Superconductivity on Hole-Doping Side of (La0.5–xNa0.5+x)Fe2As2

Article Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

pubs.acs.org/JACS

Superconductivity on Hole-Doping Side of (La0.5−xNa0.5+x)Fe2As2 Akira Iyo,*,† Kenji Kawashima,†,‡ Shigeyuki Ishida,† Hiroshi Fujihisa,† Yoshito Gotoh,† Hiroshi Eisaki,† and Yoshiyuki Yoshida† †

National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Umezono, Tsukuba, Ibaraki 305-8568, Japan IMRA Material R&D Co., Ltd., 2-1 Asahi-machi, Kariya, Aichi 448-0032, Japan

‡

ABSTRACT: (La0.5−xNa0.5+x)Fe2As2 ((La,Na)122) is an interesting system in the sense that either electrons (x < 0) or holes (x > 0) can be doped into the Fe2As2 layers, simply by changing the composition value x. However, only nonbulk superconducting samples (single crystals) with x = 0.1 have been synthesized to date. Here, we successfully synthesize polycrystalline samples with a wide hole-doping composition range of 0 ≤ x ≤ 0.35 via a conventional solid-state reaction, by tuning the reaction temperature according to x. The parent compound, (La0.5Na0.5)Fe2As2 (x = 0), is a nonsuperconductor with a resistivity anomaly at 130 K due to structural and antiferromagnetic transitions. We find that the temperature of the resistivity anomaly decreases with increasing x and that bulk superconductivity emerges for 0.15 ≤ x ≤ 0.35. The maximum transition temperature is 27.0 K, for x = 0.3. An electronic phase diagram for the hole-doping side is constructed. However, electron-doped samples (x < 0) cannot be synthesized; thus, the other half of the electronic phase diagram of (La,Na)122 requires resolution to study the electron−hole symmetry in Fe-based superconductors.

1. INTRODUCTION Since the discovery of Fe-based superconductors (FeSCs),1 a number of new FeSCs with various crystal structures have been reported.2,3 FeSC crystals are formed through alternate stacking of common superconducting (Fe 2 As 2 or Fe 2 Se 2 ) and compound-specific blocking layers. The blocking-layer diversity yields rich chemistry and physics in FeSCs.2,3 Compounds containing alkaline-earth-metal or Eu divalent ions Ae2+ (Ae = Ca, Sr, Eu, Ba) and alkali metal monovalent ions A+ (A = Na, K, Rb, Cs) in the blocking layers are important materials for FeSCs. 4−19 The compounds crystallize into 122-type (Ae 1−x A x )Fe 2 As 2 ((Ae,A)122) or 1144-type AeAFe 4 As 4 (AeA1144), depending on the difference between the ionic radii of Ae2+ and A+.4 Note that (Ae,A)122 is a solid solution between Ae122 and A122, whereas AeA1144 is a stoichiometric compound. The superconducting transition temperatures (Tc) of optimally doped (Ae,A)122 and AeA1144 (32−38 K) are less dependent on the combination of Ae and A. A new 122-type family of (La0.5−xNa0.5+x)Fe2As2 ((La,Na)122), which differs from (Ae,A)122 and AeA1144 in that trivalent La3+ and monovalent Na+ are combined in the blocking layer, was first synthesized by Yan et al.20 These researchers grew (La,Na)122 single crystals through a self-flux method. However, samples having a fixed composition of x = 0.1 ((La0.4Na0.6)Fe2As2) only were grown, despite adjustments to the growth conditions such as the charge/flux ratio. These single crystals exhibit a structural transition (tetragonalorthorhombic) at 125 K, followed by filamentary superconductivity below 30 K in resistivity. This superconductivity © XXXX American Chemical Society

