Bio-Medical Materials and Engineering 24 (2014) 651–657 DOI 10.3233/BME-130853 IOS Press
651
Computational and experimental determinations of the UV adsorption of polyvinylsilsesquioxane-silica and titanium dioxide hybrids Haiyan Wanga,*, Derong Lin a , Di Wangb , Lijiang Hua,*, Yudong Huang c, Li Liu c and Douglas A. Loy d a
School of Science, Harbin Institute of Technology, Harbin, 150001, China Key Laboratory of Bio-based Material Science and Technology, Northeast Forestry University, Harbin 150040, China c School of Chemical Engineering, Harbin Institute of Technology, Harbin, 150001, China d Department of Chemistry and Biochemistry, The University of Arizona, Tucson, AZ 85721, USA b
Abstract. Sunscreens that absorb UV light without photodegradation could reduce skin cancer. Polyvinyl silsesquioxanes are known to have greater thermal and photochemical stability than organic compounds, such as those in sunscreens. This paper evaluates the UV transparency of vinyl silsesquioxanes (VS) and its hybrids with SiO2(VSTE) and TiO2(VSTT) experimentally and computationally. Based on films of VS prepared by sol-gel polymerization, using benzoyl peroxide as an initiator, vinyltrimethoxysilane (VMS) formulated oligomer through thermal curing .Similarly, VSTE films were prepared from VMS and 5-25 wt-% tetraethoxysilane (TEOS) and VSTT films were prepared from VMS and 5-25 wt-% titanium tetrabutoxide (TTB). Experimental average transparencies of the modified films were found to be about 9-14% between 280320 nm, 67-73% between 320-350nm, and 86-89% between 350-400nm. Computation of the band gap was absorption edges for the hybrids in excellent agreement with experimental data. VS, VSTE and VSTT showed good absorption in UV-C and UV-B range, but absorbed virtually no UV-A. Addition of SiO2 or TiO2 does not improve UV-B absorption, but on the opposite increases transparency of thin films to UV. This increase was validated with molecular simulations. Results show computational design can predict better sunscreens and reduce the effort of creating sunscreens that are capable of absorbing more UV-B and UV-A. Keywords: Silsesquioxane, silica, titanium dioxide, hybrid films, ultraviolet region, sunscreen, computation of band gap
1. Background Ultraviolet radiation can be divided into categories based on wavelength: 1) UV-C (100-280 nm); 2) UV-B (280-320 nm); 3) UV-A (320-400 nm). UV-C radiation is effectively absorbed before reaching the surface of earth. Both UV-A and UV-B pass through the atmosphere and become attributable to aging skin disorders [1], and lower immunity against infection [2] and cancer [3]. Silsesquioxane (SSO) materials have long been used as protective coatings [4-8], but only recently have they been *Corresponding author. E-mail:
[email protected];
[email protected] 0959-2989/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved
652
H. Wang et al. / Computational and experimental determinations of the UV adsorption
used to block ultraviolet radiation [9-15]. Work with inorganic-modified SSO as scratch resistant, transparent coatings for optics has provided undeniable precedent that SSOs and their hybrids can be transparent to visible wavelength of light [16, 17]. In a previous paper by some of our authors, it was shown through computational and experimental methods that SSOs with tetraethoxysilane (TEOS or SiO2) and titanium tetrabutoxide (TTBO or TiO2), were transparent in the visible and infrared regions [18]. More recently, they showed that cured vinyl silsesquioxanes (VS) and hybrids with various amounts of silica could serve as a UV absorbing agent to protect plants in greenhouses from UV without attenuating the desirable visible wavelengths needed for photosynthesis [9]. 