www.acsami.org Research Article Combinatorial Screening of Cuprate Superconductors by Drop-On- Demand Inkjet Printing Albert Queraltó,* Juri Banchewski, Adria ̀ Pacheco, Kapil Gupta, Lavinia Saltarelli, Diana Garcia, Núria Alcalde, Cristian Mocuta, Susagna Ricart, Flavio Pino, Xavier Obradors, and Teresa Puig* Cite This: ACS Appl. Mater. Interfaces 2021, 13, 9101−9112 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information ABSTRACT: Combinatorial and high-throughput experimenta- tion (HTE) is achieving more relevance in material design, representing a turning point in the process of accelerated discovery, development, and optimization of materials based on data-driven approaches. The versatility of drop-on-demand inkjet printing (IJP) allows performing combinatorial studies through fabrication of compositionally graded materials with high spatial precision, here by mixing superconducting REBCO precursor solutions with different rare earth (RE) elements. The homogeneity of combinatorial Y1−xGdxBa2Cu3O7 samples was designed with computational methods and confirmed by energy-dispersive X- ray spectroscopy (EDX) and high-resolution X-ray diffraction (XRD). We reveal the advantages of this strategy in the optimization of the epitaxial growth of high-temperature REBCO superconducting films using the novel transient liquid-assisted growth method (TLAG). Advanced characterization methods, such as in situ synchrotron growth experiments, are tailored to suit the combinatorial approach and demonstrated to be essential for HTE schemes. The experimental strategy presented is key for the attainment of large datasets for the implementation of machine learning backed material design frameworks. KEYWORDS: combinatorial chemistry, high-throughput experimentation, superconducting materials, inkjet printing, transient liquid-assisted growth, cuprate superconductors, chemical solution deposition ■ INTRODUCTION High temperature superconducting (HTS) materials are to Y, which contributes to broadening the magnetic field region for applications (i.e., high irreversibility line), regarded as key players toward an effective renewable energy consequently, enhancing current densities (Jc) and pinning forces (Fp) at relevant temperatures. For instance, values of Jc > transition. The unique properties of superconductors (e.g., 7 MA cm−2 (77 K, B = self-field, on STO) and maximum Fp ∼ large current transport without losses) and the development of 16 GN m−3 (77 K, B = 2 T, on STO) have been reported for novel strategies for high power managing and transportation GdBCO and BaHfO3 nanoinclusions7 or Jc > 6 MA cm−2 (77 based on HTS are foreseen to pave the way toward cutting- K, B = self-field, on LAO) and Fp ∼ 28 GN m−3 (77 K, B = 1.5 edge and energy-efficient technologies in sectors such as clean T, on LAO) for BaHfO3-doped SmBCO.8 Despite the good electrical energy (cables, transformers, motors, generators, properties exhibited in some cases, the exploration of other RE fusion reactors), biomedicine and chemistry (magnetic is still scarce, sometimes limited by stability or synthesis resonance imaging), scientific instrumentation (ultrahigh complexity of some REBCO phases. The focus has been magnetic fields), or transportation (hybrid/electric airplanes shifted in recent years to RE mixing such as Y/Gd,6,9,10 Gd/ and ships).1−4 Yb,11,12 Y/Sm,13,14 and Dy/Gd,15 aiming to overcome such Superconducting REBa2Cu3O7 (REBCO, RE = rare earth) predicaments, as well as to shed light on the right combination films and coated conductors (CC) possess an inherent that enables best superconducting properties. advantage due to the high performances in a significantly broad spectrum of the magnetic field−temperature (B−T) diagram, hence showing promising properties for wide market Received: October 7, 2020 implementation.5 Due to historical reasons, YBCO has been Accepted: December 23, 2020 the most investigated REBCO superconductor.6 Nevertheless, Published: February 12, 2021 a wide range of RE ions that can replace Y are driving the current research interest. The main motivation is the higher critical temperature (Tc) for RE such as Gd or Sm as compared © 2021 The Authors. Published by American Chemical Society https://dx.doi.org/10.1021/acsami.0c18014 9101 ACS Appl. Mater. Interfaces 2021, 13, 9101−9112 ACS Applied Materials & Interfaces www.acsami.org Research Article The initial push of the superconductor industry has been kinetic phase diagram of REBCO films, contributing to the made with several film deposition techniques to achieve understanding of the kinetic influence of the growth process epitaxial growth; mandatory to transport large electric currents for the different compositions and, eventually, leading to the through the material. Among them, chemical solution optimal growth rates and functional properties of super- deposition (CSD) is demonstrated to be a very promising conducting coated conductors. low-cost approach for REBCO film growth with high manufacturing perspective into CC tapes, through the use of reel-to-reel compatible deposition methodologies, such as ■ EXPERIMENTAL SECTION Ink Formulation and Characterization. Metalorganic precursor inkjet printing or slot-die coating.16,17 Despite the industrial- REBCO (RE = Y and Gd) inks with a stoichiometry REBa2Cu4.66Ox ization process has seen significant advancements, reaching and a final concentration of 0.75 M were used to prepare very attractive total critical currents of ∼700 A/cm width at 77 combinatorial samples. Commercial Cu(ac)2 salts (Merck KGaA) K, the cost/performance ratio of CC technology is still limited and in-house synthesized Y/Gd and Ba propionate were employed. due to insufficient growth rates (1−10 nm/s).18 Although The selected ink stoichiometry correlates with the Ba−Cu−O eutectic liquid phase-assisted fabrication methods have been reported liquid composition (3BaCuO2-4CuO), which enables the crystal- lization of REBCO films at lower temperatures.21 First, a solvent to allow faster growth rates due to the large diffusivity of atoms mixture of propionic acid (PPA) and butanol (BuOH) in a 50:50 in the liquid,19,20 it has not been until very recently that a novel ratio was added into a flask.16 Different %v/v ethanolamine (EA) were CSD strategy, known as transient liquid-assisted growth or added to the solvent mixture. EA acts as a complexing agent for TLAG-CSD, has been reported to greatly boost the growth Cu(ac)2, improves ink solubility and homogeneity, and provides rate of epitaxial YBCO films and nanocomposites (∼100 nm/ enough viscosity to avoid excessive liquid movement during inkjet s) through the use of nonequilibrium growth conditions. A deposition. Subsequently, Ba(prop)2 and Cu(ac)2 salts were slowly rapidly induced transient Ba−Cu−O liquid promotes the fast added under continuous stirring and kept for approximately 30 min.34 Finally, Y(prop)3 or Gd(prop)3 were added to the mixture and left diffusion of RE ions to the growth front and allows a much stirring at 30 °C until a homogeneous and transparent solution was faster epitaxial growth.21 Due to the novelty of the process, obtained, i.e., typically 30−60 min. The precursor solutions were many aspects are still unexplored, such as the processing filtered through a polytetrafluoroethylene (PTFE) membrane filter parameter window to achieve a high control of epitaxial (pore size ∼0.2 μm). nucleation and crystallization, the influence of different rare Surface tension and viscosity measurements were carried out with a earth elements on the PO 2-T conditions or the liquid HAAKE RheoStress RS600 from Thermo Electron Corp, while contact angles were evaluated with a Drop Shape Analyzer DSA 100 composition. The optimization of such a large number of from KRÜ SS. The water content of the solutions is also critical for the parameters requires fast screening methods for high- deposition and final properties of the samples; therefore, it was throughput sample fabrication, combined with smart solutions evaluated using the Karl-Fischer titration method.45 All solutions have in characterization and data analysis. water percentages after the synthesis of ∼0.7% wt. Two decades ago, Evans et al. described the philosophy Drop-on-Demand Inkjet Printing of REBCO Inks. Drop-on- behind the combinatorial chemistry approach to push