Structural Coloration of Polyester Fabrics Coated with Al/TiO2 Composite Films and Their Anti-Ultraviolet Properties

Al/TiO2 composite film was successfully deposited on polyester fabrics by using magnetron sputtering techniques. X-ray photoelectron spectroscopy (XPS) and X-ray diffraction (XRD) were used to examine the deposited films on the fabrics, and the structural colors and anti-ultraviolet property of fabrics were also analyzed. The results indicated that polyester fabrics coated with Al/TiO2 composite films achieved structural colors. The reactive sputtering times of TiO2 films in Al/TiO2 composite films were 10 min, 12 min, 18 min, 20 min, 26 min, 27 min, 30 min and 45 min, respectively, the colors of corresponding fabrics were bluish violet, blue, cyan, green, yellow, yellowish red, orange and blue-green, which was consistent with the principle of the thin film interference. The structure of the TiO2 film in Al/TiO2 composite films was non-crystalline, though the fabrics were heated and maintained at the temperature of 200 °C. The anti-ultraviolet property of the fabrics deposited with Al/TiO2 composite films were excellent because of the effect of Al/TiO2 composite films.

The fastness of composite films deposited on polyester fabrics could be expressed by color fastness to washing, which was tested according to GB/T 3921-2008 [16] "textile test for color fastness to washing with soap or soap and soda".

Anti-Ultraviolet Property Test
According to GB/T18830-2009 [17], anti-ultraviolet properties of the samples were tested by ultraviolet transmittance analyzer (UV-1000F, Lapsphere, USA).The evaluation indexes of anti-ultraviolet properties included solar ultraviolet transmittance T (UVA) and T (UVB) and ultraviolet protection factor (UPF), solar UV-A spectral transmittance T (UVA), solar UV-B spectral transmittance T (UVB) and ultraviolet protection factor (UPF). Each sample was tested five times, and the average values were reported.

XPS Analysis
All samples coated with Al/TiO 2 composite films differ only in the thickness of TiO 2 film. Therefore, the XPS test results should be similar, and No. 4 sample was selected as a representative to analyze its chemical composition and valence state. Figure 1 presents the photoelectron spectroscopy of No. 4 sample. The fastness of composite films deposited on polyester fabrics could be expressed by color fastness to washing, which was tested according to GB/T 3921-2008 [16] "textile test for color fastness to washing with soap or soap and soda".

Anti-Ultraviolet Property Test
According to GB/T18830-2009 [17], anti-ultraviolet properties of the samples were tested by ultraviolet transmittance analyzer (UV-1000F, Lapsphere, USA).The evaluation indexes of antiultraviolet properties included solar ultraviolet transmittance T (UVA) and T (UVB) and ultraviolet protection factor (UPF), solar UV-A spectral transmittance T (UVA), solar UV-B spectral transmittance T (UVB) and ultraviolet protection factor (UPF). Each sample was tested five times, and the average values were reported.

XPS Analysis
All samples coated with Al/TiO2 composite films differ only in the thickness of TiO2 film. Therefore, the XPS test results should be similar, and No. 4 sample was selected as a representative to analyze its chemical composition and valence state. Figure 1 presents the photoelectron spectroscopy of No. 4 sample. In Figure 1a, only the characteristic peaks of Al, Ti, and O elements are presented in the XPS full spectrum of No. 4 sample, and there was also a handful of C, which is mainly derived from the carbon in the ultra-high vacuum chamber in the XPS equipment. Figure 1b shows the Al2p peak of Al, and it can be seen that the position of Al2p peak was 73.74 eV, which is approximately the same as the binding energy of metal Al, suggesting that Al was present in the composite films in the form of metallic aluminum [18,19]. As can be seen from Figure 1c, the positions of Ti2p1/2 and Ti2p3/2 peak  In Figure 1a, only the characteristic peaks of Al, Ti, and O elements are presented in the XPS full spectrum of No. 4 sample, and there was also a handful of C, which is mainly derived from the carbon in the ultra-high vacuum chamber in the XPS equipment. Figure 1b shows the Al2p peak of Al, and it can be seen that the position of Al2p peak was 73.74 eV, which is approximately the same as the binding energy of metal Al, suggesting that Al was present in the composite films in the form of metallic aluminum [18,19]. As can be seen from Figure 1c, the positions of Ti2p 1/2 and Ti2p 3/2 peak were 464.00 eV and 458.28 eV, respectively, and the results were consistent with the TiO 2 binding energy. It indicates that Ti was completely oxidized, and was presented in TiO 2 form [20,21]. Figure 1d shows the O1s peak of the O element. It can be seen from Figure 1d that the O1s peak position of O element was located at 530.07 eV, and the result is consistent with the binding energy of O in TiO 2 [22,23]. It also indicates that Ti existed in the composite film in the form of TiO 2 .
Consequently, the XPS analysis indicates that the composite films deposited on the polyester fabric surface were Al/TiO 2 composite films.

