Lessons from Oxypnictide Thin Films
Silvia Haindl, Martin Kidszun, Franziska Onken, Alexander Mietke, Thomas Thersleff
NNovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb
International Journal of Modern Physics Bc (cid:13)
World Scientific Publishing Company
Lessons from Oxypnictide Thin Films
Silvia Haindl
IFW Dresden, Institute for Solid State Physics, Helmholtzstr. 2001069 Dresden, Germany ∗† Martin Kidszun
IFW Dresden, Institute for Solid State Physics, Helmholtzstr. 20D-01069 Dresden, Germany † Franziska Onken
TU Dresden, Faculty of Science, Department of PhysicsD-01062 Dresden, Germany
Alexander Mietke
TU Dresden, Faculty of Science, Department of PhysicsD-01062 Dresden, Germany
Thomas Thersleff
Uppsala University, Department of Engineering Sciences, Division of Applied Materials ScienceBox 534 SE-751 21 Uppsala, Sweden † pre-print versionappears in Int. J. Mod. Phys. B 27 (2013) 1230001First experiments on the growth of oxypnictide F-doped LaFeAsO thin films indicatedan incomplete normal-to-superconducting transition and offered a work programme chal-lenging to overcome possible difficulties in their fabrication. In this regard the possibilityof an all in-situ epitaxial growth appeared to be a matter of time and growth param-eters. The following review clarifies that F-doped oxypnictide thin films are extremelydifficult to grow by in-situ PLD due to the formation of very stable impurity phasessuch as oxyfluorides (LaOF) and oxides (La O ) and the loss of stoichiometry possiblydue to incongruent evaporation of the target or re-evaporation of volatile elements atthe substrate surface. However, the review also demonstrates that the employed two-step fabrication process for oxypnictide thin films has been successfully applied in thepreparation of clean polycrystalline as well as of epitaxial thin films. Fundamental in-vestigations on the upper critical field, its temperature dependence and its anisotropycontributed to an understanding of multiband superconductivity in oxypnictides. Keywords : Fe-based superconductors; Oxypnictides; Thin Films. ∗ [email protected] † previous address: IFW Dresden, Institute of Metallic Materials, Helmholtzstr. 20 D-01069 Dres-den, Germany 1 a r X i v : . [ c ond - m a t . s up r- c on ] D ec ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb S. Haindl, M. Kidszun, F. Onken, A. Mietke, T. Thersleff
1. Introduction
Fe-based superconductors are a new class of high temperature superconductors con-taining Fe-pnictide and Fe-chalcogenide compounds with the common structuralfeature of tetrahedrally coordinated FeAs (or FeSe) layers that are supposed to beresponsible for superconductivity. In the short history of Fe-based superconductors,a superconducting transition was first observed in the oxypnictides. , At a firstglance, the fact that an Fe containing compound shows superconducting propertiesis against intuition. This may also be one of the reasons why Jeitschko and co-workers, who synthesized a plethora of Fe-arsenides and Fe-phosphides already inthe 1990s, were primarily interested in their magnetic properties. , Today, a hugenumber of different Fe-based superconductors are known and they are groupedinto families of binary Fe-chalcogenides with composition , ternary Fe-pnictidesand chalcogenides with composition or , and quaternary Fe-oxypnictideswith composition . Also more complex oxypnictides show superconductingproperties, however, this article will focus on the quaternary F-doped LnFeAsOoxypnictides (Ln = La, Sm). Several initial findings for the Fe-based superconductors have reached consen-sus: (1) The electronic band structure reveals several bands that cross the Fermilevel resulting in a complex Fermi surface with different sheets, and it became clearthat more than one electronic band is responsible for superconductivity. (2) Thehigh critical temperatures up to 55 K in the oxypnictides cannot be explained bya phonon-mediated electron-electron interaction. (3) A spin-density wave (SDW)state appears in most of the compounds and competes with the superconductingstate suggesting that spin fluctuations are responsible for the superconducting pair-ing mechanism. Therefore, Fe-based superconductors can be classified as uncon-ventional high-temperature superconductors of strong multiband nature.With a few exceptions, superconductivity appears upon charge carrier (elec-tron or hole) doping, and the undoped compound is characterized as metallic orsemi-metallic with a SDW developed. Superconductivity in the quaternary oxyp-ncitides LnOFeAs (Ln = lanthanide) can be induced, for example, by F-doping , byoxygen deficiency , by isovalent P-doping , by Co-doping , but also by externalpressure . Today, the highest critical temperature in the oxypnictides is found inF-doped SmFeAsO that shows a T c of 55 K. These high critical temperatures in theoxypnictides attracted enormous attention towards possible applications in whichthin films do play a leading role.Despite the complexity of these compounds, first attempts of oxypnictide thinfilm fabrication started in 2008 using pulsed laser deposition (PLD). , Immedi-ately, the question about the F-content and, therefore, about the doping level turnedout to be the first major experimental challenge. The first available LaFeAsO − x F x thin films neither showed any signs of superconductivity nor a T c above 20K andfull resistive transitions. Until today, an all in-situ deposition on heated substratesfailed in growing high quality superconducting oxypnictide thin films. At presentovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb
Lessons from Oxypnictide Thin Films there is increasing activity in the oxypnictide thin film growth using molecularbeam epitaxy (MBE). , , , Using MBE NdFeAsO − x F x and SmFeAsO − x F x thin films of good quality were obtained in the last years. There, F-doping wassuccessfully achieved via diffusion from a cap layer either of NdOF, NdF or SmF respectively. We will not further discuss the peculiarities of MBE grown films hereand want to refer to a summary in a recent article by Ueda et al. .This review is organized as follows: The first section is devoted to details ofoxypnictide thin film preparation using PLD in combination with a post depositionannealing heat treatment (two-step process) resulting in polycrystalline and epi-taxially grown LaFeAsO − x F x thin films. After a predominantly technical section , sections to review specific scientific problems that were investigated usingLaFeAsO − x F x thin films. In section we will focus on the upper critical field ofLaFeAsO − x F x thin films and its temperature dependence, whereas the anisotropyof the upper critical field is subject of section where the anisotropic Ginzburg-Landau scaling is used for its determination down to low temperatures. Section discusses potential applications for oxypnictide thin films. The review closes with asummary.
