Influence of growth temperature on the pinning landscape of YBa₂Cu₃O_7−δ films grown from Ba-deficient solutions

Superconducting cuprate YBa₂Cu₃O_7−δ (YBCO), or other rare earth equivalents REBa₂Cu₃O_7−d (REBCO) coated conductors hold great promise to revolutionize electrical power, generation, distribution and use, which could become more efficient and reliable, facing global challenges related to sustainable energy. In particular, REBCO materials offer a realistic vision for emerging energy and power applications to be operated at different temperature-magnetic field conditions, such as fusion reactors, electric motors, generators or highly efficient power grids [1, 2]

In this framework, many different strategies are adopted for the fabrication of nanocomposites (YBCO films with non-superconducting secondary faces) to induce artificial pinning centres with controlled dimensionality, density and size, able to pin vortices at the desired magnetic field/temperature conditions, using different growth methodologies. Moreover, simple and cost-effective processes able to be scaled for superconducting tape fabrication are strongly required [3, 4].

Chemical solution deposition (CSD) using low-fluorine trifluoroacetate (TFA) precursor solutions appears as a very promising process to fabricate REBCO coated conductors, since high performances can be obtained by a simple non-vacuum low-cost technique using environmentally friendly solutions [5–7].

Among the different parameters that may be changed to optimize the performance of CSD films, one can play with the stoichiometry of the starting solution. In this sense, Ba-deficient compositions have shown improved performances with extremely high critical current density values [8–10]. In a recent publication, we observed that by using low-fluorine Ba-deficient solutions, YBCO c-axis nucleation can be obtained at much lower temperatures (T ∼ 730 °C) than those needed in stoichiometric films, thus significantly extending the epitaxial temperature growth window [8]. The growth temperature plays an important role on the amount and nature of the pinning defects induced within the YBCO matrix, and the associated pinning anisotropy and strength [11, 12]. So, growth temperature can be used to naturally tune the defect pinning landscape of YBCO films without the need to incorporate additional secondary phases. We demonstrated in [8] that by using an optimal growth temperature of 750 °C, which gave a good compromise between percolation (good texture) and pinning performance, we can obtain very high self-field critical current values, 60 MA cm⁻² and 7 MA cm⁻², at 5 K and 77 K, respectively. Moreover, we observed that samples grown at lower temperatures showed smoother J_c(H) dependencies evidencing a higher density of defects in the YBCO matrix. In this contribution, we have carefully analysed the morphology, angular, field and temperature dependence of the critical current of these samples as compared to pristine and nanocomposite YBCO films, in order to elucidate the nature and strength of pinning centres naturally induced when growing at low temperatures. The influence of the precursor solution and growth temperature on the doping level of the films has also been evaluated.

Low-fluorine Ba-deficient precursor solutions were prepared by dissolving barium, copper acetates, and yttrium TFA with a ratio of Y:Ba:Cu equal to 1:1.5:3 in a mixture of organic solvents consisting of propionic acid (30%) and methanol while heating at 30 °C [8]. Precursor solution was deposited by spin coating with an angular velocity of 6000 rpm, angular acceleration of 6000 rpm s⁻¹ and spinning duration time of 2 min on 5 × 5 mm² (100) LaAlO₃ single-crystal substrates. The coatings were pyrolyzed up to 310 °C in a humid O₂ atmosphere and crystallized with heating ramp of 25 °C min⁻¹ up to the desired growth temperature (810 °C and 750 °C) during 3 h. Afterwards we performed an oxygenation process at 500 °C. For comparison purposes, a stoichiometric pristine YBCO film and a nanocomposite (12% BaHfO₃ (BHO) nanoparticles) were grown at 810 °C following the same process, but by using a standard stoichiometric (Y:Ba:Cu-1:2:3) TFA solution with yttrium, barium and copper TFA [13]. The thickness of all films analysed was t∼ 200 nm.

