Superlattices - ACS Publications - American Chemical Society

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article 3

3

Probing the Origin of Interfacial Carriers in SrTiO-LaCrO Superlattices Ryan B. Comes, Steven R. Spurgeon, Despoina M. Kepaptsoglou, Mark H. Engelhard, Daniel E. Perea, Tiffany C. Kaspar, Quentin M. Ramasse, Peter V Sushko, and Scott A. Chambers Chem. Mater., Just Accepted Manuscript • DOI: 10.1021/acs.chemmater.6b04329 • Publication Date (Web): 13 Jan 2017 Downloaded from http://pubs.acs.org on January 14, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Chemistry of Materials is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemistry of Materials

Probing the Origin of Interfacial Carriers in SrTiO3-LaCrO3 Superlattices

Ryan B. Comes,a,b*‡ Steven R. Spurgeon,a‡ Despoina M. Kepaptsoglou,c Mark H. Engelhard,d Daniel E. Perea,d Tiffany C. Kaspar,a Quentin M. Ramasse,c Peter V. Sushkoa and Scott A. Chambersa a

Physical and Computational Sciences Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, United States

b

Department of Physics, Auburn University, Auburn, Alabama 36849, United States

c

SuperSTEM, SciTech Daresbury Campus, Daresbury, WA44AD, United Kingdom

d

Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, Richland, Washington 99352, United States KEYWORDS Complex oxides, superlattices, x-ray photoelectron spectroscopy, x-ray absorption spectroscopy, scanning transmission electron microscopy

ABSTRACT Emergent phenomena at complex oxide interfaces could provide the basis for a wide variety of nextgeneration devices, including photovoltaics and spintronics. To date, detailed characterization and computational modeling of interfacial defects, cation intermixing, and film stoichiometry have helped to explain many of the novel behaviors observed at a single heterojunction. Unfortunately, many of the techniques employed to characterize a single heterojunction are less effective for a superlattice made up of a repeating series of interfaces that induce collective interfacial phenomena throughout a film. These repeating interfaces present an untapped opportunity to introduce an additional degree of complexity, such as confined electric fields, that cannot be realized in a single heterojunction. In this work, we explore the properties of SrTiO3-LaCrO3 superlattices to understand the role of defects, including variations in cation stoichiometry of individual layers of the superlattice, intermixing across interfaces, and interfacial oxygen vacancies. Using x-ray photoelectron spectroscopy (XPS) and scanning transmission electron microscopy electron energy-loss spectroscopy (STEM-EELS), we quantify the stoichiometry of individual layers of the superlattice and determine the degree of intermixing in these materials. By comparing these results to both density functional theory (DFT) models and STEM-EELS measurements of the Ti and Cr valence in each layer of the superlattice, we correlate different types of defects with the associated materials properties of the superlattice. We show that a combination of ab initio modeling and complementary structural characterization methods can offer unique insight into structure-property relationships in many oxide superlattice systems.

Research into epitaxial complex oxide films and interfaces over the past few decades has broadened our understanding of fundamental physics and offered applications across a wide range of useful devices. Research on perovskite oxide thin films with the general formula of ABO3 has played a major role in the field over this time. Interfaces between polar and non-polar materials are particularly intriguing for the emergent properties they exhibit.

Since the initial observation of a two-dimensional electron gas (2DEG) at the interface between films of the polar band insulator LaAlO3 (LAO) and non-polar STO substrates,1 much work has focused on understanding the phenomena that occur at polar/non-polar interfaces. In an idealized model of LAO-STO, a 2DEG is predicted to form to compensate the polar discontinuity at a nearly atomically abrupt interface, with the surface of the LAO

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

film donating an electron to the STO at the interface.2 However, investigations of interface composition have revealed that cation mixing is more extensive than originally thought, introducing a plethora of defects whose role in promoting conductivity is still poorly understood.3,4 Moreover, various works have shown that the 2DEG behavior occurs only in the case of Al-rich offstoichiometric LAO films growth under certain conditions.3,5,6 It is only through careful characterization of the film stoichiometry using Rutherford back-scattering (RBS) and x-ray photoelectron spectroscopy (XPS) that a plausible mechanism to explain the varying behaviors between samples emerged.3,5,6 These studies underscore the need for thorough interface characterization using a variety of techniques before drawing definitive conclusions about the physical mechanism(s) that lead to emergent behavior. One of the broadest fields of study of perovskite oxides is that of ferroelectricity, where materials exhibit a spontaneous remnant electrical polarization. Ferroelectric materials are used in novel technologies including ferroelectric random access memory (FeRAM).7 Epitaxial strain can be used in thin-film growth to induce a ferroelectric phase transition in a material that is otherwise paraelectric, such as SrTiO3 (STO).8 In particular, recent work has shown that when STO is grown with a slight Sr deficiency on the A-site in an ultrathin (< a few nm) film, dipoles due to TiSr antisite defects will align to yield a ferroelectric polarization.9 On the other hand, others have shown that strain-induced ferroelectricity in STO grown on (La,Sr)(Al,Ta)O3 (LSAT) is highly sensitive to stoichiometry and that ferroelectric response degrades rapidly when the Sr:Ti ratio is not unity.10 These results illustrate the wide range of structural handles to control emergent functionality, but also show that an understanding of the defects present in oxide films is necessary to explain novel phenomena observed in oxide thin films and heterostructures. Oxide superlattices in particular have generated significant interest due to the high density of interfaces present in such systems. In this sense, these composite materials may be thought of as exhibiting the behavior of a single interface, but having many interfaces in parallel produces “bulk-like” properties. Previous observations of a built-in potential gradient across LCO thin films grown on STO substrates make this combination of materials particularly interesting for further exploration.11 Our team has recently examined induced polarization in STO through the growth of STO-LaCrO3 (LCO) superlattices with asymmetric positively and negatively charged interfaces.12 We found that the alternating positive and negative charges on the interfaces produce a built-in electric field within the STO that could be used to separate photoexcited electron-hole pairs for renewable energy applications. As in the case of the induced ferroelectricity in offstoichiometry STO films9 and the defect-induced 2DEG at the LAO-STO interface,1 however, careful characterization of the possible defects present in the materials and understanding of the resulting effects induced by them is criti-

