Preparation and evaluation of high-transparent scratch-resistant thin films on plasma treated polycarbonate substrate

2.2.1 Preparation of Si sol

The SiO₂ sol was prepared by mixing TEOS with water, then adding absolute ethanol. GPTMS was added under vigorous stirring. The pH of solution was adjusted by Acetic Acid at 2 by paper indicator (Acilit, Merck). The TEOS/GPTMS/H₂O/ethanol/Acetic Acid volume ratios were 6/4/30/10/15. The amount of water was 30 cc. The mixture was stirred for 2 h and then 0.17 g BPA was added. 2 ml MI was added just before applying the coating.

2.2.2 Preparation of Al sol

Hydrolysis and condensation reactions of the metal alkoxides are very fast and must be controlled. Aluminium tri-sec-butoxide has very different reaction kinetics and must be chelated by chelating agent. The Al sol was prepared by mixing ATB with EAcAc and stirred for 30 min to chelate, after stabilizing, added absolute ethanol. The ATB/EAcAc/H₂O/ethanol/HNO₃ volume ratios were 2/1/2/15/0.2. The amount of water was 15 cc.

2.2.3 Deposition of thin films

The series of the Si-Al nanocomposites were prepared by mixing of the as-prepared Si and Al sols in different volume ratios under vigorous stirring. Then the mixed sols were stirred for 48 h at ambient temperature to complete hydrolyze reaction. Table 1 shows the codes of prepared sols. Sols were mixed in 5 vol ratios (1:1, 2:1, 3:1, 4:1 and 5:1 Si to Al sols). After mixing the thin films were applied on PC substrate freshly and after 2, 4 and 6 days. Then the coated substrates were cured at 60, 80, 100 and 120 °C. Table 2 shows the sample codes for the Sample A1. The same conditions were conducted for all samples (A1-A5) and was not listed here because enlarged the manuscript. Sols were applied on pretreated substrates by dip-coating method with a withdrawal speed of 2 mm/sec. Substrates were kept in sols for 30 s to get equilibrium with the surrounding liquid sols. Thin films were dried at room temperature for 24 h, and then heated in an oven for 120 min.

3.2.1 XRD results

The GI-XRD pattern of thin film and XRD pattern of dried sols are shown in Fig. 6. The peaks are characterized on the patterns. It can be seen that the GI-XRD pattern, due to low film thickness, belongs to the PC substrate and the peaks come from the substrate. The only peak detected in Fig. 6 (a) at 2θ = 16.8° is attributed to the Bisphenol A component with reference code of 00–022-1755 that is arisen form substrate and indicates the amorphous nature of PC (Mohamed and Abdel-Kader, 2020). For this reason, the XRD pattern of the dried sol are taken. The XRD pattern shows there are some sharp peaks that are identified precisely. However, low intensity of the peaks suggestes that the nanocrystals formed and dispersed in an amorphous matrix. The amorphous nature of the prepared matrix is predictable regarding the synthesis method and applied low temperatures as reported by others (Rahoui et al., 2013; Sowntharya and Subasri, 2013).

Close

Fig. 6

X-ray patterns of: (a) GI-XRD pattern of the prepared thin film and (b) XRD pattern of dried A2 sol.

Export to PPT

The XRD pattern shown in Fig. 6(b) proves the formation of silicone oxide (SiO₂) with the hexagonal chemical structure of quartz. The suggested reference code for the identified quartz is 01–079-1915. The peaks emerged at 2θ = 22.8°, 26.5°, 28.5°, 39.7°, 41.9°, 43.0° and 46.3° are attributed to (1 0 0), (0 1 1), (1 0 1), (1 1 0), (0 1 2), (1 1 1) and (2 0 0) respectively. The corresponding peaks for silicon oxide hydrate known as silica gel with the chemical formula of H₂Si₂O₅ with the reference code of 00–020-1049 is identified. The emerged peaks at 2θ = 14.22°, 19.3°, 21.9°, 23.6°, 29.0°, 36.2° and 48.6° are attributed to silicon oxide hydrate. As well as, the compound phase including Si and Al atoms is detected. For example, aluminum silicate hydrate with the kaolinite mineral name and Al₂Si₂O₅(OH)₄ chemical formula is characterized. The reference code is 00–001-0527. The appeared peaks at 2θ = 12.6°, 19.7°, 20.7°, 21.2°, 23.9°, 37.3°, 38.4°, 49.2° and 60.2° are attributed to (0 0 1), (0 2 0), (1 1 0), (-1–11), (0 2 2), (0 2 5), (1–31), (-2–32) and ( −1 3 4) respectively.

