Ta5GeB2: New T2 superconductor phase

We report superconductivity at TC ~3.8 K in the new ternary phase Ta5GeB2. Bulk superconductivity is confirmed by magnetization, electrical resistivity and heat capacity measurements, the results showing conventional bulk superconductivity. Ta5GeB2 is a further example of a stoichiometric T2 phase with Cr5B3 prototype structure. © 2015 Elsevier B.V. All rights reserved.


Introduction
M 5 Si 3 B and M 5 Ge 3 B (where M is a transition metal) are well known through a variety of physical properties with potential for application at high temperatures [1e3] as well as being an interesting class of materials for more fundamental investigations regarding, for example, superconductivity. Nb 5 Si 3 and Nb 5 Ge 3 are known to superconduct at 0.7 K [4] and 0.5 K [5] respectively which with doping with small radii elements such as B, N or C have T C 's enhanced to 7.8 K for Nb 5 Si 3 and 15.3 K [6] for Nb 5 Ge 3 [7]. In this case, the superconducting properties are related to a high temperature phase (with a prototype Mn 5 Si 3 with hexagonal symmetry) that is stabilized by the interstitial doping mentioned previously.
The NbeSieB ternary system also shows that it is possible to stabilize another phase, the so called T 2 phase, by substitution of Si for B. Samples with composition Nb 5 Si 3-x B x show maximum solubility for x ¼ 1 (Nb 5 Si 2 B) [8]. Interestingly, Nb 5 Si 3-x B x was found to be superconducting at certain B doping and it was the first Cr 5 B 3 prototype structure phase to be found as a superconductor. Many others compounds with the same prototype structure (Cr 5 B 3 ) were found to be superconductors as well, such as W 5 SiB 2 , (W, Ta) 5 SiB 2 and Mo 5 SiB 2 [9e11]. In Mo 5 SiB 2 the T 2 phase is stoichiometric, and its superconducting properties emerge with just one critical temperature. However, it is common to find other boron-silicide compounds that have the T 2 phase with a solubility limit of Si for B without nucleation of secondary phases and that it is possible to observe that the critical temperature is strongly dependent on the substitution level of Si.
Recent reports propose the existence of stoichiometric phases in the binary system Ta e Ge such as Ta 3 Ge, Ta 5 Ge 3 , [12e15]. Ta 3 Ge, for instance, can assume two different prototype structures, a-Ta 3 Ge and b-Ta 3 Ge. a-Ta 3 Ge is stable below 1550 C and has the Ni 3 P prototype structure with tetragonal symmetry [13], while b-Ta 3 Ge is stable above 1550 C and in spite of also having also tetragonal symmetry has the Ti 3 P prototype structure [14]. Another interesting phase is Ta 5 Ge 3, also known to exhibit two different crystal structures, one stable at high temperatures in the W 5 Si 3 prototype with tetragonal symmetry and another stable at room temperature in the Mn 5 Si 3 structure with hexagonal symmetry [16]. However when B is add to this system, the new Ta-Ge-B ternary shows two different equilibrium phases with composition of Ta 5 Ge 3 B and Ta 5 Ge 2 B at 700 C [17]. The first one consists of a Mn 5 Si 3 prototype with interstitial doping of B, the doping causing this phase to be stable at room temperature. The second one is the so called T 2 phase with Cr 5 B 3 prototype structure that is stabilized by substituting Ge for B at the 8 h Wyckoff position. Here we report results of resistivity, magnetization and specific heat as function of temperature that confirms the existence of superconductivity in Ta 5 GeB 2 as another example of the T 2 phase with Cr 5 B 3 prototype structure.

Experimental procedure
Polycrystalline samples of Ta 5 Ge 3-x B x with 0.2 Â 1.0 were prepared by solid state reaction. High purity powders of Ta, Ge and B were weighted in stoichiometric amounts, homogenized, pressed into pellets, sealed in a quartz ampoule under Ar atmosphere, heat treated at 1200 C for 100 h and finally quenched in ice water. Another heat treatment at 1800 C for 24 h was necessary to achieve a single phase sample within the X-ray diffraction resolution. This treatment was carried out in a resistive furnace (tubular Ta heating element) under argon. X-ray powder diffraction patterns were performed at room temperature with 40 kVe30 mA, Cu-Ka radiation, and Ni filter. The 2q data were collected from 10 to 90 using a step of 0.05 . X-ray diffraction data were analyzed using Rietveld refinement with the software PowderCell [18] Vesta Crystallography [19] and GSAS [20].
Physical properties were obtained using a commercial VSM-PPMS EverCool II from Quantum Design. Magnetization as a function of temperature was determined with zero field cooling (ZFC) and field cooling (FC) in an applied magnetic field of 50 Oe. Electrical resistivity as a function of temperature was measured using the standard four-probe method from 1.8 to 300 K. Here, we define the superconducting transition temperature (TC) as the temperatures corresponding to a 2% resistivity drop, a 1% magnetization drop in the ZFC measurements and a 1% heat capacity anomaly start from normal state. These measurements were done both without and in applied magnetic field in order to estimate the upper critical field. Specific heat of a polished flat sample with Ta 5 GeB 2 composition was measured in the range of 2 Ke20 K. . 1 shows the X-ray diffraction pattern of a sample annealed at 1800 C for 24 h, along with the results of the Rietveld refinement and simulated structure.

