Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Second harmonic generation in ferroelectric Ba2EuFeNb4O15-based epitaxial thin films

Open Access Open Access

Abstract

Ba2EuFeNb4O15-based epitaxial ferroelectric thin films with tetragonal tungsten bronze crystal structures are shown to have nonlinear optical properties at room temperature and are promising candidates for integrated optical frequency convertors and related applications. High quality epitaxial ferroelectric thin films of tetragonal tungsten bronze Ba2EuFeNb4O15 have been synthesized on MgO(100) by pulsed laser deposition. Structural investigation reveals that the c-oriented films are rotated in plane by ±18.4° and ±31° with respect to the substrate structure. Stable ferroelectric properties were obtained by microelectromechanical characterization. Second harmonic generation related to the spontaneous polarization in the films was studied and the independent components of the nonlinear susceptibility were determined. Ferroelectric films of Ba2EuFeNb4O15 having a tetragonal tungsten bronze structure are new candidate for room temperature nonlinear optical applications.

© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement

1. Introduction

Recently, new ferroelectric materials have been intensively studied due to their interesting nonlinear optical properties including second harmonic generation [1], electro-optical modulation [2,3], light emission [4] as well as for tunable optical filtering [5]. These characteristics derives from the fact that the electric field of intense optical pulses is sufficiently strong to act upon the intrinsic spontaneous polarization of these ferroelectric materials [68], hence allowing for the control and tuning of their nonlinear optical properties [1,9,10]. Indeed, both materials with ferroelectric and nonlinear optical properties are noncentrosymmetric, implying that ferroelectrics do necessarily exhibit nonlinear optical properties. Materials with a tetragonal tungsten bronze (TTB) have been widely studied for their ferroelectric and piezoelectric properties for the last 50 years [11,12]. It has been shown that the polar properties are related to the oxygen octahedra tilting in the TTB crystal structures, giving rise to generally good ferroelectric properties [13]. The crystal structure of TTBs consists of a network of oxygen octahedra with channels having different shapes – filled or left empty – allowing for the synthesis of a large variety of phases and doping, which, in turn, allows to optimize and control a variety of functional properties [14,15]. For instance, highly textured ferroelectric KSr2Nb5O15 ceramics with a high Curie temperature of about 170 °C exhibit a high remnant polarization of 18 µC/cm2 measured along to the c-axis [16]. Recently, we also demonstrated that TTB thin films of Ba2LnFeNb4O15 (Ln: lanthanide ion = Eu3+, Sm3+ and Nd3+) are multiferroic nanocomposites at room temperature [17]. Moreover, epitaxial TTB-Ln thin films showed an enhancement of their ferroelectric properties along the c-axis with a reduced ferroelectric fatigue compared to polycrystalline films of the same composition [18,19].

Second harmonic generation (SHG) is a nonlinear process whereby a fraction of the light passing through a nonlinear optical material gets frequency doubled [20,21]. This nonlinear process finds its application in detection and study of spontaneous polarization in ferroelectrics, as well as in probing some orientational or compositional inhomogeneities in matter [22]. SHG has also been previously demonstrated to be a valuable tool for the characterization of ferroelectrics [23]. A number of non-centrosymmetric materials with tetragonal tungsten bronze structure are ferroelectrics, and are known to have good properties for SHG [11]. Most of these materials, however, suffer from low Curie temperature that limits their use in room temperature applications. Thanks to its high Curie temperature (≈ 440 K) [24], the ferroelectric material with tetragonal tungsten bronze crystal structure Ba2EuFeNb4O15 (TTB-Eu) is a promising candidate for SHG and nonlinear optics at room temperature and above. In our previous work [17], TTB-Ln thin films have been shown not only to be ferroelectric but to be multiferroic at room temperature, thus exhibit both ferroelectric and ferromagnetic properties with a high ferroelectric Curie temperature of 440 K [14,24] and a high magnetic Curie temperature of 750 K [25,26], respectively. The various couplings between these properties offers many possibilities to control and modulate the non-linear optical properties that would not be present in classical non-linear optical materials. Moreover, we demonstrated that TTB-Eu films are characterized by a reduced ferroelectric fatigue with the polarization being reduced by only 30% after 107 cycles [18], attesting their robust ferroelectric properties.

The main goal of this study is to demonstrate that TTB-Eu films are a novel candidate for second harmonic generation at room temperature in addition of their multiferroic properties. Thus, the present paper discusses room temperature SHG in TTB-Eu epitaxial films prepared by pulsed laser deposition (PLD) in order to establish their nonlinear optical properties, by performing polarization sensitive SHG measurements. While the X-ray diffraction allowed determining the epitaxial growth of the studied films, their room temperature ferroelectric properties were determined using Piezoelectric Force Microscopy. More interesting, the independent components of the nonlinear susceptibility tensor were determined using SHG measurements. Therefore, this contribution furthers the characterization of TTB-Eu thin films, showing their room temperature nonlinear optical properties in addition of their multiferroic properties previously demonstrated.

2. Materials and methods

2.1. TTB-Eu thin films

Epitaxial Ba2EuFeNb4O15 thin films have been grown on MgO(100) single crystalline substrates by pulsed laser deposition (PLD) [18,27]. During the films deposition, a stoichiometric ceramic target of Ba2EuFeNb4O15 having a density higher than 90% of the theoretical density was ablated by an energetic pulsed KrF excimer laser beam of wavelength λ = 248 nm with pulses duration of 25 ns. The laser beam was focused on the rotating target as a spot having an area of 2 mm2, which results to a laser fluency of 2.8 J/cm2. In order to ablate the target more uniformly, the laser beam was also scanned across the target having a diameter of around 3 cm. The pulsed laser repetition rate during the ablation process was 10 Hz. The distance between the target and the MgO(100) substrate was set at 6.5 cm. The substrate, having a dimensions of 0.5 cm x 0.5 cm, was fixed on a rotating holder and heated at an optimized temperature of 850 °C, providing a good homogeneity of the composition and uniformity of the thickness as well as a good crystallization of the deposited films [18,2729]. The deposition was performed under an atmosphere of pure oxygen at a pressure of 1 mTorr.