3.2.2 TEM morphological investigation

Morphology and particle size analyses of the nanohybrid sol is conducted by TEM. TEM micrographs are took from liquid sol. The liquid sol of the composition of A2 (as mentioned in the Table 2) is attenuated by water and then used for taking micrographs. Sample preparation is done just by attenuating the liquid sol by water. Fig. 7 indicates TEM images of nanohybrid sol with different magnifications. It can be seen that the formed nanoparticles were uniform with diameters of ∼ 20 nm. These are some agglomerated nanoparticles that can be attributed to higher surface energy induced by hydroxyl groups (Abdollahi, 2014).

Close

Fig. 7

TEM images of sol–gel prepared nanohybrid A2 sol with different magnifications.

Export to PPT

3.2.3 ATR-FTIR results

ATR-FTIR method is used for studying the chemical compositions of the films. Fig. 8 shows the obtained spectra from dried sols at different aging time (fresh, 2, 4 and 6 days) to investigate the aging time effect on the films. The sols ratios effects is reported in our previous paper (Kiomarsipour, 2023). For more comparison the IR spectrum from dried TEOS + GPTMS sol without ATB is taken. The broad weak band presented at about 3200–3600 cm⁻¹ can be attributed to stretching vibration of —OH groups. The corresponding bending vibration is emerged at 1636 cm⁻¹. The C—H (CH₂) bonds of the GPTMS can absorb at 2939 cm⁻¹ and the observed band at 1280 cm⁻¹ can be assigned to the in-plane bending of C—H vibration. The carbonyl group C⚌O stretching vibration band is observed at 1716 cm⁻¹ (Chen et al., 2013). The stretching and symmetric bending of CH₃ groups bands are seen at 2874 cm⁻¹ and 1473 cm⁻¹ (Soloukhin, 2002). These C—H groups are from the glycidoxypropyl group of GPTMS. The Si—O—Si stretching vibration band at about 1040 cm⁻¹ proves the nanohybrid nature of the sols and occurrence of the condensation reaction between silanols. The band at 555 cm⁻¹ (generally in the range of 510–680 cm⁻¹) can be assigned to the stretching of Si—O—Si bonds (Handke and Kowalewska, 2011; Handke et al., 2011). It was reported that the epoxy ring in the GPTMS silane presented at 960–810 cm⁻¹ (asymmetrical ring stretching) (Liu and Berg, 2007). In TEOS + GPTMS spectrum the band of epoxy ring can be seen at around 960 cm⁻¹ but in other samples there is not. It is concluded that the ATB led to opening epoxy ring and catalyzed the ring opening reaction (Liu and Berg, 2007). As reported by Barrios et al. (Hernández-Barrios, 2017) changes occurred in the coatings after 28 days aging were intangible so it was reasonable that the changes in our coatings during 6 days are insignificant. The similarity of the spectra suggests the ability of obtaining suitable optical and mechanical properties within a wide range of chemical compositions.

Close

Fig. 8

ATR-FTIR spectra of dried TEOS + GPTMS and A24 at different aging times.

Export to PPT

3.2.4 UV/vis results

The photos of prepared sols and coatings are indicated in Fig. 9(a,b). The obtained sols were completely transparent and homogenous without phase separation and the deposited coatings were uniform, transparent and homogenous. The prepared sols remained unchanged after 18 months. The optical images of the films are shown in Fig. S1. It is clear that the coatings were free of cracks and delamination. The suitable ratios of inorganic/organic parts led to good adhesion and flexibility. The presence of GPTMS prevented cracking due to drying stresses and improved the films flexibility (Hajfarajzadeh et al., 2019). Regarding the epoxy group present in the GPTMS, it plays like the organic part in hybrid coating systems. The epoxy group promotes the coating adhesion, its flexibility and scratch resistance. In other words, organosilanes are an alternative to organic polymer binders in the hybrid coatings. These organo-alkoxysilanes precursors have organo-functional silane R’_n-Si(OR)_4-n, where R’ is an organic functional group (R’) such as amine, vinyl, epoxy, etc. The incorporation of the organic component R’ in the sol–gel coating systems allows to control the density and flexibility of the sol–gel network. This component leads to create thicker film without cracking (Perrin et al., 2020; Zandi-Zand et al., 2005; Chu et al., 1997; Hernández-Barrios, 2020; Robertson, 2003).