2. Pulsed Laser Deposition of Oxypnictide Thin Films2.1.
Principles of Pulsed Laser Deposition
Pulsed laser deposition (PLD) is a physical vapour deposition technique thatuses a high power pulsed laser beam in order to vaporize material from a targetsurface. , , Several important processes appear in PLD that can be describedseparately: (1) The interaction of the laser with the target, that is in principle avery complex process involving laser light absorption, target heating and plasmaformation. (2) The vaporized material contains ions, electrons, and clusters - alsocalled plasma plume, and it expands perpendicular to the target surface with veloc-ities typically in the range of 10 ms − . Processes of plasma expansion under highvacuum or background gas atmosphere are, therefore, considered separately due totheir longer duration compared to the duration of plasma formation. (3) Film nu-cleation and film growth on the substrate are considered as the final processes inPLD.PLD as a thin film growth method owes its rise to the successful preparationof YBa Cu O − δ (YBCO) thin films. The main advantages of the PLD processare the following: (1) High deposition and growth rates. Using PLD thin films withthicknesses in the range of nanometers up to several hundreds of nanometers can begrown in a relatively short time of 1-100 minutes. (2) A stoichiometric transfer of thetarget material to the substrate that allows the growth of complex compounds fromstoichiometric targets. (3) A highly directional material transfer from the target tothe substrate. In vacuum atmosphere the plasma plume is strongly forward directed,i.e. normal to the target surface, even if the laser incidence is not normal. (4) Themethod is applicable in the growth of many different compounds such as nitrides,ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb S. Haindl, M. Kidszun, F. Onken, A. Mietke, T. Thersleff
Fig. 1. (a)
Schematic setup for room temperature PLD consisting of a high power laser, thepulsed laser beam path and the deposition chamber. (b) (previously published as figure 2 inKidszun et al. ) Evacuated quartz tube containing two as-grown LaFeAsO − x F x films and asintered LaFeAsO . F . pellet used during the ex-situ heat treatment in the two-step fabricationprocess. complex oxides including high-temperature superconductors, metals, polymer-metalcompounds, and even fullerenes or other carbon-based materials. In addition, thepulsed process enables layer-by-layer growth and thus the synthesis of multilayerand quasi-multilayer films. The minimal PLD setup consists of a high power pulsed laser (KrF exciplex orNd:YAG lasers with pulse durations between 5 and 50 ns), optics (lense, mirrors,window transparent to laser radiation) within the laser path and a deposition systemthat consists of a vacuum chamber with the possibility for the manipulation ofsubtrate and target holders (Figure 1(a)). For a deposition at elevated temperaturessubstrate heating can be achieved by ceramic heaters directly placed above thesubstrate holder or by laser heating using diode or CO lasers. During deposition aconstant rotation of substrate and target holders is recommended in order to achievebetter film homogeneity with respect to composition and thickness.ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb Lessons from Oxypnictide Thin Films All In-Situ Depostion Not Yet Realized
Up to now it has proved difficult to grow superconducting (Ln)FeAsO − x F x (Ln =lanthanide) thin films by an all in-situ PLD process. Therefore, all relevant param-eters such as substrate temperature during deposition, repetition rate, laser fluence(or energy density at the target surface), target-substrate distance, vacuum con-ditions or the use of background gas for obtaining phase formation, texture, andoptimal superconducting properties have not been investigated systematically sofar.Besides the control of the F-content in the grown films, also the control of the As-content represents a major challenge. Stoichiometric material transfer is regardedas one of the advantages of the PLD process. However, the stoichiometric transfer islimited by a preferential ablation from the target, thermal evaporation at low energydue to different vapour pressures or segregation at the target surface as listed byArnold and Aziz. Volatile elements like As are extremely sensitive to evaporationwhich can lead to further complications regarding the thin film stoichiometry. Off-stoichiometry then can be caused either by an already off-stoichiometric materialtransfer or by re-evaporation of the volatile elements from the substrate. In this case,the used target needs to be exposed to the laser irradiation a certain time until asteady-state in the ablation process is achieved or target compositions with excessof the volatile element will be necessary in order to compete with losses. Therefore,As-enriched targets have been employed by several groups for the growth of differentFe-based superconductors. , , Two-Step Process
The successful fabrication process for oxypnictide thin films involves a two-stepprocess: (1) room temperature PLD of a polycrystalline stoichiometric target fol-lowed by (2) an ex-situ post deposition heat treatment in an evacuated quartz tubeat temperatures of 940 ◦ C to 960 ◦ C (Figure 1(b)). , Such a two-step processincluding ex-situ or in-situ annealing is generally known, for example, from earlyfabrication of MgB thin films, where the volatility of Mg , , posed problems,and also from the growth of Tl- or Hg-based cuprate thin films. , The two-stepprocess has also been successfully applied in the growth of K-doped BaFe As thinfilms. For the oxypnictide thin film growth the evacuated quartz tube contained addi-tionally a pressed pellet of F-doped LaFeAsO (or SmFeAsO) of the same stoichiom-etry during the heat treatment. This sintered pellet turned out to be vital in orderto keep the F-level and probably also the As-level within the thin film sample.Until today, an all in-situ deposition on heated substrates failed in growing highquality superconducting oxypnictide thin films. However, a comparison with thefew available oxypnictide thin film growth attempts using PLD demonstrated thesuperior success of employing the two-step process as first described by Backen etal. (Table 1). An all in-situ deposition at elevated temperatures by Hiramatsu et ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb S. Haindl, M. Kidszun, F. Onken, A. Mietke, T. Thersleff al. resulted in epitaxial but non-superconducting LaFeAsO thin films and thusemphasized the problem of F-losses. A two-step process including a post depositionheat treatment at temperatures between 900 ◦ C and 1100 ◦ C but without the addi-tional use of pellets has not been successful in the growth of oxypnictide thin films,and therefore, this approach was abandoned by the same authors. Film Deposition at Room Temperature
The room-temperature deposition uses a standard PLD setup equipped with a KrFexciplex laser (with a wavelength of 248 nm and a typical energy density of 4 Jcm − at the target surface). The setup is in principal equal to the one shown in Figure 1(a),except for the motor for substrate manipulation that was demounted. The ablationprocess takes place in a high vacuum chamber (base pressure of 10 − mbar) witha target-substrate distance of 4 cm. LaFeAsO . F . and LaFeAsO . F . targetswere synthesized based on the solid state reaction La O + 2 LaO . + 4 FeAs → . The idea ofusing an overdoped target was to compete with F-losses during the PLD process aspreliminary results indicated.Different substrates such as LaAlO , (La,Sr)(Al,Ta)O , and SrTiO were usedfor oxypnictide thin film growth experiments, however, successful film growth wasonly achieved on LaAlO (001) substrates with a lattice parameter of a = 3.82 ˚A.The lattice mismatch between the substrate lattice constant and the in-plane latticeconstant of LaFeAsO . F . ( a = 4.028 ˚A) and of LaFeAsO . F . (a = 4.018 ˚A)is about 5% and a possible source for the interface layer formation of LaOF asdescribed below in more detail. The lattice constants given here were determinedfrom powder X-ray diffraction of the powders of the pellets that were used as targetmaterials and in the ex-situ heat treatment.The Film thickness is primarily controlled by the number of pulses and the laserfluence (energy density at the target surface). Typically, film thicknesses variedbetween 100 and 1000 nm achieved by room temperature deposition.2.3.2. Ex-situ Heat Treatment
Using the above described two-step fabrication process the main process parametersconcern the heat treatment rather than deposition conditions. These parameters areannealing time, maximum temperatures and temperature profile including heatingand cooling ramps, pressure and control of the atmosphere in the sealed quartztube, thermal stability of the substrate, etc. Thus, in several fabrication series theinfluence of these parameters was investigated and stepwise improved by primarilyadjusting the temperature level and heating ramps as well as the duration of theheat treatment. Additional experiments were carried out in order to improve thepressure in the quartz tube and to investigate the role of the added LaFeAsO . F . pellet. Good growth conditions were found for temperatures around 950 ◦ C and anovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb
Lessons from Oxypnictide Thin Films reference Hiramatsu et al. Backen et al. laser source, wavelength Nd:YAG (2 ω ), 532 nm KrF, 248 nmrepetition rate 10 Hz 10 Hzenergy density 1.5 Jcm − − target composition LaFeAsO . F . LaFeAsO . F . substrates substrate temperature 700 - 880 ◦ C room temperaturematerial, lattice parameter MgO(100), 4.21 ˚A; LaAlO (100), 3.82 ˚A(La,Sr)(Al,Ta)O , 3.87 ˚AMgAl O , 8.09 ˚Aall in-situ results heteroepitaxial growth no phase formationbut not superconducting post heat treatment1) ex-situ temperature 900 - 1100 ◦ C 1060 ◦ C for 4h,optimized to 950 ◦ C-960 ◦ Cfor 5-7h , , use of a LaFeAsO − x F x pellet (cid:55) (cid:88) results not superconducting polycrystalline, T c = 11 Kpolycrystalline, T c = 28 K epitaxial growth, T c = 25 K
2) in-situ atmosphere, temperature in 5 % Ar/H at 400 ◦ C for 0.5h (cid:55) not superconducting annealing time of 5-7 hours. However, individual experiments indicated that theoptimal parameters are sensitive to the film thickness, especially when epitaxialgrowth of LaFeAsO − x F x should be achieved.An important point in the successful application of the two-step process was theuse of an additional piece of a sintered pellet of same stoichiometry (Figure 1(b)).The pellet prevents F and probably also As losses during the heat treatment ofthe as-grown films in the quartz tube. In addition, the reaction atmosphere insidethe quartz tube may facilitate the phase formation process. High critical tempera-tures in the range between 25 K and 28 K were obtained when the heat treatmenttemperature was reduced from originally 1060 ◦ C to around 950 ◦ C. , , Further-more, the two-step process resulted finally in epitaxially grown LaFeAsO − x F x andSmFeAsO − x F x thin films. , , In processes depending on a heat treatment the suitability of the substrate isnot only given by its lattice misfit, but also determined by its chemical stability.Until today, detailed chemical reactions during the heat treatment are not resolved,but it is known that F reacts with moisture and forms hydrogen fluoride, HF, thatetches the quartz tube used in the ex-situ heat treatment and/or the substrates.It is therefore regarded as necessary to avoid moisture and to evacuate the quartzovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb S. Haindl, M. Kidszun, F. Onken, A. Mietke, T. Thersleff tube before sealing.
