Tuning Oxygen Vacancy Diffusion through Strain in SrTiO3 Thin Films

Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES

Functional Inorganic Materials and Devices 3

Tuning oxygen vacancy diffusion through strain in SrTiO thin films Lucia Iglesias, Andres Gomez, Martí Gich, and Francisco Rivadulla ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b12019 • Publication Date (Web): 25 Sep 2018 Downloaded from http://pubs.acs.org on September 28, 2018

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 20 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 Applied Materials & Interfaces

Tuning Oxygen Vacancy Diffusion Through Strain in SrTiO3 Thin Films Lucia Iglesias,† Andrés Gómez,∗,‡ Mart´ı Gich,‡ and Francisco Rivadulla∗,† †Centro Singular de Investigación en Qu´ımica Biolóxica e Materiais Moleculares (CIQUS), Departamento de Qu´ımica-F´ısica, Universidade de Santiago de Compostela, 15782, Santiago de Compostela, Spain ‡Institut de Ciència de Materials de Barcelona (ICMAB-CSIC), Campus UAB, Bellaterra, Catalonia, 08193, Spain E-mail: [email protected]; [email protected]

Abstract Understanding the diffusion of oxygen vacancies in oxides under different external stimuli is crucial for the design of ion-based electronic devices, improve catalytic performance, etc. In this manuscript, using an external electric field produced by an AFM tip, we obtain the room-temperature diffusion coefficient of oxygen-vacancies in thin-films of SrTiO3 under compressive/tensile epitaxial strain. Tensile strain produces a substantial increase of the diffusion coefficient, facilitating the mobility of vacancies through the film. Additionally, the effect of tip bias, pulse time, and temperature on the local concentration of vacancies is investigated. These are important parameters of control in the production and stabilization of non-volatile states in ion-based devices. Our findings show the key role played by strain for the control of oxygen vacancy migration in thin-film oxides.

1

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 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

Keywords Strain engineering, oxygen vacancies, diffusion coefficient, resistive switching, strontium titanate, Electrostatic force microscopy, Kelvin probe force microscopy

1

Introduction

Strong electronic correlations in narrow d-bands, along with a large coupling between charge, spin and orbital degrees of freedom, are responsible for the variety of electronic and magnetic behaviors characteristic of transition metal (TM) oxides (i.e. metal-insulator transitions, superconductivity, ferroelectricity, multiferroicity, etc.). 1–3 Of particular interest within them are the oxoperovskites, given the enormous flexibility of this structure to accept chemical substitution at the cationic sites. However, the possibility to control the occupancy of the anion site in an oxide is equally interesting, although much less explored. In this regard, oxygen vacancies behave in many 3d-oxoperovskites as mobile, e-donor defects, able to induce profound changes in the structural, electronic and magnetic properties of the material. 4–6 Therefore, the possibility to control at will not only the creation of these vacancies, but also their distribution and movement inside the material using different external stimuli, is key for the design of ionic-based devices. 7 Strontium titanate, SrTiO3 (STO) is a paradigmatic example in this regard. This oxoperovskite is a quantum paraelectric insulator, 8 whose great chemical stability, diamagnetism, and high dielectric constant makes it the preferred template for the growth of thin-films of other TM oxoperovskites. Incorporation of oxygen defects may be detrimental for some applications of this oxide; for instance, high-k requirements of dielectric insulating layers for carriers density modulation can be compromised. 9,10 However, they can be also used to add new functionalities to STO, like for instance gas sensing capabilities, 11 resistive switching, 12–17 ferroelectricity, 18,19 or low temperature superconductivity. 4 In particular, resistive (switching) random-access memories (ReRAM) with controlled 2


Page 2 of 20

Page 3 of 20 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


multi-stable states are attracting a considerable attention in recent years. 20–22 The distribution of VO¨ in these devices can be modified by electrical or mechanical methods, resulting in reversible switching of their electrical resistance. Recently Kalinin et al. 23 discussed the electrical and mechanical manipulation of the local VO¨ distribution in unstrained STO thin films using an AFM tip. In a similar approach, Sharma et al. 24 were able to modulate the electrical conductivity of the 2D electron gas at the LaAlO3 /SrTiO3 interface, opening a door towards mechanically operated transistors. Here we report the effect of epitaxial strain in the diffusion coefficient, D, of VO¨ at room temperature. We followed the time dependence of the local surface potential, after the distribution of VO¨ is locally modified by an electric field produced by an AFM tip. Fitting of the experimental data to Fick’s second law of diffusion demonstrates that tensile strain enhances D considerably, while moderate compression does not show an appreciable effect with respect to unstrained films. We also studied the effect of the tip bias, pulse time, and temperature in the distribution of vacancies. These factors determine the control over VO¨ redistribution, the writing and retention times of the devices and constitute important factors for the applicability of ionic-based memresistive devices.