forward demand piezoelectric inkjet printing (DoD IJP) from Microdrop the development of inorganic materials.22 This strategy relies Technologies GmbH was used to fabricate combinatorial test pieces on single-crystal (001) SrTiO3 (STO) substrates. Prior to deposition, on the fabrication of large arrays of microsamples that are the substrates were thermally treated at 900 °C for 5 h to develop a treated together or individually and characterized using flat-terraced surface and thoroughly cleaned with acetone and adequate techniques that allow a fast evaluation of the methanol to remove any residues. processing parameters. The produced data is stored on large The IJP system possesses a 96-well plastic plate array that serves as relational databases that can be later analyzed using computa- solution containers. Each of the nozzles takes the solution from one of tional methods to find the optimal compositions and these vessels, which in our case contained YBCO and GdBCO processing conditions. Eventually, this allows researchers to precursor solutions. A multinozzle head was used to load each focus on the interpretation of results and is prospected to solution on separated nozzles (Figure 1a), and the ejection parameters (voltage and pulse length) were adjusted to achieve a speed up the discovery of materials. Nowadays, this strategy is stable drop formation, a drop diameter of ∼70 μm, and a drop volume starting to flourish, thanks to the development of data-driven of ∼180 pl (Figure 1b). Throughout this article, only two nozzles/ methods based on big data analysis tools, machine learning inks were used to fabricate the combinatorial samples. The specific algorithms, and other computational methods.23−28 High- throughput experimentation (HTE) strategies, based on the combinatorial chemistry approach, are currently being researched in many diverse fields such as solar cells,29−31 lithium-ion batteries,32−34 electronics,35−37 water splitting,38,39 and superconductivity.40,41 From several deposition techniques available for the assembly of combinatorial test pieces,40−44 drop-on-demand inkjet printing (DoD IJP) appears as an ideal technique since it allows the fabrication of complex-shaped samples suitable for high-throughput measurement procedures and also achieves locally uniform and graded compositions. In this work, we propose a strategy based on a combinatorial chemistry Figure 1. (a) DoD IJP setup, consisting of a 96-well plate array, a approach by working with a DoD IJP for the mixing of multinozzle head with four independent nozzles, a stroboscopic light/ different REBCO inks. This approach will contribute to the full camera for drop inspection, and a camera for surface inspection. (b) disclosure of optimal ultrafast TLAG-CSD processing con- Captured in-flight image drop after the adjustment of the deposition ditions through the exploration of the PO2-T nonequilibrium parameters from the camera used for drop inspection. 9102 https://dx.doi.org/10.1021/acsami.0c18014 ACS Appl. Mater. Interfaces 2021, 13, 9101−9112 ACS Applied Materials & Interfaces www.acsami.org Research Article Table 1. Physicochemical Parameters of the REBCO Inks % vol EA viscosity (cP) contact angle (deg) density (g/cm3) surface tension (mN/m) Z ejectability 0 5.0 ± 0.1 10.0 ± 0.1 0.98 ± 0.01 24.4 ± 0.1 9.2 ± 0.1 ejectable 1.14 6.8 ± 0.1 <10.0 ± 0.1 0.99 ± 0.01 25.4 ± 0.1 7.0 ± 0.1 ejectable 3.66 15.2 ± 0.1 <10.0 ± 0.1 0.99 ± 0.01 25.8 ± 0.1 3.1 ± 0.1 not ejectable printing positions were automatically generated through a Python interface that outputs an (x,y)-coordinate file with the desired pattern dimensions and drop/line pitches. ■ RESULTS AND DISCUSSION The combinatorial chemistry approach aims to improve and Sample Drying, Pyrolysis, and High-Temperature Anneal- speed up the fabrication of materials using high-throughput ing. After the deposition of the desired patterns, samples were dried deposition and characterization techniques, as well as at 60 °C to partially evaporate solvent leftovers and avoid further incorporating computational methods for the automation of liquid movement. Subsequently, a pyrolysis procedure is performed in data acquisition and processing. In this manuscript, we have a tubular furnace at a heating rate of 5 °C/min up to 240 °C and 3 studied the combinatorial approach through gradual composi- °C/min up to 500 °C in a humid oxygen flow of 0.12 l O2/min for 5 tional changes of the rare earth (RE = Y, Gd) in the REBCO min.21 After pyrolysis, all of the organic materials have been removed phase. It is worth noting that we will refer to the final from the sample and only amorphous/nanocrystalline phases remain. Y1−xGdxBCO composition throughout the manuscript even Finally, the combinatorial samples were treated at high temperatures when talking about precursor inks and deposited stripes to under various conditions using a rapid thermal annealing furnace classify the different compositions. Three stages have been (Annealsys AS Micro) to grow the REBCO phases. considered in the combinatorial DoD IJP approach for the Sample Characterization. The characterization of combinatorial investigation of these REBCO superconductors: (1) determi- samples was performed using different techniques. Surface morphol- nation of ink and deposition parameters for the deposition of ogy of the as-deposited and pyrolyzed samples was evaluated by combinatorial test pieces, (2) validation of the compositional means of optical microscopy (OM) (Leica DM1750 M), as well as homogeneity needed for combinatorial tests, and (3) profilometry (P16+ profilometer from KLA Tencor), while scanning identification of high-throughput characterization and data electron microscopy (SEM) with a QUANTA FEI 200 FEG-ESEM analysis of combinatorial test samples to extract valuable was employed to evaluate the surface morphology of grown samples. information of growth mechanisms. Energy-dispersive X-ray (EDX) spectroscopy measurements were Inkjet Printing Deposition Requirements. The use of carried out on each stripe to verify the deposited composition and DoD IJP to deposit combinatorial samples requires control of ensure an adequate merging of the REBCO inks. The structural characterization was conducted by high-resolution process parameters to ensure a uniform deposition, especially X-ray diffraction (XRD) with a Bruker D8 Discover system (Cu Kα, when combining several inks and nozzles. First of all, ink X-ray energy = 8.049 keV) equipped with a Lynxeye XE energy- formulation, drop formation, jetting parameters, and drop dispersive one-dimensional (1D) detector. The quantification of the pitches (spacing along the x- and y-directions) need to be amount of epitaxial material on each stripe was performed using a optimized. These variables are largely influenced by ink Bruker-AXS D8 Advance diffractometer (Cu Kα) equipped with a stability, as well as its physicochemical properties. Figure S1 general area detector diffraction system (GADDS). shows the viscosity and surface tension measurements for the The spatial distribution, size, and crystalline state of precursor REBCO ink with 3.66% vol EA. The inverse Ohnesorge phases in pyrolyzed films were probed via high-resolution trans- number (Z) is often used as an indication of ink ejectability mission electron microscopy (HR-TEM) and high-angle annular factor and is expressed as dark-field scanning transmission electron microscopy (STEM- HAADF). For this purpose, an FEI Tecnai F20 (S)TEM was ρlγ operated at 200 kV. For transmission electron microscopy (TEM) Z= η (1) studies, cross-sectional specimens were prepared by conventional methods: slicing, gluing, tripod mechanical polishing, and finally Ar+ ρ is the density of the liquid, η is the viscosity, γ is the surface ion milling using Gatan PIPS. tension, and l is the nozzle diameter (50 μm). Typical Z values In Situ XRD Synchrotron Measurements. In situ XRD growth ensuring good ejectability range from minimum values of 1−4 experiments were performed at the DiffAbs beamline, SOLEIL and maximum ones of 10−14.48,49 Typical values for the tested synchrotron facilities (France), using an X-ray beam energy of 18 keV REBCO inks are summarized in Table 1. The inks employed in and an XPAD (X-ray hybrid pixel area detector)46,47 area detector this work could be ejected for Z > 3.1. with