Upper critical field, critical current density and thermally activated flux flow in

CaFFe0.9Co0.1As superconductor

This article has been downloaded from IOPscience. Please scroll down to see the full text article.

2012 Supercond. Sci. Technol. 25 045004

(http://iopscience.iop.org/0953-2048/25/4/045004)

Download details:

IP Address: 141.5.13.137

The article was downloaded on 24/02/2012 at 08:22

Please note that terms and conditions apply.

View the table of contents for this issue, or go to the journal homepage for more

Home Search Collections Journals About Contact us My IOPscience

IOP PUBLISHING SUPERCONDUCTOR SCIENCE AND TECHNOLOGY

Supercond. Sci. Technol. 25 (2012) 045004 (7pp) doi:10.1088/0953-2048/25/4/045004

Upper critical field, critical currentdensity and thermally activated flux flowin CaFFe0.9Co0.1As superconductor

Chandra Shekhar1,2 , Amit Srivastava2, Pramod Kumar3,Pankaj Srivastava2 and O N Srivastava2

1 Institut fur Anorganische und Analytische Chemie, Johannes Gutenberg-Universitat, D-55099 Mainz,Germany2 Centre of Advanced Studies for Physics of Materials, Department of Physics, Banaras HinduUniversity, Varanasi 221005, India3 Department of Physics and Astronomy, Georgia State University, Atlanta, GA 30303, USA

E-mail: shekhar@uni-mainz.de and cgsbond@gmail.com

Received 2 November 2011, in final form 5 January 2012Published 23 February 2012Online at stacks.iop.org/SUST/25/045004

AbstractIn this paper, we report the synthesis, structure, transition temperature, upper critical field,critical current density and thermally activated flux flow in the CaFFe0.9Co0.1Assuperconductor. Superconductivity arises at 23 K by Co substitution at the site of Fe atomsand the upper critical field is estimated as 102 T using the Werthamer–Helfand–Hohenbergformula. The flux-flow activation energy (U0/kB) varies from 3230 K and 4190 K in a field of9 T and 1 T, respectively. At 2 K, the Jc is found to be approximately 4× 103 A cm−2 and0.3× 103 A cm−2 in zero and 6 T field, respectively. Transmission electron microscopyanalysis shows an amorphous region surrounding most of the grains which is likely to bepresent in the form of amorphous and weak link grain boundaries in this compound. It seemsthat most of the current is hindered by mis-aligned grains, amorphous grain boundaries andimpurities, which are invariably found between the grains. The presence of the weakly linkedgranules and their weakly pinned intergranular Josephson vortices are responsible for both lowJc and the Arrhenius temperature dependence of resistivity.

(Some figures may appear in colour only in the online journal)

1. Introduction

The discovery of the new family of the iron arsenide,LaO1−xFxFeAs (La-1111), superconductor [1] has stimulatedconsiderable pursuit in the scientific community in thearea of condensed matter, especially high temperaturesuperconductors and strongly correlated electron systems.Extensive research efforts have been devoted to this systemdue to the relatively high transition temperature (Tc) inthe presence of iron, a layered structure and structuralsimilarity to the cuprates. From the replacement of trivalentLa by other rare-earth elements in ROFeAs (R: rare-earthelements), Tc has been increased significantly and attainedthe record of 55 K [2–6] at ambient pressure. Another

family of the FeAs-based superconductor system has beendiscovered in Ba1−xKxFe2As2 (Ba-122) with Tc at 38 K [7,8]. The parent compounds for these superconductors belongto the ZrCuSiAs-type structure (space group P4/nmm) orThCr2Si2-type structure (space group I4/mmm), consistingof an alternating stack of (RO)+δ or (Ba)+δ and (FeAs)−δ

layers. These compounds adopt the layered structure witha single FeAs layer per unit cell of ROFeAs and two suchlayers per unit cell of BaFe2As2. It is believed that FeAslayers are responsible for the superconductivity [9, 10]. Thisis quite similar to that of cuprates, in which CuO layers areresponsible for superconductivity as in the YBCO compound.Another homologous series of FeAs-containing compounds isAFFeAs (A = Ca, Sr and Eu) in which the (FeAs)−δ layer

