Er3+/Tm3+-codoped LiNbO3 crystal was prepared by co-diffusion of stacked Er and Tm metal films coated onto surface of off-congruent, Li-deficient LiNbO3 substrate produced by Li-poor vapor transport equilibration technique. The crystalline phase on the diffused surface was analyzed by X-ray single-crystal diffraction. The Er3+ and Tm3+ profile characteristics were studied by secondary ion mass spectrometry. The emission spectra were measured under the 980 or 795 nm wavelength excitation, and the emission and absorption cross section spectra were calculated based upon McCumber theory. The lifetimes of some emissions were measured. The results show that the Er3+ and Tm3+ ions presence is in the form of LiNbO3 phase. Both ions obey to Gaussian profile with a diffusion depth 21.5 μm. In the codoping case, both ions keep their respective spectroscopic features of only doping case and do not affect each other. The codoping enables to combine the wavelength emissions of both ions and the resultant emission band in the telecommunication window around 1.5 μm is as wide as 150 nm, providing the possibility of S+C+L broadband amplification by employing commercial 980 and 795 nm laser diodes as the pump sources. The Er3+/Tm3+-codoped LN is a promising host material for integrated optics.Er3+/Tm3+-codoped LiNbO3 crystal was prepared by co-diffusion of stacked Er and Tm metal films. The crystalline phase, diffusion profile and cross section spectra of Er3+ and Tm3+ ions in the diffusion layer have been investigated. The results show that the presence of Er3+ and Tm3+ ions is in the LiNbO3 phase. Both ions follow Gaussian profile with a diffusion depth 21.5 μm. Both ions keep their respective spectroscopic features of only doping case. Excited state absorption is the dominant process for 795-nm-upconvered fluorescence of Tm3+. The codoping enables to combine the wavelength emissions of both ions and provide the possibility of S+C+L broadband amplification. Er3+/Tm3+-codoped LiNbO3 is a promising material for integrated optics.

•Cylindrical hole must be deep enough and a shallow waveguide is required.•Increasing hole radius causes blueshift, broadening and edge sharpening of band.•Non-cylindrical hole seriously affects gap, location and contrast of stop-band.•For cylindrical-conical hybrid hole, cylindrical part determines desired features.•A scheme of milling holes at bottom of a trench on waveguide surface is proposed.Effects of finite hole depth and non-cylindrical hole shape on stop-band characteristics of photonic crystal formed by air-hole square lattice in Ti-diffused LiNbO3 strip waveguide were studied theoretically. The study shows that hole depth determines the contrast of stop-band, and the hole radius and conical angle determine the bandgap and location. Cylindrical holes must be sufficiently deep so as to overlap most of waveguide mode and hence obtain a stop-band with high contrast, sharp edge and broad bandgap. Non-cylindrical holes seriously affect the stop-band features. Conical holes cause low contrast and narrow bandgap, and the stop-band shifts with the conical angle. For the cylindrical-conical hybrid holes, the cylindrical portion determines the desired features. Given the difficulty in fabricating high aspect-ratio cylindrical holes, we propose to fabricate the holes at the bottom of a shallow trench, which is introduced into waveguide surface prior to the hole milling.