Close

Fig. 9

Optical photographs of: (a) transparent silica and aluminum sols and (b) dip-coated PC substrate.

Export to PPT

The transmittance spectra of the PC substrate and TEOS + GPTMS sol–gel coating are shown in Fig. 10(a). The transmittance spectra of the nanohybrid coatings containing Si and Al nanoparticles are indicated in Fig. 10(b-d) for different conditions (temperatures, ratios and aging times). It is clear that the average %T increased from 86 % for uncoated PC to 89 % for coated PC. The preparation of new structure of Si and Al atoms leads to less refractive index than SiO₂ structure. The differences between transmittance spectra are insignificant and they are similar together. Because the chemical composition through all of the samples were kept constant their transmittance spectra showed similar behavior.

Close

Fig. 10

Transmission spectra of the sol–gel films: (a) TEOS + GPTMS coating compared with raw PC, (b) sol–gel coatings prepared at different temperatures, (c) sol–gel coatings with different sol ratios and (d) sol–gel coatings prepared at different times.

Export to PPT

Close

Fig. 10

Transmission spectra of the sol–gel films: (a) TEOS + GPTMS coating compared with raw PC, (b) sol–gel coatings prepared at different temperatures, (c) sol–gel coatings with different sol ratios and (d) sol–gel coatings prepared at different times.

Export to PPT

In order to increase the transparency, the refractive index of the coating must be less than that of the PC substrate (Chen et al., 2013). The refractive index curve versus wavelength of 4PA24 sample is indicated in Fig. 11. It is clear that refractive index of sol–gel coating was 1.481 at 775 nm and the index of PC substrate was 1.58. The enhancement of optical transparency of the nanohybrid coatings emanated of the combined effects of Si and Al atoms resulting in low refractive index (Li, 2015). This fact finally led to decrease the reflection of the incident light and increase the transmittance.

Close

Fig. 11

Refractive index (n) versus wavelength of 4PA24 sample.

Export to PPT

3.2.5 FE-SEM and optical micrographs

The optical micrographs of the films after 10 strokes are shown in Fig. 12. The images belong to various coating compositions in order to show that all of the prepared coatings have the ability to improve the PC scratch resistance. The scratch marks were clear on the bare PC while there were a few marks on the coatings.

Close

Fig. 12

Optical micrographs of the sol–gel prepared films after 10 strokes of rubbing eraser: (a) TEOS + GPTMS coating, (b) A14. (c) 4PA24, (d) 2PA32.

Export to PPT

The FE-SEM images and the EDS analysis result of the sol–gel coating are indicated in Fig. 13. Fig. 13(a-b) shows that the samples were smooth, uniform without any agglomerates, cracks, holes and delamination. The abrasion tracks of eraser strokes can be seen clearly on the uncoated PC substrate (Fig. 13(c)) while the coated area remained undamaged. It can be found that the sol–gel coatings improved the PC scratch resistance and enabled the substrate to endure scratch damages. Fig. 13(d) indicates the EDS analysis result. The presence of aluminum and silicon elements in the films are clear. It is clear that the scratch resistance improved significantly for each composition. Fig. 13(e-i) shows series of SEM elemental mapping images of the as-synthesized samples proving that Si, Al, O and C atoms were evenly dispersed and distributed throughout the composite samples. The uniform dispersion of these elements suggested that the nanoparticles were incorporated successful into the nanohybrid thin films. The sol–gel prepared coatings included well dispersed of all atoms that interlocked throughout the film uniformly.

Close

Fig. 13

FE-SEM micrographs of 4PA24 coating: (a,b) before scratch test, (c) after 10 eraser strokes and (d) the EDS elemental pattern, (e) C mapping, (f) Si mapping, (g) Al mapping, (h) O mapping and (i) their combination.

Export to PPT

Close

Fig. 13

FE-SEM micrographs of 4PA24 coating: (a,b) before scratch test, (c) after 10 eraser strokes and (d) the EDS elemental pattern, (e) C mapping, (f) Si mapping, (g) Al mapping, (h) O mapping and (i) their combination.

Export to PPT

It has been shown that the tribological performance improved by plasma treatment. Because the sol–gel films had ceramic nature and the ceramic-based oxide phase causes to decrease shear stresses and the adhesion between two contacting surfaces maybe lead to better abrasion properties (Tekdir and Yetim, 2021; Turalıoğlu, 2021).

