弾性表面波フィルタ用LiNbO_3回転Y板の最適カット

<p>As the piezoelectric substrate for elastic surface wave filters and delay lines, 131° rotated Y-cut X-propagating crystalline lithium niobate plates are widely used because of their superiority in electromechanical coupling to Rayleigh waves and low beam steering compared to other cuts. However, an unknown spurious signal generated on the substrate frequentry prevents successful experiments. For filters, this leads to such phenomena for which the attenuation in the stop band cannot be guaranteed to be sufficiently large. This paper deals mainly with the experimental suppression of the spurious component through RF pulse responses and the frequency characteristics of the rotated Y-cut plates cut at several angles near 131°. The following facts are apparent:(1)The spurious component corresponds to the slower of the two shear waves propagating along the X-axis in the semiinfinite LiNbO_3 plate. (2)The component greatry depends on the cutting angle θ and is minimum at θ=128. 86°, that is, this angle gives the cut for optimum suppression. Experiments were performed using the plate rotated θdegree from the Y-axis about the X-axis of the LiNbO_3 crystal as illustrated in Fig. 1. The specimens were obtained by cutting or rubbing down at intervals of about one degree from 123. 6° to 131. 88° and their cut angles were measured exactly by X-ray diffraction. On the surface of these specimens, uniform overlap electrodes, shown in Fig. 2 were fabricated by the photolithographic technique and on the lower part of these elecrodes, many grooves were cut in order to suppress reflection Fig. 4 shows the RF pulse response of the θ=130. 86° specimen measured by the apparatus as shown in Fig. 3. Fig. 4(a) shows the response of the Rayleigh wave component when the amplitude is maximum and the frequency is 39. 5 MHz. Fig. 4(b) shows the response of the spurious component for maximum amplitude under the condition that the Rayleigh wave component is suppressed by adhering viscoelastic tape onto the propagating path and the frequency is 40. 5 MHz. By comparing these responses, it is clear that the spurious wave propagates faster than the Rayleigh wave by about 2. 5%. Next, several frequency responses were observed using a frequency spectrum analyzer under three conditions:(a) for a free path, (b) for suppression of the Rayleigh wave component by adhering viscoelastic tape onto the path and (c) for plastic clay set on the transducer electrode beside the tape, as shown in Fig. 5. These results are shown in Figs. 6, 7 and 8. The peak values of these responses are arranged and plotted in Fig. 9. The triangles in Figs. 11 and 12 show the velocities and the effective electromechanical coupling factors, respectively, as measured by the electrodes shown in Fig. 10. In these figures, the curves marked "×" shows the theoretical values calculated using the elastic and piezoelectric constants taken from the work of Waner, Onoe and Coquin, and the cueves marked "・" show the values calculated using the constant measured by Nakagawa, Yamanouchi and Shibayama. The propagation velocity of the spurious wave obtained from the relation between the pitch of the electrodes and the measured center frequency in Figs. 6, 7 and 8 is about 4050m/sec and this agrees well with the result for the RF pulse reponse mentioned above. It is conclided, by comparing the theoretical values shown in Fig. 11 with this result, that the spurious wave corresponds to the slower of the two shear waves in the LiNbO_3 crystals. It is clear from Fig. 9 that the spurious component is minimum for a cut angle of 127. 86°for which the plane corresponds to (0, -1, 4)-plane with a Bragg diffraction angle of 32. 63° and can be suppressed by about -60 dB with respect to the Rayleigh component. On the other hand , non-beam-steering properties of the new cut crystal were confirmed by measurements using a laser probe, as is shown in Fig.</p><p>(View PDF for the rest of the abstract.)</p>