Extended Data Fig. 7: Comparison of the populations of increasing electrons and decreasing photons.

The increase of electrons (red area in Fig. 4d) and the decrease of photons (blue area in Fig. 4d) are plotted against Z_tip–mol. The number of electrons was obtained from the detected photocurrent value divided by the elementary charge e. The number of quenched photons (\({I}_{{\rm{ph}}}^{{\rm{quench}}}\)) was obtained by the following equation. \({I}_{{\rm{ph}}}^{{\rm{quench}}}=({I}_{{\rm{ph}}}^{{\rm{\det }}}-\,{I}_{{\rm{ph}}}^{{\rm{fit}}})/{\eta }_{{\rm{\det }}}\). Here, \({I}_{{\rm{ph}}}^{{\rm{\det }}}\) is the number of the photons (counts per second) detected in the experiment, and \({I}_{{\rm{ph}}}^{{\rm{fit}}}\) is the photon numbers (counts per second) deduced by the fitting curve I_ph = 1.09 × 10⁵ × exp(−6.77Z_tip–mol) for the Z_tip–mol range between 1.1 nm and 0.55 nm shown in Fig. 4d. By extrapolating the fitted curve into the Z_tip–mol value less than 0.53 nm, we estimated the expected photon intensity without the photoluminescence quenching in this region. η_det is the detection efficiency determined by the experimental set-up. By considering the collection solid angle of the lens, detection quantum efficiency of the detector, and the reflection, diffraction and transmission of the optics, η_det was estimated to be 4.9 × 10⁻⁴ for this measurement (see Methods). It was revealed that the number of electrons flowing as photocurrent and the number of quenched photons at each Z_tip–mol were comparable. These results suggest that the observed photoluminescence quenching mainly originated from the photocurrent generation. The slightly large value of the quenched photons might come from another nonradiative recombination pathway or underestimation of η_det.