Frederick Pearsall1 2 Julien Lombardi1 Stephen O'Brien1 2

1, Chemistry, City College of New York, New York, New York, United States
2, Chemistry, The Graduate Center of the City University of New York, New York, New York, United States

Frequency stable, high permittivity nanocomposite capacitors produced under mild processing conditions offer an attractive replacement to MLCCs derived from conventional ceramic firing. Here, 0−3 nanocomposites were prepared using gel-collection derived barium titanate (BTO) nanocrystals, suspended in a poly(furfuryl alcohol) matrix, resulting in a stable, high effective permittivity, low loss dielectric. The nanocrystals are produced at 60 °C, emerging as fully crystallized cubic BTO, 8 nm, with a highly functional surface that enables both suspension and chemical reaction in organic solvents. The nanocrystals were suspended in furfuryl alcohol where volume fraction of nanocrystal filler (νf) could be varied. Polymerization of the matrix in situ at 70−90 °C resulted in a nanocomposite with a higher than anticipated effective permittivity (up to 50, with νf only 0.41, from 0.5−2000 kHz), exceptional stability as a function of frequency, and very favorable dissipation factors (tan δ < 0.01, νf < 0.41; tan δ < 0.05, νf < 0.5). The increased permittivity is attributed to the covalent attachment of the poly(furfuryl alcohol) matrix to the surface of the nanocrystals, homogenizing the particle−matrix interface, limiting undercoordinated surface sites and reducing void space. XPS and FTIR confirmed strong interfacial interaction between matrix and nanocrystal surface. Effective medium approximations were used to compare this with similar nanocomposite systems. It was found that the high effective permittivity could not be attributed to the combination of two components alone, rather the creation of a hybrid nanocomposite possessing its own dielectric behavior. A nondispersive medium was selected to focus on the frequency dependent permittivity of the 8 nm barium titanate nanocrystals. Experimental corroboration with known effective medium approximations is evident until a specific volume fraction (νf ≈ 0.3) where, due to a sharp increase in the effective permittivity, approximations fail to adequately describe the nanocomposite medium.