Electrical and Thermal Transport Properties of Layered Superconducting Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ Single Crystal

Abstract

We have synthesized single crystals of iron-based superconducting Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ and performed extensive measurements on their transport properties. A remarkable difference in the behavior and a large anisotropy between in-plane and out-of-plane resistivity was observed. Disorder could explain the in-plane square-root temperature dependence resistivity, and interlayer incoherent scattering may contribute to the out-of-plane transport property. Along the ab plane, the estimated value of the coherence length is 15.5 Å. From measurements of the upper critical magnetic field H_c2 (T ≥ 20 K), we estimate H_c2(0) = 313 T. Thermal conductivity for Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ is relatively small, which can be accounted for by the disorder in the crystal and the low-charge carrier density as verified by the Hall effect.

1. Introduction

The Ca₁₀(Pt_nAs₈)(Fe₂As₂)₅ (n = 3 or 4) system was first reported to exhibit superconductivity with a wide range of transition temperatures in 2011 [1,2,3]. After that, many studies have appeared in the past seven years [4,5,6,7,8]. Ca₁₀(Pt₄As₈)(Fe₂As₂)₅ has a layered structure consisting of superconducting FeAs layers separated by the spacer layers arranged as Ca–Pt₄As₈–Ca, which is shown in Figure 1a. This superconductor has been reported to crystallize in possible space groups including P4/n (tetragonal) [2,3], P2₁/n (monoclinic) [7], and $P\overline{1}$ (triclinic) [1,3]. The structure of the Pt₄As₈ spacer layer is similar to a square lattice of As atoms, one fifth of the As atoms are replaced by substitutional Pt1 atoms [2], and the same amount of Pt2 atoms are interstitial, which leads to the displacement of As atoms from their ideal positions to form As dimers [2]. Because of the constraint from the As–As dimers and the FeAs sublattices, substitutional Pt1 atoms can sit in plane while the interstitial Pt2 atoms sit on the site either above or below the plane against Ca ions.

Most FeAs-based superconductors need doping to induce the superconductivity; Ca₁₀(Pt₄As₈)(Fe₂As₂)₅ allows doping on the Ca site or on the Fe site. In particular, Pt doping on the Fe site could bring faults and disorder into the crystal and also adjust the T_c [2,5]. By studying its structure and physical properties, the origin of superconductivity may be elucidated. The interlayer distance of Ca₁₀(Pt₄As₈)(Fe₂As₂)₅ is reported to be ~10.5 Å [2,3], which is larger than many other FeAs-based superconductors, and thus its properties are expected to be anisotropic.

The electronic structure for Ca₁₀(Pt₄As₈)(Fe₂As₂)₅ has been studied via angle resolved photoemission spectroscopy (ARPES) [9]. Generally, there is a hole pocket and an electron pocket around the zone center. There are several electron pockets around the zone corner, among which two are suggested to be contributed by the FeAs layer, and several are contributed by Pt₄As₈ layers. Thus, the Pt₄As₈ layer is suggested to be metallic. Furthermore, the d_xz and d_yz bands are not degenerate at the Brillouin zone center (Γ point) and there is only hole-like Fermi surface at the Γ point originated from d_xy orbitals, it may be caused by the interaction between the Pt₄As₈ and FeAs layers [9].

In this paper, we studied the electrical properties of Ca₁₀(Pt₄As₈)(Fe₂As₂)₅ along its c-axis in order to understand how the interaction between Pt₄As₈ and FeAs layers influences the physical properties and anisotropy in this system. In addition, thermal conductivity has not yet been reported as another approach in measuring transport properties, and thus we also provide more extensive measurements and discussion on this approach.

