The optical control of ferroic properties is a subject of fascination for the scientific community, because it involves the establishment of new paradigms for technology1,2,3,4,5,6,7,8,9. Domains and domain walls are known to have a great impact on the properties of ferroic materials1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24. Progress is currently being made in understanding the behaviour of the ferroelectric domain wall, especially regarding its dynamic control10,11,12,17,19. New research is being conducted to find effective methodologies capable of modulating ferroelectric domain motion for future electronics. However, the practical use of ferroelectric domain wall motion should be both stable and reversible (rewritable) and, in particular, be able to produce a macroscopic response that can be monitored easily12,17. Here, we show that it is possible to achieve a reversible optical change of ferroelectric domains configuration. This effect leads to the tuning of macroscopic polarization and its related properties by means of polarized light, a non-contact external control. Although this is only the first step, it nevertheless constitutes the most crucial one in the long and complex process of developing the next generation of photo-stimulated ferroelectric devices.
A ferroic material is defined by its order parameter, that is, electric polarization in ferroelectrics, magnetization in ferromagnetics, or spontaneous strain in ferroelastics, which has different but energetically equivalent switchable orientations13,14,15,16,17,18,19. This phenomenon leads to the appearance of domains of differently oriented regions that coexist in the material and are separated by domain walls17,19. In general, domain walls have symmetry and structure different from the bulk of the domains and therefore play an important role in the switching mechanism of the associated order parameter by either one or an appropriate combination of driving fields10,11,12,14,17,18,19,20,21,22,23,24. Recent studies have shown that domain walls themselves can possess additional functionalities (for example, electronic conductivity)10,21,22 and have potentiality for other interesting effects, thereby making domain walls likely candidates for active elements in future nanoelectronics1,2,3,4,5,6,7,8,9,10,11,12,17,21,22,23,24. However, the use of domain walls requires that they can be easily indexable (that is, spatially localized), manipulatable (that is, writable/erasable, change shape/orientation, and so on) and reversible by the application of external stimuli in a controllable and repeatable manner15. In regard to domain switching in ferroic materials, such as ferromagnetics or ferroelectrics, their order parameter can traditionally be switched between two opposite values by the application of an external field15,16,17,18,19,20. However, recent advances in the nanoscale properties of ferroic materials provide many more versatile possibilities for next-generation nanoelectronic devices, which demand efficient and non-invasive methods for switching ferroic domains without electrical contacts1,2,3,4,5,6,7,8,9.
It was recently demonstrated that the ability to move ferroelectric domain walls by applying a coherent light source was in fact possible16. This new concept of light–matter coupling was discovered in a classic ferroelectric single crystal by using confocal Raman microscopy (CRM). Since this stimulant effect was microscopically described, a fundamental question arises: what is the true extent of the coupling between polarized light and ferroelectric polarization? There have been few practical studies of light–matter coupling in ferroelectric materials, a phenomenon that may explain the evolution dynamics of complex domain structure and its correlations with the macroscopic response. In fact, in view of the urgent need to develop the technology towards large-scale applications (macroscopic effects), it appears surprising that research still remains almost exclusively focused on local effects. In this Letter, we present a conclusive proof that the macroscopic polarization in a BaTiO3 (BTO) single crystal can be reversibly switched by illumination with visible polarized light. Such a fascinating property of BTO may lead to a new generation of optically driven ferroelectric-based micro(nano)electronics devices.
BTO is a classic, well-known lead-free ferroelectric material that is used here as a representative polydomain crystal. Figure 1a presents an optical micrograph of the polished surface of a (100)-oriented BTO sample, where two optically different zones can be easily identified. Square 1 is a zone characterized by containing mostly a domains (Fig. 1b), while square 2 is a multidomain zone with a complex structure (Fig. 1c), and they were spectroscopically resolved by CRM (Supplementary Section 1). The domain structure is composed of out-of-plane-polarization c domains (in blue) and in-plane-polarization a domains (in red) with a head-to-head (H–H) configuration of the polarization vectors (Fig. 1d). The H–H configuration maximizes the internal stress at the domain wall, and consequently the b domains (in green) are formed to release stress1. This complex domain structure has not been resolved macroscopically until now.