Impurities, Film/Substrate Interface, and Doping
In all attempts of LaFeAsO − x F x thin film growth various impurity phases occured.Whereas LaAs formation was found in in-situ grown films by Hiramatsu et al. ,the in-situ process in Backen et al. resulted mainly in La O and LaOF impurityphases that seem to prevent any LaFeAsO − x F x phase formation. The two-step fabrication process is also not free of impurity phase formation.There, the parameters of the ex-situ heat treatment have a strong influence onthe phase formation. Typical impurity phases like LaOF and La O were identifiedin X-ray analysis carried out on the grown LaFeAsO − x F x films. Predominantly,LaOF competes with LaFeAsO − x F x phase formation as similarly seen in many at-tempts of synthesis and single crystal growth. , Analytical transmission electronmicroscopy (TEM), especially energy filtered TEM (EFTEM), resolves the main im-purity, LaOF, at the film/substrate interface and on top of the sample (Figure 2).These observations were made for the polycrystalline film (Figures 2(a,b)) as wellas for the epitaxially grown thin film (Figure 2(c)). In the latter case also the LaOFlayer at the film/substrate interface is grown epitaxially and, therefore, facilitatesthe epitaxial growth of LaFeAsO − x F x . It can be assumed that in this case theLaOF layer helps to adjust the lattice misfit between the LaAlO substrate andLaFeAsO − x F x .The intermediate part of the thin films is dense and free of detectable impuritiesand thus allows performing electrical transport measurements with a defined currentpath. This microstructure is found to be typical for the two-step process used forfilm fabrication: the accumulated impurity concentration on top of the film canbe explained by the growth process in the direction from the substrate to the topduring the ex-situ heat treatment.A determination of the F-doping level within the grown films is difficult. For anindirect estimation of the F-content the superconducting transition temperature canbe considered. T c values between 25 K and 28 K can be taken as an indicator for a F-content of approximately 10 at.%. A direct measurement of the F-content in the thinfilms commonly involves destructive chemical analysis methods. A determinationof the F-content by electron-energy-loss spectroscopy (EELS) on TEM specimensresulted in x = 0.16, but this value seems to be overestimated because of overlappingsignals from both, LaFeAsO − x F x and LaOF, phases.
3. Upper Critical Fields and Multiband Superconductivity3.1.
Multiband Superconductivity
The fact that more than one electronic band can contribute to superconductivitywas originally and independently recognized by Suhl et al. and Moskalenko in1959 shortly after the development of the well known microscopic theory by Bardeen,ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb Lessons from Oxypnictide Thin Films (a)
Overview bright field TEM image of a polycrystalline LaFeAsO − x F x thin film. Thewhite framed region is displayed in the figure below. (b) (c) (previously published as figure 5 in the supplementof Kidszun et al. ) 3-window EFTEM (of core loss edges of F(K), La(M4,5), and Fe(L2,3)resulting in an RGB image) of the epitaxially grown LaFeAsO − x F x thin film reveals LaOFimpurity phases at the film/substrate interface and on the film top. The lamellae were preparedby Focused Ion Beam (FIB) in-situ lift-out technique and thinned to electron transparency usinga 5 kV Ga + ion beam. TEM investigations were performed in a C S -corrected FEI Titan 80-300TEM operating at 300 kV at IFW Dresden. Cooper and Schrieffer (BCS theory). Due to the overlap of different electronic bandsat the Fermi level manifesting itself in different sheets of the Fermi surface, multiplegaps in the electronic excitation spectrum appear. A famous example of a two-band superconductor is MgB with superconducting gaps on two separated Fermisheets. , Multiband superconductivity in the oxypnictides was considered theoreticallysoon after their discovery. Band structure calculations predict the Fermi sur-face consisting of four to five sheets, two electron cylinders around the M-pointof the Brillouin zone, two hole cylinders and a hole pocket around the Γ-pointarising from Fe 3d orbitals. , , Indeed, the majority of the experiments foundat least two gaps different in size (∆ small ≤ large ≈ , Point-contact spectroscopy was also carried out recently on LaFeAsO − x F x andLa − x Sm x FeAsO − x F x thin films grown by the two-step method, but only one cleargap around 6 meV was identified, a second gap feature remained unclear. Apartfrom spectroscopic probes, a manifestation of the multiband nature also appears inthe temperature dependence of the upper critical field H c ( T ), which can stronglydeviate from the single band Werthamer-Helfand-Hohenberg (WHH) theory. , , ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb S. Haindl, M. Kidszun, F. Onken, A. Mietke, T. Thersleff
It is therefore not surprising that one of the first claims of experimental evidencefor multiband superconductivity in the oxypnictides was raised on the temperaturedependence of H c in LaFeAsO − x F x . A decade earlier, Gurevich developed a handy theoretical description of H c ( T )applicable to the two-band superconductor MgB . His model is based on approxi-mations for weak coupling (BCS limit) and information about the 2 × λ , λ denoting the intraband and λ , λ denoting theinterband coupling constants. , , For phonon mediated superconductors λ ij areprecisely the electron-phonon coupling constants. In the Fe-based superconductorsthe interpretation may be generalized to electron-boson coupling constants becauseof the previously mentioned high critical temperatures and the possible magneticinteraction channel.The temperature dependence of H c for an epitaxially grown LaFeAsO − x F x thin film was measured in pulsed magnetic fields up to 42 T for both directions, H (cid:107) c and H ⊥ c (as indicated by the dots in Figure 3). The sudden upwardcurvature for H (cid:107) cc ( T ) at low temperatures ( T /T c = t < T c and an increased H c due to impu-rity scattering. This so-called bilayer toy model of shunted electronic bands (firstapplied to MgB ) was qualitatively applied to the effective electron- and hole-band of the oxypnictide superconductor (lines in Figure 3(a)). A similar trend forthe temperature dependence of estimated H (cid:107) cc ( T ) was found by Hunte et al. for t < . F . sample. Application of Gurevich’s Two-Band Model to H c ( T ) For the application of a simple two-band model, the following notation should beshortened by introducing w = det Λ, s = sgn(w), λ ± = λ ± λ and λ = ( λ − +4 λ λ ) / . It can be deduced that a positive signum of w accounts for a strongerintraband coupling, whereas a negative signum of w is given for a stronger inter-band coupling. Neglecting interband scattering by nonmagnetic impurities H c ( t )is implicitely given by ln ( t ) = − (cid:20) U ( h ) + U ( h ) + λ w (cid:21) + s (cid:115) (cid:18) U ( h ) − U ( h ) − λ − w (cid:19) + λ λ w (1)with U , ( h ) = (cid:60) ( ψ [1 / i + D , /D ) h ]) − ψ (1 / H c = 2Φ k B T c th/D (cid:126) , (2)ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb Lessons from Oxypnictide Thin Films (a)