Fig
The refinement was stable and converged fast. The parameter goodness-of-fit (c 2 ) and weighted-profile reliability factor (R wp ) were 1.838 and 8.68% respectively, reasonable values for a reliable X-ray diffraction refinement. The peaks observed in 36.4 (in overlap with a peak of the T 2 phase), 40.9 and 63 (pointed out by stars in the Fig. 1) were indexed as Ta 3 Ge phase published elsewhere [12] (space group 86 and Ni3P prototype structure). A sample with this composition was prepared in other to investigate if this phase could give rise to the superconductivity observed and regarding the low temperature properties this compound is found to be a Pauliparamagnetic (data not shown). The difference between the experimental data and the refinement is shown by the blue line in Fig. 1(a). The refinement was performed by starting with the structure published for the Ta 5 Ge 3 as a-TaGe 0.5 [16] (Fig. 1(b) shows the simulated), however, some significant differences in the intensity between the observed and calculated X-ray diffraction patterns were only diminished by considering that B atoms were occupying the 8 h Wyckoff position (Fig. 1(c) shows the simulated structure). It is important to mention that neutron diffraction would be necessary to confirm that the B atoms are truly in these sites. The published data for Ta 5 Ge 3 a ¼ b ¼ 6.599 Å, c ¼ 12.01 Å for the lattice parameters and space group I4/mcm (140). Our results for the same space group are a ¼ b ¼ 6.239 Å, c ¼ 11.578 Å for the lattice parameters and with B atoms in the 8 h Wyckoff position one expects to see smaller lattice parameters since the B ionic radius is smaller than the Ge ionic radius. It is important to mention that this was the only composition where single phase tetragonal T 2 phase with Cr 5 B 3 prototype structure were observed. This behavior also occurs in the MoeSieB system [11]. These results suggest that this is another example of a stoichiometric T 2 phase. In NbeSieB system the solubility range of the substitution of Si for B is relatively large [6].
Magnetization as a function of temperature is shown in Fig. 2. Fig. 2 displays a superconducting transition close to 3.8 K in the ZFC and FC regimes and the inset shows the M vs H dependence at 2.0 K with the typical signature of type II superconductivity. The estimated superconducting volume by ZFC curve is around 80% which strongly suggests a bulk superconducting state. Resistivity as a function of temperature measurements also shows a  superconducting transition with onset temperature close to 3.8 K which is consistent with magnetization measurements as shows Fig. 3. Fig. 3(b) shows the offset temperature (R ¼ 0) around 3.5 K, attesting to the excellent quality of the polycrystalline sample obtained after annealing at 1800 C. Also shown is the resistivity as a function of temperature in different applied magnetic fields. The shift of the transition temperature as a function of applied magnetic field is typical of a real superconducting state. We used the midpoint of the transition as a criterion to define H c2 and used this data in applying the so called WHH theory [21]. Using Eq. (1) it is possible to estimate the upper critical field at zero K. Fig. 4 shows the H c2 phase diagram and the results after applying the WHH theory. Experimental data is shown as hollow circles and the calculated data as the red line.
In this diagram the upper critical field at zero Kelvin is 5190 Oe which is consistent with the M vs H measurement showed in Fig. 2. The coherence length estimate from the upper critical field at zero Kelvin is about 25,2 nm. These results strongly suggest that this compound is a new bulk superconductor with T C close to 3.8 K.
The jump close to 3.8 K is clear evidence of bulk superconductivity in this material. The Sommerfeld coefficient (g) suggests a high density state at Fermi level. The linear fit at low temperature regime gives b ¼ 0.36975 (mJ/molK 4 ), the coefficient of the phonon contribution. From this one estimates the Debye temperature of Q D ¼ 348 K. The subtraction of the phonon contribution to the specific heat allows us to separate the electronic contribution. This is shown in the inset to Fig. 5 where the jump close to 3.8 K is clearly seen and which is consistent with all measurements shown in this paper. The jump size, DC/gT C , is about 1.4, consistent with weak coupling BCS (1.43).
The electronic contribution in the superconductor state (below of T C ) also allows us to estimate the superconductor gap using the logarithmic plot shown in Fig. 6.
The linear fitting observed in Fig. 6 suggests that this material follows BCS behavior. In this case we can use the BCS prediction for the superconducting state below T C given by: From the fit we get an energy gap (D 0 ) of 0.968 meV for T/ 0 and 2D 0 /K B T c~3 .45, which is the characteristic of weak coupling BCS superconductors. Finally, these results present a new superconductor in the Cr 5 B 3 (T 2 phase) prototype series of Mo 5 SiB 2 , Nb 5 Si 3-x B x , and W 5 SiB 2 .

Conclusions
The systematic synthesis study of Ta 5 GeB 2 with Cr 5 B 3 prototype structure (so-called T 2 phase) presented in this article finds this compound to be a new superconductor in this family. Our results demonstrate that this material is a BCS superconductor with TC 3.8 K with coherence length x 0~2 5.2 nm, q D~3 48 K and gap D 0~0 .968 meV.