2

Experimental Details

SrTiO3 thin-films (with ≈ 2% Nb 5+ to compensate the presence of unintentional Sr2+ e-trap vacancies) 25,26 were grown by Pulsed Laser Deposition (PLD) (KrF, λ=248 nm and Nd:YAG λ=266 nm). The thin films were deposited on top of different substrates with pseudocubic lattice parameters ranging from 3.868˚ A of (LaAlO3 )0,3 -(Sr2 AlTaO6 )0,7 (LSAT) to 3.989˚ A of KTaO3 (KTO), in order to apply an epitaxial stress from -0.95% to +2.15%. All samples used in this work were grown in the same batch, at 800o C and 100 mTorr of oxygen pressure, with a thickness ≈17 nm. The thin films were characterized combining Electrostatic Force Microscopy (EFM), and

3



Kelvin Probe Force Microscopy (KPFM). The experiments were performed independently using two different AFM setups, a Park Systems NX10 and a Keysight 5500 AFM under both, ambient and low humidity conditions (6-7% RH), in order to suppress possible effects derived from charge injection, 27 formation of protons/hydroxyl radicals, 28 or electrochemical processes. 29 A Pt/Ir-coated metallic tip with force constant of 3 N/m was used for electrical scans in KPFM and EFM modes. The first resonance frequency of the lever was used to map the topography while the KPFM feedback was fed with the Xcomponent signal, and an excitation of 20kHz and 2VAC was applied to the tip. While KPFM is performed in single scan mode using tapping mode to acquire the topography, we performed the recording in contact mode to ensure a continuous electrical contact between the tip and the sample. We employed single-pass out of resonance KPFM, which fastened our data acquisition while avoiding crosstalk between first and second resonance mode. The feedback of the KPFM is proportional to the Contact Potential Difference, VCPD, between the conducting cantilever and the sample surface. Such VCPD can be used to map the surface potential (SP) of the samples, which is known to be related to differences in charge density distribution. 30 In the paper we followed the time dependence of the surface potential in regions previously enriched/depleted of oxygen vacancies, as a direct indication of their redistribution in the films at room temperature. This methodology allows to measure the room temperature diffusion coefficient of these ionic species, vide infra.

3

Results and Discussion

The structural properties of the films were studied by X-Ray Diffraction (XRD), and are summarized in Figure 1. The films are perfectly c-axis oriented, showing Laue oscillations around the (002) diffraction peak. This demonstrates that the films maintain a very good structural quality, irrespectively from the sign and magnitude of the lattice mismatch with the substrate.

4


Page 4 of 20

Page 5 of 20 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


On the other hand, to probe the degree of epitaxial strain induced by the substrate on the SrTiO3 films, high resolution X-Ray Reciprocal Space Maps (RSM) were performed around the (103) pseudocubic reflection of the film and substrate. As shown in Figure 1b), there is a perfect matching between the Qx of the film and substrate, independently of the degree and sign of strain. Therefore, the films are fully strained to the in-plane lattice parameter of the substrate. The epitaxial compressive (tensile) strain induced by the substrate changes the in-plane lattice parameter of the films (from (af -ab )×100/ab =-0.95% on LSAT, to +2.15% on KTO), and causes an expansion (contraction) along the out-of-plane lattice parameter. This deformation of the unit cell can be calculated assuming an elastic deformation given by the Poisson’s ratio of ν=0.2332 31 (see Fig. S1 in the supporting information). The evolution of unit cell volume follows the prediction of ν, therefore confirming that there are no important deviations in cation stoichiometry, which could compromise the analysis of the results. Moreover, the surfaces of the films are very flat, suggesting smooth layer-by-layer growth and mixed TiO2 -SrO termination, irrespective of the degree of strain (Figure S3 in the supporting information).

