Electron-Beam-Induced Antiphase Boundary Reconstructions in a

Aug 22, 2016 - The availability of aberration correctors for the probe-forming lenses makes simultaneous modification and characterization of material...
64 downloads 6 Views 7MB Size
Research Article www.acsami.org

Electron-Beam-Induced Antiphase Boundary Reconstructions in a ZrO2‑LSMO Pillar-Matrix System Dan Zhou,*,⊥ Wilfried Sigle, Marion Kelsch, Hanns-Ulrich Habermeier, and Peter A. van Aken Max Planck Institute for Solid State Research, Heisenbergstrasse 1, 70569 Stuttgart, Germany S Supporting Information *

ABSTRACT: The availability of aberration correctors for the probe-forming lenses makes simultaneous modification and characterization of materials down to atomic scale inside a transmission electron microscopy (TEM) realizable. In this work, we report on the electron-beam-induced reconstructions of three types of antiphase boundaries (APBs) in a probeaberration-corrected TEM. With the utilization of high-angle annular dark-field scanning transmission electron microscopy (STEM), annular bright-field STEM, and electron energy-loss spectroscopy, the motion of both heavy element Mn and light element O atomic columns under moderate electron beam irradiation are revealed at atomic resolution. Besides, Mn segregated in the APBs was observed to have reduced valence states which can be directly correlated with oxygen loss. Charge states of the APBs are finally discussed on the basis of these experimental results. This study provides support for the design of radiation-engineering solid-oxide fuel cell materials. KEYWORDS: electron-beam-induced reconstruction, antiphase boundary, annular bright-field imaging, oxygen loss, charge state

1. INTRODUCTION

structural and chemical modifications and characterizations down to the atomic scale. In a recent report, we showed that the formation of antiphase boundaries (APBs) provides a means for strain relaxation in a system composed of ZrO2 pillars in La2/3Sr1/3MnO3 thin films. Three types of APBs were observed and studied in detail.35,36 In the course of this work, we found that under prolonged electron irradiation a structural and chemical reconstruction of the APBs appeared. This reconstruction is studied in situ at atomic resolution using high-angle annular dark-field (HAADF) and annular bright-field (ABF) scanning transmission electron microscopy (STEM) and simultaneous electron energy-loss spectroscopy (EELS). Finally, we discuss the reconstruction mechanism of the APBs in the context of radiation effects and electrostatic coupling at the boundaries.

The radiation effects in charged particle microscopes, like scanning/transmission electron microscopes,1−5 helium ion microscopes6−9 and focused ion beam microscopes,10−12 are usually thought to introduce undesirable disorder and thus deteriorate the material. However, with controlled dose, dose rate, and beam energy, depending on the radiation mechanism,13 these radiations may have beneficial effects on nanostructured materials, like self-organization and self-assembly.14−19 Recent reports demonstrate that they can also be used to tailor the mechanical,20,21 electronic,22 and even magnetic properties6,23,24 of the material. For example, the local crystal structure and composition are critical to control many emerging phenomena in perovskite materials, like superconductivity, magneto-electric coupling, and quantum Hall effect.25−27 Herklotz et al.28 finely tuned the lattice symmetry of SrRuO3 by modifying the local octahedral bonding angle and length with controlled insertion of He atoms. A structural phase evolution from perovskite to brownmillerite to perovskite in La2/3Sr1/3MnO3 film can be fully controlled and monitored using electron-beam irradiation in a transmission electron microscope (TEM) through an incessant ordering of electron-beam-induced oxygen vacancies, as reported by Yao et al.29 Another example is the defect-controlled structural, optical, or electrical properties in two-dimensional materials, such as WSe2,30 MoS2,31 and graphene.5,15,18 Among the charged particles, electrons in scanning electron microscopy32 and transmission electron microscopy33,34 have attracted more and more interest due to the possibility of simultaneous © 2016 American Chemical Society

2. RESULTS 2.1. Geometric Arrangement Change under Electron Beam Illumination. Aberration-corrected HAADF STEM images of the three APB types before and after electron beam modification are presented in Figure 1. The details of the chemistry and structure of these APBs before electron beam modification were already presented in ref 35. To be consistent with our former reports, we keep using the denotations of APB-1, APB-2, and APB-3 and LSMO in the tetragonal notation (space Received: June 6, 2016 Accepted: August 22, 2016 Published: August 22, 2016 24177

DOI: 10.1021/acsami.6b06621 ACS Appl. Mater. Interfaces 2016, 8, 24177−24185

Research Article

ACS Applied Materials & Interfaces

Figure 1. HAADF image of a plan-view 80 mol % LSMO−20 mol % ZrO2 sample showing APB-1 and APB-2 (a) before and (b) after electron-beaminduced reconstruction and APB-3 (c) before and (d) after electron-beam-induced reconstruction. Double-arrowed magenta lines with arrows point to the same atomic columns before and after electron-beam-induced reconstructions revealing the nearby atomic column displacements.

