RESEARCH ARTICLE www.afm-journal.de On the Role of the Sr3−xCaxAl2O6 Sacrificial Layer Composition in Epitaxial La0.7Sr0.3MnO3 Membranes Pol Salles, Roger Guzman, Aleix Barrera, Martí Ramis, Jose Manuel Caicedo, Anna Palau, Wu Zhou, and Mariona Coll* 1. Introduction The possibility to fabricate freestanding single crystal complex oxide films has raised enormous interest to be integrated in next-generation electronic Next generation electronic devices are fac- ing a rapidly advancing technological de- devices envisaging distinct and novel properties that can deliver velopment to respond to the evolving unprecedented performance improvement compared to traditional needs of today’s society and its future semiconductors. The use of the water-soluble Sr3 Al2 O6 (SAO) sacrificial layer expected challenges. Within these tech- to detach the complex oxide film from the growth substrate has significantly nological advances, novel and multifunc- expanded the complex oxide perovskite membranes library. Nonetheless, the tional device concepts are starting to at- tract interest.[1–3] Complex transition-metal extreme water sensitivity of SAO hinders its manipulation in ambient oxides, in particular ABO3 perovskite ox- conditions and restricts the deposition approaches to those using high ides, are a promising family of materials vacuum. This study presents a pioneering study on the role of Ca-substitution that offer a wide range of properties in- in solution processed SAO (Sr3−x Cax Al2 O6 with x ⩽ 3) identifying a noticeable cluding ferroelectricity,[4] multiferroicity,[5] improvement on surface film crystallinity preserving a smooth surface metal-insulator transitions,[6] magnetism, morphology while favoring the manipulation in a less-restricted ambient and superconductivity,[7] to name a few.[8] These properties are affected by the ox- conditions. Then, the study focuses on the effect of the sacrificial composition ide structure, crystalline quality, and pres- on the subsequent ex situ deposition of La0.7 Sr0.3 MnO3 (LSMO) by pulsed ence of defects requiring the growth of laser deposition, to obtain epitaxial films with a variable degree of strain. such materials to be precise, constrained Finally, epitaxial and strain-free LSMO membranes with metal-insulator on specific crystalline substrates and to be transition at 290 K are delivered. This study offers a hybrid and versatile processed at high temperature[9–12] which significantly limit their unprecedented op- approach to prepare and easily manipulate crystalline perovskite oxide portunities. Nonetheless, in the last few membranes by facilitating ex situ growth on SAO-based sacrificial layer. years this field has experienced a major turn boosted by the development of syn- thetic approaches to detach these epitax- ial oxides from the growth substrate and freely manipulate them.[13–15] In fact, the fabrication of new artificial heterostructures by assembling very P. Salles, A. Barrera, M. Ramis, A. Palau, M. Coll ICMAB-CSIC dissimilar materials, their integration in conventional materi- Campus UAB, Bellaterra, Barcelona 08193, Spain als such as silicon, plastics, and 2D semiconductors,[16–18] and E-mail: mcoll@icmab.es the possibility to obtain spontaneous shaped structures [19] is R. Guzman, W. Zhou becoming a reality opening the possibility to study novel in- School of Physical Sciences terface phenomena, nanoengineer their properties, and un- University of Chinese Academy of Sciences lock strain states that might not be available by traditional Beijing 100049, China epitaxial growth.[20–22] J. M. Caicedo Catalan Institute of Nanoscience and Nanotechnology (ICN2) The use of a sacrificial layer is an appealing approach to Campus UAB, Bellaterra 08193, Barcelona, Spain fabricate and manipulate freestanding epitaxial complex ox- ides membranes.[23] This method consists of adding a buffer The ORCID identification number(s) for the author(s) of this article layer between the substrate and the complex oxide that en- can be found under https://doi.org/10.1002/adfm.202304059 sures epitaxial growth and the subsequent detachment upon © 2023 The Authors. Advanced Functional Materials published by selective etching. Recently, Sr3 Al2 O6 (SAO) has been proved Wiley-VCH GmbH. This is an open access article under the terms of the successful to be used as water-soluble sacrificial layer.[24–26] Creative Commons Attribution-NonCommercial-NoDerivs License, SAO has a pseudoperovskite structure (Pa-3) with a cubic which permits use and distribution in any medium, provided the original unit cell (a = 15.844 Å; a/4 = 3.961 Å)[27] which is close to work is properly cited, the use is non-commercial and no modifications or adaptations are made. four times the lattice constant of SrTiO3 (STO, a = 3.905 Å), a typical monocrystalline substrate used for the growth of DOI: 10.1002/adfm.202304059 different perovskite oxides.[28] The SAO structure consists of Adv. Funct. Mater. 2023, 33, 2304059 2304059 (1 of 10) © 2023 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH 16163028, 2023, 41, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adfm.202304059 by Readcube (Labtiva Inc.), Wiley Online Library on [23/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.afm-journal.de discrete Al-O networks based on rings of six AlO4 tetrahedra croscopy (STEM), x-ray diffraction (XRD) and atomic force mi- (Al6 O18− 18 ) with two nonbridging oxygen bonds with Sr2 + ions. croscopy (AFM). Then it is discussed how SCx AO composition Because Al3 + is more electronegative than Sr2 + , Sr-O bonding affects the ex situ deposition of pulsed laser deposited (PLD) has a more ionic character than Al-O. When exposed to wa- LSMO films on their crystallinity, strain, interface cation diffu- ter, the easy protonation of this Sr-O ionic bonding is the re- sion, and electrical properties. Finally, the properties of the exfo- sponsible of the fast hydrolysis of the structure, which makes liated LSMO membranes are presented, demonstrating the fea- SAO a suitable sacrificial layer to be etched with water.