First-order structural phase transition in ${\text{CaFe}}_2{\text{As}}_2$

${\text{CaFe}}_2{\text{As}}_2$ has been synthesized and found to form in the tetragonal ${\text{ThCr}}_2{\text{Si}}_2$ structure with lattice parameters $a=3.912\left(68\right)\text{Å}$ and $c=11.667\left(45\right)\text{Å}$. Upon cooling through 170 K, ${\text{CaFe}}_2{\text{As}}_2$ undergoes a first-order structural phase transition to a low-temperature orthorhombic phase with a 2–3 K range of hysteresis and coexistence. This transition is clearly evident in microscopic, thermodynamic, and transport measurements. ${\text{CaFe}}_2{\text{As}}_2$ is the third member of the ${\text{AFe}}_2{\text{As}}_2$ $\left(A=\text{Ba},\text{Sr},\text{Ca}\right)$ family to exhibit such a dramatic phase transition and is a promising candidate for studies of doping induced superconductivity.

Figure 1

(Color online) Rocking curve through the (1 1 10) reflection of the ${\text{CaFe}}_2{\text{As}}_2$ single crystal used for the x-ray diffraction study. Inset: picture of a ${\text{CaFe}}_2{\text{As}}_2$ single crystal on millimeter grid paper. The crystallographic $c$ axis is perpendicular to the plate of the crystal. The small droplets on the surface are residual Sn flux.

Figure 2

(Color online) Powder x-ray diffraction spectrum of ground single-crystal ${\text{CaFe}}_2{\text{As}}_2$. Note that Si powder was added as a standard and Sn is present from residual flux on surface of samples used.

Figure 3

Temperature-dependent electrical resistivity of ${\text{CaFe}}_2{\text{As}}_2$, with current flowing within the basal plane, for $H\parallel c$, $H=0$, and 140 kOe. Upper inset: hysteresis in temperature-dependent resistivity near the 170 K phase transition (the arrows indicate increasing and decreasing temperature scans). Lower inset: magnetoresistance for $H\parallel ab$ and $H\parallel c$ at $T=2\text{K}$.

Figure 4

Temperature-dependent magnetic susceptibility of ${\text{CaFe}}_2{\text{As}}_2$ for applied field parallel to and perpendicular to the crystallographic $c$ axis. Upper inset: enlarged transition region. Lower inset: anisotropic magnetization for applied field parallel and perpendicular to the $c$ axis for $T=300\text{K}$. As described in text, the susceptibility was determined by the difference between 30 and 50 kOe magnetization runs so as to eliminate weak ferromagnetic contributions.

Figure 5

(Color online) Temperature-dependent specific heat of ${\text{CaFe}}_2{\text{As}}_2$. Upper inset: enlargement of data in the vicinity of the 170 K phase transition. Lower inset: $C_p/T$ plotted as a function of $T^2$; line—linear fit.

Figure 6

Longitudinal $\left(00\xi \right)$ scans through the position of the tetragonal (0 0 10) reflection for selected temperatures. The lines represent fits to the data to obtain the reflection positions and intensities shown in Fig. 8. The offset between every data set is 2000 counts/s. In the coexistence range, we point out that the strong change in the position of the peaks results mainly from misalignment related to the strain occurring at the phase transition in combination with the lattice-parameter changes between both phases.

Figure 7

Transverse $\left(\xi \xi 0\right)$ scans through the position of the tetragonal (1 1 10) reflection for selected temperatures. The lines represent to the data to obtain the reflection positions shown in Fig. 8. The offset between every data set is 200 counts/s. In the coexistence range, we point out that the strong change in the position of the peaks results mainly from misalignment related to the strain occurring at the phase transition in combination with the lattice-parameter changes between both phases.

Figure 8

Lattice parameters for the tetragonal and orthorhombic phases as extracted from the data shown in Figs. 6, 7 for the (0 0 10) reflection and the (1 1 10) reflection. To allow direct comparison of both phases the lattice parameter $a$ of the tetragonal phase is given as $a_{\text{HT}}=\sqrt{2}a$. The inset shows the intensity of the reflections related to the lattice parameter $c$ (shown in Fig. 6 in both phases). The error bars represent only the relative precision of the measurements.