is understood to be induced at the crystal surface, because the resistivity drop becomes larger in magnitude and sharper in transition if the crystal is immersed in water. As noted by Yan et al., (La,Na)122 provides a material platform for the study of FeSCs, in which the electron−hole asymmetry can be studied by simply varying the La/Na composition (i.e., by varying x). It is necessary to vary x in a wide range to extensively investigate (La,Na)122. Bulk superconductivity may be induced if a sufficient number of holes can be doped into (La,Na)122. However, compared with the cases of Ae2+ and A+, the control of x is more difficult, because La3+ and Na+ differ significantly in chemical properties and valence. Metastable bulk NaFe2As2 (Na122), which is an endmember of (La,Na)122 (x = 0.5), is obtained via topochemical deintercalation of Na+ ions from NaFeAs (Na111) in liquid at room temperature.21−23 Therefore, an interesting research subject is whether (La,Na)122 samples can be synthesized through a solid-state reaction, and the amount by which the La/Na composition in the sample can be changed. In this paper, we carefully investigate the reaction temperature dependence on x and succeed in synthesizing polycrystalline (La,Na)122 samples with a wide hole-doping composition range (0.0 ≤ x ≤ 0.35). We find that bulk superconductivity emerges for 0.15 ≤ x ≤ 0.35 with a maximum Tc of 27.0 K. We report on the synthesis process, crystal structure refinement, electronic phase diagram, and superconductivity of (La,Na)122. Received: October 6, 2017

A

DOI: 10.1021/jacs.7b10656 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Article

Journal of the American Chemical Society

2. EXPERIMENTAL PROCEDURES Polycrystalline samples were synthesized through a conventional solidstate reaction in sealed stainless steel (SUS) pipes.14 LaAs, NaAs, Fe2As, and FeAs starting materials were ground at a composition of (La0.5−xNa0.5+x)Fe2As2 (x = −0.5 ≤ x ≤ 0.5), using a mortar in an N2filled glovebox. Excess As at 5 at. % was added to the mixture, taking the As evaporation from the sample during heating into account. Note that the x values given in this paper are nominal. Ground powder specimens with weights of approximately 0.16 g were pressed into pellets. Each pellet was wrapped in Ta foil to prevent contact with the SUS pipe. Each pellet was then placed into an SUS pipe with a length and outer and inner diameters of 60, 8, and 6 mm, respectively. Both ends of the SUS pipe were sealed with tube-fitting caps (Fujikin VUWJC-8). The SUS pipe was then placed in a furnace preheated to a set temperature (Ts); Ts was maintained for 2.5 h. We found that the Ts value is key to sample synthesis, and it was necessary to adjust Ts for each x. Therefore, we carefully examined Ts for each x; finally, the relation Ts (°C) = 930−200x for 0 ≤ x ≤ 0.5 was selected. A Ts of 940 °C was used for x < 0. Note that the (La,Na)122 phase did not form above 960 °C. Following heating, each SUS pipe was extracted from the furnace and quenched from Ts to room temperature by cooling with water. The (La,Na)122 specimens were stable in air for several months. Powder X-ray diffraction (XRD) patterns were measured at room temperature using a diffractometer with CuKa radiation (Rigaku, Ultima IV). The crystal structure of the sample with x = 0.2 was refined via a Rietveld analysis using BIOVIA’s Materials Studio Reflex software (version 2017 R2).24 Magnetization (M) measurements were performed under a magnetic field (H) of 10 Oe using a magneticproperty measurement system (Quantum Design, MPMS-XL7). The electrical resistivity was measured using a four-probe method.

3. RESULTS AND DISCUSSION 3.1. X-ray Diffraction Analysis. Figure 1 shows representative powder XRD patterns of (La0.5−xNa0.5+x)Fe2As2 samples, with −0.2 ≤ x ≤ 0.5. We first discuss (La0.5Na0.5)Fe2As2 (x = 0), which corresponds to the (La,Na)122 parent compound, because the formal valence of Fe is +2, as for AeFe2As2. As indexed in Figure 1, most diffraction peaks for x = 0 can be assigned to a 122-type structure (I4/mmm), except for the small peaks due to LaFeAsO and LaAs. Note that the formation of LaFeAsO is most likely due to oxygen contamination of the starting materials. No peaks attributable to 1144-type compounds (P4/mmm) (h + k + l = odd number) were observed. Because the ionic radii (VIII) of La3+ (1.16 Å) and Na+ (1.18 Å) are similar, no ordering of La and Na ions occurs in (La0.5Na0.5)Fe2As2, as in AeA1144. On the hole-doping side (x > 0), samples with similar quality to x = 0 were obtained up to x = 0.35, as is apparent from Figure 1 (the XRD pattern for x = 0.35 is not displayed). Then, for x = 0.4 and 0.45, peaks from Na111 appeared together with those from (Na,La)122 (the XRD pattern for x = 0.4 only is displayed in Figure 1). This result is natural, because Na122 cannot be synthesized by a solid-state reaction. Finally, for x = 0.5, no (La,Na)122 appeared; however, Na111 and FeAs were formed in the sample according to the formula NaFe2As2 (nominal composition) → NaFeAs + FeAs. A small and broad peak ascribed to Na122, which formed at the Na111 grain surfaces in the sample, was observed (triangle, Figure 1). For the electron-doping side (−0.5 ≤ x < 0), large diffraction peaks due to LaAs appeared, and the peak intensity increased with decreasing x, although the samples for −0.4 ≤ x ≤ − 0.1 contained (La,Na)122 phases, as shown for x = −0.1 and −0.2 in Figure 1 as typical XRD patterns. Peaks from 122-type