•A simple, compact, low-cost, highly efficient thermometer is developed.•It is based on ratiometric fluorescence technique and Er/Yb:NaYF4 microcrystals.•It consists of an LD, two lenses, one filter, one Si-photocell and one multimeter.•The sensor shows good signal-to-noise ratio, and high resolution and sensitivity.•The thermometer shows great potential for use in biological tissues and cells.We developed a highly efficient optical thermometer based on intensity ratio of upconversion green fluorescence of Er3 +/Yb3 +-codoped NaYF4 microcrystals. The sensor consists simply of a 980 nm laser diode, one narrow-band interference filter, two lenses, one Si-photocell and one multimeter, while being without use of spectrometer and additional electronics. The device not only has a simple, compact structure (hence a low cost), but also displays highly efficient sensing performance, characterized by large signal-to-noise ratio due to strong fluorescence intensity, high thermal resolution and sensitivity, which have the values 1.3 K and 1.24 × 10− 2 K− 1, respectively, at the physiological temperature 310 K. The excellent sensing performance of the device was further confirmed by the results of the measurements repeated using a spectrometer. The thermometer is highly generalized that can be applied to other luminescent materials, and shows great potential for the physiological temperature sensing in biological tissues and cells.We have developed a thermometer based on the ratiometric fluorescence technique and Er3 +/Yb3 +-codoped NaYF4 microcrystals. The device not only has a simple, compact structure (hence a low cost), but also displays highly efficient sensing performance, which is characterized by strong fluorescence intensity (hence large signal-to-noise ratio), and high thermal resolution (1.3 K) and relative sensitivity (1.24 × 10− 2 K− 1) at the physiological temperature 310 K. The excellent sensing performance of the device is confirmed by the results of the measurements by a spectrometer. The thermometer is highly generalized that can be applied to other luminescent materials, and shows great potential for the physiological temperature sensing in biological tissues and cells.Download high-res image (426KB)Download full-size image

•A simple, low-cost, highly efficient temperature sensor is developed.•It is based on FIR of down-conversion green fluorescence of Er:SrGdGa3O7 crystal.•It consists of an LED, two optical filters, two Si-photocells and two multimeters.•The sensor shows good signal-to-noise ratio and high temperature sensitivity.•The device may find its use in some harsh environments and industrial processes.We have developed a simple, low-cost and highly efficient optical temperature sensor based on intensity ratio of down-conversion green fluorescence of Er3+ ions doped in SrGdGa3O7 single crystal. The sensor consists simply of a 384 nm light emitting diode, two narrow-band interference filters, two Si-photocells and two multimeters, without use of a monochromator or spectrometer or focusing lens. It has a simple, compact structure and is low-cost. The experimental results show that the device displays highly efficient sensing performance with good signal-to-noise ratio and high thermal resolution and temperature sensitivity, which has a value of (5.5-102.1) × 10−3 K−1 in the considered temperature range of 100–430 K. In addition, the experiments were repeated using a spectrometer and consistent results are obtained.By using a UV LED ($1.5 only) as an excitation source, we have developed a simple, low-cost and highly efficient temperature sensor based on FIR of down-conversion green fluorescence of Er:SrGdGa3O7 crystal. The sensor not only has a simple structure but is also low-cost. Moreover, the device shows highly efficient sensing performance. It has not only a large signal-to-noise ratio but also a high sensitivity. The device may find its use in some special cases, such as electromagnetically and/or thermally harsh environments, and some industrial processes as well.Download high-res image (292KB)Download full-size image

•Two PhCs cascaded in a Ti:LiNbO3 strip waveguide was fabricated by FIB.•Two PhCs have the same lattice constant 376 nm and air-hole radius 150 nm.•A stop-band with a contrast ∼10 dB and a 3 dB bandwidth ∼200 nm is measured.•The experimental result is in good agreement with the simulated result.•Present work enables to implement desired PhC functions by cascading multi-PhCs.By using focused ion beam (FIB) technique, we have fabricated two air-hole-type triangular-lattice photonic crystals (PhCs) cascaded in a Ti-diffused LiNbO3 strip waveguide. To ease the problem that the FIB-milled air-holes are not sufficiently deep, which results in that the holes cannot overlap most of optical mode and hence no evident stop-band feature is observed, two measures have been taken: a 0.75-μm-deep shallow trench was milled on the waveguide surface prior to the air-holes, and a scheme of two PhCs cascaded in the same waveguide was adopted. The cascaded two PhCs have the same air-hole array of 15 × 15 with a lattice constant 376.57 nm and an air-hole radius 150 nm. Transmission spectrum of the cascaded PhCs was measured, and a stop-band with a contrast ∼10 dB and a 3 dB bandwidth of ∼200 nm is observed in the spectral region 1.0–1.2 μm, in good agreement with the simulated result. Present work opens up the possibility of implementing desired PhC functions for various photonic applications by cascading multi-PhCs with different structure parameters.By using focused ion beam (FIB) technique, we have fabricated two air-hole-type triangular-lattice photonic crystals (PhCs) cascaded in a Ti-diffused LiNbO3 strip waveguide. Two measures have been taken and proved to be effective to ease the problem of small stop-band contrast arising from the finite hole depth. First, the air-holes were milled at the bottom of a 0.75-μm-deep shallow trench which was introduced on the waveguide surface prior to the hole milling. Second, two PhCs were cascaded in a same Ti:LN strip waveguide. The two PhCs have the same air-hole array of 15 × 15 with a lattice constant 376 nm and an air-hole radius 150 nm. A stop-band with a contrast of ∼10 dB and a 3 dB bandwidth of ∼200 nm is experimentally observed. The experimental result is in good agreement with the simulated result. Present work opens up the possibility of implementing desired PhC functions for various photonic applications by cascading multi-PhCs with different structure parameters.Download high-res image (195KB)Download full-size image

