Microscopic co-existence of superconductivity and magnetism in Ba{}_{1-x}K{}_{x}Fe{}_{2}As{}_{2}

Microscopic co-existence of superconductivity and magnetism in Ba1-xKxFe2As2


It is widely believed that, in contrast to its electron doped counterparts, the hole doped compound BaKFeAs exhibits a mesoscopic phase separation of magnetism and superconductivity in the underdoped region of the phase diagram. Here, we report a combined high-resolution x-ray powder diffraction and volume sensitive muon spin rotation study of underdoped BaKFeAs ( showing that this paradigm is wrong. Instead we find a microscopic coexistence of the two forms of order. A competition of magnetism and superconductivity is evident from a significant reduction of the magnetic moment and a concomitant decrease of the magneto-elastically coupled orthorhombic lattice distortion below the superconducting phase transition.

74.70.Xa, 74.62.Dh, 74.62.En, 61.05.C-

The interplay of structural, magnetic and superconducting order parameters is one of the most intriguing aspects in iron based superconductors. In the LaFeAsO (1111) and BaFeAs (122) families, superconductivity (SC) evolves from non-superconducting parent compounds with tetragonal crystal structures that are subject to tiny orthorhombic lattice distortions below certain temperatures (). Static long-range antiferromagnetic (AF) ordering emerges at Néel temperatures () well below in LaFeAsO,de la Cruz et al. (2008) but very close to in BaFeAs.Rotter et al. (2008a) The structural and magnetic transitions of the parent compounds are suppressed and finally eliminated by doping of the FeAs layers by electrons or holes, and superconductivity emerges at certain doping levels.Zinth et al. (2011) With respect to the origin of unconventional superconductivity, the possible coexistence of magnetic and superconducting phases in the underdoped areas of the phase diagrams is of considerable interest. But the coupling of structural, magnetic and superconducting order parameters relies on microscopic phase coexistence that is often difficult to distinguish from mesoscopic phase separation. In the 122-family, microscopic co-existence of these orders is generally accepted for the electron-doped compounds Ba(FeCo)As, while conflicting reports exist for the hole doped compounds BaKFeAs.

Co-existence of the orthorhombic structure with SC has first been suggested for BaKFeAs up to 0.2 by x-ray powder diffraction,Rotter et al. (2008b) while neutron diffraction experiments additionally showed long-range AF ordering up to 0.3. Chen et al. (2009) Diffraction methods however only provide the mean structural information on a rather long spatial scale, and cannot supply conclusive information regarding phase separation. Fe-Möössbauer spectroscopy as a local probe indicated microscopic co-existence,Rotter et al. (2009) but other local probes such as SRAczel et al. (2008); Goko et al. (2009); Park et al. (2009) and NMRJulien et al. (2009) showed phase separation with non-magnetic superconducting volume fractions between 25 and 40%. However, most of these experiments were conducted with almost optimally doped BaKFeAs single crystals, despite still no growth method yields homogeneous crystals of this compound. Nevertheless, these studies manifested the paradigm that underdoped BaKFeAs exhibits mesoscopic phase separation.

In contrast to these scattered results, studies with cobalt-doped Ba(FeCo)As yielded convincing evidence for microscopic co-existence.Pratt et al. (2009) Moreover, competing order parameters became obvious by the concomitant reduction of the orthorhombic lattice distortion and magnetic moment when crossing the critical temperature.Nandi et al. (2010) This microscopic co-existence supports symmetry of the superconducting order parameterFernandes et al. (2010); Fernandes and Schmalian (2010) and gives strong evidence for unconventional superconductivity in iron arsenides.

Considering this generally accepted situation for Ba(FeCo)As, it is particularly important to clarify the intrinsic behavior of BaKFeAs, also because cobalt-doping causes additional disorder in the (FeCo)-layers, while potassium-doping hardly affects the FeAs-layers. Thus, if both orders co-exist microscopically in BaKFeAs, we rather observe the behavior of the clean superconducting FeAs-layer. Indeed, a recent neutron diffraction study with polycrystalline material supports early suggestions about microscopic co-existence Avci et al. (2011), but gives no conclusive proof regarding microscopic co-existence, because elastic neutron scattering as a bulk probe is principally unable to distinct whether the magnetic volume fraction or the magnetic moment at the iron site decreases.

In this letter, we report a combined high-resolution x-ray diffraction and muon spin rotation (SR) study with underdoped BaKFeAs (=0, 0.19, 0.23, 0.25). We unambiguously show the homogeneous co-existence of the superconducting and antiferromagnetic phase and the competition of the respective order parameters.

