SAXS studies of nanosilica/polyurethane composites

PERUZZO, P.J.; AMALVY, J.I.

Campinas, Brasil

Congreso; 18a. RAU Reunión Anual de Usuarios LNLS; 2008

Laboratorio Nacional de Luz Sincrotron - Brasil

Polymer nanocomposites materials prepared using nanosilica ¯llers have unusual properties and in these materials the polymer chains are con¯ned to nanoscale dimensions (1-10 nm). Film-forming, silica-based materials are useful as tough, abrasion-resistant coatings with increasing friction coe±cient and reduced tackiness, with no change in gloss. In this work, ¯lms of polyurethane (IPDI/PPG1000) with di®erent content (5, 10 and 15 wt%) of colloidal nanosilica (Bindzil r° CC40, 12 nm) were studied by SAXS. Samples were prepared by using two di®erent synthetic methods starting from a PU prepolymer containing vinyl end groups. In one case the PU prepolymer was dispersed in water containing the colloidal nanosilica particles and then dispersion polymerized; in the other, the PU prepolymer was dispersed in water, then colloidal nanosilica was added to the dispersion and polymerized. A mixture from the PU prepolymer dispersed in water and polymerized, was prepared with the colloidal nanosilica for comparative purpose. Films were prepared by casting the dispersion at room temperature. A monochromatic beam of wavelength 1.608 ºAwas used and the scattering intensity was registered using a sample-detector distance of 2043.72 mm. SAXS curves show oscillations with maximums at q around 0.030 and 0.055ºA-1. They are the common features found in dilute systems of nanoparticles in a homogeneous matrix. Scattering intensity increases as silica content of sample increases. Porod0s plots have an asymptotic linear regime for all the samples indicating that the structure can be described by a two-electron density model. SAXS curves for q > 0.035 ºA-1 do not show appreciable diferences for the two estrategies. This indicates that the structure is globally similar. However for the blend a noticeable smaller scattering intensity for the entire wave vector range is observed, which is more evident for the low silica content samples. For q < 0.035 ºA¡1 some diferences appear in all cases. Samples from frst and second strategies show a scattering contribution at q around 0.02-0.03 ºA, which is not observable in blends. This behavior is related partly to particle interaction and partly to the formation of a di®erent structure depending on the preparation method. This observation is consistent with TEM results.nm). Film-forming, silica-based materials are useful as tough, abrasion-resistant coatings with increasing friction coe±cient and reduced tackiness, with no change in gloss. In this work, ¯lms of polyurethane (IPDI/PPG1000) with di®erent content (5, 10 and 15 wt%) of colloidal nanosilica (Bindzil r° CC40, 12 nm) were studied by SAXS. Samples were prepared by using two di®erent synthetic methods starting from a PU prepolymer containing vinyl end groups. In one case the PU prepolymer was dispersed in water containing the colloidal nanosilica particles and then dispersion polymerized; in the other, the PU prepolymer was dispersed in water, then colloidal nanosilica was added to the dispersion and polymerized. A mixture from the PU prepolymer dispersed in water and polymerized, was prepared with the colloidal nanosilica for comparative purpose. Films were prepared by casting the dispersion at room temperature. A monochromatic beam of wavelength 1.608 ºAwas used and the scattering intensity was registered using a sample-detector distance of 2043.72 mm. SAXS curves show oscillations with maximums at q around 0.030 and 0.055ºA-1. They are the common features found in dilute systems of nanoparticles in a homogeneous matrix. Scattering intensity increases as silica content of sample increases. Porod0s plots have an asymptotic linear regime for all the samples indicating that the structure can be described by a two-electron density model. SAXS curves for q > 0.035 ºA-1 do not show appreciable diferences for the two estrategies. This indicates that the structure is globally similar. However for the blend a noticeable smaller scattering intensity for the entire wave vector range is observed, which is more evident for the low silica content samples. For q < 0.035 ºA¡1 some diferences appear in all cases. Samples from frst and second strategies show a scattering contribution at q around 0.02-0.03 ºA, which is not observable in blends. This behavior is related partly to particle interaction and partly to the formation of a di®erent structure depending on the preparation method. This observation is consistent with TEM results.