2.2. Structural and electromechanical characterizations

The films crystalline structure was investigated by X-ray diffraction using a 4-circle X-ray diffractometer (PANalytical X-Pert Pro MRD). The microelectromechanical properties at the films surface were investigated by piezoelectric force microcopy using a DI-enviroscope atomic force microscope (Bruker, Santa Barbara, CA) equipped with an MESP cantilever/tip (Bruker, Santa Barbara, CA) coated with Co/Cr (see appendix).

2.3. Polarization sensitive SHG measurements

The TTB-Eu sample was inserted in a setup that allows it to freely rotate orthogonally to the focal plane (rotating mount) in order to probe the out-of-plane components of the SHG signal, and enables a loose focalization where the plane-wave conditions are respected, such that the distortion of the light polarization due to tight focusing is avoided [30]. The different angles, planes and axis used are shown on the schematic representation of the experimental setup shown in Fig. 1. While φ represents the angle between the linear polarization and the x-axis, θ is the angle between the plane of the sample and the XY plane. A Ti:Sapph laser is used with about 1.5 W of average power corresponding to roughly 19 nJ per pulse directly focused on the TTB-Eu thin film (300 nm thick) using a 5 cm focal lens. The incident beam is linearly polarized. Another 5 cm lens re-collimates the beam and sends it onto a Glan-Thompson polarizer used as an analyzer. Two low-pass filters and a narrow bandpass filter (FF01-720/SP-25 and FF01-405/10, Semrock) allow to reject the excitation laser and a CCD camera (IDS uEye) is used to detect the remaining SHG signal. A mechanical shutter is used to ensure a time-limited illumination of the sample such as the excitation laser is authorized to pass only when the SHG signal is recorded. These SHG polarization-resolved measurements are an easy way of retrieving the components of the nonlinear susceptibility tensor, which cannot be done by a simple SHG measurement alone.

 figure: Fig. 1.

Fig. 1. Schematic of the experimental set-up. λ/2: half-wave plate, L: +5 cm lens, F: 405 ± 10 nm filter, A: analyser, S: sample (TTB-Eu).

Download Full Size | PDF

The homogeneity of the SHG signal over the thin films was verified using SHG microcopy, as explained in the Appendix A.2.

3. Results and discussion

3.1 Structure of the TTB-Eu films

The Fig. 2(a) shows a typical X-ray diffractogram of TTB-Eu thin films deposited on an MgO(100) substrate. The MgO single crystalline cubic substrate is used due to its good lattice match with respect to the TTB-Eu crystal structure (lattice mismatch ≈ - 1.38%), which is appropriate for the TTB-Eu films epitaxial growth. Due to its centrosymmetric crystal structure, the MgO does not contribute to the second harmonic generation signal, and is therefore an appropriate substrate for an accurate study of SHG in TTB-Eu films. While the intense peak observed at 42.95° in Fig. 2(a) corresponds to the (200) plane of the substrate, the peak observed at 38.61° is attributed to the presence of the kβ line of the Cu reflection that appears due to the high intensity of the (200) peak. The peak (W) at 41.07° is due to a tungsten contamination of the Cu cathode of our XRD tube. In addition to the peaks that are attributed to the substrate, two diffraction peaks at 22.17° and at 45.31° are present. These peaks correspond to the reflections on the (001) and (002) planes of the TTB-Eu structure, respectively. The presence of only the (00l) peaks of the TTB-Eu diffraction spectrum attests that the synthesized films are oriented with the c-axis of their tetragonal structure perpendicular to the substrate surface.

 figure: Fig. 2.

Fig. 2. (a) X-ray diffractogram obtained in Bragg-Brentano geometry demonstrating the synthesis of TTB-Eu films oriented with their c-axis normal to the substrate surface.

(b) X-ray diffractogram obtained in Ф-scan geometry, which is formed by 16 peaks resulting from the diffraction on the (221) planes of the TTB-Eu structure. The dashed lines represent the position of the (111) diffraction peaks of the MgO(100) substrate determined during the same measurement.

Download Full Size | PDF

In order to determine the in-plane orientations of the films unit cell relative to the substrate structure, a 360° Ф-scan measurement for the (221) diffraction planes of the TTB-Eu films (2θ = 30.27° and ψ = 41.7°) was performed. While the diffraction peaks observed in Fig. 2(b) are due to the TTB-Eu(211) planes, the dashed lines denote the positions of the (111) planes of the (100)-oriented MgO substrate (2θ = 39.985° and ψ = 54.747°). The diffractogram that represents the (221) plane is composed of 16 peaks over 360°, attesting that the growth of the TTB-Eu films is actually epitaxial on MgO substrate, but with a complex in-plane configuration. The 16 diffraction peaks are divided on two groups: (i) the first group is formed by the peaks that are making an angle of ±18.4° with the substrate peaks. (ii) The second group is composed by the peaks that are making an angle of ±31° with the substrate peaks. This result shows the presence of two different possibilities to connect the lattices of the TTB-Eu thin films and of the MgO (100) cubic substrate, i.e. of two different epitaxial relationships between film and substrate, each of them having two realizations. This results in the presence of different regions of the films, called growth domains or growth variants, where the films unit cell are rotated in-plane by an angle of ± 18.4° and ± 31° with respect to the substrate cubic unit cell (4 growth variants on a cubic substrate having itself 4 equivalent directions in-plane, resulting in the presence of regions with 16 different possible in-plane orientations, hence 16 peaks). These results confirmed what we have previously obtained and reported for epitaxial TTB-Eu thin films grown by PLD [18].