2. Objective In this study, we will extend the experimental and computational evaluations of films based on VS modified with TTBO (VSTT) to the ultraviolet wavelengths and compare their performance to films of cured VS and cured VS hybrids modified with TEOS (VSTE) examined in the previous study [9]. The experimental objectives were to prepare a set of hybrid thin films made by condensing VS with varying amounts of TTBO or TEOS and measure the transparency of the hybrid films to see how the addition of TTBO or TEOS affected the absorption of UV-A and UV-B. The computational objectives were to construct reasonable models for the hybrids that would permit the generations of accurate density of states (DOS) and electronic orbits, and predict the absorption edge in the ultraviolet region. Experimental and methods 2.1. Materials VMS was purchased from Sigma-Aldrich Corporation (St. Louis, Missouri) and used as received. TEOS, TTBO, formic acid (88 wt%), ethanol (purity 99.7%) and benzoyl peroxide (BPO) were purchased from Tianjin BoDi Chemicals Corporation (Tianjin, China) and used as received. 2.2. Preparation of VMS oligomer Solvent, "water-free" hydrolysis and condensation of VMS (0.1 mol) were carried out in ethanol (13.8 g) in beakers (100 mL) with formic acid (0.3 mmol) in a molar ratio HCOOH/Si=3. The reaction was performed in three stages, all at 35 °C in a water bath: 1) Three days in beakers sealed with polyethylene plastic wrap secured with a rubber band; 2) Three days in beakers with several small needle-sized holes made on the plastic wrap to allow water vapor to enter and solvent to evaporate; 3) Four days with the beakers opened in the air. The product was the transparent and viscous oil. The hydrolytic condensations of 5, 15, and 25 wt-% TEOS (or TTBO) were carried out in beakers placed in a water bath. Ethanol was used as solvent in a 3:1 molar ratio with Si (or Ti). The beaker was sealed with a plastic film in which the substances first reacted for a day at 40 ºC. Then needle-size holes were punctured in the plastic film and the reaction was left to continue for another day at 45 ºC. After this period, the plastic film was removed and the reaction continued for one more day at 65 ºC. 2.3. Preparation of modified films The VS and the TEOS (or TTBO) sols (5, 15 and 25 wt-%) derived from the hydrolytic condensation were diluted with ethanol in a weight ratio 1:30 and 5 mmol BPO was added into each.
H. Wang et al. / Computational and experimental determinations of the UV adsorption
653
The solution was dip coated on glass substrates (76.4 mm × 25.2 mm × 1.2 mm) at 270 mm/min. The films were cured at 80 ºC for 6 h, then at 120 °C for 2 h. Film thickness was determined by SEM to be 1-1.4 μm thick. A Shimadzu UV-3101PC scanning spectrophotometer (SSP) device was used to measure transparency of the film between 320-400 nm. 2.4. Simulations and Computations Models of VS, VSTE and VSTT (see Fig. 1) were generated with a material visualizer module of Material Studio software (MSS, Accelrys Inc) on the assumption that they are all the complete-cagetype building blocks (BB). The cells of VS, VSTE and VSTT BB were generated using building crystal tool of the MSS and crystal cell parameters were selected from the MSS database. Models of VS-BB, VSTE-BB (15 wt-%), and VSTT-BB (15 wt-%) are shown in Fig. 1. They were optimized with Geometry Optimization Program in the Dmol3 module of MSS.
(a)
(b)
(c)
Fig. 1. The models of (a) VS-BB, (b) VSTE-BB (15 wt-%), and (c) VSTT-BB (15 wt-%), generated with a material visualizer module of MSS.
