The theoretical analysis of the compound cavity of a semiconductor laser with external optical feedback is conducted. For large optical feedback, the output power from the laser and its oscillation frequency differ from those for small optical feedback. From the rate equations of the compound cavity, the conditions of the laser oscillation are derived in the presence of large optical feedback The dependencies of the output power and the laser oscillation frequency on the external-cavity length are investigated. Some new results involving laser oscillation depending on the external-cavity length are presented. The experimental results are compared with theoretical predictions. The dependence of the laser oscillation on the external-cavity length is qualitatively explained in the present model.

In silica glass, the peak position of infrared absorption spectrum at around 2260 ${\mathrm{cm}}^{\text{−1}},$ which is related to the average of distribution of Si–O–Si bonding angle, is useful for determining frozen-in temperature of glass structure, the so-called fictive temperature. We precisely determined relation between the peak position and fictive temperature up to $\text{1550\hspace{0.05em}°}\mathrm{C}$ and found that at fictive temperatures higher than $\text{1400\hspace{0.05em}°}\mathrm{C}$ the present relation differs from the relation which has been extrapolated from the previous data at fictive temperatures below $\text{1400\hspace{0.05em}°}\mathrm{C}.$ The peak position was found to simply shift to lower wave numbers with raising fictive temperature. This suggests that at fictive temperatures up to about $\text{1600\hspace{0.05em}°}\mathrm{C}$ there is no maximum in density, which has been reported to exist at around $\text{1500\hspace{0.05em}°}\mathrm{C}.$

The limit of the Rayleigh scattering and total losses in silica fiber was determined from the results of the Rayleigh scattering and structural relaxation measurements. Annealing condition during fiber drawing was optimized, and much potential to reduce the Rayleigh scattering loss has been verified. Total loss can be reduced below 0.15 dB/km at 1.55 μm, for example, even when annealing time is 0.1 s. Total loss below 0.14 dB/km can be realized by increasing the annealing time.