[29,30] sibility to integrate high quality epitaxial PLD-LSMO flakes on However, despite these wealthy properties, SAO still presents arbitrary substrates with a metal-insulator transition as high as some challenges that need to be overcome to exploit its full 290 K. potential. First, while its high reactivity with water makes it almost the 2. Results and Discussion perfect candidate for selective etching against many perovskite oxides,[23] it also helps the fast development of an amorphous 2.1. Structural Analysis capping layer when exposed to air.[31] This reactivity hinders the subsequent epitaxial growth of the perovskite oxide. The use of Figure 1a–d shows the scheme of the fabrication process of epi- a protective SrTiO3 layer [32] can partially help fixing the sac- taxial LSMO membranes. First, the Sr3−x Cax Al2 O6 (SCx AO, x ⩽ 3) rificial instability issue, although it brings some limitations to sacrificial layer is deposited and grown on a (001) SrTiO3 (STO) prepare stacked membrane heterostructures with pristine inter- substrate by solution processing (CSD), Figure 1a. Then, the faces. [33] Also, the softness of the SAO structure can give rise film is brought into a PLD chamber where an in-vacuum high- to cation interdiffusion at the interface during the high temper- temperature re-crystallization process is performed, Figure 1b, ature growth of the complex oxide, as it was reported for the followed by the PLD-LSMO deposition, Figure 1c. Finally, the La0.7 Sr0.3 MnO3 /SAO system,[34] which can hinder the membrane sacrificial layer is water-etched to obtain millimeter sized LSMO exfoliation and alters its functional properties. Finally, the growth membranes supported on a polyethylene terephthalate (PET) of complex oxides on mismatched SAO can produce cracked sheet, Figure 1d. These membranes can be subsequently trans- and wrinkled membranes upon exfoliation as a mechanism for ferred to other substrates such as glass and silicon (See Experi- strain release. [20] It is worth noting that the detachment of a mental Section for further details). faultless membrane, regardless of the nature of the sacrificial The surface crystallinity of the films at each stage of the growth layer, remains a key challenge in the preparation of freestanding process has been monitored by RHEED. Figure 1e–g shows the films.[35,36] evolution of the surface crystallinity for the system with SAO sac- A simple way to modify the properties of SAO is to engineer rificial layer. The as-grown SAO film at 800°C and exposed to air the metal-oxygen bond through cation substitution. The higher shows no spots indicating an amorphous surface. The film after the M2 + electronegativity (Ca2 + > Sr2 + > Ba2 + ), the lower its vacuum thermal annealing at 825°C results in a spotty pattern re- bonding ionicity with the nonbridging oxygens of the structure, vealing that the surface recrystallizes, as previously reported.[31] and thus it is more difficult to hydrolyze. Consequently, Ca3 Al2 O6 Then, the LSMO film is deposited by PLD on the reconstructed is dissolved much slower in water than Ba3 Al2 O6 .[37] Addition- SAO obtaining a new spotty pattern that discloses the formation ally, the difference in cation size (Ba2 + > Sr2 + > Ca2 + ) enables of an oriented LSMO film, Figure 1g. The same process has been tuning the SAO lattice parameter ranging from a/4 = 4.125 Å followed to study the behavior of the SCx AO series with x ⩽ 3 and for Ba3 Al2 O6 to a/4 = 3.816 Å for Ca3 Al2 O6 [30] providing a plat- the subsequent growth of the LSMO. From the acquired RHEED form of sacrificial layers that can accommodate many different patterns (Figure 2) it is observed that crystalline films are ob- perovskite oxides with different strain states being a powerful tained for all sacrificial layer composition after the vacuum an- tool to modulate the physical properties of the epitaxial oxide nealing and the crystallinity is transferred to the LSMO films. membrane.[20,37–39] The periodicity and intensity of the spots identified in the sac- La0.7 Sr0.3 MnO3 (LSMO) is an attractive perovskite oxide with rificial layer pattern moderately change as Ca is incorporated in extraordinary electrical and magnetic properties making the SCx AO, anticipating changes in the crystallinity and lattice pa- material an appealing candidate for spintronics and mem- rameter. Slight differences are also appreciated in the shape of ory devices.[6,40–42] The growth of high quality LSMO mem- the diffraction spots of the LSMO films, which could be attributed branes is also of crucial importance where slight variation in to a change in the surface morphology. [55] From the AFM topo- strain,[43–45] crystal orientation,[46] oxygen stoichoimetry[47,48] and graphic images it is observed that, prior to the LSMO depositions, Sr doping[49] can significantly affect its properties. The prepara- all SCx AO films are dense, smooth and homogeneous with a root- tion of flexible and epitaxial LSMO has already been attempted mean-square (rms) roughness of ≈1 nm (Figure S1a-d, Support- by high-vacuum deposition processes using mica substrates[50,51] ing Information). After the LSMO deposition, Figure S1e–h (Sup- and sacrificial layers such as YBa2 Cu3 O7 [52] and SAO.[24,34,53,54] porting Information), the films are homogeneous with smooth Here we go one step further in developing a hybrid and versa- surfaces of rms ≈2 nm and the size of the grains moderately in- tile synthetic approach for LSMO membranes, based on solution- creases for the LSMO grown on CAO, which is in agreement with processed Sr3−x Cax Al2 O6 with x ⩽ 3 (SCx AO) sacrificial lay- the shape of the RHEED pattern. ers in which ambient stability and crystallinity are improved A separate study was performed to investigate the influence upon Ca-substitution while preserving smooth surface morphol- of Ca substitution on the SCx AO ambient stability. As a case ex- ogy, demonstrated by means of reflection high energy elec- ample it is discussed the evolution for SC2 AO composition (see tron diffraction (RHEED), scanning transmission electron mi- Figure S2, Supporting Information). The as-prepared SC2 AO Adv. Funct. Mater. 