Quantification of Charge Transfer at the Interfaces ... - ACS Publications

Subscriber access provided by BUFFALO STATE

A: New Tools and Methods in Experiment and Theory

Quantification of Charge Transfer at the Interfaces of Oxide Thin Film Qingping Meng, Guangyong Xu, Huolin L. Xin, Eric A. Stach, Yimei Zhu, and Dong Su J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.9b02802 • Publication Date (Web): 03 May 2019 Downloaded from http://pubs.acs.org on May 5, 2019

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 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 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.

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 23 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

The Journal of Physical Chemistry

Quantification of Charge Transfer at the Interfaces of Oxide Thin Film Qingping Meng†, Guangyong Xu†, Huolin Xin‡, Eric A. Stach‡, Yimei Zhu*,‡, and Dong Su*,‡ †Department

of Condensed Matter Physics and Materials Science, Brookhaven National Laboratory, Upton, New York 11973, USA

‡Center

for Functional Nanomaterials, Brookhaven National Laboratory, Upton, New York, 11973, USA

ABSTRACT: The interfacial electronic distribution in transition metal oxide thin films is crucial to their interfacial physical or chemical behaviors. Core-loss electron energy-loss spectroscopy (EELS) may potentially give valuable information of local electronic density of state at high spatial resolution. Here, we studied the electronic properties at the interface of Pb(Zr0.2Ti0.8)O3 (PZT)/4.8-nm La0.8Sr0.2MnO3 (LSMO)/SrTiO3 (STO) using valance-EELS with a scanning transmission electron microscope (STEM). Modeled with dielectric function theory, the charge transfer in the vicinity of the interfaces of PZT/LSMO and LSMO/STO were determined from the shifts of plasma peaks of VEELS, agreeing with theoretical prediction. Our work demonstrates that VEELS method enables a high-efficient quantification of the charge transfer at interfaces, shedding the light on the charge transfer issues at heterogenous interfaces in physical and chemical devices.

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry 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 2 of 23

INTRODUCTION Piezoelectric and ferroelectric oxide thin films have been widely used for various sensors and actuators, tunable microwave circuits, RF devices and nonvolatile memories due to their unparalleled variety of physical properties, such as high electron mobility and high spontaneous polarization.1-3 Now, atomic-layer control of the film growth allows the study of the intrinsic physical properties of interfacial layer. The interfaces consisted of different transition metal oxides are more complex than those in simple metal and semiconductors because of strong electronic interaction, the charge transfer and atomic restructuration.4 The unusual properties may appear in these complicated interfaces. For example, it was reported that the change of electronic structure could cause the formation of dielectric dead layer at SrTiO3/SrRuO3 interface or the appearance of superconducting layer at the LaAlO3/SrTiO3 (LAO/STO) interface.5-6 Two main mechanisms， the electronically compensated and the ionically compensated interface, were suggested to describe respectively two types of interfaces, such as, AlO2-LaO-TiO2 and AlO2-SrO-TiO2, between LAO/STO.6-7 The two mechanisms generally cause the change of electronic density near interfaces. Therefore, detecting the change of electronic density at these interfaces is primary importance to help us understanding the nature of interface. Electron energy-loss spectroscopy (EELS) with a scanning transmission electron microscope (STEM) has been widely used to detect the electronic state.6,8-9 We can obtain the information about electronic structure from core-loss EELS,8 i.e. the characteristic L-edges of transition metal resulting from excitation from the 2p levels to the largely unoccupied 3d level. The information is used to quantify the number of d electrons by measuring the relative intensities, shapes and the threshold energies of L2 and L3 peaks.9 However, the theoretical calculation of the electron


