Introduction

As governments become more and more strict about regulations related to volatile organic compounds (VOCs) emission, industries have to develop more eco-friendly products such as UV, powder, and waterborne coatings.1 UV-waterborne technology presents interesting advantages as it combines the advantages of UV and waterborne technology. First, the polymerization process is achieved faster without VOCs emission. The use of water as unique solvent reduces the viscosity of coating formulations to favor better use and spray applications in more ecological and safe conditions. In comparison with high solids content UV-coatings, UV-waterborne coatings are known to be insensitive to oxygen.2–4 UV-cured coatings present excellent chemical and heat resistance and waterborne coatings give better adhesion in particular on wood.5 However, UV-waterborne coatings present some significant disadvantages which should be cited. The UV-curing of waterborne coating formulations requires previous drying to remove water which lengthens the finishing step. As a result the production line is significantly slowed down and the photocrosslinking is performed in the solid state. Furthermore, waterborne coatings present poor wetting and strict control of temperature and humidity is required to obtain a complete cure.5 Finally, most of the properties of UV-waterborne coatings are lower than those of high solids content UV-coatings.

Polyurethane-acrylates (PUA) are widely used as telechelic oligomers in UV-waterborne applications due to their excellent properties such as adhesion on substrates, flexibility, durability, impact and tensile strength, scratch and abrasion resistance, and weather stability.6 Waterborne polyurethane dispersions (WPD) usually consist of ionic polymers end-capped with acrylate double bonds and containing carboxylic groups –COO−M+ or quaternary salts –NR4 +X−.

In the past few years, nano-reinforced coatings have received more attention due to considerable improvement observed in nanocomposite mechanical properties. Improvement in abrasion, scratch, solvent and UV resistance, thermal stability, flame retardancy as well as other mechanical properties are observed following the introduction of nanoparticles in polymer matrix.7–10 Thus, the understanding of nanoparticle behavior toward polymer matrix is essential to achieve expected properties.

The most commonly used nano-reinforcement in the coating industry are oxides such as alumina (Al2O3), silica (SiO2), titania (TiO2), zinc oxide (ZnO), calcium carbonate (CaCO3), and clays such as montmorillonite (MMT). Alumina and silica were selected as nanoparticles in this work as they are known not only for their hardness on Mohs’ scale (9 and 7, respectively), but also for their excellent scratch and abrasion properties imparted to coatings upon their addition.11 A good dispersion of these oxides is difficult to achieve in non-aqueous media due to the presence of hydroxyl groups on their surface. Thus, the chemical surface has to be modified from hydrophilic to organophilic. The surface modification of silica nanoparticles by trialkoxysilanes is well known to improve the dispersibility of these nanoparticles in acrylate media.8,12,13 Therefore, it was decided to work with alumina nanoparticles presenting hydroxyl groups on their surface and silica nanoparticles modified by methacryloxypropryltrimethoxysilane (MEMO) groups.

A good dispersion of nanoparticles in polymer matrix is expected to be of great importance in achieving suitable properties. Therefore, it is important to avoid nanoparticle aggregation to promote the establishment of bonds between the inorganic nanoparticles and the organic polymer. However, in the wooden furniture and kitchen cabinet manufacturing industry, nanoscale dispersion is difficult to achieve because of the efficiency of dispersion methods commonly used in this sector of activities (high speed disperser, three roll mill, and bead mill).14

The aim of this research was to investigate the potential of nanotechnology in the wooden furniture and kitchen cabinet manufacturing industry. More specifically, this work aimed to study the effect of alumina and silica nanoparticles on the properties of varnishes type UV-cured waterborne nanocomposites coatings for wood products. Mechanical, optical, and thermal properties of varnish type UV-cured waterborne coatings based on PUA resin were analyzed and compared to those of the neat UV-cured varnish containing no nanoparticles. Hardness, scratch resistance, and gloss were assessed because these properties are of great importance in the wooden furniture and kitchen cabinet manufacturing industry where coating surfaces have to retain good properties for many years.15

Experimental

Material

The commercial aqueous polyurethane acrylate (PUA) dispersion was purchased from Bayer MaterialScience (Bayhydrol UV 2282). A photoinitiator, Irgacure 819DW (Ciba Specialty Chemicals) was added to the formulation at a typical loading of 0.92 wt%. Several additives were used to complete the formulation: a defoamer (Foamex 822, Evonik Degussa GmbH), a surfactant (Byk 348, BYK-Chemie), and a rheological modifier (Acrysol RM-2020NPR, Rohm and Haas). All compounds which are listed in Table 1 were used as received.

