Temperature-concentration phase diagram of (Ca1-xLax)10(Pt3As8)(Fe2As2)5 superconductors
N. Ni, W. E. Straszheim, D. J. Williams, M. A. Tanatar, R. Prozorov, E. D. Bauer, F. Ronning, J. D. Thompson, R. J. Cava
aa r X i v : . [ c ond - m a t . s up r- c on ] D ec Temperature-concentration phase diagram of (Ca − x La x ) (Pt As )(Fe As ) superconductors N. Ni , , W. E. Straszheim , D. J. Williams , M. A. Tanatar , R.Prozorov , E. D. Bauer , F. Ronning , J. D. Thompson and R. J. Cava Los Alamos National Laboratory,Los Alamos, NM 87544, USA Department of Chemistry, Princeton University,Princeton, NJ 08544, USA Ames Laboratory and Department of Physics and Astronomy,Iowa State University, Ames, Iowa 50011, USA (Dated: August 8, 2018)Single crystals of (Ca − x La x ) (Pt As )(Fe As ) ( x = 0 to 0.182) superconductors have beengrown and characterized by X-ray, microprobe, transport and thermodynamic measurements.Features in the magnetic susceptibility, specific heat and two kinks in the derivative of the electricalresistivity around 100 K in the x = 0 compound support the existence of decoupled structuraland magnetic phase transitions. With La doping, the structural/magnetic phase transitions aresuppressed and a half-dome of superconductivity with a maximal T c around 26 K is observed in thetemperature-concentration phase diagram. PACS numbers: 74.70.Xa, 74.25.DW, 74.25.Bt, 74.25.F-
The report of superconductivity at 26 K inLaFeAsO . F . has led to the discovery of severalfamilies of high T c iron arsenide superconductors,including the so-called 1111, 122, 111 and 42622families . The intense study of these families hasenriched our understanding of the interplay amongstructure, magnetism and superconductivity. Recentlya new iron-arsenide compound, Ca (Pt As )(Fe As ) (the so called 10-3-8 compound), has beencharacterized . This compound crystallizes in atriclinic structure with space group P -1 and has-Ca-(Pt As )-Ca-(Fe As )- layer stacking, as shownin the left inset of Fig. 1. The FeAs layer is madeof edge-sharing FeAs tetrahedra, the key structuralelement in all the Fe-pnictide superconductors. Astructural phase transition around 100 K has beenrevealed in this compound using polarized lightimaging . Although the susceptibility drop observed toaccompany the long range antiferromagnetic ordering inthe 1111 and 122 families has not been reported for the10-3-8 parent compound , recent NMR measurementsshow that this compound orders antiferromagnetically(AFM) below ∼
100 K . With Pt substitution onthe Fe sites, superconductivity up to 12 K has beenrealized . In the 1111 and 122 families, doping on theintermediary layer results in a higher T c than doping onthe FeAs layers, and thus higher T c may be expectedfor intermediary layer doping in the 10-3-8 compoundas well. Indeed, 20% La doping on the Ca sites inthis compound was found to show a T c of 30 K . Inthis paper, we report the systematic characterizationof (Ca − x La x ) (Pt As )(Fe As ) single crystalsvia X-ray diffraction, microprobe, transport andthermodynamic measurements. Due to improved qualityof the single crystals, we are able to observe a resistivity jump, a susceptibility drop, and a specific heat jump inthe parent 10-3-8 compound, supporting the existence ofboth structural and magnetic phase transitions. A T − x phase diagram for (Ca − x La x ) (Pt As )(Fe As ) ispresented.