2023, 33, 2304059 2304059 (2 of 10) © 2023 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH 16163028, 2023, 41, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adfm.202304059 by Readcube (Labtiva Inc.), Wiley Online Library on [23/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.afm-journal.de Figure 1. a–d) Sketch of the process followed to fabricate LSMO membranes: a) SCx AO is grown by CSD on a (001) STO substrate at 800 °C and exposed to air. Then, it is brought to the PLD chamber where b) an annealing treatment up to 825 °C is performed to achieve re-crystallization of the SCx AO surface and c) LSMO is grown by PLD. d) Finally, the LSMO film can be attached on a polymer support and separated from its original substrate by etching the SCx AO sacrificial layer in water. Image of a 5 × 5 mm2 LSMO membrane on a PET support after etching the sacrificial layer. e–g) In situ RHEED analysis of e) SAO after air exposure, f) SAO after vacuum annealing and g) LSMO deposited by PLD. films after being annealed in vacuum were exposed to air and analysis (XRD). Figure 3a shows the XRD 𝜃-2𝜃 analysis for the sealed in plastic bags. RHEED analyses were performed after four SCx AO systems comparing the bare STO contribution, the 7, 16, and 30 days and the samples were stored in sealed plas- SCx AO//STO and the LSMO/SCx AO//STO. The most intense tic bags between experiments. Using this procedure, the surface peak at 2𝜃= 46.5° corresponds to the (002) STO Bragg reflection quality of the samples containing Ca was preserved up to 16 days, and the shoulders appearing next to the STO after growing the Figure S2a (Supporting Information). Longer storage time (i.e. 30 different layers can be identified as (008) SCx AO and (002) LSMO, days) results in sample degradation as the spots in the RHEED indicated with ▽ and *, respectively, in Figure 3a. Focusing on pattern are significantly less intense and the surface morphol- the SCx AO//STO system, it is observed that by increasing the Ca ogy roughens dramatically. Note that by investigating the RHEED concentration, the 2𝜃 values of the (008) SCx AO reflection shifts pattern of the whole series of vacuum-annealed SCx AO films af- to higher angles. The shifts in 2𝜃 agree with a change in the cSCx AO ter seven days it is found a clear trend in surface crystallinity. The lattice parameter from cSAO /4 = 3.96 Å (x = 0) to cCAO /4 =3.81 Å (x higher the Ca concentration, the more intense the RHEED spots = 3). Texture analysis of the SCx AO films is presented in Table S1, and therefore, more stable (Figure S2b, Supporting Information), Supporting Information. in agreement with the expected diminished water sensitivity ac- After the deposition of the LSMO layer on the in situ vac- cording to the Ca doping.[30,37] Thus, this study opens the possi- uum annealed SCx AO//STO (LSMO/SCx AO//STO), the (002) bility to work in a less-strict environment for the sacrificial layers. LSMO Bragg reflection is observed for all the samples indicat- Further structural analysis to study the influence of the sac- ing that c-axis oriented growth has been achieved regardless of rificial layer composition on the film crystallinity and subse- the Ca composition in the sacrificial, in well agreement with quent growth of LSMO, was carried out by X-ray diffraction the spotty-streaky RHEED patterns shown in Figure 2e–g. It is Figure 2. RHEED analysis along the [100] STO of the different SCx O sacrificial layers a) after air exposure, b–d) after vacuum annealing and e–g) LSMO deposited by PLD. Adv. Funct. Mater. 2023, 33, 2304059 2304059 (3 of 10) © 2023 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH 16163028, 2023, 41, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adfm.202304059 by Readcube (Labtiva Inc.), Wiley Online Library on [23/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.afm-journal.de Figure 3. a) XRD 𝜃-2𝜃 scans for the LSMO/SCx AO//STO heterostructures at each step of the process. ▽ indicates the (008) SCx AO diffraction peak and * indicates the main (002) LSMO diffraction peak in each case. b) Reciprocal space map around (103) STO reflection for the different LSMO/SCx AO//STO heterostructures. (103) LSMO and (4 0 12) SCx AO reflections, indicated in each case, have been used to extract c) the lattice parameters of the SCx AO and LSMO films. d) Strain in-plane ϵxx versus strain out-of-plane ϵzz plot for Poisson’s ratio calculation of LSMO films. noted that the intensity of the (002) LSMO Bragg reflection in the Figure 3b. The most intense peak corresponds to the (103) STO SAO system is slightly lower than the rest. Also, the (008) SAO Bragg reflection, whereas less intense and defined peaks appear Bragg reflection for the same system shows a moderate shift to- for (103) LSMO and (4 0 12) SCx AO. It is noted that the higher ward higher 2𝜃 values. This could indicate that reactivity at the the Ca concentration in the sacrificial layer, the more resolved the LSMO/SAO occurred, which is in agreement with previous stud- (4 0 12) SCx AO and (103) LSMO reflections, strengthening the ies on all-vacuum deposited LSMO/SAO system.[34] Also, the het- results of the texture analysis that already indicated an increase erostructure LSMO/CAO//STO shows an additional peak that in crystalline quality. From the RSM it has been extracted the appears at 2𝜃 = 47.2° which might be attributed to the forma- in-plane (aLSMO and aSCx AO ) and out-of-plane (cLSMO and cSCx AO ) tion of a thin layer of a mixed perovskite La-Sr-Mn-Ca-Al-O at lattice parameters and compared to the respective bulk values, the interface.[56] Texture analysis of the LSMO films reveals an shown in Figure 3c. For the SCx AO films (cyan symbols) both a improvement of crystalline quality when deposited on Ca-doped and c lattice parameters decrease by increasing the Ca concen- SAO, (Table S1 and Figure S3a, Supporting Information) and tration consistent with the change observed from Figure 3a and no further secondary phases are observed in extended XRD 𝜃-2𝜃 in well agreement with Vegard’s Law. [57] On the other hand, the scans (Figure S3b, Supporting Information). Remarkably, the 2𝜃 lattice parameters of LSMO films (purple symbols) follow a dif- position of the (002) LSMO reflection shifts as a function of the ferent trend. The lattice parameters of LSMO on SAO and SC1 AO SCx AO composition suggesting a variation in the LSMO lattice are close to those of the LSMO bulk whereas those of LSMO on parameter. In order to shed light on these 2𝜃 shifts and the ones SC2 AO and CAO, diverge. Figure 3d depicts the experimental in- observed for the (008) SCx AO Bragg reflections, high-resolution plane and out-of-plane strain measured from the LSMO films on XRD reciprocal space maps (RSM) have been carried out, see the different SCx AO. The perpendicular strains (ϵxx , ϵzz ) from Adv. Funct. Mater. 