2. Material Preparation and Characterization Methods

The preparation process is similar to descriptions provided elsewhere [2,4]. To grow single crystals of Ca₁₀(Pt₄As₈)(Fe_2−xPt_xAs₂)₅, stoichiometric amounts of high purity Ca shot (99.999%, Alfa Aesar, Haverhill, MA, USA), Fe powder (99.95%, Alfa Aesar, Haverhill, MA, USA), Pt powder (99.95%, Alfa Aesar, Haverhill, MA, USA), and As powder (99.999%, Alfa Aesar, Haverhill, MA, USA) were mixed in the ratio of 17:14:9:31. The mixture was placed in an Al₂O₃ crucible (Gongtao Ceramics, Shanghai, China) and sealed in a quartz tube (Gongtao Ceramics, Shanghai, China) under vacuum. The whole ampule was heated to 700 °C in a box furnace (Henan Sante, Luoyang, China) at a rate of 140 °C/h and kept at this temperature for 5 h. It was further heated to 1100 °C at a rate of 80 °C/h where it was held for 50 h and then cooled to 1050 °C at a rate of 1.25 °C/h. It was further cooled to 500 °C in the next 100 h and finally cooled down to room temperature naturally by switching the power off. The cooling process is crucial because the mixture melts when it heats up and forms in a stable phase when it cools down. Shiny plate-like single crystals were obtained with a typical size of 3 mm × 3 mm × 0.8 mm, and a photo of the as prepared sample is shown in Figure 1b.

Crystal structure and phase purity were checked by both single-crystal and powder X-ray diffraction, which were carried out on a Rigaku-D/max X-ray diffractometer (XRD) (Rigaku, Tokyo, Japan) using Cu-Kα radiation (λ = 1.54056 Å). Powder X-ray diffraction employed the powder grounded from the as-grown single crystals. The crystal used for this experiment was selected from the same batch as that used for composition and physical properties measurements. The chemical composition of the sample was identified by Energy Dispersive X-Ray Spectroscopy (EDX) (Hitachi, Tokyo, Japan). Electrical resistivity, magnetoresistance, Hall effect, thermal conductivity, and Seebeck coefficient measurements were performed in a Quantum Design PPMS System. Both ab plane and c-axis resistivities were measured using the standard four-probe technique.

3. Results and Discussion

Figure 1c shows the X-ray diffraction spectrum for the as-grown thin, plate-like single crystal along the c-axis. Note that only (00l) peaks are observed and the pattern matches very well with that previously reported in [4]. A powder XRD spectrum is shown in Figure 1d. The result exhibits the peaks from diffractions not only (00l) planes but also (hkl) planes with nonzero h, k, l. X-ray structure determination confirms that the lattice parameters and angles for the sample are a = 8.7548(12) Å, b = 8.7642(10) Å, c = 10.69005(8) Å, and α = 94.674(9)°, β = 104.396(8)°, γ = 90.037(10)°, respectively, indicating our sample crystallizes in a triclinic structure with space group symmetry $P\overline{1}$ at room temperature. No other impurities such as FeAs and PtAs₂ were observed in any X-ray spectrum.

The chemical composition was identified by the EDX measurement. A scanning electron microscope (SEM) image measured on the surface of the single crystal is shown in the Figure 2 inset. For each sample, five locations scattered on the surface were chosen for EDX measurement. An average of these scans was calculated. The results of the different samples are consistent with each other, implying homogenous growth was acquired throughout the batch. The average deviation is 3% for Ca, 5% for Pt, and 1% for Fe. The measured composition of the single crystals is Ca₁₀Pt_5.4Fe_8.6As₁₈, corresponding to a formula of Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ if considering the Pt substitution effect.

Figure 3 displays the temperature dependence of the in-plane (ρ_ab) and out-of-plane resistivity (ρ_c) of Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ single crystals between 2 K and 300 K. The magnitude of ρ_ab and its overall features are similar to the previous reports [2]. The normal-state ρ_ab decreases with decreasing T ($\frac{d{\rho }_{\mathrm{ab}}}{dT}>0$), showing a metallic behavior. It drops sharply at T_{c_onset} = ~34 K, reaching zero-resistivity state at 31.2 K. The transition width in temperature is less than 3 K, indicating that our sample has high quality and spatial homogeneity. The residual resistivity ratio (RRR) (300K)/ρ(T_{c_onset}) is 2.4. Relatively large T_{c_onset} and small RRR reflects the presence of Pt doping on the Fe site in FeAs layers as suggested in [1,2,3]. However, the out-of-plane resistivity ρ_c behaves strikingly different with ρ_ab; it exhibits nearly independent T at higher temperatures and starts to increase with decreasing T ($\frac{d{\rho }_{\mathrm{c}}}{dT}<0$) below 200K, showing a nonmetallic behavior. It then goes through a sharp peak around 38 K before dropping to zero. The anisotropic property between ρ_ab and ρ_c has not been noted in previous reports [1,2,4]. We acknowledge that this ρ_c–T profile is reproducible and it should be an intrinsic property of Ca₁₀(Pt₄As₈)(Fe_2−xPt_xAs₂)₅.

