LETTERS

Two-band superconductivity in LaFeAsO0.89F0.11 atvery high magnetic fieldsF. Hunte1, J. Jaroszynski1, A. Gurevich1, D. C. Larbalestier1, R. Jin2, A. S. Sefat2, M. A. McGuire2, B. C. Sales2,D. K. Christen2 & D. Mandrus2

The recent synthesis of the superconductor LaFeAsO0.89F0.11 withtransition temperature Tc < 26 K (refs 1–4) has been quickly fol-lowed by reports of even higher transition temperatures in relatedcompounds: 41 K in CeFeAsO0.84F0.16 (ref. 5), 43 K inSmFeAsO0.9F0.1 (ref. 6), and 52 K in NdFeAsO0.89F0.11 andPrFeAsO0.89F0.11 (refs 7, 8). These discoveries have generatedmuch interest9,10 in the mechanisms and manifestations of uncon-ventional superconductivity in the family of doped quaternarylayered oxypnictides LnOTMPn (Ln: La, Pr, Ce, Sm; TM: Mn, Fe,Co, Ni; Pn: P, As), because many features of these materials setthem apart from other known superconductors. Here we reportresistance measurements of LaFeAsO0.89F0.11 at high magneticfields, up to 45 T, that show a remarkable enhancement of theupper critical field Bc2 compared to values expected from theslopes dBc2/dT < 2 T K21 near Tc, particularly at low temperatureswhere the deduced Bc2(0) < 63–65 T exceeds the paramagneticlimit. We argue that oxypnictides represent a new class of high-field superconductors with Bc2 values surpassing those of Nb3Sn,MgB2 and the Chevrel phases, and perhaps exceeding the 100 Tmagnetic field benchmark of the high-Tc copper oxides.

The continuing search for new superconductors has recentlyyielded a new family of oxypnictides composed of alternatingLaO12xFx and FeAs layers1–4 with Tc values of 25–28 K, which canbe raised to 40–43 K by replacing La with Ce (ref. 5) or Sm (ref. 6) orto 52 K by replacing La with Nd and Pr (refs 7, 8). Several experimentsand band structure calculations suggest unconventional supercon-ductivity in the paramagnetic Fe layer. First, ab initio calculationsindicate that superconductivity originates from the d orbitals of whatwould normally be expected to be pair-breaking magnetic Fe ions,suggesting that new non-phonon pairing mechanisms are respons-ible for the high-Tc superconducting state11,12. Second, F-dopedLaFeAsO is a semimetal, which exhibits strong antiferromagneticfluctuations and a possible spin density wave instability around150 K in the parent undoped LaFeAsO (refs 5, 13–17). And last,superconductivity may emerge on several disconnected pieces ofthe Fermi surface11,12,18,19, thus exhibiting the multi-gap pairing thathas recently attracted so much attention in MgB2 (ref. 20).

Given the importance of magnetic correlations in the doped oxy-pnictides, transport measurements at very high magnetic fields arevital to probe the mechanisms of superconductivity. Indeed, firstmeasurements of the upper critical magnetic field Bc2(T) have yieldeda slope B0c2 5 dBc2/dT < 2 T K21 near Tc, for both La- and Sm-basedoxypnictides2–6. From the conventional one-band Werthamer-Helfand-Hohenberg (WHH) theory21, such slopes already implyrather high values of the upper critical magnetic field at zero tem-perature Bc2(0) 5 0.69Tc B0c2 < 36 T for LaFeAsO0.89F0.11, ,59 T forSmFeAsO0.9F0.1, and ,72 T for PrFeAsO0.89F0.11, all well above

Bc2(0) < 30 T of Nb3Sn. However, studies of the high-field supercon-ductivity in MgB2 alloys have shown that the upward curvature ofBc2(T) resulting from multiband effects can significantly increaseBc2(0), as compared to the WHH one-band extrapolation (see ref.22 and references therein). Our high-field d.c. transport measure-ments on LaFeAsO0.89F0.11 samples show that Bc2(T) indeed exhibitssigns of two-gap behaviour similar to that in MgB2 with a value ofBc2(0) that exceeds the WHH extrapolation by ,2 times and whichalso exceeds the BCS paramagnetic limit, Bp (in tesla) 5 1.84Tc (inkelvin) 5 51.5 T for Tc 5 28 K.

