Since the discovery of high-transition-temperature (high-Tc) superconductivity in layered copper oxides, extensive effort has been devoted to exploring the origins of this phenomenon. A Tc higher than 40 K (about the theoretical maximum predicted from Bardeen–Cooper–Schrieffer theory1), however, has been obtained only in the copper oxide superconductors. The highest reported value for non-copper-oxide bulk superconductivity is Tc = 39 K in MgB2 (ref. 2). The layered rare-earth metal oxypnictides LnOFeAs (where Ln is La–Nd, Sm and Gd) are now attracting attention following the discovery of superconductivity at 26 K in the iron-based LaO1-xFxFeAs (ref. 3). Here we report the discovery of bulk superconductivity in the related compound SmFeAsO1-xFx, which has a ZrCuSiAs-type structure. Resistivity and magnetization measurements reveal a transition temperature as high as 43 K. This provides a new material base for studying the origin of high-temperature superconductivity.
LnO1-xFxFeAs adopts ZrCuSiAs-type structure. A series of equiatomic quaternary compounds LnFeAsO and LnFePO (Ln = La–Nd, Sm, Gd) with ZrCuSiAs-type structure has been reported4,5. The crystal structure of the tetragonal ZrCuSiAs-type compound SmFeAs(O,F) is shown in Fig. 1.
Polycrystalline samples with nominal composition SmFeAsO1-xFx (x = 0.15) were synthesized by conventional solid state reaction using high-purity SmAs, SmF3, Fe and Fe2O3 as starting materials. SmAs was obtained by reacting Sm chips and As pieces at 600 °C for 3 h and then 900 °C for 5 h. The raw materials were thoroughly ground and pressed into pellets. The pellets were wrapped in Ta foil, sealed in an evacuated quartz tube, and finally annealed at either 1,160 °C or 1,200 °C for 40 h.
Figure 2 shows the X-ray diffraction (XRD) pattern for a sample annealed at 1,160 °C. It is found that the peaks in the XRD pattern can be well indexed to the tetragonal ZrCuSiAs-type structure with a = 0.3932 nm and c = 0.8490 nm, except for some tiny peaks from the impurity phase SmOF. These lattice parameters are slightly smaller than the values of a = 0.3940 nm and c = 0.8501 nm for F-free SmFeAsO.
Magnetic characterization of the superconducting transitions under a magnetic field of 10 Oe for a sample annealed at 1,160 °C is shown in Fig. 3; data are shown for the zero-field cooled and field-cooled measurements. The magnetic onset for the superconducting transition is 41.8 K for the sample annealed at 1,160 °C, and 41.3 K for the sample annealed at 1,200 °C (not shown). The existence of the superconducting phase was unambiguously confirmed by the Meissner effect on cooling in a magnetic field. A superconducting volume fraction of about 50% under a magnetic field of 10 Oe was obtained at 5 K, indicating that the superconductivity is bulk in nature.
Figure 4 shows the temperature dependence of the resistivity under zero magnetic field and under fields of H = 5 and 7 T; Fig. 4a and b show data for samples annealed at 1,160 °C and 1,200 °C, respectively. Under zero magnetic field, the onset transition and midpoint temperatures of the resistive transition are respectively 43 K and 41.7 K for a sample annealed at 1,160 °C; for a sample annealed at 1,200 °C, the values are respectively 43.7 K and 41.2 K. The 90–10% transition width are 2.5 K and 3 K for the samples annealed at 1,160 °C and 1,200 °C, respectively. It is found that the onset transition temperature in susceptibility coincides with the transition midpoint temperature in resistivity. An external magnetic field of 5 or 7 T makes the transition width broader, but the onset transition temperature is not sensitive to magnetic field, indicating that the upper critical field is very high for this superconductor. Therefore, this superconductor has potential applications due to its high transition temperature and high upper critical field.
Replacement of La by Sm leads to a large increase in Tc from 26 K in LaO1-xFxFeAs (ref. 3) to 43 K in SmFeAsO1-xFx (this work). This suggests that it is possible to realize higher Tc values in such layered oxypnictides. The observed Tc of 43 K in SmFeAsO1-xFx is higher than the theoretical value predicted from Bardeen–Cooper–Schrieffer (BCS) theory1, and this provides a strong argument for considering layered oxypnictide superconductors as unconventional superconductors.
This work was supported by the Natural Science Foundation of China and by the Ministry of Science and Technology of China.
Author Contributions X.H.C. designed and coordinated the whole experiment, including the details of doping and synthesis, did some of the experiments, analysed the data and wrote the paper. T.W., G.W. and R.H.L. contributed equally to the synthesis, magnetic measurements and resistive measurements under magnetic field, H.C. did the structure analysis, and D.F.F. did some of the resistive measurements.