Standardized evaluation of chemical compositions of LiTaO3 single crystals for SAW devices using the LFB ultrasonic material characterization system

The line-focus-beam ultrasonic material characterization (LIFB-UMC) system is applied to compare and evaluate tolerances provided independently for the Curie temperature T/sub C/ and lattice constant /spl alpha/ to evaluate commercial LiTaO/sub 3/ single crystals by measuring the Rayleigh-type leaky surface acoustic wave (LSAW) velocities V/sub LSAW/. The relationships between VLSAW, and T/sub C/ and /spl alpha/ measured by individual manufacturers were obtained experimentally using 42/spl deg/YX-LiTaO/sub 3/ wafers as specimens from three crystal manufacturers. In addition, the relationship between VLSAW and SH-type SAW velocities V/sub SAW/ that are actually used for the SAW device wafers was obtained through calculations, using the chemical composition dependences of the acoustical physical constants for LiTaO/sub 3/ crystals reported previously. The result of a comparison between the T/sub C/ tolerance of /spl plusmn/3/spl deg/C and the /spl alpha/ tolerance of /spl plusmn/0.00002 nm through the common scale of VLSAW or VSAW demonstrated that the /spl alpha/ tolerance is 1.6 times larger than the T/sub C/ tolerance. Furthermore, we performed a standardized comparison of statistical data of T/sub C/ and /spl alpha/ for LiTaO/sub 3/ crystals grown by two manufacturers during 1999 and 2000, using VLSAW. The results clarified the differences of the average chemical compositions and of the chemical composition distributions among the crystal ingots between the two manufacturers. A guideline for the standardized evaluation procedure has been established for the SAW-device wafer specifications by the LFB-UMC system.     

Evaluation and improvement of optical-grade LiTaO3 single crystals by the LFB ultrasonic material characterization system

This paper describes the first demonstration for feeding back the results obtained by the line-focus-beam ultrasonic material characterization (LFB-UMC) system to the crystal growth conditions for optical-grade LiTaO₃ crystals and for achieving much improved homogeneity of chemical composition. We evaluated a commercially available optical-grade LiTaO₃ single crystal with a nominally congruent composition in detail, by measuring distributions of the velocities of leaky surface acoustic waves (LSAW) along the Y-axis direction for a Z-cut specimen plate prepared from the crystal grown in the Y-axis direction. We detected an increment of 0.66 m/s in LSAW velocity along the pulling axis direction corresponding to 0.024 mol% in Li₂O content, and the compositional gradient was +0.346×10^-3 (Li₂O-mol%)/mm. By experimentally obtaining the starting material composition dependence of the gradients, we developed a method of estimating the proper composition ratio that would lead to a more homogeneous crystal. We grew a new crystal with a Li₂O content of 48.47 mol%, resulting in a very small compositional gradient of +0.046×10^-3 (Li ₂O-mol%)/mm and a compositional homogeneity of less than 0.012 Li₂O-mol% in a Z-cut area of 50 mm×50 mm used for device substrates     

Evaluation of mass-produced commercial LiTaO3 single crystals using the LFB ultrasonic material characterization system

A mass-production line of lithium tantalate (LiTaO3) crystals with a maximum charge number of 60 for surface acoustic wave (SAW) devices was evaluated with the line-focus-beam (LFB) ultrasonic material characterization system. Some serious problems associated with chemical compositions were observed and resolved by measuring the velocities of Rayleigh-type leaky surface acoustic waves (LSAWs), VLSAW, for two groups of LiTaO3 wafers: 21 36 degrees Y X-LiTaO3 wafers selected randomly from crystal ingots grown with different charge numbers in different furnaces, and 14 42 degrees Y X-LiTaO3 wafers obtained at the top, middle, and bottom parts from 5 crystals selected from 39 crystals grown successively in the same furnace and crucible. Using the measured VLSAW and the predetermined relationship between VLSAW and Li2O concentrations, M(Li2O), we estimated the average M(Li2O) controlled in the current mass-production line to be about 48.77 mol% with a maximum difference of 0.75 mol%. The composition for each crystal ingot increased linearly about 0.04 mol% from the top to the bottom, and no dependence on the charge number was observed, as the melt composition used for the mass production was controlled through Curie temperature (TC) measurements. A nearly true congruent composition of 48.49 Li2O-mol% was obtained through the precise VLSAW data for the 42 degrees Y X-LiTaO3 wafers, that was about 0.3 mol% less than the melt composition in the production line. It was also pointed out that the TC measurement conditions, including room temperatures surrounding the measurement systems, should be re-examined for reliable production control. A guideline for more efficient mass production of the crystals has been established concerning the true congruent composition as the starting material.     