group I4/mcm, No. 140). There is a phase shift of half a {110} plane distance across the boundary along the [110] direction for APB-1 and APB-2 and along the [11̅0] direction for APB-3. The boundary planes are {110} for APB-1 and {310} for APB-2 and APB-3. The ladder-like features inside these APBs are Mn/O atomic columns with smaller occupation compared to that in the matrix region. Structures before/after irradiation are shown in Figure 1a,b for APB-1 and APB-2 and in Figure 1c,d for APB-3. The HAADF images of the matrix region show bright columns corresponding to the heavy A-site (La/Sr) atomic columns and weaker columns corresponding to the lighter B′-site (BO-site, Mn/O) atomic columns in La2/3Sr1/3MnO3. Due to the low elastic scattering cross section of oxygen atoms and the angular range used for HAADF imaging, the pure oxygen ion columns are invisible in HAADF images. Comparison of images before and after irradiation reveal some differences. First, the ladder-like appearance of Mn/O atomic column inside these APBs disappears. The arrangement of these Mn/O atomic columns becomes more like in the matrix region if the difference between A-sites and B′-sites is ignored. Second, local atomic plane shifts are observed. The original phase shift of half a {110} plane distance disappears for APB-1 and APB-2 while the boundary plane remains the same, as shown in Figure 1a,b. Lower magnification HAADF images, as included in Figure S1, confirm the disappearance of the original phase shift. For APB-3 (Figure 1c,d), another phase shift of half a {110} plane distance along the [110] direction is found in addition to the original phase shift of half a {110} plane distance along the [110̅ ] direction. This leads to an overall local phase shift of half a {100} plane distance along the [100] direction for APB-3 under moderate electron beam irradiation. The shifted atomic planes are the ones perpendicular to the phase shift direction [110] for APB-1 and APB-2 and parallel to the phase shift direction [11̅0] for APB-3. They are actually the same atomic planes and

differentiated by their relationships to the phase shift direction of these APBs before electron-beam-induced reconstructions. The lines drawn in magenta color connect two A-site atomic columns on two sides of these APBs before electron beam modification (Figure 1a,c) and are copied without any rotation to the images acquired after electron beam modification. One end of the copied lines was placed on the same A-site atomic column as before irradiation (Figure 1b,d). We find that this line also ends on A-site atomic columns on the other side of the APB, indicating no change of the atomic column positions at some distance away from the APB walls for all the APBs. Close to the APBs, the deviation from the magenta line is clearly different before and after irradiation. These modifications are achieved by atomic column shifts to opposite directions on the two sides of the APB walls, i.e., left side upward and right side downward for the situations presented in Figure 1. These local modifications occur in a region extending to about 1 nm away from the APB boundaries. It is worth noting that the plane spacing parallel to the APB plane is not influenced by the APB. Third, the number of B′-site atomic columns inside the APB walls is reduced for APB-1 and APB-2 but not for APB-3. As demonstrated in Figure 1a,b by the orange marks, there is one less B′-site column inside APB-1 walls for every 5 B′-sites columns and inside APB-2 walls for every 4 B′-site columns after irradiation. Visually, the two atomic columns at the center of the APB-1 and APB-2 have merged. The 6 B′-site columns inside APB-3 walls remain after irradiation, as given in Figure 1c,d. Last, the B′-site occupation in APB-1 and APB-2 is more uniform as judged from the more homogeneous HAADF intensity after electron-beam-reconstructions as compared to that of before. For APB-3, no visible redistribution of B′-site occupation can be observed from the HAADF images. 2.2. Elemental Distribution and Valence States. In our previous report,35 we showed that these APBs are Mn-rich. To investigate the elemental distribution and chemical states inside 24178

DOI: 10.1021/acsami.6b06621 ACS Appl. Mater. Interfaces 2016, 8, 24177−24185

Research Article

ACS Applied Materials & Interfaces

edges are normalized to the maximum intensity in the presented energy-loss range, i.e., the peak intensity at the main peak around 535 eV. The Mn-L2,3 edges are normalized to the maximum intensity in the presented range, i.e., the Mn L3 peak at about 640 eV. The intensity of the oxygen prepeak at 526 eV is clearly reduced after irradiation. The dominant reason for this prepeak intensity is the number of unoccupied metal 3d states available for mixing or hybridizing with the O 2p states; this means that the prepeak intensity decreases with a decreasing number of unoccupied 3d states.37 Therefore, the prepeak intensity reduction can be attributed to a few reasons: (i) substitution of Mn by elements with less unoccupied 3d states, like Fe, Co, or Ni,37 or even like Ga,38 whose 3d band is fully occupied, which however reduces the sum of the unoccupied 3d states hybridized with O 2p orbitals, (ii) variation of Mn valence by changing the ratio of La and Sr (since La has a valence of 3+ and Sr of 2+, an increase of the La concentration will decrease the Mn valence states and thus reduce the amount of unoccupied 3d states which will also lead to the reduction of O prepeak intensity), and (iii) formation of oxygen vacancies which hinders the mixing of O 2p with unoccupied Mn 3d states by reducing the available O 2p states. These processes can happen independently to modify the O-K prepeak intensity or happen in a combined way. Moreover, the energy separation between the prepeak and the adjacent main peak was found to decrease with lowering of the Mn valence. Our experimental data of O-K edges shown in Figure 3a reveal a decrease of the prepeak intensity, a shift of the prepeak to higher energy, and a reduced separation between the prepeak and the adjacent main peak after irradiation. No variation of the Lato-Sr ratio and no other elements substituting Mn were observed in the elemental distribution analysis. Therefore, the observed change of O-K edge can only be attributed to the creation of oxygen vacancies, which leads to a reduction of the Mn valence state, as confirmed from the Mn-L3 peak positions shown in Figure 3b. Nonlinear least-squares fitting of a Gaussian peak to the background-subtracted Mn-L3 peaks was applied to verify the valence state of Mn. Spectra from the matrix region and the APB region show Mn-L3 peak positions of 638.73 and 637. 53 eV, respectively, i.e., a difference of 0.19 and 1.39 eV, respectively, compared to the spectra collected before modification of 638.92

the APB walls after electron beam modification, EELS spectrum imaging (SI) was applied, as presented in Figure 2 for APB-1 and

Figure 2. (a) ADF image of the APB-1 area in Figure 1b in a plan-view 80 mol % LSMO−20 mol % ZrO2 sample. EELS spectrum image of (b) O-K, (c) La-M4,5, and (d) Mn-L2,3. (e) Mn-L3 peak positions obtained from Gaussian fitting of the background subtracted Mn-L3 spectrum image.