Figure 1. Powder XRD patterns of (La0.5−xNa0.5+x)Fe2As2 with −0.2 ≤ x ≤ 0.5. Peaks assigned to a 122-type compound are indicated by closed circles (diffraction indices are attached for x = 0 only). The samples with x = 0, 0.1, 0.2, 0.3, and 0.35 are almost single phase, while the other samples contain impurity phases such as NaFeAs, LaAs, and FeAs.

compounds no longer appeared for the sample with the nominal composition of LaFe2As2 (x = −0.5). 3.2. Lattice Parameters. The x dependence of the a- and c-axis lattice parameters is depicted in Figure 2. Part of the plot is divided into two regions in terms of doping polarity, that is, the electron- (x < 0) and hole-doping (x > 0) sides. The a- and c-axis lattice parameters changed almost linearly (Vegard’s law) in the 0 ≤ x ≤ 0.35 range, as indicated by the two dotted lines in Figure 2. This x range (shaded in Figure 2) coincides with that for which almost single-phase samples were obtained (Figure 1). This coincident implies that the actual x values of the (La,Na)122 phase in the samples were almost equal to the nominal compositions. In this region, the a- and c-axis lengths were shortened and elongated, respectively, as x increased. This behavior is commonly observed in hole-doped 122-type compounds4−15 and can be explained by carrier (hole) transfer from the blocking layers to the Fe2As2 layers. The lattice parameters deviated from the linear behavior at x = 0.35, which suggests that the actual x of the (La,Na)122 phase no longer agreed with the nominal values for specimens with 0.35 < x < 0.5. This means that the solubility limit on the hole-doping side exists near x = 0.35. The upper limit of the formal Fe valence in B

DOI: 10.1021/jacs.7b10656 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Article

Journal of the American Chemical Society

Figure 3. Powder XRD pattern and Rietveld refinement of (La0.2Na0.8)Fe2As2 (Obs.: observed; Cal.: calculated). Inset: Refined crystal structure illustrated using the VESTA software32 with some structural parameters.

Figure 2. Nominal x dependence of a- and c-axis lattice parameters for (La0.5−xNa0.5+x)Fe2As2 (−0.4 ≤ x ≤ 0.45). As indicated by the two dotted lines, the lattice parameters change linearly in the shaded composition region (0 ≤ x ≤ 0.35) in which almost single-phase samples are synthesized.

The refined structural parameters of (La0.2Na0.8)Fe2As2 are summarized in Table 1. The As−Fe−As bond angle of the FeAs Table 1. Atomic Coordinates and Isotropic Displacement Parameters (Uiso) of (La0.2Na0.8)Fe2As2 at Room Temperature

