Fluence-dependent femtosecond quasiparticle and ${\mathrm{Eu}}^{2+}$ spin relaxation dynamics in ${\mathrm{EuFe}}_2(\mathrm{As},\mathrm{P}{)}_2$

We investigated temperature- and fluence-dependent dynamics of the time-resolved optical reflectivity in undoped spin-density-wave (SDW) and doped superconducting (SC) ${\mathrm{EuFe}}_2{(\mathrm{As},\mathrm{P})}_2$ with emphasis on the ordered ${\mathrm{Eu}}^{2+}$ spin temperature region. The data indicate that in ${\mathrm{EuFe}}_2{(\mathrm{As},\mathrm{P})}_2$ the SDW order coexists at low temperature with the SC and ${\mathrm{Eu}}^{2+}$-ferromagnetic order. Increasing the excitation fluence leads to a slow thermal suppression of the ${\mathrm{Eu}}^{2+}$ spin order due to the crystal-lattice heating on a nanosecond time scale while the SDW order is suppressed nonthermally on a subpicosecond time scale at a higher fluence.

Figure 1

Magnetic moment along the $c$ axis as a function of temperature for both field cooled (FC) and zero-field cooled (ZFC) cases.

Figure 2

Variation of the transients across the sample surface at $T=1.5$ K and 1.5-eV pump-photon energy in ${\mathrm{EuFe}}_2{\mathrm{As}}_2$ [(a), (b), (c)] and ${\mathrm{EuFe}}_2$(${\mathrm{As}}_{0.81}{\mathrm{P}}_{0.19}$)${}_2$ [(d), (e), (f)]. Black and red lines are the transients for ${\mathcal{P}}^{+}$ and ${\mathcal{P}}^{-}$ probe polarizations, respectively, while green lines are the corresponding averages. The pump fluences used were $\mathcal{F}\sim 10\mu \mathrm{J}/{\mathrm{cm}}^2$(a) and $\mathcal{F}\sim 3\mu \mathrm{J}/{\mathrm{cm}}^2$ [(b)–(f)].

Figure 3

Photoinduced reflectivity transients at representative temperatures at $\mathcal{F}\sim 10\mu \mathrm{J}/{\mathrm{cm}}^2$ and 3.1-eV pump photon energy in ${\mathrm{EuFe}}_2{\mathrm{As}}_2$ [(a), (b)] and ${\mathrm{EuFe}}_2$(${\mathrm{As}}_{0.81}{\mathrm{P}}_{0.19}$)${}_2$ [(c), (d)]. Top and bottom panels correspond to ${\mathcal{P}}^{+}$ and ${\mathcal{P}}^{-}$ polarizations, respectively. The thin lines are the finite-excitation-pulsewidth double-exponential relaxation fits [16].

Figure 4

The relaxation time [(a), (c)] and amplitude [(b), (d)] of the sub-picosecond response as a function of temperature in ${\mathrm{EuFe}}_2{\mathrm{As}}_2$ [(a), (b)] and ${\mathrm{EuFe}}_2$(${\mathrm{As}}_{0.81}{\mathrm{P}}_{0.19}$)${}_2$ [(c), (d)] at $\mathcal{F}\sim 10\mu \mathrm{J}/{\mathrm{cm}}^2$ and 3.1-eV pump photon energy. The gray lines are the bottleneck model [16, 19] fits discussed in text.

Figure 5

The low-$T$ transients in ${\mathrm{EuFe}}_2{\mathrm{As}}_2$ at different fluences for ${\mathcal{P}}^{+}$ (a) and ${\mathcal{P}}^{-}$ (b) probe polarizations and 1.55-eV pump photon energy. To emphasize the low-$\mathcal{F}$ region the $\mathcal{F}$-normalized transients from panels (a) and (b) are shown in (c) and (d), respectively. The arrows indicate the direction of increasing fluence. Note that the overlap of the $\mathcal{F}$-normalized scans indicates a linear dependence.

Figure 6

The amplitude of the short delay extrema and the amplitude at the longest delay of the transients in ${\mathrm{EuFe}}_2{\mathrm{As}}_2$ as functions of fluence at $T=6$ K and 1.55-eV pump photon energy.

Figure 7

The low-$T$ transients in ${\mathrm{EuFe}}_2$(${\mathrm{As}}_{0.81}{\mathrm{P}}_{0.19}$)${}_2$ at different fluences for ${\mathcal{P}}^{+}$ (a) and ${\mathcal{P}}^{-}$ (b) probe polarizations and 1.55-eV pump photon energy. To emphasize the low-$\mathcal{F}$ region the $\mathcal{F}$-normalized transients from panels (a) and (b) are shown in (c) and (d), respectively. Arrows indicate the direction of increasing fluence. Note that the overlap of the $\mathcal{F}$-normalized scans indicates a linear dependence.

Figure 8

The amplitude of the sub-picosecond extrema of the transients in ${\mathrm{EuFe}}_2$(${\mathrm{As}}_{0.81}{\mathrm{P}}_{0.19}$)${}_2$ as functions of fluence at $T=5$ K and 1.55-eV pump photon energy. The inset shows the amplitude of the long delay extrema at low fluences including the data measured at 1.5 K using 3.1-eV pump-photon energy.

Figure 9

The isotropic (a) and anisotropic (b) transient components in ${\mathrm{EuFe}}_2$(${\mathrm{As}}_{0.81}{\mathrm{P}}_{0.19}$)${}_2$ at different fluences obtained from the data shown in Fig. 7. The $\mathcal{F}$-normalized transients from panels (a) and (b) are shown in (c) and (d), respectively, to emphasize the low-$\mathcal{F}$ region. Arrows indicate the direction of increasing fluence.

Figure 10

Photoinduced reflectivity transients at low $T$ and extremely low $\mathcal{F}$ in EuP-122 for ${\mathcal{P}}^{+}$ (a) and ${\mathcal{P}}^{-}$ (b) polarizations. The corresponding isotropic and anisotropic transient components are shown in (c) and (d), respectively. To reduce noise the traces were smoothed resulting in a reduction of the time resolution to $\sim 150$ fs.

Figure 11

Temperature dependence of the amplitudes of the isotropic part at 600-ps delay and the anisotropic part at 3-ps delay from Fig. 10, where the dotted and dashed vertical lines indicate $T_{\mathrm{Curie}}$ and $T_{\mathrm{c}}$ (onset) obtained from the susceptibility, respectively. The blue dotted line is the ZFC susceptibility from Fig. 1.