•Er and In/Er doping effects on electro-optic(EO) property of LiNbO3 were studied.•Both clamped and unclamped EO coefficients γ13 and γ33 were measured.•γ13 does not change in the studied Er3+ doping level range of 0–3.0 mol%.•γ33 shows a decrease tendency with Er3+ doping level, but the decrease is <5%.•In codoping hardly affects the EO coefficient in concentration range 0–2.6 mol%.A series of singly Er3+-doped and doubly In3+/Er3+-doped congruent LiNbO3 crystals were grown by Czochralski method. The Er3+-only doped crystals were grown from the growth melts doped with different concentrations of Er2O3 up to 1 mol%, and the In3+/Er3+-codoped crystals from the melts doped with fixed 0.5 mol% Er2O3 while varied In2O3 contents of 0.0, 0.5, 1.0 and 1.5 mol%. The electro-optic coefficients γ13 and γ33 of these doped crystals were measured by Mach-Zehnder interferometry. Both clamped and unclamped coefficients were measured. The results show that both γ13 and γ33 reveal small Er3+ and In3+ doping effects. In the Er3+-only doping case, γ13 almost does not change within the experimental error of 3%, γ33 reveals a degradation tendency with a rise in Er3+ doping level, but the degradation is no more than 5% in the considered Er3+ concentration range 0–3.0 mol% (in crystal). In the In3+/Er3+-codoping case, both γ13 and γ33 can be regarded as unchanged in the studied In3+ concentration range of 0–2.6 mol% (in crystal) within the error. The small Er3+ or In3+ doping effect is desired in light of electro-optic application of these crystals. It is qualitatively explained on the basis of EO coefficient model and doping effect on defect structure of LiNbO3.Er3+-only doping and In3+/Er3+ codoping effects on electro-optic property of LiNbO3 crystal have been studied by Mach-Zehnder interferometric method. Both γ13 and γ33 show little dependence on Er3+ or In3+ doping concentration. In the Er3+-only doping case, γ13 hardly change within the experimental error of 3%, γ33 reveals a degradation tendency with a rise in Er3+ doping level, but the degradation is no more than 5% in the considered Er3+ concentration of 0–3 mol%. In the In3+/Er3+-codoping case, both γ13 and γ33 can be regarded as unchanged in the studied In3+ concentration range of 0–2.6 mol% within the error. The small doping effect is desired in light of electro-optic application of these crystals. A qualitative, comprehensible explanation for the small effect is given.Download full-size image