acquisition times per image of 100 ms. The obtained images can The optimized parameters employed in this work to achieve be converted into intensity vs 2θ diffractograms considering the experimental geometry (sample and detector angular positioning) and stable drop formation are summarized in Table 2. Typically, an using a home-written Python code. The samples were heat treated in increase in the % vol EA required higher voltages to achieve a an Anton Paar DHS 1100 domed heating stage at rates of 25 °C/min, stable drop formation. Liu et al. mentioned that drop 1 bar of total pressure and 1 mbar of oxygen partial pressure. An X-ray formation is achieved when the velocity is high enough to transparent Kapton dome, equipped with a double connection to the overcome the surface tension and viscous forces,50 which vacuum pumps and gas inlet/outlet systems, was utilized to seal the supports our observations for both types of inks. heating stage. The total pressure control was achieved through The combinatorial test samples under the present study are electrovalves, while an oxygen sensor was used to preset PO2 at the composed of several stripes of different compositions desired value. Pressure values were tracked via the LabVIEW interface deposited in one single sample. This geometry has enabled using a remote connection to a pressure meter. Synchrotron us to verify the methodology employed, as described below, experiments were employed for the in situ analysis of the growth of and will allow us to extend it to more complex schemes (i.e., combinatorial samples, as it will be described in the In Situ XRD graded samples) in the future. To define the combinatorial Synchrotron Analysis section. chemistry strategy for inkjet printing, two precursor inks for 9103 https://dx.doi.org/10.1021/acsami.0c18014 ACS Appl. Mater. Interfaces 2021, 13, 9101−9112 ACS Applied Materials & Interfaces www.acsami.org Research Article Table 2. Jetting Parameters Employed for the Drop Formation (df) and Data Grid (dg) Deposition of REBCO Inks jetting parameters value voltage up (V) 140 ± 40 voltage down (V) 0 pulse length (μs) 25 ± 5 substrate temperature (°C) 25 ± 1 frequency (Hz) 1 (dg), 100 (df) drop volume (pl) 175 ± 5 drop size (μm) 215 ± 15 (dg), 70 ± 2 (df) Figure 2. OM images of the as-deposited combinatorial samples substrate/nozzle distance (mm) 1.0 ± 0.1 showing the stripe morphology for inks with a 50/50 ratio of PPA/ drop velocity (m/s) 2.0 ± 0.1 BuOH, and containing (a) 0% and (b) 1% of EA. Drop pitches of 50 μm and 95 μm along the x- and y-directions were employed. YBCO and GdBCO final compositions were loaded into two To optimize the pyrolysis conditions of combinatorial separate nozzles. These inks were tested to print 5 stripes with samples, at first, a complete film with a nominal Y1−xGdxBCO Y1−xGdxBCO ratio combinations of x = 0, 0.25, 0.5, 0.75, and 1 ratio of x = 0 was prepared by spin-coating with a subsequent on single STO substrates. Single solvent inks are usually thermal treatment (5 °C/min up to 240 °C and at 3 °C/min preferred since the evaporation rate plays a big role in the from 240 to 500 °C). This ensured the decomposition of the homogeneity of printed patterns.51 Despite this, the inks organic constituents and the formation of nanocrystalline employed in this work contained both PPA and BuOH since oxycarbonates (RE2O3, BaCO3, and CuO), which are the the addition of BuOH provided thickness in the required range and provided an adequate boiling point to ensure a compatible precursor phases for the REBCO final layer to be reached upon drying process.52 an additional second high-temperature treatment.56 TEM The identification of the best drop pitch combinations is investigations were conducted on this sample to evaluate the intermediate precursor phases present after the pyrolysis step. essential to ensure a homogeneous deposition of the different The low-magnification STEM-HAADF image of an as- Y1−xGdxBCO stripes. An average drop size of 241 ± 4 μm was pyrolyzed YBCO film (Figure 3a) reveals a film thickness of obtained after measuring the diameter of several drops ∼800 nm, and nanocrystalline phases are homogeneously deposited on single-crystalline (001) STO substrates with a distributed within the film. It has been found that any separation of 500 μm between them (Figure S2a). The coarsening or phase segregations, for example, CuO experimental testing of different drop pitches (dx = 50, 75, and segregation at the interface of multiple layers, can degrade 100 μm and dy = 50, 75, 100, 125, and 250 μm) was then the film quality and homogeneity of the grown REBCO thin performed to fine-tune the best deposition conditions (Figure film.55 In this regard, the above-pyrolyzed film, without the S2b). Small drop pitches (50 μm × 50 μm and 75 μm × 75 presence of any segregation layer, as displayed in Figure 3a, can μm) showed a large accumulation of solution, which led to provide a high-quality REBCO grown film. We expect that the inhomogeneities due to liquid movement, while larger decomposition of YBCO and GdBCO precursor solutions will combinations led to unconnected patterns. The conditions occur similarly since thermogravimetric studies have been that showed very little liquid movement and smooth surfaces performed previously on YBCO precursor films by Rasi et al.56 were dx = 50 μm and dy = 85−100 μm. These drop pitches showing that all organic precursor transformations occur below result in degrees of overlap in x = 80% and y = 60% (100 500 °C, and at that temperature only Y2O3, CuO, and BaCO3 μm)−65% (85 μm). The degree of overlap was calculated O remain in the films. Additionally, Grivel et al.57 have using the expression D × 100, where O is the overlapping investigated the decomposition of different rare earth distance between the two drops (D-dx or D-dy) and D is the propionate powder precursors, including gadolinium propio- diameter of the drop impact. nate. They observed that all rare earth propionate precursors We have also explored the inclusion of additives in ink to behave similarly, leading to amorphous nanocrystalline oxide enhance the solubility of the metalorganic precursors, increase phases at 500 °C. It is worth noting that films decompose the homogeneity, facilitate drop formation, as well as to earlier than powders due to the much more efficient gas prevent excessive drying of solvents and nozzle clogging.53,54 In transport occurring in the films as compared to powders as particular, we used ethanolamine (EA) for its known role as a reported previously.58 The low-magnification STEM-HAADF complexing agent in copper-based solutions.55 Several %v/v EA image in the bottom-right inset of Figure 3a displays the were tested to find a compromise between the increase of ink smooth surface of the as-pyrolyzed YBCO thin film. The viscosity with the amount of EA (Table 1) and the increased distinct nanocrystalline phases are identified in HR-TEM difficulty for drop ejection. The optimized solution contained a images (Figure 3b−e), which show the typical diameters of 1%v/v EA and a 50/50 PPA/BuOH ratio. Figure 2 shows two BaCO3 (orthorhombic), 10−30 nm; BaCO3 (monoclinic), 5− combinatorial samples prepared with inks containing 0%v/v and 7 nm; CuO, 10−25 nm; and Y2O3, 5−6 nm. The majority of 1%v/v EA and a 50/50 PPA/BuOH ratio in a 0.75 M solution BaCO3 is found to be orthorhombic that can easily decompose with RE-prop, Ba-prop, and Cu(ac)2 metalorganic salts. The during the TLAG-CSD process.56 Moreover, the size of addition of EA clearly shows a qualitative improvement in the nanocrystalline phases of about tens of nanometers can also homogeneity of the printed stripes, liquid movement is greatly assist in the fast reaction process during the TLAG-CSD reduced, and the stripe edges are better confined into the growth. The HR-TEM image in Figure 3f exhibits high printed dimensions (500 μm × 5 mm). homogeneity among nanocrystalline phases of BaCO 3 9104 https://dx.doi.org/10.1021/acsami.0c18014 ACS Appl. Mater. Interfaces 2021, 13, 9101−9112 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 3. (a) STEM-HAADF image of the as-pyrolyzed film, where CuO nanocrystals (in gray) are homogeneously distributed within a BaCO3 (white) matrix. The inset (top-right) shows the porosity analysis, where pores are colored in red for