Table 1 Composition of UV-waterborne nanocomposite formulations
Full size table

Alumina nanoparticles (Aeroxide Alu C, Evonik Degussa GmbH) have a particle size of 13 nm and a specific surface area of 100 m2/g.16 Silica nanoparticles were surface modified by MEMO groups. Surface modified nanosilica (Aerosil R7200, Evonik Degussa GmbH) has a particle size of 12 nm and a specific surface area of 150 m2/g.17 These alumina and surface modified silica nanoparticles used in this project whose properties are presented in Table 2 were supplied in powder form and were used as received.

Table 2 Properties of nanoalumina and surface modified nanosilica
Full size table

Sugar maple (Acer saccharum) was selected as substrate for UV-waterborne coatings in this work. This species is commonly used in Canada and USA and represents an important commercial hardwood. Its wood having close grain is used in many common applications such as furniture, flooring, interior joinery, paneling, etc. In comparison with other hardwood species present in North America, sugar maple wood possesses excellent physical properties in terms of density, hardness, bending strength, and stiffness. Sugar maple wood is appreciated not only for its working properties (machinability, sandability, paintability, and stainability) by industry, but also for its light color (ivory white).18 Flat-sawn samples in radial section were conditioned at 40% relative humidity (RH) and 20°C to reach moisture content (MC) around 8% prior to use.

Preparation of UV-waterborne nanocomposite coatings

The neat formulation based on the PUA resin, the photoinitiator and the additives was mixed and dispersed using a high speed disperser (Ragogna, Custom Machinery Ltd) where glass beads were added in order to improve the shear rate. Then water was added to adjust the viscosity of the UV-waterborne formulation. Nanoparticles were dispersed into the neat formulation at typical loadings of 1, 3, and 5 wt% then high shear mixed for 20 min. Using a roller coater wood samples were coated for the characterization of abrasion resistance and optical properties. Glass plates were coated for the characterization of the hardness of UV-waterborne nanocomposite coatings. Moreover films were cast on coated paper sheets using foam brushes to evaluate the dispersion of nanoparticles in dried/UV-cured coatings and to characterize thermal properties. First, samples were dried for 8 min at 80°C in an oven to remove water. Finally, all samples were cured on a UV-line (Sunkiss—Ayotte Techno-Gaz Inc.) under a medium pressure mercury lamp (≈236 W/cm) at a speed of 5 m/min. The UV-dose received by the samples measured using a radiometer was around 570 mJ/cm2 and temperatures ranged from 25 to 30°C. Coating thickness was measured around 50 μm by means of an ultrasonic thickness gage (Positector 100, BYK-Gardner). UV-curing was performed at ambient temperature under air atmosphere.

Characterization of UV-waterborne nanocomposite films

Attenuated total reflectance–Fourier transform infrared spectroscopy

Fourier transform infrared (FTIR) analyses were performed in attenuated total reflectance (ATR) mode with a Magna 850 infrared spectrometer (ThermoElectron) equipped with a diamond Golden Gate ATR crystal (Golden-Gate™, Specac Ltd.). The wavenumber range was set from 750 to 4000 cm−1 with a resolution of 4 cm−1. First, ATR-FTIR spectroscopy was used to confirm the grafting of MEMO on nanosilica powder. Then it was used to demonstrate the photopolymerization of dried/UV-cured waterborne coating formulations following the conversion of acrylate double bonds on coating films after UV exposure.

Transmission electron microscopy

TEM study was carried out on dried/UV-cured waterborne films to investigate the dispersion of nanoparticles in dried/UV-cured coatings. Film samples were imbedded in epoxy resin which was cured for 24 h at 37°C then for 3 days at 60°C. Ultrathin sections were mounted on nickel/formvar grids and colored with uranyl acetate and lead citrate. TEM investigation was performed on a JEOL JEM-1230 operating at 80 kV.