2023, 33, 2304059 2304059 (4 of 10) © 2023 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH 16163028, 2023, 41, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adfm.202304059 by Readcube (Labtiva Inc.), Wiley Online Library on [23/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.afm-journal.de Figure 4. Z-contrast HAADF-STEM cross-sections. a) LSMO/SAO//STO low magnification image. White arrows are used to indicate the formation of outcrops. b) Higher magnification image of one of the outcrops with its corresponding FFT as inset. The LSMO grain boundaries with misoriented grains are marked and their corresponding FFT spots highlighted. c) LSMO/CAO//STO low magnification image. d) Higher magnification image with its corresponding FFT as inset. LSMO grown on SCx AO with x=1-3 and from LSMO on STO, The nucleation of (101) LSMO grains could be favored by the follow a linear dependence according to the Poisson effect. This rough and not completely (001) epitaxial SAO surface. Electron linear trend permits to extract the corresponding Poisson’s ra- energy loss spectroscopy (EELS) analysis carried out in this re- tio, 𝜈=0.38, consistent with the values reported for other strained gion, Figure S4 (Supporting Information), indicates that there LSMO films.[44] Note that the LSMO grown on SAO does not fol- is also slight cation interdiffusion between LSMO and SAO, as low the same strain tendency (Figure 3c). In order to gain insights anticipated from XRD. It is very likely that these two factors, into this discrepancy and elucidate the possible cation interdif- misoriented LSMO grains, and cation diffusion at the interface, fusion identified from the XRD 𝜃-2𝜃 analysis, cross-sectional Z- favor the growth of an strain-free LSMO film with lower degree of contrast high angle annular dark field (HAADF) using scanning crystallinity,[59] explaining the deviation identified in Figure 3d. transmission electron microscopy (STEM) was carried out for the The LSMO/CAO//STO system displays a different landscape. systems grown on the two extreme compositions: SAO and CAO, The low magnification image, Figure 4c, confirms the presence Figure 4. of 60 nm LSMO and 10 nm CAO layers on STO. The nature First, it is investigated the LSMO/SAO//STO heterostructure, of the Z-contrast HAADF imaging evidence the successful full Figure 4a. The low magnification HAADF-STEM image allows chemical substitution of Sr by Ca, as the image contrast in the identifying the LSMO and SAO layers on STO. The 60 nm LSMO CAO layer is remarkably lower with respect to the SAO layer. film shows columnar grains, typical of the PLD growth.[58] The Note that the CAO layer is thinner than SAO (20 nm) and it is SAO film has a thickness of 20 nm with a rough interface with attributed to different wettability of the precursor solution with LSMO. It can be identified areas with dark contrast, which corre- the STO substrate. Higher magnification HAADF-STEM image sponds to amorphous SAO and areas with outcrops (white ar- at the LSMO/CAO interface, Figure 4d, confirms the growth of rows). Regarding the existence of amorphous SAO regions, it only (001) oriented films, and it is supported by the FFT (inset). reveals that the post-annealing in vacuum performed on SAO The LSMO layer has significantly less amount of structural films[31] prior to LSMO deposition does not completely recon- defects compared to LSMO/SAO (Figure 4b) probably because struct the surface crystallinity although the spotty RHEED pat- of the better crystalline quality of the CAO underneath. There- tern shown in Figure 1f. A closer look at one of these outcrops, fore, full substitution of Sr by Ca in SCx AO has an undeniable Figure 4b, shows the atomic ordering of SAO and LSMO, con- effect on improving the crystalline quality of the sacrificial and firming that it corresponds to epitaxial (001) SAO and it acts as enables the epitaxial growth of moderately strained LSMO films. a nucleation site for epitaxial (001) LSMO. From the same mag- Nevertheless, it is worth noting that in some areas of the film the nification image it is also detected the formation of an antiphase LSMO/CAO interface is not sharp and that would be in line with boundary (APB) in the center of the (001) LSMO and few (101) the previous identification of an interfacial reaction. Remarkably, oriented LSMO grains whose grain boundaries are identified the film microstructure and the interface of the LSMO/SC1 AO by dashed white lines. This is further confirmed by fast fourier system is very similar to the ones presented for the transform (FFT), red-marked in the inset image in Figure 4b. LSMO/CAO, Figure S5 (Supporting Information). Therefore, the Adv. Funct. Mater. 2023, 33, 2304059 2304059 (5 of 10) © 2023 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH 16163028, 2023, 41, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adfm.202304059 by Readcube (Labtiva Inc.), Wiley Online Library on [23/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.afm-journal.de Figure 5. a) Temperature-dependent resistivity of different LSMO/SCx AO//STO and LSMO//STO heterostructures and b) temperature derivative of the resistivity d𝜌/dT. incorporation of Ca with x = 1 in SCx AO is already causing a on the temperature-dependent resistivity.