A basic identification of the structure and crystalline orientation of the BTO crystal can be performed by X-ray diffraction (XRD) measurements. Here, we go a step further and use high-energy, high-resolution XRD data to identify in situ relative changes in the domains structure. Figure 1e shows the XRD patterns of the synchrotron light-irradiated crystal zones, which are represented by squares 1 and 2 in Fig. 1a. Both diffractograms show the expected (h00) peaks splitting, without other reflections, indicating the single-crystal nature and tetragonal crystallographic symmetry of the sample. Although similar information can be extracted from any of the peaks splitting, the difference in the peaks intensity and the d spacing makes the election of peaks opportune for data analysis. Thus, a zoom on the major intensity reflection (002)/(200), as seen in Fig. 1f, shows the dominant (100) orientation or a plane of the crystal. The substantial difference in relative intensity of the (002) and (200) peaks (the I 002/I 200 ratio), depending on the irradiated zone (0.05 and 0.24 for regions 1 and 2, respectively), clearly shows the direct relation between the XRD pattern and the local domain structure. The greatest contribution of the (002) peak in region 2 is related to the complex domain structure revealed in this zone (Fig. 1c), where the notable presence of out-of-plane domains, or c domains, makes the c plane contribution more significant.
The more complex nature of the domain structure in region 2 (Fig. 1c) is really provided by the inclusion of the b domains, which appear to stabilize the c/a domain wall stress and its H–H configuration. Thus, the a plane contribution of the b domain should be detected in XRD spectra as a splitting of (h00) peaks into (h00) and (0h0) degenerate reflections. The peak (300) is zoomed in Fig. 1g, taking into account that the greater the h index, the greater the angular separation of the degenerate peaks, thereby making it more possible to resolve the splitting. A single peak with d spacing of a 1 = 1.3369 Å is detected for region 1, which indicates little influence by b domains, as is expected from Raman measurements (Fig. 1b). In contrast, a clearly (300)/(030) degenerate reflection is observed for region 2, revealing a d spacing difference (Δd = 0.0024 Å) between a- and b-domain contributions. The inclusion of b domains slightly distorts the unit cell, so the peak (300) emerges with a higher d spacing value a 2 = 1.3373 Å, whereas the peak (030) appears with a smaller d spacing b 2 = 1.3365 Å, where b 1 < a 1 < a 2. Thus, the crystal structure resolution probes the proposed complex domain configuration.
Once it has been shown that differences in the domain structure can be detected from high-resolution synchrotron XRD measurements, it is possible to demonstrate that a light-induced local electric field is able to change the domain configuration, thereby modifying the macroscopic polarization of the crystal. First, the effect of illumination by coherent light on the evolution of the XRD pattern is analysed. Region 2 in Fig. 1a was chosen for the analysis due to its multidomain configuration. Figure 2a,b shows the light (532 nm, 40 mW laser diode) off–on influence on the I 002/I 200 ratio, the value of which increases from 0.24 to 0.30 (Supplementary Section 2). The light therefore triggers a relative increase in c domains, which is induced by domain switching. The eventuality of domain coalescence may be ruled out here, because diffuse scattering, which is directly related to the domain wall fraction of the material, remains unaltered (Supplementary Section 3). The change in the volume ratio between a and c domains should modify the volume of the b domains to rearrange the domain configuration for stress relief, as confirmed in Fig. 2d,e by analysing the effect of off–on light on the (300)/(030) degenerate reflection. The I 030/I 300 ratio rises from 0.34 to 0.46 under light illumination (Supplementary Section 2), thereby indicating a relative increase in b domains. Thus, the volume of b and c domains increases at the expense of the volume of a domains, as is schematically illustrated at the top of Fig. 2. The reversibility of the phenomenon is illustrated by the sequence of light off–on–off in Fig. 2a–c and Fig. 2d–f. A succession of time-dependent experiments showed that the XRD pattern displays only two profiles, which correspond to off–on domain configurations (Supplementary Section 4). A simplified calculation based on the XRD measurement shows that a light-induced 90° domain switching of ~7% is produced in BTO under our experimental conditions (Supplementary Section 5), thus denoting a large strain capability of the material. This result makes the optically controlled deformation of a ferroelectric material feasible.
