Infrared ellipsometry study of photogenerated charge carriers at the (001) and (110) surfaces of $\mathrm{SrTi}{\mathrm{O}}_{3}$ crystals and at the interface of the corresponding $\mathrm{LaAl}{\mathrm{O}}_{3}/\mathrm{SrTi}{\mathrm{O}}_{3}$ heterostructures

Infrared ellipsometry study of photogenerated charge carriers at the (001) and (110) surfaces of $SrTi O_{3}$ crystals and at the interface of the corresponding $LaAl O_{3} / SrTi O_{3}$ heterostructures

M. Yazdi-Rizi, P. Marsik, B. P. P. Mallett, K. Sen, A. Cerreta, A. Dubroka, M. Scigaj, F. Sánchez, G. Herranz, and C. Bernhard

Phys. Rev. B 95, 195107 – Published 3 May 2017

Abstract

With infrared (IR) ellipsometry and dc resistance measurements, we investigated the photodoping at the (001) and (110) surfaces of $SrTi O_{3}$ (STO) single crystals and at the corresponding interfaces of $LaAl O_{3} / SrTi O_{3}$ (LAO/STO) heterostructures. In the bare STO crystals, we find that the photogenerated charge carriers, which accumulate near the (001) surface, have a similar depth profile and sheet carrier concentration as the confined electrons that were previously observed in LAO/STO (001) heterostructures. A large fraction of these photogenerated charge carriers persist at low temperature at the STO (001) surface even after the ultraviolet light has been switched off again. These persistent charge carriers seem to originate from oxygen vacancies that are trapped at the structural domain boundaries, which develop below the so-called antiferrodistortive transition at $T^{*} = 105 K$ . This is most evident from a corresponding photodoping study of the dc transport in STO (110) crystals for which the concentration of these domain boundaries can be modified by applying a weak uniaxial stress. The oxygen vacancies and their trapping by defects are also the source of the electrons that are confined to the interface of LAO/STO (110) heterostructures, which likely do not have a polar discontinuity as in LAO/STO (001). In the former, the trapping and clustering of the oxygen vacancies also has a strong influence on the anisotropy of the charge carrier mobility. We show that this anisotropy can be readily varied and even inverted by various means, such as a gentle thermal treatment, UV irradiation, or even a weak uniaxial stress. Our experiments suggest that extended defects, which develop over long time periods (of weeks to months), can strongly influence the response of the confined charge carriers at the LAO/STO (110) interface.

Received 7 February 2017
Revised 26 March 2017

DOI:https://doi.org/10.1103/PhysRevB.95.195107

Physics Subject Headings (PhySH)

Dielectric properties Domains Electronic structure Optical & microwave phenomena Phase transitions Photoconductivity Structural properties Surface & interfacial phenomena Transition temperature Transport phenomena

Condensed Matter, Materials & Applied Physics

Authors & Affiliations

M. Yazdi-Rizi^1,*, P. Marsik¹, B. P. P. Mallett^1,2, K. Sen¹, A. Cerreta¹, A. Dubroka³, M. Scigaj⁴, F. Sánchez⁴, G. Herranz⁴, and C. Bernhard^1,†

¹Physics Department and Fribourg Center for Nanomaterials (FriMat), University of Fribourg, Chemin du Musée 3, CH-1700 Fribourg, Switzerland
²Robinson Research Institute, Victoria University, P.O. Box 600, Wellington, New Zealand
³Department of Condensed Matter Physics, Faculty of Science and Central European Institute of Technology, Masaryk University, Kotlářská 2, 61137 Brno, Czech Republic
⁴Institut de Ciència de Materials de Barcelona (ICMAB-CSIC), Campus de la Universitat Autònoma de Barcelona, Bellaterra 08193, Catalonia, Spain