20 30 40 50 6001 ( )( )( )( )( ) (Ca La x ) (Pt As )(Fe As ) x=0 x=0.182 I n t en s i t y ( a r b . un i t ) θ (degree) d F e A s ( Å ) x FIG. 1: The (00 l ) diffraction pattern of(Ca − x La x ) (Pt As )(Fe As ) ( x = 0 , . (Pt As )(Fe As ) .Right inset: The interlayer FeAs distance vs. doping level x . Plate-like millimeter-sized single crystals weresuccessfully grown from a CaAs-rich flux . CaAs, FeAsand LaAs precursors were made using the solid statereaction method. These precursors and Pt powder weremixed thoroughly according to the nominal ratios listedin Table I. The mixture was pressed into a pellet, putin an Al O crucible and sealed into a quartz tubeunder vacuum. The resulting ampules were heated up to TABLE I: The nominal Ca:La:Fe:Pt:As ratio in the crystalgrowth. WDS measured doping level x .(Ca − x La x ) (Pt As )(Fe As ) nominal Ca 3.5 3.45 3.4 3.35 3.3 4.1 4.2nominal La 0 0.05 0.1 0.15 0.2 0.4 0.8nominal Fe 2 2 2 2 2 2 2nominal Pt 0.4 0.4 0.4 0.4 0.4 0.4 0.4nominal As 5.5 5.5 5.5 5.5 5.5 6.5 7 x o C , held for 96 hours, slowly cooled down to 885 o C and then quenched. After rinsing off the flux usingdistilled water, single crystals were obtained. Thesesingle crystals show a layered growth habit and areeasily exfoliated and bent. In each batch, small crystals,with thickness less than 0.03 mm, were employed inthe transport measurements. The La concentration x was obtained via wavelength dispersive spectroscopy(WDS) using the electron probe microanalyzer ofa JEOL JXA-8200 electron-microprobe. WDS wasperformed on the measured transport samples toprovide a reliable determination of the electronic phasediagram as a function of composition. The resultsof the WDS measurements are summarized in TableI. These measurements directly indicate that La hasbeen successfully doped on the Ca sites, while the Ptsubstitution on the FeAs layer is well controlled: itis 0 for the x =0 compound, 0.007 for the x =0.021,0.043, 0.065, and 0.145 compounds and 0.02 for the x =0.093 and 0.182 compounds. In this paper, x refers to the WDS value. Transport and specificheat measurements were performed in a QuantumDesign (QD) Physical Properties Measurement System.Magnetic properties were measured in a QD MagneticProperties Measurement System. To easily comparethe physical properties of these superconductors withother iron arsenide superconductors, the units ofmolar susceptibility, magnetization, and heat capacitypresented are normalized to per mole-Fe .X-ray diffraction was performed on a ScintagX Advances Diffraction System employing Cu K α ( λ = 1 . A ) radiation. No FeAs, PtAs orother impurities were observed in any X-ray pattern.Figure 1 shows the (00 L ) diffraction pattern of(Ca − x La x ) (Pt As )(Fe As ) ( x = 0 , . the interlayer distance of the FeAs layerswere obtained. The right inset of Fig. 1 shows theevolution of the interlayer FeAs distance with x . Thisdistance increases monotonically with La doping from10.313(2) ˚ A in the parent compound to 10.368(2) ˚ A inthe x =0.182 compound, providing further evidence thatLa is incorporated into the structure.The physical properties of the parent compound are
50 100 150 200 250 3000.981.001.02
120 50 100 150 200 250 300050100150 ( - e m u / m o l e ) H // abH = 4 T(b)
T (K) C p ( J / m o l e - K ) (c) C p / T ( J / m o l e F e - K ) T (K ) T H = 0T d / d T ( - m - c m / K ) T H = 9TT min ( m - c m ) (a) H=0T FIG. 2: The physical properties of Ca (Pt As )(Fe As ) .(a) Electrical resistivity ρ (T) taken at H=0T. Inset: d ρ /d T vs. T taken at 0 T and 9 T; T =103 K and T =95 K. (b) χ (T) taken at 4 T with H//ab . Inset: The χ (T) from 50 to300 K. (c) C p vs. T . Inset: C p /T vs. T summarized in Fig. 2. Figure 2 (a) shows thetemperature dependence of the resistivity. The resistivityis 0.57 m Ω − cm at 300 K, which is almost twice asthat of BaFe As . ρ (T) shows a resistivity minimumat T min . T min is sample dependent, varying fromlarger than 300 K to 170 K, with an average of 210K. It is unclear wether T min comes from disorder orother mechanisms, such as charge gap formation. Withdecreasing temperature, an abrupt resistivity increasewith a bump feature occurs below ∼
100 K. No hysteresisis observed between zero field cooling and warming ρ (T)data. The inset shows the temperature derivative of theelectrical resistivity dρ/dT obtained at H = 0 and 9 T.The parent 10-3-8 compound shows two kinks at T =103 K and T = 95 K in dρ/dT . These features are alsoobserved in underdoped Ba(Fe − x Co x ) As , where thehigher temperature kink is related to the structural phasetransition and the lower temperature kink is relatedto the magnetic phase transition , suggesting that astructural phase transition in the parent 10-3-8 phasemay occur at 103 K and a magnetic phase transitionmay occur at 95 K. No change in the temperature ofthe anomalies is observed with 9 T applied field. Figure2 (b) shows the temperature dependent susceptibility χ (T) taken at 4 T. At 300 K, the susceptibility is around1 × − emu/mole, similar to that ofBaFe As . Unlike the linear temperature dependenceobserved in BaFe As at high temperatures, from300 to 100 K, the susceptibility is only weaklytemperature dependent with a minimum around 200 K.As temperature decreases, a drop in susceptibility isobserved at ∼
100 K, which is shown in the inset ofFig. 2 (b). This susceptibility drop is consistent withthe resistivity measurement and supports the existenceof a magnetic/structural phase transition despite thehighly two-dimensional crystal structure. Below 80 K, aCurie tail is observed, which may be caused by magneticimpurities. This paramagnetic contribution combineswith the intrinsic magnetism and may lead to the weaklytemperature dependent susceptibility from 200 to 300 K.Figure 2 (c) shows the temperature dependent specificheat data. A clear specific heat anomaly is observedaround 100 K, which is consistent with both ρ (T) and χ (T) data. Below 5 K, assuming there are no magneticexcitations, C p /T obeys the relation of C p /T = γ + β T , with an electronic specific heat coefficient γ =4 . -K and β = 0 .