5



Nb:STO film

b

GSO (002)pc

b

KTaO3 3.989 Å 2.15%

Nb:STO film

KTO(002)

a

GSO(332)o

Nb:STO film

DSO(002)pc

KTO(103)pc

Nb:STO(103)pc GdScO3 3.970 Å 1.66%

Intensity (a.u.)

Nb:STO(103)pc

Nb:STO(103)pc DSO(332)o

42

44

46

STO(002)

STO(103)pc STO 3.905 Å 0%

LSAT(002)

Nb:STO film

Nb:STO film

DyScO3 3.950 Å 1.15%

Qz(Å-1)

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 6 of 20

2θ(°)

Nb:STO(103)pc

LSAT(103)pc

LSAT 3.868 Å -0.95%

48

50

Nb:STO(103)pc

52 Qx(Å-1)

Figure 1: a) XRD patterns around the (002) pseudo-cubic reflection. The lattice parameter of the different substrates, as well as the theoretical lattice mismatch with the film, is shown in each panel. b) High-resolution RSM around (103)pc asymmetric reflection in the case of LSAT, STO and KTO substrates, and (332)o reflection for DSO and GSO substrates.

In order to quantify the effect of epitaxial strain on the drift mobility of VO¨ , a region on the surface of each sample was first poled with negative/positive AFM tip bias in contact mode, and then the same area was mapped by KPFM scanning at zero DC bias. An scheme of this process is depicted in Figure 2a. The highly localized electric field (∼6MV/m) generated by the AFM bias, ineherent to the nanometer scale radius of the tip, is sufficient to modify the initial homogeneous distribution of positively charged VO¨ across the film. Depending on the electric field direction, the accumulation or depletion of VO¨ produces a localized increment or decrease on the SP of the sample. 32 6


Page 7 of 20

a

+15Vdc

𝑑𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒

-15Vdc

𝑉Ö

LSAT (-0.95%)

𝑉Ö𝑟𝑖𝑖𝑖

DSO (+1.15%)

STO (0%)

GSO (+1.66%)

c 40

∆SP(%)

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


GSO

LSAT

20 DSO

STO

0 -1.0

-0.5

0.0

0.5 Strain(%)

1.0

1.5

2.0

Figure 2: a) Sketch of the KPFM method to record the sample surface with positive and negative tip bias. Positive tip bias produces a deficient VO¨ concentration, while a negative voltage leads to an enrichment of the local concentration of VO¨ . b) Characterization of the surface potential for the thin films grown on different substrates. c) Evolution of ∆SP as a function of epitaxial strain. ∆SP is defined as the difference between the minimum/maximum in the surface potential signal of the VO¨ enriched/depleted regions.

A negative voltage causes an accumulation of vacancies, which contribute with free electrons to the sample surface, producing a local decrease of the work function and therefore of the surface potential. On the contrary, a positive tip bias drives VO¨ away from the sample surface increasing SP. Significant differences in the background surface potential of the unpolarized regions are evident, depending on the magnitude of the strain. This is related to a slight different initial concentration of VO¨ in the as-grown films, and therefore, to the different initial conductivity of the samples. 33 The effect of the small amount of Nb doping used in this work

7



on the conductivity of the films is negligible (see discussion of Figure S2 in the supporting information). To minimize artifacts and offsets from the measurement conditions, all measurements were performed with the exact same probe, AC bias magnitude, frequency, tip-sample distance, and scanning/recording rate for both KPFM and topography acquisition. Following this procedure, we are able to compare how the relative change of the surface potential between VO¨ enriched/depleted regions, ∆SP, changes with strain (Figure 2 b). ∆SP increases up to ≈40% for the most stressed films, either compressive (on LSAT) or tensile (on GSO) (Figure 2 c)), with respect to the unstrained film (on STO). Therefore, epitaxial strain influences very much the drift mobility and the subsequent accumulation/depletion of VO¨ in STO. In order to determine the diffusion coefficient of the oxygen vacancies, first an area of the sample is poled with an electric field of ±10 V. Then, the electric field is set at 0 V, and the time dependence of the surface potential (by monitoring the EFM and KPFM signal) is followed for one day, every 30 min. Due to the high sensitivity of KPFM to the surface potential, the experiments were performed under ambient conditions and in low humidity (6-7% RH). The results were comparable, therefore showing the negligible effect of adsorbed species on the surface potential. The results are shown in Figure 3a-d). In the scheme of Figure 3e) we represent the whole process: charged VO¨ are attracted/repealed by the electric field, increasing/decreasing the surface potential with respect to the initial equilibrium distribution. Removing the field allows the charge to diffuse freely, blurring the effect of the charge accumulation/depletion on the surface potential. Apart from diffusion into the film bulk, the effect of surface mobility must be considered. The border of the poled regions remain well defined with time, without too much of lateral broadening, which suggests a limited in-plane diffusion of VO¨ . The KPFM profiles were fitted to a Gaussian function, and the full-width at half-maximum (FWHM) extracted from