(La,Na)122 is approximately +2.35 (x = 0.35), which is close to those for (Ae1−xNax)Fe2As2 (Ae = Ca, Sr, Ba), that is, a formal Fe valence of +3.3−3.35 for x = 0.6−0.7. Note that twice the number of holes were doped with regard to the same x for (La0.5−xNa0.5+x)Fe2As2 compared to (Ae1−xAx)Fe2As2; this is because of the difference in valence between La3+ and Na+ ions in (La,Na)122. On the electron-doping side (x < 0), the lattice parameters also deviated from the linear behavior. This result as well as the evolution in the XRD patterns for x ≤ 0 in Figure 1 means that the actual x values of the (La,Na)122 phases in the samples did not decrease considerably to below 0 under a conventional solid-state reaction. Note that some reports of indirect electron doping for 122-type compounds, that is, substitution of an aliovalent ion for an Ae or A site in Ae122 or A122, have been made.25−28 For example, electron-doped (Sr1−xLax)Fe2As2 polycrystalline superconductors have been synthesized under a pressure of 2−3 GPa.25 Further, electron-doped superconductivity has been induced in BaFe2As2 by replacing Ba2+ with La3+ using a pulsed laser deposition method.26 Thus, such unconventional and nonequilibrium processes will be required for the synthesis of electron-doped (La,Na)122 samples. 3.3. Crystal Structure Refinement. Figure 3 shows the powder XRD pattern and Rietveld refinement for an (La0.2Na0.8)Fe2As2 (x = 0.3) sample having the maximum Tc for (La,Na)122 (i.e., an optimally doped sample). First, the occupancy ratio of La3+ and Na+ was optimized to 0.21(3) and 0.79(3), respectively, by assuming that the sum of the ratio was 1. The occupancy ratio is close to the nominal composition. This analysis supports the interpretation that the actual x values of the (La,Na)122 in the samples were almost equivalent to the nominal values for 0 ≤ x ≤ 0.35. As a final refinement, a virtual chemical species comprised of 21% La3+ mixed with 79% Na+ was placed at an LaNa site with an occupancy ratio of 1. The diffraction pattern was found to be well fit to a weighted-profile reliability factor (Rwp) of 9.56% and an expected reliability factor (Re) of 10.18%, as demonstrated in Figure 3. The refined crystal structure is illustrated in the inset of Figure 3.

atom

x

y

z

1000Uiso (Å2)

LaNa Fe As

0 0 0

0 0.5 0

0 0.25 0.3659(1)

31(1) 22(1) 24(1)

tetrahedrons (αAs−Fe−As) and the As height from the Fe layers (hAs), which are important factors governing Tc in FeSCs,29−31 were obtained as 106.7(1)° and 1.428(3) Å, respectively. The a-axis length (3.8409(1) Å) is the smallest class for Fe arsenide superconductors and is almost identical to that of optimally doped (Ca0.33Na0.66)Fe2As2 (3.841 Å).10 However, the αAs−Fe−As (hAs) value of (La0.2Na0.8)122 is considerably smaller (greater) than that of (Ca0.33Na0.66)Fe2As2 (αAs−Fe−As = 107.77°; hAs = 1.401 Å), because the Fe−As bond length (dFe−As) of (La0.2Na0.8)Fe2As2 is longer than that of (Ca0.33Na0.66)Fe2As2 (2.377 Å). As a result, the αAs−Fe−As (hAs) value of (La0.2Na0.8)Fe2As2 is the smallest (greatest) for optimally doped (Ae,A)122 and AeA1144 superconductors. Note that the averaged αAs−Fe−As and hAs values are used for AeA1144, because two αAs−Fe−As and hAs values exist for that case. The characteristics of the crystal parameters may affect Tc in (La,Na)122, as described below. Space group: I4/mmm. a = 3.8409(1) Å, c = 12.3245(2) Å. V = 181.81 Å3. Z = 2. Rwp = 9.56%, Re = 10.18%, S = 1.19. Preferred orientation parameter: R0 = 1.661(4), direction = ⟨0.402, 0.138, 0.905⟩. The occupancy was fixed to 1 at all atomic sites. Selected structural parameters: dFe−As = 2.393(2) Å, αAs−Fe−As = 106.7(1)°, hAs = 1.428(3) Å. 3.4. Superconductivity. The evolution of superconductivity on the hole-doping side (x ≥ 0) of (La0.5−xNa0.5+x)Fe2As2 is clearly apparent from Figure 4a,b. In this experiment, the parent compound (x = 0) exhibited no diamagnetic magnetization. However, the sample with x = 0.1 exhibited a trace of superconductivity below 15 K (this is not clearly visible in Figure 4a). This behavior is attributable to a small superC

DOI: 10.1021/jacs.7b10656 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Article

Journal of the American Chemical Society

(109.5° and 1.38 Å, respectively) among the optimally doped (Ae,A)122 and AeA1144 superconductors. Either or both explanations may apply. 3.5. Electronic Phase Diagram. Figure 5 shows the temperature-dependent normalized resistivity of (La,Na)122

Figure 5. Temperature-dependent normalized resistivity of (La0.5−xNa0.5−x)Fe2As2 with −0.20 ≤ x ≤ 0.30. Anomalies are indicated by arrows. Inset: Composition x dependence of resistivity at 300 K.