2

Page 3 of 23 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


structure in strongly correlated materials is not accurate due to core-hole effects, the statistical ration of the spin-orbit components or other unknown correlated effects.8,10-12 Therefore, using the core-loss EELS to determine the valence state of transition metal in strongly correlated materials is only experience-based. One needs a series of reference EELS spectra from other oxides. These reference oxides must have similar crystal structure to the experimental materials and be wellknown chemical compositions, so that the valence states of transition metal in these oxides can be determined and compared with that in experimental materials. However, it is not easy to obtain these reference EELS spectra. These theoretical and experimental issues make that the measurements of localized electronic structure using the core-loss EELS become difficult for broaden applications. On the other hand, the valence EELS (VEELS) associated with the low-loss part of the spectra has not been widely used though the signal of VEELS is two-order of magnitude stronger than that of the core-loss counterpart. The main reason lies in the difficulty in the interpretation of VEELS because the intensity of VEELS is dependent not only on both valence and conduction band states,8,13 but also on other complexities, such as Cherenkov radiation, the sample thickness and the interfacial plasmons.14-18 Whereas, low-loss VEELS is dominated by collective excitations (plasmons),8,10 and the physical scenario of plasmons is clear,17 i.e., the square of the plasma frequency is directly proportional to the density of excited electron. Additionally, it was reported that the plasma EELS mapping could be achieved with subnanometer spatial resolution.19 Therefore, using plasmons to extract localized charge density is possible if other contributions can be disentangled from the bulk plasma peak. In this work, the VEELS of a Pb(Zr0.2Ti0.8)O3/4.8-nm La0.8Sr0.2MnO3/SrTiO3 (PZT/4.8-nm LSMO/STO) epitaxial system is analyzed. The change of charge density of the interfacial layer is quantitatively determined based on the shift of plasma frequency.


3


Page 4 of 23

EXPERIMENTS In this work, a 4.8-nm (12 monolayers) thick La0.8Sr0.2MnO3 (LSMO) film was grown epitaxially on a TiO-terminated SrTiO3 (STO) (001) substrate via molecular-beam epitaxy (MBE) and then a 250-nm thick PbZr0.2Ti0.8O3 (PZT) film was deposited by off-axis RF magnetron sputtering. The TEM cross-sectional specimens were prepared by a focused ion beam (FIB) system after mechanical polishing down to 30μm. The experimental details for the thin film growth, structural characterization and their properties have been described elsewhere.20-22 As shown in Figure S1 of Supporting Information, the PZT/LSMO thin film in this study is pinned at the accumulation state.23 Multiple linear least-squares (MLLS) fitting and multivariate curve resolution (MCR) methods were used to analyze the experimental EELS.24 A dedicated aberration corrected STEM (Hitachi 2700C) operated at 200kV was used for simultaneous high-angle annular dark field imaging and EELS. The convergence angle the beam is 28mrad with a focused electron probe of 1.3 Å. For the collection of core-loss spectroscopy (Ti, O, Mn, and La), a collection angle of 20 mrad and a beam current of 60 pA have been used. While for the acquisition of VEELS, the collection inner angle is 15 mrad and the beam current is about 20 pA. In our experimental condition, the energy resolution of VEELS is about 0.35 eV measured from full width at half maximum (FWHM) of zero loss peak at an energy dispersion of 0.1 eV/Ch. EELS line scans perpendicularly cross the interfaces were taken at the relative thickness (t/λ) of ~ 1.0, corresponding the thickness of sample ~ 100 nm. On this PZT/LSMO/STO sandwiched structure, we performed a VEELS line-scan from PZT to STO substrate, as shown in Figure 1. Each spectrum was deconvoluted using the Fourier-log method to remove the zero-loss peak and plural scattering of each spectrum,10 and then scaled with respect to the intensity of ADF signal. The deconvoluted process is completed using the software


4

Page 5 of 23 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


of DigitalMicrograph.24 Complex dielectric functions of PZT in position coordinates x=-7.2 nm, LSMO in x=0 nm and STO in x=7.2 nm were extracted using Kramers-Kronig analysis (KKA) from their deconvoluted spectrum. The extracted complex dielectric functions are plotted in Figure S2 of Supporting Information. The detailed descriptions about how to remove zero-loss peak and plural scattering and extract complex dielectric functions can be found in chapter 4 of Ref.10.