XRD Analysis
Due to Al/TiO 2 composite films were prepared by magnetron sputtering at room temperature, the TiO 2 film in the composite films was difficult to form a crystal structure. In general, TiO 2 film was crystallized by sputtering at high temperature or calcination after sputtering [24].
Without affecting the properties of the textile substrate, the temperature of the textile fabric substrate was raised during reactive sputtering so as to crystallize the TiO 2 films in the Al/TiO 2 composite films. Since the experimental substrate was polyester fabrics, too high a temperature would affect the fabric dimensional stability and performance. Therefore, the fabric substrate was heated and maintained the temperature at 100 • C and 200 • C respectively during RF reactive sputtering. Taking No. 7 as an example, the XRD test was performed when the temperature of the textile substrate was normal temperature, 100 • C, and 200 • C, respectively. The experimental results are shown in Figure 2. were 464.00 eV and 458.28 eV, respectively, and the results were consistent with the TiO2 binding energy. It indicates that Ti was completely oxidized, and was presented in TiO2 form [20,21]. Figure  1d shows the O1s peak of the O element. It can be seen from Figure 1d that the O1s peak position of O element was located at 530.07 eV, and the result is consistent with the binding energy of O in TiO2 [22,23]. It also indicates that Ti existed in the composite film in the form of TiO2. Consequently, the XPS analysis indicates that the composite films deposited on the polyester fabric surface were Al/TiO2 composite films.

XRD Analysis
Due to Al/TiO2 composite films were prepared by magnetron sputtering at room temperature, the TiO2 film in the composite films was difficult to form a crystal structure. In general, TiO2 film was crystallized by sputtering at high temperature or calcination after sputtering [24].
Without affecting the properties of the textile substrate, the temperature of the textile fabric substrate was raised during reactive sputtering so as to crystallize the TiO2 films in the Al/TiO2 composite films. Since the experimental substrate was polyester fabrics, too high a temperature would affect the fabric dimensional stability and performance. Therefore, the fabric substrate was heated and maintained the temperature at 100 °C and 200 °C respectively during RF reactive sputtering. Taking No. 7 as an example, the XRD test was performed when the temperature of the textile substrate was normal temperature, 100 °C, and 200 °C, respectively. The experimental results are shown in Figure 2. It can be seen from Figure 2 that the XRD patterns of the Al/TiO2 composite films deposited on polyester fabrics prepared by reactive sputtered TiO2 film at room temperature, 100 °C and 200 °C look similar. There were only the characteristic peaks of polyester fabric substrate, without the characteristic peaks of the crystalline structures of Al and TiO2, indicating that the structures of Al and TiO2 all were amorphous structures.
Normally, the metal element was easy to form a crystal structure, but the Al film in the Al/TiO2 composite film was not crystallized here. The main reason was that the preparation of the Al film used RF sputtering, and the deposition rate was very low, which affected the ordered arrangement of atoms, resulting in the amorphous structures structure of Al. The type of anatase crystal of TiO2 film was generally obtained for at temperatures between 350 °C and 500 °C, and the type of rutile crystal was formed when the temperature exceeded 500 °C [25,26]. Considering the influence of polyester fabric substrate, the temperature was only added to 200 °C, resulting in a non-crystalline structure of TiO2 film. It can be seen from Figure 2 that the XRD patterns of the Al/TiO 2 composite films deposited on polyester fabrics prepared by reactive sputtered TiO 2 film at room temperature, 100 • C and 200 • C look similar. There were only the characteristic peaks of polyester fabric substrate, without the characteristic peaks of the crystalline structures of Al and TiO 2 , indicating that the structures of Al and TiO 2 all were amorphous structures.
Normally, the metal element was easy to form a crystal structure, but the Al film in the Al/TiO 2 composite film was not crystallized here. The main reason was that the preparation of the Al film used RF sputtering, and the deposition rate was very low, which affected the ordered arrangement of atoms, resulting in the amorphous structures structure of Al. The type of anatase crystal of TiO 2 film was generally obtained for at temperatures between 350 • C and 500 • C, and the type of rutile crystal was formed when the temperature exceeded 500 • C [25,26]. Considering the influence of polyester fabric substrate, the temperature was only added to 200 • C, resulting in a non-crystalline structure of TiO 2 film.
To summarize, the XRD patterns of Al/TiO 2 composite films deposited on polyester fabrics prepared by reactive sputtering of TiO 2 film at room temperature, 100 • C and 200 • C indicated that both Al and TiO 2 in the Al/TiO 2 composite films had non-crystalline structures. TiO 2 film cannot form a crystal structure at the temperature of 200 • C, but 200 • C was the limit temperature for heating the polyester fabrics, while the performance of polyester fabrics were affected by further heating.

