Under certain conditions, light wave propagation through atmospheric turbulence can cause presence of branch points. The number of branch points in the turbulence-distorted optical field is related with the strength of atmospheric turbulence, propagation range and the wavelength of laser propagated. The phase of compensating main laser is acquired from the distorted field of beacon. But the phase of the beacon is composed of the continuous phase and discontinuous phase when there are branch points in distorted field. The discontinuous phase is related to the number and location of branch points in distorted field, and the wavelength of main laser is always different from the wavelength of the beacon. Therefore, when the wavelength of main laser propagated is determined, the wavelength of beacon must be chosen so that the turbulence-compensated power of adaptive optical system can be improved after the branch-point considered in strong scintillation. The calculated results show that the compensation effect of adaptive optical system can be slightly improved when the branch points in turbulence-distorted optical field are considered, but the correction result of adaptive optics and become much better when the beacon wavelength chosen approaches the wavelength of main laser

There are unknown impurities and defects usually in unintentionally undoped GaN film due to imperfect growth procedures so that the luminescence related to the energy levels of the unknown impurities and defects occurs. The red luminescence of undoped GaN film at 692 nm is reported. The emitting mechanisms of blue and red luminescence of undoped GaN film are studied. The energy levels of initial states of blue and red luminescence are measured directly in experiments by using a novel absorption-normalized photoluminescent excitation spectroscopy proposed by the authors. It is determined that the blue luminescence originates from the transition of donor to valence band, and the red luminescence is caused by the recombination of donor to acceptor. A complete luminescent model is built up for the blue, yellow and red luminescence of undoped GaN film. The gotten results have important reference values for determination of types of the unknown impurities and defects.

Three-component all-fiber acceleration seismometer and its digital signal processing(DSP) system are developed. The seismometer is composed of a seismic mass, six cylinders and three sets of Michelson interferometer light path. The six cylinders are critical parts among the sensor elements. An advanced quality digital signal processing(DSP) chip equipped with necessary circuits has been used in its digital signal processing system, a high accurate detection of the acceleration signal has been achieved and the environmental interference signal has been effectively compensated. Test results indicate that the accelerometer has better frequency response above the resonant frequency, and the output signal is in correspondence with the input signal. The accelerometer also has better frequency response under the resonant frequency. At last, the curve of seismometer frequency response is given.

The growing interest for the use of integrated optical(IO) amplifiers in optical fiber telecom has led to deep search for ever better rare-earth-doped glassy materials. Some properties of a set of Er~(3+)/Yb~(3+)-doped silicate glasses are reported. Planar and channel waveguides are fabricated by Ag~+-Na~+ ion exchange. Spectral measurement shows that the fluorescence emission peak for all the samples is observed at 1532 nm, with a full-width at half-maximum equal to(19 nm.) Measurements of the lifetime of the metastable state(()~4I_(13/2)) of Er~(3+) are performed upon excitation by laser pumps at 514.5 nm and 980 nm, and life time of the order of 7 ms are measured in most samples. The intensity of the up-conversion fluorescence is less in Er~(3+)/Yb~(3+) co-doped glasses than in Er~(3+)-doped ones. 5 dB net gain at the wavelength of 1536 nm has been obtained in a 3.5 cm long channel waveguide with(250 mW) pump power at 980 nm and the FWHM is 14 nm.

A new structure of polarized light interference interleaver is presented. The key component of the interleaver is birefringent analog waveplate. It is equivalent to a birefringent crystal. The equivalent birefringence can be much larger than that of a conventional birefringent crystal. The interleaver has the characteristics of good stability, high reliability that the general polarized light interference interleaver possesses. Moreover, it has much higher isolation and can more easily realize any size of frequency spacing. An interleaver with the frequency spacing of(50 GHz) is designed. The flatness width of passband and stopband are both greater than 1/5 period, and the channel isolation is -30 dB. Several parameters in practical applications of this device, such as insertion loss, polarization dependent loss and cost are analyzed and discussed as well.

A novel structure of high extinction ratio comb filter based on dual-pass Mach-Zehnder interferometer is proposed. The comb filter is structured by adding an optical isolator to the two outputs of the conventional single-pass Mach-Zehnder interferometer. The characteristic of the transmission spectrum is studied theoretically and experimentally. The numerical results show that the extinction ratio can be enhanced greatly in comparison with the conventional single-pass Mach-Zehnder interferometer in the same parameters. The improvement of the extinction ratio mainly depends on the difference between the two arms of the Mach-Zehnder interferometer. When the difference is 0.7366 mm, the extinction ratio is improved 20 dB. The experimental results show that the extinction ratio of the dual-pass Mach-Zehnder interferometer is 25.8 dB,the extinction ratio is improvedly 13 dB.