RESULTS AND DISCUSSION Figure 2a shows an annular dark-field (ADF) STEM image of PZT/LSMO/STO interface. Ti L edges, O K edge, Mn L edges and La M edges are observed at around 355 eV, 530 eV, 640 eV and 830 eV, respectively, and their spectra are plotted at Figure 2b. The integration of these core-loss EELS signals are plotted at Figure 2c, where sharp interruptions of the elements between Mn/Ti and La/Ti are observed and the elemental diffusions across the interfaces are within one unit cell. This result confirmed the epitaxial growth of the PZT on LSMO. An atomic schematic model based on this result is shown in Figure 2d. To retrieve the composition profiles for each atomic layer, we have performed MLLS fitting on the spectrum image using three references obtained from the PZT, LSMO and STO respectively. The weights of each component are plotted in Figure 3. The linear combination of these three components would match the spectrum images at the PZT (position coordinates x=-7.2 nm), LSMO (x=0 nm) and STO (x=7.2 nm) parts, but leave considerable residual signals at the interfaces of PZT/LSMO (see bottom of Figure 3). This result indicates existence of interfacial components, which has been reported in other system.25 To confirm the existence of interfacial components, we performed multivariate curve resolution (MCR) analysis with five components respectively and the results are shown in Figure S3 in Supporting Information. MCR analysis with five components clearly demonstrates that there are no significant residuals in either the spectral domain or the


5


Page 6 of 23

spatial domain demonstrating the good fit of the model. However, MCR analysis is a method of multivariate statistical analysis. The five principal components obtained from MCR analysis do not have clear physical meaning.26 Physical methods are still needed to analyze the electronic nature of the interface. The interfacial contributions to VEELS have been considered under the framework of the dielectric function theory on this sandwiched structure. According to the dielectric function theory,27-30 the VEELS in inhomogeneous medium is not only dependent on the dielectric function of the measured point, but also is influenced by that of nearby points. Consequently, the interfacial components of VEELS from the dielectric function theory can be calculated. Complex dielectric functions of PZT, LSMO and STO layers are respectively extracted using experimental measured EELS in position coordinates x=-7.2 nm at PZT, x=0 nm at LSMO, and x=7.2 nm at STO. The complex dielectric functions as the inputs, and the formulas of appendix B in Moreau et al.’s paper,30 which descripts the calculation of interfacial EELS of a sandwich structure, are used in our calculations. Figure 4a-b show respectively the experimental and calculated VEELS from x=7.2 nm to x=7.2 nm which positions are across the interfaces of PZT/LSMO and LSMO/STO. Visually comparing Figure 4a with 4b, we can find the calculated VEELS is almost consistent with the experimental spectra. However, some discrepancies are still observed in the regions highlighted by ovals in Figures 4a and 4b. The plasma peaks in the oval area of Figure 4a have more remarkable shift of peak than the ones in the corresponding area of Figure 4b with position coordinate. The direct comparisons between experimental and calculated VEELS at three special points from internal PZT to interface of PZT/LSMO are shown in Figures 4 c-e. The two curves, the calculated and experimental VEELS at internal PZT, almost overlap in Figure 4c. The result of Figure 4c confirms that the calculated EELS from dielectric function theory is correct, reliable


6

Page 7 of 23 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


and the interfacial effects is ignorable small in this position. However, Figures 4d and 4e display deviation between the calculated and experimental curves. The deviation indicates that some extra possessions near interface were missed in the current dielectric function theory. General dielectric function theory, such as Garcia-Molina et al.,28 Bolton and Chen,29 and Moreau et al.’s,30 assume that interface is an ideal geometrical plane, so that the dielectric function across the interface of two media is a Heaviside step function. However, the actual interface has interfacial strain,11, 31-33 atomic rearrangement or charge transfer.6, 21 The atomic rearrangement or charge transfer, such as the observations of Nakagawa et al. and Gariglio et al.’s in LAO/STO interface,6, 34 and Vaz et al.’s in Pb(Zr0.2Ti0.8)O3/La0.8Sr0.2MnO3 interface,21 can lead to the polar discontinuity or the change of valence state of transition metals respectively. No matter what the polar discontinuity or the change of valence state of transition metals, the charge density at these interfacial layers should be altered. Plasma theory tell us that the altered charge density can cause the shift of plasma peak.17 Consequently, the shift of plasma peak can be used to measure the change of charge density. Utilizing the relationship between electron density and physical quantities, some physical quantities may be obtained from the shift of plasma peak. For example, nanoscale temperature is mapped using the relationship between change of electron density and thermal expansion of volume.35 Im(