quantification. The inset (bottom-right) is a low-magnification cross-sectional STEM-HAADF image exhibiting the smooth surface of the thin film. (b−e) HR-TEM images identify the distinct nanocrystalline phases of Y2O3, BaCO3 (monoclinic), BaCO3 (orthorhombic), and CuO, respectively. (f) HR-TEM image shows the presence of all three nanocrystalline phases in the same region, exhibiting a homogeneous distribution. Figure 4. (a) OM images of a pyrolyzed combinatorial sample with different REBCO (RE = Y, Gd) compositions as indicated. (b) 3D profilometry surface map of a stripe with a 50/50 composition. (c) Series of profiles (gray lines) and the mean value (red line) extracted from the green region in (b). The roughness mean square (RMS) value and average height (h) have been calculated from different stripes and samples. (orthorhombic), Y2O3, and CuO, which was also confirmed Compositional Homogeneity Validation for Combi- throughout the film from the HR-TEM observation of natorial Samples. Drop positions and solution intermixing, randomly chosen regions of several films. The pore density upon deposition, are key parameters to define combinatorial of the film was calculated using image analysis (ImageJ), and it patterns (discrete or gradient type) by DoD IJP. In our case, is estimated to be approximately 1.0 ± 0.2%, as shown in the we selected a set of five equally separated stripes of different top-right inset of Figure 3a. We emphasize that the degree of compositions. The validation of the composition homogeneity porosity is remarkably low, especially for pyrolyzed CSD films, within one stripe is essential to demonstrate the capabilities of and expected to enable improved epitaxial growth of REBCO stoichiometry tuning. Figure 5 shows the visual representation in the final step of crystallization.55 We believe that the of the printing positions and drop overlapping, considering a addition of EA might be the origin of this reduced porosity and drop diameter of 240 μm and a grid size of 50 μm × 95 μm. we are currently conducting several studies to confirm it. These visualizations were generated from data obtained using a Since the only change between the different compositions is self-made Python interface, as well as employing the NumPy, related to the RE2O3 phase, combinatorial samples were Pandas, and Matplotlib libraries.59−61 The schematic repre- sentation of the drop combinations between the two inks used printed with the optimized conditions and, subsequently, for each composition is also illustrated on the right side. This subjected to the previously described pyrolysis treatment. suggests that the previously determined experimental con- Figure 4a shows a pyrolyzed combinatorial sample with stripes ditions should enable a high degree of solution intermixing and that contain nominal Y1−xGdxBCO ratios ranging from x = 0 a uniform composition distribution along the printed stripes (YBCO) to x = 1 (GdBCO) as indicated. The average given that the same miscible solvents are employed for all roughness mean square (RMS) roughness, which was solutions. It also reveals that the pattern edges may present a calculated from three-dimensional (3D) profilometry images less uniform distribution due to edge constrains. of the stripe surface on several stripes (Figure 4b), is 60 ± 20 Screening over a large number of samples is one of the main nm. A series of profiles extracted along the width of each stripe aspects that must be addressed when dealing with (Figure 4c) show that the edges are highly defined, with less combinatorial samples to promote the construction of material than 50 μm difference between the start to the end. In libraries. Among the different techniques available, it is addition, while the height of each stripe is very homogeneous, required that they possess sufficient spatial precision and an average value of 600 ± 100 nm has been calculated from resolution, as well as acquisition speed, and allow parallel or several measurements on different stripes. rapid sequential analysis.25,62,63 With this in mind, several 9105 https://dx.doi.org/10.1021/acsami.0c18014 ACS Appl. Mater. Interfaces 2021, 13, 9101−9112 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 5. Plots that illustrate the printed DoD IJP patterns for Y1−xGdxBCO composition ratios of (a) x = 0, (b) 0.25, (c) 0.5, (d) 0.75, and (e) 1, with a grid size of 50 μm × 95 μm. The (x,y) coordinates were automatically generated using a self-made Python interface. The graphs on the right side of each stripe show a schematic representation of the drop alternation between the two inks used to reach the desired composition. Yellow and green indicate the drops of the two different solutions used. Figure 6. (a) EDX elemental analysis of selected area from a stripe techniques have been employed to evaluate the composition of with a Y0.5Gd0.5BCO. The inset shows the SEM image of the combinatorial samples such as X-ray fluorescence, energy- measured region. (b) Statistical analysis of the composition measured dispersive X-ray spectroscopy (EDX), X-ray diffraction, mass by performing EDX spectroscopy on several combinatorial samples spectrometry, etc.26,27,64 In our study, we used EDX for several with stripes having Y1−xGdxBCO ratios of x = 0, 0.25, 0.5, 0.75, and 1. combinatorial samples to evaluate the composition accuracy of each stripe and validate the DoD IJP approach. Figure 6a shows the elemental analysis performed by EDX of a typical area of approximately 250 μm × 250 μm, where the acquisition was conducted for a stripe with Y0.5Gd0.5BCO. It should be recalled that the drop size was of 241 ± 4 μm in diameter and the drop pitches were dx = 50 μm and dy = 85−100 μm, which should ensure 60−80% drop overlap (Figure 5). Figure S3 shows the EDX elemental analysis and SEM images of selected areas from stripes with Y1−xGdxBCO compositions of x = 0.25 and 0.75. After the measurement of several stripes, we performed a statistical analysis that shows the average values obtained and their standard deviation for each theoretically desired composition (Figure 6b). Focusing on compositions with different percentages of Y1−xGdxBCO inks, we confirm, Figure 7. (a) High-resolution XRD diffractograms of a sample on samples annealed at 5 °C/s up to 815 and 830 °C for 2 annealed at 5 °C/s, 840 °C for 2 min and a PO2 of 1 mbar, depicting min, that the desired composition values and measured ones the (200) STO substrate peak position and the (007) peak for each do not deviate much from each other. Nevertheless, there is a Y1−xGdxBCO composition. (b) Evolution of the c-axis parameter for slightly large dispersion of values with a maximum variation of Y1−xGdxBCO samples annealed at 5 °C/s up to 835 °C and 840 °C ±15% for different samples with Y0.5Gd0.5BCO, which might for 2 min and a PO2 of 1 mbar. The (00l) REBCO peaks and associated shifts, as shown in (a), are translated into lattice parameter be influenced by the fact that growth at 50/50 was the most variations using the Nelson-Riley criteria (Figure S4). Blue lines complicated over the full range explored. indicate the theoretical values from the ICSD database, 11.701 and The evaluation of the c-axis lattice parameter for REBCO 11.683 Å for the optimally doped GdBCO and YBCO. Error bars are has been employed as another indicator to demonstrate proper obtained from the Nelson-Riley linear regression. RE mixing on the scale of a unit cell. Considering that Y1−xGdxBa2Cu3O7 is a solid solution in RE, the lattice parameter of each composition should be a weighted mean, according to their ratio in the sample.65 Following this fitting of (00l) REBCO peaks, l ≥ 5, with pseudo-Voigt profiles description, high-resolution XRD was conducted (Figure 7). and extrapolating to x = 0 following the Nelson-Riley criteria (Figure S4)66 1 jij cos2 θ cos2 θ zyz To minimize systematical errors, substrate alignment was jj z θ z{ carried out individually for each stripe prior to the measure- 2 k sin θ ment (using the STO (200) reflection at 2θ = 46.47°). The c- + axis parameter of each composition was then determined by (2) 9106 https://dx.doi.org/10.1021/acsami.0c18014 ACS Appl. Mater. Interfaces 2021, 13, 9101−9112 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 7a shows the XRD diffractograms of the (200) STO (Iepitaxial) and polycrystalline (Irandom) orientations is quantified peak together with the (007) reflection of each Y1−xGdxBCO with the following equations67 composition, used for the calculation of the c-axis lattice exp Irandom Iring × (360/Δχ ) × 4π parameter of each stripe for a sample annealed at 5 °C/s up to = 840 °C for 2 min. Figure 7b illustrates the evolution of the c- Iepitaxial exp 8 × Iepi (3) axis lattice parameter as a function of the Y1−xGdxBCO ratio for two samples annealed at 5 °C/s up to 835 °C and 840 °C Irandom + Iepitaxial = 100(%) (4) for 2 min. The trend depicted for each stripe is clearly in agreement with those obtained for the optimally doped YBCO Below 800 °C, the degree of epitaxy achieved increases with and GdBCO for single crystals, which have c-axis parameters of temperature from around 10−20% up to 50%. A higher degree 11.683 and 11.701 Å, respectively. of epitaxy is observed for Gd-rich compositions. Full epitaxial Identification of High-Throughput Characterization growth (>90%) is observed for the Gd-rich compositions (x ≥ and Data Analysis. Assessment of Epitaxial Growth 0.75) at temperatures between 815 and 830 °C, while Conditions. The validation of the combinatorial approach compositions with x ≤ 0.5 show values between 20 and presented above now allows us to explore the differences in 70%. Above 830 °C, the degree of epitaxy starts decreasing for growth conditions for Y1−xGdxBCO by investigating several Gd-rich samples, while Y-rich compositions reach epitaxy stripes at the same time. Several sets of combinatorial samples values above 90% at 845 °C. Despite this, Gd-rich were grown following the temperature route of the REBCO compositions show epitaxial fraction values between 40 and TLAG-CSD approach21 at temperatures ranging from 750 to 70% up to 850 °C, while Y-rich ones decrease much faster. 850 °C with 2 min of dwell and a heating ramp of 5 °C/s. Our Figure S6 shows the SEM images of the samples grown at 830 goal was to assess the degree of epitaxy at each condition to and 845 °C. It confirms that 830 °C is a rather good establish the optimum growth temperature for each temperature to achieve homogeneous, flat, and porous free composition. Figure 8 shows a composition vs temperature layers for Gd-rich compositions, while higher temperatures have to be achieved for Y-rich ones. The evolution of the epitaxial fraction with temperature is related to the super- saturation of the Ba−Cu−O melt, advantageously used in the nonequilibrium TLAG-CSD process to enhance growth rates. In general, at low temperatures, crystallization occurs at higher supersaturation, while at high temperatures supersaturation decreases, enabling the c-axis heteroepitaxial growth to be achieved. In these liquid mediated processes, supersaturation is dominated by the concentration of RE dissolved in the liquid. As temperature increases, this should also increase provided the heating ramp does not limit this process. Nevertheless, the optimum conditions will be different for each composition, since relative supersaturation, Δσ = (Cγ/Ce − 1), depends on Figure 8. Three-dimensional (3D) map of the epitaxial fraction as a the difference between the RE concentration in the liquid, Cγ function of the Y1−xGdxBCO composition and temperature for and equilibrium concentration, Ce, at each temperature and samples annealed at 5 °C/s for 2 min and a PO2 of 1 mbar. The time.21 It is known that Ce depends on the RE, with Ce,Gd > colored dots indicate the SEM images of specific Y1−xGdxBCO Ce,Y,68 thus confirming that lower relative supersaturation composition, depicted in Figure S6 (Supporting Information) for two conditions will likely be reached for GdBCO than for YBCO at samples annealed at 5 °C/s up to 830 and 845 °C for 2 min and a a fixed temperature, though we cannot discard variation of Cγ PO2 of 1 mbar, as indicated in the figure. with RE. All of these arguments support the experimental evidence that the highest epitaxial degree is reached with map with the degree of epitaxy as the color scale. The epitaxial GdBCO at 830 °C, while it is at 845 °C for YBCO. Overall, fraction was calculated by integrating two-dimensional (2D)- the combinatorial approach employed here has enabled to map XRD 2θ−χ scans at the position of the (005)REBCO epitaxial growth as a function of composition and temperature, reflection at χ = 0° (Iexp epi ) and χ ≠ 0° (Iring), as indicated in exp demonstrating the desired reduction in the sample number Figure S5. Then, the relationship between the epitaxial through the combinatorial strategy (13 combinatorial samples Figure 9. (a) Schematic representation of the experimental setup used for the in situ synchrotron measurements, and (b) beam alignment to determine the stripes’ position, i.e., at the intensity minima. 9107 https://dx.doi.org/10.1021/acsami.0c18014 ACS Appl. Mater. Interfaces 2021, 13, 9101−9112 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 10. In situ XRD synchrotron measurements during annealing by heating at 25 °C/min up to 860 °C. Phase evolution of the YBCO stripe (a) showing the total diffraction intensity for each crystalline phase, and (b) zoom-in of (a) for better visualization of the low-intensity phases. Figure 11. (a) Color map showing the time evolution of the intensity of the (005) epitaxial peak (2θ ∼ 16.7−16.9°), and (b) intensity increase (slope) and time difference (Δt) for crystallization of the epitaxial phase for stripes with Y1−xGdxBCO ratios of x = 0, 0.25, 0.5, 0.75, and 1, taking tGdBCO as a reference. The error bars were determined from the difference between the points used to calculate the time where epitaxial crystallization begins. instead of 78 regular samples were explored to reach the same was determined through a quick scan along the y-direction mapping resolution). with a set of defined positions, and by evaluating the scattered In Situ XRD Synchrotron Analysis. Synchrotron radiation is intensity variations (around particular 2θ values) of the a powerful tool for high-throughput experimentation, thanks to acquired XRD datasets vs the lateral position. In these the high monochromaticity, brightness, collimation, and conditions, the intensity minima indicate the center position temporal/spatial resolution, allowing rapid characterization of of each stripe, while the peaks correspond to the substrate microsized combinatorial samples.69,70 In addition, in situ (Figure 9b). The movement time between stripes was experiments can be performed if fast 2D detectors are estimated to be 1.6 s on average, considering the acquisition employed. In the case of the ultrafast TLAG-CSD growth and data transfer times. It should also be noted that substrate process, this is the only way to explore the kinetically driven positions between the stripes were also measured. Hence, a ultrafast growth mechanisms. Here, in situ XRD experiments measurement was performed on each stripe every ∼20 s, which were performed on our combinatorial samples using the setup are relatively long acquisition times. As a result, these set of described in the In Situ XRD Synchrotron Measurements experiments required the use of TLAG-CSD conditions where section. The aim was to in situ analyze the 5 stripes localized in the growth process is relatively slow (1−5 nm/s), using slower one single sample and in a single (same) growth process. Using heating ramps as compared to the ones reported before.21,56 an area detector, a certain 2θ angular range, of about 10°, is This approach allowed us, together with the combinatorial covered in a single shot/image. The sample and detector films, to resolve the REBCO crystallization process and angles were fixed and set to record scattered signal close to the intermediate reactions in time, temperature and Y1−xGdxBCO substrate and precursors diffraction peaks. The strategy compositions, all in one single experiment (Figure S7). followed was to measure the printed stripes through a “stop- Figure 10 shows the phase evolution with time and and-go” stage movement along the transverse y-direction, as temperature for an in situ XRD synchrotron experiment schematically shown in Figure 9a. The position of each stripe conducted during annealing by heating at a rate of 25 °C/min 9108 https://dx.doi.org/10.1021/acsami.0c18014 ACS Appl. Mater. Interfaces 2021, 13, 9101−9112 ACS Applied Materials & Interfaces