2.5. Computing density of states and energy-band structure The simulation and calculation for the DOS and the energy-band structure (EBS) of VS-BB, VSTEBB and VSTT-BB were performed using the Ceperley and Alder-Perdew and Zuger (CA-PZ) method in a local density approximation force field with MSS [19]. Dmol3 uses a simplified linear interpolation scheme developed by Warren [20]. This method is based on linear interpolation in parallelepipeds formed by the points of the Monkhorst-Pack set and followed by the histogram sampling of the resultant set of band energies [21]. 3. Results and discussions 3.1. Formation of hybrids The hybrids were prepared in two steps from VMS through an initial sol-gel polymerization that generated oligosilsesquixoanes. The VMS-oligomer was the viscous oil as is expected for amorphous SSOs composed of lightly interconnected cyclic tetramers and acyclic oligomers. This oligomer was used as the basic material for all three classes of hybrids studied. The baseline material was free radical cross-linked VMS oligomer. The VS resin was first dissolved in ethanol with 5 mol-% benzoyl peroxide (BPO), then casted on a glass slide, cured at 80 °C to start the radical chemistry, and was
654
H. Wang et al. / Computational and experimental determinations of the UV adsorption
finally removed of any residual solvent at 120 °C after exhausting residual BPO. Free radicals generated from the BPO will react with some of the vinyl groups, causing them to link oligomers together. Because of shift from low reactivity of VS to free radical chain polymerization and high chain, molecule weight was raised so that the VS were no longer oily, but in solid polymer state instead. The VSTE films were prepared by polymerizing TEOS in a solution of VS in ethanol. The water for the hydrolysis and condensation of TEOS came from the air over the three days required. Then, BPO (5 mol-%) was added into the VS and silica sol and the resulting mixture was cast, cured, and dried as the VS system above. Lastly, the VSTT films were prepared as with the VSTE films above, save TTBO was used instead of TEOS to prepare the solid VSTT films. 3.2. Average transparency in the ultraviolet region The VS films showed a high average transparency above 360 nm and a gradual decrease in UV transmission until modified just below 300 nm where the films become completely absorbing. The transparency had a slight increase with an additional of silica in VSTE. And with the addition of titanium dioxide in VSTT, the transparency increased even more in relation to the VS. These materials all exhibit a relatively good absorption of UV-B around 88-90%, but not as good across the UV-A region, which is around 17-21%. It is possible to get much greater variations in UV absorption with doping of titanium dioxide or titanium into non-absorbing media. Overall, the changes with addition of silica and titanium dioxide show that the VS material by itself is the dominant absorbing species and that the perturbations from the addition of silica or titanium are small. The optical band widths (Eg) approximated from a straight-line fit in the (h)1/2 versus (h) plot (Table 1). Amorphous titanium dioxide has an optical band gap of 3.75 eV [22]. Crystal titania have smaller band gaps, but are unlikely to form without hydrothermal processing. The optical band gap of VS is 3.713 eV, indicating that the light may be absorbed if its wavelength is less than 334 nm. The optical band gaps for the VSTE and VSTT films were of a similar magnitude (Table 1). The similarity in transmission profiles is attributable to the fact that both titanium dioxide and VS have similar optical band gaps, and that the silica does nothing but dilute the concentration of the UV absorbing VS constituent of the hybrid. Increasing amounts of silica and titanium dioxide have only small effect on the AT values of the VS films in the range of the 280–400 nm radiations (Table 1) with 15 wt-% having the great transparency. A higher dose of silica and titanium dioxide slightly attenuated transparency, a possible explanation is due to increased scattering of scattering particles with higher concentration in the hybrid films . Table 1 Average transparencies and Eg of VSTE and VSTT Wt-% 0 5 15 25
VSTE UV-B (280320 nm) AT (%) 11.16 12.39 12.85 12.24
UV-A (320400 nm) AT (%) 79.63 82.83 83.01 82.67
Eg (eV) 3.713 3.740 3.742 3.732
VSTT UV-B (280320 nm) AT (%) 11.16 11.88 12.21 11.83
UV-A (320400 nm) AT (%) 79.63 81.59 82.07 81.38
Eg (eV) 3.713 3.718 3.720 3.718
3.3. Computation of density of states and the energy-band structure Fig. 3 (a)-(c) show the Dmol3 output for the energy-band structure and the DOS of the VS, VSTE and VSTT building blocks, respectively. The DOS for a given band n, Nn(E) can be obtained by Tauc’s eq.
655
100 90 80 70 60 50 40 30 20 10 0 280
Transparency(%)
Transparency(%)
H. Wang et al. / Computational and experimental determinations of the UV adsorption
f-VS 5%Si 15%Si 25%Si
300
320
340 Wavelength(nm)
360
380
400
100 90 80 70 60 50 40 30 20 10 0 280
f-VS 5%Ti 15%Ti 25%Ti
300
320
340
360
380
400
Wavelength(nm)
Fig. 2. The transparency of the VSTE and VSTT, measured in the range of the UV region (280-400 nm).