The structural quality of the samples was determined by x-ray diffraction (XRD) using a 2D Bruker AXS general area detector diffraction system operating with Cu K_α and scanning electron microscopy (SEM) using an FEI Quanta 200 FEG system. Transport measurements were performed using a four-probe configuration in patterned tracks of 30 µm of width and 250 µm length, using a Quantum Design physical property measurement system (PPMS). Angular curves of the critical current density, J_c (θ), were obtained at temperatures ranging from T= 5–77 K and magnetic fields from µ₀ H= 0–9 T, considering the orientations θ = 90° for H//ab and θ = 180° for H//c. The critical current was determined with an electric field criterion of 2 µV cm⁻¹. Hall effect measurements were carried out also by the PPMS at 300 K using the Van der Pauw configuration [14] in the field range of −9 T to 9 T. The charge carrier density, n_H, was calculated at 300 K as n_H =1/(qR_H); where q is the charge of the electron and R_H the Hall coefficient [15]. High-angle annular dark-field (HAADF) scanning transmission electron microscopy (STEM) imaging was carried out on an FEI Titan ChemiSTEM probe Cs-corrected transition electron microscopy (TEM) operated at 200 kV [16].

We compare four YBCO films; two stoichiometric YBCO samples (a pristine standard film (Stoich-810) and a nanocomposite (Stoich-NC-810) both grown at 810 °C), and two Ba-deficient samples (a film grown at 810 °C (LowBa-810), and another one grown at the optimized temperature of 750 °C (LowBa-750)).

Figure 1 shows the critical current density as a function of the magnetic field H//c for these four samples at 77 K, in a log-log scale. The first thing that one can observe is that the J_c (H) curves obtained for pristine stoichiometric and the Ba-deficient sample grown at 810 °C are almost identical, indicative of a minor effect of the precursor solution used. It should be noted however that the growth temperature has a strong effect on the pinning performance. LowBa-750 shows a smother J_c(H) dependence, similar to what we obtain in the nanocomposite film grown at 810 °C by introducing artificial pinning centres (BHO nanoparticles). We observe that efficient pinning sites can be naturally induced within the YBCO matrix by reducing the growth temperature, without the need to artificially introduce secondary phases. It should be noted that although both Stoich-NC-810 and LowBa-750 samples show a similar performance, there is a crossover of the J_c(H) curves at ∼0.3 T. LowBa-750 shows a better self-field critical current density but a slightly faster J_c decay with the magnetic field, indicating that different defects are induced in the two samples.

Zoom In Zoom Out Reset image size

A comprehensive discussion of artificially engineered pinning sites induced in YBCO nanocomposites with different nature and amount of nanoparticles (BHO, BaZrO₃, Ba₂YTaO₆ and Y₂O₃) can be found in [17]. In this work we aim to analyse the superconducting performance and associated pinning landscape in pristine Ba-deficient YBCO films grown at low temperature. We use a stoichiometric pristine and a typical nanocomposite film grown at 810 °C for comparison purposes.

Figure 2(a) shows a SEM image obtained for LowBa-750 were a uniform and flat surface with no a–b needle-shape grains are observed. XRD θ–2θ scan shown in figure 2(b) indicate a good epitaxial growth with c-axis YBCO (00 l) peaks. It should be noted that a small contribution of Y₂Cu₂O₅ reflections can be identified, consistent with the barium deficiency imposed by the Ba-deficient solution.

Zoom In Zoom Out Reset image size

Figure 3 show the angular dependence of J_c (θ) obtained at 77 K–1 T, 65 K–5 T and 5 K–9 T for Stoich-810, Stoich-NC-810 and LowBa-750. By comparing the nanocomposite and the pristine sample grown at 810 °C (black open and closed symbols, respectively) we observe that effect of adding nanoparticles results in a broader ab-peak (θ =90°), attributed to the anisotropic pinning of stacking faults and higher values of J_c in the whole θ range, with a reduction of the anisotropy associated with the isotropic pinning of nanostrained regions induced at the partial dislocations at different fields and temperatures [18–20]. The J_c (θ) curves obtained for the pristine Ba-deficient sample grown at 750 °C (red symbols) show similar characteristics than that of the nanocomposite but with even much broader ab-peaks.

Zoom In Zoom Out Reset image size

In general, in YBCO films the anisotropic pinning at H//ab mainly comes from the interplay between intrinsic pinning and stacking faults. The width of the J_c (θ) ab-peak can be associated to the trapping angle θ_T that limits the staircase regime where vortices accommodate to the anisotropic defects with kinks and parallel segments [21, 22]. Beyond θ_T vortices will be straight and take the direction of the applied field, thus being unaffected by the correlated nature of the anisotropic pinning.