www.acsami.org Research Article up to 860 °C. The phase evolution is extracted from 2D-XRD demand inkjet printing was successfully used to deposit raw data images consisting of 2θ, χ and intensity information combinatorial samples in the form of stripes of different by performing a summation of counts in a particular 2θ−χ area compositions, obtaining a homogeneous surface distribution of each image and fitting of I (2θ) peaks. Figure S8 shows a during the pyrolysis step after the optimization of the ink 2D-XRD raw image of a time frame, where the (005) and composition and printing positions. The TEM microstructure (103) REBCO phases coexisted. At temperatures below 400 analysis of a complete film revealed the homogeneous °C, typical intermediate precursor phases of YBCO fluorine- distribution of nanocrystalline phases and the low porosity of free precursor films are observed,21 i.e., BaCO3 (monoclinic) the pyrolyzed stripes. The desired compositions could be and CuO. Above this temperature, the BaCO3 (monoclinic) successfully implemented and confirmed through the com- phase transforms into BaCO3 (orthorhombic) and reacts with bined use of computational methods and postcharacterization CuO to form the transient liquid. The observed BaCuO2/ techniques such as EDX and HRXRD, as evidenced by BaCu2O2 and Cu2O phases crystallize from the liquid due to elemental rare earth ratio and the c-parameter values. We have the heating rates of the experiment (25 °C/min). The YBCO also shown that epitaxial growth conditions vary for each phase starts to form at around 600 °C and is mostly randomly Y1−xGdxBCO composition, Gd-rich ones having higher values oriented at this point. At only around 790 °C, the epitaxial part of an epitaxial fraction at lower temperatures as compared to Y- is significantly increasing in intensity. It is noticeable that the rich ones. Finally, in situ XRD synchrotron experiments random YBCO phase completely disappears at expenses of the allowed to track the evolution of phases present during the epitaxial orientation at 860 °C. These experiments encouraged TLAG-CSD growth. We confirm that compositions richer in us to further study the mechanism behind this orientation Gd start at earlier times and imply faster growth processes than conversion; the mechanisms will be reported elsewhere. YBCO. The success of the in situ XRD synchrotron Additional details on the influence of the heating ramp on experiments in the analysis of combinatorial samples not the formation of intermediate precursor phases and YBCO are only shows the richness of the method but also opens paths to also of major importance and will be discussed in another the improvement of the setup and data acquisition schemes. publication since it goes beyond the scope of this work. Rapid on-the-flight measurements through compositionally Figure 11a shows a color map with the evolution of the graded samples should allow the accessibility and character- epitaxial phase (scattered intensity in the vicinity of the (005) ization of larger heating ramps and crystallization growth peak),21 but adapted to the combinatorial strategy and the rates.71,72 All-in-all, the results obtained show that the different REBCO compositions presented in this work. Note combinatorial and high-throughput approach can be now that the stripes possess different times for the initiation of moved forward to the fabrication of continuous compositional epitaxial crystallization, depending on their composition, as gradients. This will enable enhanced mapping of growth identified in Figure 11b. In particular, the stripe with 100% conditions, a fast screening of functional properties, the GdBCO starts to crystallize first, the one with 100% YBCO is construction of libraries that include, for instance, other RE the last (Δt = 324 s), and the intermediate compositions lay ions, liquid composition changes, nanoparticle percentages in between both individual REBCO phases, according to their RE nanocomposites, and the implementation of machine learning proportion. In addition, the slope of the increase in intensity of algorithms for a swift and smart optimization of TLAG-CSD the (005) epitaxial peak reveals two times faster rise for the REBCO film growth. Moreover, the methodology developed GdBCO composition as compared to YBCO. These results are here may be extended to many different functional materials in accordance with previous reports,65 where the growth rate of with different needs of fast screening and optimization. REBCO materials in liquid-assisted methods was shown to be Ultimately, the combinatorial screening will also pave the proportional to the solubility of the RE. Since the atomic way to novel material design strategies based on big data (i.e., radius of Gd is larger than that of Y, its solubility will be higher data-driven approaches), aiming not only to accelerate the and, hence, this could possibly explain its earlier crystallization innovation throughput in the field of superconducting films as compared to Y. and coated conductors but also in other fields of materials In summary, the work presented here demonstrates that science. screening over a large variety of conditions is facilitated by the fabrication of combinatorial samples. Appropriate character- ization techniques have enabled simultaneous measurement of ■ ASSOCIATED CONTENT * Supporting Information sı combinatorial samples, leading to the generation of a large The Supporting Information is available free of charge at volume of information to be analyzed and interpreted, which is https://pubs.acs.org/doi/10.1021/acsami.0c18014. vital in the implementation of a data-driven material design scheme based on “big data”. Automation tools for data Rheological measurements of REBCO inks (Figure S1); acquisition and treatment are also necessary to generate inputs additional optical images of inkjet printing patterns for machine learning algorithms, which can be used to ease the (Figure S2); EDX elemental analysis (Figure S3); interpretation of results, as well as enable the smart Nelson-Riley evaluation (Figure S4); raw 2D-XRD optimization of epitaxial growth conditions for different data (Figure S5); SEM images of samples (Figure S6); REBCO stoichiometries and, eventually, find the best func- in situ XRD synchrotron diffractograms (Figure S7); raw tional properties. 2D-XRD synchrotron image (Figure S8) (PDF) ■ CONCLUSIONS AND OUTLOOK The potential of combinatorial and high-throughput exper- ■ AUTHOR INFORMATION Corresponding Authors imentation for fast identification of the experimental Albert Queraltó − Institut de Ciència de Materials de conditions to grow different Y1−xGdxBCO compositions by Barcelona (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, the TLAG-CSD approach has been demonstrated. Drop-on- Catalonia, Spain; Email: aqueralto@icmab.es 9109 https://dx.doi.org/10.1021/acsami.0c18014 ACS Appl. Mater. Interfaces 2021, 13, 9101−9112 ACS Applied Materials & Interfaces www.acsami.org Research Article Teresa Puig − Institut de Ciència de Materials de Barcelona postdoctoral fellowship (Grant no. IJC2018-035034-I). We (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, acknowledge support of the publication fee by the CSIC Open Catalonia, Spain; Email: teresa.puig@icmab.es Access Publication Support Initiative through its Unit of Information Resources for Research (URICI). ■ Authors Juri Banchewski − Institut de Ciència de Materials de REFERENCES Barcelona (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, Catalonia, Spain (1) Shiohara, Y.; Taneda, T.; Yoshizumi, M. Overview of Materials Adrià Pacheco − Institut de Ciència de Materials de Barcelona and Power Applications of Coated Conductors Project. Jpn. J. Appl. Phys. 2012, 51, No. 010007. (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, (2) Obradors, X.; Puig, T. Coated Conductors for Power Catalonia, Spain Applications: Materials Challenges. Supercond. Sci. Technol. 