Due to the layered crystal structure of Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅, the in-plane resistivity may be understood as a net resistivity of parallel connected resistors. A schematic of such a system is shown in Figure 4a. Here, we do not consider the resistance from the Ca layer because the FeAs and Pt₄As₈ layers are the main contributors of the electronic structure around the Fermi surface [9]. The net resistance (R) can be written as:

$$R=1/\left[n\left(\frac{1}{R_{FeAs}}+\frac{1}{R_{P{t}_{4}A{s}_8}}\right)\right]$$

(1)

in which n is the number of Pt₄As₈–FeAs layers, and R_FeAs and R_Pt4As8 are the resistance of the FeAs layer and Pt₄As₈ layer, respectively. If the Pt₄As₈ layer is semiconducting or insulating, it should have a much larger resistance than the FeAs layer. Then, the net resistance R should be dominated by the resistance of the Pt₄As₈ layer and exhibit nonmetallic behavior. Yet, according to our results, the normal state ρ_ab shows metallic behavior, indicating the spacer layer Pt₄As₈ is metallic, which agrees with previous reports [2,4,9].

The normal state ρ_ab of some traditional FeAs-based superconductors shows metallic behavior and may have linear dependence on temperature due to the inelastic scattering originating from electron-phonon interactions. In some cuprates, ρ_ab has T^1.5 dependence, implying the Fermi-liquid behavior. Yet, the temperature dependence of the normal state ρ_ab of Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ is neither linear, quadratic temperature dependent, nor T^1.5 dependent, as observed in cuprates or other FeAs-based compounds [10]. The normal state ρ_ab from 50 K to 300 K can be well fitted by ${\rho }_{ab}=A+B\sqrt{T}$ with A = 0.079 ± 0.008 mΩ cm and B = 0.031 ± 0.003 mΩ cm K^−1/2. The red solid line in Figure 3a is the fit to the experimental data of the normal state ρ_ab. All reported ρ_ab for Ca₁₀Pt₄As₈(Fe₂As₂)₅ has the similar square-root temperature dependence regardless of T_c [1,2,4], and thus, it may be an intrinsic property to Ca₁₀Pt₄As₈(Fe₂As₂)₅.

The square-root temperature dependence of electrical resistivity is normally expected at low temperatures in disordered metals and degenerate semiconductors because of interference with scattering by impurities [11,12]. The studies on the Fermi surface of Ca₁₀Pt₄As₈(Fe₂As₂)₅ imply that the Pt₄As₈ layer contributes several electron pockets. Thus, there may be competition between the Pt₄As₈ layer and the negatively-charged FeAs layer for electrons, leading to possible Pt deficiency in the Pt₄As₈ layer and partial substitution of Fe by Pt in the Fe_1−xPt_xAs layer [3,5,13,14]. As a result, charge carriers in both the Pt₄As₈ and Fe_1−xPt_xAs layers could experience the effect of disorder, which may be the reason why the in-plane electrical resistivity has the square-root temperature dependence.

As shown in Figure 3b, ρ_c is much larger than ρ_ab in the normal state. It can be understood with a schematic plot of the net resistance R = n (R_FeAs + R_Pt4As8) along the out-of-plane direction shown in Figure 4b. Different from the normal state ρ_c of some cuprates which can be described using the equation ρ_c = A/T + B × T exhibiting the non-Fermi liquid behavior, the nonmetallic normal state ρ_c of Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ can be appropriately fitted using the equation ${\rho }_c=A^{\prime }+B^{\prime }\sqrt{T}+C^{\prime }{T}^{-1}$. The solid line in Figure 3b shows the fit of ρ_c from 90 K to 300 K. The fitting parameters are calculated as follows: A′ = 2.642 ± 0.009 mΩ cm, B′ = 0.308 ± 0.006 mΩ cm K^−1/2, and C′ = 667.7 ± 15 mΩ cm K.