[61–63] Thus, further significant improvement in epitaxial quality and no amorphous STEM-EELS analysis would be useful to probe the interface of interface has been devised. Overall, from a structural standpoint, our LSMO heterostructures with atomic precision to elucidate its the incorporation of Ca in SAO allows tuning the cell parameter possible role. while improving the crystalline quality and ambient stability. The epitaxy is transferred to the LSMO generating films with differ- ent strains according to the lattice mismatch and surface quality 2.3. Freestanding Epitaxial LSMO of the SCx AO underneath. Cation intermixing is spotted in some areas at the LSMO/SCx AO interface. It is noteworthy that epitax- The last step to obtain freestanding epitaxial LSMO films requires ial LSMO films can be obtained on SCx AO sacrificial layers after adhering a support layer, in our case PET, to easily manipulate the 16 days of being vacuum annealed and stored in sealed plastic membrane upon the sacrificial etching in water, Figure 6a. Also, bags (Figure S2c, Supporting Information) emphasizing the the PET can be used as a stamp to further transfer the LSMO possibility to combine ex situ and in situ deposition techniques. released membranes to other substrates with no epitaxial rela- tionship, e.g., glass or Si wafer. Importantly, we verified that the concentration of Ca in SCx AO affects the immersion time in wa- 2.2. Electrical Properties ter of the heterostructure being SAO the fastest (2 days) and CAO the slowest (2 weeks), as expected.[37] To evaluate the electrical properties of LSMO, temperature- The LSMO//PET stacks obtained from the different SCx AO dependent resistivity measurements have been carried out on compositions show smooth and homogeneous surface morphol- LSMO/SCx AO//STO heterostructures from 400 to 10K and com- ogy (Figure S1i–l, Supporting Information) and only show the pared to LSMO//STO. As shown in Figure 5a, all LSMO films (002) LSMO Bragg reflection, Figure 6b, which is centered at show a metallic behavior, however, the resistivity of the LSMO the 2𝜃 value of LSMO-bulk. Therefore, all membranes on PET on SAO is significantly higher than the rest. This performance are fully relaxed. Optical inspection of the different LSMO mem- is not surprising considering the LSMO/SAO microstructure branes disclosed the formation of cracks. The density of cracks unraveled from Figure 4 and in agreement with previous re- increases with the content of Ca in SCx AO, Figure 6e–g and it can ports on the impact of microstructure on the electrical trans- be quantified (details in Figure S6, Supporting Information) and port properties.[46,60] Focusing on LSMO on Ca-doped SAO het- compared to the strain release for each LSMO/SCx AO system.[64] erostructures and LSMO//STO, they all show similar resistivity From Figure 6c it can be observed that the least strained LSMO values at room temperature. Nonetheless, the temperature de- on SCx AO, i.e., on SC1 AO, generated least cracks. By increas- pendence of the resistivity shows variations that are more pro- ing the amount of Ca in the sacrificial, the LSMO films show nounced for LSMO//STO. Note that the 𝜌(T) data presented here moderate compressive strain and upon release they also show for LSMO/SCx AO (x = 1-3) follow the same trend and fall within more cracks. It is suggested that the strain relaxation during the the same order of magnitude of the reported values from the lit- lift-off can be responsible for the formation of cracks,[65] as pre- erature. [24,54] Also, changes in the metal to insulator transition, viously reported in BaSnO3 membranes.[37] Nevertheless, there TMI , can be appreciated from the resistance derivative curve in are other factors that could be investigated to further reduce the Figure 5b. These changes could be attributed to the presence crack density of the released membranes, including the sacrificial of different degrees of epitaxial strain, calculated in Figure 3d, layer and membrane thicknesses,[20] the use of additional cap- although no clear trend is observed for the heterostructures. ping layers[36,64] and modify the lift-off procedure.[35,66] Indeed, The interfacial reactivity is also expected to produce changes the exfoliation process is one of the most critical steps in the in the electrical properties. Indeed, it has been widely reported preparation of membranes.[67] the important role of ion stoichiometry in LSMO (oxygen de- Additional membrane analysis was performed on the LSMO ficiency, cation off-stoichiometry, Sr-segregation, interdiffusion) obtained from SC2 AO which showed the best compromise Adv. Funct. Mater. 2023, 33, 2304059 2304059 (6 of 10) © 2023 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH 16163028, 2023, 41, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adfm.202304059 by Readcube (Labtiva Inc.), Wiley Online Library on [23/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.afm-journal.de Figure 6. Freestanding LSMO on PET after the selective etching of the SCx AO sacrificial layers. a) Sketch of the process followed to transfer LSMO membranes to a PET support. b) XRD 𝜃-2𝜃 scans of the LSMO membranes supported on PET. The black bar indicates the position of the (002) LSMO Bragg reflection before the lift-off. c) Crack density and strain release of LSMO membranes as a function of the Ca concentration of the SCx AO sacrificial layer. Magnified images of the corresponding LSMO membranes on PET by etching d) SAO, e) SC1 AO, f) SC2 AO, and g) CAO. White scale bars correspond to 200 μm. between crystallinity, electrical performance (TMI =310 K), moder- lution processed SCx AO sacrificial layers with PLD deposition ate crack density, and can be released faster than from CAO. X-ray of LSMO, could be easy extended to other perovskite systems photoelectron spectroscopy (XPS) survey spectrum revealed the and beyond, such as spinels. In Figure S8 (Supporting Informa- presence of Al traces on the LSMO membrane, which was further tion) it is shown a RHEED study analogous to that presented in verified from Al2p core level spectrum (Figure S7, Supporting Figure 2 for the preparation of PLD-CoFe2 O4 on solution pro- Information). This chemical analysis reafirmed the cation inter- cessed SCx AO confirming the growth of crystalline and oriented mixing at the interface LSMO/SCx AO. CoFe2 O4 films. Then, this LSMO membrane was double-transferred onto a rigid Si substrate, Figure 7a, to perform HAADF-STEM stud- 3. Conclusion ies. Figure 7b proves the high crystallinity of the LSMO mem- brane integrated into silicon, a highly pursued goal to widen the In this study, we put forward a simple hybrid synthesis proce- possibilities of silicon-based electronics.