1 ) , the energy loss function, provides us with a good deal in the way of useful  ( )

information about the properties of the electron state.17 For example, the frequency of collective plasma oscillation directly relates to the density of electron gas and single electron transitions that depend upon the position of critical points with the band structure of the sample the most important critical points being those which define the band gap itself.36 However, the Im(


1 ) obtained  ( )

7


Page 8 of 23

from VEELS included not only collective plasma oscillations and interband excitations of single electron, but also other extra excitations, such as surface effects and Cherenkov radiation. The extra excitations significantly alter fine structures of VEELS so that the energy-loss function obtained from VEELS cannot be determined correctly. The contributions from surface effects and Cherenkov radiation generally occur from zero to several eV. For most semiconductors and insulators, their bandgap is also several eV. It means that the influence of surface effects and Cherenkov radiation should be first removed if we want to determine the bandgap of semiconductor or insulator using VEELS.15, 37 In our recent paper,38 a mathematic approach is suggested to remove surface effects and Cherenkov radiation. In this paper, we only focus on the plasmon whose energy loss is higher than 25 eV in our system. Our calculations (see Supporting Information) evidence that the VEELS with higher energy loss than 25 eV can ignore the influences from Cherenkov radiation. In addition, the relative thickness (t/λ) of our sample is about 1.0. In the thickness, the contribution of surface losses to VEELS is also ignorable.16 To retrieve plasma component from VEELS, we first use multiple Lorentzian functions to fit whole VEELS curve since the energy-loss function can be expressed by a sum-of-Lorentzian functions and similar methods are used by other researchers.17,

39

Based on the fitting, the

percentages of other excitations in energy-loss range of plasmon are obtained. Subtracting the contributions of these excitations, a pure plasma signal is extracted. The position of plasma peak (frequency) then can be accurately determined from fitting the plasma peak. Figure 5a plots the plasma frequencies as a function of position, which is extracted from the experimental (black square dot) and calculated (red circle dot) spectra. A remarkable deviation between experimental and calculated plasma frequency can be found near the interface of PZT/LSMO as shown in Figure 5a. The deviation originates from the ignoration of change of charge density in our calculated


8

Page 9 of 23 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


processes using the dielectric function theory.28~30 Therefore, utilizing the deviation of plasma frequency, the change of charge density can be determined. According to the plasma theory, the plasma frequency can be written17,40

  2 p

4e 2 N eff

(1)

m1   0 

where N eff 

m N v   f lv , m* l v

(2)

* 𝑒 is the elementary charge; m and m are respectively the mass and the effective mass of electron;

N v is the number of valence electrons which are essentially unbound in the frequency range; the

second term

f l v

lv

in N eff is due to the valance-band electrons from the interband transition.  0

describes the real interband transitions which is zero until the real excitations from the bound band are allowed.17,40 It may be zero or near zeros if the bound bands are sufficiently far removed and uncoupled from the valence bands.40 Consequently, we ignore the effect of  0 in our calculation. As described in Eq. (2), the electrons for plasma excitation are from two possible parts: the first are the electrons from conduction band whose mass is effective electron mass; and the second are the valance-band electrons whose mass is real electron mass.17,40 For insulator, such as PZT, the electrons of plasma oscillation all come from the valance bands, so the electron mass for the plasmon of PZT is equal to the electron rest mass. It means that the calculated plasma frequencies only contain the contributions from the electrons of valance bands, i.e., the second term of Eq. (2). However, near the interface of PZT/LSMO, the electrons from LSMO will transfer to the conduction band of PZT. Consequently, the plasma frequencies of experimental observations will include the contribution of both terms of Eq. (2). The change of electron density near the interface