Mechanical properties

The hardness of dried/UV-cured coatings was determined by measuring the deflection angle of a pendulum oscillating on the coating according to the standard ASTM D4366 (Pendulum hardness tester, BYK-Garner). The hardness value corresponded to the damping time of the pendulum from 6° to 3° (König pendulum).

The scratch resistance evaluation was performed with a Gardner straight line from BYK-Gardner. An abrasive pad (scotch brite no. 07447) was pulled across the coated surfaces with a load of 1 kg. The scratch resistance value was given as gloss retention at 60° after 50 cycles of reciprocating linear motion according to the following equation (1):

$$ R = {\frac{{B_{50} }}{{B_{0} }}} \times 100, $$
(1)

where R represents the gloss retention at 60°, B 50 is the gloss of the coating after a scratch wear of 50 cycles, and B 0 corresponds to the initial gloss of the coating.

The pull-off test was used to measure the adhesion strength between wood and dried/UV-cured coating film. Experiments were performed with a Positest Pull-Off Adhesion Tester (DeFelsko Corporation) according to ASTM D4541. A 20-mm diameter aluminum dolly was glued onto the coating surface with an epoxy resin. After 24 h of drying at room temperature, the test area was isolated with a cutting tool. Then a dolly was pulled away from sugar maple sample by applying a force perpendicular to the surface test. The strength required to pull the coatings away from sugar maple sample was recorded. Presence of wood fibers on the surface of the dolly was assessed for a better interpretation of numbers.

Optical properties

The influence of nanoparticles on optical properties of dried/UV-cured waterborne nanocomposite coatings was determined in terms of 60° gloss by means of a glossmeter (Micro-TRI-gloss, BYK-Gardner).

Thermal properties

The glass transition temperature (T g) was measured with a differential scanning calorimeter (DSC631e, Mettler Toledo). Dried/UV-cured waterborne coating films of about 10 mg were placed in covered aluminum pans. Experiments which were performed under air atmosphere consisted of three consecutive runs: (i) dynamic heating from −20 to 200°C at a heating rate of 20°C/min followed by an isothermal heating at 200°C for 2 min in order to remove relaxation enthalpy of PUA films; (ii) dynamic cooling down to −20°C at a cooling rate of −50°C/min followed by an isothermal cooling down at −20°C for 2 min; (iii) second dynamic heating from −20 to 200°C at a heating rate of 20°C/min to determine the glass transition temperature of dried/UV-cured coatings.

Results and discussion

Characterization of surface modified nanosilica powder

The grafting of MEMO groups onto silica nanoparticles was monitored following the appearance of characteristic bands of carbonyl and acrylate double bonds. ATR-FTIR spectra of modified (dotted line) and unmodified (solid line) nanosilica are presented in Fig. 1 whereas IR absorption peaks are listed in Table 3. The 800–4000 cm−1 (Fig. 1a) range shows two strong peaks at 803 cm−1 (Si–O–Si) and 1032 cm−1 (Si–O–Si, Si–O–C) which are typical of silica.19–22 Thus, the intensity of these peaks indicates that modified nanosilica powder is mainly based on silicon. The 1200–1900 cm−1 range (Fig. 1b) shows the appearance of new absorption peaks in modified nanosilica spectrum compared with that of unmodified nanosilica. These peaks related to MEMO groups suggest the grafting of MEMO onto modified nanosilica.19,22–25 The intensity of these peaks is lower than that of peaks related to silica which indicates that the loading in MEMO groups is lower than that in silicon. The vibration band at 1702 cm−1 is typical of the stretching of carbonyl group (C=O). The band at 1637 cm−1 is assigned to the C=C stretching mode. Peaks at 1408 and 1456 cm−1 correspond to the deformation in plane of vinyl bond and deformation of methylene, respectively. The two peaks at 1302 and 1327 cm−1 are assigned to skeletal –C–CO–O–. It is interesting to note that the spectrum of modified nanosilica also presents new characteristic bands at 2895, 2931, and 2959 cm−1 in the 2500–4000 cm−1 range which is not presented in the present study. These peaks which are not present in the spectrum of unmodified nanosilica were attributed to the stretching vibration of CH X .19–21,24