Structural Color Analysis
The color photos of the original fabric and the polyester fabrics coated with Al/TiO 2 composite film are shown in Figure 3. These photos were formed by selecting 2 cm from the central of samples and then enlarging. To summarize, the XRD patterns of Al/TiO2 composite films deposited on polyester fabrics prepared by reactive sputtering of TiO2 film at room temperature, 100 °C and 200 °C indicated that both Al and TiO2 in the Al/TiO2 composite films had non-crystalline structures. TiO2 film cannot form a crystal structure at the temperature of 200 °C, but 200 °C was the limit temperature for heating the polyester fabrics, while the performance of polyester fabrics were affected by further heating.

Structural Color Analysis
The color photos of the original fabric and the polyester fabrics coated with Al/TiO2 composite film are shown in Figure 3. These photos were formed by selecting 2 cm from the central of samples and then enlarging.  It can be clearly seen from Figure 3 that the thickness of the TiO2 film in Al/TiO2 composite films was different and the colors of the samples were also different. The original white fabric presented a variety of structural colors because their surfaces were coated with Al/TiO2 composite films. As the thickness of TiO2 films in Al/TiO2 composite films increased, the structural colors changed from purple, blue, cyan, green, yellow, orange to red, respectively. In this experiment, the thickness of the TiO2 film was mainly controlled by the reactive sputtering time. The reactive sputtering times of TiO2 films for Nos. 1-8 samples were 10 min, 12 min, 18 min, 20 min, 26 min, 27 min, 30 min and 45 min, respectively, and the corresponding fabric colors were bluish violet, blue, cyan, green, yellow, yellowish red, orange and blue-green. The results indicated that the corresponding wavelength of the samples was proportional to the thickness of the TiO2 film, which was consistent with the thin film interference theory. Figure 4 reveals the corresponding reflection spectra of samples coated with different thickness of TiO2 thin films in Al/TiO2 composite film. It can be clearly seen from Figure 3 that the thickness of the TiO 2 film in Al/TiO 2 composite films was different and the colors of the samples were also different. The original white fabric presented a variety of structural colors because their surfaces were coated with Al/TiO 2 composite films. As the thickness of TiO 2 films in Al/TiO 2 composite films increased, the structural colors changed from purple, blue, cyan, green, yellow, orange to red, respectively. In this experiment, the thickness of the TiO 2 film was mainly controlled by the reactive sputtering time. The reactive sputtering times of TiO 2 films for Nos. 1-8 samples were 10 min, 12 min, 18 min, 20 min, 26 min, 27 min, 30 min and 45 min, respectively, and the corresponding fabric colors were bluish violet, blue, cyan, green, yellow, yellowish red, orange and blue-green. The results indicated that the corresponding wavelength of the samples was proportional to the thickness of the TiO 2 film, which was consistent with the thin film interference theory. Figure 4 reveals the corresponding reflection spectra of samples coated with different thickness of TiO 2 thin films in Al/TiO 2 composite film. The spectral reflectance curves of Nos. 1 and 2 samples were close to each other, but they were slightly offset. The maximum reflection wavelength was approximate 400 nm, which is in the purple and blue wavelength range. Therefore, the colors of Nos. 1 and 2 samples were determined to be bluish violet and blue, consistent with the fabric colors presented in Figure 3b,c. The center wavelength of the reflectivity curve of the No. 3 sample was located at 420 nm, which is in the blue wavelength range. Therefore, the No. 3 sample color was cyan, which is also consistent with the fabric color in Figure 3d. The maximum reflection wavelength of No. 4 sample was 510 nm, which is in the green wavelength range, and was therefore consistent with the fabric color in Figure 3e. The maximum reflection wavelengths of No. 5 sample, No. 6 sample and No. 7 sample were 600 nm, 650 nm, and 690 nm, respectively corresponding to yellow, yellowish red, orange color, and consistent with the results in Figure 3f,g,h. The maximum reflection wavelength of No. 8 sample was 480 nm, corresponding to blue-green color, and consistent with the result in Figure 3i. L*, a*, b* and C* values for samples are shown in Table 1.     The spectral reflectance curves of Nos. 1 and 2 samples were close to each other, but they were slightly offset. The maximum reflection wavelength was approximate 400 nm, which is in the purple and blue wavelength range. Therefore, the colors of Nos. 1 and 2 samples were determined to be bluish violet and blue, consistent with the fabric colors presented in Figure 3b,c. The center wavelength of the reflectivity curve of the No. 3 sample was located at 420 nm, which is in the blue wavelength range. Therefore, the No. 3 sample color was cyan, which is also consistent with the fabric color in Figure 3d. The maximum reflection wavelength of No. 4 sample was 510 nm, which is in the green wavelength range, and was therefore consistent with the fabric color in Figure 3e. The maximum reflection wavelengths of No. 5 sample, No. 6 sample and No. 7 sample were 600 nm, 650 nm, and 690 nm, respectively corresponding to yellow, yellowish red, orange color, and consistent with the results in Figure 3f,g,h. The maximum reflection wavelength of No. 8 sample was 480 nm, corresponding to blue-green color, and consistent with the result in Figure 3i. L*, a*, b* and C* values for samples are shown in Table 1. From Table 1, it can be seen that the L* values of Nos. 1-8 samples were quite different, indicating that the lightness of the samples was significantly distinct. The chroma C* values of Nos. 1-8 samples differed greatly, with the minimum chroma value of No.4 and the maximum chroma value of No. 2 sample, also manifesting that the color and purity of the samples are different. The a* and b* values of Nos. 1-8 samples were different, indicating that the samples colors were in different positions in the color space. Figure 5 represents the distribution of chromaticity indices, a* and b*, for all the samples coated with Al/TiO 2 composite films. As shown in Figure 5, a* and b* values of Nos. 1-8 samples coated with Al/TiO2 composite films were different in the color space. According to the CIE 1976 chromaticity diagram, the colors of these samples were bluish violet, blue, cyan, green, yellow, yellowish red, orange and blue-green. The results are consistent with the experimental results shown in Figures 3 and 4.