for −0.20 ≤ x ≤ 0.30. The resistivity at 300 K is plotted against x in the inset of Figure 5. The (La0.4Na0.6)Fe2As2 single crystal is demonstrated to exhibit a structural phase transition from a high-temperature tetragonal to a low-temperature orthorhombic phase at a Ts of 125 K.20 This structural transition is accompanied by an anomaly in the temperature dependence of the electrical resistivity and antiferromagnetic ordering of the Fe moments.20 The anomalies indicated by arrows in Figure 5 correspond to the structural and magnetic transitions in (La,Na)122. The Ts of the parent compound of (La,Na)122 (x = 0) was 130 K. A slightly doped sample (x = 0.05) exhibited an anomaly at almost the same temperature (∼125 K) as the single crystal. The value of Ts further decreased to 113 K, and a trace of superconductivity appeared below 20 K for x = 0.1. Then, as apparent from Figure 5, the anomaly became unclear, and an obvious superconducting transition emerged for x = 0.15. In addition, Tc increased with x to 27.0 K (onset) for x = 0.3, which is consistent with the magnetization measurements. On the other hand, Ts was almost unchanged for −0.2 ≤ x ≤ 0, which again implies that the actual x (the formal valence of Fe) is almost equal to that of the parent compound (x = 0). Superconductivity traces were observed for x = −0.1 and −0.2 below 10−15 K. This behavior may be due to a small electrondoped (La,Na)122 superconducting component in these samples. Figure 6 shows the electronic phase diagram (plots of Ts and Tc on x for the almost single-phase samples) of (La,Na)122. Note that Tc was defined as the temperature at which the fieldcooled M/H decreased by 10% of its value at 5 K. As the number of doped holes increased, the structural transition was suppressed and superconductivity emerged; this behavior has been widely observed for FeSCs.2,3 The overall appearance of the phase diagram for the hole-doping side of (La,Na)122 is similar to that reported for the (Ae,Na)122 (Ae = Ca, Sr) system.10,14 However, it is apparent that large unsettled regions

Figure 4. Temperature-dependent M/H of (La0.5−xNa0.5−x)Fe2As2 with (a) x = 0.0−0.30 and (b) x = 0.30−0.50 below T ≤ 35 K. The magnetization was measured using zero-field-cooled (ZFC) and fieldcooled (FC) processes.

conducting component owing to the inhomogeneity of x in the sample and existing more or less in polycrystals. Thus, most of the sample with x = 0.1 was nonsuperconducting. A large diamagnetic magnetization, sufficient for a bulk superconductor, began to appear in the sample with x = 0.15, as shown in Figure 4a. From Figure 4a,b, it is apparent that Tc increased with increasing x, reaching a maximum value of 27 K at x = 0.3. The superconducting shielding volume fraction calculated using the ZFC magnetization value at 5 K for x = 0.3, without demagnetization correction, is 122%. Then, Tc remained almost unchanged for 0.3 ≤ x ≤ 0.45, while the superconducting volume fraction decreased with greater x; this also implies that the upper (solubility) limit of x is 0.35−0.4. The formal Fe valence (v) for (La0.5−xNa0.5+x)Fe2As2 with x = 0.3 is 2.3, which is almost identical to those for optimally doped (Ca1−xNax)Fe2As2 (x = 0.66, v = 2.33, Tc = 34 K) and (Sr1−xNax)Fe2As2 (x = 0.55, v = 2.275, Tc = 36.5 K).10,14 Small-magnitude superconducting diamagnetism was observed below 10 K for x = 0.5, which was most likely due to the small degree of Na122 phase observed in the XRD pattern at this x value. Note that the maximum Tc (27 K) for (La,Na)122 is considerably lower than those reported for the other optimally doped (Ae,A)122 or AeA1144 superconductors (Tc = 32−38 K).4−19 The following two explanations for this low Tc are feasible. First, the combination of La3+ and Na+ induces greater potential disorder to superconducting Fe2As2 layers than that for Ae2+ and A+. Second, the αAs−Fe−As and hAs for the optimally doped (La,Na)122 are furthest removed from the optimal value D

DOI: 10.1021/jacs.7b10656 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Article

