Studies on Lattice Vibrations of Dielectric and Magnetic Crystals by Hyper-Raman Scattering.

The experimental arrangement set up for The measurement of The .hyper-Raman spectrum is asfollows.The 1.06 <micro> m line from an acousto-optically Q-switched Nd:YAG laser (NEC) was used forthe excitation. the repetition rate was set to between 2khz （pulse width 150 ns）和6khz （200）ns). A peak power between 3 and 50 KW was employed. The laser beam was focussed into The sample by Af = 5.cm lens. The scattered light in The 90 geometry was dispersed by a polychromator （JASCO）TRS-501 and detected by the intensified photodiode array detector （Tracor Northern）controller. In this多通道检测系统the frequency region of approximately 230cm^(-1)> is measured simultaneously.In the hyper-Raman spectrum below 280 <cm^(-1)> of SrTi < o_3 >crystal. three bands at 88, 175和266 <cm^(-1)> are observed, which correspond to the <nu_1> (to，<f_(1u)>), <nu_1> (LO, <f_(1u)>) + <nu_2> (TO，<f_(1u)>) and <nu_4> (<f_(2u)>) respectively. These bands were used for the adjustment andwavenumber calibration of our spectrometer.In CsCa <C1 - 3>，following five hyper-Raman bands observed：76 < cm ^(- 1) >(< 1 >主_ lo), 114 < cm ^(- 1) >(< 2 >主_ to), 142 < cm ^(- 1) >(< 2 >主_ lo),267 <cm^(-1)> (<nu_3> TO) and 312 <cm^(-1)> (<nu_3> LO). These are compared with the data obtainedfrom the farr -infrared reflection spectrum and a Kramers-Kroning analysis. the hyper-Raman data可能更可靠，比farr -infrared databecause (1) the flectivity measurements depend on the condition of the crystal surface and (2) the数据in the low-reflectivity region influence the accuracy of the dielectric function and the TOand LO frequencies calculated from relectivity data by Kramers-Kronig relation.The hyper-Ramanspectra of several alkali halides and perovskite chlorides were measured。

The experimental arrangement set up for the measurement of the.hyper-Raman spectrum is as follows.The 1.06 <micro> m line from an acousto-optically Q-switched Nd:YAG laser (NEC) was used for the excitation. The repetition rate was set to between 2 KHz (pulse width 150 ns) and 6 KHz (200 ns). A peak power between 3 and 50 KW was employed. The laser beam was focussed into the sample by a f = 5.cm lens. The scattered light in the 90゜ geometry was dispersed by a polychromator (JASCO TRS-501) and detected by the intensified photodiode array detector (Tracor Northern) with a controller. In this multichannel detection system the frequency region of approximately 230 <cm^(-1)> is measured simultaneously.In the hyper-Raman spectrum below 280 <cm^(-1)> of SrTi <O_3> crystal. three bands at 88, 175 and 266 <cm^(-1)> are observed, which correspond to the <nu_1> (TO, <f_(1u)> ), <nu_1> (LO, <f_(1u)> ) + <nu_2> (TO, <f_(1u)> ) and <nu_4> ( <f_(2u)> ) respectively. These bands were used for the adjustment and wavenumber calibration of our spectrometer.In CsCa <C1_3> , the following five hyper-Raman bands were observed: 76 <cm^(-1)> ( <nu_1> LO), 114 <cm^(-1)> ( <nu_2> TO), 142 <cm^(-1)> ( <nu_2> LO), 267 <cm^(-1)> ( <nu_3> TO) and 312 <cm^(-1)> ( <nu_3> LO). These are compared with the data obtained from the far-infrared reflection spectrum and a Kramers-Kroning analysis. The hyper-Raman data might be more reliable than the far-infrared data, because (1) the reflectivity measurements depend on the condition of the crystal surface and (2) the data in the low-reflectivity region influence the accuracy of the dielectric function and the TO and LO frequencies calculated from relectivity data by Kramers-Kronig relation.The hyper-Raman spectra of several alkali halides and perovskite chlorides were measured.