In order to demonstrate the ferroelectric properties of the synthesized films, the variation of the longitudinal piezoelectric coefficient (ξZZ) has been studied and is shown as a function of the applied voltage (V) in Fig. 3(a). The hysteresis behavior of ξZZ vs. V indicates that the vertical component of the polarization can switch reversibly in two different states when cycling the applied voltage. The remnant piezoelectric coefficient and the piezoelectric coefficient at the saturation are 1.28 pm/V and 2.29 pm/V, respectively. The difference of coercive voltages in both positive and negative directions shows an asymmetry, which is explained by the fact that the top electrode is the AFM/PFM tip, which has both a very different geometry and a different composition than the continuous bottom electrode [18]. Indeed, During the PFM measurement an electric field is applied across the TTB film and electrically charged depletion layers are formed at the film-electrodes interfaces. Since the interfaces between films and top and bottom electrodes are different, different depletion layers appear at the top and bottom interfaces. Thus, the effect of these asymmetric depletion layers on the ferroelectric properties is responsible for the asymmetry observed in the ferroelectric hysteresis loops [31,32]. While Fig. 3(a) shows the longitudinal piezoelectric coefficient ξZZ, Fig. 3(b) reveals that the variation of the transverse piezoelectric coefficient ξXZ as a function of an applied voltage exhibits a hysteresis loop as well. It is noted that piezoelectric coefficient determined for ξXZ is around 0.34 pm/V, which is only around 15% of the value of ξZZ. The difference between ξZZ and ξXZ is mainly attributed to the c-orientation of the epitaxial TTB-Eu [18].

 figure: Fig. 3.

Fig. 3. (a) Longitudinal piezoelectric coefficient (ξZZ), and (b) transverse piezoelectric coefficient (ξXZ) versus an applied voltage. The in-plane piezoelectric coefficient ξXZ represents roughly 15% of the out-of-plane piezoelectric coefficient ξZZ.

Download Full Size | PDF

3.2 SHG response

The Fig. 4(a) shows the SHG image observed in a typical TTB-Eu thin film, where the parabolic shape of the illumination due to the laser-scanning is visible since the objective loses its focus ability at high incidence angles. Still, the observed SHG image is very homogeneous, with no defect, refractive index change, or domains visible. If the sample surface would be rough, we would see intensity variations on the SHG image, like a speckle pattern: the flat SHG signal thus suggests a low roughness and a good uniformity compared to the image resolution of ∼ 0.5 µm. The variation of the square root of the generated intensity detected by the photomultiplier tube is indicated in Fig. 4(b), both in forward (trans) and backward (epi) directions, as a function of the power (${\wp _{in}}$) of the incident laser beam. The linear behavior observed emphasizes the second order origin of this process, evidencing the second harmonic generation in the TTB-Eu film studied. Notably, since the coherence length in the backward direction (30 nm) is small compared to the film thickness (300 nm), the signal measured in the epi-direction corresponds mostly to back reflection of the forward emitted signal.

 figure: Fig. 4.

Fig. 4. (a) SHG microscopy image of the TTB-Eu thin film (forward direction). (b) Square root of the detected SHG intensity as a function of the excitation power (forward and backward). The linear behavior shows a second order process.

Download Full Size | PDF

3.3 Ratios of the independent nonlinear susceptibility components

Due to their tetragonal structure of the TTB-Eu films and their ${\textrm{C}_{4\textrm{V}}}$ symmetry [33], the only non-zero components of the nonlinear tensor of the studied films are ${d_{15}},\; {d_{24}},\; {d_{31}},\; {d_{32}}$ and ${d_{33}}$, with ${d_{15}} = {d_{24}}$ and ${d_{31}} = {d_{32}}$ [34]. Thus, the reduced tensor of TTB-Eu material is given by:

$$d = \left[ {\begin{array}{cccccc} 0&0&0&0&{{d_{15}}}&0\\ 0&0&0&{{d_{15}}}&0&0\\ {{d_{32}}}&{{d_{32}}}&{{d_{33}}}&0&0&0 \end{array}} \right]$$
Due to the fact that the electric field in our experiment is perpendicular to the propagation direction (z axis), the nonlinear polarization along the z axis $P_z^{(2 )}$ is inaccessible. The other induced nonlinear polarization can still be detected and are related to the one induced in the sample frame (see Eq. (7) in the Appendix), themselves dependent on the 2nd order nonlinear tensor elements, such that the following equations are obtained (the plane wave approximation being ensured by the loose focusing):
$$\left[ {\begin{array}{c} {P_x^{(2 )}}\\ {P_y^{(2 )}}\\ {P_z^{(2 )}} \end{array}} \right] \propto \left[ {\begin{array}{l} {{d_{15}}\; {C_\varphi }{S_\varphi }{S_\theta }}\\ {[{({{d_{15}} + {d_{32}}} )C_\theta^2S_\varphi^2 + {d_{32}}C_\varphi^2 + {d_{33}}S_\theta^2S_\varphi^2} ]{S_\theta }}\\ {[{d_{32}}({C_\varphi^2 + C_\theta^2S_\varphi^2} )+ {d_{33}}S_\theta^2S_\varphi^2 + {d_{15}}S_\varphi^2S_\theta^2]C_\theta} \end{array}} \right]$$
where Ci and Si denote respectively the contracted notations for cosine and for sine of the angle i (i =θ, φ) as described in Fig. 1. Two configurations of incident and detected polarizations are thus sufficient to extract the different tensor components.

Detecting the signal with the analyzer on the y-axis (horizontal), the polarization orientation was rotated while keeping the sample tilted at an angle ${{\theta }_0}$ = 32°, a good compromise for the value of sin(${\theta }$), hence the SHG signal, to be sufficient, while avoiding too high Fresnel reflection. The SHG detected intensity $I_y ({{\varphi },\; {\theta_0}} )\propto {({P_y^{(2 )}({{\varphi },\; {\theta_0}} )} )^2}$ is written as:

$$I_y ({{\varphi },\; {\theta_0}} )\propto {\left[ { \left( {C_\varphi^2 + \left( {1 + \frac{{{d_{15}}}}{{{d_{32}}}}} \right)C_{{\theta_0}}^2S_\varphi^2 + \frac{{{d_{33}}}}{{{d_{32}}}}S_{{\theta_0}}^2S_\varphi^2} \right){S_{{\theta_0}}}} \right]^2}$$
Another set of measurements was performed as well, keeping the polarization horizontal (p-polarization, ${\varphi _0} = 90^\circ $), and rotating the sample plane from $\theta $ = 0° to 56° so that:
$$I_{y} ({{\varphi_{0}} = 90^\circ ,\; \theta } )\propto {\left[ {\left( {\left( {1 + \frac{{{d_{15}}}}{{{d_{32}}}}} \right)C_\theta^2 + \frac{{{d_{33}}}}{{{d_{32}}}}S_{{\theta}}^2} \right){S_\theta }} \right]^2}$$
The results of these two sets are displayed on Fig. 5, where the black squares are the intensity measurements corrected for the variation of the reflection coefficient due to polarization rotation (φ) or tilting of the sample (θ). It is worth noting that any change in φ or θ will change the 3 reflection coefficients at the different interfaces: the interface air/TTB-Eu seen by the fundamental beam, the TTB-Eu/MgO and the MgO/air ones seen by the generated second-harmonic signal. When θ is varied, the incident polarization is p-polarized and the associated Fresnel coefficient [35] in reflection for one interface between medium 1 and 2 is calculated with Eq. (8) in the Appendix.