8


Page 8 of 20

Page 9 of 20

a

b

t=0h

+10V

t=2h

c

t=4h

-10V

d Amplitude (mV)

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


e 0h 2h 4h 8h 12h

0.5

0.0

-0.5 0

5 10 Distance (m)

15

Figure 3: Evolution of the EFM amplitude over time at zero DC bias. a) Immediately after removing the AFM tip bias, b) after 2 hours, and c) after 4 hours. d) Profile of the EFM amplitude recorded over 12 hours. The results correspond to a film of STO on GSO. The effect of tip bias in the surface potential is completely reversible, independently of the sign and magnitude of strain (Figure S5 of Supplementary Information). e) Scheme of the diffusion of oxygen vacancies after removal of the electric field.

these fittings at different times (see Figure S4 in the supporting information). A comparison of the FWHM and the voltage amplitude in the central part of the pooled region shows that the effect of lateral diffusion is only a small contribution to the signal. Therefore, we conclude that the main contribution to the KPFM signal is the correlated diffusion of VO¨ in the perpendicular direction to the film plane, following the behavior predicted by second Fick’s law to attain an equilibrium distribution: ∂[VO¨ ] ∂ 2 [V ¨ ] = D 2 O2 ∂t ∂ z

(1)

The absence of significant lateral diffusion, along with the small thickness of the films, suggests that we can assume a predominant diffusion along the out-of-plane direction, z (see Figure 3e). In this case, the diffusion equation can be solved in the limit of t→ ∞(see the Supplementary Information for a complete deduction of this expression): 34,35 9



∆VO¨ (t) =

8 −π22Dt e l π2

Page 10 of 20

(2)

In this expression D is the diffusion coefficient and l is the film thickness (l =17nm for all films used in this work). The fitting of the experimental data to equation (2) is shown in Figure 4 a) for two representative cases. The value of Dv2x10−17 cm2 /s for the unstrained film is in very good agreement with the value reported in bulk STO at room temperature. 23,36 In contrast, 2% of tensile strain increases D by an impressive ≈350 % with respect to the unstrained films (Figure 4 b). On the other hand, compressive strain does not show any appreciable effect on the difussion coefficient of oxygen vacancies in STO. Note that in Figure 2 we observed a similar increase in the surface potential under compressive/tensile strain. In that case we are probing the influence of an electric field (drift mobility), while in Figure 4 b) we are probing the free diffusion VO¨ . Other authors suggested that the incorporation of atmospheric oxygen to the surface could be an important contribution to the equilibration of the surface potential. 37 In our case, the diffusion coefficient of VO¨ calculated from the initially depleted/accumulated surface region shows similar results. Oxidizing the initially reduced sample surface will require adsorption of oxygen from the atmosphere, dissociation, and diffusion through the lattice. On the other hand, reducing an initially oxidized region will require diffusion of the vacancies towards the surface, recombination (probably forming peroxide-ions) and formation of molecular oxygen. These processes are expected to show different activation energies. Therefore, we can discard oxygen exchange through the surface as the main mechanism to explain the time dependent surface voltage in our case (see also the supporting information for the effect of mechanical force on the surface potential; Figure S6). On the other hand, the results shown in Figure 4b) are fully consistent with DFT calculations which predicted a continuous increase of the VO¨ diffusion with tensile strain, 38 while a compressive strain larger than ≈-4% is necessary to decrease the energy barrier for diffusion. 39 In fact, these calculations predicted a significant role of complex TiO6 rotations 10


Page 11 of 20 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


in the anisotropic oxygen diffusion in STO. We recently observed that tensile strain induces an out-of-phase rotation of the TiO6 octahedra along the c-axis, while compressive strain suppresses this rotation completely. 26 Therefore, not only the magnitude of stress, but the changes in the internal rotations of TiO6 octahedra could be playing a significant role in oxygen diffusion in STO thin-films.