9


Page 10 of 23

can be determined from the calculated and experimental plasma frequencies. From Eq. (1), the change of electron density can be written as: N 





m 2 x    pc2 x   pe 4e 2

(3)

where  pe  x  and  pc  x  are plasma frequencies in coordinate x from experimental and calculated VEELS spectra respectively. Using Eq. (3) and the effective electron mass m * is 0.09m in PZT,41 the changes of electron density with position coordinates are calculated, and shown in Figure 5b. The maximum change of electron density can be obtained as 0.23 per unit cell in PZT based on the lattice constants of PZT.42 This charge transfer from LSMO to PZT is supposed to correlate to the very large magnetoelectric coupling at LSMO/PZT interface.21 Although we cannot find the change of electron density near LSMO/PZT from other measured method, our result is comparable with other system. The nominal value in LAO/STO interface is 0.5 and the measured result by other method is 0.23 ~ 0.34.6,43 The conventional approach is utilizing L3 and L2 peaks of transition metal and K edge of oxygen to determine the valence of transition metals in their oxide.44-45 This method needs to have a series of reference spectra which come from similar structure and chemical composition to experimental materials. If the reference spectra are lacking, the application of method will be limited. Figure S4 show the pre-peak of O K edge, the L edges of Mn, and the L23 ratios in the first three layers in LSMO. The remarked change in O K edge and Mn L edges can be observed, indicating the change of the valence state of Mn. This result cannot be easily to be quantified without reference spectra, but its trend is consistent with our VEELS measurement. Delocalization of VEELS is a serious issue since it may deteriorate the spatial resolution. Muller and Silcox indicate that the plasmon at the energy-loss of 25 eV at 100 kV accelerating voltage can achieve 0.5 nm spatial resolution.19 Using their approach, we estimate that the delocalization


10

Page 11 of 23 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


distance of plasmon with the energy loss of 25 eV is about 0.6 nm at 200 kV accelerating voltage. Delocalization also affects the extraction of dielectric function of LSMO at x=0 because the thickness of LSMO layer is only 4.8 nm. We also calculate the effect of dielectric function of LSMO from the neighboring medium is only about 6% using Muller’s approach,19 indicating that using plasma peak in VEELS to probe the charge density at the interface is reliable with subnanometer spatial resolution.

CONCLUSION In summary, we use the shift of plasma peaks in VEELS to measure the change of electron density near the interface of PZT/LSMO after considering the interfacial dielectric coupling and removing the contributions from other electron excitations. Our measurement shows the maximum charge transfer is about 0.23 electrons per unit cell from the LSMO layer to the PZT layer. Comparing with the core-loss EELS method, our method offers a new approach to measure the change of electron density across the interface of heterostructured thin films.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website at DOI: Additional experimental results and theoretical analysis can be found in the Supporting Information.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]. *E-mail: [email protected].


11


Page 12 of 23

ORCID Qingping Meng: 0000-0002-3218-0726 Guangyong Xu: 0000-0003-1441-8275 Huolin Xin: 0000-0002-6521-868X Eric A. Stach: 0000-0002-3366-2153 Yimei Zhu: 0000-0002-1638-7217 Dong Su: 0000-0002-1921-6683

ACKNOWLEDGMENTS We would like to thank Drs. Nan Jiang and M. G. Han for helpful discussions, and Mr. Kissinger for aid in preparing the TEM samples. We thank Drs. C.A.F. Vaz, M. Marshall and F. Walker from Yale University for preparing the thin film samples. Q. Meng, G. Xu, and Y. Zhu were supported by DOE-BES, Materials Sciences and Engineering Division under Contract DESC0012704. The electron microscopy work used resources of the Center for Functional Nanomaterials, which is a U.S. DOE Office of Science Facility, at Brookhaven National Laboratory under Contract No. DESC0012704.