 figure: Fig. 5.

Fig. 5. (a) Measured SHG counts as a function of the angle θ (the angle between the plane of the sample and the XY plane as explained in the Fig. 1) for a p-polarization. (b) Measured SHG counts as a function of the direction of the polarization with respect to the vertical axis (X). Experimental points are black symbols, and fitting curves are in red.

Download Full Size | PDF

The measurement as a function of φ is corrected considering a varying sum of p-polarization and s-polarization at a fixed incidence angle θ0. The respective nonlinear fitting are represented by the red curves using Eq. (3) and Eq. (4), and are used to find the ratio ${d_{33}}/{d_{32}}$ = 0.13 ± 0.05 and ${d_{15}}/{d_{32}}$ = 2.6 ± 0.5. The energy band gap of TTB-Eu is about 4.35eV corresponding to a light wavelength of 285 nm, which is far enough from the SHG wavelength (405 nm) to ensure that resonant effects will not affect the nonlinear properties, such that the Kleinman condition for the permutation of indices still applies [36].

3.4. Absolute determination of the nonlinear susceptibility

The absolute value of one component of the nonlinear tensor can be extracted by comparing the SHG signals in the sample and in a 350 µm thick y-cut quartz plate under the same conditions of excitation. Again, a low focusing lens was used so that the plane-wave approximation is valid, especially for the phase-matching factor that allows to write that the SHG power generated is:

$$\wp _{2\omega }^X({\varphi = 0^{\circ },\; {\theta_0}} )= \frac{{K\wp _\omega ^{2}}}{{\varepsilon _0 c\; n_{2\omega } n_\omega ^2\lambda^2w ^2}}d_{eff}^{2}{\left( {L\; sinc\left( {\frac{L}{{{L_c}}}} \right)} \right)^2}$$
where $\wp _{\omega } $ is the average excitation power, c is the speed of light, ε0 is the vacuum permittivity, λ is the fundamental wavelength, w is the surface of focalization, $\textrm{d}_{\textrm{eff}}$ is the effective nonlinear susceptibility (depending on the geometry of excitation), $n_{\omega } $ (resp. $n_{2{\omega }} $) is the refractive indices at fundamental wavelength (resp. at the SHG wavelength), L is the sample thickness, Lc is the coherence length and K denotes a constant that contains the pulse parameters. The refractive index of the TTB-Eu thin film is measured by an optical refractometer at both fundamental and SHG wavelengths, and the measured indices are reported in Table 1. It is to be noted that the superscripts S, M and q stand for the sample under investigation (i.e. the TTB-Eu film), the MgO substrate and the quartz plate, respectively. They are in good agreement with the values found in the literature for similar structures [3639].

Tables Icon

Table 1. Ordinary refractive index [39,40] and coherence lengths for the TTB-Eu thin film, MgO substrate and reference quartz (S, M and q superscript, respectively), along with transmission coefficient of the sample (at θ = 32°).

The coherence length is defined as ${L_\textrm{c}} = \frac{\lambda }{{\pi ({ n_{2\omega } - n_\omega } )}}$. The detected SHG power has to be corrected for the Fresnel reflections (coefficient $ {\textrm{T}^\textrm{X}}$). The nonlinear coefficient d32 can then be extracted (see Eq. (9) in the Appendix for details):

$$ d_{32}^{S} = \frac{{ d_{11}^{q}}}{{\textrm{sin}\theta }}\sqrt {\frac{{ n_{2\omega }^{S}{{(n_\omega ^S)}^2}}}{{n_{2\omega }^{q}{{(n_\omega ^q)}^2}}}\frac{{{T^q}}}{{{T^S}}}\frac{{\wp _{2\omega }^{S}}}{{\wp _{2\omega }^{q}}}{{\left( {\frac{{L^q\textrm{sinc}({{L^q}/L_c^q} )}}{{L^S\textrm{sinc}(L_{\theta}^S/L_c^S)}}} \right)}^2} } $$
where $\textrm{L}_{\theta }^\textrm{S}$ is the effective thickness of the TTB-Eu thin film at the angle θ. Finally, using the ratios ${\textrm{d}_{15}}/{\textrm{d}_{32}}$ and ${\textrm{d}_{33}}/{\textrm{d}_{32}}$ previously measured, all the independent components of the nonlinear susceptibility tensor can be measured (in pm/V): ${\textrm{d}_{32}} = 0.084 \pm 0.005$, ${\textrm{d}_{15}} = 0.22 \pm 0.5$ and ${\textrm{d}_{33}} = 0.011 \pm 0.005$, as listed in Table 2.

3.5. Verification of the coefficients

Table 3 shows how the different SHG polarization can be used at a certain excitation polarization to directly measure the nonlinear susceptibility tensor components. Thus ${\textrm{d}_{15}}$ can be directly measured from the polarization in the x-direction, under a polarization of 45° from the x axis (and y axis). The calculation is the same as for Eq. (6) if $\textrm{sin}({\theta } )$ is replaced by $\sin (\theta )/2$. The ${\textrm{d}_{33}}$ component however relies on the determination of the two others: its detailed calculation can be found in the Appendix Eqs. (10) and (11).

