Mott behavior in ${\mathrm{K}}_x{\mathrm{Fe}}_{2-y}$${\mathrm{Se}}_2$ superconductors studied by pump-probe spectroscopy

We report on a signature indicating a transition to the orbital-selective Mott phase in superconducting ${\mathrm{K}}_x$${\mathrm{Fe}}_{2-y}$${\mathrm{Se}}_2$ single crystals by employing dual-color pump-probe spectroscopy. In addition to the multiexponential-decay recovery dynamics of photoinduced quasiparticles, a damped oscillatory component caused by coherent acoustic phonons emerges when the superconducting phase is suppressed at an increased temperature or excitation power. Upon raising the temperature to 150–170 K, the oscillatory component diminishes alongside a significant enhancement of the slow decay component in the recovery traces. These results can be understood as a gap opening in certain $k$ directions, indicating that a vital role is played by electron correlation in iron-based superconductors.

Figure 1

Fluence-dependent QP dynamics at the superconducting phase. (a) Temperature-dependent resistivity of a superconducting KFS single crystal. In addition to a superconducting transition at $T_C\sim 32$ K, a resistivity hump appears at $T_H\sim 180$ K. (b) Time-resolved $\Delta R/R$ recorded at 5 K with an excitation flux of ∼200 μJ/cm${}^2$. The decay curve is analyzed [Eq. (1)]. The exponential-decay component and the oscillatory component (inset) are highlighted. (c) The fluence-dependent maximum signal ($\Delta R_{max}$) and the amplitude of the oscillatory component ($A_O$). (d) The amplitude ($A_1$) and decay rate ($1/{\tau }_1$) of the fast component are plotted as functions of the incident fluence using a double-logarithm scale.

Figure 2

Temperature-dependent QP dynamics in KFS superconductors. (a) The differential reflectivity variation is plotted as functions of temperature and delay time. The maximum value of the initial reflectivity variation (b), the signal magnitude at $t\sim 4.5$ ps and the fitted amplitude of the oscillatory component (c), and the oscillatory frequency (d) are plotted vs temperature. The red line in (b) is the curve fitting to the Mattis-Bardeen formula. The dashed lines indicate the superconducting transition temperature.

Figure 3

QP dynamics at the normal state. (a) The time-resolved differential reflectivity recorded at 80 and 250 K, respectively. (b) The normalized, time-resolved differential reflectivity variation recorded at different temperatures. (c) The temperature-dependent amplitudes of the fast and slow components. The solid line is a fitting curve with a temperature-independent pseudogap. (d) The temperature-dependent amplitude ratio $(A_2/A_1)$. (e) and (f) are the traces recorded at 150 K with the pump and probe beams configured with perpendicular and parallel polarizations, respectively.

Figure 4

Temperature dependence of the oscillatory component. (a) The oscillatory component after subtracting the exponential-decay components from the total signal recorded at different temperatures. The traces are vertically shifted for clarity. Temperature-dependent amplitude (b) and frequency (c) of the oscillatory component.