Polycrystalline LaFeAsO0.89F0.11 samples were made by solid statesynthesis4. A sample ,3 3 1 3 0.5 mm3 was used for our four-probetransport measurements in the 45 T hybrid magnet at the NationalHigh Magnetic Field Laboratory, supplemented by low-field mea-surements in a 9 T superconducting magnet. Our low-field data agreewell with earlier data taken at ORNL on the same sample4, indicatingits good temporal and atmospheric stability. The 45 T hybrid magnetwas swept only from 11.5 T to 45 T owing to the static 11.5 T back-ground field generated by the outer superconducting coil of themagnet combination, while lower fields were swept from 0 T to 9 Tin a Physical Property Measurement System (PPMS) superconduct-ing magnet with parameters shown in Fig. 1 legend.

The results of our high-field measurements of the sample resist-ance as a function of magnetic field strength, R(B), are shown inFig. 1. The broad R(B) transitions are not surprising because thesample consists of misoriented anisotropic crystalline grains. Giventhe predicted resistivity ratio C 5 rc/rab < 10–15 for this layeredcompound11, the local Bc2(h) < Bc2(0)[cos2h 1 C21sin2h]21/2

should vary strongly, depending on the angle h between the c-axisin the grain and the applied field. Thus, we can identify two char-acteristic fields: the high-field onset Bmax of the superconductingtransition, and the zero-resistance, low-field onset Bmin, as illustratedby Fig. 1. The in-field transitions are shown in Fig. 1a for B perpen-dicular to the broad 1-mm-wide face of the sample, and in Fig. 1b forB applied parallel to the same face (in both cases, transport currentwas always perpendicular to B). Differences in the R(B) curves ofFig. 1 do indicate some grain texture, as the zero-resistance field Bmin

in Fig. 1b (36 T at 4.2 K) is higher than in Fig. 1a (25 T at 4.2 K),suggesting that the plate-like grains tend to align their a–b planesparallel to the broad face of the sample.

Shown in Fig. 2a are the temperature dependences of the fieldsBmin, Bmid and Bmax (all in T) evaluated at 10%, 50% and 90% of thenormal state resistance at the transition temperature, Rn(Tc), respect-ively. The interpretation of Bmin and Bmax in polycrystals can becomplicated by strong vortex pinning and the particulars of the grainmisorientation distribution. However in our case, the analysis issimplified because: (1) the measured magnetization curves are nearly

1National High Magnetic Field Laboratory, Florida State University, Tallahassee, Florida 32310, USA. 2Materials Science & Technology Division, Oak Ridge National Laboratory, OakRidge, Tennessee 37831, USA.

Vol 453 | 12 June 2008 | doi:10.1038/nature07058

903

©2008 Macmillan Publishers Limited. All rights reserved

reversible, suggesting weak pinning; and (2) the crystalline aniso-tropy of the compound results in a broad distribution of the localBc2(h) values in different grains. Thus, Bmax(T) is associated with thelarger in-plane upper critical field B

Ec2(T) < B\

c2(T)C1/2 because grainswith their a–b planes oriented along the applied field become super-conducting first upon cooling. In turn, the zero-resistance fieldBmin(T) can be interpreted as the field below which those supercon-ducting grains with Bc2(h) . B form a percolative path at our lowmeasuring current density ,0.2 A cm22. For C? 1 and negligiblethermal activation of vortices, both the effective medium and thepercolation23 theories give Bmin(T) < B\

c2(T)/pc, where pc , 1 is atemperature-independent number determined by the percolationthreshold defined by the material anisotropy and the orientationalgrain distributions of the specific sample. Accordingly, our inter-pretation of Fig. 2 is that Bmax(T) and Bmin(T) reflect the temperaturedependences of B

Ec2(T) and B\

c2(T), respectively, in the case whenthermal activation of vortices is weak, as we now discuss.

The upward curvature of Bmin(T) in Fig. 2 might also be inter-preted as the signature of the irreversibility field Bm(T) resulting frommelting of a weakly pinned vortex lattice. Thermal vortex fluctuationeffects are quantified by the Ginzburg parameter Gi 5 (8p2kBTcLa

2/jcw0

2)2/2 (ref. 24), where La is the London penetration depth, ja isthe a–b plane coherence length, w0 is the magnetic flux quantum,jc 5 jaC

21/2 is the c-axis coherence length, and kB is the Boltzmannconstant. The copper oxide high-temperature superconductors havestrong thermal fluctuations resulting in Gi < 1–1022 andBm(T)=Bc2(T). The conventional low-Tc superconductors, inwhich vortex fluctuations are negligible, have Gi < 1026–10210 andBc2 2 Bm=Bc2. The coherence length jc 5 [w0/2pBc2(0)C1/2]1/2 canbe estimated for C 5 15 and B