10953-2048/12/045004+07$33.00 c© 2012 IOP Publishing Ltd Printed in the UK & the USA

Supercond. Sci. Technol. 25 (2012) 045004 C Shekhar et al

is sandwiched by the (AF)+δ layer in place of the (RO)+δ

layer in ROFeAs compounds. The AFFeAs compounds, likeother pnictides, show the spin density wave (SDW) in between120 and 180 K [11–15]. However, the superconductivity isinduced by the partial substitution of some elements of thelanthanide series [11, 16, 17] at the A site and by the partialsubstitution of Co at the Fe site [12–14, 18–21]. Following thepartial substitution of Co at the Fe site in pnictides leads to thehighest observed Tc of 14 K for LaOFe0.85Co0.15As [18], 17 Kfor SmOFe0.9Co0.1As [19], 20 K for SrFe1.8Co0.2As2 [20]and 22 K for BaFe1.8Co0.2As2 [21], so far. Furthermore,the highest Tc is 4 K for SrFFe0.87Co0.13As [12] and 22 Kfor 10% Co-substituted CaFFeAs, i.e. CaFFe0.9Co0.1As [13].Among the transition metal doping such as Co, Cr, Cu,Ir, Mn and Ni [14, 22], only Co doping has invoked theeffective emergence of superconductivity up to 22 K and italso shows the superconducting transition for a wide range ofCo concentrations from 0.05 to 0.26 [14]. In light of thesestudies, the Co substitution is a potential way to convertFeAs-based layered compounds to superconductors.

For practical applications of a superconductor, two of themost important parameters are the upper critical field, Bc2,and the critical current density, Jc. The upper critical fieldis an intrinsic property, which has been approximated to behigher than 68 or 64 T [23, 24] in LaO0.9F0.1FeAs, 70 T inPrO0.85F0.15FeAs, over 100 T in SmFeAsO0.85F0.15 [25] and230 T in high-pressure fabricated NdFeAsO0.82F0.18 [26]. Itis well known that the Jc can be controlled by the flux pinningbehaviour. However, critical current density, pinning forceand upper critical fields of the CaFFe0.9Co0.1As compoundare not reported so far. In the present paper, we report thestudy of the structures and microstructures and investigatedthe critical current density and upper critical fields of theCaFFe0.9Co0.1As compound. In this connection, we havesynthesized a series of compounds with different Co dopingconcentrations but deal here only with the 10 at.% Co-dopedsample, due to the highest Tc and larger superconductingvolume fraction. Our result shows that Jc is quite sensitiveto the temperature, and also thermally activated flux flow isresponsible for the broadening of the transition at Tc in a highmagnetic field. The upper critical fields for CaFFe0.9Co0.1Ashave been found by means of extrapolation to be 102 and 25 Tusing the Werthamer–Helfand–Hohenberg formula.

2. Experimental details

Polycrystalline samples with the nominal composition ofCaFFe1−xCoxAs (x = 0.0, 0.05, 0.07, 0.1 and 0.13) wereprepared by conventional solid state reaction using highquality CaF2, Fe, Co and CaAs as the starting materials.The Ca and As were ground in a 1:1 ratio and pressed intopellets, then sintered at 700 ◦C for 12 h in an evacuated quartztube. The final stoichiometric materials were taken accordingto 1/2CaF2 + CaAs + (1 − x)Fe + xCo = CaFFe1−xCoxAsand were heated at 1050 ◦C for 40 h containing differentconcentrations of cobalt metal in an evacuated sealed quartztube. The x-ray diffraction of samples was performed withCu Kα radiation in the 2θ range from 5◦ to 80◦, with a step

Figure 1. Observed (solid line), calculated (symbols), Braggposition (vertical line) and difference (bottom line) XRD patterns of(a) CaFFe0.9Co0.1As and (b) CaFFeAs.

interval of 0.01◦. Microstructures of the as-obtained samplesand their morphology were studied using environmentalscanning electron microscopy (SEM, Quanta 200), operatedat 30 kV. The microstructural characterizations were carriedout by high resolution transmission electron microscopy(HRTEM, FEI Tecnai 20G2, operated at 200 kV). Thetransport and magnetic properties were measured over a widerange of temperatures and magnetic fields up to 9 T usinga physical properties measurement system (PPMS, QuantumDesign). Critical current density was calculated using theBean model. Crystal structures were refined using the Rietveldrefinement.

