained results (Fig. 10) prove that, when α increases, the impacts become more aggressive to the coating, being the loss of thickness of all organic coatings greater for the case of 60° than the 45° tests. Higher titling angles cause impacts that favor higher deformations of the surface of the coating. When further impacts take place on deformed surfaces, partial cutting of small pieces from the coating tends takes place. When, after a certain amount of previous impacts, the cutting mechanism becomes the controlling step, the highest damage occurs, and fast thickness losses of the coating are observed.

The increase of stiffness and hardness of the coatings due to nanosilica additions should foreseeably allow to achieve an decrease on the wear losses [55]. The positive effect of nanosilica addition on the wear resistance can clearly be observed for the 1% coatings, where the amount of added nanoparticles does not cause any noticeable increase on their roughness (Table 5). For these coatings, the hardening caused by the nanoadditions (Table 5) makes the surface less deformable against the impacts and delays the onset of the cutting step. However, for tests carried out at 45º, the beneficial effect of further hardening achieved by adding 2% or 3% SiO2 (Table 5) is masked by the increase on the roughness of the surface that also takes place due to the additions (Table 5). As the roughness increases, the probabilities of impacts provoking higher losses of material by cutting do, and this wear mechanism becomes soon the controlling one for rough surfaces. The time the cutting mechanism takes to the be controlling step during the wear tests is the key point for determining the final mass losses. For this reason, when nanoreinforced coatings are tested at 45°, the thickness losses for 1% are the minimum (Fig. 10), but further amount of additions has a negative effect.

On the other hand, for erosion tests carried at 60°, the striking particles already collide in way that can easily cause marked deformations after a few impacts. Hence, mass losses by cutting mechanism can soon appear in erosive test for non-rough coatings. 60° titling angle is a more aggressive condition than 45°, so the effect of the stiffness increase achieved SiO2 additions dominate over that caused by the highest Sa. In this case, the effect of the roughness on the time delay needed for the cutting mechanism to be the controlling step become less relevant. For the highest impact angle tested, the coating with the highest amount of silica tested -the one with higher stiffness- become the most wear-resistant.

Hence, the addition of 1% SiO2 can be positive for the epoxy coatings to withstand erosive stresses at moderate titling angle impacts, though, when particles collide forming higher, more aggressive angles with the material, higher concentration of nanoreinforcements would be advisable, as their effect silica on the mechanical properties of the coating becomes more critical than the modifications that they cause on the roughness.

4Conclusions

The main conclusions that can be drawn from this research are:

  • The method of mixing used can be adequate to mix hydrophilic silica nanoparticles and epoxy powder. The heat employed during mixing does not affect the initial organic powder properties, as FT-IR has proved.

  • The two methods used to study the curing kinetics of organic coatings -the Kissinger and the MFK-, show similar values for Ea. In addition, the percentage of nanoparticles slightly increases the Ea values in both cases.

  • The MFK shows that organic coatings containing nanoparticles have similar kinetics (Eavs. curing degree), compared to AR and 0% organic powders. The main difference is observed in the catalytic effect of the nanoparticles at the beginning of the curing reaction.

  • The Tg of the cured organic coatings does not vary meaningfully, which means that the nanoparticles have not influenced the curing process of the epoxy coating. Moreover, it confirms the good homogeneity of the mixture as well.

  • Silica nanoparticles have promoted the increase in the stiffness and hardness of epoxy powder coating.

  • Erosion wear is influenced by the mechanical properties of the coatings and their roughness. The addition of 1% SiO2 improves the wear resistance in the two tests carried out, though when particles collide at 60º with the samples, the lowest thickness loss was found for the epoxy with 3% SiO2.

Conflicts of interest

The authors declare no conflicts of interest.

Acknowledgments

The authors acknowledge Cubson International Consulting for their help with the coating process. This research was funded by Interreg SUDOE, through the KrEaTive Habitat project [grant number SOE1/P1/E0307].