Calibration of Curie temperatures for LiTaO/sub 3/ single crystals by the LFB ultrasonic material characterization system

The line-focus-beam ultrasonic material characterization (LFB-UMC) system is applied to a standardized comparison and evaluation of the Curie temperatures, T/sub C/, exclusively used in evaluating the chemical compositions of commercial LiTaO/sub 3/ crystals by measuring the velocities of Rayleigh-type leaky surface acoustic waves (LSAWs), V/sub LSAW/. We measured V/sub LSAW/ and T/sub C/ (standardized) under the same T/sub C/ measurement conditions for 36/spl deg/Y X-LiTaO/sub 3/ single-crystal wafers produced by four manufacturers and related the results to the T/sub C/ (individual) measured by the individual manufacturers. The relationships between V/sub LSAW/ and T/sub C/ (individual) varied from one company to another, and a single straight line of the proportional relationship between V/sub LSAW/ and T/sub C/ (standardized) was obtained for all wafers regardless of the manufacturer. These experimental results clarify that the problem associated with T/sub C/ measurements lies in the measurement conditions and the absolute accuracy of the measurement instruments. Measurements of the center frequencies of SH-type surface acoustic wave (SAW) filter devices are compared with V/sub LSAW/ measurements. A method of calibrating T/sub C/ using this ultrasonic system is proposed to establish standardized specifications of SAW-device crystal wafers.     

Determination of the true congruent composition for LiTaO3 single crystals using the LFB ultrasonic material characterization system.

The true congruent composition for LiTaO3 single crystals was determined by measuring the velocities of leaky surface acoustic waves (LSAWs) with the line-focus-beam ultrasonic material characterization (LFB-UMC) system for two 42 degrees YX-LiTa3s crystal ingots. The congruent composition determined here was 48.460 Li2O-mol%, corresponding to the LSAW velocity (42 degrees YX-LiTaO3) of 3125.3 m/s, and the absolute relationship between the LSAW velocity and chemical composition was obtained. Simulations on the variation of the melt and crystal compositions in a mass production of 100 crystals were conducted as a function of the composition of the starting material around the congruent composition. The result showed that the distributions of the melt and crystal compositions within and among the crystals varied largely with the material composition, providing the relationship of the material composition with the maximum composition variation for the 100 crystals. Based on these results, we verified the relationships between the tolerance of the material composition variation and the tolerances for the SH-type SAW velocity, LSAW velocity, and Curie temperature. The material composition needs to be constrained to within +/- 0.007 Li2O-mol% around the congruent composition to mass-produce the crystals with reliable homogeneity, satisfying the tolerance of +/- 0.01% in the SAW velocity. Furthermore, a guideline for the specification of reliable piezoelectric SAW-device wafer substrates was presented with the accurate interrelationships among the chemical composition ratio, LSAW velocity, and Curie temperature.     

Determination of the true congruent composition for LiTaO/sub 3/ single crystals using the LFB ultrasonic material characterization system

The true congruent composition for LiTaO/sub 3/ single crystals was determined by measuring the velocities of leaky surface acoustic waves (LSAWs) with the line-focus-beam ultrasonic material characterization (LFB-UMC) system for two 42/spl deg/YX-LiTaO/sub 3/ crystal ingots. The congruent composition determined here was 48.460 Li/sub 2/O-mol%, corresponding to the LSAW velocity (42/spl deg/YX-LiTaO/sub 3/) of 3125.3 m/s, and the absolute relationship between the LSAW velocity and chemical composition was obtained. Simulations on the variation of the melt and crystal compositions in a mass production of 100 crystals were conducted as a function of the composition of the starting material around the congruent composition. The result showed that the distributions of the melt and crystal compositions within and among the crystals varied largely with the material composition, providing the relationship of the material composition with the maximum composition variation for the 100 crystals. Based on these results, we verified the relationships between the tolerance of the material composition variation and the tolerances for the SH-type SAW velocity, LSAW velocity, and Curie temperature. The material composition needs to be constrained to within /spl plusmn/0.007 Li/sub 2/O-mol% around the congruent composition to mass-produce the crystals with reliable homogeneity, satisfying the tolerance of /spl plusmn/0.01% in the SAW velocity. Furthermore, a guideline for the specification of reliable piezoelectric SAW-device wafer substrates was presented with the accurate interrelationships among the chemical composition ratio, LSAW velocity, and Curie temperature.     