Page 2 of 12

cal. Unfortunately, measurements of superlattice stoichiometry cannot be easily performed with the same precision as in a single heterojunction because of the lack of depth resolution. For example, due to interfacial intermixing, the large number of interfaces in superlattices, and the close overlap in atomic number of Cr and Ti,13 RBS models would have large uncertainties in the stoichiometry of the individual layers of the superlattice. Alternative techniques employing XPS and cross-sectional electron microscopy to measure and model the composition of the superlattice are therefore extremely valuable. In this work, we examine the composition, structure and electronic properties of STO-LCO superlattices grown by molecular beam epitaxy on (La,Sr)(Al,Ta)O3 (LSAT) (001) substrates. We have previously shown that a polarization within the STO layers of the superlattice throughout the samples is induced by alternating positively charged TiO20-LaO+ and negatively charged CrO2-SrO0 interfaces in the structures.12 However, we also observe free carriers and conductivity in the system that was not expected based on our first-principles modeling . Using ex situ XPS sputter depth-profiling, we measure the stoichiometry of individual layers throughout the film via comparisons to an RBS-calibrated LCO reference film and a single-crystal STO substrate. We then compare these results to cross-sectional scanning transmission electron microscopy electron-energy loss spectroscopy (STEMEELS) measurements of the Ti and Cr valence and the LaCr stoichiometry to link how defects in individual superlattice layers mediate the overall composite behavior. In light of density functional theory (DFT) modeling, we explore the origin of interface-induced free carriers in these superlattices. We broadly categorize the interfacial defects into two categories: those which suppress a builtin electrostatic field and those which do not. We show that this combined approach is highly useful for gaining deeper insight into the behavior of superlattice materials in general. This knowledge provides an approach to control the magnitude of the field by tuning the growth protocols and, vice versa, enabling formation of preferential defects via control of field-producing structural blocks. Results Interfacial Phenomena We have conducted extensive STEM-EELS mapping to provide a detailed local picture of interfacial composition and valence. Figures 1(a-b) show a representative highangle annular dark field (HAADF) STEM image and simultaneously acquired EELS chemical map of the entire heterostructure, a superlattice with repeating units of 6 STO unit cells and 3 LCO unit cells (STO6-LCO3). These images confirm the structure and regularity of the LCO and STO layers, and reveal no apparent structural defects or imperfections in the film. Moreover, they directly confirm that the asymmetric termination of each layer is preserved throughout the superlattice. Figures 1(c-d) show line profiles averaged in plane for the integrated Ti L23 edge intensity and Ti L3 t2g – eg crystal field splitting


Page 3 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


( ), respectively, while Figures 1(e-f) show similar line profiles for the integrated Cr L23 edge intensity and the O K – Cr L3 edge energy loss separation ( ), respectively. The former two profiles (Fig. 1(c,e)) indicate that Ti is present in all 3 u.c. of each LCO layer and that there is a finite Cr concentration present in all 6 u.c. of each STO layer, albeit at a lower level than the intermixed Ti. This finding is further supported by Ti L23, O K, and Cr L23 edge EEL spectra extracted from an STO / LCO / STO region in the middle of the film, shown in Figures 1(g-h). Such behavior is consistent with our earlier studies of LCO-STO interfaces11 and suggests that Ti out-diffusion increases the thermodynamic stability of the interface. The differing Ti and Cr intermixing concentrations are primarily attributable to the difference in thickness between the STO (6 u.c.) and LCO (3 u.c.) repeating layers.

havior is consistent with our earlier study23 and suggests that Cr may partially reduce to compensate the Ti4+ cations in the adjacent STO layer, though other defects (such as antisites) may also drive reduction.

To determine whether this intermixing has any effect on valence, we have measured changes in the Ti and Cr edges, which result from transitions from spin-orbit split 2p1/2 and 2p3/2 core levels to unoccupied portions of the valence band with 3d character.14 The Ti L2,3 edge is further split into t2g and eg states by the crystal field and is highly sensitive to local valence state.15 While the white-line ratio method is a well-accepted means of characterizing valence, it is difficult to compare to the literature, owing to the range of background subtraction methods used, and can sometimes show a complex spatial dependence.16 By measuring the separation of the Ti L3 and L2 edges, we are able to explore trends in valence independent of energy drift, without the need for simultaneous background fitting. We map at each pixel of the spectrum image and average in the plane of the film to generate the line profile shown in Figure 1(d). This measurement indicates an average separation of ~ 6.5 eV, similar to values reported in the literature (~6.1–6.7 eV).17,18 We find that decreases in the vicinity of each LCO layer to a value near that of Ti3+, and furthermore, that the overall profile trends slightly toward a lower separation near the film surface; this result supports a slight reduction in Ti valence. One possible explanation for this is that excess La3+, accumulated during LCO layer growth, has diffused into the STO layer and substituted for Sr2+, acting as an electronic donor and resulting in Ti valence changes.19 However, we note that Ti coordination can also significantly affect the Ti edge fine structure,20 and more work is needed to disentangle octahedral rotations from intermixing effects. The use of the white-line ratio method is also problematic for the Cr L23 edge, where the O K edge continuum overlaps with the onset of the Cr L3 peak, precluding an accurate background subtraction. We therefore map the energy loss separation between the Cr L3 peak and O K edge pre-peak ( ), which we and others have previously shown to be an accurate indicator of Cr valence.21,22 Figure 1(f) shows a line profile of that has been integrated in the plane of the film. The separation across each LCO layers falls well within the range for Cr3+ (~46.25–46.75 eV), but we see indications of a slight reduction in Cr valence at each STO interface. 21 This be-



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 12

sults are shown in Figure 2. The Drude peak at low binding energy for all three superlattices indicates that free carriers are present in the system. The presence of free carriers is confirmed by electrical transport measurements, which yield sheet resistances of 1-20 kΩ/square at