The square-root temperature dependent component in ρ_c is due to the in-plane scattering, which may be intrinsic as Anderson proposed for high-T_c cuprates [15] or extrinsic due to a possible stacking fault [2]. The T⁻¹ term can be attributed to interlayer incoherent scattering, similar to the properties of high-T_c cuprates [15] and (Sr₄V₂O₆)Fe₂As₂ [16]. It implies the spacer layer Ca₁₀Pt₄As₈ is not superconducting in nature, so Josephson junctions are formed along the c-axis of Ca₁₀Pt₄As₈(Fe₂As₂)₅ by stacking of the FeAs superconducting layer, the Ca₁₀Pt₄As₈ semiconducting layer, and the FeAs superconducting layer.

The resistivity anisotropy γ is calculated by $\gamma =\sqrt{{\rho }_c/{\rho }_{ab}}$ and it increases from ~4.1 at 300 K to ~7.9 around T_c as displayed in Figure 5. The γ of Ca₁₀Pt₄As₈(Fe₂As₂)₅ is larger than that of LaFeAsO_0.9F_0.1 [17]. Relatively large anisotropy and a conspicuous difference in the behavior of ρ_c and ρ_ab suggests rgw strong 2D nature of electronic structures in Ca₁₀Pt₄As₈(Fe₂As₂)₅, which is rarely encountered in traditional Fe-based superconductors.

The resistivity anisotropy (γ) in the superconducting state of Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ was evaluated through resistivity measurements with an applied magnetic field H. Figure 6 shows the temperature dependence of ρ_ab and ρ_c under H along both the c-axis and the ab plane.

In the case of I//ab and H//c-axis, it can be seen that the zero resistivity transition temperature (T_c0) decreases with the rising applied magnetic field accompanied by an increase in the transition width. The T_c0 is 18 K and the transition width ∆T_c is 16 K for H = 14 T, considering T_{c_onset} is 34 K. This type of superconducting transition broadening with increasing magnetic field (H) is rarely seen in Fe-based superconductors but very common in cuprates due to the presence of strong thermal vortices fluctuations [18].

Interestingly, the superconducting transition width ∆T_c does not change with an applied field when I//c and H//ab plane (Figure 6b) and H pushes both T_{c_onset} and T_c0 to the lower temperature by the same amount. This shift is also much smaller than that of ∆T_c for cases I//ab and H//c.

As shown in Figure 7, the H_c2 (T) of Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ was extracted by identifying the T_c(H) at which resistivity drops to 90% ρ_n, 50% ρ_n, and 10% ρ_n. ρ_n is the normal resistivity before superconducting transition, as indicated by dashed lines in Figure 6.

The temperature dependence of the H_c2 anisotropy parameter ${\Gamma }_H=\frac{H_{c2}^{ab}}{H_{c2}^c}$ obtained from Figure 7 is presented in Figure 8. Γ_H taken at 90% resistivity drop reaches 8 near T_c, which is very close to the normal-state γ we obtained previously in the resistivity measurements. Γ_H varies from 4 to 6 using a 50% criteria as compared to 122 type. For example, with Γ_H = ~2 for BaFe₂As₂ [19,20], the H_c2 anisotropy of Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ is much larger.

It also can be noted from Figure 7b that $H_{c2}^{ab}\left(T\right)$ increases almost linearly with decreasing temperature, while $H_{c2}^c\left(T\right)$ exhibits an upward shape with a steep increase at low temperatures. This phenomena quite possibly originates from the multi-band effect such as in the NdFeAsO_0.7F_0.3 and MgB₂ systems [21,22]. Thus, the positive curvature of $H_{c2}^c\left(T\right)$ reflects a multi-band nature in electrical structure for Ca₁₀(Pt₄As₈)(Fe₂As₂)₅.

Upper critical field H_c2 is an important parameter for all superconductors especially in their practical applications. We estimated H_c2 at 0 K using the Werthamer–Helfand–Hohenberg (WHH) approximation [23]:

$$H_{c2}\left(0\right)=-0.69T_c\times {\left|\frac{d{H}_{c2}}{dT}\right|}_{T_c}$$

(2)

The corresponding coherence lengths are calculated via the Ginburg Landau (GL) formula:

$$\{\begin{array}{c}{\xi }_{\mathrm{ab}}\left(0\right)=\sqrt{\frac{{\mathsf{\phi }}_0}{2\mathsf{\pi }{H}_{c2}^c\left(0\right)}}\\ {\xi }_{\mathrm{c}}\left(0\right)=\frac{{\mathsf{\phi }}_0}{2{\mathsf{\pi }\mathsf{\xi }}_{\mathrm{ab}}{H}_{c2}^{ab}\left(0\right)}\end{array}$$

(3)

where ${\phi }_0=2.07\times {10}^{-15}$ Wb. The obtained results are listed in

Table 1. The upper critical field and coherence length of Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅.

