Thermo-optical coefficient

The sign and magnitude of the resulting thermal nonlinear refraction coefficient (which is, actually, a pure linear effect [219]) depend on the thermo-optical coefficient 9 /9r of the material. This coefficient has sometimes been assimilated to the one of the surrounding host only [132, 218], but we have recently shown that, due to local field enhancement at the SPR, they can be very different - even for weakly concentrated media exactly as for the pure electronic nonlinear properties as demonstrated in Section 3.2.4. Moreover, an absorptive thermo-optical effect, which is always disregarded in the literature, can occur parallel to the refractive one. These conclusions will be published soon. 

Here, Cs is the coefficient of strain, which is related to the strain optic coefficient Pe- The coefficient of temperature is made up of the thermal expansion coefficient of the optical fibre, ota, and the thermo-optic coefficient, 0. These parameters are slightly dependent on the FBG manufacturer, but typical values are 0.79 for Cs, and 0.55E - 6 and 8.6E - 6 for a and respectively. 

Detailed investigation on the optical characteristics, including the electro-optic phase modulation, electric hysteresis property, and thermo-optic coefficient, of transparent PMN-PT electro-optic ceramics have been conducted [229]. A polarization independent PMNT electro-optic switch by using s -shifted fiber Sagnac interferometer stmcture was constracted and analyzed experimentally. Some switch performances, including thermal characteristic and different switching frequency response, were also realized. 

Figure 10.13 shows experimental setup for the optical characteristic measurement of PMNT ceramics [133]. The size of PMN-PT ceramic sample was 5 mm X 2 mm x 1 mm for length x width x thickness. Ti/Pt/Au layers were sputtered on both surfaces of the ceramics as electrodes. Two collimators were used to collimate the incident beam and receive the transmission beam. The output beam was detected by using an optical spectrometer and phase demodulation. Because the PMN-PT electro-optic ceramics have a large refractive index, i.e., n = 2.465, the ceramic samples could be considered as a Fabry-Perot (FP) resonator, which can be used to measure the electric hysteresis and thermo-optic coefficient. The applied voltage generated a transverse electro-optic effect for the transmission light beam. 

The optical loss of the FPPEs had relatively low values (less than 0.27 dB/cm at 1310mn). In addition, the thermo-optic coefficient (dn/dT) values of the FPPEs at 1310 and 1550nm (TE mode) ranged from -0.97 X 10- to -1.33 X10-4 °C and from -0.96 x 10 4 to -1.29 X10-4 °C, respectively, indicating its suitability as an optical waveguide material. 

In R25, R26 and R27 the reader flnds detailed descriptions of phase-conjugation, self-diffraction and optical bistability realized with the help of thermal effects in nematics. The absorption of the cells were controlled by adding dyes to the nematic or coating metallic layers onto the substrates. Due to the large thermo-optic coefficients very low operating power-levels could be used. As an example, Lloyd and Wherrett observed optical bistability at an input power as low as 20/iiy. ... 

That the critical switch power is inversely proportional to the magnitude of thermo-optic coefficient dn/dT) was confirmed by experiment, and led to the choice of nematic phase liquid crystals as suitable, low absorption, spacer materials. Their high thermo-optic coefficient dn /dT — —2xl0 K at room temperature) produced all optical bistability at critical power levels of 140//W, allowing the demonstration of submilliwatt laser diode operation of these devices. 

Study of the refractive indices of the liquid crystal 4-cyano-4 -pentylbiphenyl (K 15, SCBorPCB) over its nematic range, has shown an increase in magnitude of the thermo-optic coefficient as the nematic-isotropic phase transition is approached. From (1) it is clear that if dT/dP is constant over this range, then a commensurate reduction in operating power should result. 

In samples prepared later on with better optical quality W. Schmid observed a different behaviour [120]. The damping became stronger with lower repetition rates, i.e. higher peak intensity. No thermo-optic effect was observed. This can be well explained by the advances in crystal preparation solvent inclusions were reduced and the crystal quality improved, resulting in a lower absorption coefficient at the same, near-resonant wavelength (e.g. 720 nm). So, the absorbed heat load and the thermo-optical coefficients were reduced concurrently. 

---