Fig. 1
figure 1

ATR-FTIR spectra of modified (dotted line) and unmodified (solid line) nanosilica powders within (a) the 800–4000 cm−1 range, (b) the 1200–1900 cm−1 range

Full size image
Table 3 IR peak positions of modified and unmodified silica nanoparticles
Full size table

Morphology of UV-waterborne nanocomposite coatings

UV-curing process of UV-waterborne coating formulations

The photopolymerization of UV-waterborne PUA formulations upon UV exposure was studied by means of ATR-FTIR spectroscopy on dried/UV-cured waterborne coatings by following the decrease of the characteristic C=C band at 1636 cm−1. The C=O band at 1724 cm−1 was used as invariant band to normalize the spectra. Figure 2 presents the IR spectrum of the neat formulation before and after drying/UV curing and clearly shows the disappearance of the characteristic C=C band at 1636 cm−1 following UV exposure. This behavior illustrates the conversion of acrylate double bonds according to a free-radical polymerization process.2,3,8,9 It is interesting to note the marked decrease of the strong peak at 3349 cm−1. The decrease of this peak ascribed to hydroxyl groups (–OH) from water molecules is mainly due to the drying process required prior to UV-curing.2

Fig. 2
figure 2

ATR-FTIR spectrum of the neat formulation containing no nanoparticles before (solid line) and after (dotted line) drying and UV exposure

Full size image

Dispersion of nanoparticles in UV-waterborne coatings

Figure 3 shows TEM images of the dispersion of 3 wt% of nanoalumina (Fig. 3a) and nanosilica (Fig. 3b) in dried/UV-cured waterborne nanocomposite coatings. Pictures demonstrate that nanometer scale dispersion is not completely reached as aggregates larger than nanometer size (<100 nm) are present. This behavior is observed in all UV-waterborne nanocomposite coatings as soon as 1 wt% of nanoparticles is added. Nanometer scale complete dispersion is difficult to reach because of the powder form of nanoparticles which already presents aggregates when received as showed in Fig. 4.25 The dispersion of nanoparticles in polymer matrix can be improved by adjusting some parameters such as the mixing method, the addition and the type of dispersants or the nanoparticle loading. Research with regard to ultrasonication effect on the dispersion of nanoparticles in UV-waterborne nanocomposite coatings have been realized by our research group. After high speed mixing, nanoparticles were dispersed for 20 min using an ultrasound probe (750 W). Formulations were placed in an ethylene glycol bath at 5°C and the temperature in formulations did not exceed 40°C. Figure 5 shows TEM images of dried/UV-cured waterborne nanocomposite coatings containing of 3 wt% of nanosilica dispersed by means of high speed disperser (Fig. 5a) and ultrasonication (Fig. 5b). Despite the continuous presence of aggregates, ultrasonication seems more efficient than high speed mixing as the number and the size of aggregates decrease following ultrasonication in comparison to high speed mixing. While larger nanosilica aggregates observed in dried/UV-cured waterborne coating containing 3 wt% of nanosilica dispersed by means of high speed mixing has sizes near around 14 μm, larger nanosilica aggregates observed in dried/UV-cured waterborne coating containing 3 wt% of nanosilica dispersed by means of ultrasonication were around 6 μm in size.

Fig. 3
figure 3

TEM images of dried/UV-cured waterborne nanocomposite coatings containing (a) 3 wt% of nanoalumina and (b) 3 wt% of modified nanosilica

Full size image
Fig. 4
figure 4

TEM image of nanoalumina powder

Full size image
Fig. 5
figure 5

TEM images of dried/UV-cured waterborne nanocomposite coatings containing 3 wt% of nanosilica dispersed by means of (a) high shear mixing and (b) ultrasonication

Full size image

Table 4 shows the mean size of the largest aggregates observed in dried/UV-cured waterborne nanocomposite coatings containing nanoalumina and nanosilica. The dispersion of modified nanosilica in UV-waterborne PUA nanocomposite coatings is better than that of nanoalumina as the size of nanosilica aggregates decreased in comparison with that of nanoalumina aggregates. At 1 wt% content, the size of nanoalumina aggregates is twice larger than that of nanosilica aggregates. The surface properties of nanoparticles are probably responsible of this general trend. Indeed, a good dispersion of hydrophilic nanoparticles is difficult to achieve in non-aqueous media and surface modification of nanoparticles by trialkoxysilanes is well known to improve the dispersibility of these nanoparticles in acrylate media especially for nanosilica.8,12,13 As nanosilica used in this work presents MEMO group on its surface, it can be expected that the dispersion of surface modified nanosilica in UV-waterborne coating will be better than that of nanoalumina presenting hydroxyl groups on its surface.