Polycrystalline samples of BaKFeAs were synthesized by heating stoichiometric mixtures of the elements (purities ) in alumina crucibles sealed in silica tubes under purified argon. In order to avoid potassium evaporation, alumina inlays were used. The mixtures were heated to 873 K (50 K/h) and kept at this temperature for 15 h. The products were homogenized in the crucible, enclosed in a silica tube and annealed at 923 K for 15 h. Finally the powders were cold pressed into pellets, sintered for 20 h at 1023 K and cooled to room temperature by switching off the furnaces. Lattice parameters were obtained by temperature-dependent x-ray powder diffraction (Co,Cu,Mo--radiation) and Rietveld refinements using the TOPAS package Cheary et al. (2004). Fig. 1 a) shows a typical XRD pattern. Only traces of impurity phases were detected. Ba:K ratios was examined by refinement of the site occupancy parameters and cross checked by ICP-AAS chemical analysis. SR measurements have been performed using the GPS and Dolly spectrometers located at the M3 and E1 beamlines of the Swiss Muon Source at the Paul Scherrer Institut, Switzerland. The data have been analyzed using the MUSRFIT package Suter and Wojek (2011).

Figure 1: a) X-ray powder diffraction pattern (blue) and Rietveld fit (red) of BaKFeAs (). b) Temperature dependence of the (112) reflections of for  0.19, 0.23, and 0.25. c) a- and b-axis cell parameters as a function of temperature.

X-ray powder patterns of the samples revealed the known structural phase transitions from tetragonal to orthorhombic symmetry. In agreement with our earlier studies Rotter et al. (2008b), also Ref.Avci et al. (2011) showed that the orthorhombic distortion depends on the potassium concentration and is finally absent if 0.3. Fig.1 b) shows the temperature dependency of the (112) reflections. While the clear splitting, or at least broadening of the peak is visible at = 0.19 and 0.23, it is apparently absent at = 0.25. However, a closer inspection reveals the onset of peak broadening below 70 K also in this case. From this we obtained the tetragonal to orthorhombic transition temperatures  K, 98 K, 84 K and 70 K for , 0.19, 0.23, and 0.25 respectively. The lattice parameters obtained from Rietveld-refinements are shown in Fig.1 c). It is obvious that potassium doping of BaFeAs reduces the transition temperature and also the extent of the lattice parameter splitting, which is still visible at = 0.25 where is already 32.6 K.

The main goal of this study is to clarify how magnetism and superconductivity coexist in the underdoped region of the BaKFeAs phase diagram. For this reason x-ray, AC-susceptibility, and SR measurements have been performed on the very same samples to investigate their superconducting and magnetic properties, respectively.

The AC-susceptibility of finely ground powder samples were measured between 3.4 K and 45  K at 8 Oe and 1.333 kHz. Diamagnetic signals were detected below  K (= 0.19), 28.5 K ( = 0.23) and 32.6 K ( = 0.25) as shown in Fig. 2 a). The superconducting volume fractions of all samples are close to 100% and prove bulk superconductivity.

Figure 2: a) Magnetic susceptibility measurement of BaKFeAs with  = 0.19, 0.23, and 0.25 showing 100% diamagnetic shielding. b) Structural, magnetic and superconducting phase diagram of BaKFeAs. c) Magnetic volume fraction as a function of temperature obtained from wTF-SR measurements.

Muon spin rotation measurements in a weak transverse field (wTF-SR) provide an easy means to measure the magnetic volume fraction. In Fig. 2 c) the magnetic volume fractions obtained by such measurements in  Oe are shown for various BaKFeAs samples (, 0.19 and 0.23) as a function of temperature. For all samples a transition to a magnetic state is observed. From this the magnetic transition temperature where 50% of the volume is magnetic has been determined to  K, 97 K and 83 K for the three samples respectively. The magnetic volume fraction reaches 100% for all three samples and, most remarkable, does not change below the superconducting . Therefore, these results, together with the 100% superconducting shielding signal observed in the AC-susceptibility measurements, prove the microscopic coexistence of magnetism and superconductivity in the orthorhombic phase of our samples. The structural, magnetic and superconducting transition temperatures are compiled in the phase diagram depicted in Fig. 2 b).