3. Results and discussion

3.1. Crystal structure

The as-synthesized samples were subjected to the grossstructural characterization employing the x-ray diffractiontechnique. A typical x-ray diffraction (XRD) pattern ofCaFFeAs and CaFFe0.9Co0.1As is presented in figure 1. Itcan be seen that the samples are nearly single phase. Rietveldrefinement results confirm a ZrCuSiAs-type tetragonalstructure with space group P4/nmm having lattice parametersa = 3.877 A, c = 8.596 A and a = 3.884 A, c = 8.554 A forpure and 10% Co-doped samples, respectively. These valuesare very close (a little higher) to the reported standard latticeparameter values [19, 20] which indicate that Co atoms havebeen substituted into FeAs layers successfully.

2

Supercond. Sci. Technol. 25 (2012) 045004 C Shekhar et al

Figure 2. Resistivity behaviour of CaFFeAs and CaFFe0.9Co0.1Aswith temperature. The dotted line clearly shows the suppression ofSDW in the doped sample. Inset shows the enlarged view of thetransition point.

3.2. Superconducting transition temperature

The resistivity behaviour with temperature of pure and 10%Co-substituted CaFFeAs samples have been presented infigure 2. For CaFFeAs, the resistivity increases slightly withdecrease in temperature, but below ∼130 K, the resistivitydrops steeply, implying a SDW anomaly to be at 130 Kin the CaFeAsF sample. This SDW relates the structuraldistortion followed by the anti-ferromagnetic order of Fespins as revealed by neutron diffraction experiments [27].However, the 10% Co-substituted CaFeAsF sample showedTc at 22.7 K, as shown in the inset of figure 2. It is also pointedout that the maxima point of pure and the minima point ofdoped samples in the resistivity curves lie nearly at the sametemperature (see the dotted line in figure 2). This shows thatSDW is completely suppressed and superconductivity arisesdue to Co doping. The superconducting transition width,1Tc (Tc,on − Tc), is found to be nearly 2.3 K. In orderto confirm this, the superconductivity and superconductingvolume fraction, and the magnetic susceptibility (χ ) forthe CaFFe0.9Co0.1As sample were investigated at 5 and20 mT in both zero-field-cooled (ZFC) and field-cooled(FC) conditions, as shown in figure 3. The susceptibilitybecomes negative below 23 K, as shown in the inset offigure 3. The sharp diamagnetic superconducting transitionfrom the Meissner effect indicates good sample quality, anda high superconducting shielding volume fraction revealsthe bulk nature of this superconductor. Assuming thetheoretical density of roughly 6.68 g cm−3 for the perfectdiamagnetism, we estimated the shielding and Meissnerfractions, which are found to be about 90% and 2%,respectively, at 2 K. It should be noted that since theMeissner fraction is determined by pinning and penetrationeffects, its interpretation is quite ambiguous on polycrystallinesamples. The onset diamagnetic superconducting transitiontemperature is the same as that of the onset transitiontemperature on the corresponding resistivity curve. The

Figure 3. Temperature dependence of the susceptibility measuredin zero-field-cooled (ZFC) and field-cooled (FC, 50 and 2000 Oe)conditions. Inset shows enlarged view of transition point.