Table 4 Mean size of the largest aggregates observed in the dried/UV-cured waterborne nanocomposite coatings and their glass transition temperature
Full size table

Effect of nanoparticle content on mechanical properties of UV-waterborne coatings

König hardness

Figure 6 shows König hardness of dried/UV-cured waterborne nanocomposite coatings as a function of nanoparticle content. The neat coating containing no nanoparticles presents a damping time value of 183 ± 5 s, whereas nanocomposite coatings present damping times lower than 120 s. Thus, the addition of nanoparticles is found to significantly decrease the hardness of nanocomposite coatings. This behavior is surprising taking the hardness on Mohs scale of alumina and silica (9 and 7, respectively) into account. The presence of aggregates in UV-waterborne nanocomposite coatings, as confirmed with TEM images (Fig. 3), is certainly responsible for this trend. Because of their larger-than-nanometer size, aggregates can strongly increase the surface roughness of nanocomposite coatings. As hardness was found to strongly depend on the surface roughness, lower hardness of UV-waterborne nanocomposite coatings could be expected because of the presence of aggregates.26 Nevertheless it is possible to improve the König hardness of such coatings using more efficient mixing methods such as ultrasonication as discussed previously.

Fig. 6
figure 6

König hardness of dried/UV-cured waterborne nanocomposite coatings containing nanoalumina and nanosilica

Full size image

While Table 4 shows the mean size of the largest aggregates observed in dried/UV-cured waterborne nanocomposite coatings contained as a function of the mixing method, Fig. 7 shows the influence of the mixing method on König hardness of dried/UV-cured waterborne nanocomposite coatings containing nanoalumina. As discussed previously in the case of nanosilica, Table 4 clearly demonstrates that ultrasounds markedly decrease the size of aggregates. As a result, the damping time of UV-waterborne nanocomposite coatings containing nanoalumina dispersed by means of ultrasonication increases up to 144 ± 4% as soon as 1 wt% is added (Fig. 7). Not only does the size of aggregates decrease following the use of ultrasounds as dispersing method, but also the König hardness increases in comparison to UV-waterborne nanocomposite coatings formulated using the same nanoparticles loading mixed with a high speed disperser. It is interesting to notice that the damping times of UV-waterborne coatings prepared with ultrasounds are still lower than that of the reference coating. Despite the fact that the size of aggregates decreases following ultrasonication, aggregates remain larger than nanometer size. As a result König hardness of UV-waterborne coatings prepared with ultrasounds is lower than that of the reference coating containing no nanoparticles.

Fig. 7
figure 7

Influence of the mixing method on König hardness of dried/UV-cured waterborne nanocomposite coatings containing nanoalumina

Full size image

Scratch resistance

Results of scratch resistance of dried/UV-cured waterborne nanocomposite coatings are summarized in Fig. 8. The neat coating containing no nanoparticles presents a low scratch resistance as the 60° gloss retention is only 22 ± 1%. The positive effect of nanoalumina and nanosilica on scratch resistance of UV-waterborne nanocomposite coatings is clearly demonstrated as the 60° gloss retention increases upon the addition of nanoparticles. The incorporation of 5 wt% of nanoalumina and nanosilica leads to gloss retention of 50 and 91%, respectively.