Figures S2 and S3 for APB-2 and APB-3, respectively. These figures reveal that the Mn enrichment is still present after structural modification. No diffusion of A-site atoms into the APB is observed. Figure 3a,b presents the O-K and Mn-L2,3 edges of the region inside the APB walls before and after electron-beam modification and the LSMO matrix region close to the APB walls after electron-beam modification, as detailed in the figures. The O-K

Figure 3. Comparison of (a) O-K and (b) Mn-L2,3 near-edge fine structures from the APB region before electron beam modification (labeled by “APB before”), the APB region after electron beam modification (labeled by “APB after”), and LSMO matrix region near the APB walls after electron beam modification (labeled by “LSMO”). 24179

DOI: 10.1021/acsami.6b06621 ACS Appl. Mater. Interfaces 2016, 8, 24177−24185

Research Article

ACS Applied Materials & Interfaces

Figure 4. (a) Starting structure model of APB-1, as derived in ref 35. Possible structure of APB-1 after electron beam modifications: (b) nonpresence and (c) presence of oxygen columns between B′-sites inside the APB. Simultaneously acquired (d) HAADF image and (e) ABF image of a region including APB-1 after electron beam illumination of the plan-view 80 mol % LSMO−20 mol % ZrO2 sample. Line profiles along the blue line in (f) the HAADF image and (g) the ABF image. The line profiles are integrated along the horizontal direction for 3 pixels. The magenta lines are drawn to help identify the APB walls and thus confirm the structure of APB-1 after electron-beam-modifications, as shown in the Figure S4.

eV. Comparison with literature data shows that this shift to lower energies can be interpreted as a reduced valence state of Mn.39−42 The comparison also reveals that the APB region is influenced stronger and more rapidly by electron-beam irradiation than the LSMO matrix region. Application of the same nonlinear leastsquares fitting of a Gaussian peak to the Mn spectrum image from these three APB regions shows a consistent gradual decrease of Mn valence state from matrix to boundary regions, as shown in Figure 2e. 2.3. O Position Determination. To retrieve the oxygen atomic positions in and at the APBs after electron-beam modifications, simultaneous HAADF and ABF images are acquired. It should be noted that in HAADF images atoms appear bright on a dark background, whereas in ABF imaging atoms appear dark on a bright background. As shown in Figure 4 for APB-1, the disappearance of a phase shift of half {110} plane distance under electron beam illumination could give two possible structures, as given in Figure 4b,c, starting from the structure shown in Figure 4a. The difference of these two structures relies on the oxygen columns between two B′-sites in the boundary. Qualitatively speaking, this difference can occur if all the oxygen atoms in the B′-sites columns inside the APBs move together with Mn. If so, one would obtain the final structure shown in Figure 4b. Otherwise, one would obtain the structure shown in Figure 4c. The results from simultaneously acquired HAADF and ABF images are shown in Figure 4d−g. From the line profile drawn across 3 B′sites inside the APB, the existence of pure oxygen columns can be recognized from the slight damping of the intensity between neighboring B′-sites in the ABF line profile. These results confirm the structure to be as depicted in Figure 4c.

The situation for APB-2 is shown in Figure 5. The simultaneously acquired HAADF and ABF images confirm the

Figure 5. (a) Starting structure model of APB-2, as derived in ref 35. (b) Structure model of APB-2 after electron beam modifications.

structure after electron modifications to be as shown in Figure 5b. Visually, the two B′ columns and the two pure O columns, marked by light blue frames in Figure 5a, merge to one B′ column and one pure O column, respectively. The homogeneous HAADF intensity of B′ columns inside the wall, as shown in Figure 1b, revealed the further redistribution of B′ column inside the APB walls. Figure 6 shows details of APB-3 after electron-beam-induced reconstructions. Derived from the heavy atomic column arrangement from HAADF images, the structure of APB-3 is shown in Figure 6b,c, with differences in the presence of oxygen 24180

DOI: 10.1021/acsami.6b06621 ACS Appl. Mater. Interfaces 2016, 8, 24177−24185

Research Article

ACS Applied Materials & Interfaces

Figure 6. (a) Starting structure model of APB-3, as derived in ref 35. Possible structure of APB-3 after electron beam modifications: (b) nonpresence and (c) presence of oxygen columns between B′-sites along the ⟨110⟩ direction inside the APB. Simultaneously acquired (d) HAADF image and (e) ABF image of a region including APB-3 after electron beam illumination of the plan-view 80 mol % LSMO−20 mol % ZrO2 sample. Line profiles are shown along the magenta line in (f) the HAADF image and (g) the ABF image and along the blue line in (h) the HAADF image and (i) the ABF image. The line profiles are integrated along the vertical direction for 3 pixels.