Figure 1. Cross-sectional STEM-EELS mapping of local composition and valence. a) STEM-HAADF image of the overall film structure acquired during EELS mapping. b) Composition map of the Ti L23, Cr L23, and La M45 (red, green, and blue, respectively) edges constructed using multiple linear least squares fitting (MLLS). c) Line profile of the integrated Ti L23 edge intensity. d) Line profile of the Ti L3 and L2 edge separation ( ), used to estimate local Ti valence; this profile falls within the 4+ bounds of a Ti state (marked by the dashed lines). e) Line profile of the integrated Cr L23 edge intensity. f) Line profile of the O K and Cr L3 edge energy loss separation ( ), used to estimate local Cr valence; 3+ this profile falls within the bounds of a Cr state (marked by the dashed lines). g-h) O K / Cr L23 and Ti L23 edge spectra, respectively, extracted across the LCO / STO / LCO layers in the middle of the superlattice (dashed lines added to guide eye).

room temperature when measured using the van der Pauw method.24 The Drude peak intensity scales with interface density, with the smallest peak found in the STO8-LCO4 film that has the lowest interface density, suggesting that the carriers are primarily present at the interfaces of the superlattice. We also note that the shoulder between 2.5 and 3.5 eV scales with interface density. This shoulder can be attributed to Cr-O-Ti bonds across the interface, as we have previously observed in La, Cr co-doped SrTiO3 films that exhibit visible light absorption.25 The observed conductivity and free carrier concentrations are not predicted to occur in the idealized superlattice,12 suggesting that they may be the result of defects or non-idealities during the film growth process. These could include variations in cation stoichiometry in the layers of the superlattice or interfacial oxygen vacancies. In the following sections, we explore the effects of both of these mechanisms on the polarization in the superlattice by detailed characterization and ab initio modeling.

To determine whether free carriers are indeed present at the buried interfaces, spectroscopic ellipsometry was performed on the STO6-LCO3 superlattice shown in Figure 1, along with a pristine STO thin film on LSAT and STO8-LCO4 and STO4-LCO2 superlattices. These re-


Page 5 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemistry of Materials sources during the course of the growth which results in a slight La excess, inducing preferential La diffusion towards the film surface. There may also be a slight Cr excess in the bottom three layers of the superlattice, though this is within the uncertainty of the measurement. In both cases, however, the results suggest that the La:Cr ratio was within ~5% of the ideal 1:1, which is the general uncertainty with which the stoichiometry can be quantified in our LCO films.28 These variations in stoichiometry, while small within the level of control we generally observe in MBE-grown films, can have significant effects on the electronic behavior.

Figure 2. Extinction coefficient, k, derived from spectroscopic ellipsometry for the STO4-LCO2, STO6-LCO3, and STO8-LCO4 superlattices, and an STO reference.

Film Stoichiometry To probe the stoichiometry of the superlattice, depthprofile x-ray photoelectron spectroscopy measurements were performed on the STO6-LCO3 superlattice. The results were compared to measurements made using the same sputter energy on a LaCrO3 reference film grown on a SrTiO3 substrate with the La:Cr ratio verified to be 1.00(5) using Rutherford back scattering spectrometry measurements. The Sr 3d, Ti 2p, Cr 2p and La 3d peak areas were measured and compared for both samples and are summarized in Figure 3. The STO ratios were averaged using a simple boxcar average over a period of one layer of the superlattice, while the LCO ratios were measured at the point of greatest peak area for the La 3d peak. This approach for LCO takes into account the different escape depths for photoemitted electrons from the high binding energy La 3d (~834 eV) and Cr 2p (~570 eV) from the 3 u.c. thin layers.26 In the case of the thicker 6 u.c. STO layers, the escape depths are greater for Ti 2p (~459 eV) and Sr 3d (~133 eV), meaning that a boxcar average over one period of the superlattice provides a reasonable measure of the stoichiometry through the entire depth of the layer. The gradual decay in peak intensity in Figure 3(b) is most likely due to roughening of the film surface during Ar+ sputtering, which reduces electron escape depth and overall peak intensity.26 A similar trend of overall peak intensity in the Ti 2p and Sr 3d ratio is observed for the STO peaks in Figure 3(a) (not shown). In Figure 3(a), we see that the STO ratio remains with the measurement range for stoichiometric STO throughout the growth process. We cannot rule out slight variations in the cation stoichiometry, but these variations are ~1-2% in Sr:Ti ratio. This is consistent with the calibration process performed based on RHEED oscillations.27 In the LCO layer (Figure 3(b)), we see good agreement with the reference ratio in the bulk of the superlattice, but with a slight trend toward La-rich conditions near the superlattice surface. This could be due to drift in the effusion cell

Figure 3. Sputter depth-profile XPS analysis of (a) SrTiO3 and (b) LaCrO3 stoichiometry throughout STO6-LCO3 film.

To study the effect of cation off-stoichiometry, DFT models of various superlattice structures were constructed assuming different substitutional effects of the excess ions. We assume substitutional rather than anti-site defects, based on consideration of the growth process that we employ and previous models of the most energetically stable defects in LCO.13 The films are grown via sequential shuttering, such that alternating AO and BO2 planes are deposited. Such an approach is likely to result in defects that will remain primarily on the film surface, as has been



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

observed through analysis of cation rearrangement in Ruddlesden-Popper Sr2TiO4 films grown by molecular beam epitaxy with in situ x-ray scattering measurements.29 In this work and others, the authors found that excess cations preferentially diffuse to the surface of the film where defect energies are lower, leaving the stoichiometry in the bulk of the film closer to the ideal.30 This again suggests that, provided the overall A-site to B-site ratio is close to the ideal one-to-one, excess A or B ions will be present on the surface. This type of behavior has been observed in MBE-grown LaAlO36 and SrTiO331 thin films previously through RHEED analysis of surface reconstructions and can be used to calibrate the cation stoichiometry, as we have done in the present work. Given the layer-by-layer growth process and the relatively long deposition times for each unit cell (86 seconds/unit cell), this diffusion to the surface is likely to be facile. Collectively, these results suggest that cation vacancy concentrations are likely to be low when compared to other defects that may be present in the system. This also means that any variations in the stoichiometry for either LCO or STO are likely to produce substitutional dopants in the neighboring layers while leaving the overall A-to-B site ratio close to the ideal value. We refer to this as “offstoichiometry” within a single superlattice unit in the models that follow. Substitutional doping due to off-stoichiometry in the LCO or STO layers of the superlattice can occur at either the nominally positively-charged LaO+-TiO2 interface or at the negatively-charged SrO-CrO2- interface. The structural model for these defects and corresponding geometrical parameters of the superlattice are summarized in Figure 4(a-b), with (a) showing the effects of offstoichiometry within the STO layer and (b) showing the effects of off-stoichiometry within the LCO layer. Substitutional dopants at both interfaces are considered for each case, with the +/- notation used to describe the interface at which the substitution is made (i.e. CrTi+ to indicate Cr substituted for Ti at the LaO+/TiO2 interface). The graphs show the difference in the B-O bond lengths along the c-axis (growth direction), which is directly correlated with a built-in electric field in the material. A