Criteria in Determining Hc2	$\frac{\mathit{d}{\mathit{H}}_{\mathit{c}2}^{\mathit{c}}}{\mathit{d}\mathit{T}}{\mathit{T}}_{\mathit{C}}\left(\mathit{T}/\mathit{K}\right)$	${\mathit{H}}_{\mathit{c}2}^{\mathit{c}}\left(0\right)$ (Tesla) (WHH Approach)	$\frac{\mathit{d}{\mathit{H}}_{\mathit{c}2}^{\mathit{a}\mathit{b}}}{\mathit{d}\mathit{T}}{\mathit{T}}_{\mathit{C}}\left(\mathit{T}/\mathit{K}\right)$	${\mathit{H}}_{\mathit{c}2}^{\mathit{a}\mathit{b}}\left(0\right)$ (Tesla) (WHH Approach)	ξab(0) (Å)	ξc(0) (Å)
0.9 ρn	−6.5 ± 0.2	138 ± 5	−14.1 ± 0.1	313 ± 6	15.5 ± 0.2	6.8 ± 0.3
0.5 ρn	−0.89 ± 0.1	19.7 ± 2	−4.4 ± 0.1	99 ± 2	40.9 ± 0.5	8.1 ± 0.3
0.1 ρn	−0.74 ± 0.1	16.4 ± 1	−3.8 ± 0.1	84 ± 3	44.8 ± 0.4	8.7 ± 0.5

WHH: Werthamer–Helfand–Hohenberg.

. Although all derived values depend on how H_c2(T) is extracted, the large difference of $H_{c2}^{ab}$ and $H_{c2}^c$ implies Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ exhibits a large anisotropy in its superconducting state. Furthermore, the values of ξ_c(0) derived from different criteria are all less than 10.69005(8) Å, i.e., the length of c-axis of the unit cell, suggesting that Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ meets the dimensional requirement of fabricating the intrinsic Josephson junctions (iJJs).

For a normal metal with Fermi liquid behavior, the Hall coefficient (R_H) is independent of temperature. The situation is more complex if the material has multi-band or non-Fermi liquid behavior such as, for example, cuprates or heavy fermions. Their Hall coefficient exhibits strong temperature and doping dependencies. The $H_{c2}^c\left(T\right)$ curves show the multi-band nature of Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅, so the Hall effect was measured by applying a magnetic field H⊥I//ab, and the data is presented in Figure 9.

The transverse R_H remains negative at all temperatures above T_c, indicating that the charge carrier is dominated by electrons. The magnitude of R_H increases with decreasing temperature, suggesting a multi-band and non-Fermi liquid behavior. Even though the doping in the FeAs layers is considered to be isovalent (i.e., Pt²⁺ replaces Fe²⁺) [2,5], it may still influence R_H. Our R_H(T) is similar with previous reports [1,2], but it exhibits stronger temperature dependence, which suggests that the carrier concentration cannot be solely determined by the Pt concentration in this material.

If a single band model is adopted, the corresponding carrier concentration (n) could be calculated with R_H via n = −1/eR_H. The temperature dependence of n is shown as the red line in Figure 9. Generally, n decreases monotonically with decreasing T. The calculated n is about 2.5 × 10²² cm⁻³ at 300 K. It is comparable to other FeAs-based superconductors; for example, n of SrFe₂As₂ is about 1.52 × 10²² cm⁻³ at 300 K [24]. The carrier concentration indicates that the normal state of Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ behaves as a good metal. Before superconducting transition, the carrier concentration drops to 0.5 × 10²² cm⁻³; however, it is still an order of magnitude larger than that of Ca10-3-8 (~10²¹ between 2 K and 300 K) [2]. This is because the one extra Pt atom in the Pt₄As₈ intermediary layer exceeds the Zintl valence satisfaction requirement and introduces redundant electrons into the system, leading to enhanced metallicity and a relatively higher T_c.