3.2.6 Adhesion results

Adhesion test was conducted by cross cut method that a pressure sensitive tape was applied on the cross pattern on the coating. It was clear that the films had good adhesion (5B) despite coatings aged for 6 days. Their adhesion reduced to 3B-4B due to more cross linking occurred in the coating. The good adhesion strength resulted by DBD treatment of the PC. Fig. S2 shows optical micrographs after the adhesion test.

Tiringer et al. (Tiringer, 2021) studied the effect of chemistry, roughness and wettability of the different pre-treated surfaces on the adhesion of a hybrid sol–gel coating by measuring the pull-off strengths. They used four different substrates pretreatments (bare, HNO₃, KOH and boiling water). They showed that the highest pull-off strength was obtained after boiling DI water pretreatment, that can be related to the formation of a pseudoboehmite layer. They found that more surface roughness led to more adhesion strength (Robertson, 2003).

3.2.7 Pencil hardness results

Pencil hardness test is a good measurement for evaluating coating's scratch resistance under heavy load. Describing the scratch mechanisms of a thin film on a soft substrate under a pencil lead is a complicated issue. In order to evaluate the scratch resistance, pencil hardness tests was carried out according to the ASTM D3363 standard. The optical micrographs of the pencil scratch are shown in Fig. S3. The pencil hardness is the grade of the pencil that does not leave a visible scratch mark on the coating. The pencil hardness of the PC substrates (4B) increased to 3H for the sample 4PA24. Improvement the PC hardness from 4B to 3H (8 grade increment) was very incredible. This pencil hardness increment obtained with mono-layer coating with thickness of about 775 nm. One of the most effective factors on the pencil hardness is the film thickness. Increasing coating thickness leads to increase the pencil hardness. The previous reports showed the pencil hardness lower than 1H in the thin films (Chen, 2008). In acrylates reinforced by colloidal silica hardness of 5/6H was obtained by 50 µm thickness (Barletta, 2016) and 3H with 15 µm thickness by the nanohybrid coatings on PC (Sowntharya, 2012).

All the hybrid coatings deposited on the PC exhibited the pencil hardness of HB at thickness of 1.7 µm (Suriano, 2017) and H for colloidal silica content of 36.0 vol% with thickness within the range of 5 ± 0.5 μm (Chen, 2008). The incorporation of Al₂O₃ and ZrO₂ nanoparticles in the TEOS-GPTMS sol system improved the pencil hardness of coatings from 2H (pristine PC) to 2H at 5 µm coating thickness (Chantarachindawong, 2016) and up to 3H with 19.6 µm coating thickness (Shin, 2009). The pencil hardness in nanohybrid coatings made of titanium tetraisopropoxide and an acrylic modified silane was 3H at 7.2 µm thickness (Sowntharya and Subasri, 2013). It seems that the boehmite nanoparticles led to a higher hardness and scratch resistance in the sol–gel prepared coatings on the PC (Schottner et al., 2003).

3.2.8 Mechanisms of coating failure

Fig. S4 shows the optical micrographs of the typical pencil scratch marks with 5H pencil. It can be seen that the coating cracking was the dominant mode of the failure. It can be concluded from the fracture patterns that coating damage was caused by tensile cracking behind the indenter. There was little or no coating detachment at either the leading edge or the trailing end of the scratch, indicating that the adhesion between the film and the substrate was very strong due to the plasma activated PC surface.

The failure mechanisms of films have been completely expressed by Wu et al. (Wu, 2008). After applying hard pencil, the static pressure imposed by sharp narrow pencil tip on the small contact area caused to gouge creation. Movement of the hard pencil lead on the film led to the creation of the cylindrical scratch mark. The gouge failure caused by the hardness of the coating and the strength of the substrate. Micro-cracks are present on the pencil trail both inside and outside of the impression. The film failure results by film cracking and substrate deformation. In gouge failure mechanism the film endures tensile stress while the PC polymeric substrate undergoes a plastic deformation. The plastic deformation starts as soon as the pressure imposed by pencil tip exceeds the substrate yield strength and consequently a gouge failure is formed on the surface coating. The high vertical pressure exerted by pencil lead led to plastic deformation of substrate at the front edge. Behind the pencil tip, the film undergoes tensile stress that is produced by the friction effect. Coatings will be cracked if this tensile stress exceeds the strength of the coating material. In the meantime, of scratch formation, the tensile stress created in the film outside and near the scratch impression by substrate plastic deformation. The good adhesion strength between the films and plasma treated PC substrates led to no film delamination (Wu, 2008). It must be noticed that the pencil hardness grade is similar in thin coatings (≥5 μm thickness) and the improvement of pencil hardness is very limited (Wu, 2008).