there is more reduction of Mn valence states inside the APB than the LSMO matrix region under the same electron beam illumination, as shown in Figure 3. Therefore, the modifications observed in this paper can be ascribed to the interplay between the strain imposed by the formation of unpaired misfit dislocations, changes in oxygen vacancy content and cation mobility, and the ordering of oxygen vacancies. The process can be described as follows. The rupture of bonds between oxygen and A-sites and oxygen and Mn leads to oxygen loss inside the boundaries during electron beam irradiation. Owing to oxygen loss, A-site and Mn atoms are less strongly bound and have lower displacement energy and thus can move under further electron beam irradiation toward a stable state. For APB-2 and APB-3, a common feature in the boundaries before irradiation is the presence of two or more pure closely spaced oxygen columns. This might cause enhanced responsiveness to the electron beam than is the case for the other oxygen atoms or columns. The occupancy of the B′-sites inside the APBs can be achieved from the background-subtracted signal in HAADF images.49 From the background-subtracted signal ratio between B′ columns in the APB with those in the LSMO matrix region, where the B′ occupancy is assumed to be unity, we find Mn occupancy for APB-1 to be 0.57 ± 0.24 and 0.83 ± 0.08 before and after irradiation, for APB-2 to be 0.65 ± 0.26 and 0.89 ± 0.24 before and after irradiation, and for APB-3 to be 0.73 ± 0.17 and 0.72 ± 0.35 before and after irradiation. As already mentioned in Section 3.1, for APB-1, 5 B′-sites are observed to “merge” to 4 B′sites. If only oxygen is lost during irradiation, the Mn occupancy after irradiation is 0.71 (5 times 0.57 divided by 4), which is close to but lower than the measured value of 0.83. For APB-2,

columns between neighboring B′-sites atomic columns inside the APB walls. The results from simultaneously acquired HAADF and ABF images are shown in Figure 6d−i. From the line profiles, the existence of pure oxygen columns can be recognized from the arrowed slight damping of the intensity between neighboring Aand B′-sites (Figure 6g) or neighboring B′-sites (Figure 6i) in the ABF line profile. These results confirm the structure to be as represented in Figure 6c.

3. DISCUSSIONS 3.1. Occurrence of the Electron-Beam-Induced Reconstructions. As reported before, Mn oxides are sensitive to electron beam illumination,29,43,44 especially when a focused electron beam in an aberration-corrected TEM is used. In the case of transitional metal-oxides, radiolysis is believed to be the dominant damage process and occurs via the Knotek-Feibelman mechanism.45,46 The incident electron creates a hole on the highest occupied cation core-level which is filled by an electron from the neighboring oxygen anion. This is accompanied by the emission of one or more Auger electrons leaving back a neutral or even positively charged O atom. The O atom is then repelled by the surrounding metal ions and ejected from its original site. The process continues until the material becomes sufficiently conducting to screen the positive oxygen ion. The described rupture of Mn−O bonds, probably assisted by strain and local electric-fields, can cause the movement of both Mn and O atom columns. The existence of defects, like an interface, and strain can speed up the process described above. Oxygen vacancy formation/ ordering is found to drive the lattice mismatch accommodation/ strain relief mechanism.47,48 A good proof of this claim is that 24181

DOI: 10.1021/acsami.6b06621 ACS Appl. Mater. Interfaces 2016, 8, 24177−24185

Research Article

ACS Applied Materials & Interfaces

3.2. Atomic Occupation and Valence Discussions. From the results above, we can exclude heavy atom loss during irradiation. Similar to our previous report,35 assuming pure ionicity, we get the charge sequence of planes parallel to boundary planes for these three boundaries after electron-beaminduced reconstructions, as shown in Table 1. The illustrations of getting these planes for APB-2 and APB-3 can be found in the Figure S5. Columns in the boundary region are highlighted by different colors. To differentiate from the original structure, where we let OMn and OO be the average occupancies of Mn and O inside the APBs relative to the ones in the bulk, here we let OMn′ and OO′ be the average occupancies of Mn and O inside the boundaries relative to the ones in the bulk and VMn′ to be the average Mn valence state in the APB region. To achieve zero total charge across the APB, the following conditions have to be met:

considering the merge and redistribution, the calculated Mn occupancy should be 0.87 if no Mn is lost, which is close to the experimentally measured value. For APB-3, the value should be the same, as calculated and observed. Inspection of a time-series of HAADF images (Movie S1) shows that the observable A-site and B′-site atomic column movement stops at about 150 s. Quantitative analysis of this time-series of HAADF images gives the Mn occupancies shown in Figure 7, showing a decrease until

Figure 7. Mn occupancy of APB-3 measured from the time-series HAADF images during reconstructions.

APB‐1:

2 + 3 × ( −4OO′ + OMn′VMn′) = 0 3

(1)

APB‐2:

2 + 3 × ( −4OO′ + OMn′VMn′) = 0 3

(2)

APB‐3:

2 + 6 × ( −4OO′ + OMn′VMn′) = 0 3

(3)

Therefore,

about 150 s, then a gradual increase until about 400 s, and kept constant within error consideration afterward. These results could be correlated with the ordered to disordered and again to ordered atomic column arrangement under electron beam illumination. Prolonged electron-beam irradiation without changing probe size or dwell time over 20 min does not further modify the structures. However, a large increase of dose rate by increasing probe size, condenser aperture, or dwell time leads to much darker contrast in the APBs indicating loss of Mn atoms and even amorphization in the boundaries.

