LaFeAsO 1−x F x thin films: high upper critical fields and evidence of weak link behavior
S. Haindl, M. Kidszun, A. Kauffmann, K. Nenkov, N. Kozlova, J. Freudenberger, T. Thersleff, J. Haenisch, J. Werner, E. Reich, L. Schultz, B. Holzapfel
LLaFeAsO − x F x thin films: high upper critical fields and evidence of weak link behavior. S. Haindl * , M. Kidszun, A. Kauffmann, K. Nenkov, N. Kozlova, J. Freudenberger,T. Thersleff, J. Hänisch, J. Werner, E. Reich, L. Schultz, and B. Holzapfel IFW Dresden, P. O. Box 270116, D–01171 Dresden, Germany.
Superconducting LaFeAsO − x F x thin films were grown on single crystalline LaAlO substrateswith critical temperatures (onset) up to 28 K. Resistive measurements in high magnetic fields up to40 T reveal a paramagnetically limited upper critical field, µ H c (0) around 77 T and a remarkablesteep slope of − . TK − near T c . From transport measurements we observed a weak link behaviorin low magnetic fields and the evidence for a broad reversible regime. PACS numbers: 74.25.Dw 74.25.Qt 74.25.Sv 74.78.Bz
Since the discovery [1–3] of high temperature super-conductivity in the R FeAsO − x F x ( R = rare earth) ironpnictides, the so called ‘1111’ phase or family, and sub-sequently in the intermetallic Ba − x K x Fe As , the ‘122’phase, an intensive investigation of this material classstarted [4, 5] and related compounds of the FeSe and theLiFeAs–structure, respectively the ‘11’ and ‘111’ phase,have been found [6, 7]. The quaternary iron pnictidesexhibit high superconducting transition temperatures upto 55 K, high upper critical fields, and a multiband char-acter. Consequently their classification between MgB and the cuprates is heavily discussed to date. This sand-wich position and the vicinity of magnetic order [8] makesthem a good candidate not only for new phenomena insuperconductivity but also for an understanding of thepairing mechanism in high temperature superconductorsincluding the cuprates.Similarities between the cuprate high temperature su-perconductors and the iron pnictides have already beenpointed out, and the analogy is mainly based upon thelayered structure, the charge carrier doping, and the vicin-ity of an antiferromagnetic phase [4, 9]. On the otherhand, there are crucial differences between the cupratesand the iron pnictides. Very recent theoretical resultson the basis of tight–binding calculations highlight themulti–band character of the electronic band structure,and a clear two–dimensional behavior of the band struc-ture near the Fermi energy was found for the ‘1111’ phase[10]. In addition, a possible Pauli limitation of the uppercritical field is still under discussion [11, 12].In case of the ‘1111’ phase most of the experimentalinvestigation were undertaken using polycrystals and theavailable NdFeAsO − x F x and SmFeAsO − x F x single crys-tals [13, 14]. The first superconducting LaFeAsO − x F x thin film has been grown on LaAlO substrate by pulsedlaser deposition (PLD) [15]. Other thin films are reportedfor the ‘122’ phase in Ref. [16] and for the ‘11’ structurein Ref. [17]. PLD is regarded as a versatile tool for thinfilm fabrication [18], and one of its advantages is the sto-ichiometric transfer of target material to the substrate.However, it seems to remain a difficult task to obtainsuperconductivity in the case of the iron based fluorinedoped oxypnictide thin films. A crucial point is the sto-ichiometric control of the fluorine content in the grown Fig. 1. Bright field TEM image (orientation contrast) of theLaFeAsO − x F x thin film. The film is polycrystalline with asignificant portion of grains growing elongated perpendicularto the substrate surface. No pores were observed. samples and poses a challenge to the growth of the quater-nary compounds. In addition, due to the high reactivityof the rare earth elements (especially La), the suppressionof oxide phases is a key issue for thin film deposition.We succeeded in the fabrication of a LaFeAsO − x F x thinfilm with optimal critical temperature and a completesuperconducting transition. High–quality thin films aremandatory for electronic devices, multilayers or Josephsonjunctions from the new superconductors, which may leadto new effects based on the interplay of superconductiv-ity and magnetism. Detailed results of superconductingLaFeAsO − x F x single crystals are not reported so far,therefore the deposition of thin films is of enormous inter-est for fundamental studies. For instance, the behavior ofgrain boundaries is still an open question, whether theyact as pinning sites or as weak links.We present the first detailed transport investigation ofsuperconducting LaFeAsO − x F x thin films in magneticfields up to 40 T. Our measurements demonstrate a highupper critical field and Pauli limitation. In our transportmeasurements we observe small critical current densitiesand weak link behavior in this iron pnictide superconduc-tors.Thin films were grown at room temperature on singlecrystalline LaAlO (001) (LAO) substrates by standard a r X i v : . [ c ond - m a t . s up r- c on ] J a n (cid:114) ( m (cid:87) c m ) Temperature (K)