Figure 4: a) Time decay of the surface potential due to VO¨ diffusion in STO grown on GSO and LSAT. Lines are fitting to equation (2). b) Strain dependence of the diffusion coefficient of VO¨ . The dashed line is a guide to the eye.

The dramatic dependence of D with strain has important implications for the design of ionic-conducting devices, like resistive switching memories: tensile strain would favour the writing process in such devices, but would also reduce the retention state due to the increase in the VO¨ mobility. Therefore, ion-based devices of this type will be adversely affected by tensile strain. To study the effect of diffusion and the possibility to define multiple stable neighboring states, we recorded side-by-side areas with different voltages, under compressive (LSAT, figure 5 a)) and tensile stress (GSO, figure 5 b)). The results demonstrate clear contrast boundaries between neighboring areas poled with ± voltages, with excellent retention between them. However, due to the effect of strain on D, the signal decays ≈50% after 3 hours on the sample grown on LSAT, and ≈75% in GSO (Figure 5 c)). This demonstrates again the negative effect of tensile strain in the stabilization and retention of the multipotential

11



surface states. a

LSAT

LSAT

t=0h

t=3h

GSO

GSO

t=0h

t=3h

-5V +5V -10V +10V

b -5V +5V -10V +10V

c

1.0 EFM Amplitude (a.u.)

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 12 of 20

LSAT

0.5 GSO

0.0

0

1

2 t (h)

3

4

Figure 5: Record of different potential surface states with a variable voltage from -10V to 10V. a) EFM amplitude on LSAT immediately after the record and 3 hours later. b) Same record on the sample deposited on GSO. c) Normalized evolution of the EFM amplitude for both samples during 4.5 hours. After 3 hours the surface potential decays ≈50% from the initial value in LSAT, and ≈75% in GSO. Dashed lines are guides to the eye.

Apart from low writing voltage/time, and high retention, competitive non-volatile ionic devices also require a high thermal stability. We performed successive poling experiments applying a constant pulse time with different tip bias, and viceversa, maintaining the tip in contact with the sample surface. The results for topography and KPFM surface potential at 25o C are shown in Figure 6 a). Note that the impact of VO¨ accumulation is negligible in the topography, irrespective of the tip bias or the pulse time. In Figure 6 b) the profile of the KPFM image evidences the existence of a threshold voltage (v5-6 V) above which 12


Page 13 of 20

there is a notable increment of the local accumulation of VO¨ . Similarly, a threshold pulse time of 500 ms increases significantly the local concentration of VO¨ . The existance of these threshold voltages/times indicates an activation energy for ionized vacancies to detach from lattice defects, most probably cationic vacancies, which can affect the drift mobility beyond

b

Normalized SP (V)

strain. 40

Normalized SP (V)

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


0

-1

-2 0

4.8 V 7.2 V 9.5 V 15 V

200

400 600 Distance (nm)

800

400 600 Distance (nm)

800

0

-1

-2 0

100ms 200ms 400ms 1000ms

200

Figure 6: Influence of tip bias, pulse time and temperature on the VO¨ distribution on GSO. a) Topography (left) and surface potential (right) of four-dot array, which were performed by pausing the tip for 0.4s and varying the tip bias from 4.8 V to 15 V. Dots recorded while keeping the voltage constant at 15 V and varying the pulse time for 0.1 to 1 s (right). b) Resulting profiles of the normalized surface potential for the dots poling at constant time (up) and constant voltage (down), respectively. c) 3D-plot of the surface potential, related to the VO¨ distribution, after different pulse times (+15 V) applied at 25o C and 60o C.

