Robust Ferroelectricity in Silicon Dioxide upon Intercalation of Ammonia

TL;DR

This paper demonstrates that intercalating ammonia into silicon dioxide induces robust, reversible ferroelectricity, potentially enabling silicon-compatible ferroelectric devices with novel quantized polarization effects.

Contribution

It introduces a new method to induce ferroelectricity in SiO2 via NH3 intercalation, revealing unconventional polarization mechanisms and quantized ferroelectricity in silicon-based materials.

Findings

NH3 intercalation into SiO2 is exothermic and induces large polarization.

Reversible ferroelectricity can be achieved through N-Si bond reformation.

Unconventional quantized ferroelectricity may occur under strong electric fields.

Abstract

The nanoelectronic applications of current ferroelectrics have been greatly impeded by their incompatibility with silicon. In this paper we propose a way to induce ferroelectricity in silicon dioxide (SiO2), which is still the most widely used dielectric material in silicon-based chips. We show first-principles evidence that the intercalation of NH3 molecules into crystalline SiO2 is exothermic, where NH3 molecules form quasi-bonds with SiO2, giving rise to large and robust polarizations. In general, such polarization can be reversed via the reformation of N-Si bondings, which is multiaxial so vertical ferroelectricity may emerge in their thin-films of any facets. When the applied external electric field is large enough, however, the system may exhibit unconventional quantized ferroelectricity of unprecedented magnitude, where NH3 may migrate for multiple lattice constants like mobile ions in ion conductors. Compared with ion conductors with charged mobile ions and ion vacancies that may lead to current leakage, herein the intercalated systems can be denoted as neutral ion conductors where both pristine SiO2 and SiO2 filled with NH3 are insulating. Similar ferroelectricity may exist in various SiO2 crystalline polymorphs, its amorphous phase, and other porous structures intercalated by NH3. Our findings may not only resolve the bottleneck issues for the compatibility of ferroelectrics and silicon, but also develop unconventional mechanisms of ferroelectricity.