1 Introduction

The thallium-based superconductor is one of the high-temperature superconductors (HTSC) with relatively higher transition temperatures among the cuprates. Several phases exist depending on the number of CuO and TlO layers. In terms of phase formation and electrical behavior, the Tl-1212 phase is one of the most interesting. The Tl-1212 phase consists of TlBa2CaCu2O7 and TlSr2CaCuO7.

TlBa2CaCu2O7 superconducts at 80 K and is readily prepared by heating at 900 °C [1]. On the other hand, TlSr2CaCuO7 is difficult to produce in pure form. Some studies showed that it is superconducting above 80 K with proper substitution [2], whilst other reports showed that it was not superconducting [3,4,5]. This was due to the high formal average Cu valence (+ 2.5) and over-doped hole states [3]. For the highest transition temperature in the Tl-1212 phase, the optimum average Cu valence ranges from + 2.25 to + 2.35 [6, 7]. This can be achieved by substituting the higher valence ions with an appropriate ionic radius for the Tl-1212 phase. For example, rare-earth element substitution greatly improved the transition temperatures up to 100 K and stabilized the Tl-1212 phase [8,9,10].

Interestingly, including both Ba and Sr with a 1:1 ratio effectively enhanced the superconducting properties of various phases of thallium-based superconductors. In the Tl-2223 phase, transition above 100 K was achieved, and a single Tl-2223 phase was more easily obtained for Tl2(BaSr)Ca2Cu3O10 compared with Tl2Ba2Ca2Cu3O10 [11]. When heated above 865 °C, Tc increased to 102 K with a single Tl-2223 phase [12]. For the Tl-1212 phase, both Ba and Sr bearing with a 1:1 ratio resulted in Tc ranging from 90–108 K with a highly pure Tl-1212 phase [13,14,15,16,17].

Due to the significant enhancement of Tc and phase formation, several studies on the effect of elemental substitution on TlBaSrCaCu2O7 were conducted. Substitution of Se (non-metal) at the Tl site of Tl1-xSexBaSrCaCu2O7 enhanced Tc up to 104 K for x = 0.3 sample [17]. Several reports on Pb substitution at the Tl site showed that the optimized composition was Tl1.6Pb0.4BaSrCaCu2O7 with zero-transition temperature, Tc-zero = 98 K [18,19,20,21]. The onset transition temperature, Tc-onset and Tc-zero were then improved to 118 K and 110 K, respectively when heated at 970 °C [22]. Ga substituted at the Ca site in TlBaSrCa1-xGaxCu2O7 increased Tc-onset for a small amount of substitution (x = 0.1–0.2) [15]. However, substituting transition metals such as W [16] and Ta [14, 23] at the Tl site reduced Tc-onset and Tc-zero and weakened the intergrain coupling and flux pinning strength. These elements are placed next to one another in period VI of the periodic table. Substitution of Te (metalloid) at the Tl site of Tl1-xTexBaSrCaCu2O7 was also reported [17].

Besides that, there are several studies related to Ge substitution in Tl-based superconductors. For example, Ge-substituted (Cu0.5Tl0.5)Ba2Ca3(Cu4−yGey)O12 showed the improvement of transition temperatures and higher quality of diamagnetism was obtained for post-annealed samples [24]. The enhancement of superconductivity was observed in the Ge-substituted (Cu0.5Tl0.5)Ba2Ca2Cu3-yGeyO10-δ due to CuO2/GeO2 planes [25]. Substitution of Ge at Cu site in Tl0.85Cr0.15Sr2CaCu2-xGexO7-δ increased Tc-zero from 98 K (x = 0) to 100 K (x = 0.1) [26]. Studies on fluctuation-induced conductivity of Cu0.5Tl0.5Ba2Ca2Cu2Ge1O10-δ increased 3D fluctuation and improved diamagnetism, attributed to the improvement of phonon density [27].

Studies on metalloid element substitution in TlBaSrCaCu2O7 are limited to Te substitution. Thus, it is interesting to investigate the interaction of another metalloid—Ge with TlBaSrCaCu2O7. Ge is a multivalent element that consists of Ge+2 and Ge+4. The ionic radius of Ge+2 and Ge+4 is 0.73 Å and 0.53 Å, respectively which is smaller than Tl+1 (1.5 Å) and Tl+3 (0.885 Å). Substitution of higher valence state elements (Ge+4) at the Tl site of the Tl-1212 phase may help optimize the average Cu valence. Furthermore, Ge doping in TlBaSrCaCu2O7 may improve our understanding of CuO2/GeO2 planes effects on the superconductivity of the Tl-1212 phase.