References
[1]
Q Shi, W Huang, Y Zhang, Y Zhang, Y Xu, G. Guo.
Curing of polyester powder coating modified with rutile nano-sized titanium dioxide studied by DSC and real-time FT-IR.
J Therm Anal Calorim, 108 (2012), pp. 1243-1249
[2]
M Barletta, A Gisario, F Trovalusci, S. Vesco.
Visual appearance and scratch resistance of high performance thermoset and thermoplastic powder coatings.
Prog Org Coat, 76 (2013), pp. 244-256
[3]
P Castell, M Wouters, H Fischer, G. De With.
Kinetic studies of a UV-curable powder coating using photo-DSC, real-time FTIR and rheology.
J Coat Technol Res, 4 (2007), pp. 411-423
[4]
MKI Khan, MAI Schutyser, K Schroën, RM. Boom.
Electrostatic powder coating of foods - State of the art and opportunities.
J Food Eng, 111 (2012), pp. 1-5
[5]
L Halász, K. Belina.
An investigation into the curing of epoxy powder coating systems.
J Therm Anal Calorim, 119 (2015), pp. 1971-1980
[6]
M Barletta, A Gisario, L Lusvarghi, G Bolelli, G Rubino.
On the combined use of scratch tests and CLA profilometry for the characterization of polyester powder coatings: Influence of scratch load and speed.
Appl Surf Sci, 254 (2008), pp. 7198-7214
[7]
M Sharifi, M Ebrahimi, S. Jafarifard.
Preparation and characterization of a high performance powder coating based on epoxy/clay nanocomposite.
Prog Org Coat, 106 (2017), pp. 69-76
[8]
ZT Khodair, AA Khadom, HA. Jasim.
Corrosion protection of mild steel in different aqueous media via epoxy/nanomaterial coating: Preparation, characterization and mathematical views.
J Mater Res Technol, (2018), pp. 1-12
[9]
H Shi, F Liu, L Yang, E. Han.
Characterization of protective performance of epoxy reinforced with nanometer-sized TiO2 and SiO2.
Prog Org Coat, 62 (2008), pp. 359-368
[10]
J Yu, W Zhao, Y Wu, D Wang, R. Feng.
Tribological properties of epoxy composite coatings reinforced with functionalized C-BN and H-BN nanofillers.
Appl Surf Sci, 434 (2018), pp. 1311-1320
[11]
D Piazza, AF Baldissera, SR Kunst, ES Rieder, LC Scienza, CA Ferreira, et al.
Influence of the addition of montmorillonite in an epoxy powder coating applied on carbon steel.
Mater Res, 18 (2015), pp. 897-903
[12]
JM Salla, JM Morancho, X Ramis, A. Cadenato.
Isothermal degradation and thermooxidative degradation of an epoxy powder coating.
J Therm Anal Calorim, 80 (2005), pp. 163-169
[13]
M Fernández-Álvarez, F Velasco, A. Bautista.
Effect on wear resistance of nanoparticles addition to a powder polyester coating through ball milling.
J Coat Tecnhol Res, 15 (2018), pp. 771-779
[14]
P Jawahar, R Gnanamoorthy, M. Balasubramanian.
Tribological behaviour of clay - thermoset polyester nanocomposites.
Wear, 261 (2006), pp. 835-840
[15]
G Gheno, R Ganzerla, M Bortoluzzi, R. Paganica.
Accelerated weathering degradation behaviour of polyester thermosetting powder coatings.
Prog Org Coat, 101 (2016), pp. 90-99
[16]
M Zouari, M Kharrat, M Dammak, M. Barletta.
A comparative investigation of the tribological behavior and scratch response of polyester powder coatings filled with different solid lubricants.
Prog Org Coat, 77 (2014), pp. 1408-1417
[17]
M Conradi, A Kocijan, D Kek-merl, M Zorko, I. Verpoest.
Mechanical and anticorrosion properties of nanosilica-filled epoxy-resin composite coatings.
Appl Surf Sci, 292 (2014), pp. 432-437
[18]
AY Grigoriev, GV Vaganov, VE Yudin, NK Myshkin, IN Kovaleva, IV. Gofman, et al.
