Strontium titanate SrTiO3 is known as a quantum paraelectric material, and its lattice dynamics and unusual dielectric character have been studied extensively. The cubic (Oh) structure above the structural phase transition temperature (
Photo-induced effect in dielectric materials is an attractive topic. Some kind of ferroelectric materials such as SbSI (Ueda et al., 1967) and BaTiO3 (Volk et al., 1973; Godefroy et al., 1976) are known to show photo-induced effects. In this decade, much interest has been paid on the giant enhancement in dielectric constants under ultraviolet (UV) illumination and DC electric field in quantum paraelectrics, strontium titanate SrTiO3 and potassium tantalate KTaO3 (Takesada et al., 2003; Hasegawa et al., 2003; Katayama et al., 2003), because weak light illumination gives rise to an intense response in dielectricity.
The two models shown Fig. 1, the ferroelectric cluster model (Takesada et al., 2003; Hasegawa et al., 2003; Katayama et al., 2003) and the conductive-region model (Homes et al., 2001; Katayama et al., 2003), have been proposed to explain the origin of the giant dielectric constants. At present, however, it is still not clear which model is better. In the ferroelectric cluster model, the photo-induced ferroelectric region has a huge dipole moment, where it is expected that a photo-induced polar domain generates spatial lattice distortion. In the conductive-region model, on the other hand, the superposition of insulative and photo-induced conductive regions, which is characterized by the boundaries between the two regions, makes the apparent dielectric constants to be enormous.
Giant dielectric response has been observed in some types of nonferroelectric materials (Homes et al., 2001; Wu et al., 2002; Dwivedi et al., 2010). The enormous increase in dielectric constants is attributed to the formation of barrier layer capacitors and the resultant Maxwell-Wagner polarization or interfacial polarization. This giant dielectric response often occurs in materials with grains surrounded by the insulating grain boundary and is explained by the conductive-region model.
According to the measurement of dielectric constants, a doped crystal Sr1-
In Ca-doped SrTiO3, a UV illumination causes a shift of the ferroelectric phase transition temperature toward the lower side (Yamada & Tanaka, 2008). The
In the present study, we performed three types of experiment in pure and Ca-doped SrTiO3;
stationary birefringence measurement in UV light and DC electric fields,
transient birefringence measurement in UV light and pulsed electric fields, and
transient absorption and birefringence measurements after the optical pulse excitation using the pump-probe technique.
The photo-induced dynamics of the lattice distortion, the dielectric polarization, and the relaxed excited state in SrTiO3 is studied in comparison with the lattice distortion in the doping-induced ferroelectric phase of Ca-doped SrTiO3. We discuss which model explains the experimental results better.
The experiments are performed on single crystals of pure and Ca-doped SrTiO3 with the Ca concentration of
2. Lattice distortion in the UV and DC fields in Ca-doped SrTiO3
The stationary birefringence is studied to investigate the static properties of the lattice distortion generated by the UV illumination in comparison with that generated by the ferroelectric deformation.
2.1. Birefringence measurement in the UV light and DC electric fields
The schematic diagram of the birefringence measurement in the UV light and DC electric fields is shown in Fig. 2. The change in birefringence is detected as the change in the polarization of a linearly polarized probe light provided by a Nd:YAG laser (532 nm). The source of UV illumination is provided by the second harmonics (380 nm, 3.3 eV) of the output from a mode-locked Ti-sapphire laser, whose energy is larger than the optical band gap of SrTiO3 (3.2 eV). The intensity of UV illumination is 1.6 mW/mm2. Since the repetition rate of the UV pulses is 80 MHz, this UV illumination can be considered to be continuous in the present experiment. The UV beam is illuminated on the gap between two Au electrodes. The electrodes with a gap of 0.8 mm are deposited on a (100) surface of the samples by spattering. A DC electric field, whose amplitude is 375 V/mm, is applied between the two electrodes. The DC electric field is applied parallel to  direction of the crystal.