Measured temperature dependence of the upper critical field for an epitaxially grownLaFeAsO − x F x thin film (with T c = 25 K). Both major directions H c (cid:107) c and H c ⊥ c are shown.Qualitatively, a simple toy model can explain the measured temperature dependence of the uppercritical field. A schematic description of two shunted electronic bands is given in the inset. (b) The same measured upper critical field values as in (a). Solid curves are specific solutions fromGurevich’s two-band model. A HRTEM image resolves the layered structure of the LaFeAsO − x F x phase in the epitaxially grown film (inset, previously publisehd as inset in figure 1(b) in Kidszun et al. ). Magnetotransport measurements in pulsed magnetic fields up to 42 T were carried outat IFW Dresden by A. Kauffmann and N. Kozlova in collaboration with J. Freudenberger. (Datapreviously published as figure 12 in the supplement of Kidszun et al. ) where h is the critical field parameter, t = T /T c is the reduced temperature, D = (cid:126) / m , and ψ ( x ) is the digamma function (with (cid:60) denoting the real part). Thecomplex argument in U , ( h ) accounts for paramagnetic effects. Neglecting themgives U , ( h ) = ψ (1 / D , /D ) h ) - ψ (1 / , , The diffusivities of the two bands D and D play an important role becauseaccording to their ratio, η = D /D , three cases can be distinguished: η < η = 1 (resembling the one-band solution), and η >
1. For η < H c results from the case η > λ ij , andthe band diffusivities, D i , (i, j indicating the band indices) is necessary to calculate H c ( t ) with the help of equations (1) and (2). Here, impurity scattering is consideredonly by the band diffusivities and their ratio. More parameters or estimated valuesfor an additional 2 × code which findsall possible solutions based on equation (1) within a certain error range by varyingthe coupling constants. An initial step width of 0.01 for the coupling constants, λ ij , in the range between 0.1 and 1 was used. To decrease the computing time theovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb S. Haindl, M. Kidszun, F. Onken, A. Mietke, T. Thersleff calculations were adjusted to minimize the deviation from the experimental datapoints. Assumptions on the diffusivities were made previous to the calculations for H (cid:107) cc . The diffusivity ratio between the two bands was set initially to η = 0.01 inorder to adjust the sharp upward curvature of H (cid:107) cc (T) at low temperatures.As demonstrated by the lines in Figure 3(b) a simulation of the measured H c ( T )data points is possible using Gurevich’s model, however, the inversion of the prob-lem, i.e. a quantification of the coupling constants determined from H c ( T ) is im-possible. There is no unique solution but rather a whole set of solutions. This setof solutions can however be characterized by λ · λ > λ · λ , i.e. the intrabandcoupling dominates over the interband coupling in an effective two-band model.Several points are noteworthy: (1) A drawback of the described procedure isthe high number of parameters. Already in the two band scenario there are fourcoupling constants, λ ij , and two diffusivity parameters, D i , ( i, j = 1,2) with theimpurity scattering matrix already neglected. (2) A full description would of courseinvolve a 4 × × × H c ( T ) of polycrystallineLaFeAsO − x F x where all coupling constants were set to λ ij = 0.5 ( ∀ i, j ) in orderto reduce the number of unknown parameters. There, an upward curvature in H c ( T ) was observed for low F-doping levels ( x = 0.05) and high F-doping levels( x = 0.14), modelled with small diffusivity ratios η = 0.1 (for x = 0.05) and η =0.01 (for x = 0.14). (4) There is also no information about the pairing symmetry,especially s ++ and s ± cannot be distinguished primarily from H c ( T ) since onlythe product λ · λ enters the equations and, therefore, cancels any sign of λ ( λ respectively). Additional assumptions such as preferential interband couplingdue to spin fluctuations for s ± symmetry of the superconducting order parameterare neglected here, but have been applied in more complete descriptions involvingmore than two bands. Similar calculations using Gurevich’s model were done todescribe the temperature dependence of the upper critical field measured for singlecrystals of NdFeAsO . F . . Different coupling constant matrices (motivated bydifferent pairing scenarios) were discussed there. A clear prediction however failed,because of the lack of H c data in the low temperature and high field regions, wherethe scenarios could be distinguished.
4. Critical Current Scaling and H c Anisotropy4.1.
Anisotropic Ginzburg Landau Scaling Theory
Uniaxial electronic systems (superconductors with a layered structure) can be de-scribed by the effective electronic mass anisotropy, γ m = ( m (cid:107) c /m ⊥ c ) / , where m (cid:107) c denotes the effective electron mass perpendicular to the layered structureand m ⊥ c denotes the effective electron mass parallel to the layers. It is wellknown that in the frame of the anisotropic Ginzburg-Landau theory the equationovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb Lessons from Oxypnictide Thin Films γ m = H ⊥ cc /H (cid:107) cc = H (cid:107) cc /H ⊥ cc ( γ Hc = γ Hc ) holds for an anisotropic single bandsuperconductor. The angular dependence ( θ denoting the angle between the ap-plied magnetic field H and the c -axis of the uniaxial superconductor) of the uppercritical field can thus be written as H c ( θ ) = H (cid:107) cc ( cos ( θ ) + γ − m sin ( θ )) / . (3)In the original work of Blatter, Geshkenbein, and Larkin it was shown that ascaling approach exists which maps a physical quantity, Q , obtained for an isotropicsuperconductor to the solution of an anisotropic superconductor by following ascaling rule that can be written as Q ( θ , H , T , ξ , λ , γ , δ ) = s Q · Q (cid:63) ( (cid:15) θ H , γT , ξ , λ , γδ ).The following quantities are used: θ is defined above, T is the temperature(strength of thermal fluctuations), ξ and λ are the coherence length and the pen-etration depth (values for the ab -plane), δ is the (scalar) disorder strength and (cid:15) θ = ( cos ( θ ) + γ − m sin ( θ )) / with γ m the effective electronic mass anisotropy. Forapplied magnetic fields the scaling factor s Q is equal to 1/ (cid:15) θ , which leads to theabove result for Q ≡ H c ( θ ) (with Q (cid:63) = H (cid:107) cc ) in 3.Upon the assumption of uncorrelated disorder (also called random pinning ), thescaling rule for the in-plane critical current density is given by Q ≡ J ipc ( θ, H ) = J H (cid:107) cc ( (cid:15) θ H ) with J H (cid:107) cc = J (cid:63)c ( ≡ Q (cid:63) ) being the critical current density for the (equiv-alent) isotropic superconductor. , This relation predicts the angular dependenceof the in-plane critical current density. Blatter and Geshkenbein write : Hencechanging the direction of the magnetic field leads to a dependence of the in-planecritical current density J ipc ( θ, H ) on the angle θ through the combination (cid:15) θ H , re-sulting in sharp maxima of J ipc ( θ, H ) when the field is aligned with the supercon-ducting planes and a sharpening of these maxima with increasing field amplitudeH. This scaling rule has been used in the discussion of the angular dependence ofthe in-plane critical current densities by several authors, and it is valid in the weakcollective pinning regime. , , The scaling method has also been used to extractuncorrelated pinning effects from correlated ones. , , The presence of correlated defects (and thus correlated pinning ) might imposelimitations on the application of the scaling procedure: The mass anisotropy forYBa Cu O − δ is known to be in the range of γ m = 5 to 7. Civale found a valueof γ m = 5 and Guti´errez et al. reported a value of γ m = 7 from the appliedscaling theory on J ipc ( θ, H ) for YBCO thin films. However, as reported by the sameauthors scaling was obtained in another YBCO thin film containing secondaryphases of BaZrO (BZO) only for a value of γ m = 1.5, which was explained as decreased intrinsic anisotropy for the BZO/YBCO composite. In the latter case,a discussion of pinning energy anisotropy would be meaningful, because there, the in-plane critical current density, J ipc ( θ, H ), is not only affected by the effectiveelectronic mass anisotropy, but depends also strongly on extended pinning centers.ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb S. Haindl, M. Kidszun, F. Onken, A. Mietke, T. Thersleff