Friction and wear of powder coatings of epoxy composites with alumosilicate nanoparticles.
J Frict Wear, 33 (2012), pp. 101-107
[19]
M Kalaee, S Akhlaghi, A Nouri, S Mazinani, M Mortezaei, M Afshari, et al.
Effect of nano-sized calcium carbonate on cure kinetics and properties of polyester/epoxy blend powder coatings.
Prog Org Coat, 71 (2011), pp. 173-180
[20]
S Palraj, M Selvaraj, K Maruthan, G. Rajagopal.
Corrosion and wear resistance behavior of nano-silica epoxy composite coatings.
Prog Org Coat, 81 (2015), pp. 132-139
[21]
S Zhou, L Wu, J Sun, W. Shen.
The change of the properties of acrylic-based polyurethane via addition of nano-silica.
Prog Org Coat, 45 (2002), pp. 33-42
[22]
M Malaki, Y Hashemzadeh, M. Karevan.
Effect of nano-silica on the mechanical properties of acrylic polyurethane coatings.
Prog Org Coat, 101 (2016), pp. 477-485
[23]
Y Kang, X Chen, S Song, L Yu, P Zhang.
Friction and wear behavior of nanosilica-filled epoxy resin composite coatings.
Appl Surf Sci, 258 (2012), pp. 6384-6390
[24]
SM Mirabedini, A. Kiamanesh.
The effect of micro and nano-sized particles on mechanical and adhesion properties of a clear polyester powder coating.
Prog Org Coat, 76 (2013), pp. 1625-1632
[25]
D Piazza, DS Silveira, NP Lorandi, EJ Birriel, LC Scienza, AJ. Zattera.
Polyester-based powder coatings with montmorillonite nanoparticles applied on carbon steel.
Prog Org Coat, 73 (2012), pp. 42-46
[26]
AA Esmailpour, N Mostoufi, R. Zarghami.
Effect of temperature on fluidization of hydrophilic and hydrophobic nanoparticle agglomerates.
Exp Therm Fluid Sci, 96 (2018), pp. 63-74
[27]
Q Huang, H Zhang, J. Zhu.
Flow properties of fine powders in powder coating.
Particuology, 8 (2010), pp. 19-27
[28]
J Abenojar, MA Martínez, F Velasco, V Pascual-Sánchez, JM. Martín-Martínez.
Effect of Boron Carbide Filler on the Curing and Mechanical Properties of an Epoxy Resin.
J Adhes, 85 (2009), pp. 216-238
[29]
J Abenojar, MA Martínez, M Pantoja, F Velasco, JC. Del Real.
Epoxy Composite Reinforced with Nano and Micro SiC Particles: Curing Kinetics and Mechanical Properties.
J Adhes, 88 (2012), pp. 418-434
[30]
HE. Kissinger.
Reaction Kinetics in Differential Thermal Analysis.
Anal Chem, 29 (1957), pp. 1702-1706
[31]
P Taylor, S Vyazovkin, CA. Wight.
Isothermal and non- isothermal kinetics of thermally stimulated reactions of solids.
Int Rev Phys Chem, (2013), pp. 37-41
[32]
S Vyazovkin, CA. Wight.
Model-free and model-fitting approaches to kinetic analysis of isothermal and nonisothermal data.
Thermochim Acta, (2002), pp. 340-341
[33]
F Velasco, G Blanco, A Bautista, MA. Martínez.
Effect of welding on local mechanical properties of stainless steels for concrete structures using universal hardness tests.
Constr Build Mater, 23 (2009), pp. 1883-1891
[34]
W Chok, W Guo, RA Sims, G Tebbetts, DL Wankum, T Chasser, et al.
Effect of Particle Size Distribution on Transfer Efficiency and Appearance in Powder Coatings.
Part Sci Technol, 16 (2007), pp. 239-248
[35]
P Ratanatriwong, S. Barringer.
Particle size, cohesiveness and charging effects on electrostatic and nonelectrostatic powder coating.
J Electrostat, 65 (2007), pp. 704-708
[36]
AS Biris, MK Mazumder, CU Yurteri, RA Sims, J Snodgrass, S. De.
Gloss and texture control of powder coated films.
Part Sci Technol, 19 (2001), pp. 199-217
[37]
PR. Horinka.
Powder particle size : Its effects on coating line performance.