Development of an improved calibration method for the LFB ultrasonic material characterization system

We investigated standard specimens for accurately calibrating the line-focus-beam ultrasonic material characterization (LFB-UMC) system without system dependencies. We evaluated several types of lithium tantalate (LiTaO3) substrates using two LFB-UMC systems with different device/system characteristics to measure and calibrate the propagation characteristics of the leaky surface acoustic waves (LSAWs), and analyzed the variations between the calibrated results. We concluded from this analysis that, by selecting materials with the cut surfaces and propagation directions of standard specimens that are identical to the objects to be calibrated, calibration errors resulting from different performance characteristics between the two systems could be nearly eliminated. Also, analytical errors caused by the effects of spectra with two close peaks (another propagation wave mode), one of the most common problems of characterization in the past, could be eliminated at the same time by this method.     

Evaluation method of TiO2-SiO2 ultra-low-expansion glasses with periodic striae using the LFB ultrasonic material characterization system.

Experimental procedures and standard specimens for characterizing and evaluating TiO2-SiO2 ultra-low expansion glasses with periodic striae using the line-focus-beam (LFB) ultrasonic material characterization system are discussed. Two types of specimens were prepared, with specimen surfaces parallel and perpendicular to the striae plane using two different grades of glass ingots. The inhomogeneities of each of the specimens were evaluated at 225 MHz. It was clarified that parallel specimens are useful for accurately measuring velocity variations of leaky surface acoustic waves (LSAWs) excited on a water-loaded specimen surface associated with the striae. Perpendicular specimens are useful for obtaining periodicities in the striae for LSAW propagation perpendicular to the striae plane on a surface and for precisely measuring averaged velocities for LSAW propagation parallel to the striae plane. The standard velocity of Rayleigh-type LSAWs traveling parallel to the striae plane for the perpendicular specimens was numerically calculated using the measured velocities of longitudinal and shear waves and density. Consequently, a reliable standard specimen with an LSAW velocity of 3308.18 +/- 0.35 m/s at 23 degrees C and its temperature coefficient of 0.39 (m/s)/degrees C was obtained for a TiO2-SiO2 glass with a TiO2 concentration of 7.09 wt%. A basis for the striae analysis using this ultrasonic method was established.     

Accurate measurements of the acoustical physical constants of LiNbO (3) and LiTaO(3) single crystals

The acoustical physical constants (elastic constant, piezoelectric constant, dielectric constant, and density) of commercial surface acoustic wave (SAW)-grade LiNbO(3) and LiTaO(3) single crystals were determined by measuring the bulk acoustic wave velocities, dielectric constants, and densities of many plate specimens prepared from the ingots. The maximum probable error in each constant was examined by considering the dependence of each constant on the measured acoustic velocities. By comparing the measured values of longitudinal velocities that were not used to determine the constants with the calculated values using the previously mentioned constants, we found that the differences between the measured and calculated values were 1 m/s or less for both LiNbO(3) and LiTaO(3) crystals. These results suggest that the acoustical physical constants determined in this paper can give the values of bulk acoustic wave velocities with four significant digits.     

Evaluation method of TiO/sub 2/-SiO/sub 2/ ultra-low-expansion glasses with periodic striae using the LFB ultrasonic material characterization system