APB‐1: OO′ =

1 1 OMn′VMn′ + 4 18

(4)

APB‐2: OO′ =

1 1 OMn′VMn′ + 4 18

(5)

APB‐3: OO′ =

1 1 OMn′VMn′ + 4 36

(6)

The relationship between original occupancy and occupancy after radiation is as follows:

Table 1. Charge Sequence Across the APB Walls after Electron-Beam-Induced Reconstructionsa

a

For APB-1, the direction from left to right corresponds to [11̅0] and the direction from top to bottom to [1̅1̅0]. For APB-2 and APB-3, the direction from left to right corresponds to [1−30] and the direction from top to bottom to [3̅1̅0]. 24182

DOI: 10.1021/acsami.6b06621 ACS Appl. Mater. Interfaces 2016, 8, 24177−24185

Research Article

ACS Applied Materials & Interfaces APB‐1: 3OMn′ = 4OMn

(7)

APB‐2: 3OMn′ = 4OMn

(8)

APB‐3: OMn′ = OMn

(9)

two sides of the APBs of the LSMO matrix region move locally, accompanying the movement of B′-sites inside the APBs. From combined HAADF and ABF imaging, we obtain structure models with all positions of cation and anion atomic columns. Moreover, despite the change of the number of B′-sites inside the APBs, all APBs are confirmed to lose no heavy atoms, i.e., Mn, as analyzed from the background subtracted HAADF signal ratio. The mechanism of electron-beam-induced reconstructions is supposed to be driven by the strengthened radiolysis at the interface region. Due to the loss of oxygen under electron beam irradiation and therefore rupture of bonds between oxygen and heavy atoms, the structures went from order to disorder and to order again by local area atomic column movement to reach a new stabilized state. From charge analysis, we conclude that none of the APBs’ electric fields depend on the elemental occupancies once charge neutrality is achieved.

Here, we use x to represent the reduced valence, i.e., 10 VMn′ = VMn − x = 3 − x . We also have APB‐1: OMn =

9 1 OO − 10 20

(10)

APB‐2: OMn =

6 1 OO − 5 20

(11)

) 1 30

APB‐3: OMn = OO −

(12)

5. SAMPLE PREPARATION AND TEM DETAILS Zirconium oxide (ZrO2) and lanthanum strontium manganese oxide (La2/3Sr1/3MnO3, LSMO) were codeposited epitaxially on (001) singlecrystalline lanthanum aluminum oxide (LaAlO3, LAO) substrate by pulsed laser deposition. Stoichiometric amounts of LSMO and ZrO2 according to (1−x)LSMO + xZrO2, with x = 0.2, were used. Details of the material growing process can be found in ref 50. The plan-view specimens for transmission electron microscopy (TEM) studies were prepared by grinding and dimpling followed by low-temperature (at liquid nitrogen temperatures) argon ion thinning with a precision ion polishing system (PIPS, Gatan, model 691) to achieve electron transparency. High-angle annular dark-field (HAADF) images, electron energy-loss (EEL) spectrum images, and annularbright-field (ABF) images from plan-view specimens were obtained using a probe-aberration-corrected JEOL ARM200F microscope operated at 200 kV, equipped with a cold-field emission gun (CFEG) and a Gatan GIF Quantum ERS imaging filter with dual-EELS acquisition capability. The experimental convergence angle was 28 mrad for HAADF and EELS imaging. The corresponding inner and outer collection semiangles for HAADF were set to 75−310 mrad. The inner and outer collection semiangles for annular dark-field (ADF) images acquired simultaneously during EELS spectrum imaging with Gatan ADF detector were 67−166 mrad, and the collection angle for EELS spectrum imaging was 67 mrad. Multivariate statistical analysis (MSA) was performed to reduce the noise of the EEL spectra with weighted principle-component analysis (PCA). The time-series HAADF image stacks were acquired with 20 μs/pixel, 1000 μs flyback time, and 512 × 512 pixels per frame resulting in 5.75 s total acquisition time per frame. We acquired 30 frames per series. The convergence angle was 28 mrad, and the collection angle was 75−310 mrad. The corresponding electron beam current was 25 pA with the condenser lens set to spot size 8. To minimize electron illumination on the interface region before image acquisitions, we did focusing and astigmatism correction on a nearby area remote from the interface. Modification of electron dose rate by convergence angle, condenser lens setting, and collection time per pixel can change the time of structural variation and imaging acquisition time. The experimental convergence angle was 20.4 mrad for HAADF and ABF imaging. The corresponding inner and outer collection semiangles for HAADF were 75−310 mrad and for ABF were 11−23 mrad.

Then, we have APB‐1: OO′ = OO −

3 ⎛⎜ 1 ⎞⎟ x OO − ⎝ 10 18 ⎠

(13)

APB‐2: OO′ =

4 2 ⎛ 1 ⎞ ⎟ OO − x⎜OO − 3 5 ⎝ 24 ⎠

(14)

APB‐3: OO′ =

5 1 ⎛ 1 ⎞ ⎟ OO − x⎜OO − 6 4 ⎝ 30 ⎠

(15)

To estimate how much oxygen is lost in each building block, we calculate for APB‐1: ΔO = 6OO − 6OO′ =

⎛9 1 ⎞ ⎜ O − ⎟x O ⎝5 10 ⎠

(16)

APB‐2: ΔO = 8OO − 6OO′ =

⎛ 12 1 ⎞ ⎜ ⎟x O − ⎝ 5 O 10 ⎠

(17)

⎛ 1 ⎞⎟ APB‐3: ΔO = 10OO − 12OO′ = ⎜3OO − x ⎝ 10 ⎠

(18)

Suppose VMn′ = 2 (assuming MnO), i.e., x =

4 , 3

then

APB‐1: ΔO =

12 2 OO − 5 15

(19)

APB‐2: ΔO =

16 2 OO − 5 15

(20)

APB‐3: ΔO = 4OO −

2 15

(21)

From eqs 4−6, we can rewrite the charge sequence in the third line of each APB in Table 1, respectively. All APBs do not depend on the occupancies. This is different from the situations before electron beam modifications, where the dipole moments of APB1 and APB-3 depend on the atom occupancy to minimize the electric field and that of APB-2 does not.