95 mJ/mole-Fe -K ,corresponding to a Debye temperature of 256 K. T (K) (a) R / R (b) T c (c) πχ T (K)
H = 5 mTH // ab
T (K) - R H ( c m / C ) x=0 H ⊥ ab|H| = 9T T (K) R / R FIG. 3: The evolution of the (Ca − x La x ) (Pt As )(Fe As ) series with doping. (a) negative Hall coefficient - R H (T)for x = 0 and 0.145 compositions. (b) The temperaturedependent R/R K . Inset: The amplified R/R K . (c) TheZFC and FC 4 πχ taken at 5 mT with H along the ab plane.(d) The temperature-concentration phase diagram. The evolution of the (Ca − x La x ) (Pt As )(Fe As ) series with doping is presented in Fig. 3. Figure 3(a) shows the temperature dependent Hall coefficientfor the x =0 and x =0.145 compounds. The negative Hall coefficient indicates the dominant role of electrons.A dramatic slope change of Log | R H | , indicating agap opening related to the structural/magnetic phasetransition, is observed in the parent compound, butnot in the x =0.145 compound. Within the single bandmodel, the carrier concentration is determined from n = -1/ eR H , which leads to n x =0300 K =8.7 × cm − and n x =0 . K =1.06 × cm − . Using the unit cell volume V= 788.1 ˚ A , the estimated extra carrier concentrationdue to the La doping is ( n x =0 . K - n x =0300 K ) × V =1.5/unit cell. This number is consistent with theWDS measurement assuming one La atom adds oneelectron. Figure 3 (b) shows the temperature dependentnormalized resistivity
R/R K . The high temperatureresistive bump is suppressed to 87 K in the x = 0 . x ≥ . , thisimplies that no structural phase transition occurs. Forthe samples with x ≤ . T min . Although T min is sample dependent, its average value decreases withincreasing doping and disappears at x = 0 . T min is: 210 ±
30K for x = 0, 200 ±
15 K for x = 0 . ±
30 K for x = 0 . ±
40 K for x = 0 .
065 and 70 ± x = 0 . x ≥ . T c first appears at 13.3 K in the x = 0 .
043 sample,rises to 26.1 K in the x = 0 .
145 sample, and thendecreases to 22.1 K in the x = 0 .
182 sample. Figure 3 (c)presents ZFC and FC susceptibility data taken at 5 mTwith
H//ab . The criterion to infer T c is shown in thefigure. Although the transitions are broader comparedto Ba(Fe − x Co x ) As single crystals, the large shieldingfraction is comparable to the 122 series, which indicatesbulk superconductivity. A small Meissner fraction is acommon feature in Fe-pnictide superconductors and isattributed to flux pinning. The temperature-compositionphase diagram, constructed from the above physicalproperties, is shown in Fig. 3 (d). With La doping, thestructural/magnetic phase transitions are suppressed andsuperconductivity occurs, with a maximum T c around26.1 K at x = 0 . x = 0 .
03, it is not yetclear whether there is a coexistence region of AFM andsuperconductivity, nor is it clear how T evolves withdoping. Further work is necessary to resolve these issues.To study the superconducting state of the La doped10-3-8 superconductors in detail, a representative samplewith x = 0 .