Fig. 8
figure 8

Scratch resistance of dried/UV-cured waterborne nanocomposite coatings containing nanoalumina and nanosilica

Full size image

Overall, the scratch resistance of coatings containing nanoalumina is lower than that of coatings containing nanosilica. More than 200% of scratch resistance improvement could be reached using only 1 wt% of nanosilica. The surface modification of silica nanoparticles and the free volume change (see following T g data) are probably responsible for this behavior. As explained previously, the grafting of trialkoxysilanes onto silica nanoparticles provides a better dispersion of nanoparticles by reducing the number and the size of aggregates. As a result UV-waterborne nanocomposite coatings containing nanosilica have lower free volume and present better scratch resistance in comparison to UV-waterborne nanocomposite coatings containing nanoalumina. Results with regard to the positive effect of nanoparticles on scratch resistance are consistent with observations reported in literature especially in nanosilica case.27,28

Adhesion

Figure 9 shows results of adhesion strength of dried/UV-cured waterborne nanocomposite coatings as a function of nanoparticle content. The neat coating shows adhesion strength of 6.4 ± 0.4 MPa. Compared with the neat coating, the addition of alumina nanoparticles had no effect on adhesion as adhesion strength values are 7.0 ± 0.6, 6.4 ± 0.5, and 6.4 ± 0.5 MPa for 1, 3, and 5 wt% nanoalumina content, respectively. All adhesion strength values of UV-waterborne coatings containing nanosilica are higher than those of the neat coating. The addition of 1, 3, and 5 wt% of silica nanoparticles leads to a marked adhesion strength improvement around 36, 41, and 20%, respectively. It seems that the surface modification of silica nanoparticles which increases the number of acrylate functions and consequently the number of reactive groups is responsible for these results. The increase of acrylate double bond amount allows the improvement of the crosslink density which contributes to the increase of the adhesion of UV-waterborne coatings containing silica nanoparticles. Moreover the viscosity of formulations increases with nanoparticle content because of aggregation effect. Thus, it is possible that the addition of 5 wt% of nanosilica significantly increases the viscosity of the corresponding formulation. As a result the adhesion strength of the UV-waterborne nanocomposite coating containing 5 wt% of nanosilica is lower than that of UV-waterborne coatings containing 1 and 3 wt% of nanosilica.

Fig. 9
figure 9

Adhesion strength of sugar maple samples coated with dried/UV-cured waterborne nanocomposite coatings containing nanoalumina and nanosilica

Full size image

Among all adhesion theories existing, mechanical interlocking seems to be the more appropriate to explain observed results.29 This theory suggests that good adhesion is mainly due to the penetration of liquid coating in the pores of wood by capillarity and its subsequent solidification.30 Thus, adhesion depends not only on porosity properties of the wood substrate, but also on rheological properties of the liquid coating. Sugar maple (Acer saccharum) wood presents vessels with an average diameter between 40 and 60 μm.31 Despite the fact that the addition of nanoparticles probably leads to increase the viscosity of UV-waterborne coating formulations, the presence of nanoparticle aggregates does not seem to affect adhesion properties probably because they were small enough to penetrate in the pores of sugar maple wood.

It is interesting to note that no adhesion rupture was observed across the interface wood/coating during adhesion tests and wood fibers pulled off could be observed on dolly surface. This more suitable case reports a good adhesion between sugar maple wood and UV-waterborne coatings.1 However, coating residue could be present occasionally on the wood surface and/or dolly surface.

Effect of nanoparticle content on appearance properties of UV-waterborne coatings

Gloss

As can be seen in Fig. 10, the 60° gloss of dried/UV-cured waterborne nanocomposite coatings decreases with the increase of nanoparticle content as soon as 1 wt% of nanoparticles is introduced. Less than 40% of the neat coating gloss remains following the addition of 5 wt% of nanoparticles. The reduction of gloss is attributed to an enhancement of surface roughness due to the presence of the particles and aggregates. Such surface roughness leads to decrease the diffuse reflection of visible light.32 The residual 60° gloss of UV-waterborne coatings containing nanoalumina is lower than that of UV-waterborne coatings containing nanosilica, except at 1 wt% content. As the aggregation effect of nanoalumina is more important than that of nanoalumina, the increase of surface roughness of UV-waterborne coatings containing nanoalumina is more considerable. The residual 60° gloss of UV-waterborne coating containing 1 wt% of nanosilica is more important than that of UV-waterborne containing 1 wt% of nanoalumina. The matting effect of nanosilica is probably responsible for this trend. Indeed, silica is well known for its matting effect imparted to coatings upon its addition.33 Therefore, it is possible that the matting effect of nanosilica becomes predominant at lower loadings and that nanosilica aggregation effect becomes predominant at higher loadings.