Temperature (K) N o r m a li z ed r e s i s t an c e R / R N (cid:109) H = Fig. 2. The resistivity shows metallic behavior and supercon-ductivity below a temperature of 28 K with a heavy broadeningof the superconductive transition in an external magnetic field(shown up to 9 T). on–axis PLD using a KrF laser (Lambda Physik) with awavelength of λ = 248 nm, a pulse duration of τ = 30 ns,and an energy density of (cid:15) ≈ Jcm − at the target sur-face. Vacuum conditions in the deposition chamber were p base = 10 − mbar. The nominal target composition isLaFeAsO . F . . An ex–situ post annealing step at950 ◦ C for 4 hours in a sealed quartz tube followed theroom temperature deposition. Thin film preparation de-tails can be found in Ref. [15]. A further reduction ofthe oxygen partial pressure, compared to previous experi-ments, results in a complete superconducting transitionto zero resistivity and reduces the fraction of impurityphases (LaOF, La O ). Structural investigation by trans-mission electron microscopy (TEM) reveals a homoge-neous film without pores. The cross section bright fieldimage (Fig. 1), where the orientation contrast dominates,in combination with Fast Fourier Transformation (FFT)shows polycrystallinity and columnar grains elongatedperpendicular to the substrate surface. From detailedelectron energy loss spectroscopy (EELS) as well as en-ergy dispersive x–ray analysis (STEM–EDX) we concludephase homogeneity in the superconducting layer examinedby TEM.The ‘1111’ phase was identified by X–ray diffraction(Bragg–Brentano) using Co K α radiation showing (003),(110) and (200) as the three strongest reflections. Thelattice constants derived are a = 4 . Å and c = 8 . Å. Acut piece with dimensions of 1 mm × T c, = 28 K and residual resistivity ratio(RRR) of 4.1 (inset in Fig. 2) was used for resistive mea-surements. No spin density anomaly was found around150 K, and from the comparison with the temperaturedependence of the resistivity in polycrystals for differentfluorine doping levels, a fluorine content of above ∼ is estimated. Four probe electrical transport measure-ments were carried out in a commercial Physical Property no r m a li z ed r e s i s t an c e R / R N µ H (T) (a) µ H ( T ) t = T/T C pulsedfield staticfield T C,90 T C,10
WHH slope:-6.2 TK -1 (b) Fig. 3. (a) The pulsed field data (circles) and correspondingflattened curves are shown up to 40 T for different tempera-tures. (b) The magnetic phase diagram of the LaFeAsO − x F x thin film. T c, and T c, of measurements in pulsed fieldsand static fields are given. The low field T c, from PPMSmeasurements fit the pulsed field data well and exhibit a slopenear T c of − . TK − (dashed line). The upper critical fielddata deviates from the WHH model. Measurement System (PPMS) up to magnetic fields of9 T (100 µA dc current) as well as up to 14 T (100 µA ac amplitude). Both R ( T ) and R ( µ H ) data have beenrecorded in static fields.Resistive measurements, R ( T ) , show well defined andfully developed transitions from the normal to the super-conducting state with R = 0 (Fig. 2). The broadeningof the superconductive transitions in applied magneticfields can be ascribed to the effects of i ) a polycrystallinestructure of the thin film, and ii ) low critical currentdensities. Pulsed field measurements, R ( µ H ) , up to 40 Twere performed in a cryostat equipped with a solenoid,which is operated in pulsed mode (with an ac amplitude of95 µ A and a frequency of 10 kHz). A detailed descriptionof the pulsed field system can be found elsewhere [19]. E ( µ V / mm ) J (A/cm²) µ H (T) Temperature (K)0.1 T0 T2 K5 K10 K - C r i t i c a l c u rr en t den s i t i e s ( A c m ) - C r i t i c a l c u rr en t den s i t i e s ( A c m ) (a)(b) Fig. 4. (a) Temperature dependence of the critical currentdensities in zero field ( • ) and in a small applied field of 0.1 T( ◦ ). The inset shows the E ( J ) –measurements in zero field fordifferent temperatures (4, 5, 6, 10, 15, 17 and 19 K) and the1 µ Vmm − criterion for evaluation.(b) Field dependence of thecritical current densities for temperatures of 2 K, 5 K and 10K. The fits (lines) correspond to the behavior of weakly linkedgrains. Heating of the sample during the pulse is negligible, aschecked by several test pulses with successively increasingmaximum field. All measurements were carried out foran orientation of the film surface perpendicular to themagnetic field direction.The resistive data of the pulsed field measurementsfor different temperatures show a linear increase in anextended field interval (Fig. 3 (a)). There is a qualitativeagreement between the pulsed field data and the low field R ( µ H ) data from the PPMS measurements up to 14 T, in T c, . Deviations between low–field (PPMS) and pulsedfield measurements can be observed for the evaluation of T c, at low temperatures. The collapse of the current den-sity at higher fields and temperatures, addressed furtherbelow, can explain the observed behavior well if one takesinto account that the measurement current of 100 µ A ex-ceeds the critical current. The additional perturbationof possibly pinned flux by the application of an ac cur-rent in the pulsed field measurements versus a dc currentin the PPMS measurements further supports this inter-pretation. The upper critical field obtained by the T c, criterion (Fig. 3 (b)) shows a remarkable slope near T c of | dµ H c dT | T c = 6 . TK − . This high value exceeds the previ-ously ones for LaFeAsO − x F x [11, 12] and is at the lowerlimit of the values found in NdFeAsO − x F x single crys-tals [9, 20]. The Werthamer–Helfand–Hohenberg (WHH)estimation of the upper critical field at zero temperatureyields . T c | dµ H c dT | T c = 120 T [21]. The deviation fromthe WHH model with increasing applied magnetic fieldsindicates a Pauli limited upper critical field which canbe estimated with a coupling constant λ = 0 . to around µ H P = (1 + λ ) µ H P BCS = 77
