Synthesis and Properties of CaFe2 As2 Single Crystals
F. Ronning1 , T. Klimczuk1,2 , E.D. Bauer1 , H. Volz1 , J.D. Thompson1
Los Alamos National Laboratory, Los Alamos, New Mexico 87545, USA
Faculty of Applied Physics and Mathematics,
arXiv:0806.4599v2 [cond-mat.str-el] 10 Jul 2008
Gdansk University of Technology, Narutowicza 11/12, 80-952 Gdansk, Poland
(Dated: July 10, 2008)
We report the synthesis and basic physical properties of single crystals of CaFe2 As2 , an isostructural compound to BaFe2 As2 which has been recently doped to produce superconductivity. CaFe2 As2 crystalizes in the ThCr2 Si2 structure with lattice parameters a = 3.887(4) ? and c = A 11.758(23) ?. Magnetic susceptibility, resistivity, and heat capacity all show a ?rst order phase A transition at T0 = 171 K. The magnetic susceptibility is nearly isotropic from 2 K to 350 K. The heat capacity data gives a Sommerfeld coe?cient of 8.2 ?À 0.3 mJ/molK2 , and does not reveal any evidence for the presence of high frequency (> 300 K) optical phonon modes. The Hall coe?cient is negative below the transition indicating dominant n-type carriers.
The discovery of superconductivity in the oxypnictide compounds RFeAs(O1?x Fx ) (R = La, Ce, Pr, Nd, Sm, Gd) with the ZrCuSiAs structure type has stimulated a wealth of activity around the globe [1, 2, 3, 4, 5, 6, 7]. The similarity of this system and the cuprates with respect to the layered structure and the phase diagram with carrier doping, suggests that the physics may be similar. The fact that the parent compounds are metallic indicates additional similarity to the heavy fermion superconductors [8, 9, 10]. If the FeAs layers are critical for the relatively high superconducting transition temperatures then it does not come as a big surprise that superconductivity was found in the related ThCr2 Si2 structure. Hole-doping by potassium in BaFe2 As2 and SrFe2 As2 produces superconductivity up to Tc = 38 K.[13, 18, 19]. Given the relatively high transition temperatures and that single crystals appear relatively easier to synthesize than those in the RFeAsO family, the ThCr2 Si2 structure may be a more ideal system for elucidating the physics of these new Fe-based superconductors.
Currently, the AFe2 As2 compounds are known to be stable with divalent A = Ba, Sr, and Eu[11, 20], of which only the A = Ba and Sr compounds so far have been doped to produce superconductivity [13, 18, 19]. In this paper we report on the synthesis and basic physical properties of single crystals of CaFe2 As2 . A sharp ?rst order anomaly is observed by susceptibility, heat capacity and electrical transport measurements at T0 = 171 K. Single crystals of CaFe2 As2 were grown in Sn ?ux in the ratio Ca:Fe:As:Sn=1:2:2:20. The starting elements were placed in an alumina crucible and sealed under vacuum in a quartz ampoule. The ampoule was placed in a furnace and heated to 500 ? C at 100 ? C hr?1 , and held at that temperature for 6 hours. This sequence was repeated at 750 ? C, 950
C and at a maximum temperature of 1100 ? C, with hold times of 8 hr., 12 hr., and 4 hr,
respectively. The sample was then cooled slowly (?? 4? C hr?1 ) to 600 ? C, at which point the excess Sn ?ux was removed with the aid of a centrifuge. The resulting plate-like crystals of typical dimensions 5 x 5 x 0.1 mm3 are micaceous and ductile and are oriented with the c-axis normal to the plate. CaFe2 As2 crystallizes in the ThCr2 Si2 tetragonal structure (space group no. 139) (Fig. 1) with lattice parameters a = 3.887(4) ? and c = 11.758(23) A ? as revealed by the powder x-ray di?raction pattern shown in Fig. 2. In this structure, A layers of Ca are capped by Fe-As tetrahedra along the c-axis. These Fe-As tetrahedra are the common structural units to the Fe-based RFeAsO and AFe2 As2 superconductors. Magnetic measurements were performed from 1.8 K to 300 K using a commercial SQUID 2
FIG. 1: Crystal structure of CaFe2 As2
101 004 112
0 5 10 15 20 25 30 35 40 45 50 55 60
FIG. 2: Powder X-ray di?raction pattern (CuK?Á radiation) for CaFe2 As2 . Vertical bars at the bottom represent the Bragg peak positions for the ThCr2 Si2 tetragonal (I4/mmm) structure with re?ned cell parameters a = 3.887(4) ? and c = 11.758(23) ?. Miller indices for each peak are A A shown, and a peak from the Sn ?ux is marked with an arrow.
magnetometer. Speci?c heat measurements were carried out using an adiabatic method in a commercial cryostat from 2 K to 300 K. Electrical transport measurements were performed using a LR-700 resistance bridge with an excitation current of 1 mA, on samples for which platinum leads were spot welded. The magnetic susceptibility ?Ö(T ) of CaFe2 As2 measured in a magnetic ?eld H = 5 T with H||ab and H||c is shown in Fig. 3. The susceptibility is essentially isotropic over the entire measured temperature range. Close to T0 =172 K a sharp drop is evident in ?Öab and in ?Öc , albeit slightly smaller, likely indicating a structural transition that is similar to those observed in BaFe2 As2 [12, 14, 18] and LaFeAsO.