•Electro-optic (EO) property of In3+/Er3+-doped LiNbO3 (LN) crystal is studied.•In3+/Er3+ codoping has small effect on the EO property of LN.•The small effect is desired from the viewpoint of the electro-optic application.•The small effect is explained in terms of EO coefficient and defect models of LN.Clamped and unclamped electro-optic coefficients γ13 and γ33 of In3+/Er3+-codoped LiNbO3 crystals, which were grown by Czochralski method from the melts containing 0.5 mol% Er2O3 while varied In2O3 contents of 0.0, 0.5, 1.0 and 1.5 mol%, were measured by Mach-Zehnder interferometry. The results show that In3+/Er3+ codoping does not cause change of γ13 and γ33, and both γ13 and γ33 can be regarded as unchanged in the studied In3+ concentration range of 0–2.6 mol% (in crystal) within the experimental error of 3%. The small doping effect is desired in light of the electro-optic application of the crystal. A qualitative, comprehensible explanation for the small effect is given on the basis of the EO coefficient model of LiNbO3 and doping effect on the defect structure of LiNbO3.Electro-optic (EO) property of In3+/Er3+-doped LiNbO3 crystal has been studied. The results show that In3+/Er3+ codoping does not cause change of clamped and unclamped EO coefficients γ13 and γ33, and the coefficient can be regarded as identical in the studied In3+ concentration range of 0–2.6 mol% (in crystal) within the experimental error of 3%. The little doping effect is desired in light of the electro-optic application of the crystal. A qualitative, comprehensible explanation for the little effect is given on the basis of the EO coefficient model of LiNbO3 and doping effect on the defect structure of LiNbO3.Download high-res image (231KB)Download full-size image

Composition effect on electro-optic (EO) properties of a LiNbO3 (LN) single-crystal has been investigated in a Li2O-content range of 47.0–49.95 mol%. Some non-congruent LN crystals with different Li2O-contents were prepared by performing Li-deficient or Li-rich vapour transport equilibration treatments on as-grown congruent LN crystals. Unclamped EO coefficients γ13 and γ33 of these samples were measured by a Mach–Zehnder interferometric method. The measurements show that in the Li-deficient regime both γ13 and γ33 increase by ∼8% as Li2O-content decreases from the congruent 48.6 mol% to the 47.0 mol% in the Li-deficient regime. The feature is desired for the EO application of the Li-deficient crystal. In the near-stoichiometric regime, both γ13 and γ33 reveal a non-monotonic dependence. As the Li2O-content increases from the 48.6 mol%, the EO coefficient decreases. Around Li2O-content 49.5 mol%, a minimum is reached. After that, the EO coefficient recovers slowly. At the stoichiometric composition, it recovers to a value close to that at the congruent point. Comparison shows that different crystal growth methods give rise to different defect structure features and hence different composition effects.We demonstrate composition effect on electro-optic property of LiNbO3 in both Li-deficient and NS regimes. In the Li-deficient regime, both γ13 and γ33 increase almost linearly by ∼8% as the Li2O content decreases from the congruent point 48.6–47.0 mol%. The feature is desired for the EO application of the Li-deficient crystal. Such crystal allows to utilize its merits of higher rare-earth ion solubility and larger EO coefficients γ13 and γ33, and is therefore a more promising substrate material than the congruent one for developing a more efficient active EO device. In the NS regime, both γ13 and γ33 reveal a non-monotonic dependence with a minimum coefficient at the Li2O content 49.5 mol%. Comparison shows that different crystal growth methods cause different defect structure features and hence different composition effects on EO property of LN.Download full-size image

Temperature characteristics of 980-nm-upconverted green 530 and 550 nm fluorescence of Er3+-doped SrGdGa3O7 single crystal have been studied for temperature sensing purposes. To achieve that, a simple experimental setup has been developed that consists of a 980 nm laser diode, two interference filters, two Si-photocells and two multimeters, while without use of monochromator or focusing lens for fluorescence collection. The study shows that the Er3+-doped SrGdGa3O7 crystal displays strong green emission intensity (hence good signal-to-noise ratio) and high temperature sensitivity of fluorescence intensity ratio of the 530 and 550 nm emissions, which is (5.6–104.0) × 10−3 K−1 in the considered temperature range 100–430 K, depending on the temperature. In addition, the experiments were repeated using a spectrometer and consistent results of temperature characteristics were obtained. Present study shows that the Er3+-doped SrGdGa3O7 crystal is a promising host material for optical temperature sensing.Characteristics of 980-nm-upconverted green 530 and 550 nm fluorescence of Er3+-doped SrGdGa3O7 single crystal have been studied for temperature sensing purposes. A simple experimental setup has been developed that consists of a laser diode, two interference filters, two Si-photocells and two multimeters, without use of monochromator or focusing lens for fluorescence collection. The study shows that the Er:SrGdGa3O7 crystal displays stronger green emission intensity and hence better signal-to-noise ratio, as well as higher temperature sensitivity. The results of temperature characteristics are consistent with those obtained from the experiments repeated using a spectrometer. It is concluded that the Er:SrGdGa3O7 crystal is a promising substrate material for both lasers and temperature sensing.Download high-res image (306KB)Download full-size image