Fig. 4. (a)
Angular dependence of the critical current densities at T = 4.2 K for different appliedmagnetic fields (exemplarily shown for µ H = 1 T, 3 T, 5 T, 7 T and 9 T). θ = 180 ◦ correspondsto the configuration H (cid:107) c. (b) Scaling of J c ( θ ) by using the scaling law for the effective magneticfield H eff = (cid:15) θ H = H ( cos ( θ ) + γ − sin ( θ )) . . In the original theory γ corresponds to γ m = m (cid:107) c /m ⊥ c the effective electronic mass anisotropy. Here, γ Scaling is evaluated for overlapping(i.e. scaling ) J c data. (c) Re-mapping exemplarily shown for J c (4.2 K, 7 T). The red line of J c ( θ )indicates the J c -anisotropy due to the mass anisotropy (if γ Scaling corresponds to γ m ) as it isobtained by the anisotropic Ginzburg-Landau scaling as shown in (b). Additional J c anisotropiesfor θ = 90 ◦ and θ = 270 ◦ that can not be described by the scaling theory originate from surfaceand intrinsic pinning effects. (d) Comparison between γ Scaling obtained from (b) and the measured H c -anisotropy and evidence for γ Scaling = γ Hc . Under the assumption of decoupled electron andhole bands we evaluate γ m = 4.8 for the electronic band dominating at high temperatures and γ m =3.2 for the band dominating at low temperatures. T c of the film is 25 K. All transport criticalcurrent measurements were carried out in a commercial Physical Property Measurement System(PPMS) from Quantum Design in magnetic fields up to 9 T. (Data previously published as figure2 in Kidszun et al. ) In-Plane Critical Current Densities and H c Anisotropy
It is well established, that the Fe-based superconductors are multiband superconduc-tors and thus the relationship between γ m and γ Hc (or the anisotropy of the coher-ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb Lessons from Oxypnictide Thin Films (a)
Bright field TEM overview image of the microstructure of the epitaxially grownLaFeAsO − x F x thin film (with the c -axis oriented out-of-plane ). The only extended secondaryphases are LaOF layers at the film/substrate interface and on top of the oxypnictide thin film.Their influence to surface pinning effects can be seen for magnetic fields applied perpendicular tothe c -axis. (b) HRTEM micrograph of the LaFeAsO − x F x phase. (c) The Fast-Fourier-Transform(FFT) of image (b) indicates a highly crystalline film. The b - and c -axis can be measured andcorrespond with the values obtained by X-Ray diffraction. (d) Inverse FFT was filtered for spatialfrequencies and superimposed with an opacity of 50% on top of image (b). This serves as anoise reduction technique to emphasize the crystalline structure of the film. The investigatedTEM-lamella was prepared by Focused Ion Beam (FIB) in-situ lift-out technique and thinnedto electron transparency using a 5 kV Ga + ion beam. All TEM investigations were performedin a C S -corrected FEI Titan 80-300 TEM operating at 300 kV at IFW Dresden. (Previouslyunpublished.) ence lengths, γ ξ and also the relationship between γ m and γ Hc (or the anisotropyof the penetration depths, γ λ , respectively) is generally unknown. As it was al-ready shown for the two-band superconductor MgB , the anisotropies of upperand lower critical field are temperature dependent and, in general, do not match: H ⊥ cc /H (cid:107) cc (cid:54) = H (cid:107) cc /H ⊥ cc ( γ ξ (cid:54) = γ λ ). , , , The anisotropic Ginzburg-Landau scaling approach was applied to the epitax-ially grown LaFeAsO − x F x thin film (Figure 4). The measured angular depen-dence of the in-plane critical current densities, J ipc ( H ) is obtained upon rotationovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb S. Haindl, M. Kidszun, F. Onken, A. Mietke, T. Thersleff of the thin film in an external magnetic field where θ denotes the angle betweenthe magnetic field direction and the c -axis of the film (Figure 4(a)). Note, that wedo not probe the anisotropy of the critical current densities obtained from in-plane transport and c -axis transport measurements here.Surprisingly, the scaling procedure works excellent in a wide temperature in-terval, despite the fact, that the oxypnictides are multiband superconductors (Fig-ure 4(b)). For reasons of clarity, we write the scaling factor as (cid:15) θ = ( cos ( θ ) + γ − sin ( θ )) . and introduce technically γ Scaling . In order to find the overlap of J c ( (cid:15) θ H ) curves for a given temperature, the only parameter that has to be varied isthe scaling parameter, γ Scaling . Two questions arise here: Is the scaling parameter, γ Scaling governed by the H c anisotropy ( γ Hc , γ ξ ) or by the H c anisotropy ( γ Hc , γ λ )? And, what is the relationship between the technically introduced γ Scaling andthe mass anisotropy, γ m ?First, the scaling parameter γ Scaling obtained for different temperatures can beplotted as a function of temperature (Figure 4(d)). Comparison between γ Scaling andthe anisotropy of the upper critical field from pulsed magnetic field measurements(Figure 3) shows an overlap at temperatures around 15 K. At present, furthermeasurements are carried out in order to test the relationship between γ Scaling and γ Hc in a broader temperature interval. Very recently van der Beek et al. pointedout that two effects cause magnetic field and angular dependence of J ipc ( θ, H ).The first effect is the anisotropy of the pinning energy due to the anisotropy ofthe vortex line energy. The second effect is the anisotropy of the vortex core sizes(i.e. the anisotropy of the coherence lengths). In the case of core pinning of smalland point-like defects (size comparable to the coherence length) the second effectdominates the angular dependence of J ipc ( θ, H ) leading to γ Scaling = γ Hc .Second, γ Scaling is interpreted as mass anisotropy γ m in the limits for T → T c and T →
0. A consistent explanation for this correspondence can be delivered, when theprevious obtained results from the temperature dependence of H c are considered. H (cid:107) cc ( T ) was discussed (in analogy to MgB ) in the frame of decoupled effectiveelectron and hole bands. The obtained γ m values in the limits of T → T c (moreprecisely, T → T c, , with T c, denoting the critical temperature of the strongerband) and T → T → T c, , with T c, denoting the critical temperature of theweaker band) would, therefore, correspond to effective electronic mass anisotropiesof the dominating electronic bands. This result is also supported within a two-band Ginzburg-Landau theory, where in the discussed limits (neglecting impurityscattering) γ m, = ( γ Hc ) | T c, and γ m, = ( γ Hc ) | T c, . The temperature dependence of the discussed anisotropies is commonly regardedto be a general feature of multiband superconductivity. The correspondence between γ m and γ Hc can be understood qualitatively in terms of a decoupled bands scenario.The above investigated epitaxially grown LaFeAsO − x F x thin film with T c = 25 Kshows a residual resistance ratio (RRR) of 6.8. It is important to note that thefilm is free of large extended defects (particularly, defects oriented parallel to the c -axis): There is no indication for correlated defects in the microstructure of theovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb Lessons from Oxypnictide Thin Films film (Figure 5). Epitaxial growth is confirmed via X-ray analysis and TEM analysis(Figures 6 and 7). Again, the presence of only small point-like defects supportsstrongly the observation that the scaling parameter in the anisotropic Ginzburg-Landau scaling equals γ Hc . Fig. 6. (a)