H || ab
H || c
FIG. 3: Magnetic susceptibility ?Ö(T ) of CaFe2 As2 measured in a magnetic ?eld H = 5 T for H||ab (black circles) and H||c (blue triangles). A structural transition is indicated by the sharp drop at T0 = 172 K (dashed line) in ?Öc and ?Öab .
The heat capacity presented in ?gure 4 reveals a very sharp symmetric anomaly consistent with a ?rst order phase transition at 172 K (upon warming). In the top inset, the relaxation curve of sample temperature versus time is shown. While a constant heat is applied to the sample
it steadily increases in temperature as dictated by the sample heat capacity and the thermal link to the bath. The plateau in the curve indicates an abrupt increase in the heat capacity as well as the latent heat associated with the ?rst order transition, sharply de?ned in temperature at 171.8 ?À 0.1 K. The low temperature heat capacity is presented in the lower inset. Below 10 K the heat capacity data can be ?t to C = ?ÃT + ?ÂT 3 + ?ÁT 5 . This gives an electronic speci?c heat coe?cient of ?Ã = 8.2 ?À 0.3 mJ/mol K2 . Assuming that the
171.0 170.0 0 5 10 15 20 25
)s( emit ) K lom/Jm( T/C
C (mJ/mol K)
60 40 20 0 0 20 40
0 0 50 100 150 200 250 300
FIG. 4: Speci?c heat versus temperature is shown for CaFe2 As2 . The dashed line represents a simple lattice estimate as described in the text. The top inset displays the relaxation curve at the transition temperature. The lower inset displays the low temperature heat capacity. The solid line is a ?t to C/T = ?Ã + ?ÂT 2 + ?ÁT 4 .
T 3 term is entirely due to acoustic phonons, from the ?Â coe?cient = 0.383 ?À 0.018 mJ/mol K4 we extract a Debye temperature ??D = 292 K. The dashed curve in the ?gure gives the lattice contribution to the speci?c heat based upon a simple Debye model using ??D = 292 K. While this is certainly an oversimpli?cation of the exact phonon density of states, the fact that this gives a reasonable account of the data at high temperatures indicates that there are relatively few high frequency optical phonon modes with energies above 300 K. This is in contrast to the case of phonon-mediated superconductor MgB2 , where analysis of the heat capacity indicates the presence of phonon modes
up to 750 K. As with susceptibility and heat capacity, the resistivity data presented in ?gure 5 contains a clear ?rst order phase transition at 170 K. The jump indicates either an increase in scattering or a decrease in the number of carriers below the transition relative to above it. The samples have a RRR (= ?Ñ(300 K)/?Ñ(4 K)) of 10. A small partial superconducting transition at 3.8 K is due to small Sn inclusions. Data for current parallel to the c-axis on 4 samples (not shown) have a qualitatively similar temperature dependence, but range in absolute magnitude from 50 to 1000 times larger than the in-plane data, possibly a consequence of weakly coupled micaceous layers leading to large variations in the magnitude
25 24 23 22 21 165 170 175
0.00 0 50 100 150 200 250 300
FIG. 5: In-plane resistivity (I ab) and the Hall coe?cient of CaFe2 As2 as a function of temperature. The inset illustrates the thermal hysteresis at the transition observed in resistivity for current along the c-axis.
of the c-axis resistivity. The inset demonstrates the thermal hysteresis expected for a ?rst order phase transition. Also shown in Fig. 5 is the Hall coe?cient. The dominant carrier below the 171 K transition is electron-like. There is a role-over at 15 K which may be due to either the multiband nature of these systems, or due to localization e?ects. With a resolution of 2x10?11 ?-cm/Oe we can not say whether the dominant carrier type at room temperature is also electron-like.
TABLE I: Comparison of properties of known AFe2 As2 compounds where A is a divalent atom. Compound CaFe2 As2 EuFe2 As2 SrFe2 As2 BaFe2 As2
a (?) A 3.887(4) 3.911(1) 3.927(6) 3.962(6)
c (?) A 11.758(23) 12.110(4) 12.37(2) 13.04(2)
??D (K) 292
?Ã (mJ/molK2 ) 8.2(3)
T0 (K) 171 195
refs. this work [20, 21, 22] [11, 16, 17] [11, 12, 14, 15]
Ref  report ?Ã = 33 mJ/mol K2 , but the ?gure indicates a value between 11 and 3.3 mJ/mol K2 .