The change in the polarization of the probe light is detected by a polarimeter. The construction of the polarimeter is shown in Fig. 3. The polarimeter (Kohmoto et al., 2000; Jones, 1976) detects the rotation of polarization plane of a light beam. A linearly-polarized beam is split by a polarized beam splitter (PBS) and incident on the two photodiodes (PD) whose photocurrents are subtracted at a resistor (R). When the polarized beam splitter is mounted at an angle of 45° to the plane of polarization of the light beam, the two photocurrents cancel. If the plane of polarization rotates, the two currents do not cancel and the voltage appears at the resistor.
In the present experiment, the birefringence generated by the lattice deformation is detected as the change in polarization of the probe beam using a quarterwave plate and a polarimeter. The birefringence generated in the sample changes the linear polarization before transmission to an elliptical polarization after transmission. The linearly-polarized probe beam is considered to be a superposition of two circularly-polarized components which have the opposite polarizations and the same intensities. The generated birefringence destroys the intensity balance between the two components. The two circularly-polarized beams are transformed by the quaterwave plate to two linearly-polarized beams whose polarizations are crossed each other, and the unbalance of circular polarization is transformed to the unbalance of linear polarization or the rotation of polarization plane. This rotation is detected by the polarimeter as the signal of the lattice deformation.
2.2. UV intensity dependence of the birefringence
The ultraviolet intensity dependence of the change in birefringence in Ca-doped SrTiO3 is shown in Fig. 4, where the temperature is 6 K and the polarization plane of the probe light is along the  and  axes, with which the lattice distortion along the  and  axes are detected, respectively. The birefringence increases nonlinearly as the UV intensity is increased. As is shown in Fig. 4(a), the change in birefringence appears at very weak UV intensity in the polarization plane only along the  axis, rises rapidly, and holds almost a constant value above 0.5 mW/mm2. Figure 4(b), where the horizontal axis is in a logarithmic scale, indicates that the structural deformation begins at the UV intensity of 10-3 mW/mm2. The change in birefringence for the probe polarization along the  axis is much larger than that along the  axis. These facts imply that the UV illumination causes Ca-doped SrTiO3 to undergo a first-order-like structural deformation and generates a lattice distortion along the  axis as a result of the competition between the UV-induced and ferroelectric deformations, and its threshold value is very small.
Figure 5 schematically shows the direction of the local lattice distortion in pure and Ca-doped SrTiO3. The observed direction of the lattice distortion in Ca-doped SrTiO3 generated by the UV illumination is the same as that in the case of pure SrTiO3 (Nasu, 2003).
2.3. Temperature dependence of the birefringence in the UV and DC fields
We investigated the temperature dependence of the change in birefringence for Ca-doped SrTiO3 in the combination of two external fields, UV light (UV) and DC electric (DC) fields. The experimental result is shown in Fig. 6 where the polarization plane of the probe light is along the  and  axes. The sample is in the four types of fields; neither UV nor DC (no field), only DC (DC), only UV (UV), and both UV and DC (UV+DC). The changes in birefringence for the probe polarization along the  axis are much larger than those along the  axis. This means that the optical anisotropy is generated along the  axis. For the probe polarization along the  axis without the DC electric field, the change in birefringence for no field is similar to that for UV, as is seen in Fig. 6, while under the DC1
electric field the change for DC is different from that for UV+DC. The difference arises from that of the macroscopic optical anisotropy generated along the  axis by the UV illumination.
