RESUMO
Ferroelectric domain walls provide a fertile environment for novel materials physics. If a polarization discontinuity arises, it can drive a redistribution of electronic carriers and changes in band structure, which often result in emergent 2D conductivity. If such a discontinuity is not tolerated, then its amelioration usually involves the formation of complex topological patterns, such as flux-closure domains, dipolar vortices, skyrmions, merons, or Hopfions. The degrees of freedom required for the development of such patterns, in which dipolar rotation is a hallmark, are readily found in multiaxial ferroelectrics. In uniaxial ferroelectrics, where only two opposite polar orientations are possible, it has been assumed that discontinuities are unavoidable when antiparallel components of polarization meet. This perception has been borne out by the appearance of charged conducting domain walls in systems such as hexagonal manganites and lithium niobate. Here, experimental and theoretical investigations on lead germanate (Pb5 Ge3 O11 ) reveal that polar discontinuities can be obviated at head-to-head and tail-to-tail domain walls by mutual domain bifurcation along two different axes, creating a characteristic saddle-point domain wall morphology and associated novel dipolar topology, removing the need for screening charge accumulation and associated conductivity enhancement.