093 was chosen. Figure 4 (a) shows thespecific heat measured at H = 0 T and 9 T with H ⊥ ab . A specific heat jump can be observed, confirmingbulk superconductivity. The inset shows C p /T vs. T taken at H = 0 T. The inferred residual γ is 5.8mJ/mole-Fe K and the Debye temperature is 257 K.Figure 4 (b) shows a plot of ( C Tp − C Tp ) /T vs. T.Using an equal entropy construction shown in Fig. 4(b), the resulting ∆ C p /T c is 13 mJ/mole-Fe K and T c
12 16 20 2402468101210 20 3001510 2010 200.00.7 x=0.093
H = 0 T H = 9 T(a) T (K) C p / T ( J / m o l - K ) T (K) C p / T ( m J / m o l - K ) H // ab
10% 10% 50% (d) H ab H c ( T ) T c (K)C p /T c =13 mJ/mole-Fe K T c =21.9 K(b) T (K) ( C p0 T - C p9 T ) ( m J / m o l e - K ) T c
10% H // ab
T (K) 50% (c) H ab R / R H= 9 T7 T5 T3 T2 T1 T0.5 T0 T
FIG. 4: Physical properties of the superconducting state inthe x =0.093 10-3-8 compound. (a) C p /T vs. T in H = 0 T(open squares) and 9 T (solid line). Inset: C p /T vs. T in 0T. (b) ( C Tp − C Tp ) /T vs. T. T c is inferred using the equalentropy construction shown in figure (c) R/R K taken atH=0, 0.5, 1, 2, 3, 5, 7, 9 T with H along and perpendicular tothe ab plane. (d) H c vs. T obtained with the 10% and 50%criteria from the resistivity data in (c). is 21.9 K. These values fall onto the Budko-Ni-Canfield(BNC) log-log plot reasonably well , adding one moreexample to the BNC scaling, which reveals ∆ C p /T c isproportional to T c for a large number of 122, 111, 1111based superconductors. The calculated γ n is 16 ± K , leading to ∆ C p /T c γ n ∼ .
8. Figure4 (c) shows the suppression of T c under an appliedmagnetic field, in which the resistive transition becomes much broader, indicating strong thermal fluctuations ofvortices in this compound. Since T c is suppressed byless than 0.1 K using the 90% criterion when 9 T isapplied along ab plane, only the 50% and 10% criteriaare employed to infer T c under field. The derivedupper critical field H c (T) is summarized in Fig. 4(d). The orbital limiting H c (0) can be calculatedvia the WHH equation, -0 . T c dH c /dT | T c . Using the50% criterion, the estimated H //abc (0) ∼
400 T and H ⊥ abc (0) ∼
60 T; using the 10% criterion, the estimated H //abc (0) ∼
70 T and H ⊥ abc (0) ∼
10 T. H //abc (0)obtained from 50% criterion is almost 4 times of thatin SmFeAsO . F . with a T c of 40 K, implying a verylarge gap formation. Although the H ⊥ abc curve inferredfrom the 50% criterion shows roughly linear behavior, the H ⊥ abc curve inferred from the 10% criterion shows upwardcurvature, which is common in cuprate and multigap1111 superconductors .In conclusion, we have characterized superconducting(Ca − x La x ) (Pt As )(Fe As ) ( x = 0 to 0.182) singlecrystals. With La doping, the structural/magnetic phasetransitions around 100 K in the pure 10-3-8 compoundare suppressed. Bulk superconductivity occurs at 13.3K at 4.4% doping, rises to 26.1 K at 14.5% doping, andthen decreases to 22.1 K at 18.2% doping.Work at Los Alamos was performed under theauspices of the U.S. Department of Energy, Office ofScience, Division of Materials Science and Engineering.Work at Princeton University was supported by theAFOSR MURI on superconductivity. Work at AmesLaboratory (WES, MAT, RP) was supported by theU.S. Department of Energy, Office of Basic