In the ferroelectric phase of Ca-doped SrTiO3, the direction of the local lattice distortion is diagonal along the  axis (Bednorz & Muller, 1984) as shown in Fig. 5(b). There are six equivalent diagonal sites where the distortion directions are , [1-10], , [01-1], , and [10-1]. In no field, it is expected that the six local sites distribute randomly as shown in Fig. 7(a), and no optical anisotropy is generated. The observed birefringence change Δ
In the DC electric field along the  axis, on the other hand, the six local diagonal sites in the ferroelectric phase are not equivalent as shown in Fig. 7(b). The two diagonal sites,  and [01-1] which are perpendicular to the  axis, are more unstable in the DC elected field along the  axis than the other four diagonal sites, , [1-10], , and [10-1]. The random distribution of the four diagonal sites generate a macroscopic optical anisotropy along the  axis. This explains the observed large increase of the birefringence change Δ
As discussed in section 2.2, the lattice distortion generated by the UV illumination is axial along the  axis. In no DC electric field, its direction distributes randomly among the three equivalent directions, , , and  as shown in Fig. 7(c), where no macroscopic optical anisotropy is expected. The observed similar behavior between the birefringence changes Δ
In the DC electric field along the  axis, the three axial sites generated by the UV illumination are not equivalent as shown in Fig. 7(d). The two axial sites  and , which are perpendicular to the  axis, are more unstable than the other axial site . This axial site can also contribute to generate a macroscopic optical anisotropy along the  axis. The UV illumination changes some part of the local distortion from the diagonal site along the  axis to the axial site along the  axis. The UV illumination decreases the birefringence change in the DC electric field; Δ
3. Lattice distortion in the UV and pulsed electric fields in pure and Ca-doped SrTiO3
3.1. Transient birefringence measurement in the UV light and pulsed electric fields
The transient birefringence in a pulsed electric field is studied to probe whether or not the lattice distortion generated by the UV illumination is affected by the electric field. In the experiment of transient birefringence, the change in birefringence is monitored by the probe light of 532 nm from the Nd:YAG laser, and its time evolution is obtained from the waveform digitized on an oscilloscope. A pulsed electric field, whose amplitude is 150 V/mm and pulse width is 1.2 ms, is applied between the two Au electrodes.
3.2. Transient birefringence in SrTiO3
Figure 8(a) shows the change Δ
Figure 8(b) shows the change in birefringence in SrTiO3 under the UV illumination. The dielectric response of relaxation type is observed at all temperatures. As the temperature is decreased, the relaxation rate becomes increased.
3.3. Transient birefringence in Ca-doped SrTiO3
The change Δ
3.4. Temperature dependence of the transient birefringence amplitude
The temperature dependences of the transient birefringence amplitude Δ
For both samples, it is clear that the transient birefringence amplitude for the probe polarization along the  axis is reduced by the UV illumination, which indicates that the lattice deformation is reduced by the UV illumination and that the lattice distortion generated by the UV illumination is not affected by the electric field. This result is not consistent with the ferroelectric cluster model, where a large change in birefringence is expected in the pulsed electric field.
In Ca-doped SrTiO3, judging from the rising temperature of the transient birefringence amplitude, the ferroelectric phase transition temperature is shifted toward the lower temperature side under the UV illumination. This is consistent with the dielectric measurement (Yamada & Tanaka, 2008) and the coherent phonon experiment (Koyama et al., 2010). The doped Ca ions behave as permanent dipoles, and ferroelectric clusters are formed around the Ca dipoles with the high polarizability of the host crystal. The ferroelectric transition is caused by the ordering of the randomly distributed Ca dipoles. The ordering is prevented by the photo-excited carriers generated by the UV illumination.
4. Dynamics of the relaxed excited state after the optical pulse excitation in SrTiO3
Figure 11 illustrates the electronic states in SrTiO3. The UV illumination, whose photon energy is larger than the band-gap energy of SrTiO3 (3.2 eV), excites the electrons from the ground state to the excited state. The excited electrons are self-trapped in the relaxed excited state where the optically induced lattice distortion is created (Nasu, 2003). In the relaxed excited state, a broad absorption band appears (Hasegawa & Tanaka, 2001; Okamura et al., 2006).
Figure 12 shows the emission spectra observed under two types of optical excitation; the pulse excitation of 790 nm and the UV illumination of 380 nm. As is seen, the two spectra are almost the same as each other. This result indicates that the same relaxed excited state is generated by the pulse excitation through a multi-photon absorption process. We observed the dynamics of the optically induced lattice distortion in the relaxed excited state generated by ultrashort pump pulses. The transient absorption and birefringence after the optical pulse excitation, which is originated in the generation of the relaxed excited state, is studied by the pump-probe technique to investigate the dynamical properties of the optically induced lattice distortion.
4.1. Pump-probe measurement after the optical pulse excitation
The dynamics of the optically induced lattice distortion was observed by transient absorption and birefringence with the pump-probe technique. The lattice distortion is generated in the relaxed excited state by a linearly polarized pump pulse through a multi-photon absorption process. Induced time-dependent absorption or anisotropy of refractive index after the pulse excitation is detected as the change of the transmission or polarization of a linearly polarized probe pulse. The time evolution of the signal was observed by changing the optical delay between the pump and probe pulse.