A thermodynamic study was performed on the growth of Ga3+-doped LiNbO3 (LN) thin film. Some Ga3+-doped LN thin films were grown on the LN surface by thermal diffusion of Ga2O3 film in the temperature range of 1000–1100 °C. After growth, the Ga3+ ion in the grown thin film was profiled, and its diffusion-growth characteristics were studied. The temperature dependences of diffusion-growth coefficient and solubility were quantified. These dependences are crucial to the design and growth of a Ga3+-doped LN thin film for various photonic applications. A comparison with the case of Ti4+ diffusion-growth, which induces an increase of LN refractive index and hence formation of an optical waveguide, shows that Ga3+ grows considerably faster than Ti4+ because of its smaller ionic radius. In addition, Ga3+-doping contribution to LN refractive index was studied by measuring and comparing the refractive indices at Ga3+-grown and Ga3+-free parts of the crystal surface. The results show that the contribution is small.

•Diffusion growth of Ga3+-doped LiTaO3 thin film were studied thermodynamically.•Diffusion constant is 1.41 · 10−6 m2/s and activation energy is 237.2 kJ/mol.•Solubility constant is 22.9 · 1026 ions/m3 and enthalpy of solution is 28.9 kJ/mol.•Ga3+ dopant has small effect on LiTaO3 refractive index.•Ga3+ growth can be described by a Fick-type equation with a constant diffusivity.A thermodynamic study was performed on diffusion growth of Ga3+-doped LiTaO3(LT) thin film for integrated optics. Some Ga3+-doped LT thin films were grown on LT surface by in-diffusion of homogeneously coated Ga2O3 film at the temperature range of (1273 to 1473) K. After growth, the refractive indices at Ga3+-doped and un-doped surface parts were measured by prism coupling technique and Li composition there was evaluated from the measured refractive indices. The results show that Ga3+ dopant has small effect on the LT index. Li2O out-diffusion is not measurable. The Ga3+ profile in the grown thin film was analysed by secondary ion mass spectrometry. It is found that the grown Ga3+ ions follow a complementary error function profile. A thermodynamic model for Ga3+ diffusion growth is suggested and verified experimentally. From the measured Ga3+ profiles, some thermodynamic parameters were obtained. These include diffusivity, diffusion constant, chemical activation energy, solubility, solubility constant and enthalpy of solution. These parameters are crucial to design and growth of a desired Ga3+-doped LT thin film for integrated optics. Comparison shows that Ga3+ grows in LT two orders faster than Ti4+ because of its smaller radius. Ga3+ diffusivity/solubility in LT is one order/1.5-fold lower than in LiNbO3 because Ta has a larger atomic mass and hence a lower mobility than Nb.Diffusion growth of Ga3+-doped LiTaO3(LT) thin film was studied thermodynamically. Some Ga3+-doped LT thin films were grown on LT surface by in-diffusion of homogeneously coated Ga2O3 film at the temperature range of (1273 to 1473) K. The Ga3+ profile in the grown thin film was analyzed by secondary ion mass spectrometry. Form the measured Ga3+ profiles, some thermodynamic parameters were obtained. These include diffusivity, diffusion constant, chemical activation energy, solubility, solubility constant and enthalpy of solution. These parameters are crucial to design and growth of a Ga3+-doped LT thin film with desired Ga3+ profile for integrated optics application. A thermodynamic model is suggested for the growth and verified experimentally.