^*meghdad.yazdi@unifr.ch
^†christian.bernhard@unifr.ch

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 95, Iss. 19 — 15 May 2017

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
DC resistance and IR spectroscopy study of the photo-induced charge carriers at the surface of a STO (001) single crystal. (a) Evolution of the dc resistance during and after the illumination with UV light. Shown by open symbols are the periods during which the ellipsometry spectra in (b)–(e) have been recorded. (b)–(e) Difference spectrum of the ellipsometric angle, Ψ, as measured at different times during and after the UV illumination [at $t > 0$ , as indicated by open symbols in (a)] and in the initial, dark state at $t < 0$ , i.e., $Δ Ψ = Ψ (t > 0) - Ψ (t < 0)$ . The solid lines show the best fits with the model of a conducting surface layer with a graded depth profile of the charge carriers. (f) Depth profiles of the charge carrier concentration obtained from the best fits in (b)–(e). The other fit parameters are listed in Table 1.
Reuse & Permissions
Figure 2
Difference spectra of the ellipsometric angle, $Δ Ψ = Ψ (t ≫ 0, T) - Ψ (t < 0, T)$ , in the vicinity of the highest LO mode of a STO (001) single crystal after and before UV illumination at $T = 10 K$ , as measured at different temperatures upon warming. The inset shows the temperature dependence of the magnitude of the dip around $865 c m^{- 1}$ , which is an indicator for the sheet carrier density, $N_{s}$ .
Reuse & Permissions
Figure 3
Photodoping effect at 30 K on the dc resistance of a STO (110) single crystal in the weakly twinned pristine state (red dashed line) in a monodomain state with a weak stress applied along the [1-10] direction (blue solid line) and in a polydomain state with the stress along [001] (green line with open squares). Also shown for comparison is the corresponding curve of a heavily twinned STO (001) crystal (cyan line with cross symbols). (a) Time dependence of the resistance during and right after the UV illumination that was started at $t^{start} = 0$ and stopped at $t^{stop} = 101$ minutes, as indicated by the vertical arrows. (b) Magnification of the time dependence right after the UV radiation has been switched off.
Reuse & Permissions
Figure 4
Difference spectra of the ellipsometric angle, $Δ Ψ = Ψ (LAO / STO) - Ψ (STO)$ , of a LAO/STO (001) heterostructure with respect to a bare STO (001) substrate (without photodoping) showing the Berreman mode in the vicinity of the highest LO phonon of STO and its variation upon photodoping. (a) Comparison of the spectra at 30 K in the initial, dark state (red solid line), during UV illumination (blue line with solid circles), shortly after the UV radiation has been switched off (cyan line with solid triangles) and after the sample has been warmed to room temperature and cooled again (in the dark) to 30 K (green dashed line). (b) Experimental data (red circles) and best fit (red line) to the data in the initial dark state at 30 K prior to the UV illumination. The inset shows the obtained depth profile of the carrier density. (c) Data (blue symbols) and best fit (blue line) of the spectrum obtained during UV illumination. (d) Data (cyan symbols) and best fit (cyan line) of the spectrum as obtained at 30 K shortly after the end of the UV illumination. The parameters obtained from the best fits in (b)–(d) are listed in Table 2.
Reuse & Permissions
Figure 5
Difference spectra of the ellipsometric angle, $Δ Ψ = Ψ (LAO / STO) - Ψ (STO)$ , of LAO/STO (110) with respect to a bare STO (110) substrate (without photodoping) showing the Berreman mode in the vicinity of the highest LO phonon of STO and its variation upon UV illumination and thermal annealing. (a) Spectra in the pristine state prior to any UV illumination with the plane of incidence of the IR light along the [001] (red circles) and [1-10] direction (blue triangles). The best fits obtained with a model that allows for an in-plane anisotropy of the charge carrier mobility are shown by the corresponding solid lines. Inset: Spectra of the optical conductivity at the position of the so-called $R$ mode, which has a much larger oscillator strength along [1-10] than along [001]. (b) Corresponding spectra [as in (a)] after the sample has been subject to a moderate thermal treatment for 1 hour at $120^{\circ} C$ and $10^{- 6} mbar$ . The strong anisotropy of the Berreman mode is now inverted as compared to the pristine state in (a). (c) Spectra obtained shortly after a UV illumination (for 45 minutes) measured along the [001] (red circles) and [1-10] directions (blue triangles). Shown by the green diamond symbols is a spectrum after heating to 300 K and cooling again (in dark condition) to 10 K. (d) Comparison of the experimental data (blue triangles) and the best fit (blue solid line) of the spectrum along [1-10] shortly after the UV illumination. (e) Ellipsometric spectra showing the absence of a Berreman mode after the sample has been annealed for 2 hours at $450^{\circ} C$ in flowing oxygen gas atmosphere. (f) Ellipsometric spectra of the Berreman mode that has reoccurred about 1 year after the annealing treatment in (e). The parameters obtained from the best fits in (a), (b), (d), and (f) are listed in Table 3.
Reuse & Permissions
Figure 6
Effect of an external stress on the difference spectra of the ellipsometric angle, $Δ Ψ = Ψ (LAO / STO) - Ψ (STO)$ , of a LAO/STO (110) heterostructure with respect to a STO (110) substrate. The measurements have been performed after the ones shown in Fig. 5. A weak uniaxial stress of about 2.3 MPa was applied along either the [1-10] or [001] direction by mounting the sample in a mechanical clamp and cooling it to 10 K. (a) Spectra showing the anisotropic response along the [1-10] and [001] directions with the stress applied along [1-10] (blue dashed line and red solid line) or [001] (magenta line with open triangles and cyan dashed line with open diamonds). (b) Corresponding spectra of the $R$ mode, which show the anisotropy of the oscillator strength along [1-10] and [001] from which the preferred orientation of the structural domains has been deduced. (c) Data (open symbols) and best fits (solid lines) for the $Δ Ψ$ spectra at 10 K after cooling with uniaxial stress along [1-10]. (d) Corresponding data (symbols) and best fits (solid lines) obtained after warming the sample to 120 K. (e) Data (open symbols) and best fits (solid lines) for the $Δ Ψ$ spectra at 10 K after cooling with the uniaxial stress along [001]. (f) Corresponding data (symbols) and best fits (solid lines) obtained after warming the sample to 120 K. The parameters obtained from the best fits in (c)–(f) are listed in Table 4.
Reuse & Permissions

