We calculate the D→η transition form factor in light-cone sum rules by taking improved current correlators to avoid the pollution from the twist-3 wave function. We get consistent results of the D+→η(′)l+νl decays with the experimental data. By comparing the difference between the results of the branching ratios of D+→η(′)l+νl from a two-pole parameterization model and from a BZ parameterization model, we find that the two-pole model and the BZ model are comparably believable. One way is supposed for the determination of the η-η mixing angle from the dependence of the branching ratios of D+→η(′)l+νl decays on the η-η mixing angle.
We calculate the D → η transition form factor in light-cone sum rules by taking improved current correlators to avoid the pollution from the twist-3 wave function. We get consistent results of the D+ → η^(')l+νl decays with the experimental data. By comparing the difference between the results of the branching ratios of D+ → η^(')l+νl from a two-pole parameterization model and from a BZ parameterization model, we find that the two-pole model and the BZ model are comparably believable. One way is supposed for the determination of the η-η mixing angle from the dependence of the branching ratios of D+ → η^(')l+νl decays on the η-η mixing angle.