X-ray Phi-scan of the (112) reflection of the epitaxially grown LaFeAsO − x F x (La-1111) phase. The full-width-half-maximum (FWHM) is approximately 1 ◦ . (b) X-ray diffractionpattern ( θ − θ scan) of the epitaxially grown LaFeAsO − x F x thin film. (00 l ) reflections of theLaFeAsO − x F x phase are indexed. Reflections from the LaAlO substrate are denoted by S . TheX-ray analysis of the thin film was carried out using an X’Pert Philips X-ray Diffractometer withCu K α radiation in (a) and Co K α radiation in (b). (c) A Selected Area Diffraction (SAD) of theoxypnictide film on the LaFeAsO − x F x phase and the LaOF phase indicates epitaxial growth. Nofurther secondary phases are observed. The investigated TEM-lamella was prepared by FocusedIon Beam (FIB) in-situ lift-out technique and thinned to electron transparency using a 5 kVGa + ion beam. All TEM investigations were performed in a C S -corrected FEI Titan 80-300 TEMoperating at 300 kV at IFW Dresden. ((a) and (b) were previously published as figures 2(b) and(1) in Kidszun et al. ; (c) is previously unpublished.)
5. Possible Applications For Oxypnictides
The high transition temperatures of the oxypnictide superconductors triggered thequestion about their possible use for superconducting applications. Similar to thecuprates, the small coherence length in the oxypnictides can be regarded as respon-sible for a weak-link behaviour of the grain boundaries . Therefore, critical currentdensities in polycrystalline LaFeAsO − x F x films are about two orders of magnitudeovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb S. Haindl, M. Kidszun, F. Onken, A. Mietke, T. Thersleff
Fig. 7. (a)
HRTEM at the epitaxial LaOF/LaFeAsO − x F x (La-1111) interface confirms therelationship La-1111[100] (cid:107) LaOF[100] (FFT insets of the two regions). The interface is exceedinglyclean with no secondary phases or reaction layers and is almost atomically sharp. (b)
HRTEMat the LAO/LaOF interface reveals epitaxial growth of the LaOF phase (FFT insets of the tworegions). A reaction layer appears within the LAO substrate. This layer can be indexed as LAOalbeit with broad spatial frequencies suggesting structural disorder within the LAO crystal lattice.A likely explanation for this is the diffusion of F into the LAO substrate, which could substituteon the O site. The investigated TEM-lamella was prepared by Focused Ion Beam (FIB) in-situ lift-out technique and thinned to electron transparency using a 5 kV Ga + ion beam. All TEMinvestigations were performed in a C S -corrected FEI Titan 80-300 TEM operating at 300 kV atIFW Dresden. (Previously unpublished.) lower (approx. 10 Acm − at 4.2 K) compared to critical current densities in epi-taxially grown LaFeAsO − x F x films (approx. 7 · Acm − at 4.2 K). Comparablevalues for critical current densities have been measured in LaFeAsO − x F x powder-in-tube (PIT) wires. SmFeAsO − x F x thin films and wires show slightly highercritical current densities compared to LaFeAsO − x F x .ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb Lessons from Oxypnictide Thin Films The weak-link behaviour seems to be a general property for Fe-based super-conductors, as it was reported also for Co-doped BaFe As thin films grown onbicrystal substrates. , Therefore, any possible high-current application as thinfilm conductor or PIT wire has to solve the technological problem of (biaxial) grainalignment.In contrast to other Fe-based superconductors of composition or , the oxypnictides have an extended spacer oxide layer between the conductingFeAs layers. This gives rise to the possible formation of intrinsic Josephson junctionsin these compounds, which can be applied in emission sources for Terahertz radi-ation. In collaboration with Paul M¨uller from University Erlangen-N¨urnberg firstexperiments on current injection perpendicular to the FeAs layers in mesa struc-tures of 5 × µ m in lateral size for a 100 nm thin LaFeAsO − x F x film were carriedout. The few published results are still controverse. On the one hand, Kashiwaya etal. reported an intrinsic Josephson effect for PrFeAsO . single crystals. On theother hand, c -axis transport measurements in oxygen deficient SmFeAsO . sin-gle crystals have not yet confirmed the presence of intrinsically Josephson coupledlayers.
6. Summary
Today, after five years of oxypnictide thin film growth, an all in-situ
PLD fabricationhas still not been successful. The very few attempts so far resulted in the formationof impurity phases and stable oxides and oxyfluorides, but not in the growth ofthe so-called phase. Instead, a two-step process consisting of a conventionalroom temperature PLD and a subsequent ex-situ post deposition heat treatmentwas successfully employed in the growth of LaFeAsO − x F x and SmFeAsO − x F x thin films. The main process parameters are however governed by the heat treat-ment rather than by the PLD itself. Therefore, important PLD parameters for anoxypnictide thin film growth have not yet been investigated in detail. Besides, thereproducibility of the oxypnictide thin film growth would be improved for an all in-situ PLD process. The growth of oxypnictide thin films thus stays challenging.Polycrystalline and even epitaxial thin film growth was achieved by the two-step process. Still, the very stable LaOF impurity phase is present in the films.Microstructural investigations revealed that the LaOF phase is located generally atthe film/substrate interface and on top of the film with the superconducting phase sandwiched in between. The overall clean and dense LaFeAsO − x F x layer canbe perfectly analyzed by electrical transport measurements.As demonstrated for an epitaxially grown LaFeAsO − x F x thin film, theanisotropic Ginzburg-Landau scaling of the angular dependent critical current den-sity can be applied. The resulting scaling parameter that corresponds to the squareroot of the effective electronic mass anisotropy in the limits for T → T c and T → S. Haindl, M. Kidszun, F. Onken, A. Mietke, T. Thersleff of multiband superconductivity. The multiband behaviour can also be probed inthe temperature dependence of the upper critical field itself, which we were able todescribe for LaFeAsO − x F x by a simplified two band model.Finally, the applicability of oxypnictides in superconducting devices faces tech-nological limitations. Oxypnictides are not strongly two-dimensional and have effec-tive electronic mass anistropies in the order of 5 to 7. On the one hand, electronicapplications based on intrinsic Josephson junctions will be more difficult to realizethan in Bi Sr CaCu O . On the other hand, applications based on high currentsand pinning will still be affected by thermal fluctuations and flux flow at highertemperatures near T c . In addition, the weak-link behaviour of grain boundaries is adisadvantage because grains have to be large and aligned in order to achieve a highcritical current flow in power applications. Acknowledgements
The authors would like to thank M. Langer, J. Werner and G. Behr for target prepa-ration, E. Reich, A. Reisner, K. Nenkov, U. Besold, and J. H¨anisch for technicalassistance. Furthermore, the authors would like to thank S. F¨ahler, G. Fuchs, andS.-L. Drechsler for fruitful discussions. Oxypnictide thin film growth and charac-terization was funded by the German Research Foundation (DFG) under projectnumbers HA5934/1-1 and HA5934/3-1.