Figure 4. Temperature dependence of resistivity under differentmagnetic fields. The onset transition point shifts weakly with themagnetic field. The dashed lines indicate the 10% and 90% points ofthe onset resistivity.

resistivity and magnetization measurements demonstrate that10% Co-substituted CaFeAsF is a bulk superconductor at∼23 K. This is the highest achieved Tc in Co-doped 1111pnictide compounds. Higher Tc might be expected due tothe correlation between Tc and the structure of the FeAstetrahedron. Recently, Zhao et al [28] found that the highestTc of FeAs-based compounds can be obtained when theFe–As–Fe bond angle is very close to the standard valueof 109.47◦, as expected for a perfect FeAs tetrahedron.The efficient way to increase Tc in FeAs-based systemsis to optimize the Fe–As–Fe angle. It can be noticed thatthe Fe–As–Fe bond angle in CaFFeAs is 108.55◦, whichis relatively close to the standard value than for the otherpnictides [29–32]. Therefore, CaFFeAs can be considered asa promising system for maximizing Tc by substitution.

3.3. Upper critical fields

Figure 4 presents the temperature dependence resistivity ofa 10% Co-doped sample at different magnetic fields, B, up

3

Supercond. Sci. Technol. 25 (2012) 045004 C Shekhar et al

to 9 T. It can be seen that the onset Tc,on drops very slowlywith increasing magnetic field. However, the Tc shifts quicklyto lower temperatures with the increase in magnetic fieldand the transition width becomes wider with increasing B.This is understandable in terms of flux creep in granularpolycrystalline materials. The Tc shifts quickly to lowertemperatures, with the increase in magnetic field implyingthe weak links between the grains as well as the vortex flowbehaviour, while the slow drop of Tc,on with increase inmagnetic field, which is controlled by the upper critical fieldof the individual grains, predicts the relatively higher value ofthe upper critical field. The upper critical field, Bc2, is definedas the field at which the resistivity increases and approachesthe normal state resistivity. Here, we have used a criterion of90% of normal resistivity at the onset temperature and theBc2 defined in this way refers to the case of a field parallelto the ab plane, Bab

c2 . The 90% and 10% points of normal stateresistivity can be regarded as the upper critical field parallelto the ab plane, Bab

c2 , and parallel to the c plane, Bcc2 [24]. The

values of the Babc2 and Bc

c2, determined through the aforesaidmanner has now been plotted with temperature, are presentedin figure 5. The slope dBab

c2/dT for 90% ρn is estimated tobe −6.45 T K−1. This value is larger than that for La-1111(dBc2/dT = −2 T K−1) [23] and for Nd-11111 (dBc2/dT =−5.8 T K−1) [25, 26]. Similarly the slope dBc

c2/dT for 10%ρn is −1.6 T K−1 at T ≤ 13 K. The Bc2(0) can be estimatedusing a single-band Werthamer–Helfand–Hohenberg (WHH)model [33, 34] which is hardly applicable to iron-basedsuperconductors [24, 26]. This model is commonly used toevaluate the T = 0 limit of the Bc2(0). Taking the orbitaleffects only into account (where the Maki parameter α is0), the breakdown of superconductivity predicted by thefollowing formula: Bab

c2(0) = −0.69Tc(dBc2/dT)Tc, is 102 Tfor the 90% ρn fields and Bc

c2(0) is 25 T for the 10% ρn fields.The upper critical fields for CaFFe0.9Co0.1As have been foundby means of extrapolation to be 102 and 25 T using the WHHformula. Therefore, our result gives a rough estimate of Bc2(0)because of the limit of the applied magnetic field. It may benoted that high values of Bc2(0) can be achieved by: (i) strongband scattering, (ii) small Fermi velocities and (iii) strongcoupling. Strong coupling can be excluded empirically forFe–As superconductors [35, 36]. It appears that the strongband scattering is responsible for high values of Bc2(0). Leeet al [37] have studied the behaviour of Bc2 and measured it ina field up to 60 T in Sm-based pnictides, which is determinedby the complex interplay of a two-band nature and the Pauliparamagnetic effect depending on the direction of the appliedmagnetic field with respect to the crystal axes. However,Bc2 shows only Pauli paramagnetic behaviour in As-deficientLa-based pnictides [38]. Finally, we can say that the WHHapproximation could not be simply applied in this material.