Different domain configurations should lead to different values of functional properties, such as dielectric permittivity, piezoelectric constants and so on. Figure 3 shows a light-induced change in the macroscopic dielectric permittivity of the BTO crystal. A notable variation in the real permittivity value occurs when the ferroelectric crystal is illuminated (Fig. 3a), which is related to the difference in its macroscopic polarization. Little differences in imaginary permittivity are observed (Fig. 3b), because no additional contribution to the imaginary electrical displacement exists. Taking into account that synchrotron radiation is in fact a light source, the dark condition in the capacitance experiment implies a real light off condition. Thus, the observed change in permittivity is a conclusive proof of light-controlled macroscopic polarization in ferroelectrics.
As concerns the physical mechanism responsible for this effect, the photovoltaic effect can be discarded (Supplementary Section 6). The formation of a stack of b domains at the boundary between the a and c domains with a H–H configuration leads to the formation of an asymmetric potential in this region8,25,26. Furthermore, this results in the appearance of compensation ‘locally free charges’ at the domain walls to balance the pressure of the bound polarization charge at the wall27. In the presence of the asymmetric potential, the a.c. electric field of the light can induce a net displacement of the charges in one direction due to a ratchet effect28. This displacement of the compensation charge will unbalance the pressure on the wall, hence inducing a force responsible for the domain wall motion (Supplementary Section 6).
So far, we have shown that a reversible optical change of the configuration of ferroelectric domains is possible. The next step is to consider the implications of our results for future technological applications. It is important to point out that no spatial confinement of light is needed to tune the macroscopic polarization of the material and that the effect emerges with relatively low light intensity. Light-driven electric polarization switching will be able to compete with electric polarization switching by an electric field or by applying strain. Furthermore, optical control of macroscopic polarization may enable us to establish a paradigm for a new generation of ferroelectric devices, in particular as photo-actuators.
The BTO single crystal was supplied by PI-KEM. The crystal was grown with (100) orientation (that is, in the a plane) using top-seeded solution growth method. A 5 mm × 5 mm × 0.5 mm sample was polished using 1 μm diamond paste and cleaned with ethanol and acetone before characterization. No additional chemical and/or thermal treatments were employed to reveal the domain structure, thereby preventing topographic artefacts because of nanoroughness.
Confocal Raman microscopy
The domain mappings were performed using a confocal Raman microscope (Witec alpha-300R). Raman spectra were collected using a 532 nm excitation laser and a ×100 objective lens (NA = 0.95). The incident laser power was 40 mW. The lateral and vertical resolutions of the confocal microscope were ~250 nm and 500 nm, respectively. The spectral resolution of the Raman mode was as low as 0.02 cm−1. The piezoelectric scanning allowed steps of 3 nm and 0.3 nm in the vertical direction, providing a suitable spatial resolution. Under the microscope objective, the domain walls of the sample were adjusted perpendicular to the x axis and parallel to the y axis of the piezo-driven scan table. The light was focused onto the sample surface, with the light polarization parallel to the surface, and the scan direction (indicated as the x axis) was kept perpendicular to the a–c domain wall. The domain structure of the sample was controlled by capturing sequential depth scans in two distinct areas separated by 1 mm. A position reference was established using a significant point on the sample surface visible with the optical microscope. Raman images for the depth scan (500 μm width, 1.8 μm height) equated to 1,000 × 36 spectra, and required ~2 h for acquisition of the whole cross-section. Accumulated spectra were analysed using Witec Control Plus Software.
