80% Valley Polarization of Free Carriers in Singly Oriented Single-Layer ${\mathrm{WS}}_2$ on Au(111)

We employ time- and angle-resolved photoemission spectroscopy to study the spin- and valley-selective photoexcitation and dynamics of free carriers at the $\overline{\mathit{K}}$ and ${\overline{\mathit{K}}}^{\prime }$ points in singly oriented single-layer ${\mathrm{WS}}_2/\mathrm{Au}(111)$. Our results reveal that in the valence band maximum an ultimate valley polarization of free holes of 84% can be achieved upon excitation with circularly polarized light at room temperature. Notably, we observe a significantly smaller valley polarization for the photoexcited free electrons in the conduction band minimum. Clear differences in the carrier dynamics between electrons and holes imply intervalley scattering processes into dark states being responsible for the efficient depolarization of the excited electron population.

Figure 1

Time-resolved ARPES data of ${\mathrm{WS}}_2/\mathrm{Au}(111)$ taken at $\overline{\mathit{K}}$ for excitation with linearly polarized 2.10 eV pump pulses. (a) ARPES snapshots recorded before optical excitation (left) and at $\mathrm{\Delta }t=50\mathrm{fs}$ (right). Top (bottom) panels show the conduction (valence) band signal. Color scales have been adjusted separately for top and bottom panels to account for the significantly different photoemission intensities from CB and VB. The inset in the bottom left panel shows the hexagonal Brillouin zone of SL ${\mathrm{WS}}_2$. (b) Comparison of energy distribution curves (EDCs) derived from the ARPES data shown in (a). Photoemission intensity was integrated over a momentum window of $0.2{\AA }^{-1}$. The green line shows the difference calculated from the two EDCs.

Figure 2

Detection of valley-selective excitation of the ${\mathrm{WS}}_2$ layer using circularly polarized light. (a) Schematic illustration of the optical selection rules of single-layer ${\mathrm{WS}}_2$ for valley-selective excitation at $\overline{\mathit{K}}$ and ${\overline{\mathit{K}}}^{\prime }$. (b) Difference photoemission intensity maps at $\overline{\mathit{K}}$ and ${\overline{\mathit{K}}}^{\prime }$ obtained from trARPES spectra recorded upon excitation with ${\sigma }^{-}$ and ${\sigma }^{+}$ polarized pump pulses at $\mathrm{\Delta }t=50\mathrm{fs}$. The top (bottom) panels show conduction (valence) band data. The offset of the signals with respect to $k_y=0$ arises from the pronounced momentum dependence of the photoemission cross section of VB and CB excitation [see Fig. 1]. (c) Comparison of normalized difference EDCs at $\overline{\mathit{K}}$ and ${\overline{\mathit{K}}}^{\prime }$ derived from the data shown in (b). The signal was integrated over a momentum window of $0.35{\AA }^{-1}$ and normalized to the VBM peak value.

Figure 3

Photoemission (PE) signal of CB and UVB at $\overline{\mathit{K}}$ and ${\overline{\mathit{K}}}^{\prime }$ as a function of the angle of the quarter-wave plate in the pump beam ($\mathrm{\Delta }t=50\mathrm{fs}$). For the evaluation of the UVB data, an equilibrium state spectrum ($\mathrm{\Delta }t\ll 0$) was subtracted from the excited state spectrum. CB (UVB) traces are normalized to maximum (minimum) PE signal. The error bars of the experimental data account for the uncertainties in determining the signal background not originating from the valley population. The solid lines are the results of the fits of Eq. (1) to the experimental data. The errors in the fits (shaded areas) account for error propagation of the fitting results and the uncertainties in determining ${\widehat{S}}_3$. The top panel displays the normalized Stokes parameter ${\widehat{S}}_3$ determined from the Stokes polarimeter measurements of the pump pulse [29] (red line) in comparison to an ideally polarized pump pulse (dashed gray line).

Figure 4

Intervalley scattering of free carriers in the valence and conduction band. (a) Comparison of the temporal evolution of normalized PE intensities in the CB (left) and the UVB (right) at $\overline{\mathit{K}}$ for excitation with (predominantly) ${\sigma }^{-}$ and ${\sigma }^{+}$ circularly polarized light. The solid lines are the results of a fit of a rate equation model to the experimental data as described in the text. For better comparison, the data are normalized to the maximum or minimum transient intensity, respectively. (b) Schematic illustration of intervalley scattering from ${\overline{\mathit{K}}}^{\prime }$ to $\overline{\mathit{K}}$ for direct photoexcitation at ${\overline{\mathit{K}}}^{\prime }$ (excitation with ${\sigma }^{+}$ circularly polarized light). Phonon emission (blue arrow) accounts for energy conservation in the intervalley scattering process in the CB. The time constants ${\tau }_A$ and ${\tau }_{\overline{\mathit{K}}-{\overline{\mathit{K}}}^{\prime }}$ describe the decay of excited carriers according to the used rate equation model [29].