Page 6 of 12

schematic showing how the difference in B-O bond lengths in figure is calculated is provided in the Supporting Information (Figure S4). Any LaSr substitution would electron-dope the STO layer, while a SrLa substitution would hole-dope the LCO layer. Each could conceivably screen the electric field and eliminate the polarization. CrTi substitutions replace a d0 Ti4+ cation with a d2 Cr4+ ion with an available hole at the top of the valence band, while TiCr substitutions replace a d3 Cr3+ with a d1 Ti3+, which adds an electron to the LCO layer near the conduction band of the neighboring STO layer. In both cases, these substitutions may add carriers to the system at room temperature, as the holes and electrons are weakly bound. Somewhat surprisingly, we find from the models that the built-in polarization is preserved in the majority of cases. Only in cases of hole-doping at the positivelycharged interface (SrLa+ or CrTi+) do we see screening of the electric field. Indeed, in both cases a hole at the positively-charged interface will be electrostatically repelled towards the negatively charged interface. The substitutional CrTi ion at the positively charged interface adopts the charge states of 3+ while a Cr in LCO at the negatively charged interface adopts a 4+ charge state. A SrLa substitution at the positively charged interface yields an analogous charge redistribution. Both substitutions cancel out the electric field due to the hole on the Cr4+ ion at the negatively charged interface, which screens the polarization. This eliminates the polarization in the material, as can be observed in the case of the SrLa+ and CrTi+ bond distortion curves in Figure 4. Furthermore, these configurations are found to have a lower total free energy than the opposite case of hole-doping at the negatively charged interface, suggesting that a screened polarization is energetically favorable. Interestingly, the converse does not hold, in that an electron donated from LaSr or TiCr substitutions does not screen the field, regardless of the interface where the defect occurs. Given that we experimentally observe cation distortions and a built-in field in good agreement with the theoretical predictions,12 this suggests that any free carriers are likely to be electrons rather than holes, which would screen the field.


Page 7 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. Substitutional dopant models of impact of (a) variations in SrTiO3 stochiometry; (b) variations in LaCrO3 stoichiometry; and (c) oxygen vacancies induced by Cr-rich LaCrO3 layers. Key: La- Purple circles; Sr- Yellow circles; Cr- Red octahedral; Ti- Green octahedral; OLight blue circles. The difference in B-site bond length is equal to the length of the B-O bond between the apical oxygen above the B site ion in the figure minus the length of the B-O bond between the apical oxygen below the B site ion.

The observed polarization in slightly Cr-rich LCO layers, which may be present deeper within the film based on our XPS depth profile, raises questions as to the effects that we observe, since hole-doping is expected to screen the electric field in the Cr-rich regime where holes would be expected. Given the instability of octahedrallycoordinated Cr(IV)32 and the presence of a hole at the top of the valence band that can readily be neutralized through the formation of an oxygen vacancy, a defect pair formed by a substitutional CrTi and an oxygen vacancy needs to be considered. We have modeled additional configurations for each CrTi substitutional case with a neighboring oxygen vacancy at two locations: in the equatorial BO2 plane, and in the neighboring axial AO plane found to be most energetically favorable. These configurations are shown in Figure 4(c), along with the original CrTi substitutional cases. We note that a single oxygen vacancy adds two electrons to the system but that only one Cr ion is present to be reduced, so an additional electron is added to the system. In each case, through analysis of the charge and spin distributions in the models, we find that the substitutional Cr ion is reduced to Cr3+ as expected. Furthermore, we see that the electric field and bond length distortions are preserved. The additional electron contributes charge over both neighboring Ti and Cr, reducing each towards Ti3+ and Cr2+. Given that we have observed no spectroscopic evidence in either our previous XPS analysis or in the EELS data presented in this work for Cr4+, we suggest that in the case of excess Cr in the

LCO layers that interfacial oxygen vacancies form and the polarization is preserved. Oxygen Vacancies Direct measurement of interfacial oxygen vacancies is notoriously difficult, but their presence is often inferred from the observed chemical valence or electronic behavior of these materials.33 Our observations of free carriers and evidence of interfacial Ti3+ suggest that oxygen vacancies are likely to be present, either due to cation offstoichiometry, such as excess Cr in the LCO layers, or due to incomplete STO oxidation at the interface. To understand the effects of oxygen vacancies in an otherwise ideal lattice, DFT models of various oxygen-deficient configurations were constructed with oxygen vacancies in each AO and BO2 plane in the superlattice. Oxygen vacancies were found to be most energetically stable near the interfaces compared to the bulk, though a side-by-side energy comparison between a fully oxidized film and the oxygen deficient structures is not possible using this method to determine if oxygen vacancy formation would be preferred to the fully oxidized case. The results of these simulations are shown in Figure 5, with schematics of the four oxygen deficient structures shown in (a) and Bader ionic charge34 and the bond length differences shown in (b) and (c) respectively.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 12