Tables Icon

Table 2. Nonlinear susceptibility components of TTB-Eu thin film: while d15 and d33 are determined by curve fitting, the d32 is calculated using Eq. (6).

Tables Icon

Table 3. Induced SHG polarization in specific directions, excited by a specific polarization (first line), and the component of the nonlinear tensor it allows to retrieve

The nonlinear components verification is reported in Table 4. The value of ${\textrm{d}_{33}}$ is roughly the same as measured before. Here, ${\textrm{d}_{15}}$ is a bit higher but the difference lies within the uncertainty of the measurement. Finally, the components of the nonlinear tensor determined here are reduced compared to what is reported in literature for other TTBs [41]. This difference is mainly explained by the fact that the material studied in [41] is a single crystal while the material studied in this report is an epitaxial films having different growth domains or crystallographic variants. Thus, the dij components of TTB-Eu films are in fact representing the contributions of a set of single crystalline domains (small crystals) that are composing these epitaxial films, whose distribution we do not know neither control. Even though the components of the nonlinear tensor of the TTB-Eu epitaxial films are small, this study shows that thin films with TTB structure possess room temperature nonlinear optical properties in addition of their ferroelectric properties. On the other hand, it has been shown that ceramics and films based on TTB-Eu structure can be multiferroic due to the spontaneous formation of barium hexaferrite as a magnetic secondary phase [14]. Thus, TTB-Eu films present different functional properties simultaneously at room temperature, and the coupling between the different properties of this room temperature multifunctional material allows for many ways to control and modulate its non-linear optical properties.

Tables Icon

Table 4. Verification of the nonlinear susceptibility components of TTB-Eu thin film by direct measurement.

4. Conclusion

Tetragonal tungsten bronze Ba2EuFeNb4O15 thin films were deposited on MgO(100) substrates by PLD. XRD results confirmed the highly oriented epitaxial growth of the TTB-Eu films with four different in-plane orientation, for which the TTB-Eu unit cell is rotated in plane by ${\pm} $ 18.4° and ${\pm} $ 31° with respect to the unit cell of the substrate MgO crystal. PFM results showed stable microelectromechanical response confirming the ferroelectric nature of the epitaxial TTB-Eu thin films. These films were shown to exhibit a SHG signal, and their nonlinear susceptibility tensor components were measured using polarization-resolved SHG by two different approaches. The absolute values of the three independent components of the nonlinear optical susceptibility tensor d15, d32 and d33 were accessed using a SHG measurement in the same conditions in a reference quartz plate: the obtained tensor values showed good agreement for both cases. This study furthers the characterization of TTB-Eu thin films, showing their room temperature nonlinear optical properties in addition of their multiferroic properties previously demonstrated, whose coupling provides new ways to control and modulate its non-linear optical properties.

Appendix

A.1 PFM measurement

During the investigation of the microelectromechanical properties, the out-of-plane and the in-plane ferroelectric polarization at the films surface was manipulated by applying an alternating voltage of 1V at 20 kHz between the conductive tip and the Pt-coated MgO substrate. The surface-induced piezoelectric vibrations were detected using a lock-in amplifier from Signal Recovery (model 7265, Wokingham, UK).

A.2 SHG microscopy

In order to verify the uniformity of the SHG signal over the thin films, SHG microcopy images were acquired. Ti:Sapph laser (150 fs pulses duration, wavelength of 810nm, 80MHz repetition rate) is used to excite the sample, with an average power kept under 0.8 W to avoid thermal degradation (pulse energy below 10 nJ). The sample is placed on a custom laser-scanning multiphoton microscope and illuminated by an air immersion objective (20X, 0.75NA, Olympus), and the SHG signal is detected on photomultiplier tube (R6357, Hamamatsu) set at 700V. The exposure time was set to 12.5 µs and 10 images were recorded to perform averaging. For a complete description of the set-up see [42,43].

A.3 Equation details

The second-order nonlinear light polarization at the detector is related to the one in the sample frame by (see Fig. 1):

$$\begin{array}{c} {P_x^{(2 )} = P_{X^{\prime}}^{(2 )}}\\ {P_y^{(2 )} = {C_\theta }P{{_{Y^{\prime}}^{(2 )}}} - {S_\theta }P_{Z^{\prime}}^{(2 )}}\\ {P_z^{(2 )} = {C_\theta }P{{_{Z^{\prime}}^{(2 )}}} + {S_\theta }P_{Y^{\prime}}^{(2 )}} \end{array}$$
The Fresnel reflection coefficient under p-polarization is calculated as [35]:
$${R_P} = {\left|{{n_1}\sqrt {1 - {{\left( {\frac{{{n_1}}}{{{n_2}}}\textrm{sin}\theta } \right)}^2}} - {n_2}\textrm{cos}\theta /{n_1}\sqrt {1 - {{\left( {\frac{{{n_1}}}{{{n_2}}}\textrm{sin}\theta } \right)}^2}} + {n_2}\textrm{cos}\theta } \right|^2}$$
where ni are the optical index of medium i (i = 1, 2).

Combining Eq. (5) for the quartz and TTB-Eu samples:

$$\frac{{\wp _{2\omega }^S({\varphi = 0{^\circ}, \theta } )}}{{\wp _{2\omega }^q({\varphi = 0{^\circ}, {\theta}} )}} = \frac{{{T^S}}}{{{T^q}}}{\left( {\frac{{d_{32}^{S}\textrm{sin}\theta }}{{d_{11}^{q}}}} \right)^2}\frac{{n_{2\omega }^{q}{{(n_\omega ^q)}^2}}}{{n_{2\omega }^{S}{{(n_\omega ^S)}^2}}}{\left( {\frac{{L^S\textrm{sinc}(L_\theta^S/L_c^S) }}{{L^q\textrm{sinc}({{L^q}/L_c^q} )}}} \right)^2} $$
The ${\textrm{d}_{33}}$ component is extracted from ${P_y}({\varphi = 90^\circ ,{\theta }} )$ such as:
$$d_{\textrm{eff}} = ({d_{15}} + {d_{32}}){S_\theta }C_\theta ^2 + {d_{33}}S_\theta ^3\mathop \Leftrightarrow \limits \; d_{33} = \frac{1}{{\textrm{si}{\textrm{n}^3}{\theta }}}[{{d_{\textrm{eff}}} - ({d_{15}}\textrm{ + }{d_{32}})\textrm{sin}{\theta} {\textrm{cos}^2}{\theta }} ]$$
Now ${d_{\textrm{eff}}}$ is calculated with Eq. (5), and ${\textrm{d}_{33}}$ is obtained as:
$$d_{33} = \frac{1}{{si{n^3}\theta }}\left[ {d_{11}^{q}\sqrt {\frac{{n_{2\omega }^{S}{{ {(n_\omega^S} )}^2}}}{{n_{2\omega }^{q}{{ {(n_\omega^q} )}^2}}}\frac{{{T^q}}}{{{T^S}}}\frac{{\wp_{2\omega }^{S}}}{{\wp_{2\omega }^{q}}}{{\left( {\frac{{L^{q}sinc({{L^q}/L_c^q} )}}{{L^{S}sinc(L_\theta^S/ {L_c^S} )}}} \right)}^2} } - ({d_{15}}\textrm{ + }{d_{32}})sin\theta co{s^2}\theta } \right]$$

Funding

Canada Foundation for Innovation; Natural Sciences and Engineering Research Council of Canada; Fonds de recherche du Québec – Nature et technologies; Natural Sciences and Engineering Research Council of Canada (RGPIN 261662).

Disclosures

The authors declare no conflicts of interest.

References

1. S. A. Denev, T. T. A. Lummen, E. Barnes, A. Kumar, and V. Gopalan, “Probing ferroelectrics using optical second harmonic generation,” J. Am. Ceram. Soc. 94(9), 2699–2727 (2011). [CrossRef]  

2. P. Castera, D. Tulli, A. M. Gutierrez, and P. Sanchis, “Influence of BaTiO3 ferroelectric orientation for electro-optic modulation on silicon,” Opt. Express 23(12), 15332 (2015). [CrossRef]  

3. W. J. Hu, Z. Wang, W. Yu, and T. Wu, “Optically controlled electroresistance and electrically controlled photovoltage in ferroelectric tunnel junctions,” Nat. Commun. 7(1), 1–9 (2016). [CrossRef]  

4. M. İlhan, M. K. Ekmekçi, A. Mergen, and C. Yaman, “Synthesis and Optical Characterization of Red-Emitting BaTa2O6:Eu3+ Phosphors,” J. Fluoresc. 26(5), 1671–1678 (2016). [CrossRef]  

5. A. Guarino, G. Poberaj, D. Rezzonico, R. Degl’Innocenti, and P. Günter, “Electro-optically tunable microring resonators in lithium niobate,” Nat. Photonics 1(7), 407–410 (2007). [CrossRef]  

6. T. Miyamoto, H. Yada, H. Yamakawa, and H. Okamoto, “Ultrafast modulation of polarization amplitude by terahertz fields in electronic-type organic ferroelectrics,” Nat. Commun. 4(1), 1–9 (2013). [CrossRef]  

7. J. Li, B. Nagaraj, H. Liang, W. Cao, C. H. Lee, and R. Ramesh, “Ultrafast polarization switching in thin-film ferroelectrics,” Appl. Phys. Lett. 84(7), 1174–1176 (2004). [CrossRef]  

8. X. Chen, P. Karpinski, V. Shvedov, K. Koynov, B. Wang, J. Trull, C. Cojocaru, W. Krolikowski, and Y. Sheng, “Ferroelectric domain engineering by focused infrared femtosecond pulses,” Appl. Phys. Lett. 107(14), 1–4 (2015). [CrossRef]  

9. Y. Zhang, J. Martinez-Perdiguero, U. Baumeister, C. Walker, J. Etxebarria, M. Prehm, J. Ortega, C. Tschierske, M. J. O’Callaghan, A. Harant, and M. Handschy, “Laterally azo-bridged H-shaped ferroelectric dimesogens for second-order nonlinear optics: Ferroelectricity and second harmonic generation,” J. Am. Chem. Soc. 131(51), 18386–18392 (2009). [CrossRef]  

10. K. Kintaka, M. Fujirnura, T. Suhara, and H. Nishihara, “High-Efficiency LiNbO3 Waveguide Second-Harmonic Generation Devices with Ferroelectric-Domain-Inverted Gratings Fabricate by Applying Voltage,” J. Lightwave Technol. 14(3), 462–468 (1996). [CrossRef]  

11. E. O. Chi, A. Gandini, K. M. Ok, L. Zhang, and P. S. Halasyamani, “Syntheses, Structures, Second-Harmonic Generating, and Ferroelectric Properties of Tungsten Bronzes: A6M2M′8O30 (A) Sr2+, Ba2+, or Pb2+ M) Ti4+, Zr4+, or Hf4+ M′) Nb5+or Ta5+),” Chem. Mater. 16(19), 3616–3622 (2004). [CrossRef]  

12. P. V. Lenzo, E. G. Spencer, and A. A. Ballman, “Electro-optic coefficients of ferroelectric strontium barium niobate,” Appl. Phys. Lett. 11(1), 23–24 (1967). [CrossRef]  

13. I. Levin, M. C. Stennett, G. C. Miles, D. I. Woodward, A. R. West, and I. M. Reaney, “Coupling between octahedral tilting and ferroelectric order in tetragonal tungsten bronze-structured dielectrics,” Appl. Phys. Lett. 89(12), 2–5 (2006). [CrossRef]  

14. M. Josse, O. Bidault, F. Roulland, E. Castel, A. Simon, D. Michau, R. Von der Muhll, O. Nguyen, and M. Maglione, “The Ba2LnFeNb4O5 “tetragonal tungsten bronze”: Towards RT composite multiferroics,” Solid State Sci. 11(6), 1118–1123 (2009). [CrossRef]  

15. H. Kalhori, M. Coey, I. Abdolhosseini Sarsari, K. Borisov, S. B. Porter, G. Atcheson, M. Ranjbar, H. Salamati, and P. Stamenov, “Oxygen Vacancy in WO3 Film-based FET with Ionic Liquid Gating,” Sci. Rep. 7(1), 12253 (2017). [CrossRef]  

