Contents lists available at ScienceDirect Chemical Engineering Science journal homepage: www.elsevier.com/locate/ces Oxygen concentration and conversion distributions in a layer-by-layer UV- cured film used as a simplified model of a 3D UV inkjet printing system Kentaro Taki a, ⁎ , Yoshihito Watanabe b , Tadao Tanabe c , Hiroshi Ito d , Masahiro Ohshima b a Advanced Reactive Systems Lab, Chemical and Material Engineering Course, School of Natural Systems, College of Science and Engineering, Kanazawa University, Rm. 1C416, Kakumacho, Kanazawa, Ishikawa 920-1192, Japan b Department of Chemical Engineering, Kyoto University, Japan c Department of Materials Science, Graduate School of Engineering, Tohoku University, Japan d Department of Polymer Science and Engineering, Yamagata University, Japan ARTICLE INFO Keywords: Photopolymerization Oxygen inhibition UV curing Additive manufacturing ABSTRACT Three-dimensional (3D) ultraviolet (UV) inkjet printers represent a versatile technology for creating complex functional structures. During their operation, 3D objects are formed by repeating cycles of drawing a UV-curable resin with inkjet nozzles and then solidifying it with UV irradiation. In this study, the activity performed by a 3D UV inkjet printer was simulated by spin casting a 33 μm thick layer of UV-curable resin (containing diurethanedi- methacrylate and 1-hydroxycyclohexyl phenyl ketone compounds mixed at a weight ratio of 99:1) onto a Si wafer followed by photopolymerization for 2 s at a UV irradiation of 10 mW cm -2 . Afterwards, the second resin layer with a thickness of 33 μm was spun-cast onto the first layer and photopolymerized under the same conditions. The conversion distribution of C=C bonds in the UV-curable resin was investigated via confocal laser Raman microscopy and numerical calculations, which took into account the kinetics of photopolymerization and oxygen inhibition reactions. The confocal laser Raman microscopy technique provided a unique distribution of the C=C bond conversion across the film depth. Thus, the conversion magnitude at a depth of 0 μm was zero and increased to 0.2 at 6 μm. Afterwards, the slope of the conversion distribution plot became moderate until the conversion reached the value of 0.43 at a film depth of 28 μm. Between the film depths of 28 and 38 μm, the conversion remained constant with a variation not exceeding 0.03. After that, the conversion value increased again, reaching the magnitude of 0.48 at a depth of 50 μm and remained constant in the region between 50 and 56 μm (with a variation not exceeding 0.04). At higher depths, the graph slope became moderate again, and the conversion value increased gradually to 0.51 at 66 μm, after which the silicon wafer was reached. As a result, two different plateaus were observed on the conversion distribution plot: between 28 and 38 μm and between 50 and 56 μm (the corresponding conversion variation in these regions was below 0.05). The obtained experimental data were in good agreement with the results of numerical calculations, which attributed the existence of the two plateaus on the plot of the C=C bond conversion distribution to the formation of an oxygen-lean point. In addition, the effects of the UV intensity, irradiation time, lamination time, photoinitiator concentration, and concentration of dissolved oxygen on the oxygen concentration and conversion distributions across the depth direction have been examined. The obtained results revealed that the increases in the UV intensity, irradiation time, and photoinitiator concentration as well as the decrease in the initial dissolved oxygen concentration effectively increased the conversion of C=C bonds in the resin film and decreased the thickness of an unpolymerized layer. http://dx.doi.org/10.1016/j.ces.2016.10.050 Received 10 February 2016; Received in revised form 27 September 2016; Accepted 15 October 2016 ⁎ Corresponding author. E-mail address: taki@se.kanazawa-u.ac.jp (K. Taki). Abbreviations: I, Initiator radical; I-OO, Initiator peroxide radical; I-M n , Macroradical; I-M n -OO, Peroxide macroradical; M, Monomer; PI, Photoinitiator; O 2 , Oxygen molecule; A p , Parameter of the Goodner-Bowman model; A t , Parameter of the Goodner-Bowman model; A ν, , Absorbance measured at wavenumber ν; D O , Oxygen diffusion coefficient; D O0 , Oxygen diffusion coefficient in monomer; f, Free volume fraction; f a , Rational factor of Raman spectroscopy and FT-IR techniques; f cp , Critical free volume fraction of propagation; f ct , Critical free volume fraction of termination; f m , Free volume fraction of monomer; f p , Free volume fraction of polymer; I(z), UV light intensity; I 0 , UV light intensity at the upper film surface; [IR], Concentration of initiator radicals; k i , Rate coefficient for the reaction of initiator radicals with C=C bonds; k io , Rate coefficient for the reaction of initiator radicals with oxygen species; k O , Rate coefficient for the oxygen inhibition reaction; k p0 , Rate coefficient for the propagation reaction without diffusion; k p , Propagation rate coefficient; k t0 , Rate coefficient for the termination reaction without diffusion; k t , Binary termination constant; H,, Film thickness; [M], Concentration of C=C bonds; [MR], Concentration of macroradicals; [O 2 ] eqb , Equilibrium concentration of oxygen; [PI], Photoinitiator concentration; R,, Reaction diffusion parameter; R i , Initiation rate; R Si ,, UV light reflection ratio; t, Time; T gm , Glass transition temperature for the monomer thermal expansion coefficient; T gp , Glass transition temperature for the polymer thermal expansion coefficient; x, Vertical coordinate across the UV light direction; x A , Functional monomer conversion; z,, Horizontal coordinate across the UV light direction; α, Attenuation parameter for the diffusion coefficient; α m , Thermal expansion coefficient for monomer; α p , Thermal expansion coefficient for polymer; ϵ, Molar absorption coefficient for photoinitiator; ϕ, Quantum yield; ϕ m , Monomer volume fraction Chemical Engineering Science 158 (2017) 569–579 0009-2509/ © 2016 Elsevier Ltd. All rights reserved. Available online 05 November 2016 crossmark