where the polarization is screened. In this case, the oxygen vacancy donates two electrons to the system, and the population analysis indicates that they are distributed with one electron lying on a Ti3+ ion at each interface. This electronic configuration neutralizes the electric field and completely screens the polarization, with no change in the Cr valence. Conversely, in the case of an oxygen vacancy in the AO planes at each interface, one electron is donated to the nearest-neighbor Ti ion, while the other lies on the neighboring Cr ion, as can be seen by the purple square and yellow diamond with charge of ~1.2e in the LCO region of Figure 7(c). Similar Bader charge behavior was predicted in the DFT model for the CrTi substitution at the negatively charged interface with an oxygen vacancy in the SrO plane (CrTi-VOax in Figure 4(c)). This Bader charge corresponds to a 2+ formal charge on the Cr ion, in agreement with the trend observed in the Cr STEM-EELS data in Figure 1 and observed previously.23 Discussion Our experimental and computational results have allowed us to produce quantitative estimates of the stoichiometry of each layer of a STO-LCO superlattice. Based on the detailed STEM-EELS analysis, we can correlate the film stoichiometry with the observed interfacial valence phenomena and the built-in polarization to better understand the source of free carriers in the system. Our depthprofile XPS analysis showed that the STO stoichiometry is essentially ideal throughout the superlattice. While we have not considered the effects of cation vacancies or anti-site defects in the STO, these would not be expected to suppress the observed polarization9. However, we do not observe any measurable deviation in the STO stoichiometry, suggesting that these defects are unlikely to be the main driver of our observed polarization.

Figure 5. a) Models of interfacial oxygen vacancy locations; b) Bader charges of atoms occupying the lattice bsites; c) Bond distortions in B-site cations with oxygen vacancies at locations shown in (a). Key: La- Purple circles; Sr- Yellow circles; Cr- Red octahedral; Ti- Green octahedral; O- Light blue circles. The difference in Bsite bond length is equal to the length of the B-O bond between the apical oxygen above the B site ion in the figure minus the length of the B-O bond between the apical oxygen below the B site ion.

We find that the polarization is preserved in three of the four cases, with an oxygen vacancy in the TiO2 plane adjacent to the SrO-CrO2 negatively-charged interface (referred to as TiO2- in Figure 5(c)) as the lone exception

Near the film surface, we detect a possible excess of La ions relative to Cr in the LCO layers of a few atomic percent in the depth profile XPS analysis. In this region, we also observe a more pronounced trend of Ti reduction at the Ti L23 edge, indicating that excess La has donated electrons in the STO layer. This trend alone could be sufficient to explain the free carriers that are observed. Our ab initio simulations indicate that free electrons donated by excess La may reduce the polarization near the interfaces, but do not eliminate the effect, which is in agreement with our previous experimental observations.12 Beyond effects due to cation stoichiometry, the observed valence changes near the interface suggest that oxygen vacancies may be present even in regions where the cation stoichiometry is nearly perfect. Our models indicate that these vacancies would reduce both the Ti and Cr ions near the interface, in agreement with our STEM-EELS observations. It is also likely that any Cr excess in an LCO layer would promote the formation of oxygen vacancies. In this case, the vacancy would donate one electron to reduce the excess Cr ion to a 3+ formal valence, while the second electron from the vacancy could either compensate a second excess Cr ion or contribute a free electron to the system. This type of behavior is quali-


Page 9 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


tatively similar to what has been observed in the case of Al-rich LAO films grown on STO, where carrier concentrations increased in the cases of Al-rich films.5,6 Conclusion Using complementary x-ray spectroscopy and electron microscopy techniques, we have shown a new approach to characterize the stoichiometry of a LaCrO3-SrTiO3 superlattice on a layer-by-layer basis. This understanding of the properties of each layer of the superlattice allows us to study the effects of various defects and non-idealities in the materials using a combination of density functional theory modeling and electron energy-loss spectroscopy. We find that excess La ions in the LCO layer of the superlattice contribute free carriers to the neighboring STO layers. Meanwhile, excess Cr ions promote the formation of interfacial oxygen vacancies. These vacancies produce a combination of Cr2+ and Ti3+ ions while also preventing the formation of a Cr4+ hole that would screen the built-in polarization in the superlattice. Methods Film Synthesis Epitaxial superlattices were synthesized via molecular beam epitaxy using atomic fluxes from metal effusion cells containing each cation element using a technique described previously.12 The effusion cells were initially calibrated using a quartz crystal rate monitor to ~5% precision to achieve a growth rate of 43 seconds per atomic layer (AO or BO2 plane). Prior to superlattice growth, a homoepitaxial SrTiO3 film was grown via sequential shuttering to calibrate the Sr:Ti cation flux ratios to within 12%.27,35 A LaCrO3 film was then grown and the surface reconstruction in RHEED pattern was monitored to fine tune the La:Cr stoichiometry to better than 5% precision.6 The (La,Sr)(Al,Ta)O3 (LSAT) (001) substrate was then transferred into the chamber and cleaned in atomic oxygen plasma at 750 °C for 10 minutes. Growth was performed in 3×10-6 Torr O2 at 700 °C to prevent overoxidation of LCO.28 This atmosphere is sufficient to fully oxidize STO films in our experience and in other groups’ work.29 A 2-3 nm STO buffer layer was deposited by shuttering to achieve a B-site terminated film surface followed by the growth of superlattice by shuttering. The sample was cooled to room temperature at 15 °C/min in the same O2 environment. X-ray Photoelectron Spectroscopy Angle-resolved measurements were performed on the sample after growth by transferring under ultra-high vacuum to an appended XPS chamber with a Scienta R3000 analyzer and monochromated Al Kα source. An electron flood gun was used to prevent charging due to the insulating nature of the substrate. The Cr and Ti 2p core levels were measured at normal emission and over an angular range from 0° to 80° off-normal to vary the degree of surface sensitivity. Measurements were also made on the Sr 3d and La 4d peaks and showed analogous behavior, though incomplete charge compensation by the flood gun affected the quality of some of the data. Depth profile measurements of the Cr 2p, Ti 2p, La 3d, and Sr 3d were