Experimental procedures and standard specimens for characterizing and evaluating TiO/sub 2/-SiO/sub 2/ ultralow expansion glasses with periodic striae using the line-focus-beam (LFB) ultrasonic material characterization system are discussed. Two types of specimens were prepared, with specimen surfaces parallel and perpendicular to the striae plane using two different grades of glass ingots. The inhomogeneities of each of the specimens were evaluated at 225 MHz. It was clarified that parallel specimens are useful for accurately measuring velocity variations of leaky surface acoustic waves (LSAWs) excited on a water-loaded specimen surface associated with the striae. Perpendicular specimens are useful for obtaining periodicities in the striae for LSAW propagation perpendicular to the striae plane on a surface and for precisely measuring averaged velocities for LSAW propagation parallel to the striae plane. The standard velocity of Rayleigh-type LSAWs traveling parallel to the striae plane for the perpendicular specimens was numerically calculated using the measured velocities of longitudinal and shear waves and density. Consequently, a reliable standard specimen with an LSAW velocity of 3308.18 /spl plusmn/ 0.35 m/s at 23/spl deg/C and its temperature coefficient of 0.39 (m/s)//spl deg/C was obtained for a TiO/sub 2/-SiO/sub 2/ glass with a TiO/sub 2/ concentration of 7.09 wt%. A basis for the striae analysis using this ultrasonic method was established.     

Temperature coefficients of SAW delay and velocity for Y-cut and rotated LiNbO(3)

Measurements that indicate that the surface-acoustic-wave (SAW) temperature coefficients of delay and velocity over approximately room temperature to +100 degrees C for the popular cuts of lithium niobate are presented. These values of delay coefficient are close to the value previously given for Y-cut LiNbO(3) but are significantly different from values given for the rotated cuts.     

Development of the line-focus-beam ultrasonic material characterization system

A line-focus-beam ultrasonic material characterization (LFB-UMC) system has been developed to evaluate large diameter crystals and wafers currently used in electronic devices. The system enables highly accurate detection of slight changes in the physical and chemical properties in and among specimens. Material characterization proceeds by measuring the propagation characteristics, viz., phase velocity and attenuation, of Rayleigh-type leaky surface acoustic waves (LSAWs) excited on the water-loaded specimen surface. The measurement accuracy depends mainly upon the translation accuracy of the mechanical stages used in the system and the stability of the temperature environment. New precision mechanical translation stages have been developed, and the mechanical system, including the ultrasonic device and the specimen, has been installed in a temperature-controlled chamber to reduce thermal convection and conduction at the specimen. A method for precisely measuring temperature and longitudinal velocity in the water couplant has been developed, and a measurement procedure for precisely measuring the LSAW velocities has been completed, achieving greater relative accuracy to better than ±0.002% at any single chosen point and ±0.004% for two-dimensional measurements over a scanning area of a 200-mm diameter silicon single-crystal substrate. The system was developed to address various problems arising in science and industry associated with the development of materials and device fabrication processes     

High-accuracy standard specimens for the line-focus-beam ultrasonic material characterization system

We prepared standard specimens for the line-focus-beam ultrasonic material characterization system to obtain absolute values of the propagation characteristics (phase velocity and attenuation) of leaky surface acoustic waves (LSAWs). The characterization system is very useful for evaluating and analyzing specimen surfaces. The calibration accuracy of these acoustic parameters depends on the accuracy of acoustical physical constants (elastic constants, piezoelectric constants, dielectric constants, and density) determined for standard specimens. In this paper, we developed substrates of non piezoelectric single crystals (viz., gadolinium gallium garnet [GGG], Si, and Ge) and an isotropic solid (synthetic silica [SiO₂] glass) as standard specimens. These specimens can cover the phase velocity range of 2600 to 5100 m/s for Rayleigh-type LSAWs. To determine the elastic constants with high accuracy, we measured velocities by the complex-mode measurement method and corrected diffraction effects. Measurements of bulk acoustic properties (bulk wave velocity and density) were conducted around 23°C, and bulk wave velocities were obtained with an accuracy of within ±0.004%. We clearly detected differences in acoustic properties by comparing the obtained results with the previously published values; the differences were considered to be due to differences of the specimens used. We also detected differences in acoustic properties among four SiO₂ substrates produced by different manufacturers     

Extraordinary-mode refractive-index change produced by the linear electro-optic effect in LiNbO(3) and reverse-poled LiNbO(3)

To examine aspects of an integrated photonic electric-field sensor, we calculate electro-optically induced refractive-index change in regular and reverse-poled LiNbO(3). Specifically, for y-propagating extraordinary modes, we determine how index change depends on electric-field magnitude and direction. To accomplish this, changes in index-ellipsoid shape and orientation are found by the use of a numerical eigenvalue procedure to diagonalize the impermeability tensor; then, refractive index is calculated by the use of a vector reference-frame transformation and a small perturbation approximation. A general formula is inferred from calculations for specific field directions. Electro-optic coefficients for reversepoled LiNbO(3) are obtained by application of a tensor reference-frame transformation to those of LiNbO(3). The index-calculation procedure has utility beyond the problem that is considered.     