The pump pulse is provided by a Ti:sapphire regenerative amplifier whose wavelength, pulse energy, and pulse width at the sample are 790 nm, 2 μJ, and 0.2 ps, respectively. In the experiment in the picosecond region, the probe pulse is provided by an optical parametric amplifier whose wavelength, pulse energy, and pulse width are 700 nm, 0.1 μJ, and 0.2 ps, respectively. The repetition rate of the pulses is 1 kHz. The time evolution of the signal is observed by changing the optical delay between the pump and probe pulses. The pump pulse is switched on and off shot by shot by using a photoelastic modulator, a quarter-wave plate, and a polarizer, and the output from the polarimeter is lock-in detected to improve the signal-to-noise ratio. In the experiment in the millisecond region, a continuous-wave probe light provided by a laserdiode, whose wavelength is 660 nm, is used. The linearly polarized pump and probe beams are nearly collinear and perpendicular to the (001) surface of the sample, and are focused on the sample in a temperature-controlled refrigerator.
The transmission intensity is detected by a photodiode. The induced anisotropy of refractive index is detected by the polarimeter with a quarter-wave plate as the polarization change of the probe pulse. The plane of polarization of the probe pulse is tilted by 45° from that of the pump pulse. The two different wavelengths for the pump and probe lights and pump-cut filters are used to eliminate the leak of the pump light from the input of the photodiode or polarimeter.
4.2. Transient absorption in the picosecond region
It is found that the transmission is decreased after the optical pulse excitation. The transient absorption signal in SrTiO3 have the component rising in the picosecond region as shown in Fig. 13. This rise time, which is of the order of 100 ps, corresponds to the generation time of the relaxed excited state.
4.3. Transient absorption in the millisecond region
The temperature dependence of the transient absorption in the millisecond region in SrTiO3 is shown in Fig. 14. The transient absorption signals have the decay time in the millisecond region. As the temperature is decreased, the signal amplitude is increased. The averaged occupation probability in the relaxed excited state strongly depends on the temperature and is large at low temperatures, where the photo-induced giant dielectricity is observed.
4.4. Transient birefringence in the millisecond region
The time evolution of the lattice distortion is observed by transient birefringence measurement with the pump-probe technique. The temperature dependence of the transient birefringence in the millisecond region in SrTiO3 is shown in Fig. 15. The observed signal curves of the transient birefringence and their temperature dependence show similar behavior to those of the transient absorption.
The photo-induced conductivity in SrTiO3, where the resistivity drops rapidly at low temperatures, has also been reported (Katsu et al., 2000). The photo-induced lattice distortion, absorption, and conductivity can be interpreted to have the same origin in the relaxed excited state. The experimental results suggest that the photo-induced effect in SrTiO3 is consistent with the conductive-region model.
Giant dielectric response in nonferroelectric materials has been reported for grains surrounded by the insulating grain boundary and is explained by the Maxwell-Wagner model, namely the conductive-region model (Homes et al., 2001; Wu et al., 2002; Dwivedi et al., 2010). SrTiO3 crystal is a stoichiometric material and has no grain structure. It may have a superposition of insulative and conductive regions under the UV illumination at low temperatures, whose structure may be related to the quantum paraelectricity under the dark illumination. However, the detail of the microscopic structure is unknown at present.
We observed the temperature dependences of the stationary birefringence, the transient birefringence in the pulsed electric field, and the transient absorption and birefringence after the optical pulse excitation to investigate the photo-induced effect in pure and Ca-doped SrTiO3.
The first-order-like structural deformation is found in the UV intensity dependence in Ca-doped SrTiO3, which begins at a very weak UV intensity of 10-3 mW/mm2. The generated lattice distortion is along the  axis as in the case of pure SrTiO3. The shift of the ferroelectric phase transition temperature toward the lower temperature side is observed under the UV illumination.