made using a PHI Quantera Scanning X-ray Microprobe XPS system with a focused monochromatic Al Kα (1486.7 eV) source. Ar+ ions with energies of 500 eV were used to sputter through the superlattice slowly, with data acquired at a density of one to two scans per perovskite unit cell in the superlattice (~2-3 Å/scan). Depth profile measurements through an STO single crystal and LCO film with stoichiometry calibrated using Rutherford back scattering were used to determine the peak sensitivity factors. Background subtraction was performed systematically for both angle-resolved and depth-profile measurements using a traditional Shirley background37 and the peak area ratios were then calculated over the relevant energy range. Scanning Transmission Electron Microscopy Samples were prepared for TEM using an FEI Helios NanoLab Dual-Beam Focused Ion Beam (FIB) microscope and a standard lift out procedure, with initial cuts made at 30 kV and final polishing done at 5 kV / 5.5º and 2 kV / 6º incidence angle.37 An additional reference sample for bulk valence states was prepared from a thick (~40 nm) LaCrO3 film deposited on SrTiO3. EELS maps and STEMHAADF images were collected on a CS-corrected Nion UltraSTEM 100 operated at 100 keV, with a convergence angle of 30 mrad. The HAADF inner and outer collection semi-angles were 82 and 190 mrad, respectively. Additional mapping was performed on a CS-corrected Nion UltraSTEM 100MC operated at 100 keV, with a convergence angle of 30 mrad. The EELS collection semi-angle was 44 mrad. Spectra were collected with 0.5 eV ch-1, 0.3 eV ch-1, 0.2 eV ch-1 dispersions, yielding effective energy resolutions of 1.5, 0.9, and 0.6 eV, respectively. No plural scattering correction was performed since zero loss measurements confirm that the samples are sufficiently thin (/ ≈ 0.5 MFP). Maps of and were calculated using the SI Tools plugin38 for Digital Micrograph. Density Functional Theory We employed the density functional theory (DFT) with the exchange-correlation functionals by Perdew–Burke– Ernzerhof39 and modified for solids (PBEsol)40 to examine ideal [SrTiO3]6/[LaCrO3]3.12 A plane wave basis set with the energy cutoff of 500 eV and the projected augmented waves method41 implemented in the Vienna Ab initio Simulation Program (VASP) were employed.42,43 The heterostructures were represented using the periodic model and the √2a0×√2a0 lateral cell, where a0 corresponds to the lattice constant of a cubic perovskite. The 4×4×1 Monkhorst-Pack k-grid centered at the Γ point was used in all calculations. To explore the effects of oxygen deficiency, cation intermixing, and off-stoichiometry on the structure and properties of the [SrTiO3]6-[LaCrO3]3 system various structural models were constructed with oxygen vacancies, substitutional dopants and intermixed interfaces We first examined the stability of a single oxygen vacancy and the vacancy-induced charge redistribution as a function of its position in the hetero-structure. The charge re-distribution was analyzed using the method



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

proposed by Bader.34 To examine the effect of the vacancies on the internal electrostatic field, we employed the same methodology as in the idealized model. Similar calculations and analysis were performed for the Ti ⇔ Cr and Sr ⇔ La intermixed cation configurations as well as for the off-stoichiometric configurations for both n- and p-type interfaces. In the latter case we considered defects formed by excess Ti occupying the Cr site (TiCr) in LCO and vice versa (CrTi) and, similarly for the A-site of the lattice, SrLa and LaSr. CrTi substitutions with a correlated oxygen vacancy were also modeled.

ASSOCIATED CONTENT Supporting Information. Experimental and computational studies of cation intermixing in the behavior of these samples; atom probe tomography of the superlattices to study cation intermixing and stoichiometry. This material is available free of charge via the Internet at http://pubs.acs.org.

(2)

(3)

(4)

(5)

(6)

AUTHOR INFORMATION Corresponding Author * [email protected]; [email protected]

(7)

Present Addresses †If an author’s address is diﬀerent than the one given in the affiliation line, this information may be included here.

(8)

Author Contributions ‡These authors contributed equally. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

(9)

Funding Sources Any funds used to support the research of the manuscript should be placed here (per journal style). (10)

ACKNOWLEDGMENT RBC was supported by the Linus Pauling Distinguished Postdoctoral Fellowship at Pacific Northwest National Laboratory (PNNL LDRD PN13100/2581). SRS, MEB and SAC were supported by the U.S. Department of Energy (DOE), Basic Energy Sciences (BES), Division of Materials Sciences and Engineering under Award #10122. PVS was supported by the LDRD Program at PNNL. A portion of this research was performed using EMSL, a national scientific user facility sponsored by the Department of Energy's Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory. Electron microscopy was carried out in part at the SuperSTEM Laboratory, the U.K. National Facility for Aberration-Corrected STEM, which is supported by the Engineering and Physical Sciences Research Council (EPSRC). The research leading to these results has received funding from the European Union Seventh Framework Programme under Grant Agreement 312483 - ESTEEM2 (Integrated Infrastructure Initiative–I3).

REFERENCES (1)

Ohtomo, A.; Hwang, H. Y. A High-Mobility Electron Gas at the LaAlO3/SrTiO3 Heterointerface. Nature 2004, 427, 423–426.

(11)

(12)

(13)

(14)

(15)