3.4. Thermally activated flux flow

The broadening of the ρ(T) (just above the Tc) in amagnetic field for superconductors is interpreted in termsof a dissipation of energy caused by the motion ofvortices [39–41]. This interpretation is based on the fact that,

Figure 5. Upper critical fields, Bc2 versus T as determined from the10% and 90% points of the onset resistivity from figure 4.

for the low-resistivity region, the resistivity is caused by thecreep of vortices so that the ρ(T) dependences are thermallyactivated and are usually described by an Arrhenius equation:

ρ(B,T) = ρ0 exp(−

U0

kBT

)where U0 is the thermally activation flux-flow (TAFF) energy,which can be obtained from the slope of the linear part ofan Arrhenius plot, ρ0 is a field-independent pre-exponentialfactor and kB is Boltzmann’s constant. The best fitted ln ρversus T−1 plot to the experimental data, as shown infigure 6, yields the values of the activation energy rangingfrom U0/kB = 3230 K and 4190 K in fields of 9 T and1 T, respectively. The flux-flow activation energy generallyvaries from 3000 to 300 K with fields from 1 to 9 T inthe BiSrCaCuO system [42] and, in the case of MgB2, it isaround 10 000 K in a field ≤1 T and down to 300 K in afield of 10 T [40], Then the resistivity measurements of thesuperconducting transition of the different superconductingmaterials give insight into the flux pinning properties sothat TAFF differ for different materials. Further, the powerlaw field dependence of the activation energy U0(B) ∝ B−n

with the exponent n ≤ 1 which is usually observed for otherlayered systems [39–45]. Figure 7 presents the magnetic fielddependence of the activation energy U0 of CaFF0.9Co0.1FeAs.We can see that the values of U0 for CaFF0.9Co0.1FeAsdrop weakly with the field for B ≤ 3 T, scaled as B−0.05,and then decrease as B−0.15 for B > 3 T. In order to geta broad insight, the variation of U0 is basically due to thepinning of the flux line. However, it is found that n = 1/3for B < 3 T and n = 1/4 for B > 3 T in BaSrCaCuO [42]and n = 1/6 for B < 2 T and n = 1/3 for B > 2 T inMgB2 film [40]. A relatively slow decrease of U0(B) inlow fields implies that weakly pinned intergranular Josephsonvortices dominate (single vortex pinning) [46] followed by aquick decrease of U0(B) in the field, which could be relatedto a crossover to a collective flux creep regime [47]. Thisdissipative phenomenon is associated with the vortex motioninside the material.

4

Supercond. Sci. Technol. 25 (2012) 045004 C Shekhar et al

Figure 6. Arrhenius plot of the resistivity. The activation energy U0in a field is given by the slope from linear fitting.

Figure 7. Magnetic field dependence of the activation energy U0.

3.5. Critical current density

Magnetization hysteresis loops (M–H loop) at 2, 7, 15 and25 K are measured by superconducting quantum interferencedevice magnetometry (SQUID) as shown in figure 8.However, the M–H loop at 25 K (above Tc) is measuredto know the background. The inset shows the paramagneticbackground, due to a small amount of impurity phases,similar to earlier studies [1] at 25 K. After the subtractionof the background, the complete hysteresis loops are shownin figure 8. The hysteresis loops are a combination of twodifferent contributions, Meq or Mrev and Mirr, where Meqor Mrev and Mirr are the equilibrium or reversible andirreversible magnetizations, respectively. Here, the values ofMeq and Mirr at 2 K are 0.97 emu g−1 and 1.13 emu g−1,respectively. The sample showed a high Bab

c2 (5 K) of 102 T;only small hysteresis loops were found as compared toother polycrystalline iron oxypnictides [48, 49]. These smallhysteresis loops of the present sample reveal either weak fluxpinning and/or weak intergranular coupling. However, strongintergranular coupling implies a very high hysteresis loop inthe case of MgB2 [50]. The magnetic field dependence of the

Figure 8. Superconducting contribution to the magnetic moment inhysteresis loops as a function of field at 2, 7 and 15 K. Thesuperconducting signal was isolated from the paramagneticbackground determined above Tc, i.e. 25 K. Inset shows theparamagnetic background determined above Tc, i.e. 25 K.