[68–71] The temperature- dure combining solution processing and pulsed laser deposition dependent resistivity of the LSMO membrane prepared from to easily prepare epitaxial complex oxide membranes. First, it SC2 AO was also evaluated by transferring it to a glass sub- is investigated the influence of gradual Ca-substitution in solu- strate. Electrical measurements were performed by patterning tion processed SAO films (SCx AO with x⩽3) to effectively en- gold contacts on top of a flake with no cracks. The resistivity gineer its crystallinity, structure, and robustness preserving a curves obtained before and after the transferring are very similar smooth surface. Moreover, it has been demonstrated that vac- (Figure 7c) , indicating that there is almost no degradation of the uum annealing of SCx AO after air exposure reconstructs the film during the transferring process. The small differences in the surface crystallinity and increases their stability in air offering LSMO resistivity before and after the release could be attributed a less-strict environment to prepare and manipulate the sacri- to the strain relaxation of the film.[24] ficial layers. The composition of SCx AO influences the prop- Finally, it is important to note that the hybrid process here erties of the subsequent epitaxial LSMO film. Ca substitution presented to prepare epitaxial membranes which combines so- in SCx AO favors better crystallinity inducing slight compressive Adv. Funct. Mater. 2023, 33, 2304059 2304059 (7 of 10) © 2023 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH 16163028, 2023, 41, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adfm.202304059 by Readcube (Labtiva Inc.), Wiley Online Library on [23/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.afm-journal.de Figure 7. a) LSMO membranes transferred from a PET support to other substrates such as SiO2 (300 nm)/Si or glass, which were used to make the electrical contact of the transferred LSMO. b) Z-contrast HAADF-STEM cross-section of LSMO//SiO2 interface. c) Temperature-dependent resistivity of the released LSMO membrane. Upper inset shows the temperature derivative of the resistivity d𝜌/dT. Lower inset corresponds to a magnified image of a LSMO flake with the contacts for the resistivity measurements. strain that is totally released upon LSMO exfoliation generating Crystal Structure: X-ray Diffraction (XRD) measurements were per- cracks. Improving the functional complex oxide/sacrificial inter- formed with Cu-K𝛼 using a Bruker-AXS (model A25 D8 Discover). Re- face quality and improving the lift-off process is still required, flection high-energy electron diffraction (RHEED) was performed in the PLD chamber during SCx AO re-crystallization and LSMO deposition with but epitaxial LSMO membranes can be obtained while maintain- incidence of electrons along the [100] STO at a glancing angle of 1–2°. ing the metallic behavior identified prior the lift-off. Therefore, Aberration-corrected Scanning Transmission Electron Microscopy (STEM) this study provides an innovative and versatile platform to synthe- imaging was performed using a Nion HERMES-100, operated at 60 kV, size complex oxide freestanding films and heterostructures that at the University of Chinese Academy of Sciences, Beijing, China. High- can be integrated in a wide variety of substrates in the search for angle annular dark-field (HAADF) images were acquired using an an- novel and enhanced properties envisaging many technological nular detector with collection semi-angle of 75–210 mrad. To minimize the possible beam-induced structural damage on the SAO films, im- benefits. ages were acquired with reduced beam current (10 pA). Electron energy loss spectroscopy (EELS) measurements were performed using a collec- tion semi-angle of 75 mrad, an energy dispersion of 0.9 eV per chan- nel, and a probe current of ≈10 pA. Cross-sectional STEM specimens 4. Experimental Section were prepared using the standard focused ion beam (FIB) lift-out pro- cess in a Thermo Fisher Scientific FIB system. Protective amorphous Synthesis of SCx AO Sacrificial Layer by CSD: SCx AO sacrificial layers carbon and thin Pt layers were applied over the region of interest be- with different ratios of Ca (x = 0, 1, 2, 3) were epitaxially grown on (001) fore milling. To minimize the sidewall damage and sufficiently thin the STO single-crystal substrates by chemical solution deposition. For each specimen for electron transparency, final milling was carried out at a composition, a solution was prepared with stoichiometric amounts of voltage of 2 kV. Sr(NO3 )2 , strontium nitrate (>99%), Ca(NO3 )2 , calcium nitrate (>99%), Surface Morphology: Atomic Force Microscopy (AFM) was used to and Al(NO3 )3 ·9H2 O, hydrated aluminum nitrate (>98%), which were dis- study the surface morphology and roughness of the samples. To- solved in Milli-Q water with citric acid, C6 H8 O7 (>99%), to obtain a 0.1 m pographic images of 5 × 5 μm2 were acquired with a Keysight solution. This solution was spin-coated on STO and annealed in a tubular 5100 AFM instrument and analyzed by Mountains8 software. An op- furnace at 800 °C under 0.6 l·min−1 O2 flow to finally obtain a 20 nm thick tical microscope Leica DM1750 M was used to take magnified op- film. This route was previously developed and optimized for SAO sacrificial tical images of the transferred membranes and analyze the crack layer and it is described with further details elsewhere.[26] density. Ex Situ Synthesis of Epitaxial LSMO by PLD: Following the growth of Electrical Properties: Resistivity curves, 𝜌(T) were measured using the SCx AO, upon sample air-exposure, the film was introduced in the PLD Van der Pauw configuration[72] in a Physical Properties Measurement Sys- chamber to perform an in-vacuum annealing up to 825 °C for 30 min at tem (PPMS) from Quantum Design. 