Ec2 (0) 5 60 T in Fig. 2 as 1.2 nm, which

for Tc 5 26 K and La 5 215 nm, extracted from recent NMR mea-surements25, yields Gi 5 3.4 3 1024, a value close to Gi 5 2.1 3 1024

of clean MgB2, but a value some 30 times smaller than Gi for the leastanisotropic high-temperature superconductor, YBa2Cu3O72x, lead-

0

1

2

3

0 10 20 30 40 50

15

10 6

4.2

B (T)

R (m

Ω)

T (K) = 25

20

17

90%

50%

10%

B ⊥ broad sample face

a

0

1

2

3

10 20 30 40 50

2.5

10

7

4.2

B (T)

90%

50%

10%

T (K) = 15B || broad sample face

b

Figure 1 | Variation of LaFeAsO0.89F0.11 sample resistance with appliedmagnetic field at fixed temperatures in the range 4.2 K to 25 K. a, Theresistance R(B) for different temperatures taken in swept fields in the 45 Thybrid magnet at a measuring current of 1 mA and from 0 to 9 T in asuperconducting magnet at a measuring current of 5 mA for B appliedperpendicular to the broad face of the sample. The current density was,0.2 A cm22 in the hybrid magnet measurements and 5 times higher in the

PPMS superconducting magnet measurements. b, R(B) data taken in thehybrid magnet for B parallel to the broad sample face, which we believe has ahigher fraction of a–b-oriented grains. The normal state resistivity at 30 K isestimated to be 0.15 mV cm with an uncertainty of ,15%. The horizontaldashed lines indicate 10%, 50% and 90% of the resistive transition relative tothe normal state resistance at the transition temperature, Rn(Tc),respectively.

0

10

20

30

40

50

0 5 10 15 20 25 30T (K)

90%50%

10%

WHH

a

0 0.2 0.4 0.6 0.8 10

10

20

30

40

50

60

70

T/Tc

Bc2

(T

)

b

Figure 2 | Upper critical field–temperature phase diagram ofLaFeAsO0.89F0.11. a, The measured fields Bmax(T) and Bmin(T) (black andred squares, respectively) along with the midpoint transition fields (bluecircles). The filled and open symbols correspond respectively to the paralleland perpendicular field orientations. The dotted line shows the WHH curvedefined by the slope of Bmin(T) at Tc. b, Bmax(T) (black squares) and Bmin(T)

(red squares) plotted as functions of the reduced temperature T/Tc. The datapoints above 45 T were extracted by linear extrapolation of R(B) at B , 45 Tto R(B) 5 0.9Rn(Tc), as shown by dashed lines in Fig. 1. The lines correspondto Bc2(T) calculated from the two-gap theory for the parameters described inthe text.

LETTERS NATURE | Vol 453 | 12 June 2008

904

©2008 Macmillan Publishers Limited. All rights reserved

ing us to classify LaFeAsO0.89F0.11 as an ‘intermediate-Tc’ supercon-ductor. We estimate Bm based on a theory of vortex fluctuations inmoderately anisotropic superconductors24 (see SupplementaryInformation), which shows that Bc2(T) and Bm(T) at T , 0.5Tc differonly by ,20%. Therefore, we conclude that the temperature depend-ence of the resistive transition field Bmin(T) at T , 0.5Tc reflects thebehaviour of B\

c2(T) rather than that of the melting field Bm(T).As is evident from Fig. 2, B\

c2(T) exhibits a significant upwardcurvature, which is much less pronounced for BE

c2(T). Because thisbehaviour is similar to that observed in dirty MgB2 films22, we sug-gest, in agreement with the ab initio calculations11,12, that supercon-ductivity in LaFeAsO0.89F0.11 results from two bands: a nearly two-dimensional electron band with high in-plane diffusivity D1 and amore isotropic heavy hole band with smaller diffusivity D2. Theupward curvature of Bc2(T) is then controlled by the ratio g 5 D2/D1: for g= 1, the upward curvature is pronounced, while for g < 1,Bc2(T) exhibits a more traditional WHH-like behaviour21. For fields

along the a–b plane, the parameter g should be replaced with g 5 D2/

[ D(ab)1 D

(c)1 ]1/2, allowing strong anisotropy D

(ab)1 ?D

(c)1 to significantly

increase g for Bjjab as compared to Bjjc (ref. 22). In this case theupward curvature of B