Figure 9. Magnetic field dependence of the critical current densityat different temperatures.

critical current density Jc derived from the irreversible partsof the hysteresis loop width using the extended Bean’s model:

Jc =201M

Va(1− a/3b), a < b

where 1M is the width of the hysteresis loop measured inemu, V is the volume of the sample in cm3, a and b are therespective sample dimensions in cm and Jc is in A cm−2.For estimation of Jc, the full sample dimensions of 2.0 ×1.0 × 0.5 mm3 were taken. The Jc versus magnetic fieldat different temperatures extracted from the hysteresis loopwidths using the Bean model is shown in figure 9. At 2 K,the Jc is approximately 4 × 103 A cm−2 at zero field andthen decreases to 1.5× 103 A cm−2 at 0.3 T. The Jc increasesslightly with an increase in magnetic field above 0.6 T to amaximum of 1.6×103 A cm−2. This peak effect has also beenobserved in SmO1−xFxFeAs [48] and CeO1−xFxFeAs [51]compounds. It is more clearly visible in figure 9. It can be

5

Supercond. Sci. Technol. 25 (2012) 045004 C Shekhar et al

Figure 10. (a) Back-scattered SEM image showing FeAs impurity phase and voids which are indicated by arrows. (b) TEM image showinga thick amorphous region is surrounded around the grain which might be responsible for forming thick and weak grain boundaries.

observed from figure 9 that Jc at 2 K rapidly decreases upto a field of 0.2 T. This indicates that impurities grain switchoff when a small magnetic field is applied and behave likeJosephson junctions [52].

3.6. Microstructures

The Jc is one of the important parameters which areused to characterize the technologically important of thepolycrystalline superconductors. The grain boundaries inthe samples are assumed to play a crucial role indetermining the Jc. Therefore, in order to explore themicrostructural characteristics and their possible correlationwith superconducting properties, we have carried out SEMand TEM studies of the Co-doped CaFFeAs sample. Theextensive microstructural investigations reveal the followingspecial characteristics as shown in figure 10 and figure 10(a)shows back-scattered SEM of a polished surface of theCo-doped CaFFeAs sample. This sample consists of largergrains having sizes in the range of 15–40 µm, and hasbeen identified as Co-doped CaFFeAs grains. These grainsare much larger than those reported previously [52–54].However, some FeAs is also visible at some points whichare marked by arrows in figure 10(a). These FeAs may bepresent sometimes between the boundaries of the grains andthus these interrupt the grain-to-grain supercurrent paths.Further, study of the microstructure of the samples has alsobeen performed using TEM. Figure 10(b) shows a typicalgrain of the Co-doped CaFFeAs sample. Numerous randomlyselected grains have been examined in this approach and,during this course, it has been observed that there is anamorphous layer present around most of the grains in thissample. A typical example of this observation is presented infigure 10(b). A detailed observation of a single grain has beenperformed and an amorphous layer of several nanometres(a region between the arrows in figure 10(b)) in thicknessaround individual grains has been noticed. Evidence of asimilar amorphous layer was also observed in polycrystallineSr0.6K0.4Fe2As2 and YBa2Cu3O7−δ [54, 55]. Generally, thisamorphous layer forms thick and weak grain boundaries(GBs) which hinder most of the current. Finally, we havetried to summarize the correlation of transport properties

with microstructures. The transport properties are effectivelygoverned by grain boundaries. As discussed above, somerecent reports have also suggested that most of the GBsof iron-based superconductors are weakly linked [56, 57]in a similar way to GBs in the cuprate superconductors.The presence of weakly linked granules and their weaklypinned intergranular Josephson vortices are responsible forthe TAFF resistivity, which leads to both low Jc and theArrhenius temperature dependence of resistivity. For furtherimprovement in Jc, the synthesis conditions and tailoring ofthe materials need to be modified by some doping/admixingthrough which intergranular Josephson vortices can easily bepinned.

