Nonet symmetry in \eta, \eta^{\prime} and B\to K\eta,K\eta^{\prime} decays

Abstract

The nonet symmetry scheme seems to describe rather well the masses and $\eta -{\eta }^{\prime }$ mixing angle of the ground state pseudo-scalar mesons. It is expected that nonet symmetry should also be valid for the matrix elements of the pseudo-scalar densitty operators which play an important role in charmless two-body B decays with $\eta$ or ${\eta }^{\prime }$ in the final state. Starting from the divergences of the SU(3) octet and singlet axial vector currents, we show that nonet symmetry for the pseudo-scalar mass term implies nonet symmetry for the pseudo-scalar density operators. In this nonet symmetry scheme, we find that the branching ratio $B\to PP,PV$, with $\eta$ in the final state agrees well with data, while those with ${\eta }^{\prime }$ are underestimated, but by increasing the $B\to {\eta }^{\prime }$ form factor by $40-50$, one could explain the tree-dominated $B^{-}\to {\pi }^{-}{\eta }^{\prime }$ and $B^{-}\to {\rho }^{-}{\eta }^{\prime }$ measured branching ratios. With this increased form factor and with only a moderate annihilation contribution, we are able to obtain $62\times 10^{-6}$ for the penguin-dominated $B^{-}\to K^{-}{\eta }^{\prime }$ branching ratios, quite close to the measured value. This supports the predicted value for the $B\to {\eta }^{\prime }$ form factor in PQCD and light-cone sum rules approach. A possible increase by 15% of $<0|\bar{s}i{\gamma }_{5}s|s\bar{s}>$ for ${\eta }_0$ would bring the predicted $B^{-}\to K^{-}{\eta }^{\prime }$ branching ratio to $69.375\times 10^{-6}$, very close to experiment.

Comments: 13 pages, minor typos corrected (references), PRD version