Regarding the temperature dependence of the VO¨ distribution, we compared write/read voltage measurements performed at room temperature (25o C) and at 60o C, using different pulse times with same tip bias of +15V (Figure 6 c)). Poling of the sample and subsequent KPFM scan are separated by 15 minutes. The required pulse time to achieve a measurable 13



change of the surface potential increases substantially with temperature, reflecting the rapid increase of D(T). This result shows the low retention of the multipotential states, which can be erased at moderate temperatures. In summary, we have shown the dramatic effect of epitaxial strain on the diffusion coefficient of VO¨ in STO. Tensile strain enhances the mobility of VO¨ through the films, being therefore detrimental to the stable definition of multi-stable states in ionic devices. Additionally, we have proved that other parameters such as tip bias, pulse time and temperature can also control the VO¨ distribution, offering an alternative tool to write/erase multi-stable states in ion-based devices.

Supporting Information Available The following files are available free of charge. .Additional experimental details about the structural and morphological characterization of thin films is provided in the supporting information. A Complete deduction of the Fick’s second law is also included.

Acknowledgement This work was supported by the Ministerio de Econom´ıa y Competitividad of Spain under project numbers MAT2016-80762-R and SEV-2015-0496 (Severo Ochoa Program Grant), and Xunta de Galicia (Centro Singular de Investigación de Galicia accreditation 2016-2019) and the European Union (European Regional Development Fund – ERDF). L. I. also acknowledges the Ministerio de Econom´ıa y Competitividad of Spain for a FPI grant.

References (1) Goodenough, J. B. Perspective on Engineering Transition-Metal Oxides. Chem. Mater. 2014, 26, 820–829. 14


Page 14 of 20

Page 15 of 20 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


(2) Khomskii, D. I. Transition Metal Compounds; Cambridge University Press, 2014. (3) Goodenough, J. B. Magnetism and the Chemical Bond ; John Wiley and Sons, 1963. (4) Schooley, J. F.; Hosler, W. R.; Cohen, M. L. Superconductivity in Semiconducting SrTiO3 . Phys. Rev. Lett. 1964, 12, 474–475. (5) Hu, L.; Luo, X.; Zhang, K. J.; Tang, X. W.; Zu, L.; Kan, X. C.; Chen, L.; Zhu, X. B.; Song, W. H.; Dai, J. M.; Sun, Y. P. Oxygen Vacancies-Induced Metal-Insulator Transition in La2/3 Sr1/3 VO3 Thin Films: Role of the Oxygen Substrate-to-Film Transfer. Appl. Phys. Lett. 2014, 105, 111607. (6) Zhu, D.; Cao, Q.; Qiao, R.; Zhu, S.; Yang, W.; Xia, W.; Tian, Y.; Liu, G.; Yan, S. Oxygen Vacancies Controlled Multiple Magnetic Phases in Epitaxial Single Crystal Co0.5 (Mg0.55 Zn0.45 )0.5 O1−v Thin Films. Sci. Rep. 2016, 6, 24188. (7) Kalinin, S. V.; Spaldin, N. A. Functional Ion Defects in Transition Metal Oxides. Science 2013, 341, 858–859. (8) Rowley, S. E.; Spalek, L. J.; Smith, R. P.; Dean, M. P. M.; Itoh, M.; Scott, J. F.; Lonzarich, G. G.; Saxena, S. S. Ferroelectric Quantum Criticality. Nat. Phys. 2014, 10, 367–372. (9) Fuchs, D.; Schneider, C. W.; Schneider, R.; Rietschel, H. High Dielectric Constant and Tunability of Epitaxial SrTiO3 Thin Film Capacitors. J. Appl. Phys. 1999, 85, 7362–7369. (10) Yamada, T.; Astafiev, K. F.; Sherman, V. O.; Tagantsev, A. K.; Su, D.; Muralt, P.; Setter, N. Structural and Dielectric Properties of Strain-Controlled Epitaxial SrTiO3 Thin Films by Two-Step Growth Technique. J. Appl. Phys. 2005, 98, 054105. (11) Hodak, S. K.; Supasai, T.; Wisitsoraat, A.; Hodak, J. H. Design of Low Cost Gas Sensor Based on SrTiO3 and BaTiO3 Films. J. Nanosci. Nanotechnol. 2010, 10, 7236–7238. 15