The mutual influence of co-diffused Ti4+ and Sc3+ in LiNbO3 single-crystal was studied. Sc3+/Ti4+ co-doped LN plates were prepared by annealing (at 1060 °C in wet O2) Z-cut congruent LiNbO3 substrates coated with stacked Sc2O3 and Ti films, which have different thicknesses and coating sequences. After the annealing, the refractive index and composition at the undoped and Sc3+-only doped surface parts were characterized by prism coupling technique. The Sc3+ and Ti4+ profiles were analyzed by secondary ion mass spectrometry. The results show that the Sc3+-doping has little contribution to the LiNbO3 refractive index. Li2O out-diffusion is not measurable. In the only diffusion case, the Sc3+ diffuses five times slower than the Ti4+. In the Sc3+/Ti4+ co-diffusion case, the Ti4+ assists and controls the Sc3+ diffusion. The Sc3+ diffusivity increases linearly with the initial Ti-metal thickness. In contrast, the Sc3+ affects the Ti4+ diffusion little. Neither the Sc2O3 film thickness nor the coating sequence influences the diffusivity of the two ions. The co-diffusion features are explained. A model for the co-diffusion system is suggested and verified experimentally. The Ti:Sc:LiNbO3 planar and strip waveguide are optically characterized. The results show that the Sc3+ presence does not affect the waveguide characteristics and the waveguide is optical-damage-resistant.

It is crucial to find an appropriate solvent for composition analysis of LiNbO3 crystal by a chemical method, such as inductively coupled plasma atomic emission spectroscopy. We have comparatively studied several solvents for LiNbO3 crystal, including HF acid, KHSO4, B2O3, LiBO2, and NaVO3. The results show that as the NaVO3 is used as the solvent, the solubility of LiNbO3 is as high as 1 g/g at 1000 °C. The dissolving is quite fast. Neither solute nor solvent is lost from the melting during the dissolving procedure. A clear high-concentration solution is obtained. Moreover, it is verified experimentally that such a solution is valid for composition analysis of LiNbO3 crystal by a chemical method. In contrast, the other solvents suffer from one problem or another. We conclude that NaVO3 is an appropriate solvent for chemical analysis of LiNbO3.

Influence of factors on growth of off-congruent LiNbO3 single-crystal by Li-rich/Li-poor chemical vapor transport equilibration (VTE) was studied. These factors include the molar ratio of raw materials Li2CO3 and Nb2O5 for preparing two-phase crucible, the hollow space volume for VTE growth, and the period of use of two-phase crucible. To achieve the goal, a number of off-congruent LiNbO3 plates were grown by Li-rich/Li-poor VTE technique using four two-phase crucibles having different molar ratios of raw materials, hollow space volumes, and periods of use. The VTE-induced crystal composition alteration was characterized by measurement of birefringence. A comparison of results of different crucibles shows that in the Li-rich VTE case the first factor has less effect, while the latter two may cause a difference of Li2O content as much as 0.4 mol %. Larger hollow space volume and older crucible tend to degrade the VTE growth efficiency. Instead, the growth by Li-poor VTE is less affected by the three factors because of much lower Li2O gas pressure and hence slow growth rate in comparison with that by Li-rich VTE.

The influence of factors on the growth of the Er3+/Tm3+-codoped LiNbO3 (LN) crystal by codiffusion of stacked Er and Tm metal thin films was studied. These factors include the thicknesses and the coating sequence of the Er and Tm metal films, and the growth is characterized by the diffusivity and solubility. To achieve the goal, Er3+/Tm3+-codoped LN crystals were grown by codiffusion in air at 1130 °C of stacked Er and Tm metal thin films coated onto the surface of Z-cut congruent LN plates. The metal films have different thicknesses and coating sequences. After the growth, the Er3+ and Tm3+ profiles were analyzed by secondary ion mass spectrometry. The Er3+/Tm3+ diffusivity and surface concentration were obtained from the measured profiles. The diffusivity is influenced by neither the thickness nor the coating sequence of metal films. The solubility consists of Er3+ and Tm3+ components, and the two components equal the respective concentrations in the diffusion reservoir and change with the initial metal film thickness. Nevertheless, their sum, i.e., the solubility, remains a constant at a given temperature and shows a weaker effect of either the crystal cut or the coating sequence of Er and Tm films, together with the two components. In addition, the emission characteristics of the Er3+/Tm3+ diffusion-codoped LN were investigated. The results show that the codoping allows the combination of the wavelength emissions of both ions, and the resultant emission band in the telecommunication window around 1.5 μm is as wide as 150 nm, providing the possibility of S+C+L broadband amplification by employing the commercial 980 and 795 nm laser diodes as the pump sources.