A different method of probing conduction mechanism is provided by thermal conductivity. We measured the temperature dependence of thermal conductivity and the Seebeck coefficient for Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ from 2 K to 350 K, and the results are shown in Figure 10a.

The Seebeck coefficient (S) is negative at the measured temperature range above T_c, with a value of −25.5604 μV/K at 300 K and a minimum value of −29.1343 μV/K near 154 K resulting from a phonon-drag contribution. This confirms that the electron-type charge carrier dominates in Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅, which is consistent with the Hall effect measurement. Then, S starts to increase with decreasing temperature and reaches zero steeply at the superconducting transition temperature. In addition, the Seebeck curve shows no anomalous enhancements associated with the crystal structure or spin density wave (SDW) transitions widely observed in undoped compounds such as SmFeAsO and BaFe₂As₂ [25,26], suggesting that there were no corresponding transitions in our sample at the temperatures measured.

With regard to the thermal conductivity (κ), it drops monotonously above T_c when temperature is lower than T_c, κ and decreases sharply with the opening of the superconducting gap. Before the drastic drop, κ first displays an abrupt increase with a bump feature upon entering the superconducting (SC) state, and a similar behavior is commonly observed in the cuprate superconductors, e.g., YBa₂Cu₃O₇ [27] and Bi₂Sr₂CaCu₂O₈ [28]. The enhancement of κ below T_c reflects the increase of the phonon mean free path by the condensation of charge carriers. Phonons then cease to dissipate their momentum in collisions with such a condensate.

The electronic contribution to thermal conductivity (${\kappa }_e$) above T_c could be evaluated by the Wiedemann–Franz law ${\kappa }_{el}=\sigma LT$, where σ is the electrical conductivity of a metal and T is the temperature, L, known as the Lorenz number, which is equal to:

$$L=\frac{{\pi }^2}{3}{\left(\frac{{\mathrm{k}}_{\mathrm{B}}}{\mathrm{e}}\right)}^2=2.44\times {10}^{-8} {\mathrm{W}\mathrm{\Omega }\mathrm{K}}^{-2}$$

(4)

in which e is elementary charge and k_B is the Boltzmann constant. The calculated results are presented in Figure 10b. ${\kappa }_{el}$ is smaller by about 5 orders of magnitude than κ as shown in Figure 10a. Thus, heat in Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ is mainly carried by phonons, and the electron contribution can be negligible. This is wholly different from the copper oxide superconductors. Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ is a relatively low-charge carrier density system.

It is worth noting that the κ value for Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ is smaller than that for LaFeAsO_0.89F_0.11 [29]. We know that the main scattering mechanisms for phonons in crystal are carriers and structural defects, and intrinsic phonon–phonon scattering only exists in clean materials. While in the Ca₁₀(Pt₄As₈)(Fe₂As₂)₅ system, both the off-centered Pt atoms in the Pt₄As₈ plane and the substitutional Pt atoms in the FeAs plane introduce disorder into the crystal, which causes phonons to be strongly scattered and the crystal lattice vibration to localize, resulting in the rather smaller thermal conductivity of Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅.

4. Conclusions

In conclusion, X-ray diffraction, resistivity, Hall effect, Seebeck coefficient, and thermal conductivity measurements were performed on high quality Ca₁₀(Pt₄As₈)((Fe_0.86Pt_0.14)₂As₂)₅ single crystals. We observed metallic in-plane resistivity but non-metallic out-of-plane resistivity for Ca₁₀Pt₄As₈(Fe₂As₂)₅. The anisotropic property is unusual, and its normal state resistivity exhibits a large anisotropy (~8) near T_c, making it one of the most anisotropic FeAs-based superconductors. The normal state in-plane resistivity has a square-root temperature dependence which is intrinsic to Ca₁₀Pt₄As₈(Fe₂As₂)₅. The interlayer incoherent scattering contributes to the out-of-plane transport property. A large coherence length along the ab plane and upper critical field were observed. Disorder and low-charge carrier density in the crystal may account for the relatively small thermal conductivity. The layered structure and the relatively higher transition temperature with the large electrical transport anisotropy of Ca₁₀Pt₄As₈(Fe₂As₂)₅ implies it may be a new good candidate for and have potential application in the fabrication of high frequency microelectronic devices such as next generation intrinsic Josephson junctions (IJJs).