16. S. Alkoy, C. Duran, and D. A. Hall, “Electrical properties of textured potassium strontium niobate (KSr2Nb5O15) ceramics fabricated by reactive templated grain growth,” J. Am. Ceram. Soc. 91(5), 1597–1602 (2008). [CrossRef]  

17. T. Hajlaoui, C. Harnagea, and A. Pignolet, “Influence of lanthanide ions on multiferroic properties of Ba2LnFeNb4O15 (Ln = Eu3+, Sm3+ and Nd3+) thin films grown on silicon by pulsed laser deposition,” Mater. Lett. 198, 136–139 (2017). [CrossRef]  

18. T. Hajlaoui, L. Corbellini, C. Harnagea, M. Josse, and A. Pignolet, “Enhanced ferroelectric properties in multiferroic epitaxial Ba2EuFeNb4O15 thin films grown by pulsed laser deposition,” Mater. Res. Bull. 87, 186–192 (2017). [CrossRef]  

19. T. Hajlaoui, C. Harnagea, D. Michau, M. Josse, and A. Pignolet, “Highly oriented multiferroic Ba2LnFeNb4O5-based composite thin films with tetragonal tungsten bronze structure on silicon substrates,” J. Alloys Compd. 711, 480–487 (2017). [CrossRef]  

20. K. M. Ok, E. O. Chi, and P. S. Halasyamani, “Bulk characterization methods for non-centrosymmetric materials: Second-harmonic generation, piezoelectricity, pyroelectricity, and ferroelectricity,” Chem. Soc. Rev. 35(8), 710–717 (2006). [CrossRef]  

21. R. W. Boyd, Nonlinear Optics (Academic Press, 2008).

22. Y. Uesu, S. Kurimura, and Y. Yamamoto, “Optical second harmonic images of 90° domain structure in BaTiO3 and periodically inverted antiparallel domains in LiTaO3,” Appl. Phys. Lett. 66(17), 2165–2167 (1995). [CrossRef]  

23. K. J. Choi, M. Biegalski, Y. L. Li, A. Sharan, J. Schubert, R. Uecker, P. Reiche, Y. B. Chen, X. Q. Pan, V. Gopalan, L. Q. Che, D. C. Schlom, and C. B. Eom, “Enhancement of ferroelectricity in strained BaTiO3 thin films,” Science 306(5698), 1005–1009 (2004). [CrossRef]  

24. F. Roulland, M. Josse, E. Castel, and M. Maglione, “Influence of ceramic process and Eu content on the composite multiferroic properties of the Ba6-2xLn2xFe1 + xNb9-xO30 TTB system,” Solid State Sci. 11(9), 1709–1716 (2009). [CrossRef]  

25. M. Sharma, S. C. Kashyap, H. C. Gupta, M. C. Dimri, and K. Asokan, “Enhancement of Curie temperature of barium hexaferrite by dense electronic excitations,” AIP Adv. 4(7), 077129 (2014). [CrossRef]  

26. D. Chen, I. Harward, J. Baptist, S. Goldman, and Z. Celinski, “Curie temperature and magnetic properties of aluminum doped barium ferrite particles prepared by ball mill method,” J. Magn. Magn. Mater. 395, 350–353 (2015). [CrossRef]  

27. T. Hajlaoui, C. Chabanier, C. Harnagea, and A. Pignolet, “Epitaxial Ba2LnFeNb4O15-based multiferroic nanocomposite thin films with tetragonal tungsten bronze structure,” Scr. Mater. 136, 1–5 (2017). [CrossRef]  

28. T. Hajlaoui, N. Émond, C. Quirouette, B. Le Drogoff, J. Margot, and M. Chaker, “Metal–insulator transition temperature of boron-doped VO2 thin films grown by reactive pulsed laser deposition,” Scr. Mater. 177, 32–37 (2020). [CrossRef]  

29. T. Hajlaoui, “Synthèse et étude des couches minces composites multiferroïques à base des bronzes de tungstène quadratiques,” Université du Québec (2017).

30. M. Gu, “Advanced optical imaging theory,” Springer Sci. Bus. Media. 75 (2000).

31. D. Damjanovic, “Ferroelectric, dielectric and piezoelectric properties of ferroelectric thin films and ceramics,” Rep. Prog. Phys. 61(9), 1267–1324 (1998). [CrossRef]  

32. Y. Xiao, V. B. Shenoy, and K. Bhattacharya, “Depletion layers and domain walls in semiconducting ferroelectric thin films,” Phys. Rev. Lett. 95(24), 247603 (2005). [CrossRef]  

33. E. Castel, M. Josse, D. Michau, and M. Maglione, “Flexible relaxor materials: Ba2PrxNd1−xFeNb4O15 tetragonal tungsten bronze solid solution,” J. Phys.: Condens. Matter 21(45), 452201 (2009). [CrossRef]  

34. R. L. Sutherland, “Elements of the Theory of Nonlinear Optics,” in Handbook of Nonlinear Optics, 2 Ed. (Marcel Dekker, Inc, 2003), p. 963.

35. M. Born and E. Wolf, “Basic Properties of the Electromagnetic Field,” in Principles of Optics (Pergamon Press, 1980), pp. 38–41.

36. M. Eßer, M. Burianek, P. Held, J. Stade, S. Bulut, C. Wickleder, and M. Mühlberg, “Optical characterization and crystal structure of the novel bronze type CaxBa1−xNb2O6 (x = 0.28; CBN-28),” Cryst. Res. Technol. 38(6), 457–464 (2003). [CrossRef]  

37. G. Ghosh, “Dispersion-equation coefficients for the refractive index and birefringence of calcite and quartz crystals,” Opt. Commun. 163(1-3), 95–102 (1999). [CrossRef]  

38. C. W. Teng, J. F. Muth, U. Ozgur, M. J. Bergmann, H. O. Everitt, A. K. Sharma, C. Jin, and J. Narayan, “Refractive indices and absorption coefficients of Mg x Zn 1− x O alloys,” Appl. Phys. Lett. 76(8), 979–981 (2000). [CrossRef]  

39. R. E. Stephens and I. H. Malitson, “Index of refraction of magnesium oxide,” J. Res. Natl. Bur. Stand. (1934). 49, 2360 (1952).

