Fabrication and characterization of SmO0.7F0.2FeAs bulk with a transition temperature of 56.5 K

The superconductivity of iron-based superconductor SmO0.7F0.2FeAs was investigated. The SmO0.7F0.2FeAs sample was prepared by the two-step solid-state reaction method. The onset resistivity transition temperature is as high as 56.5 K. X-ray diffraction (XRD) results show that the lattice parameters a and c are 0.39261 and 0.84751 nm, respectively. Furthermore, the global Jc was more than 2.3 × 10 A/cm at T = 10 K and H = 9 T, which was calculated by the formula of Jc = 20ΔM/[a(1  a/(3b))]. The upper critical fields, Hc2 ≈ 256 T (T = 0 K), was determined according to the Werthamer-Helfand-Hohenberg formula, indicating that the SmO0.7F0.2FeAs was a superconductor with a very promising application.


Introduction
Since the discovery of layered copper oxide superconductors [1][2], some researches have been focused on exploring higher T c materials. MgB 2 , which was discovered in the early 2001 [3][4], aroused people to search for new superconductors due to its unique performance compared with the traditional low-temperature and high-temperature oxide conductors. Recently, superconductivity in iron and nickelbased layered quaternary compounds has been reported: They were LaOFeP (T c -4 K) [5], LaONiP (T c -3 K) [6], and LaOFeAs (T c -26 K) [7]. Moreover, the T c of iron-based superconductors has been greatly improved. It has been increased as high as 50 K in ReO 1x F x FeAs, with the replacement of La by other rare-earth elements such as Pr, Nd, Sm, Ce, and Gd [8][9][10][11][12][13][14][15][16]. These quaternary superconductors have the tetragonal layered ZrCuSiAs structure, P4/nmm symmetry, and alternating FeAs and ReO layers [7][8]17], similar to cuprates. In this paper, we reported the preparation and superconductivity of SmO 0.7 F 0.2 FeAs, whose onset resistivity transition temperature is as high as 56.5 K.

Sample preparation
The Sm chip, As, Fe, and Fe 2 O 3 powders were all with purity better than 99.99%; the purity of FeF 3 was 97%. The precursor SmAs was home-synthesized by reacting mixtures of Sm chips and As powder in an evacuated quartz tube (the vacuum was better than 10 -5 Pa) at 600 ºC for 5 h, then followed by heating to 900 ºC holding for 20 h. The excess of 1 wt.% was added to compensate for the loss of As by volatilization [14].
The SmO 0.7 F 0.2 FeAs sample was prepared by the two-step solid-state reaction method with the starting materials Fe, Fe 2 O 3 , FeF 3 , and the obtained SmAs powders wiht the ratio being 14:7:2:30. The stoichiometric mixture of the starting materials was ground thoroughly and pressed into pellets. The obtained pellets were sintered in an evacuated quartz tube at 900ºC for 5h, then 1200ºC for 40 h, and then the samples were furnace-cooled to room temperature. The vacuum in the evacuated quartz tube was at the order of 10 -5 Pa.
The phase purity and structural identification were characterized by powder X-ray diffraction (XRD) analysis on an MXP18A-HF type diffractometer with Cu K α radiation from 20° to 80° with a step of 0.02°. The resistivity of the sample was measured by the standard four-probe method from 30 to 300 K. The DC magnetization was measured using a Quantum Design MPMS XL-1 system. Fig. 1 shows the comparison of XRD patterns for both nominal SmOFeAs and SmO 0.7 F 0.2 FeAs samples; the nominal SmOFeAs sample was also synthesized by the two-step solid-state reaction method. The XRD results indicate that the main phase of non-doped and F-doped samples have the same structure with slight impurity phases. The major impurity phases such as unreacted SmAs are observed in our present samples, in addition to slight impurity phase of SmOF in F-doped SmO 0.7 F 0.2 FeAs sample. The impurity phases have been identified to be some by-products, which do not have superconductivity. The lattice parameters for the nominal SmO 0.7 F 0.2 FeAs are a = 0.39261 nm, c = 0.84751 nm, which are smaller than those of nominal SmO 0.7 F 0.3 FeAs in Ref. [18]. For the SmO 0.7 F 0.2 FeAs, due to the lack of F, its lattice constants are even smaller. In Fig. 2, it can be seen from the inset that a clear superconducting onset transition (T c (onset)) occurred at 56.5 K, and a zero resistivity transition (T c (zero)) at 53 K for the nominal SmO 0.7 F 0.2 FeAs was obtained, which is about 1.5 K higher than that of Ref. [12]. This may be due to the smaller lattice constant of SmO 0.7 F 0.2 FeAs sample described above. The SmO 0.7 F 0.2 FeAs sample has a metallic behavior when the temperature goes up to 300 K. The residual resistivity ratio RRR =ρ (300 K) /ρ (55 K) was about 5.3, and the transition width ΔT is about 3 K, which shows that the SmO 0.7 F 0.2 FeAs sample has a high purity, but the fish tail indicate the existence of some non-superconducting impurities. For an experimental cycle, the SmO 0.7 F 0.2 FeAs sample was cooled down to 30 K in zero field cooling (ZFC), and the data were gathered when warmed in an applied field, then cooled again under an applied field (field cooling, FC), and measured when warming up. The DC-susceptibility data (measured under a magnetic field of 1591.5 A/m) of the SmO 0.7 F 0.2 FeAs are shown in Fig. 3. The magnetic onset T c for the SmO 0.7 F 0.2 FeAs sample is at 56.5 K, which is consistent with the resistivity result. The existence of a superconducting phase is confirmed by the Meissner effect by cooling it in a magnetic field while the resistance goes to zero.