The graphs are annotated to show the orbital identification (s, p, or d) of the main peaks in the DOS. This information is often used for a quick visual analysis of the electronic structure and characteristics (such as the width of the band, the energy gap in insulator, etc.). The Fermi level is shown in the graphs with a vertical dash line. All energies are relative to the Fermi level (or the top of the valence band in the case of insulators or semiconductors). Peaks to the right are higher in energy and peaks to the left are lower in energy than the Fermi level. The labels along the X axis of the energy-band structure graph correspond to the standard definitions of high symmetry points for a given lattice type, for instance, the point is denoted by a G. With the DOS and EBS graphs, the width between the valence band and the conduction band (the forbidden band) of VS-BB, VSTE-BB and VSTT-BB can be determined: 0.1629, 0.1707 and 0.1691 Ha (4.43, 4.64 and 4.60 eV), respectively. The greater width of the band gaps in VSTE-BB and VSTT (relative to VS) can be explained by greater contributions from oxygen p orbitals with greater energy differences between bonding and anti-bonding orbitals. The VSTT system has additional contributions to the conduction band from the 3d orbits that results in a slightly narrower gap than that observed with silica. The differences between the computational and experimental band gaps are likely rooted in the difficulties in obtaining accurate and experimental structures of these amorphous hybrids, especially when complicated by SiO2 or TiO2 structures and radical generated networks. 4. Conclusions Vinylsilsesquioxane thin films can be used to block a significant amount of UV-B and a lesser amount of UV-A. The addition of in-situ generated silica or titanium dioxide have minuscule effects on the absorption of UV. Although they have little to do with improving the sunscreen characteristics of the materials, silica or titanium dioxide can be added as reinforcing agents to strengthen the hybrids without attenuating the UV absorbing ability of the films. The VS hybrids were prepared by a relatively simple sol-gel polymerization followed by in-situ polymerization of silica or titanium dioxide from TEOS or TTOB, and then casting and free radical curing of the VS, VSTE or VSTT thin films. The films were found to be optically transparent above 360 nm with calculated optical band gaps near 3.7 eV. Molecular modeling was used to calculate DOSs and EBSs that were a reasonable approximation of what was observed experimentally, but different enough to highlight the need for continued basic research into the amorphous structure and soft matter structures such as these hybrids.
656
H. Wang et al. / Computational and experimental determinations of the UV adsorption
(a)
Energy (Ha)
Density of States (electrons/Ha)
0.3
1200
s f
0.2
p Sum
d
1000
0.1 0.0
800
-0.1 -0.2
600
-0.3 -0.4
400
-0.5 -0.6
200
-0.7
0 -0.8
-0.8 G
F
Q
Z
G
-0.6
-0.4
-0.2
0.0
0.2
0.4
Energy (Ha)
(b)
Density of States (electrons/Ha)
Energy (Ha) 0.3
1000 s f
0.2 0.1
p Sum
d
800
0.0 -0.1
600
-0.2 -0.3
400
-0.4 -0.5
200
-0.6 -0.7 -0.8 G
F
Q
Z
G
0 -0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
Energy (Ha)
(c)
Density of States (electrons/Ha)
Energy (Ha) 0.3
900
0.2
s f
800
0.1
p Sum
d
700
0.0 -0.1
600
-0.2
500
-0.3
400
-0.4
300
-0.5
200
-0.6
100
-0.7 -0.8 G
F
Q
Z
G
0 -0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
Energy (Ha)
Fig. 3. The Dmol3 outputs for the energy-band structure and DOS of (a) VS, (b) VSTE (15 wt-%) and (c) VSTT (15 wt-%).