A thermodynamic study on Li-poor vapor transport equilibration (VTE) in MgO-doped LiNbO3 crystal was carried out. A thorough thermodynamic VTE model is established that considers all of possible factors including the depth relative to crystal surface, VTE temperature and time. To solve the model, the composition on crystal surface was studied as a function of VTE temperature and time. To achieve that, a number of Li-deficient MgO(5 mol% in melt):LiNbO3 crystals were produced by the VTE under different temperatures from 1010 to 1130 °C for different durations up to 395 h, the crystalline phase contained was determined by X-ray powder diffraction and their surface compositions were determined from the measured birefringence. The results show that the Li-deficient crystal keeps the LiNbO3 phase. The surface Li2O-content has an Arrhenius relationship to the VTE temperature and a square-root dependence on the VTE duration. Based upon the VTE temperature and time dependences of surface Li2O-content, the solution to the thermodynamic VTE is obtained and verified experimentally. The solution is crucial to design and produce a Li-deficient MgO:LiNbO3 crystal with desired Li2O-content profile. By using the solution, one can predict the Li2O-content depth profile after a certain VTE time for a given VTE temperature.Highlights► Li-poor VTE temperature and time dependence of surface Li2O content. ► A unified expression for VTE-induced surface Li2O-content reduction. ► A thorough VTE thermodynamic model. ► An ierfc solution to VTE model is obtained and verified experimentally.

•High-temperature fluxing salt of LiNbO3 (LN) crystal by KPO3 solvent is reported.•The solubility of LN is as much as 3 g g−1 at 1050 °C and follows the Van't Hoff law.•The solution obtained is valid for LN composition analysis using chemical method.•The fluxing melt is amorphous with Nb presence in Lindqvist ion and P in free ion.•Flexible P–O bond, opened structure and high viscosity make KPO3 the adequate solvent.We report that potassium meta-phosphate (KPO3) is an adequate solvent for high-temperature fluxing salt of LiNbO3 crystal. As the KPO3 is used as the solvent, the solubility of LiNbO3 is as high as 3 g g−1 at 1050 °C. The dissolving is fast. Neither solute nor solvent evaporates from the melt during the dissolving procedure. A clear solution is obtained and verified valid for crystal composition analysis using chemical method of inductively coupled plasma atomic emission spectroscopy. To help for understanding the dissolving mechanism, the obtained fluxing melt product was further characterized using X-ray diffraction, nuclear magnetic resonance and Raman scattering spectroscopy. The results show that the melt is amorphous. The P ion in the melt is in the form of low condensation and the Nb ion is likely in the form of Lindqvist ion. Finally, the dissolving mechanism is discussed.An interesting and crucial finding that potassium meta-phosphate (KPO3) is an adequate high-temperature solvent for composition analysis of LiNbO3 optical single-crystal using chemical method.

The influence of Li-poor vapor-transport equilibration (VTE) on the surface Li2O content of initially congruent X- and Z-cut LiNbO3 crystal plates was studied against the VTE temperature and time. The VTE-induced surface-Li2O-content reduction was evaluated from the measured birefringence. The results show that the reduction and VTE temperature follow the traditional Arrhenius law with a surface-Li2O-content alteration constant of (1.0 ± 0.2) × 108/(1.6 ± 0.2) × 1010 mol % and an activation energy (2.2 ± 0.2)/(2.8 ± 0.2) eV for the X/Z-cut plate, and the reduction has a square-root dependence on the VTE time, ΔCX = 0.15t0.5 for the X-cut plate and ΔCZ = 0.167t0.5 for the Z-cut plate. A generalized empirical expression that relates the reduction to both the VTE temperature and duration is presented. The expression is useful for producing an off-congruent, Li-deficient LiNbO3 plate with the desired surface Li2O content via adjustment of the VTE temperature and duration. On the basis of the known VTE time dependence on the surface-Li2O-content reduction, a solution to the Li+ out-diffusion equation, an integral of the error function complement, is obtained and verified by previously reported experimental results. The results also show that the VTE displays slight anisotropy and is slightly faster along the optical axis direction of the crystal. The Li-poor VTE is a slow process. At 1100 °C, the Li-poor VTE time required for the surface Li2O content reaching the Li-deficient boundary is about 400/323 h for the X/Z-cut plate.