EnergyScience, Division of Materials Sciences and Engineering.Ames Laboratory is operated for the U.S. Departmentof Energy by Iowa State University under Contract No.DE-AC02-07CH11358. Dr. Ni acknowledges the MarieCurie Fellowship at Los Alamos National Laboratory.The authors thank Mr. Eunsung Park, Dr. Xin Lu andDr. Ryan Baumbach for useful discussions. Yoichi Kamihara, Takumi Watanabe, Masahiro Hirano,and Hideo Hosono, J. Am. Chem. Soc., 130, 3296 (2008) Marianne Rotter, Marcus Tegel, and Dirk Johrendt, Phys.Rev. Lett., 101, 107006 (2008) 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 Commun.,148538 (2008). Xiyu Zhu, Fei Han, Gang Mu, Peng Cheng, Bing Shen,Bin Zeng, and Hai-Hu Wen, Phys. Rev. B 79, 220512(R)(2009) N. Ni, J. M. Allred, B. C. Chan, and R. J. Cava, Proc.Natl. Acad. Sci. (USA) 108, E1019 (2011) S. Kakiya, K. Kudo, Y. Nishikubo, K. Oku, E. Nishibori,H. Sawa, T. Yamamoto, T. Nozaka, and M. Nohara, J.Phys. Soc. Jpn. 80, 093704 (2011) C. Lohnert, T. Sturzer, M. Tegel, R. Frankovsky, G.Friederichs, and D. Johrendt, Angew. Chem. Int. Ed. 50,9195 (2011). K. Cho, M. A. Tanatar, H. Kim, W. E. Straszheim, N. Ni,R. J. Cava, and R. Prozorov, Phys. Rev. B 85, 020504(R)(2012) Z. J. Xiang, X. G. Luo, J. J. Ying, X. F. Wang, Y. J. Yan,A. F. Wang, P. Cheng, G. J. Ye, and X. H. Chen Phys.Rev. B 85, 224527 (2012) T. Zhou, G. Koutroulakis, J. Lodico, Ni Ni, J. D.Thompson, S. E. Brown, and R. J. Cava, arXiv:1212.3901,unpublished Tobias St¨urzer, Gerald Derondeau, and Dirk Johrendt,Phys. Rev. B 86, 060516(R) (2012) T. J. B. Holland and S. A. T. Redfern, Mineralogical
Magazine 61, 65 (1997) M. A. Tanatar, N. Ni, A. Thaler, S. L. Budko, P. C.Canfield, and R. Prozorov, Phys. Rev. B 82, 134528 (2010) N. Spyrison, M. A. Tanatar, Kyuil Cho, Y. Song,Pengcheng Dai, Chenglin Zhang, and R. Prozorov, Phys.Rev. B 86, 144528 (2012) N. Ni, A. Thaler, J. Q. Yan, A. Kracher, E. Colombier, S.L. Budko, P. C. Canfield, and S. T. Hannahs, Phys. Rev.B 82, 024519 (2010) D. K. Pratt, W. Tian, A. Kreyssig, J. L. Zarestky, S.Nandi, N. Ni, S. L. Budko, P. C. Canfield, A. I. Goldman,and R. J. McQueeney, Phys. Rev. Lett. 103, 087001 (2009) G. M. Zhang, Y. H. Su, Z. Y. Lu, Z. Y. Weng, D. H. Lee,and T. Xiang, Europhys. Lett., 86, 37006 (2009) X. F. Wang, T. Wu, G. Wu, H. Chen, Y. L. Xie, J. J. Ying,Y. J. Yan, R. H. Liu, and X. H. Chen, Phys. Rev. Lett.,102, 117005 (2009) Sergey L. Budko, Ni Ni, and Paul C. Canfield, Phys. Rev.B 79, 220516(R) (2009) G. R. Stewart, Rev. Mod. Phys. 83, 1589 (2011) J. S. Kim, B. D. Faeth, and G. R. Stewart, Phys. Rev. B86, 054509 (2012) We asumme the phonon contribution of the x = 0 sampleand x = 0 .
145 sample obey the law of corresponding states.Then the electronic specific heat of the x = 0 .
145 sampleis obtained using C x =0 . e = C x =0 . p − A · C x =0 phonon , where C x =0 phonon is obtained by subtracting the electronic part from C x =0 p . A and γ x =0 . n are thus determined by enforcing R T C γ x =0 . n dT = R T C C x =0 . e /T dT . Hyun-Sook Lee, Marek Bartkowiak, Jae-Hyun Park,Jae-Yeap Lee, Ju-Young Kim, Nak-Heon Sung, B. K. Cho,Chang-Uk Jung, Jun Sung Kim, and Hu-Jong Lee, Phys.Rev. B 80, 144512 (2009) A. P. Mackenzie, S. R. Julian, G. G. Lonzarich, A.Carrington, S. D. Hughes, R. S. Liu, and D. C. Sinclair,Phys. Rev. Lett. 71, 1238 (1993)25