This work aims to study the effects of Ge doping on the superconductivity of the Tl-1212 phase with starting formula Tl1-xGexBaSrCaCu2O7 for x = 0–0.6. Results from X-ray diffraction (XRD), scanning electron microscopy (SEM), and resistivity measurements versus temperature and AC susceptibility are reported. The effects of Ge on the phase, lattice parameter, microstructure, transition temperature and critical current density are discussed.

2 Experimental Details

The samples were prepared by the solid-state reaction method with nominal starting composition of Tl1-xGexBaSrCaCu2O7 for x = 0 to 0.6. Appropriate amounts of BaCO3, SrCO3, CaO, and CuO with high purity (≥ 99.0%) were weighed in a stoichiometric ratio (2:1:2) and mixed completely using an agate mortar. The mixed powders were heated at 900 °C for 24 h with several intermittent grindings to obtain a homogenous black powder. Appropriate amounts of Tl2O3 and GeO2 were then added to the precursor, mixed, and then pressed into pellets of 13 mm diameter and 2 mm thickness. Excess 10% of Tl2O3 was added to compensate for the loss of thallium during the heating process. The pellets were then heated in a tube furnace at 970 °C with flowing oxygen for 4 min followed by furnace cooling to room temperature.

The samples were analyzed by the powder X-ray diffraction (XRD) method using a Bruker D8 Advance Diffractometer with CuKα source (λ = 1.5418 Å) to identify the resultant phase formation. The ICDD data bank PDF 00–046-0777 was used as a reference for TlBaSrCaCu2O7 (Tl-1212) phase. The phase volume fraction was estimated by using the ratio of the sum of intensities of each major peak for every phase. The lattice parameters of the Tl-1212 phase were calculated by the Pawley refinement using X'Pert HighScore software. A JEOL Model JSM6010PLUS/LA scanning electron microscope (SEM) was used to observe the microstructure of the samples.

The temperature-dependent DC electrical resistance measurements were performed using the four-probe method with silver paint contact in conjunction with a CTI Cryogenics Model 22 closed-cycle refrigerator. All samples were in pellet form with 13 mm diameter and 2 mm thickness, and the average distance between contacts was about 2 mm. A constant current source between 1 and 100 mA was used. The resistivity at room temperature ρ298 K was measured via the Van der Pauw method.

An AC susceptometer from Cryo Industry model number REF-1808-ACS was used to measure the susceptibility. For the susceptibility measurements, the samples were cut into bar shapes with a cross-section of 2 mm × 2 mm and length 5 mm. An AC signal with frequency 295 Hz and magnetic field of 5 Oe parallel to the samples’ surface was applied. The critical current density, Jc at the peak temperature, Tp of χ″ was determined using the Bean model [28, 29] with the formula Jc(Tp) = H/(lw)1/2 where H is the applied magnetic field, l and w are the dimensions of the cross-section of the bar-shaped sample.

3 Results and Discussion

The experimental and refined XRD patterns of Tl1-xGexBaSrCaCu2O7 for x = 0–0.6 are shown in Fig. 1. The XRD patterns were refined by the Pawley method using X'Pert HighScore software. The Rp, Rwp, and Rexp parameters and the goodness of fit, χ2 as shown in Table 1 were used as the numerical criteria of refinement. The patterns revealed the presence of the Tl-1212 phase in all samples. The Tl-1212 phase was indexed based on the tetragonal unit cell with the space group of P4/mmm (123). The x = 0–0.2 samples showed a single Tl-1212 phase. However, the volume fraction of the Tl-1212 phase decreased as Ge content was increased for x ≥ 0.3 samples. Formation of the impurity phases such as TlSr2Ca2Cu3O9 (Tl-1223), TlSr2CuO5 (Tl-1201), Ca0.3Sr0.7CuO2 (CSCO) was also observed for x ≥ 0.3 samples. The signature peak for TlSr2Ca2Cu3O9 (Tl-1223) phase at 2θ = 5.709° was detected for x = 0.3 and 0.4 samples. In Ga substituted TlBaSrCaCu2O7 superconductor, the Tl-1223 phase was also observed in higher substituted samples [15]. Besides that, in our Ge-doped samples, the volume fraction of the Tl-1201 phase increased with increasing Ge content. Peaks corresponding to GeO2 were also spotted at 2θ = 42.47° and 42.95°. This study implied that a small amount of Ge-doping did not affect the Tl-1212 phase, but further doping suppressed the Tl-1212 formation and resulted in the formation of other impurity phases.

Fig. 1
figure 1figure 1figure 1

Experimental and refined XRD patterns of Tl1-xGexBaSrCaCu2O7 for (a) x = 0–0.2, (b) x = 0.3–0.4 and (c) x = 0.5–0.6. The plot shows the Pawley refinement (red line) to the observed intensity (black line). The blue line shows the difference between observed and calculated intensity. The vertical line symbol with different colors represents the expected positions of Bragg reflections for each phase

Full size image
Table 1 Structural parameters of Tl1-xGexBaSrCaCu2O7 for x = 0–0.6
Full size table

The lattice parameters for the non-doped sample were a = 3.830 Å and c = 12.344 Å. There was no systematic change in lattice parameters a observed in all samples. However, lattice parameter c and unit cell volume for all Ge-doped samples were much smaller than the non-doped sample. The ratio of lattice parameters a/c was used to calculate the internal lattice constraint [30]. As shown in Table 1, the a/c ratio for all doped samples was much higher than the non-doped sample.

Figure 2 displays the SEM micrographs of Tl1-xGexBaSrCaCu2O7 for x = 0, 0.1 and 0.4 samples. The x = 0 sample showed a closed-packed microstructure with some voids. Partial melting was also visible in some parts of the micrograph. The Ge-doped samples (x = 0.1 and 0.4) showed a randomly oriented plate-like structure with more defined grain boundaries compared to the non-doped sample. Ge-doping at x = 0.4 resulted in smaller grain sizes and higher porosity microstructures.

Fig. 2
figure 2

SEM micrographs of Tl1-xGexBaSrCaCu2O7 for (a) x = 0, (b) x = 0.1 and (c) x = 0.4

Full size image

The normalized electrical resistance versus temperature curves of Tl1-xGexBaSrCaCu2O7 for x = 0–0.6 is shown in Fig. 3. All samples demonstrated metallic normal state behavior above Tc-onset. The non-doped sample showed Tc-onset = 100 K and Tc-zero = 75 K. Tc-onset and Tc-zero for all doped samples were between 90 and 94 K, and 67 and 81 K, respectively. The highest Tc-onset (94 K) and Tc-zero (81 K) for all doped samples was for x = 0.4 sample. The x = 0.4 sample also showed a resistive anomaly at 188 K, which is probably associated with a semimetal-metal transition. The doping of Ge suppressed Tc-onset but generally improved Tc-zero.

Fig. 3
figure 3

Electrical resistance versus temperature of Tl1-xGexBaSrCaCu2O7 for x = 0 to 0.6

Full size image

The transition width ΔTc for the non-doped sample was the widest (25 K). All substituted samples exhibited smaller ΔTc (11–16 K) than the non-doped sample except for the x = 0.6 sample. This indicated that Ge-doping generally narrowed the transition width of Tl-1212 as a result of improved homogeneity of transition temperatures for individual superconducting grains. The electrical resistivity at room temperature ρ298 K for x = 0 was 3.31 mΩ-cm. In general, ρ298 K increased as the amount of Ge-doping increased. Further doping of Ge (x = 0.5 and 0.6) substantially increased ρ298 K. Another factor that may have contributed to this result is the formation of impurity phases, which reduced the volume fraction of the Tl-1212 phase as Ge content increased, primarily in higher doping samples. Table 2 lists the Tc-onset, Tc-zero, ΔTc, Tcχ′, Tp and Jc(Tp) of Tl1-xGexBaSrCaCu2O7 for x = 0–0.6.

Table 2 Tc-onset, Tc-zero, ΔTc, Tcχ′, Tp and Jc(Tp) of Tl1-xGexBaSrCaCu2O7 for x = 0–0.6
Full size table

Figure 4 shows the complex AC susceptibility measurements (χ = χ’ + iχ”) of Tl1-xGexBaSrCaCu2O7 for x = 0–0.6. The bulk sample susceptibility transition temperature, Tcχ’ is defined as the temperature at which a sudden drop in the real part of susceptibility occurs. The Tcχ’ also represents the beginning of bulk diamagnetic shielding. The imaginary part of susceptibility, χ” is associated with the nature of flux pinning strength and grain connection. A small peak in the χ” part of susceptibility at higher temperatures is associated with intrinsic losses, while a larger peak at lower temperatures is associated with coupling losses. Our results revealed no intrinsic loss peak in all samples, which could be attributed to the low applied magnetic field, Hac (5 Oe). Besides that, the intrinsic peak is often obscured by the coupling peak due to the strong coupling between the grains [31].

Fig. 4
figure 4

AC susceptibility (χ = χ’ + iχ”) versus temperature of Tl1-xGexBaSrCaCu2O7 for (a) x = 0 to 0.3 and (b) x = 0.4 to 0.6

Full size image

The highest Tcχ’ (91 K) was for the non-doped sample. For all doped samples, Tcχ’ was around 50 to 81 K. In general, Tcχ’ decreased with increasing Ge content showing that Ge suppressed the diamagnetic transition of the Tl-1212 phase. For lower doping samples (x = 0–0.3), Tcχ’ was slightly above Tc-zero, while for higher doping samples, Tcχ’ was below Tc-zero. This showed that samples with lower Ge-doping induced the macroscopic current loops which shield the magnetic field better than samples with a higher doping [32].

Magnetic shielding began at Tcχ’ and full penetration occurs at peak temperature, Tp. The degree of Tp shifting reflects the intergrain coupling of the samples. When Tp shifted to higher temperatures, the intergrain coupling was enhanced. Tp was the highest for the x = 0.3 sample which was 74 K. This indicated that Ge enhanced the intergrain coupling in this value. Tp shifted to lower temperatures for higher doping samples (x = 0.4–0.6), demonstrating that a higher amount of Ge-doping suppressed the intergrain coupling. Higher Ge-doping (x = 0.4) changed the morphology of the Tl-1212 phase (Fig. 2 (c)) by hindering grain growth and increasing porosity. This weakened the intergrain coupling of the Tl-1212 phase as observed in AC susceptibility data. For ceramic superconductors, the intergrain critical current is also primarily related to the Josephson junction rather than the flux pinning energy [33]. Thus, higher doping of Ge also weakened the Josephson current between the grains of Tl-212 phase.

Besides that, the x = 0.1–0.3 samples exhibited a second intergrain losses peak at lower temperatures. Tp for the second intergrain peak was from 53 to 57 K. This demonstrated the existence of other superconducting phases such as TlSr2Ca2Cu3O9 (Tl-1223) and Ca0.3Sr0.7CuO2 (CSCO) as observed in the XRD pattern. Also, a second superconducting transition in the real part of susceptibility was feebly observed in this temperature range. The Bean's model [28, 29] can be used to calculate intergrain critical current density at Tp, Jc(Tp) because the magnitude of Hac equals the penetrated flux at Tp. Using this model, the formula Jc(Tp) = H/(lw)1/2 has been used to approximate the intergrain critical current density at peak temperature, Tp for cuprate-based superconductors [17, 33,34,35]. Jc(Tp) for all samples was between 15 and 18 A cm−2.

Figure 5 shows the Tc-onset, Tc-zero, Tcχ’ and Tp of Tl1-xGexBaSrCaCu2O7 for x = 0–0.6. Tc-onset and Tcχ’ were suppressed when Ge was doped in Tl-1212 phase superconductor. Further doping (x = 0.4–0.6) resulted in an abrupt decrease in Tcχ’ and Tp, indicating that the superconductivity of the Tl-1212 phase was severely suppressed. On the other hand, Tc-zero was enhanced when Ge was doped in the Tl-1212 phase. Besides that, previous studies showed that the existence of GeO2/CuO2 layers due to the Ge substitution at the Cu site resulted in the enhancement of transition temperatures and magnetic properties of Tl-based superconductors [25, 26]. This implied that the GeO2/CuO2 layers were not formed in the Tl-1212 phase.

4 Conclusion

Fig. 5
figure 5

Tc-onset, Tc-zero, ΔTc, Tcχ′ and Tp versus x

Full size image

The effects of Ge-doping on Tl1-xGexBaSrCaCu2O7 for x = 0–0.6 have been studied. Ge doping resulted in forming a single Tl-1212 phase for x = 0 to 0.2 samples. Ge-doping suppressed Tc-onset and Tcχ’ but enhanced Tc-zero and narrowed the transition width ΔTc. The x = 0.3 sample showed the highest Tp indicating the enhancement of the intergrain coupling. The microstructure revealed a decrease in grain size and an increased porosity for higher doping samples. This led to a decrease in Tp, which indicated the weakening of intergrain coupling for higher-doped samples. Further research can be performed by substituting other multivalent metalloid elements with various ionic radii at the Tl-site of Ba and Sr bearing Tl-1212 phase in order to improve the superconducting properties.