Disappearance of static magnetic order and evolution of spin fluctuations in ${\text{Fe}}_{1+\delta }{\text{Se}}_x{\text{Te}}_{1-x}$

We report neutron-scattering studies on static magnetic orders and spin excitations in the Fe-based chalcogenide system ${\text{Fe}}_{1+\delta }{\text{Se}}_x{\text{Te}}_{1-x}$ with different Fe and Se compositions. Short-range static magnetic order with an in-plane wave vector near the (0.5,0) (using the two-Fe unit cell), together with strong low-energy magnetic excitations is found in all nonsuperconducting samples for Se doping up to 45%. When the static order disappears and bulk superconductivity emerges, the spectral weight of the magnetic excitations shifts to the region of reciprocal space near the in-plane wave vector (0.5, 0.5), corresponding to “collinear” spin correlations. Our results suggest that there is a strong correlation between superconductivity and the character of the magnetic order/fluctuations in this system. Excess Fe appears to be important for stabilizing the magnetic order that competes with superconductivity.

Figure 1

(Color online) (a) Schematic of the collinear C-type AF spin structure with scattering intensities mainly around (0.5,0.5). The square shows a unit cell with two Fe atoms. (b) Schematic of the bicollinear E-type AF spin structure, which contribute mainly to scattering intensities near (0.5,0). (c) The schematic of the neutron-scattering measurements in the $\left(HK0\right)$ zone. Dashed lines denote linear scans performed across (i) (0.5,0) and (ii) (0.5,0.5) in the text.

Figure 2

(Color online) ZFC magnetization measurements by SQUID for SC30 (red), NSC30 (blue), NSC45 (green), and SC50 (black). The inset shows the same data from the nonsuperconducting samples with different scales.

Figure 3

(Color online) Elastic neutron-scattering measurements performed on SC30 (red circle), NSC30 (blue triangle), NSC45 (green square), and SC50 (black diamond) near (0.5,0,0.5). [(a) and (d)] Intensity profiles along [100] direction ($H$ scans) at $T=4\text{K}$. [(b) and (e)] Scans along [001] direction ($L$ scans) at $T=4\text{K}$. The horizontal bars represent the $H$ and $L$ resolutions at [(a) and (b)] SPINS and [(d) and (e)] BT-9. The solid lines are based on least-square fits to the data with single Gaussian peak and sloping background. (c) shows the magnetic peak intensity (from fitted Gaussian peak intensity) at (0.5,0,0.5) vs temperature. Corresponding scans measured at $T=34\text{K}$ are used as background and have been subtracted from all the data shown. The inset in (c) shows the 4 K (open) and 34 K (close) raw data used in (b) for NSC45. [(d)–(f)] Data from Ref. 28. The error bars represent the square root of the number of counts.

Figure 4

(Color online) Constant-$Q$ scans at (0.5,0,0) taken at (a) 4 K (1.5 K for SC30 and NSC30) and (b) 25 K. A background, determined from constant-energy scans as in Fig. 5 has been subtracted from all data sets. The error bars correspond to the square root of the number of counts.

Figure 5

(Color online) Magnetic excitations for ${\text{Fe}}_{1+\delta }{\text{Se}}_x{\text{Te}}_{1-x}$ measured around (0.5,0,0). The left and right columns show the magnetic peak profiles for 4 K (1.5 K for SC30 and NSC30) and 25 K, respectively. Constant-energy scans at (0.5, $K$, 0) [as shown by (i) in Fig. 1c] at [(a) and (b)] $\hslash \omega =0.5\text{meV}$, [(c) and (d)] $\hslash \omega =2\text{meV}$, and [(e) and (f)] $\hslash \omega =5\text{meV}$. The solid lines are based on least-square fits to the data with one main Gaussian peak at $K=0$, plus sometimes two small Gaussian peaks at $K\pm 0.5$, which represent intensities from $Q=\left(0.5,0.5\right)$, and a constant background. The fitted $K$-independent background has been subtracted from all data sets. The error bars correspond to the square root of the number of counts.

Figure 6

(Color online) Constant-$Q$ scans at (0.5,0.5,0) taken at (a) 4 K (1.5 K for SC30 and NSC30) and (b) 25 K. A background, determined from constant-energy scans as in Fig. 7 has been subtracted from all data sets. The error bars corresponds to the square root of the number of counts.

Figure 7

(Color online) Magnetic excitations for ${\text{Fe}}_{1+\delta }{\text{Se}}_x{\text{Te}}_{1-x}$ measured around (0.5,0.5,0). The left and right columns show the magnetic peak profiles for 4 K (1.5 K for SC30 and NSC30) and 25 K, respectively. Constant-energy scans at (0.5,0.5,0), taken along the transverse direction [as shown by (ii) in Fig. 1c] at [(a) and (b)] $\hslash \omega =5\text{meV}$, [(c) and (d)] $\hslash \omega =6.5\text{meV}$, and [(e) and (f)] $\hslash \omega =12\text{meV}$. The solid lines are based on least-square fits to the data with two Gaussian peaks and a constant background. The fitted constant background has been subtracted from all data sets. The error bars corresponds to the square root of the number of counts.