Synchrotron radiation high-resolution XRD
Measurements were carried out at the SpLine CRG BM25 beamline at the European Synchrotron Radiation Facility in Grenoble, France. The X-ray beam wavelength was set to 0.5636 Å (22 keV) with an energy resolution ΔE/E of ∼10−4. The beam spot size was adjusted to 0.5 mm × 0.5 mm, based on the area of the two main regions of the BTO single crystal previously determined by CRM. The home-designed θ–2θ diffractometer had a custom-designed multicrystal detector stage. This stage incorporated 10 single point detectors (NaI(TI) scintillation) operating in parallel, and each point detector was equipped with a Ge(111) single-crystal analyser in a θ–2θ configuration. Both the diffractometer and the multi-analyser detector set-up produced a significant resolution improvement over the powder diffraction station while retaining a relatively low acquisition time, because simultaneous parallel detection was accomplished. Instrument calibration and wavelength refinement were performed with a silicon standard. The sample was mounted in a flat configuration with the domains walls aligned perpendicular to the X-ray beam path, and the sample was not rotated. The diffractometer was appropriately aligned to select the area of incidence of the beam spot. XRD patterns were recorded over the angular range 7.5–27.5° (2θ) with a step size of 0.0025° (this implies a d spacing resolution of ∼10−4 Å) and a time per step of 1 s. Some measurements were performed only around a certain peak to verify data repeatability. The angular resolution and time per step remained unchanged. For experimentation in illumination conditions, a laser diode (Thorlabs) was coupled to the θ–2θ diffractometer. The power and wavelength of the light source were 40 mW and 532 nm, respectively. The light spot was adjusted at 1.0 mm × 1.0 mm using an iris. The light irradiation time was precisely controlled using an electronic switch. The sample was illuminated in the area denoted as region 2 in Fig. 1a. The reversibility of the phenomenon was investigated by measuring XRD patterns for a set of on–off light cycles.
Macroscopic dielectric measurement
To carry out the dielectric measurements, a thin layer of indium tin oxide (ITO) with a total thickness of 200 nm was deposited on both sides of the BTO sample. Two gold pins were connected to the front and back electrodes via a mechanical contact. The admittance of the BTO single crystal was tested using an impedance analyser (HP4294A) in the frequency range 100 Hz to 100 kHz at room temperature. For experimentation under illumination conditions, a laser diode (Thorlabs) operating at a wavelength of 532 nm and a power of 40 mW was used. The spot size was adjusted to 1 mm × 1 mm, as in the case of XRD measurements. The light irradiation time was controlled using an electronic switch.
The data that support the plots within this paper and other findings of this study are available from the corresponding authors upon reasonable request.
Rubio-Marcos, F., Del Campo, A., Marchet, P., Romero, J. J. & Fernández, J. F. Ferroelectric domain wall motion induced by polarized light. Nat. Commun. 6, 6594 (2015).
Manz, S. et al. Reversible optical switching of antiferromagnetism in TbMnO3. Nat. Photon. 10, 653–656 (2016).
Iurchuk, V. et al. Optical writing of magnetic properties by remanent photostriction. Phys. Rev. Lett. 117, 107403 (2016).
Ying, C. Y. J. et al. Light-mediated ferroelectric domain engineering and micro-structuring of lithium niobate crystals. Laser Photon. Rev. 6, 526–548 (2012).
Boes, A. et al. Direct writing of ferroelectric domains on strontium barium niobate crystals using focused ultraviolet laser light. Appl. Phys. Lett. 103, 142904 (2013).
Guo, R. et al. Non-volatile memory based on the ferroelectric photovoltaic effect. Nat. Commun. 4, 1990 (2013).
Sando, D. et al. Large elasto-optic effect and reversible electrochromism in multiferroic BiFeO3. Nat. Commun. 7, 10718 (2016).
Yang, S. Y. et al. Above-band gap voltages from ferroelectric photovoltaic devices. Nat. Nanotech. 5, 143–147 (2010).
Choi, K. J. et al. Enhancement of ferroelectricity in strained BaTiO3 thin films. Science 306, 1005–1008 (2004).
