Figure 1

(a) Schematics of a double-barrier single-molecule junction. (b) Molecular structure of CuPc. (c) STM topographic images of three different conformations of CuPc (M1, M2, and M3) adsorbed on the oxide surface. The scan size is 37 Å by 37 Å, $V_b=2.0\mathrm{V}$, and $I=0.1\mathrm{n}\mathrm{A}$. (d) $dI/dV$ spectra measured with the STM tip positioned over the center of the three molecules in (c). The conductance onsets at opposite bias polarities are marked by I (broad band onset) and II (sharp peak onset). The spectra for M1, M2, and oxide are offset for clarity. The $dI/dV$ signal was recorded with the lock-in technique. The tunnel gap was set at $V_b=2.0\mathrm{V}$ and $I=0.1\mathrm{n}\mathrm{A}$. The bias modulation was 10 mV (rms) at 400 Hz. (e) Plot showing the correlation of onset type I and onset type II for 70 different CuPc molecules adsorbed on oxide surface (squares) compared to the correlation of the conductance peaks (pluses).Reuse & Permissions

Figure 2

(a),(b) Diagram showing the conduction through the affinity (${\mathrm{C}\mathrm{u}\mathrm{P}\mathrm{c}}^{-}$) level of a CuPc under negative ($V_b<0$) and positive ($V_b>0$) sample bias, respectively. The electrostatic potential profiles in the junction under the influence of sample bias are shown by the solid lines. (c) Numerical simulation of $I(V)$, normalized at 2 V, based on rate equations with the inclusion of vibronic coupling for different ratios of tunneling rates [18, 19]. ${\Gamma }_1$, ${\Gamma }_2$ are the tunneling rates through the vacuum gap and oxide film, respectively. The two insets show the $dI/dV$ spectra, scaled and offset, for opposite bias polarities. For clarity, one of the closely spaced curves for $V_b>0$ is shown. Parameters are $g=3$, $\hslash {\omega }_0=30\mathrm{m}\mathrm{e}\mathrm{V}$, $kT=0.1\hslash {\omega }_0$, $E_0=0.15\mathrm{V}$, and onset ratios are 5.8, 5.2, and 4.6 for curves 1, 2, and 3, respectively, where $g$ is the Franck-Condon parameter and $\hslash {\omega }_0$ is the vibrational energy. Curves of similar shapes were obtained for different vibrational energies.Reuse & Permissions

Figure 3

(a),(b) $dI/dV$ spectra for different tip-molecule separations ($z$) above molecule M1 (Fig. 1) for $V_b<0$ and $V_b>0$, respectively. $Z$ offsets, referenced to the gap set with $V_b=2.0\mathrm{V}$, $I=0.1\mathrm{n}\mathrm{A}$, were adjusted incrementally from $-1.1$ to $+1.9\AA$. The spectra (except the topmost spectra) are scaled by different factors and offset for clarity. (c) $d^{2}I/d{V}^2$ spectrum measured concurrently with the second spectrum from the top in (b). (d) Positions of the peaks from (c) as a function of the number of the peak. Linear fitting of the data gives the spacing of $29\pm 2\mathrm{m}\mathrm{V}$. (e) Peak positions observed in (a) as a function of $z$ offset. The open squares, circles, and triangles correspond to the first, second, and third peaks marked in (a), respectively. (f) Peak intensities from (a) (open circles) and (b) (solid squares) as a function of $z$ offset. The solid squares correspond to the band intensity maximum in (b), while the open circles correspond to the first peak intensity in (a).Reuse & Permissions

Figure 4

(a),(b) $dI/dV$ spectra with different tip-molecule separations over the molecule M3 (Fig. 1) for $V_b<0$ and $V_b>0$, respectively. The spectra (except the topmost spectra) are scaled by different factors and offset for clarity.Reuse & Permissions