Figs. 4(a) and 4(b) show the typical M-H hysteresis loops
of SmO 0.7 F 0.2 FeAs at 10 and 30 K, respectively. The gap in the loop drops dramatically in the low-field region with increased applied field, but it rapidly reaches a slowly decreasing state over a wide range of applied magnetic field. The rapid decrease in the wide hysteresis loop is a typical abnormal behavior for granular superconductors, since the weak superconductivity in the grain boundaries is extremely sensitive to the magnetic field. The global J c of the samples is calculated on the basis of the following formula: where ΔM is the height of the magnetization loop and a and b are the dimensions of the sample perpendicular to the magnetic field with a < b. Fig. 5 presents the J c -H curves of SmO 0.7 F 0.2 FeAs at 10 and 30 K, respectively. It shows that J c drops drastically in the low-field region with the increase of the applied field from 0 to 3 T, which illustrates that the grain boundaries are a weak-link behavior for the granular SmO 0.7 F 0.2 FeAs sample, but J c rapidly reduces and then becomes stable over a wide range of applied magnetic field from 3 T to 9 T. The J c of SmO 0.7 F 0.2 FeAs is more than 2.3 × 10 5 A/cm 2 at 10 K and has a very weak dependence on the field, which shows that SmO 0.7 F 0.2 FeAs has a fairly large pinning force in the grain. But at high temperature, J c drops drastically with the increase of the applied field. This shows that the magnetic flux pinning ability has been badly damaged in the sample at the high emperature. Fig. 6 shows the temperature dependence of electrical resistivity of the superconducting SmO 0.7 F 0.2 FeAs sample at different magnetic fields. An empirical criterion for the onset of superconductivity at a given applied field would be the deviation from the apparent linear behavior of the normal-state resistivity (ρ n ). Using this criterion, we define transition temperatures on c T , mid c T , and 10 c T corresponding to 90%, 50%, and 10% of ρ n , respectively. With the increase of the magnetic field from 0 to 9 T, both the onset transition point ( on c T ) and the zero-resistance point shift toward lower temperature, with on c T by 2 K, while the latter by 10 k. This is understandable since the latter is determined by the weak links between the grains and the vortex flow of magnetic flux lines, while the former is controlled by the upper critical field of the individual grains. The temperature dependence of H c2 , H irr , and H peak determined from the 90%, 50%, and 10% of ρ n criteria is shown in Fig. 7. The values of (dH c2 /dT) Tc for 10% , 50%, and 90% of ρ n , are 1.3, 2.4, 4.1 TK 1 , respectively. It needs to be mentioned that the slope (dH c2 /dT) Tc determined with different criteria is a linear function of the curve shown in Fig. 7. At low temperatures, well below T c , where the flux flow dominates resistivity, an empirical relation ρ/ρ n = H/H c has been shown. Using the conventional single-band Werthamer-Helfand-Hohenberg  is the flux quantum,  is the coherence length, and t = T/T c is the reduced temperature. From the above relation, one obtains The value of H c2 (0) obtained from the G-L equation is about 203 T for the 90% ρ n criterion. Apparently, the estimated values of the upper critical fields obtained from the WHH theory (256 T) and the G-L theory (203 T) are much higher than the BCS paramagnetic limit BCS p H = 1.84T c in the weak-coupling regime. However, the BCS model underestimates H p in the presence of strong e-ph coupling. In such a case, the actual Pauli paramagnetic limit is (1 where λ is the e-ph coupling constant. Using the value of λ~1.5 as estimated from the ρ(T) curve of a PrO 0.6 F 0.12 FeAs sample [19], the paramagnetic limit H p increases to ~240 T. Such a high value of zero-temperature upper critical field is comparable with that observed in cuprate superconductors.

Conclusion
The F-doped SmO 0.7 F 0.2 FeAs was prepared by the two-step solid-state reaction method. The optimal nominal SmO 0.7 F 0.2 FeAs is found to have the highest T c at 56.5 K (onset) in the Sm-system. The XRD results indicate a clear shrinkage of lattice parameters in the F-doping sample compared with that of the undoped one. The electrical resistivity and the DC-susceptibility (measured under a magnetic field of 1591.5 A/m) of the SmO 0.7 F 0.2 FeAs show that a clear superconducting onset transition (T c (onset)) occurred at 56.5 K and a zero resistivity transition (T c (zero)) at 53 K for the nominal SmO 0.7 F 0.2 FeAs, which is the optimal T c in this Sm-system. The transition width ΔT is about 3 K, meaning that our sample has a high quality. The typical M-H hysteresis loops at 10 K show that the rapid decrease in the wide range of the hysteresis loop is not a typical behavior of granular superconductors since the weak superconductivity in the grain boundaries is extremely sensitive to the magnetic field. The global J c was calculated on the basis of J c = 20ΔM/[a(1  a/(3b))]. At 10 K, the J c of SmO 0.7 F 0.2 FeAs is more than 2.3 × 10 5 A/cm 2 , which has a very weak dependence on the field, showing that SmO 0.7 F 0.2 FeAs has a fairly large pinning force in the grain. According to the conventional single-band Werthamer-Helfand-Hohenberg (WHH) theory, H c2 (0) is about 256 T.