12

Page 13 of 23 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


REFERENCES 1. Schlom, D. G.; Chen, L.; Pan, X.; Schmehl, A.; Zurbuchen, M. A. A Thin Film Approach to Engineering Functionality into Oxides. J. Am. Ceram. Soc. 2008, 91, 2429-2454. 2. Tagantsev, A. K.; Sherman, V. O.; Astafiev, K. F.; Venkatesh, J.; Setter, N. Ferroelectric Materials for Microwave Tunable Applications. J. Electroceram. 2003, 11, 5-66. 3. Setter, N.; Damjanovic, D.; Eng, L.; Fox, G.; Gevorgian, S.; Hong, S.; Kingon, A.; Kohlstedt, H.; Park, N. Y.; Stephenson, G. B., et al. Ferroelectric Thin Films: Review of Materials, Properties, and Applications. J. Appl. Phys. 2006, 100, 051606. 4. Hwang, H. Y.; Iwasa, Y.; Kawasaki, M.; Keimer, B.; Nagaosa, N.; Tokura, Y. Emergent Phenomena at Oxide Interfaces. Nat. Mater. 2012, 11, 103-113. 5. Stengel, M.; Spaldin, N. A. Origin of the Dielectric Dead Layer in Nanoscale Capacitors. Nature, 2006, 443, 679-682. 6. Nakagawa, N.; Hwang, H. Y.; Muller, D. A. Why Some Interfaces cannot be Sharp. Nat. Mater. 2006, 5, 204-209. 7. Chambers, S. A. Understanding the Mechanism of Conductivity at the LaAlO3/SrTiO3 (001) Interface. Surf. Sci. 2011, 605, 1133-1140. 8. Rez, P.; Muller, D. A. The Theory and Interpretation of Electron Energy Loss Near-Edge Fine Structure. Annu. Rev. Mater. Res. 2008, 38, 535-558. 9. Jang, H. W.; Felker, D. A.; Bark, C. W.; Wang, Y.; Niranjan, M. K.; Nelson, C. T.; Zhang, Y.; Su, D.; Folkman, C. M.; Baek, S. H., et al. Metallic and Insulating Oxide Interfaces Controlled by Electronic Correlations. Science, 2011, 331, 886-889. 10. Egerton, R. F. Electron Energy-Loss Spectroscopy in the Electron Microscope, 3rd ed; Springer, New York 2011.


13


Page 14 of 23

11. Aruta, C.; Ghiringhelli, G.; Bisogni, V.; Braicovich, L.; Brookes, N. B.; Tebano, A.; Balestrino, G. Orbital Occupation, Atomic Moments and Magnetic Ordering at Interfaces of Manganite Thin Films. Phys. Rev. B 2009, 80, 014431. 12. 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. 13. Van Benthem, K.; Elsässer, C.; French, R. H. Bulk Electronic Structure of SrTiO3: Experiment and Theory. J. Appl. Phys. 2001, 90, 6156-6164. 14. Chen, C. H.; Silcox, J.; Vincent, R. Electron-Energy Losses in Silicon: Bulk and Surface Plasmons and Čerenkov Radiation. Phys. Rev. B, 1975, 12, 64-71. 15. Stöger-Pollach, M.; Franco, H.; Schattschneider, P.; Lazar, S.; Schaffer, B.; Grogger, W.; Zandergen, H. W. Čerenkov Losses: A Limit for Bandgap Determination and Kramers-Kronig Analysis. Micron, 2006, 37, 396-402. 16. Erni, R.; Browning, N. D. The Impact of Surface and Retardation Losses on Valence Electron Energy-Loss Spectroscopy. Ultramicroscopy, 2008, 108, 84-99. 17. Pines, D. Elementary Excitations in Solids, W. A. Benjamin, INC. New York, 1963. 18. Ritchie, R. H. Plasma Losses by Fast Electrons in Thin Films, Phys. Rev. 1957, 106, 874-881. 19. Muller, D. A.; Silcox, J. Delocalization in Inelastic Scattering. Ultramicroscopy, 1995, 59, 195-213. 20. Su, D.; Meng, Q.; Han, M-G; Vaz, C. A. F.; Segal, Y.; Walker, F. J.; Sawicki, M.; Broadbridge, C.; Ahn, C. H. Origin of 90° Domain Wall Pinning in Pb(Zr0.2Ti0.8)O3 Heteroepitaxial Thin Films. Appl. Phys. Lett. 2011, 99, 102902. 21. Vaz, C. A. F.; Hoffman, J .; Segal, Y.; Reiner, J. W.; Grober, R. D.; Zhang, Z.; Ahn, C. H.;