Propagation of transverse bulk and surface acoustic waves in LiNbO (3) variable time-delay devices

Experimental measurements are reported on voltage-controlled acoustic time-delay lines operating at 1 GHz in the nearly pure shear-horizontal (S-H) mode in 38 degrees rotated Y-cut LiNbO(3). The high-acoustic velocity (4800 m/s) in conjunction with the large electroacoustic effect exhibited by this orientation allows high-frequency operation and optimum time-delay tuning sensitivity with a planar, single surface, device geometry. The authors demonstrate fractional time delay of 0.3x10(-6) V(-1 ) for surface electrodes that produce an in-plane E-field. However, the simultaneous excitation and propagation of both a leaky surface-acoustic wave (LSAW) and surface skimming bulk wave (SSBW), both as (nearly pure) S-H waves in these devices, seriously restricts the extent to which it is possible to maximize the time delay modulation sensitivity by reducing electrode gap spacing as done in similar SAW devices. The LSAW and surface-skimming body wave (SSBW) propagate at nearly the same velocity on a free surface, and perturbation of their velocity and relative attenuation rates by surface electrodes causes pronounced interference effects between the two modes for some device geometries.     

Phases of the SAW reflection and transmission coefficients for short reflectors on 128 degree LiNbO3

We study numerically the phase of surface acoustic waves reflected by or transmitted through short reflectors comprising only 1-3 aluminium electrodes on 128 degree YX-cut lithium niobate (LiNbO3). The electrodes have a finite thickness, and they are either open-circuited or grounded. The center-to-center distance between adjacent electrodes d corresponds roughly either to half of the characteristic wavelength d proportional to lambda0/2 or to d proportional to lambda0, for the reflectors operating at the fundamental and second harmonic modes, respectively. We use software based on the finite-element and boundary-element methods (FEM/BEM) for numerical experiments with a tailored test structure having 3 interdigital transducers (IDTs), simulating experimental conditions with an incident wave and reflected and transmitted surface acoustic wave (SAW). Using artificial enhancement of time resolution in conjunction with the fast Fourier transform (FFT) and time-gating, calculation of the Y-parameters in a relatively wide frequency range allows us to determine the phase of the reflection and transmission coefficients.     

BaTiO3 (BTO)–CaCu3Ti4O12 (CCTO) substrates for microwave devices and antennas

The solid state procedure was used to produce bulk ceramics of BTO (BaTiO₃), CCTO (CaCu₃Ti₄O₁₂) and BTO_0.5–CCTO_0.5 that were studied in the medium-frequency (MF) range (100 Hz–1 MHz) and in the microwave range of frequencies. The presence of BTO is decreasing the dielectric constant (ε_r) of the BTO–CCTO composite. The CCTO and BTO samples present a strong tendency to the increase of the loss with frequency. The BTO substrates are presenting higher values of the ε_r in the range of 1–4 GHz (around 140). For pure CCTO the dielectric constant is around 37.6. Similar behaviour observed at the MF range, that the higher dielectric constant is also associated to the higher loss is also present in the microwave region. The study of a planar microstrip antenna, that uses the BTO _X –CCTO_(1−X) ceramic as a high ε_r substrate was done. Therefore, these measurements confirm the potential use of such materials for small high dielectric planar antennas (HDA). These materials are also very promising for capacitor applications and certainly for microelectronics, microwave devices (cell mobile phones for example), where the miniaturization of the devices is crucial.     

Real-time photorefractive interferometer for dynamic phase perturbation by self-interference in LiNbO(3)

We propose a simple real-time system and demonstrate its use for measuring dynamic optical phase perturbation. In this system we used a 0.1-wt. % Fe:LiNbO(3) to record the self-interference grating with incident light. The system is a new kind of real-time holographic interferometer. After rise time in the interferometer, the speed for showing the fringes is as fast as that of dynamic phase perturbations. Characteristics of the interferometer are proposed and examined.