Journal of the American Chemical Society

(4) Rotter, M.; Tegel, M.; Johrendt, D.; Schellenberg, I.; Hermes, W.; Pöttgen, R. Phys. Rev. B: Condens. Matter Mater. Phys. 2008, 78, 020503. (5) Lv, B.; Gooch, M.; Lorenz, B.; Chen, F.; Guloy, A. M.; Chu, C. W. New J. Phys. 2009, 11, 025013. (6) Sasmal, K.; Lv, B.; Lorenz, B.; Guloy, A. M.; Chen, F.; Xue, Y.-Y.; Chu, C.-W. Phys. Rev. Lett. 2008, 101, 107007. (7) Bukowski, Z.; Weyeneth, S.; Puzniak, R.; Moll, P.; Katrych, S.; Zhigadlo, N. D.; Karpinski, J.; Keller, H.; Batlogg, B. Phys. Rev. B: Condens. Matter Mater. Phys. 2009, 79, 104521. (8) Shirage, P. M.; Miyazawa, K.; Kito, H.; Eisaki, H.; Iyo, A. Appl. Phys. Express 2008, 1, 081702. (9) Wu, G.; Chen, H.; Wu, T.; Xie, Y. L.; Yan, Y. J.; Liu, R. H.; Wang, X. F.; Ying, J. J.; Chen, X. H. J. Phys.: Condens. Matter 2008, 20, 422201. (10) Zhao, K.; Liu, Q. Q.; Wang, X. C.; Deng, Z.; Lv, Y. X.; Zhu, J. L.; Li, F. Y.; Jin, C. Q. Phys. Rev. B: Condens. Matter Mater. Phys. 2011, 84, 184534. (11) Cortes-Gil, R.; Clarke, S. J. Chem. Mater. 2011, 23, 1009−1016. (12) Cortes-Gil, R.; Parker, D. R.; Pitcher, M. J.; Hadermann, J.; Clarke, S. J. Chem. Mater. 2010, 22, 4304. (13) Avci, S.; Allred, J. M.; Chmaissem, O.; Chung, D. Y.; Rosenkranz, S.; Schlueter, J. A.; Claus, H.; Daoud-Aladine, A.; Khalyavin, D. D.; Manuel, P.; Llobet, A.; Suchomel, M. R.; Kanatzidis, M. G.; Osborn, R. Phys. Rev. B: Condens. Matter Mater. Phys. 2013, 88, 094510. (14) Shinohara, N.; Tokiwa, K.; Fujihisa, H.; Gotoh, Y.; Ishida, S.; Kihou, K.; Lee, C. H.; Eisaki, H.; Yoshida, Y.; Iyo, A. Supercond. Sci. Technol. 2015, 28, 062001. (15) Anupam; Paulose, P. L.; Ramakrishnan, S.; Hossain, Z. J. Phys.: Condens. Matter 2011, 23, 455702. (16) Qi, Y.; Gao, Z.; Wang, L.; Wang, D.; Zhang, X.; Ma, Y. New J. Phys. 2008, 10, 123003. (17) Iyo, A.; Kawashima, K.; Kinjo, T.; Nishio, T.; Ishida, S.; Fujihisa, H.; Gotoh, Y.; Kihou, K.; Eisaki, H.; Yoshida, Y. J. Am. Chem. Soc. 2016, 138, 3410. (18) Liu, Y.; Liu, Y. B.; Tang, Z. T.; Jiang, H.; Wang, Z. C.; Ablimit, A.; Jiao, W. H.; Tao, Q.; Feng, C. M.; Xu, Z. A.; Cao, G. H. Phys. Rev. B: Condens. Matter Mater. Phys. 2016, 93, 214503. (19) Kawashima, K.; Kinjo, T.; Nishio, T.; Ishida, S.; Fujihisa, H.; Gotoh, Y.; Kihou, K.; Eisaki, H.; Yoshida, Y.; Iyo, A. J. Phys. Soc. Jpn. 2016, 85, 064710. (20) Yan, J.-Q.; Nandi, S.; Saparov, B.; Č ermák, P.; Xiao, Y.; Su, Y.; Jin, W. T.; Schneidewind, A.; Brückel, Th.; et al. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 91, 024501. (21) Gooch, M.; Lv, B.; Sasmal, K.; Tapp, J. H.; Tang, Z. J.; Guloy, A. M.; Lorenz, B.; Chu, C. W. Phys. C 2010, 470, S276−S279. (22) Todorov, I.; Chung, D. Y.; Claus, H.; Malliakas, C. D.; Douvalis, A. P.; Bakas, T.; He, J.; Dravid, V. P.; Kanatzidis, M. G. Chem. Mater. 2010, 22, 3916−3925. (23) Friederichs, G. M.; Schellenberg, I.; Pöttgen, R.; Duppel, V.; Kienle, L.; Günne, J. S.; Johrendt, D. Inorg. Chem. 2012, 51, 8161− 8167. (24) Dassault Systemes, BIOVIA Corp. Materials Studio Reflex website. http://accelrys.com/products/collaborative-science/bioviamaterials-studio/analytical-and-crystallization-software.html (accessed January 4, 2016). (25) Muraba, Y.; Matsuishi, S.; Kim, S. W.; Atou, T.; Fukunaga, O.; Hosono, H. Phys. Rev. B: Condens. Matter Mater. Phys. 2010, 82, 180512R. (26) Katase, T.; Iimura, S.; Hiramatsu, H.; Kamiya, T.; Hosono, H. Phys. Rev. B: Condens. Matter Mater. Phys. 2012, 85, 140516. (27) Saha, S. R.; Butch, N. P.; Drye, T.; Magill, J.; Ziemak, S.; Kirshenbaum, K.; Zavalij, P. Y.; Lynn, J. W.; Paglione, J. Phys. Rev. B: Condens. Matter Mater. Phys. 2012, 85, 024525. (28) Lv, B.; Deng, L.; Gooch, M.; Wei, F.; Sun, Y.; Meen, J. K.; Xue, Y.-Y.; Lorenz, B.; Chu, C.-W. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 15705.