14

Page 15 of 23 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


Walker, F. J. Origin of the Magnetoelectric Coupling Effect in Pb(Zr0.2Ti0.8)O3/La0.8Sr0.2MnO3 Multiferroic Heterostructures. Phys. Rev. Lett. 2010, 104, 127202. 22. Vaz, C. A. F.; Segal, Y.; Hoffman, J.; Walker, F. J.; Ahn, C. H. Growth and Characterization of PZT/LSMO Multiferroic Heterostructures. J. Vac. Sci. Technol. B, 2010, 28, C5A6. 23. Han, M.-G.; Marshall, M. S. J.; Wu, L.; Schofield, M. A.; Aoki, T.; Twesten, R.; Hoffman, J.; Walker, F. J.; Ahn, C. H.; Zhu, Y. Interface-Induced Nonswitchable Domains in Ferroelectric Thin Films. Nat. Commun. 2014, 5, 4693. 24. Gatan, Inc. Pleaston, CA, US, Digital Micrograph, http://www.gatan.com/use-mlls-fittingapproach-resolve-overlapping-edges-eels-spectrum-atomic-level 25. Borisevich, A. Y.; Chang, H. J.; Huijben, M.; Oxley, M. P.; Okamoto, S.; Niranjan, M. K.; Burton, J. D.; Tsymbal, E. Y.; Chu, Y. H.; Yu, P., et al. Suppression of Octahedral Tilts and Associated Changes in Electronic Properties at Epitaxial Oxide Heterostructure Interfaces. Phys. Rev. Lett. 2010, 105, 087204. 26. Dobigeon, N.; Brun, N. Spectral Mixture Analysis of EELS Spectrum-Images. Ultramicroscopy 2012, 120, 25-34. 27. Kröger, E. Transition Radiation, Cerenkov Radiation and Energy Losses of Relativistic Charged Particles Traversing Thin Foils at Oblique Incidence. Z. Phys. 1970, 235, 403-421. 28. Garcia-Molina, R.; Gras-Marti, A.; Howie, A.; Ritchie, R. H. Retardation Effects in the Interaction of Charged Particle Beams with Bounded Condensed Media. J. Phys. C 1985, 18, 5335-5345. 29. Bolton, J. P. R.; Chen, M. Electron Energy Loss in Multilayered Slabs. II. Parallel Incidence. J. Phys: Condens. Matter 1995, 7, 3389-3403. 30. Moreau, P.; Brun, N.; Walsh, C. A.; Colliex, C.; Howie, A. Relativistic Effects in Electron-


15


Page 16 of 23

Energy-Loss-Spectroscopy Observations of the Si/SiO2 Interface Plasmon Peak. Phys. Rev. B 1997, 56, 6774-6781. 31. Pinta, C.; Fuchs, D.; Merz, M.; Wissinger, M.; Arac, R.; Löhneysen, H. V.; Samartsev, A.; Nagel, P.; Schuppler, S. Suppression of Spin-State Transition in Epitaxially Strained LaCoO3. Phys. Rev. B 2008, 78, 174402. 32. Van der Merwe, J. H. On the Stresses and Energies Associated with Inter-Crystalline Boundaries. Proc. Phys. Soc. A 1950, 63, 616-637. 33. Meng, Q.; Wu, L.; Zhu, Y. Phonon Scattering of Interfacial Strain Field Between Dissimilar Lattices. Phys. Rev. B 2013, 87, 064102. 34. Gariglio, S.; Gabay, M.; Triscone, J.-M. Research Update: Conductivity and Beyond at the LaAlO3/SrTiO3 Interface. APL Mater. 2007, 4, 060701. 35. Mecklenburg, M.; Hubbard, W. A.; White, E. R.; Dhall, R.; Cronin, S. B.; Aloni, S.; Regan, B. C. Nanoscale Temperature Mapping in Operating Microelectronic Devices. Science, 2015, 347, 629-632. 36. Rafferty, B.; Brown, L. M. Direct and Indirect Transitions in the Region of the Band Gap Using Electron-Energy-Loss Spectroscopy. Phys. Rev. B 1998, 56, 10326-10337. 37. Stöger-Pollach, M.; Schattschneider, P. The Influence of Relativistic Energy Losses on Bandgap Determination Using Valence EELS. Ultramicroscopy, 2007, 107, 1178-1185. 38. Meng, Q; Wu, L; Xin, H. L.; Zhu, Y. Retrieving the Energy-Loss Function from Valence Electron Energy-Loss Spectrum: Separation of Bulk- Surface-Losses and Cherenkov Radiation. Ultramicroscopy, 2018, 194, 175-181. 39. Reed, B. W.; Chen, J. M.; MacDonald, N. C.; Silcox, J.; Bertsch, G. F. Fabrication and STEM/EELS Measurements of Nanometer-Scale Silicon Tips and Filaments. Phys. Rev. B