References
1. Y. Kamihara, H. Hiramatsu, M. Hirano, R. Kawamura, H. Yanagi, T. Kamiya, H.Hosono,
J. Am. Chem. Soc. , 10012 (2006).2. Y. Kamihara, T. Watanabe, M. Hirano, H. Hosono,
J. Am. Chem. Soc. , 3296(2008).3. B. I. Zimmer, W. Jeitschko, J. H. Albering, R. Glaum, M. Reehuis,
J. Alloys Comp. , 238 (1995).4. P. Quebe, L. J. Terb¨uchte, W. Jeitschko,
J. Alloys Comp. , 70 (2000).5. A. L. Ivanovskii,
Russ. Chem. Rev. , 1 (2010).6. A. L. Ivanovskii, Physics-Uspekhi , 12 (2008).7. D. J. Singh, M.-H. Du, Phys. Rev. Lett. , 237003 (2008).8. Z.-A. Ren, W. Lu, J. Yang, W. Yi, X-L. Shen, Z.-C. Li, G.-C. Che, X.-L. Dong, L.-L.Sun, F. Zhou, Z.-X. Zhao,
Chin. Phys. Lett. , 2215 (2008).9. L. Boeri, O. V. Dolgov, A. A. Golubov, Phys. Rev. Lett. , 026403 (2008).10. C. de la Cruz, et al. (2008),
Nature , 899 (2008).11. Z.-A. Ren, G.-C. Che, X.-L. Dong, J. Yang, W. Lu, W. Yi, X-L. Shen, Z.-C. Li, L.-L.Sun, F. Zhou, Z.-X. Zhao,
EPL , 17002 (2008).12. C. Wang, S. Jiang, Q. Tao, Z. Ren, Y. Li, L. Li, C. Feng, J. Dai, G. Cao, Z.-A. Xu, EPL Phys. Rev. B , 054521(2009).14. H. Okada, K. Igawa, H. Takahashi, Y. Kamihara, M. Hirano, H. Hosono, K. Matsub-ayashi, Y. Uwatoko, J. Phys. Soc. Jpn. , 113712 (2008). ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb Lessons from Oxypnictide Thin Films
15. H. Hiramatsu, T. Katase, T. Kamiya, M. Hirano, H. Hosono,
Appl. Phys. Lett. ,162504 (2008).16. E. Backen, S. Haindl, T. Niemeier, R. H¨uhne, T. Freudenberg, J. Werner, G. Behr, L.Schultz, B. Holzapfel, Supercond. Sci. Technol. , 122001 (2008).17. T. Kawaguchi, H. Uemura, T. Ohno, R. Watanabe, M. Tabuchi, T. Ujihara, K. Tak-enaka, Y. Takeda, H. Ikuta, Appl. Phys. Express , 093002 (2009).18. T. Kawaguchi, H. Uemura, T. Ohno, M. Tabuchi, T. Ujihara, K. Takenaka, Y. Takeda,H. Ikuta, Appl. Phys. Lett. , 042509 (2010).19. S. Ueda, S. Takeda, S. Takano, A. Yamamoto, M. Naito, Appl. Phys. Lett. , 232505(2011).20. S. Ueda, S. Takeda, S. Takano, A. Mitsuda, M. Naito, Jpn. J. Appl. Phys. , 010103(2012).21. S. Ueda, T. Yamagishi, S. Takeda, S. Agatsuma, S. Takano, A. Mitsuda, M. Naito, Physica C , 1167 (2011).22. D. B. Chrisey, G. H. Hubler, eds., Pulsed Laser Deposition of Thin Films. (John Wileyand Sons, Inc., 1994)23. R. Eason, ed., Pulsed Laser Deposition of Thin Films: Applications-Led Growth ofFunctional Materials. (Wiley Interscience, Hoboken, 2007)24. J. T. Cheung, H. Sankur,
Crit. Rev. Solid State Mat. Sci. , 63 (1988).25. D. Dijkkamp, T. Venkatesan, X. D. Wu, S. A. Shaheen, N. Jisrawi, Y. H. Min-Lee,W. L. McLean, M. Croft, Appl. Phys. Lett. , 619 (1987).26. C. B. Arnold, M. J. Aziz, Appl. Phys. A , S23 (1999).27. J. Schou, Appl. Surf. Sci. , 5191 (2009).28. S. Lee, J. Jiang, J. D. Weiss, C. M. Folkman, C. W. Bark, C. Tarantini, A. Xu,D. Abraimov, A. Polyanskii, C. T. Nelson, Y. Zhang, S. H. Baek, H. W. Jang, A.Yamamoto, F. Kametani, X. Q. Pan, E. E. Hellstrom, A. Gurevich, C. B. Eom, D. C.Larbalestier,
Appl. Phys. Lett , 212505 (2009).29. E.-M. Choi, S.-G. Jung, N. H. Lee, Y.-S. Kwon, W. N. Kang, D. H. Kim, M.-H. Jung,S.-I. Lee, L. Sun, Appl. Phys. Lett. , 062507 (2009).30. M. Kidszun, S. Haindl, T. Thersleff, J. Werner, M. Langer, J. Hnisch, E. Reich, L.Schultz, B. Holzapfel, EPL , 57005 (2010).31. A. Brinkman, D. Mijatovic, G. Rijnders, V. Leca, H. J. H. Smilde, I. Oomen, A.A. Golubov, F. Roesthuis, S. Harkema, H. Hilgenkamp, D. H. A. Blank, H. Rogalla, Physica C , 1 (2001).32. D. H. A. Blank, H. Hilgenkamp, A. Brinkman, D. Mijatovic, G. Rijnders, H. Rogalla,
Appl. Phys. Lett. , 394 (2001).33. H. Y. Zhai, H. M. Christen, L. Zhang, M. Paranthaman, C. Cantoni, B. C. Sales, P.H. Fleming, D. K. Christen, D. H. Lowndes, J. Mater. Res. , 2759 (2001).34. W. Y. Lee, V. Y. Lee, J. Salem, T. C. Huang, R. Savoy, D. C. Bullock, S. S. P. Parkin, Appl. Phys. Lett. , 329 (1988).35. C. C. Tsuei, A. Gupta, G. Trafas, D. Mitzi, Science , 1259 (1994).36. N. H. Lee, S.-G. Jung, D. H. Kim, W. N. Kang,
Appl. Phys. Lett , 202505 (2010).37. S. Haindl, M. Kidszun, A. Kauffmann, K. Nenkov, N. Kozlova, J. Freudenberger, T.Thersleff, J. H¨anisch, J. Werner, E. Reich, L. Schultz, B. Holzapfel, Phys. Rev. Lett. , 077001 (2010).38. M. Kidszun, S. Haindl, E. Reich, J. H¨anisch, K. Iida, L. Schultz, B. Holzapfel,
Super-cond. Sci. Technol , 022002 (2010).39. J. Karpinski, N. D. Zhigadlo, S. Katrych, Z. Bukowski, P. Moll, S. Weyeneth, H.Keller, R. Puzniak, M. Tortello, D. Daghero, R. Gonnelli, I. Maggio-Aprile, Y. Fasano,Ø. Fischer, K. Rogacki, B. Batlogg, Physica C , 370 (2009). ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb S. Haindl, M. Kidszun, F. Onken, A. Mietke, T. Thersleff