ASSOCIATED CONTENT

S Supporting Information *

4. CONCLUSIONS Electron-beam-induced reconstructions of the structure, composition, and charge state of three APBs in LSMO/ZrO2 were observed and analyzed. For all the APBs, the Mn segregation inside the boundaries is present before and after irradiation but with a reduced valence state after irradiation. Atomic columns on

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b06621. HAADF images; ADF image; EELS spectrum images; illustration of plane components (PDF) HAADF time series movie (AVI) 24183

DOI: 10.1021/acsami.6b06621 ACS Appl. Mater. Interfaces 2016, 8, 24177−24185

Research Article

ACS Applied Materials & Interfaces



(16) Lin, X. M.; Parthasarathy, R.; Jaeger, H. M. Direct Patterning of Self-Assembled Nanocrystal Monolayers by Electron Beams. Appl. Phys. Lett. 2001, 78 (13), 1915−1917. (17) Maas, D. J.; van Veldhoven, E.; Chen, P.; Sidorkin, V.; Salemink, H. W. M.; van der Drift, E.; Alkemade, P. F. a. Nanofabrication with a Helium Ion Microscope. Proc. SPIE 2010, 7638, 763814. (18) Bell, D. C.; Lemme, M. C.; Stern, L. a; Williams, J. R.; Marcus, C. M. Precision Cutting and Patterning of Graphene with Helium Ions. Nanotechnology 2009, 20 (45), 455301. (19) Qian, H. X.; Zhou, W. Self-Organization of Ripples on Ti Irradiated with Focused Ion Beam. Appl. Surf. Sci. 2012, 258 (6), 1924− 1928. (20) Peng, B.; Locascio, M.; Zapol, P.; Li, S.; Mielke, S. L.; Schatz, G. C.; Espinosa, H. D. Measurements of near-Ultimate Strength for Multiwalled Carbon Nanotubes and Irradiation-Induced Crosslinking Improvements. Nat. Nanotechnol. 2008, 3 (10), 626−631. (21) Kis, A.; Csányi, G.; Salvetat, J.-P.; Lee, T.-N.; Couteau, E.; Kulik, a J.; Benoit, W.; Brugger, J.; Forró, L. Reinforcement of Single-Walled Carbon Nanotube Bundles by Intertube Bridging. Nat. Mater. 2004, 3 (3), 153−157. (22) Gómez-Navarro, C.; de Pablo, P. J.; Gómez-Herrero, J.; Biel, B.; Garcia-Vidal, F. J.; Rubio, a; Flores, F. Tuning the Conductance of Single-Walled Carbon Nanotubes by Ion Irradiation in the Anderson Localization Regime. Nat. Mater. 2005, 4 (7), 534−539. (23) Lee, H.; Miyamoto, Y.; Yu, J. Possible Origins of Defect-Induced Magnetic Ordering in Carbon-Irradiated Graphite. Phys. Rev. B: Condens. Matter Mater. Phys. 2009, 79 (12), 1−4. (24) Esquinazi, P.; Spemann, D.; Höhne, R.; Setzer, A.; Han, K.-H.; Butz, T. Induced Magnetic Ordering by Proton Irradiation in Graphite. Phys. Rev. Lett. 2003, 91 (22), 227201. (25) Hwang, H. Y.; Iwasa, Y.; Kawasaki, M.; Keimer, B.; Nagaosa, N.; Tokura, Y. Emergent Phenomena at Oxide Interfaces. Nat. Mater. 2012, 11 (2), 103−113. (26) Yu, P.; Chu, Y. H.; Ramesh, R. Oxide Interfaces: Pathways to Novel Phenomena. Mater. Today 2012, 15, 320−327. (27) Zubko, P.; Gariglio, S.; Gabay, M.; Ghosez, P.; Triscone, J.-M. Interface Physics in Complex Oxide Heterostructures. Annu. Rev. Condens. Matter Phys. 2011, 2 (1), 141−165. (28) Herklotz, A.; Wong, A. T.; Meyer, T.; Biegalski, M. D.; Lee, H. N.; Ward, T. Z. Controlling Octahedral Rotations in a Perovskite via Strain Doping. Sci. Rep. 2016, 6 (February), 26491. (29) Yao, L.; Majumdar, S.; Akaslompolo, L.; Inkinen, S.; Qin, Q. H.; Van Dijken, S. Electron-Beam-Induced Perovskite-BrownmilleritePerovskite Structural Phase Transitions in Epitaxial La2/3Sr1/3MnO3 Films. Adv. Mater. 2014, 26 (18), 2789−2793. (30) Stanford, M. G.; Pudasaini, P. R.; Belianinov, A.; Cross, N.; Noh, J. H.; Koehler, M.; Mandrus, D. G.; Duscher, G.; Rondinone, A. J.; Ivanov, I. N.; Ward, T. Z.; Rack, P. D. Focused Helium-Ion Beam Irradiation Effects on Electrical Transport Properties of Few-Layer WSe2: Enabling Nanoscale Direct Write Homo-Junctions. Sci. Rep. 2016, 6, 27276. (31) Fox, D.; Zhou, Y.; Maguire, P.; O’Neill, A.; O’Coileain, C.; Gatensby, R.; Glushenkov, A. M.; Tao, T.; Duesberg, G. S.; Shvets, I.; Abid, M.; Abid, M.; Wu, H.-C.; Chen, Y.; Coleman, J. N.; Donegan, J. F.; Zhang, H. Nanopatterning and Electrical Tuning of MoS2 Layers with a Sub-Nanometre Helium Ion Beam. Nano Lett. 2015, 15, 5307. (32) Fox, D.; O’Neill, A.; Zhou, D.; Boese, M.; Coleman, J. N.; Zhang, H. Z. Nitrogen Assisted Etching of Graphene Layers in a Scanning Electron Microscope. Appl. Phys. Lett. 2011, 98 (24), 243117. (33) Lee, J.; Zhou, W.; Pennycook, S. J.; Idrobo, J.-C.; Pantelides, S. T. Direct Visualization of Reversible Dynamics in a Si6 Cluster Embedded in a Graphene Pore. Nat. Commun. 2013, 4 (April), 1650. (34) Lin, Y.-C.; Dumcenco, D. O.; Huang, Y.-S.; Suenaga, K. Atomic Mechanism of the Semiconducting-to-Metallic Phase Transition in Single-Layered MoS2. Nat. Nanotechnol. 2014, 9 (5), 391−396. (35) Zhou, D.; Sigle, W.; Kelsch, M.; Habermeier, H.-U.; van Aken, P. A. Linking Atomic Structure and Local Chemistry at ManganeseSegregated Antiphase Boundaries in ZrO2 -La2/3Sr1/3MnO3 Thin Films. Adv. Mater. Interfaces 2015, 2 (15), 1500377.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Present Address ⊥