16

Page 17 of 23 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


1999, 60, 5641-5652. 40. Pillipp, H. R.; Ehrenreich, H. Optical Properties of Semiconductors. Phys. Rev. 1963, 129, 1550-1560. 41. Shin, J. C.; Hwang, C. S.; Kim, H. J.; Park, S. O. Leakage Current of Sol-Gel Derived Pb(Zr,Ti)O3 Thin Films Having Pt Electrodes. Appl. Phys. Lett. 1999, 75, 3411-3413. 42. Frantti, J.; Lappalainen, J.; Eriksson, D.; Lantto, V.; Nishio, S.; Kakihana, M.; Ivanov, S.; Rundlöf, H. Neutron Diffraction Studies of Pb(ZrxTi1-x)O3 Ceramics. Jpn. J. Appl. Phys. 2000, 39, 5697-5703. 43. Cantoni, C.; Gazquez, J.; Granozio, F. M.; Oxley, M. P.; Varela, M.; Lupini, A. R.; Pennycook, S. J.; Aruta, C.; Di Uccio, U. S.; Perna, P., et al. Electron Transfer and Ionic Displacements at the Origin of the 2D Electron Gas at the LAO/STO Interface: Direct Measurements with Atomic-Column Spatial Resolution. Adv. Mater. 2012, 24, 3952-3957. 44. Varela, M.; Oxley, M. P.; Luo, W.; Tao, J.; Watanabe, M.; Lupini, A. R.; Pantelides, S. T.; Pennycook, S. J. Atomic-Resolution Imaging of Oxidation States in Manganites. Phys. Rev. B 2009, 79, 085117. 45. Wang, Z. L.; Bentley, J.; Evans, N. D. Mapping the Valence States of Transition-Metal Elements Using Energy-Filtered Transmission Electron Microscopy. J. Phys. Chem. B 1999, 103, 751-753.


17


Page 18 of 23

Figure 1. Spectrum image of VEELS scanning crossing the PZT/LSMO/STO structure.


18

Page 19 of 23 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 2. (a) Cross-sectional ADF-STEM image of the PZT/LSMO/STO structure. Green line is moved route the electron beam. (b) The core-loss EELS spectrum image of Ti L, O K, Mn L and La M edge along the green line shown in (a). (c) Relative intensities of elements Ti, O, Mn, and La crossing the PZT/LSMO/STO structure. The sharp interruptions of the elements between Mn/Ti and La/Ti are observed. It indicates that the elemental diffusions crossing the interfaces are within one unit cell. d) An atomic schematic model of the PZT/LSMO/STO structure.


19


Page 20 of 23

Figure 3. Concentration by MLLS fitting with three components of PZT, LSMO and STO. The bottom is the residual signal after fitting. The significant residual signals are found at interface of PZT/LSMO.


20

Page 21 of 23 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. Experimental (a) and calculated (b) VEELS spectra. The inconsistency between experimental and calculated spectra is found in the oval areas of the two figures. The experimental spectra (a) in its oval area are evenly distribution with position coordinate, but the calculated spectra (b) in its oval area are unevenly. It means that the plasma frequencies of experimental observation are more obvious shift than that of calculated results with position coordinates. (c) ~ (e) the experimental and calculated VEELS are directly compared in some special position coordinates. (c) x=-7.20 nm is far away from the PZT/LSMO interface in PZT, (d) x=-3.36 nm near the PZT/LSMO interface, but still in PZT, and (e) x=-2.40 nm at PZT/LSMO interface.


21


(a)

Page 22 of 23

(b)

Figure 5. (a) the plasma frequency in function of position coordinate. The black squares are from experimental spectra and the red circles are from calculated spectra. (b) The change of valence electron density calculated from the shift of the plasma frequencies with position coordinate. Squares are the calculated results using Eq. (3). Solid lines are fitting curve using quadratic function.


22

Page 23 of 23 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


TOC Graphic


23

Quantification of Charge Transfer at the Interfaces ... - ACS Publications

Recommend Documents