H. Wang et al. / Computational and experimental determinations of the UV adsorption
657
5. Acknowledgements Financial supports are acknowledged from The Aerospace Supported Fund (GN: 2012-HT-HGD11), from The Invitation of Foreign Experts Program of The Chinese Foreign Experts Bureau (GN: GDW20122300078), from Basic Science Foundation of Central University of China (DL11BB01) and from 9th One-Thousand Expert Plan of The Organization Ministry of The Central Government, China. References [1] [2] [3] [4] [5]
[6]
[7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]
E.C. Naylor, R.E.B. Watson and M.J. Sherratt, Molecular aspects of skin ageing, Maturitas 69 (2011), 249–256. M. Norval and G.M. Halliday, The consequences of UV-induced immunosuppression for human health, Photochem Photobiol 87 (2011), 965–977. L. Andreassi, UV exposure as a risk factor for skin cancer, Expert Review of Dermatology 6 (2011), 445–454. B. Simionescu, M. Aflori and M. Olaru, Protective coatings based on silsesquioxane nanocomposite films for building limestones, Constr Build Mater 23 (2009), 3426–3430. S.H. Phillips, R.I. Gonzalez, R.L. Blanski, B.D. Viers and G.B. Hoflund, Hybrid inorganic/organic reactive polymers for severe environment protection, 47th International Sampe Symposium and Exhibition, Vol 47 (2002), Books 1 and 2, 1488–1496. D. Wang, X.D. Chen, X.W. Zhang, Y. Liu and L.J. Hu, Enhancement corrosion resistance of (γ-methacryloxypropyl)silsesquioxane hybrid films and its validation by gas-molecule diffusion coefficients using MD simulation, J Sol-Gel Sci Tech 49 (2009), 293–300. G. Xie, P. Wang and L. Hu, Enhancement corrosion resistance of (γ-glycidoxylpropyl)-silsesquioxane hybrid films and its validation by gas-molecule diffusion coefficients using MD simulation, Mater Sci Forum 610-613 (2009), 190–197. P. Chen, L. Hu, X. Zhang and D. Sun, Enhanced corrosion resistance for silsesquioxane coatings by diglycidyl ether of biphenol A, Mater Sci-Poland 25 (2007), 843–849. P. Chen, T. Liu, D. Wang, Y. Liu and L. Hu, Silsesquioxane materials as sun protection factor ingredients and as films for greenhouse covers, Materials Science Forum 610-613 (2009), 104–108. X. Chen, D. Wang, X. Li, Y. Liu and L. Hu, POSS/TiO2 nanohybrids as sun protection factors for skin changes caused by ultraviolet A radiation, Materials Science Forum 610-613 (2009), 1039–1043. L. Hu, X. Chen, D. Wang and Y. Liu, Silsesquioxane materials as sun protection factor ingredients and as films for greenhouse covers, PMSE Preprints 100 (2009), 271–272. S.S. Kim, V. Kim and Y.B. Kim, Preparation and characterization of polysilsesquioxane particles containing UVabsorbing groups, Macromolecular Research 20 (2012), 437–446. Y.B. Kim and K-S. Yoon, Organic-inorganic hybrid UV blocking powders based on polysilsesquioxane, Proceeding of NSTI Nanotech 2007, 4 (2007), 254–256. P.G. Parejo, M. Zayat and D. Levy, Photostability and retention of UV absorber molecules in sol-gel hybrid UVprotective coatings, J Sol-Gel Sci Tech 53 (2010), 280–286. D. Wang, P. Zhu, S. Wei and L. Hu, Influence of molecular structure on the optical property of POSS: a DFT calculation based quantum chemistry calculation, Materials Science Forum 689 (2011), 58–63. W.X. Que, Z. Sun, Y. Zhou, Y.L. Lam, S.D. Cheng and Y.C. Chan, Preparation of hard optical coatings based on an organic/inorganic composite by sol-gel method, Mater Lett 42 (2000), 326–330. P. Etienne, J. Denape, J.Y. Paris, J. Phalippou and R. Sempere, Tribological properties of ormosil coatings, J Sol-Gel Sci Tech 6 (1996), 287–297. X-D. Chen, D. Wang, D-R. Lin, Q. Zhang and L-J. Hu, Transparency of modified vinyl-silsesquioxane films and its validation by computing energy-band structure, J Mater Res 25 (2010) 76–81. J. Hernandez-Paredes, D. Glossman-Mitnik, H.E. Esparza-Ponce, M.E. Alvarez-Ramos and A. Duarte-Moller, Band structure, optical properties and infrared spectrum of glycine-sodium nitrate crystal, J Mol Struc 875 (2008), 295–301. M.C. Warren, G.J. Ackland, B.B. Karki and S.J. Clark, Phase transitions in silicate perovskites from first principles, Mineral Mag 62 (1998), 585–598. H.J. Monkhorst and J.D. Pack, Special Points for Brillouin-Zone Integrations, Phys Rev B 13 (1976) , 5188–5192. S. Valencia, J.M. Marín and G. Restrepo, Study of the Bandgap of Synthesized Titanium Dioxide Nanoparticules Using the Sol-Gel Method and a Hydrothermal Treatment, The Open Materials Science Journal 4 (2010), 9–14.