Page 10 of 12

Pentcheva, R.; Pickett, W. E. Ionic Relaxation Contribution to the Electronic Reconstruction at the n-Type LaAlO3/SrTiO3 Interface. Phys. Rev. B 2008, 78, 205106. Chambers, S. A.; Engelhard, M. H.; Shutthanandan, V.; Zhu, Z.; Droubay, T. C.; Qiao, L.; Sushko, P. V.; Feng, T.; Lee, H. D.; Gustafsson, T.; Garfunkel, E.; Shah, A. B.; Zuo, J.-M.; Ramasse, Q. M. Instability, Intermixing and Electronic Structure at the Epitaxial Heterojunction. Surf. Sci. Rep. 2010, 65, 317–352. Qiao, L.; Droubay, T. C.; Varga, T.; Bowden, M. E.; Shutthanandan, V.; Zhu, Z.; Kaspar, T. C.; Chambers, S. A. Epitaxial Growth, Structure, and Intermixing at the LaAlO3/SrTiO3 Interface as the Film Stoichiometry Is Varied. Phys. Rev. B 2011, 83, 85408. Breckenfeld, E.; Bronn, N.; Karthik, J.; Damodaran, A. R.; Lee, S.; Mason, N.; Martin, L. W. Effect of Growth Induced (Non)Stoichiometry on Interfacial Conductance in LaAlO3/SrTiO3. Phys. Rev. Lett. 2013, 110, 196804. Warusawithana, M. P.; Richter, C.; Mundy, J. A.; Roy, P.; Ludwig, J.; Paetel, S.; Heeg, T.; Pawlicki, A. A.; Kourkoutis, L. F.; Zheng, M.; Lee, M.; Mulcahy, B.; Zander, W.; Zhu, Y.; Schubert, J.; Eckstein, J. N.; Muller, D. A.; Hellberg, C. S.; Mannhart, J.; Schlom, D. G. LaAlO3 Stoichiometry Is Key to Electron Liquid Formation at LaAlO3/SrTiO3 Interfaces. Nat. Commun. 2013, 4, 2351. Scott, J. F.; Araujo, C. A. P. de. Ferroelectric Memories. Science 1989, 246, 1400–1405. Haeni, J. H.; Irvin, P.; Chang, W.; Uecker, R.; Reiche, P.; Li, Y. L.; Choudhury, S.; Tian, W.; Hawley, M. E.; Craigo, B.; Tagantsev, A. K.; Pan, X. Q.; Streiffer, S. K.; Chen, L. Q.; Kirchoefer, S. W.; Levy, J.; Schlom, D. G. RoomTemperature Ferroelectricity in Strained SrTiO3. Nature 2004, 430, 758–761. Lee, D.; Lu, H.; Gu, Y.; Choi, S.-Y.; Li, S.-D.; Ryu, S.; Paudel, T. R.; Song, K.; Mikheev, E.; Lee, S.; Stemmer, S.; Tenne, D. A.; Oh, S. H.; Tsymbal, E. Y.; Wu, X.; Chen, L.Q.; Gruverman, A.; Eom, C. B. Emergence of RoomTemperature Ferroelectricity at Reduced Dimensions. Science 2015, 349, 1314–1317. Haislmaier, R. C.; Engel-Herbert, R.; Gopalan, V. Stoichiometry as Key to Ferroelectricity in Compressively Strained SrTiO3 Films. Appl. Phys. Lett. 2016, 109, 32901. Chambers, S. A.; Qiao, L.; Droubay, T. C.; Kaspar, T. C.; Arey, B. W.; Sushko, P. V. Band Alignment, Built-In Potential, and the Absence of Conductivity at the LaCrO3/SrTiO3(001) Heterojunction. Phys. Rev. Lett. 2011, 107, 206802. Comes, R. B.; Spurgeon, S. R.; Heald, S. M.; Kepaptsoglou, D. M.; Jones, L.; Ong, P. V.; Bowden, M. E.; Ramasse, Q. M.; Sushko, P. V.; Chambers, S. A. Interface-Induced Polarization in SrTiO3-LaCrO3 Superlattices. Adv. Mater. Interfaces, 2016, 3, 1500779. Qiao, L.; Zhang, K. H. L.; Bowden, M. E.; Varga, T.; Shutthanandan, V.; Colby, R.; Du, Y.; Kabius, B.; Sushko, P. V.; Biegalski, M. D.; Chambers, S. A. The Impacts of Cation Stoichiometry and Substrate Surface Quality on Nucleation, Structure, Defect Formation, and Intermixing in Complex Oxide Heteroepitaxy–LaCrO3 on SrTiO3(001). Adv. Funct. Mater. 2013, 23, 2953–2963. Varela, M.; Oxley, M. P.; Luo, W.; Tao, J.; Watanabe, M.; Lupini, A. R.; Pantelides, S. T.; Pennycook, S. J. AtomicResolution Imaging of Oxidation States in Manganites. Phys. Rev. B 2009, 79, 85117. Ohtomo, A.; Muller, D. A.; Grazul, J. L.; Hwang, H. Y. Artificial Charge-Modulationin Atomic-Scale Perovskite Titanate Superlattices. Nature 2002, 419, 378–380.


Page 11 of 12 (16)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(17)

(18)

(19)

(20)

(21)

(22)

(23)

(24)

(25)

(26) (27)

(28)

(29)