50 nm thick gold contacts were pre- an oxygen partial pressure PO2 of 0.1 mbar to restore the SCx AO surface pared by standard photolithography combined by lift-off. In the case of the crystallinity.[31] Then, 60 nm thick LSMO films were epitaxially grown in transferred LSMO, patterned contacts were placed on top of a single flake situ by PLD at 725 °C and a PO2 of 0.1 mbar using a KrF excimer laser. (see Figure 7c). The same PLD growing conditions were used to prepare LSMO directly on a (001) STO substrate. On the other hand, a similar PLD process was followed to grow CoFe2 O4 on SAO//STO, Figure S8 (Supporting Informa- Supporting Information tion). Supporting Information is available from the Wiley Online Library or from Membrane Transfer: The transfer of the LSMO thin films deposited on the author. SCx AO//STO was done adhering a polyethylene terephthalate (PET) poly- mer support to the film before immersing the entire heterostructure in Milli-Q water. Once the SCx AO sacrificial layer was completely etched, the Acknowledgements PET was mechanically separated achieving the membrane of LSMO at- tached on it. For specific analysis, the membrane held on PET was subse- This work was funded by MICIN/AEI/10.13039/ 501100011033/FEDER quently transferred to silicon or glass. through the projects Severo Ochoa FUNFUTURE CEX2019-00917-S, Adv. Funct. Mater. 2023, 33, 2304059 2304059 (8 of 10) © 2023 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH 16163028, 2023, 41, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adfm.202304059 by Readcube (Labtiva Inc.), Wiley Online Library on [23/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.afm-journal.de CEX2021-001214-S and PID2020-114224RBI00, TED2021-130402B-I00. [15] H. Yoon, T. K. Truttmann, F. Liu, B. E. Matthews, S. Choo, Q. Su, V. The authors also acknowledge the financial support from the 2020 Saraswat, S. Manzo, M. S. Arnold, M. E. Bowden, J. K. Kawasaki, S. J. Leonardo Grant for Researchers and Cultural Creators BBVA Foundation, Koester, S. R. Spurgeon, S. A. Chambers, B. Jalan, Sci. Adv. 2022, 8, the i-link A20346-CSIC project, the National Key RD Program of China 51eadd5328. (2018YFA0305800), and the Beijing Outstanding Young Scientist Program [16] K. T. Kang, J. Park, D. Suh, W. S. Choi, Adv. Mater 2019, 31, 1803732. (BJJWZYJH01201914430039). The project that gave rise to these results [17] G. Rijnders, D. H. A. Blank, Nature 2005, 433, 369. received the support of a fellowship (P.S.) from “la Caixa” Foundation (ID [18] S. Puebla, T. Pucher, V. Rouco, G. Sanchez-Santolino, Y. Xie, V. 100010434). The fellowship code was LCF/BQ/DI19/11730026. M.R. ac- Zamora, F. A. Cuellar, F. J. Mompean, C. Leon, J. O. Island, M. Garcia- knowledges the support from FPI fellowship PRE2021-099672. The au- Hernandez, J. Santamaria, C. Munuera, A. Castellanos-Gomez, Nano thors are thankful to the Thin film scientific service at ICMAB and to Lett. 2022, 22, 7457. Dr. J. Santiso from ICN2 for the fruitful discussions on structure proper- [19] C. Guo, H. Huang, Microstructures 2022, 2, 21. ties. This work had been done in the framework of the doctorate in material [20] S. S. Hong, M. Gu, M. Verma, V. Harbola, B. Y. Wang, D. Lu, A. science of the Autonomous University of Barcelona (P.S. and M.R.). Vailionis, Y. Hikita, R. Pentcheva, J. M. Rondinelli, H. Y. Hwang, Sci- ence 2020, 368, 71. [21] Z. Dong, H. Chen, M. Qi, J. Shen, W. Liu, E. Guo, D. Li, Y. Zhang, Z. Conflict of Interest Wu, Laser Photonics Rev. 2022, 16, 2100454. The authors declare no conflict of interest. [22] S. Cai, Y. Lun, D. Ji, P. Lv, L. Han, C. Guo, Y. Zang, S. Gao, Y. Wei, M. Gu, C. Zhang, Z. Gu, X. Wang, C. Addiego, D. Fang, Y. Nie, J. Hong, P. Wang, X. Pan, Nat. Commun 2022, 13, 5116. Data Availability Statement [23] F. M. Chiabrera, S. Yun, Y. Li, R. T. Dahm, H. Zhang, C. K. R. Kirchert, D. V. Christensen, F. Trier, T. S. Jespersen, N. Pryds, Ann. Phys. 2022, Research data are not shared. 524, 2200084. [24] D. Lu, D. J. Baek, S. S. Hong, L. F. Kourkoutis, Y. Hikita, H. Hwang, Nat. Mater. 2016, 15, 1255. Keywords [25] D. Ji, S. Cai, T. R. Paudel, H. Sun, C. Zhang, L. Han, Y. Wei, Y. Zang, M. Gu, Y. Zhang, W. Gao, H. Huyan, W. Guo, D. Wu, Z. Gu, E. Y. Tsymbal, freestanding oxides, interfaces, La0.7 Sr0.3 MnO3 , sacrificial layers, solution P. Wang, Y. Nie, X. Pan, Nature 2019, 570, 87. processing, Sr3 Al2 O6 [26] P. Salles, I. Caño, R. Guzman, C. Dore, A. Mihi, W. Zhou, M. Coll, Adv. Mater. Interfaces 2021, 8, 2001643. Received: April 12, 2023 [27] J. A. Alonso, I. Rasines, J. L. Soubeyroux, Inorg. Chem. 1990, 29, 4768. Revised: May 18, 2023 [28] Y.-Y. Pai, A. Tylan-Tyler, P. Irvin, J. Levy, Rep. Prog. Phys. 2018, 81, Published online: June 15, 2023 036503. [29] A. S. Brand, J. W. Bullard, Langmuir 2017, 33, 9645. [30] A. K. Prodjosantoso, B. J. Kennedy, B. A. Hunter, Aust. J. Chem. 2000, 53, 195. [1] M. Coll, J. Fontcuberta, M. Althammer, M. Bibes, H. Boschker, A. [31] P. Salles, R. Guzmán, D. Zanders, A. Quintana, I. Fina, F. Sánchez, W. Calleja, G. Cheng, M. Cuoco, R. Dittmann, B. Dkhil, I. El Baggari, Zhou, A. Devi, M. Coll, ACS Appl. Mater. Interfaces 2022, 14, 12845. M. Fanciulli, I. Fina, E. Fortunato, C. Frontera, S. Fujita, V. Garcia, [32] D. Li, C. Adamo, B. Y. Wang, H. Yoon, Z. Chen, S. S. Hong, D. Lu, Y. S. Goennenwein, C.-G. Granqvist, J. Grollier, R. Gross, A. Hagfeldt, Cui, Y. Hikita, H. Y. Hwang, Nano Lett. 2021, 21, 4454. G. Herranz, K. Hono, E. Houwman, M. Huijben, A. Kalaboukhov, D. [33] Y. Li, C. Xiang, F. M. Chiabrera, S. Yun, H. Zhang, D. J. Kelly, R. T. Keeble, G. Koster, L. Kourkoutis, et al., Appl. Surf. Sci. 2019, 482, 1. Dahm, C. K. R. Kirchert, T. E. L. Cozannet, F. Trier, D. V. Christensen, [2] Q. Lin, H. Huang, Y. Jing, H. Fu, P. Chang, D. Li, Y. Yao, Z. Fan, J. T. J. Booth, S. B. Simonsen, S. Kadkhodazadeh, T. S. Jespersen, N. Mater. Chem. C 2014, 2, 1233. Pryds, Adv. Mater. 2022, 34, 2203187. [3] Y. S. Rim, S. Bae, H. Chen, N. De Marco, Y. Yang, Adv. Mater. 2016, [34] D. J. Baek, D. Lu, Y. Hikita, H. Y. Hwang, L. F. Kourkoutis, APL Mater. 28, 4415. 