(12) Li, J. K.; Ma, C.; Jin, K. J.; Ge, C.; Gu, L.; He, X.; Zhou, W. J.; Zhang, Q. H.; Lu, H. B.; Yang, G. Z. Temperature-Dependent Resistance Switching in SrTiO3 . Appl. Phys. Lett. 2016, 108, 10–15. (13) Pan, X.; Shuai, Y.; Wu, C.; Luo, W.; Sun, X.; Yuan, Y.; Zhou, S.; Ou, X.; Zhang, W. Resistive Switching Behavior in Single Crystal SrTiO3 Annealed by Laser. Appl. Surf. Sci. 2016, 389, 1104–1107. (14) Janousch, M.; Meijer, G. I.; Staub, U.; Delley, B.; Karg, S. E.; Andreasson, B. P. Role of Oxygen Vacancies in Cr-Doped SrTiO3 for Resistance-Change Memory. Adv. Mater. 2007, 19, 2232–2235. (15) Wu, S.; Ren, L.; Qing, J.; Yu, F.; Yang, K.; Yang, M.; Wang, Y.; Meng, M.; Zhou, W.; Zhou, X.; Li, S. Bipolar Resistance Switching in Transparent ITO/LaAlO3 /SrTiO3 Memristors. ACS Appl. Mater. Interfaces 2014, 6, 8575–8579. (16) Moon, T.; Jung, H. J.; Kim, Y. J.; Park, M. H.; Kim, H. J. Research Update: Diode Performance of the Pt/Al2 O3 /Two-Dimensional Electron Gas/SrTiO3 Structure and its Time-Dependent Resistance Evolution. APL Mater. 2017, 5, 042301. (17) Park, J.; Kwon, D.-H.; Park, H.; Jung, C. U.; Kim, M. Role of Oxygen Vacancies in Resistive Switching in Pt/Nb-doped SrTiO3 . Appl. Phys. Lett. 2014, 105, 183103. (18) 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. Room-Temperature Ferroelectricity in Strained SrTiO3 . Nature 2004, 430, 758–761. (19) Jang, H. W.; Kumar, A.; Denev, S.; Biegalski, M. D.; Maksymovych, P.; Bark, C. W.; Nelson, C. T.; Folkman, C. M.; Baek, S. H.; Balke, N.; Brooks, C. M.; Tenne, D. A.; Schlom, D. G.; Chen, L. Q.; Pan, X. Q.; Kalinin, S. V. , Gopalan, V. , Eom, C. B., Ferroelectricity in Strain-Free SrTiO3 Thin Films. Phys. Rev. Lett. 2010, 104, 197601. 16


Page 16 of 20

Page 17 of 20 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


(20) Waser, R.; Dittmann, R.; Staikov, C.; Szot, K. Redox-Based Resistive Switching Memories Nanoionic Mechanisms, Prospects, and Challenges. Adv. Mater. 2009, 21, 2632– 2663. (21) Waser, R.; Aono, M. Nanoionics-Based Resistive Switching Memories. Nat. Mater. 2007, 6, 833–840. (22) Sawa, A. Resistive Switching in Transition Metal Oxides. Mater. Today 2008, 11, 28– 36. (23) Das, S.; Wang, B.; Cao, Y.; Rae Cho, M.; Jae Shin, Y.; Mo Yang, S.; Wang, L.; Kim, M.; Kalinin, S. V.; Chen, L.-Q.; Noh, T. W. Controlled Manipulation of Oxygen Vacancies Using Nanoscale Flexoelectricity. Nat. Commun. 2017, 8, 615. (24) Sharma, P.; Ryu, S.; Burton, J. D.; Paudel, T. R.; Bark, C. W.; Huang, Z.; Ariando,; Tsymbal, E. Y.; Catalan, G.; Eom, C. B.; Gruverman, A. Mechanical Tuning of LaAlO3 /SrTiO3 Interface Conductivity. Nano Lett. 2015, 15, 3547–3551. (25) Sarantopoulos, A.; Ferreiro-Vila, E.; Pardo, V.; Magén, C.; Aguirre, M. H.; Rivadulla, F. Electronic Degeneracy and Intrinsic Magnetic Properties of Epitaxial Nb : SrTiO3 Thin Films Controlled by Defects. Phys. Rev. Lett. 2015, 115, 166801. (26) Iglesias, L.; Sarantopoulos, A.; Magén, C.; Rivadulla, F. Oxygen Vacancies in Strained SrTiO3 Thin Films: Formation Enthalpy and Manipulation. Phys. Rev. B 2017, 95, 1–7. (27) Villeneuve-Faure, C.; Makasheva, K.; Boudou, L.; Teyssedre, G. Charge Injection in Thin Dielectric Layers by Atomic Force Microscopy: Influence of Geometry and Material Work Function of the AFM Tip on the Injection Process. Nanotechnology 2016, 27, 245702.