T, where µ H P BCS = 1 . T c is the BCS paramagnetic limit (for weak coupling).Transport current measurements for zero field and µ H = 0 . T evaluated for an electrical field criterionof 1 µ Vmm − (Fig. 4 (a)) show a fast decrease of the criti-cal current density, J c , with increasing temperature. Thetemperature dependence of J c is proportional to (1 − t ) . ,with t = T /T c, where T c, is taken from the offset of theresistive transition. The observed small values of J c at lowtemperatures in the thin film (< 2 kAcm − ) are compara-ble with the results in LaFeAsO . F . powder–in–tube(PIT) wires [22]. There are several possible explanationsto the small absolute values of the critical current den-sities, and the polycrystalline structure of the film is astrong candidate for the critical current limitation. Ex-periments on bicristalline YBa Cu O − δ thin films havedemonstrated that high–angle grain boundaries are ableto reduce the critical current by a factor of 10 [23]. Al-though there is no pronounced weak–link behavior seen inthe E ( J ) –curves (see inset in Fig. 4 (a)), the field depen-dence of the critical current densities (Fig. 4 (b)) suggestsstrongly a weak link behavior due to grain boundaries(compare Fig. 1) parallel to the applied field direction.The fit to the experimental data follows well the descrip-tion of weakly linked grains with J c ∝ (cid:0) µ HB (cid:63) (cid:1) − with B (cid:63) being a characteristic field of the weak links [24]. Inaddition, a linear relation between voltage and current hasbeen observed at increased magnetic fields and tempera-tures, which indicates a dominant reversible regime in themagnetic phase diagram. The irreversibility or depinningline is normally estimated by the offset of the criticaltemperature in R ( T ) measurements (Fig. 2), but detailedmagnetization measurements have to be made in orderto confirm this interpretation in the discussed thin films.Since the nature of the vortex matter in the quaternaryiron pnictides is still unclear, and the possibility of vortexpancakes, for example, due to the alternating stackingof FeAs and LaO − x F x layers cannot be excluded at themoment, further investigation of flux pinning are of highinterest. The problem of granularity and the enormouslycomplex analysis of a vortex dynamical regime has beenpointed out in Ref. [25] for polycrystals. Therefore, a dif-ference in inter–grain and intra–grain vortex dynamics canalso be expected in the polycrystalline iron pnictide thinfilms. The dissipative signature in the resistive transitionspoints towards a strong similarity with Bi Sr CaCu O (Bi–2212) or other highly two–dimensional cuprate su-perconductors [26], although the mass anisotropy in theoxypnictides is definitive lower (for instance, γ ≈ − inRef. [20]). In the first instance, this is quite intriguing, butcompletely reasonable with regard to the two–dimensionalconfinement of the charge carriers within the FeAs layer.Certainly, the investigation of the irreversible propertiesand the vortex matter in the new iron pnictides will be anecessary task in order to fully understand the magneticphase diagram of these materials. The grain boundaries inthe investigated thin film do not play the role of effectivepinning centers since the coherence length obtained fromthe Ginzburg–Landau relation, µ H c = Φ πξ , is only2.1 nm. Consequently, the increase of the critical currentdensities in the new superconducting iron pnictides de-pends drastically on the possibility of the introduction ofpinning sites into the material as well as on the reductionof the number of high–angle grain boundaries.To conclude, we have demonstrated the growth of su-perconducting LaFeAsO − x F x thin films with a criticaltemperature of 28 K and a steep increase ( − . TK − ) ofthe upper critical field. The Pauli limited upper criticalfield, µ H c (0) , was estimated to ∼ T in accordancewith experimental data on polycrystalline bulk material.From transport current measurements there is evidenceof a broad dissipative (reversible) regime. The field de-pendence of the critical current densities support the factof a weakly linked network in accordance to the veryshort coherence length, ξ , similar to the cuprates. Dueto the fact that grain–boundaries in this material actas weak links epitaxial thin films will be necessary forfundamental experimental investigation, including phasesensitive tests [27], the fabrication of Josephson junctionsand multilayers with new interesting properties. Acknowledgments.