Figure 6. Electronic phase diagram of hole-doping side (x ≥ 0) of (La0.5−xNa0.5−x)Fe2As2. Ts (closed circles) and Tc (closed squares) are plotted for the almost single-phase samples (0 ≤ x ≤ 0.35). The Tc of metastable Na122 (25 K) is indicated by a closed triangle. We failed to synthesize samples with −0.5 ≤ x < 0 and 0.35 < x ≤ 0.5 by solid-state reaction under ambient pressure.

(−0.5 ≤ x < 0, 0.35 < x ≤ 0.5) remain, which are due to difficulties affecting the sample synthesis.

4. SUMMARY We have successfully synthesized (La,Na)122 samples with a wide hole-doping composition range (0 ≤ x ≤ 0.35). It was found that careful tuning of the reaction temperature is essential for successful synthesis of (La,Na)122 samples containing La3+ and Na+, which have broadly different chemical properties and valence. We investigated the composition dependence of the superconducting and antiferromagnetic (structural) transition temperatures and constructed the holedoping side of the electronic phase diagram for (La,Na)122. However, the other half of the electronic phase diagram remains unresolved. To construct the entire phase diagram of (La,Na)122, it is necessary to synthesize electron-doped samples using unconventional or nonequilibrium processes such as high-pressure and thin film methods.

■

AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Akira Iyo: 0000-0002-9610-647X Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS This work was partially supported by JSPS KAKENHI grant numbers JP16H06439 and JP26247057. REFERENCES

(1) Kamihara, Y.; Watanabe, T.; Hirano, M.; Hosono, H. J. Am. Chem. Soc. 2008, 130, 3296−3297. (2) Tanabe, K.; Hosono, H. Jpn. J. Appl. Phys. 2012, 51, 010005. (3) Chen, X.; Dai, P.; Feng, D.; Xiang, T.; Zhang, F. C. Nat. Sci. Rev. 2014, 1, 371−395. E

DOI: 10.1021/jacs.7b10656 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Article

Journal of the American Chemical Society (29) Lee, C. H.; Iyo, A.; Eisaki, H.; Kito, H.; Fernandez-Diaz, M. T.; Ito, T.; Kihou, K.; Matsuhata, H.; Braden, M.; Yamada, K. J. Phys. Soc. Jpn. 2008, 77, 083704. (30) Lee, C. H.; Kihou, K.; Iyo, A.; Kito, H.; Shirage, P. M.; Eisaki, H. Solid State Commun. 2012, 152, 644−648. (31) Mizuguchi, Y.; Hara, Y.; Deguchi, K.; Tsuda, S.; Yamaguchi, T.; Takeda, K.; Kotegawa, H.; Tou, H.; Takano, Y. Supercond. Sci. Technol. 2010, 23, 054013. (32) Momma, K.; Izumi, F. J. Appl. Crystallogr. 2008, 41, 653.

F

DOI: 10.1021/jacs.7b10656 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX