Gd-123 is regarded as a promising material in the field of high temperature superconductors HTSCs. It possesses high superconducting transition temperature (around 90 K) [1] , high critical current density values at high magnetic fields [2] [3] , high ability to trap the magnetic field up to 2.0 T at 77 K [4] , relative low crystalline anisotropy and well developed pinning properties in the magnetic field [5] . However, Gd-123 showed a rapid decrease in its J_c due to the movement of the non-superconducting vortices, as both the temperature and the magnetic field increase. An effective way to overcome this problem is the enhancing of its flux pinning capabilities by nanoparticles addition [6] - [8] . This enhancement of the flux pinning consequently improves the critical current density [9] - [11] by causing a strong interaction between the flux line networks and the nanosized magnetic particles. This is possible if the size of these nanoparticles is higher than the coherence length ξ and lower than penetration depth λ. A wide variety of HTSCs were pinned [12] - [20] and most of the results showed that the superconducting parameters were improved at low amounts of nanosized additions. On the other hand, high amounts of addition reduced the formation of superconducting phase.

Ion beam analysis (IBA) including RBS and PIXE has many contributions in the field of the high temperature superconductors. PIXE technique has the advantage that the cross section for X-ray production is large and the background contribution is low. So, it is a highly sensitive method for the multi-elemental analysis and a large number of elements may be seen simultaneously. In addition, it may be a very useful complementary technique when combined with other ion beam methods. This high precision allows us to inspect any contamination during the preparation process. Furthermore, RBS is essential for fabrication of thin-film specimens, performing fundamental irradiation effect studies [21] . It is well known that the physical properties of high-temperature superconductors are strongly dependent on the elemental content, especially O-stoichiometry, and then RBS plays a very important role in sample investigation and studies. Many literatures have been conducted for studying PIXE and RBS for a large family of superconductors [22] - [25] . These studies found that the real stoichiometry, as wt%, obtained from PIXE spectra was nearly close to the starting values. Moreover, the oxygen-content, obtained by non-Rutherford backscattering using 3 MeV H⁺, was found to be very close to the nominal ones.

In this work, we investigate the effect of ZnFe₂O₄ and CoFe₂O₄ addition to the Gd-123 phase. The elemental composition is determined by PIXE and RBS techniques, while the phase analysis, volume fraction, theoretical density and porosity are determined from XRD analysis. The superconductivity investigation is conducted via electric resistivity and critical current density measurements. The variations of T_c and J_c with x for both nanoferrite additions are investigated and discussed.

Nanosized ZnFe₂O₄ and CoFe₂O₄ powder were prepared by chemical Co-precipitation method as discussed in previous work [26] [27] . Superconducting samples of type (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ and (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ, 0.0 ≤ x ≤ 0.4 wt%, were prepared by the conventional solid-state reaction technique. First, the starting materials Gd₂O₃, BaCO₃ and CuO (purity ≥ 99.9) were crushed in an agate mortar and were sifted using a 125 μm sieve to get a homogeneous mixture. Then, the powder was subjected to calcinations process in air at 840˚C and 880˚C for 24 h each. The resultant powder was ground and sifted, and x wt% of nanosized ZnFe₂O₄ or CoFe₂O₄ were added to the resulting powder. The powder was mixed for one hour to ensure the homogenous distribution of nanoferrite inside the sample. Subsequently, the mixed powder was pressed in a disc form (1.5 cm in diameter and about 0.3 cm in thickness). Afterwards, discs were sintered in air at 930˚C with a heating rate of 4˚C/min, and held at this temperature for 24 h. Then, the samples were cooled by a rate of 1˚C/min down to 450˚C and kept at this temperature for 10 h under oxygen flow to control the oxygen-content of the final compounds.

The prepared samples were characterized by XRD using Bruker D8 advance powder diffractometer with Cu-K_α radiation (λ = 1.54056 Å) in the range 4^ᵒ ≤ 2θ ≤ 70^ᵒ. The grain morphology of the samples was identified using Jeol scanning electron microscope JSM-5300, operated at 25 - 30 kV, with a resolution power of 4 nm. The relative volume fractions of Gd-123 and BaCuO₂ phases are estimated by comparing the intensities of the whole pattern [28] , according to the relations

where is the peak intensity of the present phases above the baseline.

The bulk densities (r) of (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ and (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ samples were determined by the immersion method using Archimedes principle with toluene as a solvent. By measuring the mass of the sample and the mass of the displaced liquid, with density ρ_toulene, the density of the sample can be determined using the equation:

The theoretical density and the porosity P are given by:

where N, M, N_A and V represent the number of atoms per unit cell = 1, molecular weight of the sample, Avogadro’s number and the volume of the orthorhombic unit cell respectively and is given by: (where a, b and c represent the lattice parameters).

The elemental composition of the samples was measured by the conventional vacuum PIXE. The 1.7 MV tandem accelerator of the Lebanese Atomic Energy Commission [29] was used to deliver 3 MeV proton beam on the samples with 1 μC of fluence. For ion beam analysis, a mass of 0.3 g of each sample was grinded very well to ensure homogeneity and hence a thick target pellet of approximate dimensions 1 × 1 × 0.2 cm³ was formed. The target chamber contains Silicon drifted detector (SDD) formed of high purity silicon with a very low leakage current leading to high energy resolutions (down to 123 eV for Mn K_α wavelength). The SDD allows the detection of elements with Z > 10. In order to obtain one spectrum for all the elements which compose the sample, an Aluminum filter of 100 μm thickness was placed in front of SDD X-ray absorber. The real stoichiometric amount of the elements can be calculated using the following equation:

where is the real element-content as wt%, M_rT is the total molecular weight and M_rx is the molecular weight of the element. Simultaneously, elastic backscattering measurements were taken under normal incident beam. A partially depleted PIPS (Passivity Implanted Planar Silicon) detector from Canberra, with 14 keV energy resolutions and 25 mm² active areas, detected the backscattered particles of the 3 MeV proton beam, at a scattering angle θ of 165˚ and solid angle of 5.45 × 10⁻³ sr.

The electrical resistivity of the prepared samples was measured by the conventional four-probe technique from room temperature down to the zero superconducting transition temperature (T₀) via closed cryogenic refrigeration system employing helium gas as a working medium. The samples had the shape of parallelepipeds of approximate dimensions 15 × 2 × 2 mm³, and the connections of the copper leads with the sample were made using a conductive silver paint. The temperature of the sample was monitored by a Chromel versus Fe-Au thermocouple and stabilized with the aid of a temperature controller to within ±0.1 K. A typical excitation of 1 mA was used to avoid heating effects on the samples. The potential drop across the sample was measured using a Keithley-181 digital nano-voltmeter. The voltage was determined by taking the average of the values measured with the normal and the reverse directions of applied current to eliminate the thermo-electric voltages of the junction leads. The transport critical current density was measured by the conventional four probe technique at 77 K using liquid nitrogen flow.

The XRD patterns of (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ and (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ (x = 0.00, 0.01, 0.06 and 0.40 wt%) are shown in Figure 1. All of the diffraction patterns peaks are indexed by the main orthorhombic phase with a space group P_mmm. An insignificant peak belonging to the non-superconducting phase BaCuO₂ appears at 2θ = 30˚ [30] . The values of the lattice parameters a, b and c are calculated and are listed in Table 1 for different values of x. The lattice parameters a, b and c do not show systematic variation with respect to the pure sample. These results indicate clearly that lattice parameters a, b, and c are not affected by nanoferrite addition which implies that any substitution by the nanoferrites does not occur. So the nanoferrites do not enter inside the crystal sites and sit only at the grains boundaries. This means that nanoferrites addition serves as interface between

the superconducting matrix and does not affect the structure of HTSCs. Similar results were observed by Awad [31] for studying the effect of MgO nanoparticles addition into (Cu_0.25Tl_0.75-1234) phase, and by Roumié et al. [13] for studying the effect of Fe₂O₃ nanoparticles addition into the Bi_1.8Pb_0.4Sr₂Ca₂Cu₃O_10+δ phase. The experimental densities, and porosity of (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ and (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ (0.00 ≤ x ≤ 0.40 wt%) are listed in Table 1. It is clear that P increases up to x = 0.06 wt% and then it decreases with further increase in x for (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ phase. Whereas P increases for 0.00 ≤ x ≤ 0.40 wt% for (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ phase.

The porosity of both phases varied from 19% to 30%, confirming the porous nature of ceramic materials. The relative volume fractions were calculated using all peak intensities of Gd-123 and BaCuO₂ phases according to Equations (1) and (2) and their values with x of both additions are listed in Table 1. It is obvious that the relative volume fraction of the main phase, Gd-123, increases with increasing nanosized ZnFe₂O₄ addition up to 0.06 wt% and then it decreases with further increase in x. On the other hand, the increase in the relative volume fraction of the impurity phase BaCuO₂ is observed for x < 0.06 wt%. The decrease of the relative volume fraction suggested that the presence of nanoparticles can affect the viscosity of the transient liquid formed at the reaction temperature and its homogeneity which in turn affect the formation rate of the Gd-123 phase. Also, the ability of nanosized ZnFe₂O₄ particles to prevent, to some extent, the spatial phase growth may be another reason. The enhancement rate of the relative volume fraction of Gd-123 added with nanosized ZnFe₂O₄ is 1.06%, while the retardation rate is 1.76%. Table 1 also shows that nanosized CoFe₂O₄ addition increases the relative volume fraction of the main phase from 97.36% to 98.33% as x varies from 0.0 to 0.01 wt% and then it decreases with further increase in x, but its value still greater than that for the free sample up to x = 0.06 wt%. This means that higher concentrations of nanosized CoFe₂O₄ (<0.06 wt%) have a significant role in changing the reaction rate which slightly retards the Gd-123 phase formation. This is attributed to the ability of the nanosized CoFe₂O₄ to prevent the spatial phase growth. The enhancement rate of relative volume fraction of Gd-123 phase added with nanosized CoFe₂O₄ is 0.99%, while the retardation rate is 2.12%.

Typical PIXE spectra of the samples GdBa₂Cu₃O_7−δ, (ZnFe₂O₄)_0.1 GdBa₂Cu₃O_7−δ and (CoFe₂O₄)_0.1 GdBa₂- Cu₃O_7−δ are shown in Figures 2(a)-(c), respectively. The elements Cu, Zn and Fe are identified by the K X rays lines, whereas Gd and Ba are identified by L X-rays lines. No contaminations are observed in the spectra, which verify the high purity of all investigated samples. Co was not detected in the spectra, this is probably due to the overlapping of its K-X ray lines (K_α = 6.924 keV) with the L-X-ray lines of heavier and much abundant Gd (L_β1 = 7.102 keV). The real elements-content is calculated according to Equation (6) and listed in Table 2 for (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ and (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ.

It is clear that the stoichiometric ratio of the different elements are nearly close to the stoichiometric starting compositions. Taking into account the statistical and fit errors, the uncertainties of the PIXE measurements are: ±3% for Gd, ±5% for Ba, ±8% for Cu, +10% for Zn and Fe. Combining these errors, the indirect PIXE determination of O-stoichiometry by completing the stoichiometry to 100 wt% of different constituents that appeared in PIXE spectra, is inaccurate and differs from the expected value (as shown in Table 2). The Ba stoichiometric

ratios are slightly higher than their nominal values resulting from the formation of BaCuO₂ as impurities. These results are consistent with those obtained from XRD measurements. The error percentages of Zn and Fe elements are higher than that of the rest elements. This is attributed to the superposition of Zn and Fe peaks with Cu peaks. On the other side, the stoichiometric ratios of Zn and Fe show a systematic and continuous increase up to x = 0.1 wt% as shown in Figure 3.

Figures 4(a)-(c) show the RBS spectra of the samples GdBa₂Cu₃O_7−δ, (ZnFe₂O₄)_0.1 GdBa₂Cu₃O_7−δ and (CoFe₂O₄)_0.1 GdBa₂Cu₃O_7−δ, respectively using 3 MeV proton elastic backscattering by fitting a non-Rutherford backscattering cross-section for O [22] in the SIMNRA simulations. The actual O-content, as wt% and stoichiometric ratios, for all samples, was determined and listed in Table 3. It was found that the actual O-content is very close to the starting nominal one. The deviation between PIXE and RBS results for O as wt% is also listed in Table 3. It is clear that the oxygen content determined from RBS measurements is closer to the nominal values than that determined from PIXE measurements. This observation is not surprising since RBS is more accurate in determining the O-content. However, the deviation between PIXE and RBS results is very acceptable (around 4%). For both additions, the oxygen content remains constant up to x = 0.1 wt% and then increases slightly with x. This is consistent with the invariance of the lattice parameters determined from XRD measurements.

The temperature dependence of the electrical resistivity for GdBa₂Cu₃O_7−δ, (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ (x = 0.00, 0.01, 0.06 and 0.40 wt%) and for (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ (x = 0.00, 0.06 and 0.40 wt%) are shown in Figure 5. All samples behave like a metal in the normal state followed by a superconducting transition at T_c, and zero resistivity at temperature T₀. The normal metallic-like behavior above T_c can be interpreted by the liquid model that provides a natural explanation of anomalous features of transport properties in several high-temper- ature superconductors. It reflects the spin charge separation in CuO₂-planes that results in the longitudinal transport relaxation rate 1/τ ~ T [32] . A small curvature in the electrical resistivity appears above the superconducting transition temperature that characterizes the superconducting Cooper-pairs fluctuations [33] or the opening of the spin-gap in high-temperature superconductivity [34] [35] due to magnetic impurities. The electrical resistivity data of the prepared samples are well fitted, in the temperature range from 300 K down to 2 T_c, according to equation:

The room temperature electrical resistivity, residual resistivity and resistivity temperature coefficient for different x wt% are listed in Table 4. It is well known that is an indicator for the disorder of cations and Oxygen vacancies which increase the number of scattering centers [36] and is an indicator for the purity of the sample. It is clear that both and of (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ decrease with the increase of nanosized ZnFe₂O₄ addition from x = 0.0 wt% to x = 0.06 wt% and then increase for x > 0.06 wt%. On the other side, and of (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ slightly decrease as x increase from 0.0 to 0.01 wt% and then increase with further increase in x. The decrease in and is an indication to the reduction of the grain boundaries and defects and the improvement in the quality and purity of the samples. Whereas, the increase of and may strongly related to the impurities and scattering of grain boundaries. These results are consistent with those obtained through XRD and relative volume fraction percentage calculations. It is clear that slightly changes with x for (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ and (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ. So, nanosized ZnFe₂O₄ and CoFe₂O₄ do not affect the charge carriers-concentrations because the nanosized ZnFe₂O₄ and

CoFe₂O₄ do not enter the crystal structure of Gd-123 phase. The increase in with x > 0.06 and 0.02 wt% for (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ and (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ, respectively could be due to the unsuspected temperature-dependent scattering (magnetic impurities scattering) contribution by nanosized ZnFe₂O₄ and CoFe₂O₄ [37] .

The superconducting transition temperature T_c, which displays the superconducting transition within the grains, is determined as the temperature corresponding to the crest in dρ/dT curve [38] . In addition, the values of T_c for different nanosized ZnFe₂O₄ and CoFe₂O₄ additions are shown in Figure 6. It is obvious that for (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ, T_c increases from 91.77 K to 97.64 K as x varies from 0.0 to 0.06 wt%, then it decreases for x > 0.06 wt%. The increases of T_c with x can be explained due to the increase in the volume fraction of Gd-123 and the decrease of porosity up to x = 0.06 wt%. As a matter of fact, the suppression of T_c is most likely due to the trapping of mobile free carriers, pair-breaking due to spin interaction [39] [40] and the non-uniform distribution of nanoparticles at grain-boundaries of the host Gd-123 superconductor which increases the grain boundaries resistance. The enhancement rate of T_c for Gd-123 phase added with nanosized ZnFe₂O₄ is 6.5%, while the retardation rate is 6.8%. In the other hand, T_c decreases from 91.77 K to 78 K for (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ as x increases from 0.0 to 0.4 wt%. The suppression of T_c is most likely due the progressive decoupling of the superconducting grains with microscopic in homogeneities such as secondary phases, cracks, and grain boundaries that are non-uniformly filled with the nanoparticles [27] . This behavior can also be interpreted as a result of the ferromagnetism role of nanosized CoFe₂O₄ which would cancel the supercurrents and destroy the long-range coupling of Cu ion in the CuO layer. The retardation rate of T_c for Gd-123 phase added with nanosized CoFe₂O₄ is 15%.

The transport critical current density J_c, for all samples, was measured using a criterion of 1 mV/cm. The variation of J_c for (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ and (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ with x wt% is shown in Figure 7. It is

clear that, for (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ, J_c increases as x increases from 0.0 to 0.06 wt% and then it decreases with further increases in x. Whereas, J_c is enhanced with x up to 0.01 wt% for (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ and then it decreases with further increase in x, but its value remain greater than that for the free sample until x = 0.04 wt%. The increase in J_c may be attributed to the increase in the volume fractions of Gd-123 phase or the diffusion of nanosized ZnFe₂O₄ and CoFe₂O₄ in the added samples over the pore surfaces, grain boundaries and twin boundaries. This can increase flux pinning and coupling between grains, leading to increase the critical current density. While, the decrease in J_c is attributed to the suppression of volume fractions of Gd-123 phase. Another probable reason is the large agglomerations between the grains and not through the boundaries and voids due to the excessive addition of nanosized ZnFe₂O₄ and CoFe₂O₄ nanoferrites which deteriorate the intergranular critical current density [41] . The samples added with 0.06 wt% nanosized ZnFe₂O₄ and 0.01 wt% nanosized CoFe₂O₄, increase J_c by 60.6% and 40.5%, respectively. Therefore the addition of nanosized ZnFe₂O₄ is more significant than that of nanosized CoFe₂O₄ in improving the flux pinning and critical current density of the prepared samples. This result could be related to the ferromagnetic properties of nanosized CoFe₂O₄ which contributes to the breaking of Cooper pairs representing the supercurrent [42] .

The PIXE analysis for (ZnFe₂O₄)_xGdBa₂Cu₃O_7−δ and (CoFe₂O₄)_xGdBa₂Cu₃O_7−δ superconducting phases, with 0.0 ≤ x ≤ 0.4 wt%, has successfully determined the actual contents of Gd, Ba and Cu which are very close to the nominal contents. Zn and Fe showed a systematic increase with x, while Co was not detected due to the overlapping of its KX-rays with the LX-rays of Gd. On the other hand, the PIXE technique has not successfully determined the O-stoichiometry. The O-stoichiometry obtained by non-Rutherford backscattering using 3 MeV proton beam was very close to the nominal values and its value increased for high additions. T_c increased with increasing ZnFe₂O₄ content up to x = 0.06 wt% and then decreased for further increase in x. This behavior indicates that the optimal ZnFe₂O₄ addition is x = 0.06 wt% leading to a best enhancement in T_c and J_c attributed to the enhancement of the volume fraction of Gd-123 phase and the reduction in the voids between grains. The addition of CoFe₂O₄ caused a deterioration in T_c and J_c for x > 0.01 wt% resulting from the ferromagnetic nature of CoFe₂O₄ which may destroy the long-range coupling of Cu ion in the CuO layer.

This work was performed in the Superconductivity and Metallic-glass lab, Physics Department, Faculty of Science, Alexandria University, Alexandria, Egypt with cooperation with the accelerator Laboratory, Lebanese Atomic Energy Commission, CNRS, Beirut, Lebanon.

Hadi Basma,Salem Marhaba,Mohamad Roumié,Mohamad Albast,Ramadan Awad,Ali Abualy, (2015) Ion Beam Analysis and Electric Properties of GdBa₂Cu₃O_7-δ Added with Nanosized Ferrites ZnFe₂O₄ and CoFe₂O₄. Materials Sciences and Applications,06,828-840. doi: 10.4236/msa.2015.69085