D.Z.: Department of Materials Science and Engineering, University of Wisconsin-Madison, 1509 University Avenue, 53706, Madison, Wisconsin, United States. E-mail: dzhou36@ wisc.edu.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work has been initiated by Prof. J. Zhang (Department of Physics, Shanghai University). Samples have been prepared by Dr. Y. Gao during his PhD work at Max Planck Institute for Solid State Research (MPI-FKF). Careful proofreading of the manuscript by Dr. A. Mark (MPI-FKF) is gratefully acknowledged. The research leading to these results has received funding from the European Union Seventh Framework Program [FP/ 2007/2013] under grant agreement no. 312483 (ESTEEM2).



REFERENCES

(1) Pantano, C. G.; Madey, T. E. Electron Beam Damage in Auger Electron Spectroscopy. Appl. Surf. Sci. 1981, 7, 115−141. (2) Kumar, S.; Adams, W. W. Electron Beam Damage in High Temperature Polymers. Polymer 1990, 31 (1), 15−19. (3) Isaacson, M.; Johnson, D.; Crewe, A. V. Electron Beam Excitation and Damage of Biological Molecules; Its Implications for Specimen Damage in Electron Microscopy. Radiat. Res. 1973, 55 (2), 205−224. (4) Mkhoyan, K. A.; Silcox, J. Electron-Beam-Induced Damage in Wurtzite InN. Appl. Phys. Lett. 2003, 82 (6), 859−861. (5) Teweldebrhan, D.; Balandin, A. A. Modification of Graphene Properties due to Electron-Beam Irradiation. Appl. Phys. Lett. 2009, 94, 013101. (6) Makarova, T. L.; Shelankov, A. L.; Serenkov, I. T.; Sakharov, V. I. Magnetism in Graphite Induced by Irradiation of Hydrogen or Helium Ions - A Comparative Study. Phys. Status Solidi B 2010, 247 (11−12), 2988−2991. (7) Fox, D.; Zhou, Y. B.; O’Neill, A.; Kumar, S.; Wang, J. J.; Coleman, J. N.; Duesberg, G. S.; Donegan, J. F.; Zhang, H. Z. Helium Ion Microscopy of Graphene: Beam Damage, Image Quality and Edge Contrast. Nanotechnology 2013, 24 (33), 335702. (8) Livengood, R.; Tan, S.; Greenzweig, Y.; Notte, J.; McVey, S. Subsurface Damage from Helium Ions as a Function of Dose, Beam Energy, and Dose Rate. J. Vac. Sci. Technol. B Microelectron. Nanom. Struct. 2009, 27 (6), 3244. (9) Dutta, N. J.; Buzarbaruah, N.; Mohanty, S. R. Damage Studies on Tungsten due to Helium Ion Irradiation. J. Nucl. Mater. 2014, 452 (1− 3), 51−56. (10) Andrzejczuk, M.; Plocinski, T.; Zielinski, W.; Kurzydlowski, K. J. TEM Characterization of the Artefacts Induced by FIB in Austenitic Stainless Steel. J. Microsc. 2010, 237 (3), 439−442. (11) Rubanov, S.; Munroe, P. R. FIB-Induced Damage in Silicon. J. Microsc. 2004, 214, 213−221. (12) Tang, L. J.; Zhang, Y. J.; Bosman, M.; Woo, J. Study of Ion Beam Damage on FIB Prepared TEM Samples. In Proceedings of the International Symposium on the Physical and Failure Analysis of Integrated Circuits; IPFA: London, 2010. (13) Egerton, R. F.; Li, P.; Malac, M. Radiation Damage in the TEM and SEM. Micron 2004, 35 (6), 399−409. (14) Krasheninnikov, a V; Banhart, F. Engineering of Nanostructured Carbon Materials with Electron or Ion Beams. Nat. Mater. 2007, 6 (10), 723−733. (15) Song, B.; Schneider, G. F.; Xu, Q.; Pandraud, G.; Dekker, C.; Zandbergen, H. Atomic-Scale Electron-Beam Sculpting of near-DefectFree Graphene Nanostructures. Nano Lett. 2011, 11 (6), 2247−2250. 24184