Figure 4 shows the normalized ab-peak measured at different magnetic fields for the LowBa-750 sample. Note that the width of the peak decreases as the field increases tending to a saturation at high fields. Field dependence of the trapping angle obtained for Stoich-810, Stoich-NC-810 and LowBa-750 at 77 K and 25 K is shown in figure 5. For both temperatures, the values of the trapping angle obtained for Stoich-NC-810 and LowBa-750 samples are much higher than those for Stoich-810, indicating that one can promote the formation of stacking faults either by introducing nanoparticles in a nanocomposite film or by reducing the growth temperature in a pristine film. It is worth noting that at low temperatures (figure 5(b)) the pristine stoichiometric sample grown at 810 °C (Stoich-810) shows a fast θ_T decrease with H and becomes constant above ∼4 T. Field independent θ_T can be associated with a single vortex pinning regime mainly dominated by intrinsic pinning with a very large matching field [21]. Samples with a high number of staking faults (Stoich-NC-810 and LowBa-750) do not reach the θ_T saturated region fingerprint of an extension of the pinning region dominated by stacking faults.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Figure 6 shows cross-section HAADF STEM images obtained for LowBa-750 at different magnifications, where we clearly observe the presence of a huge density of stacking faults as compared with a stoichiometric pristine film, were for a 200 × 200 N m² cross-section image less than 20 stacking faults are found [18].

Zoom In Zoom Out Reset image size

To better elucidate the nature and strength of pinning sites induced in the Ba-deficient pristine YBCO film grown at 750 °C we analyse the contribution of different type of defects at different regions of the H–T phase diagram. A first classification of pinning centres may be performed considering their temperature dependence of the critical current density, J_c(T). Weak pinning decay as an exponential function of T (equation (1)) [23] whereas strong pinning follows an exponential function of T² (equation (2)) [24]

$$\begin{align}J_{\text{c}}^{\,{\text{weak}}}\left( T \right) = J_{\text{c}}^{\,{\text{weak}}}\left( 0 \right){\text{ exp}}\left( { - \frac{T}{{{T_0}}}} \right)\end{align} \tag{ 1 }$$

$$\begin{align}J_{\text{c}}^{\,{\text{strong}}}\left( T \right) = J_{\text{c}}^{\,{\text{strong}}}\left( 0 \right){\text{ exp}}\left[ { - 3{{\left( {\frac{T}{{{T^{\,*}}}}} \right)}^2}} \right]\end{align} \tag{ 2 }$$

were $J_{\text{c}}^{\,{\text{weak}}}$(0) and $J_{\text{c}}^{\,{\text{strong}}}$(0) are the contributions to J_c at 0 K and T₀ and T* the characteristic pinning energies of weak and strong pinning defects, respectively.

Another classification may be performed considering their shape. Isotropic pinning sites, with a pinning length that do not depend on the magnetic field orientation, and anisotropic pinning sites which are correlated along a certain field direction. These two kind of defects may be identified by using the Blatter scaling approach on a series of J_c(θ,H) curves plotted as a function of the effective magnetic field, H_eff (equation (3)) [25]

$$\begin{align}{H^{\,{\text{eff}}}} \sim H{\left[ {{{\cos }^2}\theta + {\gamma ^{ - 1}}{{\sin }^2}\theta } \right]^{1/2}}\end{align} \tag{ 3 }$$

were θ is the angle between the magnetic field and the c-axis and γ the anisotropy parameter.

By combining equations (1)–(3) and adjusting angular, field and temperature dependence of J_c(H,T,θ), assuming a direct summation of pinning centres, one can determine and quantify three different pinning contributions: isotropic-weak (I-W), isotropic-strong (I-S), and anisotropic-strong (A-S) [26].

Figures 7 (a)–(c) plot log J_c versus T² at μ₀ H//c = 1 T, μ₀ H//c = 9 T and μ₀ H//ab = 9 T. Linear fits to these plots correspond to strong pinning centres (equation (2)) which control J_c in the intermediate temperature regime. At low temperatures for H//c the J_c values of Stoich-810 and Stoich-NC-810 deviate from the linear fit below ∼30 K at 1 T and 20 K at 9 T pointing out a region dominated by weak pinning defects [18]. In the case of LowBa-750 the linear dependence is observed down to lower temperatures, indicating that strong pinning centres control the pinning performance in a wider temperature region: 5 K–60 K at 1 T and 15 K–55 K at 9 T. At this low-intermediate temperature regime this sample shows the best J_c performance. At high temperatures, a fast decay of J_c is obtained in all the samples due to thermal fluctuations close to T_c. As pointed out above a crossover of the J_c(T) curves of Stoich-NC-810 and LowBa-750 is obtained at 68 K–1 T and 65 K–9 T H//c which may be associated to a slightly higher irreversibility line of the former [18].