Rare-earth-doping and various post-growth heat treatments effects on OH− absorption in singly Er- and Yb-doped, doubly Er/Mg- and Er/Yb-doped LiNbO3 crystals have been investigated. The OH− absorption in the rare-earth-doped crystals shows definite anisotropy and is stronger in a Z-cut plate. Rare-earth-doping causes considerable OH− absorption reduction. MgO-codoping induces additional reduction. Slight/heavy rare-earth-doping causes slight/considerable spectral shape change. Thermal reduction treatment causes OH− absorption vanishing whether the crystal is originally congruent or near-stoichiometric. Further oxidation treatment renders the absorption reemerged. Li-rich vapor transport equilibration (VTE) treatment causes significant absorption decrease. For a slightly rare-earth-doped Li-rich VTE-treated crystal, its spectral feature is similar to that of the undoped near-stoichiometric crystal; for a heavily rare-earth-doped VTE-treated crystal, its spectral shape preserves the feature of the heavily rare-earth-doped as-grown crystal. For the Er/Mg-codoped crystal studied, the Li-rich VTE also induces an abrupt conversion of peaking position from 3483 cm−1 to 3535 cm−1. The absorption in the Li-rich VTE-treated crystals also reveals aging effects in spectral shape, peaking position and absorption intensity. Li-poor VTE induces significant absorption increase. In addition, the effect of OH− presence on lifetime of Er3+ emission at 1.5 μm was also studied. The results show that the effect is small for the as-grown congruent crystals and slight for Li-rich VTE-treated ones.Graphical abstractHighlights► Rare-earth-doping, heat treatments and aging effects on OH− absorption in LiNbO3. ► OH− absorption reveals considerable rare-earth-doping and MgO-codoping effects. ► OH− band vanishes after thermal reduction and reemerges after further oxidation. ► Li-rich/Li-poor VTE induces significant OH− absorption decrease/increase. ► Rare-earth-doped Li-rich VTE crystals reveal considerable aging effects.

We have measured OH absorption spectra of a 0.47-mm-thick Z-cut MgO (5 mol% in melt):LiNbO3 crystal subjected to post-growth Li-poor vapor transport equilibration (VTE) treatments at 1100 °C for different durations ranging from zero to 395 h. These spectra allow the evolution of OH absorption characteristics with prolonged VTE to be followed. After 2 h of VTE process an additional absorption appears at 3483 cm− 1. A transition regime that the original 3536 cm− 1 and new 3483 cm− 1 absorptions simultaneously appear exists within the VTE duration range of 2–16 h. In this regime, the 3536 cm− 1 absorption becomes weak gradually and eventually disappears around 16 h while the 3483 cm− 1 absorption increases remarkably with the prolonged VTE. The presence of transition regime gives a hint that the 3483 cm− 1 absorption is due to the VTE-induced formation of a new center. Based upon the spectral features, we have suggested a simple three-layer (two on surface and one in the center of crystal plate) model to describe the depth profile of the photorefractive damage MgO concentration threshold in the crystal in the transition regime. A Li out-diffusion theory is suggested and correlated with the model. To support the Li out-diffusion theory, some additional experiments have been done. These include the depth profile characterization of VTE-induced Li2O content reduction and the measurement of the surface Li2O content as a function of the VTE duration. A quantitative analysis and discussion shows that the model is well supported by the experimental results.Highlights► Evolution of OH absorption in Mg:LN with Li-poor VTE duration is clearly followed. ► VTE induces a new center resulting in absorption at 3483 cm− 1. ► The VTE duration for crystal fully below photorefractive threshold is ~ 16 h. ► Three-layer model is suggested and supported by experimental results. ► Li out-diffusion theory is established and correlated with three-layer model.