Stripes: Why hole rich lines are antiphase domain walls?

For stripes of hole rich lines in doped antiferromagnets, we investigate the competition between anti-phase and in-phase domain wall ground state configurations. We argue that a phase transition must occure as a function of the electron/hole filling fraction of the domain wall. Due to {\em transverse} kinetic hole fluctuations, empty domain walls are always anti-phase. At arbitrary electron filling fraction ($\delta$) of the domain wall (and in particular for $\delta \approx 1/4$ as in LaNdSrCuO), it is essential to account also for the transverse magnetic interactions of the electrons and their mobility {\em along} the domain wall. We find that the transition from anti-phase to in-phase stripe domain wall occurs at a critical filling fraction $0.28<\delta_{c}<0.30$, for any value of $\frac{J}{t}<{1/3}$. We further use our model to estimate the spin-wave velocity in a stripe system. Finally, relate the results of our microscopic model to previous Landau theory approach to stripes.