The authors would like to thankMarco Langer, Ulrike Besold, Margitta Deutschmann andespecially Stefan Pofahl for technical support and TetyanaShapoval for critical reading of the manuscript. We ac-knowledge S. Baunack, C. Deneke and O. G. Schmidt forsupport with the preparation of the TEM sample. * Corresponding author: [email protected] [1] Y. Kamihara, T. Watanabe, M. Hirano, H. Hosono, J.Am. Chem. Soc. , 3296 (2008).[2] M. Rotter, M. Tegel, D. Johrendt, Phys. Rev. Lett. ,107006 (2008).[3] 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,Eur. Phys. Lett. , 17002 (2008).[4] S. Uchida, J. Phys. Soc. Jpn. , Suppl. C, 9 (2008).[5] M. V. Sadovskii, Physics Uspekhi , 1201 (2008).[6] F.-C. Hsu, J.-Y. Luo, K.-W. Yeh, et al., Proc. Nat. Acad.Sci. (USA) , 14262 (2008).[7] X.-C. Wang, Q. Q. Liu, Y. X. Lv, W. B. Gao, L. X. Yang,R. C. Yu, F. Y. Li, C. Q. Jin, Solid State Comm. ,538 (2008).[8] C. de la Cruz, Q. Huang, J. W. Lynn, J. Li, W. RatcliffII, J. L. Zarestky, H. A. Mook, G. F. Chen, J. L. Luo, N.L. Wang, P. Dai, Nature , 899 (2000).[9] C. Xu, S. Sachdev, Nature Physics , 898 (2008).[10] H. Eschrig, K. Koepernik, Phys. Rev. B , 104503 (2009).[11] G. Fuchs, et al., Phys. Rev. Lett. , 237003 (2008).[12] 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).[13] Y. Jia, P. Cheng, L. Fang, H. Luo, H. Yang, C. Ren, L.Shan, C. Gu, H.-H. Wen, Appl. Phys. Lett. , 032503(2008).[14] N. D. Zhigadlo, S. Katrych, Z. Bukowski, S. Weyeneth,R. Puzniak, J. Karpinski, J. Phys.: Condens. Matter ,342202 (2008).[15] E. Backen, S. Haindl, T. Niemeier, R. Hühne, T. Freuden-berg, J. Werner, G. Behr, L. Schultz, B. Holzapfel, Super-cond. Sci. Technol. , 122001 (2008).[16] S. A. Baily, Y. Kohama, H. Hiramatsu, B. Maiorov, F. F.Balakirev, M. Hirano, H. Hosono, Phys. Rev. Lett. ,117004 (2009).[17] M. K. Wu, et al., Physica C , 340 (2009).[18] D. B. Chrisey, G. H. Hubler (eds.): Pulsed laser depositionof thin films. John Wiley and Sons, Inc. (1994).[19] H. Krug et al., Physica B , 605 (2001).[20] J. Jaroszynski, et al., Phys. Rev. B , 174523 (2008).[21] N. R. Werthamer, E. Helfand, P. C. Hohenberg, Phys.Rev. , 295 (1966).[22] Z. Gao, L. Wang, Y. Qi, D. Wang, X. Zhang, Y. Ma,Supercond. Sci. Technol. , 105024 (2008).[23] H. Hilgenkamp, J. Mannhart, Rev. Mod. Phys. , 485(2000).[24] K.-H. Müller, D. N. Matthews, R. Driver, Physica C ,339 (1992).[25] M. Polichetti, M. G. Adesso, D. Zola, J. Luo, G. F. Chen,Z. Li, N. L. Wang, C. Noce, S. Pace, Phys. Rev. B ,224523 (2008).[26] T. T. M. Palstra, B. Batlogg, L. F. Schneemeyer, J. V.Waszczak, Phys. Rev. Lett. , 1662 (1988).[27] C.-T. Chen, C. C. Tsuei, M. B. Ketchen, Z.-A. Ren, Z. X.Zhao, arXiv: 0905.3571arXiv: 0905.3571