40. R. W. McCallum, J.-Q. Yan, G. E. Rustan, E. D. Mun, Y. Singh, S. Das, R. Nath, S.L. Bud’ko, K. W. Dennis, D. C. Johnston, P. C. Canfield, M. J. Kramer, A. Kreyssig,T. A. Lograsso, A. I. Goldman,
J. Appl. Phys. , 123912 (2009).41. H. Suhl, B. T. Matthias, L. R. Walker,
Phys. Rev. Lett. , 552 (1959).42. V. A. Moskalenko Fiz. Met. Metalloved , 503 (1959) [ Phys. Met. Metallogr. (USSR) , 25 (1959)].43. J. Nagamatsu, N. Nakagawa, T. Muranaka, Y. Zenitani, J. Akimitsu, Nature , 63(2001).44. A. Y. Liu, I. I. Mazin, J. Kortus,
Phys. Rev. Lett. , 087005 (2001).45. S. Leb`egue, Phys. Rev. B , 035110 (2007).46. H. Eschrig, K. Koepernik, Phys. Rev. B , 104503 (2009).47. A summary of gap values for Fe-based superconductors can be found in D. V. Ev-tushinsky, et al. New Journal of Physics , 055069 (2009).48. R. S. Gonnelli, et al. Phys. Rev. B , 184526 (2009).49. Yu. G. Naidyuk, O. E. Kvitnitskaya, I. K. Yanson, G. Fuchs, S. Haindl, M. Kidszun,L. Schultz, B. Holzapfel, Supercond. Sci. Technol. , 065010 (2011).50. E. Helfand, N. R. Werthamer, Phys. Rev. Lett. , 686 (1964).51. E. Helfand, N. R. Werthamer, Phys. Rev. , 288 (1966).52. N. R. Werthamer, E. Helfand, P. C. Hohenberg,
Phys. Rev. , 295 (1966).53. F. Hunte, J. Jaroszynski, A. Gurevich, D. C. Larbalestier, R. Jin, A. S. Sefat, M.A. McGuire, B. C. Sales, D. K. Christen, D. Mandrus,
Nature , 903 (2008). Thesample investigated in this publication was a polycrystalline bulk where a 10% R n criterion ( T c ) and a 90% R n criterion ( T c ) were used to estimate H (cid:107) cc and H ⊥ cc . R n denotes the normal state resistance above T c .54. A. Gurevich, Phys. Rev. B , 184515 (2003).55. V. Braccini, et al. Phys. Rev. B , 012504 (2005).56. V. Braccini, et al. Phys. Rev. B , 179902e (2005).57. Matlab Version 7.0 (2004), (computer software), The MathWorks Inc., Natick, Mas-sachusetts (USA).58. A. Gurevich, Physica C , 160 (2007).59. Y. Kohama, Y. Kamihara, S. A. Baily, L. Civale, S. C. Riggs, F. F. Balakirev, T.Atake, M. Jaime, M. Hirano, H. Hosono,
Phys. Rev. B , 144527 (2009).60. G. A. Ummarino, M. Tortello, D. Daghero, R. S. Gonnelli, Phys. Rev. B , 172503(2009).61. J. Jaroszynski, et al. Phys. Rev. B , 174523 (2008).62. M. Tinkham, Introduction to Superconductivity. (2nd ed., McGraw-Hill 1996, p.319f).63. G. Blatter, V. B. Geshkenbein, A. I. Larkin, Phys. Rev. Lett. , 875 (1992).64. G. Blatter, M. V. Feigel’man, V. B. Geshkenbein, A. I. Larkin, V. M. Vinokur, Rev.Mod. Phys. , 1125 (1994).65. G. Blatter, V. B. Geshkenbein, Vortex Matter. (In: K. H. Bennemann, J. B. Ketterson(eds.), The Physics of Superconductors, Vol. 1, Springer 2003, p. 787).66. D. Braithwaite et al., J. Low. Temp. Phys. , 1 (1993).67. X. Xu, J. Fang, X. Cao, K. Li, W. Yao, Z. Qi, Solid State Comm. , 501 (1994).68. L. Civale, B. Maiorov, A. Serquis, J. O. Willis, J. Y. Coulter, H. Wang, X. Jia, P. N.Arendt, M. Jaime, J. L. MacManus-Driscoll, M. P. Maley, S. R. Foltyn, J. Low. Temp.Phys. , 87 (2004).69. J. Guti´errez, T. Puig, X. Obradors,
Appl. Phys. Lett. , 162514 (2007).70. J. Guti´errez, A. Llord´es, J. G´azques, M. Gibert, N. Rom´a, S. Ricart, A. Pomar, F.Sandiumenge, N. Mestres, T. Puig, X. Obradors, Nat. Mat. , 367 (2007).71. V. G. Kogan, S. L. Bud’ko, Physica C , 131 (2003). ovember 11, 2018 13:55 WSPC/INSTRUCTION FILE ws-ijmpb
Lessons from Oxypnictide Thin Films
72. Z. X. Shi, M. Tokunaga, T. Tamegai, Y. Takano, K. Togano, H. Kito, H. Ihara,
Phys.Rev. B , 104513 (2003).73. A. Gurevich, S. Patnaik, V. Braccini, K. H. Kim, C. Mielke, X. Song, L. D. Cooley,S. D. Bu, D. M. Kim, J. H. Choi, L. J. Belenky, J. Giencke, M. K. Lee, W. Tian, S. Q.Pan, A. Siri, E. E. Hellstrom, C. B. Eom, D. C. Larbalestier, Supercond. Sci. Technol. , 278 (2004).74. L. Lyard, P. Sazb´o, T. Klein, J. Marcus, C. Marcenat, K. H. Kim, B. W. Kang, H. S.Lee, S. I. Lee, Phys. Rev. Lett. , 057001 (2004).75. M. Kidszun, S. Haindl, T. Thersleff, J. H¨anisch, A. Kauffmann, K. Iida, J. Freuden-berger, L. Schultz, B. Holzapfel, Phys. Rev. Lett. , 137001 (2011).76. C. J. van der Beek, M. Konczykowski, R. Prozorov,
Supercond. Sci. Technol. ,084010 (2012).77. I. N. Askerzade, Physics-Uspekhi , 1003 (2006).78. Z. Gao, L. Wang, Y. Qi, D. Wang, X. Zhang, Y. Ma, Supercond. Sci. Technol ,105024 (2008).79. T. Katase, Y. Ishimaru, A. Tsukamoto, H. Hiramatus, T. Kamiya, K. Tanabe, H.Hosono, Nature Commun. , 409 (2011).80. H. Kashiwaya, K. Shirai, T. Matsumoto, H. Shibata, H. Kambara, M. Ishikado, H.Eisaki, A. Iyo, S. Shamoto, I. Kurosawa, S. Kashiwaya, Appl. Phys. Lett. , 202504(2010).81. J.-H. Park, H.-S. Lee, H.-J. Lee, B. K. Cho, Y.-J. Doh, J. Korean Phys. Soc. ,58