Ec2(T) does become less pronounced than for

B\c2(T), in agreement with the data in Fig. 2.To check the self-consistency of our interpretation, we fit B\

c2 (T) inFig. 2, using the two-gap theory outlined in SupplementaryInformation. We took g 5 D2/D1 5 0.08 and the interband BCScoupling constants l12 5 l21 5 0.5, but the results are relativelyinsensitive to the choice of either l12 and l21 (including their sign)or the intraband coupling constants l11 and l22. For example, Fig. 2shows a rather good fit for the case of strong interband repulsionl12 l21 ? l11 l22 suggested in ref. 12. However, this theory also showsthat the fit remains nearly as good, even if we assume that intrabandpairing is significant, l11l22 . l12l21, with all coupling constantsbeing of the same order of magnitude. Thus, our experimental datado not yet enable us to unambiguously distinguish between differentpairing scenarios suggested in the literature, but they do indicate asignificant difference in the effective masses in the electron and holebands. For Bjjab, we rescaled the parameter g R gC1/2 < 0.31, takingthe estimate C 5 rc/rab < 15 suggested in ref. 11, which describesB

Ec2(T) well, as is also evident from Fig. 2. Therefore the two-gap

scenario is qualitatively consistent with our experimental data.The newly discovered LaFeAsO0.89F0.11 exhibits exceptionally high

Bc2, and this obviously non-optimized material has been quicklysynthesized in bulk form showing zero resistance above 35 T at2.5 K. Moreover, given the high dBc2/dT values of 2–3 T K21, whichhave also been observed in the Sm-based (Tc 5 43 K; ref. 6) and Pr- orNd-based oxypnictides (Tc 5 52 K), it seems reasonable to expectBE

c2(T) and B\c2(T) values 1.5–2 times higher than LaFeAsO0.89F0.11,

which puts BEc2(0) above 100 T. Thus doped oxypnictides appear as a

new family of high-field superconductors, for which extensivepulsed-field measurements in the 100 T range will be required to fullyreveal the novel physics of competing superconducting and magneticorders. It is tempting to think that they may also have great import-ance for high-field applications.

Received 2 April; accepted 5 May 2008.Published online 28 May 2008.

1. Kamihara, Y., Watanabe, T., Hirano, M. & Hosono, H. Iron-based layeredsuperconductor La[O1-xFx]FeAs (x 5 0.05-0.12) with Tc 5 26 K. J. Am. Chem. Soc.130, 3296–3297 (2008).

2. Zhu, X., Yang, H., Fang, L., Mu, G. & Wen, H.-H. Upper critical field, Hall effect andmagnetoresistance in the iron-based layered superconductor LaO0.9F0.1-dFeAs.Preprint at Æhttp://arxiv.org/abs/0803.1288v1æ (2008).

3. Chen, G. F. et al. Superconducting properties of Fe-based layered superconductorLaO0.9F0.1-d FeAs. Preprint at Æhttp://arxiv.org/abs/0803.0128v1æ (2008).

4. Sefat, A. S. et al. Electron correlations in the superconductor LaFeAsO0.89F0.11 withlow carrier density. Phys. Rev. B 77, 174503 (2008).

5. Chen, G. F. et al. Superconductivity at 41K and its competition with spin-density-wave instability in layered CeO1-xFx FeAs. Preprint at Æhttp://arxiv.org/abs/0803.3790v2æ (2008).

6. Chen, X. H. et al. Superconductivity at 43 K in samarium-arsenide oxidesSmFeAsO1-xFx. Preprint at Æhttp://arxiv.org/abs/0803.3603v1æ (2008).

7. Ren, Z.-A. et al. Superconductivity in iron-based F-doped layered quaternarycompound Nd[O1-xFx]FeAs. Preprint at Æhttp://arxiv.org/abs/0803.4234æ (2008).

8. Ren, Z.-A. et al. Superconductivity at 52K in iron-based F-doped layeredquaternary compound Pr[O1-xFx]FeAs. Preprint at Æhttp://arxiv.org/abs/0803.4283æ (2008).

9. Zimmer, B. I., Jeitschko, W., Albering, J. H., Glaum, R. & Reehuis, M. The rare earthtransition metal phosphide oxides LnFePO, LnRuPO and LnCoPO with ZrCuSiAstype structure. J. Alloy. Comp. 229, 238–242 (1995).