4. Conclusion

Based on the above results and discussion it can beconcluded that the transition temperature of 23 K in theCaFFe0.9Co0.1As compound is highest compared to anytransition-element-doped 1111 compound. The upper criticalfield is 102 T. At 2 K, the Jc is evaluated to be approximately4 × 103 A cm−2 and 0.3 × 103 A cm−2 in zero field and 6 Tfield, respectively. Therefore, it needs further improvement ofJc and Bc2 by improving the grains’ connectivity and creatingdisorder in this compound, respectively.

Acknowledgments

One of the authors (CS) is grateful to Professor Dr B Buchnerfor an invitation as a guest scientist at IFW-Dresden, Germany.The financial support from DST-UNANST and CSIR isgratefully acknowledged. AS is grateful to UGC—New Delhi(India) for providing a Teacher Fellowship under the FIPScheme. CS is also grateful to UGC for the award of aDr D S Kothari postdoctoral fellowship.

References

[1] Kamihara Y, Watanabe T, Hirano M and Hosono H 2008J. Am. Chem. Soc. 130 3296

[2] Chen G F, Li Z, Wu D, Li G, Hu W Z, Dong J, Zheng P,Luo J L and Wang N L 2008 Phys. Rev. Lett. 100 247002

[3] Ren Z A et al 2008 Europhys. Lett. 83 17002[4] Ren Z A et al 2008 Europhys. Lett. 82 57002

6

Supercond. Sci. Technol. 25 (2012) 045004 C Shekhar et al

[5] Chen X H, Wu T, Wu G, Liu R H, Chen H and Fang D F 2008Nature 453 761

[6] Peng C, Lei F, Huan Y, Yu Z X, Gang M, Qian L H,Sheng W Z and Hu W H 2008 Sci. China Ser. G 51 719

[7] Rotter M et al 2008 Phys. Rev. Lett. 101 107006[8] Goldman A I et al 2008 Phys. Rev. B 78 100506[9] Singh D J and Du M H 2008 Phys. Rev. Lett. 100 237003

[10] Haule K et al 2008 Phys. Rev. Lett. 100 226402[11] Zhu X, Han F, Cheng P, Mu G, Shen B, Fang L and

Wen H-H 2009 Europhys. Lett. 85 17011[12] Matsuishi S, Inoue Y, Nomura T, Hirano M and

Hosono H 2008 J. Phys. Soc. Japan 77 113709[13] Matsuishi S, Inoue Y, Nomura T, Yanagi H, Hirano M and

Hosono H 2008 J. Am. Chem. Soc. 130 14428[14] Matsuishi S, Inoue Y, Nomura T, Kamihara Y, Hirano M and

Hosono H 2009 New J. Phys. 11 025012[15] Han F et al 2008 Phys. Rev. B 78 180503[16] Wu G et al 2009 J. Phys.: Condens. Matter 21 142203[17] Cheng P, Shen B, Mu G, Zhu X, Han F, Zeng B and

Wen H-H 2009 Europhys. Lett. 84 67003[18] Sefat A S, Huq A, McGuire M A, Jin R, Sales B and

Mandrus D 2008 Phys. Rev. B 78 104505[19] Cao C-G, Wang C, Zhu Z-W, Jiang S, Luo Y-K, Chi S, Ren Z,

Tao Q, Wang Y and Xu Z 2009 Phys. Rev. B 79 054521[20] Jasper A L, Schnelle W, Geibel C and Rosner H 2008 Phys.