Fig. 10
figure 10

Residual 60° gloss of dried/UV-cured waterborne nanocomposite coatings containing nanoalumina and nanosilica

Full size image

Effect of nanoparticle content on thermal properties of UV-waterborne coatings

Glass transition temperature

Thermal analyses were performed on dried/UV-cured waterborne nanocomposite coatings by means of differential scanning calorimetry. It is interesting to notice that a relaxation enthalpy peak could be observed during the first dynamic heating. As the UV-curing occurs with a fast rise in temperature, polymer chains reticulate quasi-immediately. With this exothermic heat of dried/UV-cured waterborne coatings, polymer chains reorganize themselves in order to adopt a configuration thermodynamically more stable. This thermal dilatation indicates an enthalpy relaxation. As a result a second dynamic heating is necessary to evaluate the glass transition of polymer coatings. The low slope of the glass transition reveals a high reticulation degree. Moreover, the low glass transition temperature (<100°C) is characteristic of thermoset polymers. All dried/UV-cured waterborne nanocomposite coatings studied follow similar thermal profile.

Glass transition temperature values of dried/UV-cured waterborne nanocomposite coatings are listed in Table 4. Whereas the neat coating presents a T g around 64°C, the addition of nanoparticles causes a rise of the glass transition temperature from 13 to 22%. This increase is due to interactions of nanoalumina and nanosilica with PUA matrix. More specifically, the addition of nanoparticles tends to increase the T g of nanocomposite coatings because of the hindering of macromolecular chains mobility at the interface around nanoparticles. According to DSC instrument resolution and sensitivity, it can be admitted that T g values of dried/UV-cured waterborne coatings containing 3 and 5 wt% of nanoalumina and nanosilica are similar to T g values of dried/UV-cured waterborne coatings containing 1 wt% of nanoalumina and nanosilica, respectively. It can be concluded that nanoparticle loading has no significant effect on glass transition temperature value.

Glass transition temperatures of dried/UV-cured waterborne nanocomposite coatings containing nanosilica are slightly higher than those of dried/UV-cured waterborne nanocomposite coatings containing nanoalumina. The presence of MEMO groups which increases the dispersion of nanosilica in acrylate media can explain this result. As T g is related to free volume in composite coatings, it is possible that dried/UV-cured waterborne nanocomposite coatings with nanosilica have lower free volume which results in higher glass transition temperatures.34

Conclusion

The effect of alumina and silica nanoparticles on the properties of UV-waterborne nanocomposite coatings was investigated. Mechanical (König hardness, scratch resistance, adhesion), optical (gloss), and thermal (glass transition temperature) properties were analyzed and compared to the neat coating. A good dispersion was expected to be the key to achieve suitable properties. TEM demonstrated that nanoparticle dispersion in UV-cured coatings was not efficient as aggregates larger than nanometer size were present. As a result the addition of nanoparticle affected the hardness of UV-waterborne nanocomposite coatings. Indeed, the surface roughness of UV-waterborne coatings increased following nanoparticle addition because of the presence of aggregates. However, the use of more powerful mixing method such as ultrasonication enabled the hardness property to increase. Despite the presence of nanoparticle aggregates, gloss retention analysis demonstrated a clear improvement of scratch resistance. Moreover, aggregates did not affect the adhesion between UV-waterborne coatings and Sugar maple (Acer saccharum) samples. As expected gloss of UV-waterborne coatings lower proportionally to nanoparticle content. DSC experiments revealed nanoparticles raised T g of nanocomposite coatings because of the hindering of macromolecular chains at the interface around the nanoparticles. Finally, results of UV-waterborne coatings containing nanosilica demonstrated the efficiency of surface modification. Not only did the amount and the size of aggregates decrease following grafting trialkoxysilanes on nanosilica, but also the number of acrylates functions and consequently the number of reactive groups increased. Indeed scratch resistance, adhesion and T g clearly increased as soon as 1 wt% of surface modified nanosilica were added. This work revealed the aptitude of alumina and silica nanoparticles to improve numerous properties despite the presence of aggregates. Excellent scratch resistance of UV-waterborne nanocomposite coatings was observed especially with nanosilica. This property essential to efficiently protect surface substrates such as wood is well suited in surface coating applications especially in the wooden furniture and kitchen cabinet manufacturing industry.