From the temperature dependence of the change in birefringence in Ca-doped SrTiO3 in the four types of fields, where a UV light field is on or off and a DC electric field is on or off, it is found that a structural deformation is generated along the  axis under the UV illumination and its magnitude is of the same order as that of the ferroelectric deformation along the  axis.
It is found that the photo-induced lattice distotion is not affected by the electric field. The photo-induced lattice distotion, absoption, and conductivity are generated in the relaxed excited state, whose lifetime is in the millisecond region. These facts suggest that the photo-induced giant electricity in SrTiO3 is consistent with the conductive-region model rather than the ferroelectric cluster model.
SrTiO3 is a very peculiar material in which the dielectricity of the Maxwell-Wagner type appears in a stoichiometric material only under UV illumination at low temperatures. The origin of the enormous dielectric constants under UV illumination is considered to be different from that under dark illumination, although the latter is still not made clear.
We would like to thank Dr. Y. Yamada and Prof. K. Tanaka for providing us the samples of Ca-doped SrTiO3.
Bednorz J. G. Muller K. A. 1984Sr1-xCaxTiO3: An XY quantum ferroelectric with transition to randomness,
Bianchi U. Kleeman W. Bednorz J. G. 1994 Raman scattering of ferroelectric Sr1-xCaxTiO3, x=0.007,
Dwivedi G. D. Tseng K. F. Chan C. L. Shahi P. Lourembam J. Chatterjee B. Ghosh A. K. Yang H. D. Chatterjee S. 2010Signature of ferroelectricity in magnetically ordered Mo-doped CoFe2O4,
Godefroy G. Jullien P. Cai L. 1976Photoconduction in doped BaTiO3 single crystals,
Hasegawa T. Mouri S. Yamada Y. Tanaka K. 2003 Giant photo-induced dielectricity in SrTiO3,
Hasegawa T. Tanaka K. 2001 Photo-induced polaron states in strontium titanate,
Homes C. C. Vogt T. Shapiro S. M. Wakimoto S. Ramirez A. P. 2001 Optical response of high-dielectric-constant perovskite-related oxide,
Jones R. V. 1976 Rotary ‘aether drag’,
Katayama I. Ichikawa Y. Tanaka K. 2003 Critical behaviors of photoinduced giant permittivity in potassium tantalate
Katsu H. Tanaka H. Kawai T. 2000 Anomalous Photoconductivity in SrTiO3,
Kohmoto T. Fukuda Y. Kunitomo M. Isoda K. 2000Observation of ultrafast spin-lattice relaxation in Tm2+-doped CaF2 and SrF2 crystals by optical means,
Koyama Y. Moriyasu T. Okamura H. Yamada Y. Tanaka K. Kohmoto T. 2010 Doping-induced ferroelectric phase transition in strontium titanate: Observation of the birefringence and the coherent phonons under the ultravioletillumination,
Muller K. A. Burkard H. 1979SrTiO3: An intrinsic quantum paraelectric below 4 K
Nasu K. 2003 Photogeneration of superparaelectric large polarons in dielectrics with soft anharmonic T1u phonons,
Okamura H. Matsubara M. Tanaka K. Fukui K. Terakami M. Nakagawa H. Ikemoto Y. Moriwaki T. Kimura H. Nanba T. 2006 Photogenerated carriers in SrTiO3 probed by mid-infrared absorption,
Takesada M. Yagi T. Itoh M. Koshihara S. 2003 A gigantic photoinduced dielectric constant of quantum paraelectric perovskite oxides observed under a weak DC electric field,
Ueda S. Tatsuzaki I. Shindo Y. 1967 Change in the dielectric constant of SbSI caused by illumination,
Volk T. R. Grekov A. A. Kosonogov N. A. Fridkin V. M. 1973Influence of illumination on the domain structure and the Curie temperature of BaTiO3,
Vugmeister B. E. Glinchuk M. P. 1990 Dipole glass and ferroelectricity in random-site electric dipole systems
Wu J. Nan C. W. Lin Y. Deng Y. 2002 Giant dielectric permittivity observed in Li and Ti doped NiO,
Yamada Y. Tanaka K. 2008 Mechanism of photoinduced dielectric response in ferroelectric Sr1-xCaxTiO3,