Seidel, J. et al. Conduction at domain walls in oxide multiferroics. Nat. Mater. 8, 229–234 (2009).
McGilly, L. J., Yudin, P., Feigl, L., Tagantsev, A. K. & Setter, N. Controlling domain wall motion in ferroelectric thin films. Nat. Nanotech. 10, 145–150 (2015).
Agar, J. C. et al. Highly mobile ferroelastic domain walls in compositionally graded ferroelectric thin films. Nat. Mater. 15, 549–556 (2016).
Kwak, B. S. et al. Strain relaxation by domain formation in epitaxial ferroelectric thin films. Phys. Rev. Lett. 68, 3733–3736 (1992).
Li, D. & Bonnell, D. A. Controlled patterning of ferroelectric domains: fundamental concepts and applications. Annu. Rev. Mater. Res. 38, 351–368 (2008).
Eerenstein, W., Mathur, N. D. & Scott, J. F. Multiferroic and magnetoelectric materials. Nature 442, 759–765 (2006).
Mathur, N. A desirable wind up. Nature 454, 591–592 (2008).
Catalan, G., Seidel, J., Ramesh, R. & Scott, J. F. Domain wall nanoelectronics. Rev. Mod. Phys. 84, 119 (2012).
Béa, H. & Paruch, P. A way forward along domain walls. Nat. Mater. 8, 168–169 (2009).
Bibes, M. Nanoferronics is a winning combination. Nat. Mater. 11, 354–357 (2012).
Matsukura, F., Tokura, Y. & Ohno, H. Control of magnetism by electric fields. Nat. Nanotech. 10, 209–220 (2015).
Feig, L. et al. Controlled stripes of ultrafine ferroelectric domains. Nat. Commun. 5, 4677 (2014).
Sluka, T., Tagantsev, A. K., Bednyakov, P. & Setter, N. Free-electron gas at charged domain walls in insulating BaTiO3. Nat. Commun. 4, 1808 (2013).
Salje, E. K. H. Multiferroic domain boundaries as active memory devices: trajectories towards domain boundary engineering. ChemPhysChem. 11, 940–950 (2010).
Farokhipoor, S. et al. Artificial chemical and magnetic structure at the domain walls of an epitaxial oxide. Nature 515, 379–383 (2014).
Seidel, J. et al. Efficient photovoltaic current generation at ferroelectric domain walls. Phys. Rev. Lett. 107, 126805 (2011).
Liu, S. et al. Ferroelectric domain wall induced band gap reduction and charge separation in organometal halide perovskites. J. Phys. Chem. Lett. 6, 693–699 (2015).
Mokrý, P., Tagantsev, A. K. & Fousek, J. Pressure on charged domain walls and additional imprint mechanism in ferroelectrics. Phys. Rev. B 75, 094110 (2007).
Pérez-Junquera, A. et al. Crossed-ratchet effects for magnetic domain wall motion. Phys. Rev. Lett. 100, 037203 (2008).
This work is supported by the Ministry of Economy, Industry and Competitiveness (MINECO, Spanish Government) project MAT2013-48009-C4-P and by the Spanish National Research Council (CSIC) under project NANOMIND CSIC 201560E068. The authors acknowledge ESRF, The European Synchrotron, CSIC, MINECO and the SpLine CRG BM25 beamline staff for provision of synchrotron radiation and assistance during XRD measurements. F.R.-M. acknowledges MINECO for a ‘Ramon y Cajal’ contract (RyC-2015-18626), co-financed by the European Social Fund.
The authors declare no competing financial interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Rubio-Marcos, F., Ochoa, D.A., Del Campo, A. et al. Reversible optical control of macroscopic polarization in ferroelectrics. Nature Photon 12, 29–32 (2018). https://doi.org/10.1038/s41566-017-0068-1
npj 2D Materials and Applications (2021)
Nature Communications (2021)
Scripta Materialia (2020)
Physical Sciences Reviews (2020)
Journal of the American Chemical Society (2020)