2014, Kapil Gupta − Institut de Ciència de Materials de Barcelona 27, No. 044003. (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, (3) Ito, S.; Hashizume, H.; Yanagi, N.; Tamura, H. Advanced High- Catalonia, Spain Temperature Superconducting Magnet for Fusion Reactors: Segment Lavinia Saltarelli − Institut de Ciència de Materials de Fabrication and Joint Technique. Fusion Eng. Des. 2018, 136, 239− Barcelona (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, 246. Catalonia, Spain (4) Haran, K. S.; Kalsi, S.; Arndt, T.; Karmaker, H.; Badcock, R.; Diana Garcia − Institut de Ciència de Materials de Barcelona Buckley, B.; Haugan, T.; Izumi, M.; Loder, D.; Bray, J. W. High Power (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, Density Superconducting Rotating MachinesDevelopment Status Catalonia, Spain and Technology Roadmap. Supercond. Sci. Technol. 2017, 30, No. 123002. Núria Alcalde − Institut de Ciència de Materials de Barcelona (5) Jha, A. K.; Matsumoto, K. Superconductive REBCO Thin Films (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, and Their Nanocomposites: The Role of Rare-Earth Oxides in Catalonia, Spain Promoting Sustainable Energy. Front. Phys. 2019, 7, 82. Cristian Mocuta − Synchrotron SOLEIL, L’Orme des (6) Cayado, P.; Erbe, M.; Kauffmann-Weiss, S.; Jung, A.; Hänisch, J.; Merisiers Saint-Aubin, 91192 Gif-sur-Yvette, France; Holzapfel, B. Chemical Solution Deposition of Y1−xGdxBa2Cu3O7−δ- orcid.org/0000-0001-5540-449X BaHfO3 Nanocomposite Films: Combined Influence of Nanoparticles Susagna Ricart − Institut de Ciència de Materials de Barcelona and Rare-Earth Mixing on Growth Conditions and Transport (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, Properties. RSC Adv. 2018, 8, No. 42398. Catalonia, Spain; orcid.org/0000-0003-4196-2081 (7) Cayado, P.; Erbe, M.; Kauffmann-Weiss, S.; Bühler, C.; Jung, A.; Flavio Pino − Institut de Ciència de Materials de Barcelona Hänisch, J.; Holzapfel, B. Large Critical Current Densities and Pinning Forces in CSD-Grown GdBa2Cu3O7−x-BaHfO3 Nano- (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, composite Films. Supercond. Sci. Technol. 2017, 30, No. 094007. Catalonia, Spain (8) Miura, S.; Yoshida, Y.; Ichino, Y.; Tsuruta, A.; Matsumoto, K.; Xavier Obradors − Institut de Ciència de Materials de Ichinose, A.; Awaji, S. Vortex Pinning at Low Temperature under Barcelona (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, High Magnetic Field in SmBa2Cu3Oy Superconducting Films with Catalonia, Spain; orcid.org/0000-0003-4592-7718 High Number Density and Small Size of BaHfO3 Nano-Rods. Complete contact information is available at: Supercond. Sci. Technol. 2015, 28, No. 114006. (9) Majkic, G.; Pratap, R.; Xu, A.; Galstyan, E.; Higley, H. C.; https://pubs.acs.org/10.1021/acsami.0c18014 Prestemon, S. O.; Wang, X.; Abraimov, D.; Jaroszynski, J.; Selvamanickam, V. Engineering Current Density over 5 kA mm-2 at Author Contributions 4.2 K, 14 T in Thick Film REBCO Tapes. Supercond. Sci. Technol. This manuscript was written through contributions of all 2018, 31, No. 10LT01. authors. All authors have given approval to the final version of (10) Dong, Z.; Ding, F.; Zhang, H.; Shang, H.; Huang, D.; Xu, W.; the manuscript. Li, T.; Zou, Q.; Gu, H. Preparation of High Performance YGdBCO Films by Low Fluorine TFA-MOD Process. J. Rare Earths 2020, 38, Notes 755−762. The authors declare no competing financial interest. (11) Matsushima, K.; Taka, C.; Nishida, A. Variations of ■ ACKNOWLEDGMENTS The authors acknowledge the European Research Council for Superconducting Transition Temperature in YbBa2Cu3O7−δ Ceramics by Gd substitution. J. Phys.: Conf. Ser. 2018, 969, No. 012059. (12) Zhang, S.; Xu, S.; Fan, Z.; Jiang, P.; Han, Z.; Yang, G.; Chen, Y. Broad Temperature Study of RE-Substitution Effects on the In-Field the ULTRASUPERTAPE project (ERC-2014-ADG-669504) Critical Current Behavior of REBCO Superconducting Tapes. and EU COST action for CA16218 (NANOCOHYBRI). We Supercond. Sci. Technol. 2018, 31, No. 125006. also acknowledge financial support from the Spanish Ministry (13) Miura, M.; Kato, T.; Yoshizumi, M.; Yamada, Y.; Izumi, T.; of Economy and Competitiveness, and Spanish Ministry of Hirayama, T.; Shiohara, Y. Rare Earth Substitution Effects and Science, Innovation and Universities through the “Severo Magnetic Field Dependence of Critical Current in Y1−xRExBa2Cu3Oy Ochoa” Program for Centers of Excellence in R&D (SEV- Coated Conductors with Nanoparticles (RE=Sm, Gd). Appl. Phys. 2015-0496 and CEX2019-000917-S), SUMATE project Express 2009, 2, No. 023002. (RTI2018-095853-B-C21, co-financed by the European (14) Miura, M.; Kato, T.; Yoshizumi, M.; Yamada, Y.; Izumi, T.; Regional Development Fund). Synchrotron SOLEIL is Hirayama, T.; Shiohara, Y. Magnetic Field Dependence of Critical Current and Microstructure in TFA-MOD Y1−xSmxBa2Cu3Oy with acknowledged for granting the beamtime (under project no. Nanoparticles for Coated Conductors. IEEE Trans. Appl. Supercond. 20190274) and P. Joly for setup assistance at the DiffAbs 2009, 19, 3275−3278. beamline. We also acknowledge support from the Catalan (15) Pavan Kumar Naik, S.; Muralidhar, M.; Jirsa, M.; Murakami, M. Government with 2017-SGR-1519 and the energy Catalan Enhanced Flux Pinning of Single Grain Bulk (Gd, Dy)BCO network E4S. A.Q. thanks the Spanish Ministry of Science, Superconductors Processed by Cold-Top-Seeded Infiltration Growth Innovation and Universities for his “Juan de la Cierva” Method. Mater. Sci. Eng., B 2020, 253, No. 114494. 9110 https://dx.doi.org/10.1021/acsami.0c18014 ACS Appl. Mater. Interfaces 2021, 13, 9101−9112 ACS Applied Materials & Interfaces www.acsami.org Research Article (16) Pop, C.; Villarejo, B.; Pino, F.; Mundet, B.; Ricart, S.; de Palau, (34) Liu, P.; Guo, B.; An, T.; Fang, H.; Zhu, G.; Jiang, C.; Jiang, X. M.; Puig, T.; Obradors, X. Growth of All-Chemical High Critical High Throughput Materials Research and Development for Lithium Current YBa2Cu3O7−δ Thick Films and Coated Conductors. Super- Ion Batteries. J. Materiomics 2017, 3, 202−208. cond. Sci. Technol. 2019, 32, No. 015004. (35) Heinselman, K. N.; Lany, S.; Perkins, J. D.; Talley, K. R.; (17) Lei, M.; Zhang, Y.; Zhao, Y. Performance of SCO Buffers Zakutayev, A. Thin Film Synthesis of Semiconductors in the Mg− Prepared by Dip Coating and Slot-Die Coating Methods. J. Supercond. Sb−N Materials System. Chem. Mater. 2019, 31, 8717−8724. Novel Magn. 2016, 29, 2587−2590. (36) Talley, K. R.; Sherbondy, R.; Zakutayev, A.; Brennecka, G. L. (18) Solovyov, V.; Dimitrov, I. K.; Li, Q. Growth of Thick Review of High-Throughput Approaches to Search for Piezoelectric YBa2Cu3O7 Layers Via a Barium Fluoride Process. Supercond. Sci. Nitrides. J. Vac. Sci. Technol., A 2019, 37, No. 060803. Technol. 2013, 26, No. 013001. (37) Han, Y.; Bauers, S.; Zhang, Q.; Zakutayev, A. High-Throughput (19) Yun, K. S.; Choi, B. D.; Matsumoto, Y.; Song, J. H.; Kanda, N.; Fabrication and Semi-Automated Characterization of Oxide Thin Itoh, T.; Kawasaki, M.; Chikyow, T.; Ahmet, P.; Koinuma, H. Vapor− Film Transistors. Chin. Phys. B 2020, 29, No. 018502. Liquid−Solid Tri-Phase Pulsed-Laser Epitaxy of RBa2Cu3O7−y Single- (38) Jiang, C.; Wang, R.; Parkinson, B. A. Combinatorial Approach Crystal Films. Appl. Phys. Lett. 2002, 80, 61−63. to Improve Photoelectrodes Based on BiVO4. ACS Comb. Sci. 2013, (20) Kursumovic, A.; Tomov, R. I.; Hühne, R.; MacManus-Driscoll, 15, 639−645. J. L.; Glowacki, B. A.; Evetts, J. E. Hybrid Liquid Phase Epitaxy (39) Ndione, P. F.; Ratcliff, E. L.; Dey, S. R.; Warren, E. L.; Peng, Processes for YBa2Cu3O7 Film Growth. Supercond. Sci. Technol. 2004, H.; Holder, A. M.; Lany, S.; Gorman, B. P.; Al-Jassim, M. M.; 17, 1215−1223. Deutsch, T. G.; Zakutayev, A.; Ginley, D. S. High-Throughput (21) Soler, L.; Jareño, J.; Banchewski, J.; Rasi, S.; Chamorro, N.; Experimental Study of Wurtzite Mn1−xZnxO Alloys for Water Splitting Guzman, R.; Yáñez, R.; Mocuta, C.; Ricart, S.; Farjas, J.; Roura- Applications. ACS Omega 2019, 4, 7436−7447. Grabulosa, P.; Obradors, X.; Puig, T. Ultrafast Transient Liquid (40) Yuan, J.; Stanev, V.; Gao, C.; Takeuchi, I.; Jin, K. Recent Assisted Growth of High Current Density Superconducting Films. Advances in High-Throughput Superconductivity Research. Super- Nat. Commun. 2020, 11, No. 344. cond. Sci. Technol. 2019, 32, No. 123001. (22) Evans, J. R. G.; Edirisinghe, M. J.; Coveney, P. V.; Eames, J. (41) Zhigadlo, N. D.; Iranmanesh, M.; Assenmacher, W.; Mader, W.; Combinatorial Searches of Inorganic Materials Using the Ink-Jet Hulliger, J. Exploring Multi-Component Superconducting Com- Printer: Science, Philosophy and Technology. J. Eur. Ceram. Soc. pounds by a High-Pressure Method and Ceramic Combinatorial 2001, 21, 2291−2299. Chemistry. J. Supercond. Novel Magn. 2016, 30, 79−84. (23) Agrawal, A.; Choudhary, A. Perspective: Materials Informatics (42) Barber, Z. H.; Blamire, M. G. High Throughput Thin Film and Big Data: Realization of the “Fourth Paradigm” of Science in Materials Science. Mater. Sci. Technol. 