CommuniCation(1 of 5) 1600888 UV barrier films on the flexible substrate.Furthermore, multilayer barrier films exhibit high Fresnel reflection loss at visible wavelengths (≈400 ≤ λ ≤ ≈700 nm) as a result of differences in the refractive indices of barrier layers. [27] By using an optical design to create a film that is impermeable to both water and UV, these problems will be overcome.In this letter, we propose a new type of multifunctional barrier film that prevents penetration by both water and UV light (Figure 1). To enable a low-cost and simple process, we developed a fabrication approach that uses UV-distributed Bragg reflector (UV-DBR) structures to combines thin-film passivation and a UV reflector. Layers of passivation materials (Si x N y , SiO 2 ) were alternated ("SN/SO pairs") in the UV-DBR structure, so water transmission was reduced. Simulation was conducted to optimize the thickness of SiO 2 and Si x N y to block UV light (λ < 400 nm) while maintaining visible transparency. According to finite difference time domain (FDTD) simulations, UV light can be remarkably reflected by the UV-DBR because of the welldesigned Bragg reflector structure. The UV-DBR structure with 10 SN/SO pairs reduced UV transmittance UT (wavelengths 200 ≤ λ ≤ 400 nm) to ≈1.6% while maintaining visible transmittance VT ≈ 78.9%.To design the UV-DBR structure, optical simulation was performed using Macleod (Thin Film Center, Inc.), which is based on the characteristics matrix method for optical analysis involving a multilayer structure. The optical constants of Si x N y and SiO 2 were measured using spectroscopic ellipsometry. Si x N y has an energy bandgap >4 eV, so it can protect against high-energy UV radiation (λ < 300 nm). [28] Therefore, optical simulation was conducted to optimize the thicknesses of SiO 2 and Si x N y to block UV (300 < λ < 400 nm). To determine the optimum thickness, five kinds of UV-DBR structure were considered (Figure 2a). In the design of the UV-DBR samples, a DBR structure with 10 SN/SO pairs (10-pair UV-DBR) was used, and center wavelengths were 310 ≤ λ c ≤ 390 nm. At 350 ≤ λ c ≤ 390 nm, UT was minimized to ≈3.0%. However, at λ c = 310 and 330 nm, 5.3 ≤ UT ≤ 7.9%. The average visible transmittance VT avg gradually decreased from 83.7% to 75.4% as λ c increased (Figure 2b). λ c = 350 nm was chosen as the best compromise between blocking of UV and maintenance of VT (81.8%). To obtain these values, the thicknesses of Si x N y and SiO 2 layers were calculated as t = λ/4n, where n is the refractive index of Si x N y (n = 2.3) and SiO 2 (n = 1.6) layers. For Si x N y , t = 41 nm; for SiO 2 , t = 54 nm. To study the angular dependency of UV-DBR, its UT and VT (Figure 2c) were calculated for incident angles 0°-80° to the normal. Three structures Si x N y (41 nm, Figure S2, Supporting Information), SiO 2 (54 nm, Figure S2, Supporting Information), and 10-pair UV-DBR (Figure 2c) on glass were simulated. Both Si x N y and SiO 2 samples had high angular UT (Si x N y = 49.6%, SiO 2 = 84.7% at 0°),