2017, 5, 096108. [4] J. F. Scott, Science 2007, 315, 954. [35] J. Bouaziz, C. Cancellieri, B. Rheingans, L. P. H. Jeurgens, F. La [5] P. Machado, M. Scigaj, J. Gazquez, E. Rueda, A. Sánchez-Díaz, I. Fina, Mattina, Adv. Mater. Interfaces 2022, 10, 2201458. M. Gibert-Roca, T. Puig, X. Obradors, M. Campoy-Quiles, M. Coll, [36] L. Gong, M. Wei, R. Yu, H. Ohta, T. Katayama, ACS Nano 2022, 16, Chem. Mater. 2019, 31, 947. 21013. [6] Z. Liao, F. Li, P. Gao, L. Li, J. Guo, X. Pan, R. Jin, E. W. Plummer, J. [37] P. Singh, A. Swartz, D. Lu, S. S. Hong, K. Lee, A. F. Marshall, K. Nishio, Zhang, Phys. Rev. B 2015, 92, 125123. Y. Hikita, H. Y. Hwang, ACS Appl. Electron. Mater. 2019, 1, 1269. [7] X. Obradors, T. Puig, Supercond. Sci. Technol. 2014, 27, 044003. [38] R. Xu, J. Huang, E. S. Barnard, S. S. Hong, P. Singh, E. K. Wong, T. [8] R. L. Z. Hoye, J. Hidalgo, R. A. Jagt, J.-P. Correa-Baena, T. Fix, J. L. Jansen, V. Harbola, J. Xiao, B. Y. Wang, S. Crossley, D. Lu, S. Liu, H. MacManus-Driscoll, Adv. Energy Mater. 2021, 12, 59. Y. Hwang, Nat. Commun. 2020, 11, 3141. [9] S. A. Chambers, Adv. Mater. 2010, 22, 219. [39] Z. Lu, Y. Yang, L. Wen, J. Feng, B. Lao, X. Zheng, S. Li, K. Zhao, B. Cao, [10] G. Koster, M. Huijben, G. Rijnders, Epitaxial Growth of Complex Metal Z. Ren, D. Song, H. Du, Y. Guo, Z. Zhong, X. Hao, Z. Wang, R.-W. Li, Oxides, 1st Ed., Elsevier, Amsterdam 2015. npj Flex Electron 2022, 6, 9. [11] J. L. MacManus-Driscoll, M. P. Wells, C. Yun, J.-W. Lee, C.-B. Eom, D. [40] J.-H. Park, E. Vescovo, H.-J. Kim, C. Kwon, R. Ramesh, T. Venkatesan, G. Schlom, APL Mater. 2020, 8, 040904. Nature 1998, 392, 794. [12] M. Brahlek, A. S. Gupta, J. Lapano, J. Roth, H.-T. Zhang, L. Zhang, R. [41] Y. Wu, Y. Suzuki, U. Rüdiger, J. Yu, A. D. Kent, T. K. Nath, C. B. Eom, Haislmaier, R. Engel-Herbert, Adv. Funct. Mater. 2018, 28, 1702772. Appl. Phys. Lett. 1999, 75, 2295. [13] D. Pesquera, A. Fernández, E. Khestanova, L. W. Martin, J. Phys.: Con- [42] M. Cesaria, A. P. Caricato, G. Maruccio, M. Martino, J. Phys. Conf. Ser. dens. Matter 2022, 34, 383001. 2011, 292, 012003. [14] S. Han, Y. Meng, Z. Xu, J. S. Kim, Y. Li, I.-P. Roh, H. Ahn, D.-H. Kim, [43] Y. Takamura, R. V. Chopdekar, E. Arenholz, Y. Suzuki, Appl. Phys. Lett. S.-H. Bae, ACS Appl. Electron. Mater. 2023, 5, 690. 2008, 92, 162504. Adv. Funct. Mater. 2023, 33, 2304059 2304059 (9 of 10) © 2023 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH 16163028, 2023, 41, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adfm.202304059 by Readcube (Labtiva Inc.), Wiley Online Library on [23/10/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.afm-journal.de [44] C. Adamo, X. Ke, H. Q. Wang, H. L. Xin, T. Heeg, M. E. Hawley, [59] Y. Chen, J. A. Tilka, Y. Ahn, J. Park, A. Pateras, T. Zhou, D. E. Savage, W. Zander, J. Schubert, P. Schiffer, D. A. Muller, L. Maritato, D. G. I. McNulty, M. V. Holt, D. M. Paskiewicz, D. D. Fong, T. F. Kuech, P. Schlom, Appl. Phys. Lett. 2009, 95, 112504. G. Evans, J. Phys. Chem. C 2019, 123, 7447. [45] B. Wang, L. You, P. Ren, X. Yin, Y. Peng, B. Xia, L. Wang, X. Yu, S. Mui [60] J. Santiso, J. Roqueta, N. Bagués, C. Frontera, Z. Konstantinovic, Q. Poh, P. Yang, G. Yuan, L. Chen, A. Rusydi, J. Wang, Nat. Commun. Lu, B. Yildiz, B. Martínez, A. Pomar, L. Balcells, F. Sandiumenge, ACS 2013, 4, 2778. Appl. Mater. Interfaces 2016, 8, 16823. [46] J. Fontcuberta, M. Bibes, B. Martínez, V. Trtik, C. Ferrater, F. Sánchez, [61] J. D. Burton, E. Y. Tsymbal, Phys. Rev. Lett. 2011, 106, M. Varela, J. Appl. Phys. 1999, 85, 4800. 157203. [47] M. Sirena, N. Haberkorn, M. Granada, L. B. Steren, J. Guimpel, J. [62] H. Zenia, G. A. Gehring, G. Banach, W. M. Temmerman, Phys. Rev. B Appl. Phys. 2009, 105, 033902. 2005, 71, 024416. [48] C. Ge, K.-J. Jin, L. Gu, L.-C. Peng, Y.-S. Hu, H.-Z. Guo, H.-F. Shi, J.-K. [63] Z. Liao, J. Zhang, Appl. Sci. 2019, 9, 144. Li, J.-O. Wang, X.-X. Guo, C. Wang, M. He, H.-B. Lu, G.-Z. Yang, Adv. [64] Q. Wang, H. Fang, D. Wang, J. Wang, N. Zhang, B. He, W. Lü, Crystals Mater. Interfaces 2015, 2, 1500407. 2020, 10, 733. [49] Z. Fang, I. V. Solovyev, K. Terakura, Phys. Rev. Lett. 2000, 84, 3169. [65] H. Taghinejad, A. A. Eftekhar, P. M. Campbell, B. Beatty, M. [50] J. Huang, H. Wang, X. Sun, X. Zhang, H. Wang, ACS Appl. Mater. Taghinejad, Y. Zhou, C. J. Perini, H. Moradinejad, W. E. Henderson, Interfaces 2018, 10, 42698. E. V. Woods, X. Zhang, P. Ajayan, E. J. Reed, E. M. Vogel, A. Adibi, NPJ [51] W. Hou, S. Zhao, T. Wang, M. Yao, W. Su, Z. Hu, Z. Zhou, M. Liu, 2D Mater. Appl. 2018, 2, 1. Appl. Surf. Sci. 2021, 563, 150074. [66] B. Zhang, C. Yun, J. L. MacManus-Driscoll, Nano-Micro Lett. 2021, 13, [52] C.-C. Chiu, Y.-W. Chang, Y.-C. Shao, Y.-C. Liu, J.-M. Lee, S.-W. Huang, 39. W. Yang, J. Guo, F. M. F. de Groot, J.-C. Yang, Y.-D. Chuang, Sci. Rep. [67] A. Castellanos-Gomez, V. Singh, H. S. J. van der Zant, G. A. Steele, 2021, 11, 5250. Ann. Phys. 2015, 527, 1-227. [53] D. J. Baek, D. Lu, Y. Hikita, H. Y. Hwang, L. F. Kourkoutis, ACS Appl. [68] U. K. Bhaskar, N. Banerjee, A. Abdollahi, Z. Wang, D. G. Schlom, G. Mater. Interfaces 2017, 9, 54. Rijnders, G. Catalan, Nat. Nanotechnol. 2016, 11, 263. [54] Z. Lu, J. Liu, J. Feng, X. Zheng, L.-h. Yang, C. Ge, K.-j. Jin, Z. Wang, [69] A. Gómez, J. M. Vila-Fungueiriño, R. Moalla, G. Saint-Girons, J. R.-W. Li, APL Mater. 2020, 8, 051105. Gázquez, M. Varela, R. Bachelet, M. Gich, F. Rivadulla, A. Carretero- [55] D. O’Connor, B. Sexton, R. Smart, Surface Analysis Methods in Mate- Genevrier, Small 2017, 13, 1701614. rials Science, 2 Ed., Springer, Berlin 2003. [70] J. M. Vila-Fungueiriño, J. Gázquez, C. Magén, G. Saint-Girons, R. [56] A. M. Deml, V. Stevanović, A. M. Holder, M. Sanders, R. O’Hayre, C. Bachelet, A. Carretero-Genevrier, Sci. Technol. Adv. Mater. 2018, 19, B. Musgrave, Chem. Mater. 2014, 26, 6595. 702. [57] A. R. Denton, N. W. Ashcroft, Phys. Rev. A 1991, 43, 3161. [71] D. P. Kumah, J. H. Ngai, L. Kornblum, Adv. Funct. Mater. 2020, 30, [58] O. I. Lebedev, G. V. Tendeloo, S. Amelinckx, H. L. Ju, K. M. Krishnan, 1901597. Philos. Mag. A 2000, 80, 673. [72] L. J. Van der Pauw, Philips Tech. Rev. 1958, 13, 1. Adv. Funct. Mater. 2023, 33, 2304059 2304059 (10 of 10) © 2023 The Authors. Advanced Functional Materials published by Wiley-VCH GmbH