Zoom In Zoom Out Reset image size

In figures 7(d)–(f) we show the weights of the three basic pinning contributions of J_c(T) for LowBa-750. At intermediate fields, μ₀ H//c = 1 T, the most important contribution comes from I-S pinning which can be associated to nanostrain induced to accommodate the stacking faults [20]. At higher fields, μ₀ H//c = 9 T, we observe that an important contribution of I-W defects appears at low temperature. These defects may be associated to Cu–O cluster vacancies appearing at the intergrowths [27]. At μ₀ H//ab = 9 T the A-S contribution coming from stacking faults is clearly governing the pinning performance.

In addition to the effect of the vortex pinning landscape, the superconducting properties of cuprates, strongly depend on the carrier concentration. In particular, the critical current density can be significantly enhanced by achieving the overdoped state due to an increase of the condensation energy [15, 28]. Figure 8 shows the average values of the self-field critical current density as a function of their carrier density, n_H, obtained for several pristine stoichiometric films grown at 810 °C (four samples) and Ba-deficient films grown at 810 °C (three samples) and 750 °C (four samples). The n_H value obtained for standard films grown at 810 °C with stoichiometric TFA solution with J_c ^sf ∼ 4–5 MA cm⁻² is the one expected for optimally doped films n_H ∼ 8.10²¹ cm⁻³ [29]. For the films grown with Ba-deficient solution however, we observe that the values of n_H appear at the overdoped region. Films grown at 810 °C show similar J_c ^sf values than those obtained for standard stoichiometric but slightly higher n_H∼ 11.10²¹ cm⁻³. Interestingly, Ba-deficient films grown at 750 °C show the highest charge carrier density with n_H∼ 14 × 10²¹ cm⁻³ and large values of J_c ^sf ∼ 6–7 MA cm⁻² can be obtained. The enhancement of the carrier density, and associated J_c ^sf, can be correlated with a high density of stacking faults, suggesting that the induced disorder and nanostrain within the YBCO matrix may enhance the oxygen diffusion thus allowing to easily achieve an underdoped state.

Zoom In Zoom Out Reset image size

The pinning performance of Ba-deficient samples grown at low temperature (750 °C) has been characterised through transport measurements and compared with a standard pristine film and a nancomposite with BHO nanoparticles grown at higher temperature (810 °C). The results show that by reducing the growth temperature very high self-field critical current density values are obtained with smoth J_c(H) dependences similar those obtained in nanocomposites. By measuring the temperature, field and angular dependence of J_c we have found that a highly effective pinning landscape is induced in the films grown at 750 °C mainly associated to the formation of a large number of stacking faults, observed by TEM images. Moreover, those films show very high values of J_c ^sf which may be associated to a large carrier concentration at the overdoped state. Consequently, the use of Ba-deficient solutions showing a large c-axis epitaxial YBCO growth window appears as a very appealing way to tune both pinning landscape and the carrier density of YBCO pristine films and thus maximize their superconducting performance. It is worth pointing out that the addition of nanoparticles in the YBCO matrix not only generates pinning due to induced stacking faults and the associated strain but it is also possible to achieve a synergetic effect, with additional pinning coming from small nanoparticles [30, 31]. Thus, the combined effect of Ba-deficient precursor solution, low temperature growth and the addition of small nanoparticles may boost the pining performance of the films.

This work has been carried out within the framework of the EUROfusion Consortium and has received funding from the Euratom Research and Training Programme 2014–2018 and 2019–2020 under Grant Agreement No. 633 053. The views and opinions expressed herein do not necessarily reflect those of the European Commission. TEM analysis was funded from the EU Horizon 2020 research and innovation program under grant agreement 823717—ESTEEM3. The authors acknowledge financial support from Spanish Ministry of Economy and Competitiveness through the 'Severo Ochoa' Programme for Centres of Excellence in R&D (SEV-2015-0496 and CEX2019-000917-S), SuMaTe RTI2018-095853-B-C21, co-financed by the European Regional Development Fund), the Catalan Government with Grant 2017-SGR-1519 and the EU COST action NANOCOHYBRI CA16218. We also acknowledge the Scientific Services at ICMAB. J A like to thank the UAB PhD program in Materials Science.

All data that support the findings of this study are included within the article (and any supplementary files).