The orthorhombic distortion in terms of the structural order parameter is shown in Fig. 3 a). In the = 0.19 sample, achieves a clear maximum at the superconducting transition temperature close to 23 K and then decreases to tower temperatures. Higher potassium concentrations further decrease to 84 K while again coincides with at 28.5 K ( = 0.23). This trend continues to = 0.25 with , = 32.6 K and . This behavior is similar to Ba(FeCo)As,Nandi et al. (2010) however, we do not observe the further linear decrease of at lower temperatures back to a quasi-tetragonal structure, but rather saturation of . Also in contrast to the Co-doped material, we find that the effect becomes smaller with increasing potassium concentrations . The reason for that is not yet clear. We suggest that the stronger effect in the case of Co-doping may be connected with the fact, that magnetic ordering is weakened not only by the electron-doping, but additionally by the disorder that is introduced by the cobalt-atoms at the iron sites. Thus the competition of superconductivity and antiferromagnetism for the same electrons may affect the (FeCo)As layers more efficiently than the clean FeAs-layers in the K-doped material.

To elucidate further the magnetic properties of BaKFeAs  zero field (ZF) SR measurements have been performed. The ZF-SR spectra shown in Fig. 4 exhibit well defined muon spin precessions at below even for the sample indicating a long-range ordered magnetic phase.

As already observed in other Fe-based superconductors Klauss et al. (2008); Jesche et al. (2008); Maeter et al. (2009); Goko et al. (2009) the ZF spectra are composed of two distinct precession frequencies which has been interpreted as two magnetically inequivalent muon stopping sites in the structure. The data can be well fitted with a damped cosine functional form indicating a commensurate magnetic structure.Yaouanc and de Réotier (2011) Please note, that the in under-doped samples of the related Ba(FeCo)As  family only a strongly over-damped oscillation can be observed Bernhard et al. (2009); Marsik et al. (2010). This indicates that the doping in the Ba layer causes considerably less disorder into the magnetic system. Another difference is that in the Co-doped systems SR spectra consistent with incommensurate order have been found. ZF-SR allows to precisely determine the temperature dependence of the magnetic order parameter (Fe moment) which is proportional to the measured SR frequency. The observed SR frequency is shown in Fig. 3 b) together with the orthorhombicity parameter deduced from the XRD measurements.

Figure 3: Orthorhombicity parameter and ZF-SR frequency (magnetic order parameter) of BaKFeAs as a function of temperature.
Figure 4: Zero field SR spectra for characteristic temperatures (above, at, and below the transition) for BaKFeAs powder samples with  = 0.0, 0.19, and 0.23.

The magnetic order parameter shown in Fig. 3 b) decreases alike the orthorhombicity as a function of potassium doping. Most remarkable however is the decrease of the magnetic order parameter (Fe moment) below the superconducting clearly visible in the inset of Fig. 3 b). Here we would like to mention that SR as a local probe is able to measure the magnetic volume fraction (as shown above) and the size of the ordered moment (via the ZF-SR frequency) separately unlike it is done in scattering experiments where the product of both quantities is measured. Taking all data together it is obvious that all investigated samples remain 100% magnetic, but that the ordered Fe magnetic moment as well as the orthorhombicity decrease below the superconducting . In other words, superconductivity and magnetism coexist on a microscopic scale, but compete for the same electrons in the underdoped region of the BaKFeAs phase diagram.

In summary, our results prove the paradigm of phase-separation in BaKFeAs wrong. Instead we find compelling evidence of microscopic co-existence of superconductivity with magnetic ordering from combined x-ray and SR data. The competition for the same electrons reduces the magnetic moment below , while the magnetic fraction remains 100 % according to volume-sensitive SR measurements. The response of the structural and magnetic order parameters at is weaker than in Co-doped Ba(FeCo)As. Since K-doping introduces no disorder in the superconducting FeAs-layer, we suggest that we rather observe its intrinsic behavior.

Thanks to Marianne Rotter for the low temperature X-ray diffraction measurements and to Marcus Tegel for support with Rietveld refinements. This work was financially supported by the German Research Foundation (DFG) within the priority program SPP1458, project No. JO257/6-1. Part of this work has been performed at the Swiss Muon Source at the Paul Scherrer Institut, Switzerland.