Rev. Lett. 101 207004[21] Sefat A S, Jin R, McGuire M A, Sales B C, Singh D J and

Mandrus D 2008 Phys. Rev. Lett. 101 117004[22] Qi Y, Wang L, Gao Z, Wang D, Zhang X, Zhang Z and

Ma Y 2009 Phys. Rev. B 80 054502[23] Narduzzo A et al 2008 Phys. Rev. B 78 012507[24] Hunte F, Jaroszynski J, Gurevich A, Larbalestier D C, Jin R,

Sefat A S, McGuire M A, Sales B C, Christen D K andMandrus D 2008 Nature 453 903

[25] Senatore C, Flukiger R, Cantoni M, Wu G, Liu R H andChen X H 2008 Phys. Rev. B 78 054514

[26] Wang X L, Ghorbani S R, Peleckis G and Dou S X 2008 Adv.Mater. 21 236

[27] Xiao Y, Su Y, Mittal R, Chatterji T, Hansen T, Kumar C M N,Matsuishi S, Hosono H and Brueckel Th 2009 Phys. Rev. B79 060504

[28] Zhao J et al 2008 Nature Mater. 7 953[29] Si Q and Abrahams E 2008 Phys. Rev. Lett. 101 076401[30] de la Cruz C et al 2008 Nature 453 899[31] McQueen T M et al 2008 Phys. Rev. B 78 024521

[32] Rotter M et al 2008 Phys. Rev. B 78 020503[33] Helfand E and Werthamer N R 1966 Phys. Rev. 147 288[34] Werthamer N R, Helfand E and Hohenberg P C 1966 Phys.

Rev. 147 295[35] Chen T Y et al 2008 Nature 453 1224[36] Drechsler S-L et al 2008 Phys. Rev. Lett. 101 257004[37] Lee H-S, Bartkowiak M, Park J-H, Lee J-Y, Kim J-Y,

Sung N-H, Cho B K, Jung C-U, Kim J S and Lee H-J 2009Phys. Rev. B 80 144512

[38] Fuchs S-L et al 2008 Phys. Rev. Lett. 101 237003[39] Jaroszynski J et al 2008 Phys. Rev. B 78 174523[40] Sidorenko A et al 2005 Phil. Mag. 85 1783[41] Wang X L et al 2005 J. Appl. Phys. 97 10B114[42] Palstra T M, Batlogg B, van Dover R B, Schneemeyer L F and

Waszchak J V 1990 Phys. Rev. B 41 6621[43] Yeshurn Y and Malozemoff A P 1988 Phys. Rev. Lett. 60 2202[44] Fogel N Y, Cherkasova V G, Koretzkaya O A and

Sidorenko A S 1997 Phys. Rev. B 55 85[45] Brunner O, Antognazza L, Triskone J M, Mieville L and

Fischer O 1991 Phys. Rev. Lett. 67 1354[46] Blatter G, Feigel’man M V, Geshkenbein V B, Larkin A I and

Vinokur V M 1994 Rev. Mod. Phys. 66 1125[47] Yeshurun Y and Malozemoff A P 1988 Phys. Rev. Lett.

60 2202[48] Wang L, Gao Z, Qi Y, Zhang X, Wang D and Ma Y 2009

Supercond. Sci. Technol. 22 015019[49] Senatore C, Flukiger R, Cantoni M, Wu G, Liu R H and

Chen X H 2008 Phys. Rev. B 78 054514[50] Larbalestier D C 2001 Nature 410 186[51] Shahbazi M, Wang X L, Shekhar C, Srivastava O N and

Dou S X 2010 Supercond. Sci. Technol. 23 105008[52] Kametani F et al 2009 Appl. Phys. Lett. 95 142502[53] Qi Y P, Zhang X P, Gao Z S, Zhang Z Y, Wang L,

Wang D L and Ma Y W 2009 Physica C 469 717[54] Wang L, Qi Y, Gao Z, Wang D, Zhang X and Ma Y 2010

Supercond. Sci. Technol. 23 025027[55] Matsumoto Y, Hombo J and Yamaguchi Y 1990 Appl. Phys.

Lett. 56 1585[56] Durrell J H, Eom C-B, Gurevich A, Hellstrom E E,

Tarantini C, Yamamoto A and Larbalestier D C 2011 Rep.Prog. Phys. 74 124511

[57] Polichetti M, Zola D, Luo J-L, Chen G-F, Li Z, Wang N-L,Noce C and Pace S 2012 Supercond. Sci. Technol.25 025010

7