Chemistry of Materials Gulec, A.; Phillips, P. J.; Klie, R. F. Position-Sensitive Change in the Transition Metal L-Edge Fine Structures. Appl. Phys. Lett. 2015, 107, 143111. Shah, A. B.; Ramasse, Q. M.; May, S. J.; Kavich, J.; Wen, J. G.; Zhai, X.; Eckstein, J. N.; Freeland, J.; Bhattacharya, A.; Zuo, J. M. Presence and Spatial Distribution of Interfacial Electronic States in LaMnO3-SrMnO3 Superlattices. Phys. Rev. B 2010, 82, 115112. Leapman, R. D.; Grunes, L. A.; Fejes, P. L. Study of the L23 Edges in the 3d Transition Metals and Their Oxides by Electron-Energy-Loss Spectroscopy with Comparisons to Theory. Phys. Rev. B 1982, 26, 614–635. Shah, A. B.; Ramasse, Q. M.; Zhai, X.; Wen, J. G.; May, S. J.; Petrov, I.; Bhattacharya, A.; Abbamonte, P.; Eckstein, J. N.; Zuo, J.-M. Probing Interfacial Electronic Structures in Atomic Layer LaMnO3 and SrTiO3 Superlattices. Adv. Mater. 2010, 22, 1156–1160. Zhang, J.; Visinoiu, A.; Heyroth, F.; Syrowatka, F.; Alexe, M.; Hesse, D.; Leipner, H. S. High-Resolution Electron Energy-Loss Spectroscopy of BaTiO3∕SrTiO3 Multilayers. Phys. Rev. B 2005, 71, 64108. Kaspar, T. C.; Schreiber, D. K.; Spurgeon, S. R.; McBriarty, M. E.; Carroll, G. M.; Gamelin, D. R.; Chambers, S. A. Built-In Potential in Fe2O3-Cr2O3 Superlattices for Improved Photoexcited Carrier Separation. Adv. Mater. 2015, 28, 1616-1622. Arévalo-López, Á. M.; Alario-Franco, M. Á. Reliable Method for Determining the Oxidation State in Chromium Oxides. Inorg. Chem. 2009, 48, 11843–11846. Colby, R.; Qiao, L.; Zhang, K. H. L.; Shutthanandan, V.; Ciston, J.; Kabius, B.; Chambers, S. A. Cation Intermixing and Electronic Deviations at the Insulating LaCrO3/SrTiO3(001) Interface. Phys. Rev. B 2013, 88, 155325. Bierwagen, O.; Pomraenke, R.; Eilers, S.; Masselink, W. T. Mobility and Carrier Density in Materials with Anisotropic Conductivity Revealed by van Der Pauw Measurements. Phys. Rev. B 2004, 70, 165307. Comes, R. B.; Smolin, S. Y.; Kaspar, T. C.; Gao, R.; Apgar, B. A.; Martin, L. W.; Bowden, M. E.; Baxter, J. B.; Chambers, S. A. Visible Light Carrier Generation in Co-Doped Epitaxial Titanate Films. Appl. Phys. Lett. 2015, 106, 92901. Hofmann, S. Compositional Depth Profiling by Sputtering. Prog. Surf. Sci. 1991, 36, 35–87. Haeni, J. H.; Theis, C. D.; Schlom, D. G. RHEED Intensity Oscillations for the Stoichiometric Growth of SrTiO3 Thin Films by Reactive Molecular Beam Epitaxy. J. Electroceramics 2000, 4, 385–391. Qiao, L.; Droubay, T. C.; Bowden, M. E.; Shutthanandan, V.; Kaspar, T. C.; Chambers, S. A. LaCrO3 Heteroepitaxy on SrTiO3(001) by Molecular Beam Epitaxy. Appl. Phys. Lett. 2011, 99, 61904-061904–3. Lee, J. H.; Luo, G.; Tung, I. C.; Chang, S. H.; Luo, Z.; Malshe, M.; Gadre, M.; Bhattacharya, A.; Nakhmanson, S. M.; Eastman, J. A.; Hong, H.; Jellinek, J.; Morgan, D.;

(30)

(31)

(32)

(33)

(34) (35)

(36)

(37)

(38) (39)

(40)

(41) (42)

(43)

Fong, D. D.; Freeland, J. W. Dynamic Layer Rearrangement during Growth of Layered Oxide Films by Molecular Beam Epitaxy. Nat. Mater. 2014, 13, 879–883. Feng, Z.; Hong, W. T.; Fong, D. D.; Lee, Y.-L.; Yacoby, Y.; Morgan, D.; Shao-Horn, Y. Catalytic Activity and Stability of Oxides: The Role of Near-Surface Atomic Structures and Compositions. Acc. Chem. Res. 2016, 49, 966– 973. Nie, Y. F.; Zhu, Y.; Lee, C.-H.; Kourkoutis, L. F.; Mundy, J. A.; Junquera, J.; Ghosez, P.; Baek, D. J.; Sung, S.; Xi, X. X.; Shen, K. M.; Muller, D. A.; Schlom, D. G. Atomically Precise Interfaces from Non-Stoichiometric Deposition. Nat. Commun. 2014, 5, 4530. Zhang, K. H. L.; Sushko, P. V.; Colby, R.; Du, Y.; Bowden, M. E.; Chambers, S. A. Reversible Nano-Structuring of SrCrO3-δ through Oxidation and Reduction at Low Temperature. Nat. Commun. 2014, 5, 4669. Park, J.; Cho, B.-G.; Kim, K. D.; Koo, J.; Jang, H.; Ko, K.T.; Park, J.-H.; Lee, K.-B.; Kim, J.-Y.; Lee, D. R.; Burns, C. A.; Seo, S. S. A.; Lee, H. N. Oxygen-Vacancy-Induced Orbital Reconstruction of Ti Ions at the Interface of LaAlO3/SrTiO3 Heterostructures: A Resonant Soft-X-Ray Scattering Study. Phys. Rev. Lett. 2013, 110, 17401. Bader, R. F. W. Atoms in Molecules: A Quantum Theory; Clarendon Press: Oxford, 1994. Comes, R. B.; Sushko, P. V.; Heald, S. M.; Colby, R. J.; Bowden, M. E.; Chambers, S. A. Band-Gap Reduction and Dopant Interaction in Epitaxial La,Cr Co-Doped SrTiO3 Thin Films. Chem. Mater. 2014, 26, 7073–7082. Shirley, D. A. High-Resolution X-Ray Photoemission Spectrum of the Valence Bands of Gold. Phys. Rev. B 1972, 5, 4709–4714. Giannuzzi, L. A.; Drown, J. L.; Brown, S. R.; Irwin, R. B.; Stevie, F. A. Applications of the FIB Lift-out Technique for TEM Specimen Preparation. Microsc. Res. Tech. 1998, 41, 285–290. Mitchell, D. R. G. SITools v 2.0 http://sitools2.sourceforge.net/ (accessed Dec 22, 2015). Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865–3868. Perdew, J. P.; Ruzsinszky, A.; Csonka, G. I.; Vydrov, O. A.; Scuseria, G. E.; Constantin, L. A.; Zhou, X.; Burke, K. Restoring the Density-Gradient Expansion for Exchange in Solids and Surfaces. Phys. Rev. Lett. 2008, 100, 136406. Blöchl, P. E. Projector Augmented-Wave Method. Phys. Rev. B 1994, 50, 17953–17979. Kresse, G.; Furthmüller, J. Efficiency of Ab-Initio Total Energy Calculations for Metals and Semiconductors Using a Plane-Wave Basis Set. Comput. Mater. Sci. 1996, 6, 15–50. Kresse, G.; Joubert, D. From Ultrasoft Pseudopotentials to the Projector Augmented-Wave Method. Phys. Rev. B 1999, 59, 1758–1775.



Page 12 of 12

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

12

Superlattices - ACS Publications - American Chemical Society

Recommend Documents