  1. C. de la Cruz, Q. Huang, J. W. Lynn, J. Li, W.-R. Ratcliff-II, J. L. Zarestky, H. A. Mook, G. F. Chen, J. L. Luo, N. L. Wang, et al., Nature 453, 899 (2008).
  2. M. Rotter, M. Tegel, I. Schellenberg, W. Hermes, R. Pöttgen, and D. Johrendt, Phys. Rev. B 78, 020503 (2008a).
  3. V. Zinth, T. Dellmann, H.-H. Klauss, and D. Johrendt, Angew. Chem. Int. Ed. 50, 7919 (2011).
  4. M. Rotter, M. Pangerl, M. Tegel, and D. Johrendt, Angew. Chem. Int. Ed. 47, 7949 (2008b).
  5. H. Chen, Y. Ren, Y. Qiu, W. Bao, R. H. Liu, G. Wu, T. Wu, Y. L. Xie, X. F. Wang, Q. Huang, et al., Europhys. Lett. 85, 17006 (2009).
  6. M. Rotter, M. Tegel, I. Schellenberg, F. M. Schappacher, R. Pöttgen, J. Deisenhofer, A. Gunther, F. Schrettle, A. Loidl, and D. Johrendt, New J. Phys. 11, 025014 (2009).
  7. A. A. Aczel, E. Baggio-Saitovitch, S. L. Budko, P. C. Canfield, J. P. Carlo, G. F. Chen, P. Dai, T. Goko, W. Z. Hu, G. M. Luke, et al., Phys. Rev. B 78, 214503 (2008).
  8. T. Goko, A. A. Aczel, E. Baggio-Saitovitch, S. L. Bud’ko, P. C. Canfield, J. P. Carlo, G. F. Chen, P. Dai, A. C. Hamann, W. Z. Hu, et al., Phys. Rev. B 80, 024508 (2009).
  9. J. T. Park, D. S. Inosov, C. Niedermayer, G. L. Sun, D. Haug, N. B. Christensen, R. Dinnebier, A. V. Boris, A. J. Drew, L. Schulz, et al., Phys. Rev. Lett. 102, 117006 (2009).
  10. M. H. Julien, H. Mayaffre, M. Horvatic, C. Berthier, X. D. Zhang, W. Wu, G. F. Chen, N. L. Wang, and J. L. Luo, Europhys. Lett. 87, 37001 (2009).
  11. D. K. Pratt, W. Tian, A. Kreyssig, J. L. Zarestky, S. Nandi, N. Ni, S. L. Bud’ko, P. C. Canfield, A. I. Goldman, and R. J. McQueeney, Phys. Rev. Lett. 103, 087001 (2009).
  12. S. Nandi, M. G. Kim, A. Kreyssig, R. M. Fernandes, D. K. Pratt, A. Thaler, N. Ni, S. L. Budko, P. C. Canfield, J. Schmalian, et al., Phys. Rev. Lett. 104, 057006 (2010).
  13. R. M. Fernandes, D. K. Pratt, W. Tian, J. Zarestky, A. Kreyssig, S. Nandi, M. G. Kim, A. Thaler, N. Ni, P. C. Canfield, et al., Phys. Rev. B 81, 140501 (2010).
  14. R. M. Fernandes and J. Schmalian, Phys. Rev. B 82, 014521 (2010).
  15. S. Avci, O. Chmaissem, E. A. Goremychkin, S. Rosenkranz, J. P. Castellan, D. Y. Chung, I. S. Todorov, J. A. Schlueter, H. Claus, M. G. Kanatzidis, et al., Phys. Rev. B 83, 172503 (2011).
  16. R. W. Cheary, A. A. Coelho, and J. P. Cline, J. Res. Natl. Inst. Stand. Technol. 109, 1 (2004).
  17. A. Suter and B. M. Wojek, Physics Procedia, accepted (2011).
  18. H.-H. Klauss, H. Luetkens, R. Klingeler, C. Hess, F. Litterst, M. Kraken, M. M. Korshunov, I. Eremin, S.-L. Drechsler, R. Khasanov, et al., Phys. Rev. Lett. 101, 077005 (2008).
  19. A. Jesche, N. Caroca-Canales, H. Rosner, H. Borrmann, A. Ormeci, D. Kasinathan, H. H. Klauss, H. Luetkens, R. Khasanov, A. Amato, et al., Phys. Rev. B 78, 180504 (2008).
  20. H. Maeter, H. Luetkens, Y. G. Pashkevich, A. Kwadrin, R. Khasanov, A. Amato, A. A. Gusev, K. V. Lamonova, D. A. Chervinskii, R. Klingeler, et al., Phys. Rev. B 80, 094524 (2009).
  21. A. Yaouanc and P. D. de Réotier, MUON SPIN ROTATION, RELAXATION and RESONANCE (Oxford University Press, 2011).
  22. C. Bernhard, A. J. Drew, L. Schulz, V. K. Malik, M. Rössle, C. Niedermayer, T. Wolf, G. D. Varma, G. Mu, H.-H. Wen, et al., New Journal of Physics 11, 055050 (2009).
  23. P. Marsik, K. W. Kim, A. Dubroka, M. Rössle, V. K. Malik, L. Schulz, C. N. Wang, C. Niedermayer, A. J. Drew, M. Willis, et al., Phys. Rev. Lett. 105, 057001 (2010).
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
Add comment
Loading ...
This is a comment super asjknd jkasnjk adsnkj
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test description