Powder Coating, 6 (1995), pp. 37-45
[38]
AQ Barbosa, LFM Da Silva, J Abenojar, JC Del Real, RMM Paiva, A. Öchsner.
Kinetic analysis and characterization of an epoxy/cork adhesive.
Thermochim Acta, 604 (2015), pp. 52-60
[39]
L Gherlone, T Rossini, V. Stula.
Powder coatings and differential scanning calorimetry: The perfect fit.
Prog Org Coat, 34 (1997), pp. 57-63
[40]
GM Odegard, A. Bandyopadhyay.
Physical aging of epoxy polymers and their composites.
J Polym Sci Part B Polym Phys, 49 (2011), pp. 1695-1716
[41]
N Causse, E Dantras, C Tonon, M Chevalier, H Combes, P Guigue, et al.
Enthalpy relaxation phenomena of epoxy adhesive in operational configuration: Thermal, mechanical and dielectric analyses.
J Non Cryst Solids, 387 (2014), pp. 57-61
[42]
J Abenojar, J Tutor, Y Ballesteros, JC del Real, MA. Martínez.
Erosion-wear, mechanical and thermal properties of silica filled epoxy nanocomposites.
Compos Part B Eng, 120 (2017), pp. 42-53
[43]
A Burkanudeen, P. Ramesh.
Novel latent epoxy curing agent for secondary insulation in electrical rotors and stators.
IEEE Trans Dielectr Electr Insul, 19 (2012), pp. 1791-1798
[44]
DE Lee, HW Kim, B Kong, HO. Choi.
A study on the curing kinetics of epoxy molding compounds with various latent catalysts using differential scanning calorimetry.
J. Appl. Polym. Sci., 45252 (2017), pp. 1-9
[45]
V Arabli, A. Aghili.
The effect of silica nanoparticles, thermal stability, and modeling of the curing kinetics of epoxy/silica nanocomposite.
Adv Compos Mater, 24 (2015), pp. 561-577
[46]
A Omrani, LC Simon, AA. Rostami.
The effects of alumina nanoparticle on the properties of an epoxy resin system.
Mater Chem Phys, 114 (2009), pp. 145-150
[47]
I Isik, U Yilmazer, G. Bayram.
Impact modified epoxy/montmorillonite nanocomposites: Synthesis and characterization.
Polymer, 44 (2003), pp. 6371-6377
[48]
A Allahverdi, M Ehsani, H Janpour, S. Ahmadi.
The effect of nanosilica on mechanical, thermal and morphological properties of epoxy coating.
Prog Org Coat, 75 (2012), pp. 543-548
[49]
N Saba, M Jawaid, OY Alothman, Z. Almutairi.
Evaluation of dynamic properties of nano oil palmempty fruit bunch filler/epoxy composites.
J Mater Res Technol, (2018), pp. 8-13
[50]
X Shi, TA Nguyen, Z Suo, Y Liu, R. Avci.
Effect of nanoparticles on the anticorrosion and mechanical properties of epoxy coating.
Surf Coat Technol, 204 (2009), pp. 237-245
[51]
SX Zhou, LM Wu, J Sun, WD. Shen.
Effect of nanosilica on the properties of polyester-based polyurethane.
J Appl Polym Sci, 88 (2003), pp. 189-193
[52]
N Domun, H Hadavinia, T Zhang, T Sainsbury, GH Liaghat, S. Vahid.
Improving the fracture toughness and the strength of epoxy using nanomaterials – a review of the current status.
Nanoscale, 7 (2015), pp. 10294-10329
[53]
HMI Clark, KK. Wong.
Impact angle, particle energy and mass loss in erosion by dilute slurries.
[54]
V Javaheri, D Porter, V. Kuokkala.
Slurry erosion of steel – Review of tests, mechanisms and materials Slurry erosion of steel – Review of tests, mechanisms and materials.
[55]
MH Moradi, M Aliofkhazraei, M Toorani, A Golgoon, AS. Rouhaghdam.
SiAlON–epoxy nanocomposite coatings: Corrosion and wear behavior.
J Appl Polym Sci, 133 (2016), pp. 1-13
Copyright © 2019. The Authors
Journal of Materials Research and Technology

Subscribe to our newsletter

Article options
Tools
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.