Brie?y, we compare our results with currently available data on other members of the AFe2 As2 compounds with A = Ba, Sr, and Eu, listed in Table 1. We note that the lattice constants monotonically decrease from BaFe2 As2 to SrFe2 As2 to EuFe2 As2 to CaFe2 As2 , as 6
one would expect given the smaller ionic radii of Ca2+ versus Eu2+ and Sr2+ versus Ba2+ . The higher Debye temperature for CaFe2 As2 (292 K) is consistent with the smaller unit cell volume. However, the structural/SDW phase transition is not monotonic with cell volume even within the group IIA of the periodic table ranging from 80-140 K for BaFe2 As2 , and 195-205 K in SrFe2 As2 compared with 170 K in CaFe2 As2 . (EuFe2 As2 orders at 195 K, and the Eu moments order at 20 K.) The reason for this is not understood, but possibly indicates the sensitivity of the transition to details of the electronic structure. Similarly, the anisotropy of the susceptibility of CaFe2 As2 measured at 5 T is nearly isotropic over the entire temperature range measured, both above and below the transition. Although the overall magnitude is similar to that observed in SrFe2 As2  the anisotropy is qualitatively di?erent. In our case the low temperature Curie-Weiss tail is isotropic, and thus could originate from an impurity contribution. The isotropic behavior above the transition is more consistent with that observed in BaFe2 As2 . The role of trace amounts of ferromagnetic impurity phases such as Fe2 As, will be studied in more detail to determine the intrinsic behavior of the susceptibility. Finally, whether the current samples are a?ected by Sn substitution as suggested for single crystals of BaFe2 As2 grown by a similar technique  must still be investigated. We have synthesized single crystals of CaFe2 As2 , which possesses a ?rst order transition at 170 K, which is likely a combined structural and magnetic transition. Given that superconductivity has been found by doping the isostructural Ba and Sr compounds[13, 18, 19] we believe that the Ca compound is also a likely candidate for the presence of
superconductivity upon chemical substitution. At the completion of this
work we became aware of two other papers reporting the synthesis of CaFe2 As2 . Single crystals of CaFe2 As2 grown using self ?ux and Sn ?ux methods gave results similar to ours, and indeed superconductivity was found upon Na doping.
We acknowledge useful discussions with B. Scott. Work at Los Alamos National Laboratory was performed under the auspices of the U.S. Department of Energy.
 Y. Kamihara, T. Watanabe, M. Hirano, H. Hosono, J. Am. Chem. Soc. 130, 3296 (2008).  X.H. Chen, et. al., Nature 453, 761 (2008).  Z.-A. Ren, et. al., Chin. Phys. Lett. 25, 2215 (2008).  G.F. Chen, et. al., Phys. Rev. Lett. 100, 247002 (2008).  Z.-A. Ren, et. al., Europhys. Lett. 82, 57002 (2008).  Z.-A. Ren, et. al., arXiv:0803.4283 , (2008).  P. Cheng, et. al., Science in China G 51, 719 (2008).  N.D. Mathur, et. al., Nature 394, 39 (1998)  T. Park, et. al., Nature 440, 65 (2006)  H.Q. Yuan, et. al., Science 302, 2104 (2003)  M. P?sterer, G. Nagorsen, Z. Naturforsch. B: Chem. Sci. 35, 703 (1980).  M. Rotter, M. Tegel, D. Johrendt, I. Schellenberg, W. Hermes, R. Poettgen, arXiv:0805.4021 , (2008).  M. Rotter, M. Tegel, D. Johrendt, arXiv:0805.4630 , (2008).  N. Ni, et. al., arXiv:0806.1874 , (2008).  J.K. Dong, et. al., arXiv:0806.3573 , (2008).  J.-Q. Yan, et. al., arXiv:0806.2711 , (2008).  C. Krellner, et. al., arXiv:0806.1043 , (2008).  G.F. Chen, et. al., arXiv:0806.1209 , (2008).  K. Sasmal,et. al., arXiv:0806.1301 , (2008).  R. Marchand, W. Jeitschko, J. Solid State Chem. 24, 351 (1978).  H. Ra?us, et. al., J. Phys. Chem. Solids. 54, 135 (1993).  H.S. Jeevan, Z. Hossain, C. Geibel, P. Gegenwart, arXiv:0806.2876 , (2008).  J. Lashley, et. al., cryogenics 43, 369 (2003)  Ch. W??lti, et. al., Phys. Rev. B 64, 172515 (2001) a
 I. Nowik, I. Felner, arXiv:0806.4078 , (2008).  G. Wu, et. al., arXiv:0806.4279 , (2008).  N. Ni, et. al., arXiv:0806.4328 , (2008).