As a consequence, polyester fabrics coated with Al/TiO2 composite films presented structural colors resulted from the effect of nano-composite films, and the colors varied along with the change of the TiO2 film thickness. With the increase of the TiO2 film thickness, the colors of fabrics changed from purple, blue, cyan, green, yellow, orange to red. When the thicknesses of the TiO2 films continued to increase, the colors of the fabrics changed regularly followed by these seven colors, and the results were consistent with the rules of thin film interference. In this experiment, the reactive sputtering times of TiO2 films for the samples were 10 min, 12 min, 18 min, 20 min, 26 min, 27 min, 30 min and 45 min, respectively, and the corresponding fabric colors were bluish violet, blue, cyan, green, yellow, yellowish red, orange and blue-green.
Compared with the polyester fabrics coated with Ag/TiO2 composite film samples, the color rules of fabrics coated with nano-composite films were the same, though the underlying metal in composite films was different. The corresponding wavelength of colors was linearly proportional to the thickness of TiO2 films [27].
Due to the different refractive indexes of metallic Ag and Al, the sputtering efficiency of the TiO2 film was affected, resulting in the difference between the sputtering time and the corresponding wavelength of the fabric color. For example, the reactive sputtering times of the TiO2 films in Ag/TiO2 composite films deposited on polyester fabrics were 1 min, 3 min, and 4 min, correspondingly, the reactive sputtering times of the TiO2 films in Al/TiO2 composite films deposited on polyester fabrics were 12 min, 20 min, and 26 min, respectively, the corresponding fabrics colors were blue, green, and yellow [25]. As a result, the deposition efficiency of the TiO2 films in Al/TiO2 composite films was lower than that in Ag/TiO2 composite films.
According to the experimental results, the color fastness to washing of polyester fabrics coated with Al/TiO2 composite films was 5, so the color fastness to washing of samples was very good. It indicated that Al/TiO2 composite films and polyester fabrics were combined strongly, and the composite films were firm, and not easy to fall off. As shown in Figure 5, a* and b* values of Nos. 1-8 samples coated with Al/TiO 2 composite films were different in the color space. According to the CIE 1976 chromaticity diagram, the colors of these samples were bluish violet, blue, cyan, green, yellow, yellowish red, orange and blue-green. The results are consistent with the experimental results shown in Figures 3 and 4.
As a consequence, polyester fabrics coated with Al/TiO 2 composite films presented structural colors resulted from the effect of nano-composite films, and the colors varied along with the change of the TiO 2 film thickness. With the increase of the TiO 2 film thickness, the colors of fabrics changed from purple, blue, cyan, green, yellow, orange to red. When the thicknesses of the TiO 2 films continued to increase, the colors of the fabrics changed regularly followed by these seven colors, and the results were consistent with the rules of thin film interference. In this experiment, the reactive sputtering times of TiO 2 films for the samples were 10 min, 12 min, 18 min, 20 min, 26 min, 27 min, 30 min and 45 min, respectively, and the corresponding fabric colors were bluish violet, blue, cyan, green, yellow, yellowish red, orange and blue-green.
Compared with the polyester fabrics coated with Ag/TiO 2 composite film samples, the color rules of fabrics coated with nano-composite films were the same, though the underlying metal in composite films was different. The corresponding wavelength of colors was linearly proportional to the thickness of TiO 2 films [27].
Due to the different refractive indexes of metallic Ag and Al, the sputtering efficiency of the TiO 2 film was affected, resulting in the difference between the sputtering time and the corresponding wavelength of the fabric color. For example, the reactive sputtering times of the TiO 2 films in Ag/TiO 2 composite films deposited on polyester fabrics were 1 min, 3 min, and 4 min, correspondingly, the reactive sputtering times of the TiO 2 films in Al/TiO 2 composite films deposited on polyester fabrics were 12 min, 20 min, and 26 min, respectively, the corresponding fabrics colors were blue, green, and yellow [25]. As a result, the deposition efficiency of the TiO 2 films in Al/TiO 2 composite films was lower than that in Ag/TiO 2 composite films.
According to the experimental results, the color fastness to washing of polyester fabrics coated with Al/TiO 2 composite films was 5, so the color fastness to washing of samples was very good. It indicated that Al/TiO 2 composite films and polyester fabrics were combined strongly, and the composite films were firm, and not easy to fall off. Table 2 shows the anti-ultraviolet properties of Nos. 1-8 samples coated with Al/TiO 2 composite films. The smaller the values of T (UVA) and T (UVB), the anti-ultraviolet property is better. Conversely, the greater the ultraviolet protection factor (UPF), the anti-ultraviolet property is higher. As can be seen from Table 2, the results of anti-ultraviolet property for Nos. 1-8 samples were similar, and T (UVA) and T (UVB) were less than 4%, and UPF was more than 30.

Anti-Ultraviolet Property Analysis
According to the national standard GB/T18830-2002 [28], when UPF > 30 and T (UVA) < 5%, the samples can be called anti-UV textiles. It can be seen that all the polyester fabrics coated with Al/TiO 2 composite films meet the requirements and belong to the UV protection product. The main reason was that the high reflectivity of the Al film reflects ultraviolet light, and the TiO 2 film in the Al/TiO 2 composite film also has the effect of UV protection. Duo to high refractive and high photoactivity, nano-TiO 2 , as an excellent ultraviolet protection agent, can absorb ultraviolet ray, and reflect and scatter ultraviolet rays, and transmit visible light.
In summary, the anti-ultraviolet property of polyester fabrics coated with Al/TiO 2 composite films can be greatly improved resulting from the influence of Al/TiO 2 composite films.

Conclusions
Al/TiO 2 composite film was successfully deposited on polyester fabrics using the magnetron sputtering technique. Due to the effect of nano-composite films, polyester fabrics coated with Al/TiO 2 composite films achieves structure coloration. The reactive sputtering times of TiO 2 films in Al/TiO 2 composite films were 10 min, 12 min, 18 min, 20 min, 26 min, 27 min, 30 min and 45 min, respectively, the colors of corresponding fabrics were bluish violet, blue, cyan, green, yellow, yellowish red, orange and blue-green, which was consistent with the rules of the thin film interference. XPS analysis shows that the composited films deposited on the surface of the polyester fabrics substrate were Al/TiO 2 composite films. The XRD patterns of Al/TiO 2 composite films deposited on polyester fabrics prepared by reactive sputtering of TiO 2 film at room temperature, 100 • C and 200 • C indicated that the structures of Al films and TiO 2 films in Al/TiO 2 composite films were non-crytalline structures. The anti-ultraviolet property of polyester fabrics coated with Al/TiO 2 composite films were all excellent resulting from the influence of Al/TiO 2 composite films.

Conflicts of Interest:
The authors declare no conflict of interest.