DOI: 10.1021/acsami.6b06621 ACS Appl. Mater. Interfaces 2016, 8, 24177−24185

Research Article

ACS Applied Materials & Interfaces (36) Habermeier, H.-U.; van Aken, P. A. Manganese Segregation at Antiphase Boundaries Connecting ZrO2 Pillars in ZrO2−La2/3Sr1/3MnO3 Pillar−Matrix Structures. Microsc. Microanal. 2015, 21 (Supplementary 3), 2067−2068. (37) De Groot, F. M. F.; Grioni, M.; Fuggle, J. C.; Ghijsen, J.; Sawatzky, G. A.; Petersen, H. Oxygen 1s X-Ray-Absorption Edges of TransitionMetal Oxides. Phys. Rev. B: Condens. Matter Mater. Phys. 1989, 40 (8), 5715−5723. (38) Lafuerza, S.; Subías, G.; García, J.; Di Matteo, S.; Blasco, J.; Cuartero, V.; Natoli, C. R. Origin of the Pre-Peak Features in the Oxygen K-Edge X-Ray Absorption Spectra of LaFeO3 and LaMnO3 Studied by Ga Substitution of the Transition Metal Ion. J. Phys.: Condens. Matter 2011, 23 (32), 325601. (39) Marris, H.; Deboudt, K.; Flament, P.; Grobety, B.; Giere, R. Fe and Mn Oxidation States by TEM-EELS in Fine-Particle Emissions from a Fe-Mn Alloy Making Plant. Environ. Sci. Technol. 2013, 47 (19), 10832−10840. (40) Tan, H.; Verbeeck, J.; Abakumov, A.; Van Tendeloo, G. Oxidation State and Chemical Shift Investigation in Transition Metal Oxides by EELS. Ultramicroscopy 2012, 116, 24−33. (41) Wang, Z. L.; Yin, J. S.; Jiang, Y. D. EELS Analysis of Cation Valence States and Oxygen Vacancies in Magnetic Oxides. Micron 2000, 31 (5), 571−580. (42) Wang, Y. Q.; Maclaren, I.; Duan, X. F. EELS Analysis of Manganese Valence States in Rare-Earth Manganites (La1‑xYx)0.5MnO3. Mater. Sci. Eng., A 2001, 318 (1−2), 259−263. (43) Livi, K. J. T.; Lafferty, B.; Zhu, M.; Zhang, S.; Gaillot, A. C.; Sparks, D. L. Electron Energy-Loss Safe-Dose Limits for Manganese Valence Measurements in Environmentally Relevant Manganese Oxides. Environ. Sci. Technol. 2012, 46 (2), 970−976. (44) Garvie, L. A. J.; Craven, A. J. Electron-Beam-Induced Reduction of Mn4+ in Manganese Oxides as Revealed by Parallel EELS. Ultramicroscopy 1994, 54 (1), 83−92. (45) McCartney, M.; Crozier, P.; Weiss, J.; Smith, D. J. Electron-BeamInduced Reactions at Transition-Metal Oxide Surfaces. Vacuum 1991, 42 (4), 301−308. (46) Knotek, M. L.; Feibelman, P. J. Stability of Ionically Bonded Surfaces in Ionizing Environments. Surf. Sci. 1979, 90 (1), 78−90. (47) Gazquez, J.; Bose, S.; Sharma, M.; Torija, M. A.; Pennycook, S. J.; Leighton, C.; Varela, M. Lattice Mismatch Accommodation via Oxygen Vacancy Ordering in Epitaxial La0.5Sr0.5CoO3‑δ Thin Films. APL Mater. 2013, 1 (1), 012105. (48) Donner, W.; Chen, C.; Liu, M.; Jacobson, A. J.; Lee, Y. L.; Gadre, M.; Morgan, D. Epitaxial Strain-Induced Chemical Ordering in La0.5Sr 0.5CoO3‑δ Films on SrTiO3. Chem. Mater. 2011, 23 (4), 984−988. (49) Klenov, D. O.; Stemmer, S. Contributions to the Contrast in Experimental High-Angle Annular Dark-Field Images. Ultramicroscopy 2006, 106 (10), 889−901. (50) Gao, Y.; Cao, G.; Zhang, J.; Habermeier, H. U. Intrinsic and Precipitate-Induced Quantum Corrections to Conductivity in La2/3Sr1/3MnO3 Thin Films. Phys. Rev. B: Condens. Matter Mater. Phys. 2012, 85 (19), 1−8.

24185

DOI: 10.1021/acsami.6b06621 ACS Appl. Mater. Interfaces 2016, 8, 24177−24185