2013, 24, 757−770. (43) McGinn, P. J. Thin-Film Processing Routes for Combinatorial Materials Science. APL Mater. 2016, 4, No. 053208. (24) Himanen, L.; Geurts, A.; Foster, A. S.; Rinke, P. Data-Driven Materials Investigations-A Review. ACS Comb. Sci. 2019, 21, 501− 515. Materials Science: Status, Challenges, and Perspectives. Adv. Sci. (44) Shin, D. H.; Woo, S.; Yem, H.; Cha, M.; Cho, S.; Kang, M.; 2019, 6, No. 1900808. Jeong, S.; Kim, Y.; Kang, K.; Piao, Y. A Self-Reducible and Alcohol- (25) Green, M. L.; et al. Fulfilling the Promise of the Materials Soluble Copper-Based Metal-Organic Decomposition Ink for Printed Genome Initiative with High-Throughput Experimental Method- Electronics. ACS Appl. Mater. Interfaces 2014, 6, 3312−3319. ologies. Appl. Phys. Rev. 2017, 4, No. 011105. (45) Calleja, A.; Ricart, S.; Palmer, X.; Luccas, R. F.; Puig, T.; (26) Hattrick-Simpers, J. R.; Gregoire, J. M.; Kusne, A. G. Obradors, X. Water Determination of Precursor Solutions with Perspective: Composition−Structure−Property Mapping in High- Oxidant Cations by the Karl Fischer Method: The YBCO-TFA Case. Throughput Experiments: Turning Data into Knowledge. APL Mater. J. Sol-Gel Sci. Technol. 2010, 53, 347−352. 2016, 4, No. 053211. (46) Basolo, S.; Bérar, J.-F.; Boudet, N.; Breugnon, P.; Caillot, B.; (27) Hattrick-Simpers, J. R.; et al. An Inter-Laboratory Study of Zn- Clemens, J.-C.; Delpierre, P.; Dinkespiler, B.; Koudobine, I.; Meesen, Sn-Ti-O Thin Films Using High-Throughput Experimental Methods. C.; Menouni, M.; Mouget, C.; Pangaud, P.; Potheau, R.; Vigeolas, E. ACS Comb. Sci. 2019, 21, 350−361. XPAD: Pixel Detector for Material Sciences. IEEE Trans. Nucl. Sci. (28) Vasudevan, R. K.; Choudhary, K.; Mehta, A.; Smith, R.; Kusne, 2005, 52, 1994−1998. G.; Tavazza, F.; Vlcek, L.; Ziatdinov, M.; Kalinin, S. V.; Hattrick- (47) Pangaud, P.; Basolo, S.; Boudet, N.; Bérar, J.-F.; Chantepie, B.; Simpers, J. Materials Science in the Artificial Intelligence Age: High- Clemens, J.-C.; Delpierre, P.; Dinkespiler, B.; Medjoubi, K.; Hustache, Throughput Library Generation, Machine Learning, and a Pathway S.; Menouni, M.; Morel, C. XPAD3-S: A Fast Hybrid Pixel Readout from Correlations to the Underpinning Physics. MRS Commun. 2019, Chip for X-Ray Synchrotron Facilities. Nucl. Instrum. Methods Phys. 9, 821−838. Res., Sect. A 2008, 591, 159−162. (29) Bag, M.; Jiang, Z.; Renna, L. A.; Jeong, S. P.; Rotello, V. M.; (48) Jang, D.; Kim, D.; Moon, J. Influence of Fluid Physical Venkataraman, D. Rapid Combinatorial Screening of Inkjet-Printed Properties on Ink-Jet Printability. Langmuir 2009, 25, 2629−2635. Alkyl-Ammonium Cations in Perovskite Solar Cells. Mater. Lett. 2016, (49) Mattana, G.; Loi, A.; Woytasik, M.; Barbaro, M.; Noël, V.; Piro, 164, 472−475. B. Inkjet-Printing: A New Fabrication Technology for Organic (30) Sánchez-Díaz, A.; Rodríguez-Martínez, X.; Córcoles-Guija, L.; Transistors. Adv. Mater. Technol. 2017, 2, No. 1700063. Mora-Martín, G.; Campoy-Quiles, M. High-Throughput Multi- (50) Liu, Y.; Derby, B. Experimental Study of the Parameters for parametric Screening of Solution Processed Bulk Heterojunction Stable Drop-On-Demand Inkjet Performance. Phys. Fluids 2019, 31, Solar Cells. Adv. Electron. Mater. 2018, 4, No. 1700477. No. 032004. (31) Samoilenko, Y.; Yeung, G.; Zakutayev, A.; Reese, M. O.; (51) Yunker, P. J.; Still, T.; Lohr, M. A.; Yodh, A. G. Suppression of Wolden, C. A. In Combinatorial Study of MZO Emitters for CdTe-Based the Coffee-Ring Effect by Shape-Dependent Capillary Interactions. Solar Cells, 2019 IEEE 46th Photovoltaic Specialists Conference Nature 2011, 476, 308−311. (PVSC), 16−21 June 2019; pp 2498−2502. (52) Villarejo, B. Synthesis of Defect Free YBa2Cu3O7‑x Films over 1 μm (32) Brown, C. R.; McCalla, E.; Watson, C.; Dahn, J. R. by CSD Using Inkjet Printing; Universitat Autònoma de Barcelona: Combinatorial Study of the Li-Ni-Mn-Co Oxide Pseudoquaternary Bellaterra, 2018. System for Use in Li-Ion Battery Materials Research. ACS Comb. Sci. (53) Calvert, P. Inkjet Printing for Materials and Devices. Chem. 2015, 17, 381−391. Mater. 2001, 13, 3299−3305. (33) Carey, G. H.; Dahn, J. R. Combinatorial Synthesis of Mixed (54) Gómez-Núñez, A.; Alonso-Gil, S.; López, C.; Roura, P.; Vilà, A. Transition Metal Oxides for Lithium-Ion Batteries. ACS Comb. Sci. Role of Ethanolamine on the Stability of a Sol−Gel ZnO Ink. J. Phys. 2011, 13, 186−189. Chem. C 2017, 121, 23839−23846. 9111 https://dx.doi.org/10.1021/acsami.0c18014 ACS Appl. Mater. Interfaces 2021, 13, 9101−9112 ACS Applied Materials & Interfaces www.acsami.org Research Article (55) Obradors, X.; Puig, T.; Ricart, S.; Coll, M.; Gazquez, J.; Palau, A.; Granados, X. Growth, Nanostructure and Vortex Pinning in Superconducting YBa2Cu3O7 Thin Films Based on Trifluoroacetate Solutions. Supercond. Sci. Technol. 2012, 25, No. 123001. (56) Rasi, S.; Soler, L.; Jareño, J.; Banchewski, J.; Guzmán, R.; Mocuta, C.; Kreuzer, M.; Ricart, S.; Roura, P.; Farjas, J.; Obradors, X.; Puig, T. Relevance of the Formation of Intermediate Non-Equilibrium Phases in YBCO Film Growth by Transient Liquid Assisted Growth. J. Phys. Chem. C 2020, 124, 15574−15584. (57) Grivel, J. C. Thermal Decomposition of RE(C2H5CO2)3·H2O (RE = Dy, Tb, Gd, Eu and Sm. J. Therm. Anal. Calorim. 2014, 115, 1253−1264. (58) Rasi, S.; Ricart, S.; Obradors, X.; Puig, T.; Roura, P.; Farjas, J. Thermal Decomposition of Yttrium Propionate: Film and Powder. J. Anal. Appl. Pyrolysis 2018, 133, 225−233. (59) Van der Walt, S.; Colbert, S. C.; Varoquaux, G. The Numpy Array: A Structure for Efficient Numerical Computation. Comput. Sci. Eng. 2011, 13, 22−30. (60) Reback, J.; et al. pandas-dev/pandas: Pandas 1.0. 5; Zenodo, 2020. (61) Hunter, J. D. Matplotlib: A 2D Graphics Environment. Comput. Sci. Eng. 2007, 9, 90−95. (62) Maier, W. F. Early Years of High-Throughput Experimentation and Combinatorial Approaches in Catalysis and Materials Science. ACS Comb. Sci. 2019, 21, 437−444. (63) Hanak, J. J. The “Multiple-Sample Concept” in Materials Research: Synthesis, Compositional Analysis and Testing of Entire Multicomponent Systems. J. Mater. Sci. 1970, 5, 964−971. (64) Mao, S. S.; Burrows, P. E. Combinatorial Screening of Thin Film Materials: An Overview. J. Materiomics 2015, 1, 85−91. (65) Vegard, L. Die Konstitution Der Mischkristalle Und Die Raumfüllung Der Atome. Z. Phys. 1921, 5, 17−26. (66) Nelson, J. B.; Riley, D. P. An Experimental Investigation of Extrapolation Methods in the Derivation of Accurate Unit-Cell Dimensions of Crystals. Proc. Phys. Soc. 1945, 57, 160−177. (67) Llordés, A.; Palau, A.; Gázquez, J.; Coll, M.; Vlad, R.; Pomar, A.; Arbiol, J.; Guzmán, R.; Ye, S.; Rouco, V.; Sandiumenge, F.; Ricart, S.; Puig, T.; Varela, M.; Chateigner, D.; Vanacken, J.; Gutiérrez, J.; Moshchalkov, V.; Deutscher, G.; Magen, C.; Obradors, X. Nanoscale Strain-Induced Pair Suppression as a Vortex-Pinning Mechanism in High-Temperature Superconductors. Nat. Mat. 2012, 11, 329−336. (68) Shiohara, Y.; Endo, A. Crystal Growth of Bulk High-Tc Superconducting Oxide Materials. Mater. Sci. Eng., R 1997, 19, 1−86. (69) Gregoire, J. M.; Van Campen, D. G.; Miller, C. E.; Jones, R. J.; Suram, S. K.; Mehta, A. High-Throughput Synchrotron X-Ray Diffraction for Combinatorial Phase Mapping. J. Synchrotron Radiat. 2014, 21, 1262−1268. (70) Lyu, Y.; Liu, Y.; Cheng, T.; Guo, B. High-Throughput Characterization Methods for Lithium Batteries. J. Materiomics 2017, 3, 221−229. (71) Cook, P. K.; Mocuta, C.; Dufour, E.; Languille, M.-A.; Bertrand, L. Full-Section Otolith Microtexture Imaged by Local-Probe X-Ray Diffraction. J. Appl. Crystallogr. 2018, 51, 1182−1196. (72) Gueriau, P.; Réguer, S.; Leclercq, N.; Cupello, C.; Brito, P. M.; Jauvion, C.; Morel, S.; Charbonnier, S.; Thiaudière, D.; Mocuta, C. Visualizing Mineralization Processes and Fossil Anatomy Using Synchronous Synchrotron X-Ray Fluorescence and X-Ray Diffraction Mapping. J. R. Soc. Interface 2020, 17, No. 20200216. 9112 https://dx.doi.org/10.1021/acsami.0c18014 ACS Appl. Mater. Interfaces 2021, 13, 9101−9112