10. Quebe, P., Terbuchte, L. J. & Jeitschko, W. Quaternary rare earth transition metalarsenide oxides RTAsO (T5Fe, Ru, Co) with ZrCuSiAs type structure. J. Alloy.Comp. 302, 70–74 (2000).

11. Singh, D. J. & Du, M. H. LaFeAsO1-xFx: A low carrier density superconductor nearitinerant magnetism. Preprint at Æhttp://arxiv.org/abs/0803.0429v1æ (2008).

12. Mazin, I. I., Singh, D. H., Johannes, M. D. & Du, M. H. Unconventional sign-reversalsuperconductivity in LaFeAsO1-xFx. Preprint at Æhttp://arxiv.org/abs/0803.2740v1æ (2008).

13. Cao, C., Hirschfeld, P. J. & Cheng, H.-P. Coexistence of antiferromagnetism withsuperconductivity in LaO1-xFxFeAs: effective Hamiltonian from ab initio studies.Preprint at Æhttp://arxiv.org/abs/0803.3236æ (2008).

14. Dai, X., Fang, Z., Zhou, Y. & Zhang, F. Even parity, orbital singlet and spin tripletpairing for superconducting La(O1-xFx)FeAs. Preprint at Æhttp://arxiv.org/abs/0803.3982v1æ (2008).

15. Dong, J. et al. Competing orders and spin-density wave instability inLa(O1-xFx)FeAs. Preprint at Æhttp://arxiv.org/abs/0803.3426v1æ (2008).

16. Ma, F. & Lu, Z.-Y. Iron-based superconductor LaO1-xFxFeAs superconductor: anantiferromagnetic semimetal. Preprint at Æhttp://arxiv.org/abs/0803.3286v1æ(2008).

17. McGuire, M. A. et al. Evidence for the spin density wave in LaFeAsO. Preprint atÆhttp://arxiv.org/abs/0804.0796æ (2008).

18. Kuroki, K. et al. Unconventional superconductivity originating from disconnectedFermi surfaces in LaO1-xFxFeAs. Preprint at Æhttp://arxiv.org/abs/0803.3325v1æ(2008).

19. Lebegue, S. Electronic structure and properties of the Fermi surface of thesuperconductor LaOFeP. Phys. Rev. B 75, 035110 (2007).

20. Nagamatsu, J., Nakagawa, N., Muranaka, T., Zenitani, Y. & Akimitsu, J.Superconductivity at 39 K in magnesium diboride. Nature 410, 63–64 (2001).

21. Werthamer, N. R., Helfand, E. & Hohenberg, P. C. Temperature and puritydependence of the superconducting critical field, Hc2. III. Electron spin and spin-orbit effects. Phys. Rev. 147, 295–302 (1966).

22. Gurevich, A. Limits of the upper critical field in dirty two-gap superconductors.Physica C 456, 160–169 (2007); Enhancement of the upper critical field bynonmagnetic impurities in dirty two-gap superconductors. Phys. Rev. B 67, 184515(2003).

23. Glazman, L. I., Koshelev, A. I. & Lebed’, A. G. Resistive transition and critical fieldsof superconducting ceramics. Sov. Phys. JETP 67, 1235–1241 (1988).

24. Blatter, G. & Geshkenbein, V. in The Physics of Superconductors Vol. 1 (edsBennerman, K. H. & Ketterson, J. B.) 800 (Springer, Berlin, 2003).

25. Ahilan, K. et al. 19F NMR investigation of LaAsFeO0.89F0.11 superconductor.Preprint at Æhttp://arxiv.org/abs/0804.4026æ (2008).

Supplementary Information is linked to the online version of the paper atwww.nature.com/nature.

Acknowledgements Work at the NHMFL was supported by IHRP under NSFCooperative Agreement, by the State of Florida, by the DOE, by the NSF FocusedResearch Group on Magnesium Diboride (FRG), and by AFOSR. Work at ORNLwas supported by the Division of Materials Sciences and Engineering, Office ofBasic Energy Sciences. We are grateful for discussions with G. Boebinger,E. Hellstrom, P. Lee, J. Jiang, and C. Tarantini at the NHMFL.

Author Information Reprints and permissions information is available atwww.nature.com/reprints. Correspondence and requests for materials should beaddressed to F.H. (hunte@asc.magnet.fsu.edu).

NATURE | Vol 453 | 12 June 2008 LETTERS

905

©2008 Macmillan Publishers Limited. All rights reserved