17



(28) Rowicka, E.; Kashyn, D.; Reagan, M. a.; Hirano, T.; Paramonov, P. B.; Dolog, I.; Mallik, R. R.; Lyuksyutov, S. F. Influence of Water Condensation on Charge transport and Electric Breakdown Between an Atomic Force Microscope Tip, Polymeric, and Semiconductor CdS Surfaces. Curr. Nanosci. 2008, 4, 166. (29) Collins, L.; Belianinov, A.; Somnath, S.; Balke, N.; Kalinin, S. V.; Jesse, S. Full Data Acquisition in Kelvin Probe Force Microscopy: Mapping Dynamic Electric Phenomena in Real Space. Sci. Rep. 2016, 6, 30557. (30) Melitz, W.; Shen, J.; Kummel, A. C.; Lee, S. Kelvin Probe Force Microscopy and its Application. Surf. Sci. Rep. 2011, 66, 1–27. (31) Ledbetter, H.; Lei, M.; Kim, S. Elastic Constants, Debye Temperatures, and ElectronPhonon Parameters of Superconducting Cuprates and Related Oxides. Phase Transitions 1990, 23, 61–70. (32) Girard, P. Electrostatic Force Microscopy: Principles and some Applications to Semiconductors. Nanotechnology 2001, 12, 485–490. (33) Kubicek, M.; Cai, Z.; Ma, W.; Yildiz, B.; Hutter, H.; Fleig, J. Tensile Lattice Strain Accelerates Oxygen Surface Exchange and Diffusion in La1−x Srx CoO3−d Thin Films. ACS Nano 2013, 7, 3276–3286. (34) Crank, J. The Mathematics of Diffusion; Clarendon Press, 1975. (35) Merkle, R.; Maier, J. How is Oxygen Incorporated into Oxides? A Comprehensive Kinetic Study of a Simple Solid-State Reaction with SrTiO3 as a Model Material. Angew. Chem., Int. Ed. 2008, 47, 3874–3894. (36) De Souza, R. A.; Metlenko, V.; Park, D.; Weirich, T. E. Behavior of Oxygen Vacancies in Single-Crystal SrTiO3 : Equilibrium Distribution and Diffusion Kinetics. Phys. Rev. B 2012, 85, 1–11. 18


Page 18 of 20

Page 19 of 20 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


(37) Andrä, M.; Gunkel, F.; Bäumer, C.; Xu, C.; Dittmann, R.; Waser, R. The Influence of the Local Oxygen Vacancy Concentration on the Piezoresponse of Strontium Titanate Thin Films. Nanoscale 2015, 7, 14351–14357. (38) AL-Hamadany, R.; Goss, J. P.; Briddon, P. R.; Mojarad, S. A.; O’Neill, A. G.; Rayson, M. J. Impact of Tensile Strain on the Oxygen Vacancy Migration in SrTiO3 : Density Functional Theory Calculations. J. Appl. Phys. 2013, 113, 224108. (39) Al-Hamadany, R.; Goss, J. P.; Briddon, P. R.; Mojarad, S. A.; Al-Hadidi, M.; O’Neill, A. G.; Rayson, M. J. Oxygen Vacancy Migration in Compressively Strained SrTiO3 . J. Appl. Phys. 2013, 113, 0–8. (40) Szot, K.; Speier, W.; Bihlmayer, G.; Waser, R. Switching the Electrical Resistance of Individual Dislocations in Single-Crystalline SrTiO3 . Nat. Mater. 2006, 5, 312–320.

19


8


Page 20 of 20

D (10-17cm2/s)

KTO 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

6 GSO

4 DSO

2

STO LSAT

0

-1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 Strain (%)

VÖ

STO thin film ACS Paragon Plus Environment

Strain

Tuning Oxygen Vacancy Diffusion through Strain in SrTiO3 Thin Films

Recommend Documents