No Arabic abstract
A deep understanding of the character of superconductivity in the recently discovered Fe-based oxypnictides ReFeAsO1-xFx (Re = rare-earth) necessarily requires the determination of the number of the gaps and their symmetry in k space, which are fundamental ingredients of any model for the pairing mechanism in these new superconductors. In the present paper, we show that point-contact Andreev-reflection experiments performed on LaFeAsO1-xFx (La-1111) polycrystals with Tc ~ 27 K and SmFeAsO0.8F0.2 (Sm-1111) ones with Tc ~ 53 K gave differential conductance curves exhibiting two peaks at low bias and two additional structures (peaks or shoulders) at higher bias, an experimental situation quite similar to that observed by the same technique in pure and doped MgB2. The single-band Blonder-Tinkham-Klapwijk model is totally unable to properly fit the conductance curves, while the two-gap one accounts remarkably well for the shape of the whole experimental dI/dV vs. V curves. These results give direct evidence of two nodeless gaps in the superconducting state of ReFeAsO1-xFx (Re = La, Sm): a small gap, Delta1, smaller than the BCS value (2Delta1/kBTc ~ 2.2 - 3.2) and a much larger gap Delta2 which gives a ratio 2Delta2/kBTc ~ 6.5 - 9. In Sm-1111 both gaps close at the same temperature, very similar to the bulk Tc, and follow a BCS-like behaviour, while in La-1111 the situation is more complex, the temperature dependence of the gaps showing remarkable deviations from the BCS behaviour at T close to Tc. The normal-state conductance reproducibly shows an unusual, but different, shape in La-1111 and Sm-1111 with a depression or a hump at zero bias, respectively. These structures survive up to T* ~ 140 K, close to the temperatures at which structural and magnetic transitions occur in the parent, undoped compound.
Point-contact (PC) Andreev-reflection (AR) measurements of the superconducting gap in iron-oxipnictide ReFeAsO_{1-x}F_x (Re=La, Sm) films have been carried out. The value of the gap is distributed in the range 2Delta simeq 5-10 meV (for Re=Sm) with a maximum in the distribution around 6 meV. Temperature dependence of the gap Delta(T) can be fitted well by BCS curve giving reduced gap ratio 2Delta /kT_c^*simeq 3.5 (here T_c^* is the critical temperature from the BCS fit). At the same time, an expected second larger gap feature was difficult to resolve distinctly on the AR spectra making determination reliability of the second gap detection questionable. Possible reasons for this and the origin of other features like clear-cut asymmetry in the AR spectra and current regime in PCs are discussed.
Systematic studies of the NdFeAsOF superconducting energy gap via the point-contact Andreev-reflection (PCAR) spectroscopy are presented. The PCAR conductance spectra show at low temperatures a pair of gap-like peaks at about 4 - 7 mV indicating the superconducting energy gap and in most cases also a pair of humps at around 10 mV. Fits to the s-wave two-gap model of the PCAR conductance allowed to determine two superconducting energy gaps in the system. The energy-gap features however disappear already at T* = 15 to 20 K, much below the particular Tc of the junction under study. At T* a zero-bias conductance (ZBC) peak emerges, which at higher temperatures usually overwhelms the spectrum with intensity significantly higher than the conductance signal at lower temperatures. Possible causes of this unexpected temperature effect are discussed. In some cases the conductance spectra show just a reduced conductance around the zero-bias voltage, the effect persisting well above the bulk transition temperature. This indicates a presence of the pseudogap in the system.
FeSe single crystals have been studied by soft point-contact Andreev-reflection spectroscopy. Superconducting gap features in the differential resistance dV/dI(V) of point contacts such as a characteristic Andreev-reflection double-minimum structure have been measured versus temperature and magnetic field. Analyzing dV/dI within the extended two-gap Blonder-Tinkham-Klapwijk model allows to extract both the temperature and magnetic field dependence of the superconducting gaps. The temperature dependence of both gaps is close to the standard BCS behavior. Remarkably, the magnitude of the double-minimum structure gradually vanishes in magnetic field, while the minima position only slightly shifts with field indicating a weak decrease of the superconducting gaps. Analyzing the dV/dI(V) spectra for 25 point contacts results in the averaged gap values <Delta_L> = 1.8+/-0.4meV and <Delta_S>=1.0+/-0.2 meV and reduced values 2<Delta_L>/kTc=4.2+/-0.9 and 2<Delta_S>/kTc=2.3+/-0.5 for the large (L) and small (S) gap, respectively. Additionally, the small gap contribution was found to be within tens of percent decreasing with both temperature and magnetic field. No signatures in the dV/dI spectra were observed testifying a gapless superconductivity or presence of even smaller gaps.
The newly discovered oxypnictide family of superconductors show very high critical temperatures of up to 55K. Whilst there is growing evidence that suggests a nodal order parameter, point contact Andreev reflection spectroscopy can provide crucial information such as the gap value and possibly the number of energy gaps involved. For the oxygen deficient NdFeAsO0.85 with a Tc of 45.5K, we show that there is clearly a gap value at 4.2K that is of the order of 7meV, consistent with previous studies on oxypnictides with lower Tc. Additionally, taking the spectra as a function of gold tip contact pressure reveals important changes in the spectra which may be indicative of more complex physics underlying this structure.
Using intrinsic multiple Andreev reflections effect (IMARE) spectroscopy, we studied SnS contacts in the layered oxypnictide superconductors Sm$_{1-x}$Th$_x$OFeAs with various thorium doping and critical temperatures $T_C = 21-54$ K. We observe a scaling between both superconducting gaps and $T_C$. The determined BCS-ratio for the large gap $2Delta_L/k_BT_C = 5.0-5.7$ and its eigen BCS-ratio (in a hypothetical case of zero interband coupling) $2Delta_L/k_BT_C^L = 4.1-4.6$ both exceeding the weak-coupling limit 3.52, and for the small gap $2Delta_S/k_BT_C = 1.2-1.6$ remain nearly constant within all the $T_C$ range studied. The temperature dependences $Delta_{L,S}(T)$ agree well with a two-band BCS-like Moskalenko and Suhl model. We prove intraband coupling to be stronger than interband coupling, whereas and Coulomb repulsion constants $mu^{ast}$ are finite in Sm-based oxypnictides.