Physical Review B

covering condensed matter and materials physics

Infrared ellipsometry study of photogenerated charge carriers at the (001) and (110) surfaces of $SrTi O_{3}$ crystals and at the interface of the corresponding $LaAl O_{3} / SrTi O_{3}$ heterostructures

M. Yazdi-Rizi, P. Marsik, B. P. P. Mallett, K. Sen, A. Cerreta, A. Dubroka, M. Scigaj, F. Sánchez, G. Herranz, and C. Bernhard

Phys. Rev. B 95, 195107 – Published 3 May 2017

Abstract

Physics Subject Headings (PhySH)

Authors & Affiliations

Article Text (Subscription Required)

References (Subscription Required)

Issue

Access Options

Physical Review B

covering condensed matter and materials physics

Infrared ellipsometry study of photogenerated charge carriers at the (001) and (110) surfaces of SrTiO3 crystals and at the interface of the corresponding LaAlO3/SrTiO3 heterostructures

M. Yazdi-Rizi, P. Marsik, B. P. P. Mallett, K. Sen, A. Cerreta, A. Dubroka, M. Scigaj, F. Sánchez, G. Herranz, and C. Bernhard

Phys. Rev. B 95, 195107 – Published 3 May 2017

Abstract

Physics Subject Headings (PhySH)

Authors & Affiliations

Article Text (Subscription Required)

References (Subscription Required)

Issue

Access Options

Authorization Required

Other Options

Download & Share

Images

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Log In

Search

Article Lookup

Paste a citation or DOI

Enter a citation

Infrared ellipsometry study of photogenerated charge carriers at the (001) and (110) surfaces of $SrTi O_{3}$ crystals and at the interface of the corresponding $LaAl O_{3} / SrTi O_{3}$ heterostructures