40. C. Z. Tan, “Determination of refractive index of silica glass for infrared wavelengths by IR spectroscopy,” J. Non-Cryst. Solids 223(1-2), 158–163 (1998). [CrossRef]  

41. J. J. E. Reid, “Resonantly enhanced, frequency doubling of an 820 nm GaAlAs diode laser in a potassium lithium niobate crystal,” Appl. Phys. Lett. 62(1), 19–21 (1993). [CrossRef]  

42. M.-A. Houle, Microscopie à Génération de Seconde Harmonique Appliquée à l’imagerie de l’ostéo-Arthrite (Université du Québec, Institut national de la recherche scientifique., 2013).

43. M. A. Houle, C. A. Couture, S. Bancelin, J. Van der Kolk, E. Auger, C. Brown, K. Popov, L. Ramunno, and F. Légaré, “Analysis of forward and backward Second Harmonic Generation images to probe the nanoscale structure of collagen within bone and cartilage,” J. Biophotonics 8(11-12), 993–1001 (2015). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (5)

Fig. 1.
Fig. 1. Schematic of the experimental set-up. λ/2: half-wave plate, L: +5 cm lens, F: 405 ± 10 nm filter, A: analyser, S: sample (TTB-Eu).
Fig. 2.
Fig. 2. (a) X-ray diffractogram obtained in Bragg-Brentano geometry demonstrating the synthesis of TTB-Eu films oriented with their c-axis normal to the substrate surface.
Fig. 3.
Fig. 3. (a) Longitudinal piezoelectric coefficient (ξZZ), and (b) transverse piezoelectric coefficient (ξXZ) versus an applied voltage. The in-plane piezoelectric coefficient ξXZ represents roughly 15% of the out-of-plane piezoelectric coefficient ξZZ.
Fig. 4.
Fig. 4. (a) SHG microscopy image of the TTB-Eu thin film (forward direction). (b) Square root of the detected SHG intensity as a function of the excitation power (forward and backward). The linear behavior shows a second order process.
Fig. 5.
Fig. 5. (a) Measured SHG counts as a function of the angle θ (the angle between the plane of the sample and the XY plane as explained in the Fig. 1) for a p-polarization. (b) Measured SHG counts as a function of the direction of the polarization with respect to the vertical axis (X). Experimental points are black symbols, and fitting curves are in red.

Tables (4)

Tables Icon

Table 1. Ordinary refractive index [39,40] and coherence lengths for the TTB-Eu thin film, MgO substrate and reference quartz (S, M and q superscript, respectively), along with transmission coefficient of the sample (at θ = 32°).

Tables Icon

Table 2. Nonlinear susceptibility components of TTB-Eu thin film: while d15 and d33 are determined by curve fitting, the d32 is calculated using Eq. (6).

Tables Icon

Table 3. Induced SHG polarization in specific directions, excited by a specific polarization (first line), and the component of the nonlinear tensor it allows to retrieve

Tables Icon

Table 4. Verification of the nonlinear susceptibility components of TTB-Eu thin film by direct measurement.

Equations (11)

Equations on this page are rendered with MathJax. Learn more.

d = [ 0 0 0 0 d 15 0 0 0 0 d 15 0 0 d 32 d 32 d 33 0 0 0 ]
[ P x ( 2 ) P y ( 2 ) P z ( 2 ) ] [ d 15 C φ S φ S θ [ ( d 15 + d 32 ) C θ 2 S φ 2 + d 32 C φ 2 + d 33 S θ 2 S φ 2 ] S θ [ d 32 ( C φ 2 + C θ 2 S φ 2 ) + d 33 S θ 2 S φ 2 + d 15 S φ 2 S θ 2 ] C θ ]
I y ( φ , θ 0 ) [ ( C φ 2 + ( 1 + d 15 d 32 ) C θ 0 2 S φ 2 + d 33 d 32 S θ 0 2 S φ 2 ) S θ 0 ] 2
I y ( φ 0 = 90 , θ ) [ ( ( 1 + d 15 d 32 ) C θ 2 + d 33 d 32 S θ 2 ) S θ ] 2
2 ω X ( φ = 0 , θ 0 ) = K ω 2 ε 0 c n 2 ω n ω 2 λ 2 w 2 d e f f 2 ( L s i n c ( L L c ) ) 2
d 32 S = d 11 q sin θ n 2 ω S ( n ω S ) 2 n 2 ω q ( n ω q ) 2 T q T S 2 ω S 2 ω q ( L q sinc ( L q / L c q ) L S sinc ( L θ S / L c S ) ) 2
P x ( 2 ) = P X ( 2 ) P y ( 2 ) = C θ P Y ( 2 ) S θ P Z ( 2 ) P z ( 2 ) = C θ P Z ( 2 ) + S θ P Y ( 2 )
R P = | n 1 1 ( n 1 n 2 sin θ ) 2 n 2 cos θ / n 1 1 ( n 1 n 2 sin θ ) 2 + n 2 cos θ | 2
2 ω S ( φ = 0 , θ ) 2 ω q ( φ = 0 , θ ) = T S T q ( d 32 S sin θ d 11 q ) 2 n 2 ω q ( n ω q ) 2 n 2 ω S ( n ω S ) 2 ( L S sinc ( L θ S / L c S ) L q sinc ( L q / L c q ) ) 2
d eff = ( d 15 + d 32 ) S θ C θ 2 + d 33 S θ 3 d 33 = 1 si n 3 θ [ d eff ( d 15  +  d 32 ) sin θ cos 2 θ ]
d 33 = 1 s i n 3 θ [ d 11 q n 2 ω S ( n ω S ) 2 n 2 ω q ( n ω q ) 2 T q T S 2 ω S 2 ω q ( L q s i n c ( L q / L c q ) L S s i n c ( L θ S / L c S ) ) 2 ( d 15  +  d 32 ) s i n θ c o s 2 θ ]
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.