The reconstruction problem of temperature field of plasma with limited-view data is studied. The reconstruction experiment of the arc plasma tempreture is fulfilled combining the optical computed tomography with the absolute spectral intensity method in the plasma diagnosis. An optical computed tomography algorithm based on maximum entropy and optimization is discussed in detail theoretically. Its reconstruction results for asymmetrical temperature field with the simulation computation are studied. The effects of noise, view numbers and distribution are analyzed and this algorithm is compared with ART algorithm. The results show the average error of the new algorithm is 0.03% and that of ART is 1.77% with the data of two orthogonal views.The average error of the new algorithm is 1.77% and that of ART is 2.02% with the data of four space evenly views. As a measurement sample, an arc temperature field with spectrum method is reconstructed.

The method of counting for circular ring objects in frequency domain is proposed. A binarized image with circular rings is processed by Fourier transformation and filtered in frequency domain, then inverse Fourier transformation is carried on. After selecting a suitable threshold, the circular rings become solid particles individually. The number of circular rings is obtained through counting the solid particles. The mathematical model is established and the experimental results for both simulated and real images are also given. The method is robust to noise and the detection for partially occluded circular rings is also available. The method can also be applied to the situation that some objects are overlapped. The method is more effective when the size of the objects in the image is similar to each other.

A new data processing method by means of Fourier transform has been developed. In Fourier transform, frequency leakage will surely be caused because of the space-limited extension of fringe pattern in Fourier transform. It will cause measurement error. The reason why frequency leakage is caused is analyzed, the corresponding numerical simulation is presented in detail, and the preliminary experiment is carried out. On the basis of the analysis, the numerical simulation and the experiment, it is proposed that external interpolation continuation of the fringe pattern to any integer times will greatly improve the measurement accuracy. The data processing method will promote the application of moiré deflectometry in micro-scale heat transfer and flow measurement.

The total internal reflection and the system of photon scanning tunneling microscope(PSTM) are simulated using finite difference time domain(FDTD) method. The basic principle of photon scanning tunneling microscope is that the surface evanescent wave generated under total internal reflection condition and the surface evanescent wave is explored. The incident wave is set throughout the problem space in photon scanning tunneling microscope into three-wave included the incident wave, reflective wave and transmitted evanescent wave. The intensity distribution of near field of sphere samples is calculated. And the scanning images can be obtained when the probe of sphere scans along a line above the sphere samples. The results show that the finite difference time domain method is an efficient tool for photon scanning tunneling microscope.

Using series expansion, the Z-scan characteristics of thin optically nonlinear medium are analyzed. Through discussing series expansion and Fresnel intergral method, It is verified that Gaussian decomposition method is equivalent to Fresnel integral method for(Z-scan) measurements of a thin medium using a Gaussian light beam even for a large nonlinear phase shift, and clarify some misunderstandings. Meanwhile, the relationship of normalized transmittance with upper limiter of Fresnel integral is analyzed, the reason for the oscillation of series summation is given and a criterion for minimum summation number of series needed to eliminate the oscillation is suggested while using a Gaussian decomposition method. The situations that are suitable to be applied for Gaussian decomposition method and Fresnel intergral method are discussed. The conclusion given can be used to choose the appropriate and high efficient analytic method in experimental and theoretical analysis.

The analytical study on transmitted intensity′s perturbance of the Glan-Taylor prism with varying rotation angle is given. It is found that the variation of the transmitted intensity with the rotation angle deviates from Malus law, exhibiting a cyclic fluctuation with the rotation angle. The occurrence of the disturbance is explained by interferance between the monochromatic light passing through the Glan-Taylor prism and the factor of the crystal-air interface transmissivity, which together lead to the relation between the transmitted intensity and the incident angle. The notion of the perturbance factor is introduced, and the influences of the factor discussed systematically. Theoretical results meet well with the experiment. Effective measures for weakening the disturbance are presented. The discussions here supply theoretical basis for the design,production and use of the polarizing prisms.

The concept of coaxial di-nondiffracting beam is proposed. If two nondiffracting beams with small difference of convergence angle and common axis exist at same time, they will interference with each other to form a light field structure that varies periodically along the propagation direction. It is a new kind of space distribution of light beam. The cross-section of this beam is still rings fringe pattern with common center that can be used for precision alignment. However in different longitudinal position, the intensity distribution of these rings is different and varies periodically. The periodicity of the coaxial di-nondiffracting beam can be used for the positioning along the propagation direction. It is shown that the annular di-slit, axicon consisting of two conic surface and sine phase ring grating are capable of generating coaxial di-nondiffracting beam in theory and experiments. The diffraction pattern characteristics of coaxial di-nondiffracting beam along both transverse direction and longitudinal direction are analyzed. The experimental result fits to theoretical prediction basically.

The whole-geometric-approximation-method(WGAM) is presented to compute the forward light intensity. Since Fraunhofer diffraction and geometric scattering are combined, such a method is applicable under anomalous or normal diffraction regime. Due to the consideration of interference between the diffraction, refraction and the reflection rays, the calculable angle range is extended to 0°～60°. Because of its high calculation efficiency and wide validity range, WGAM is especially effective for light pattern calculation of the large particles with size parameter α larger than 40. Numerical calculations show that the results of WGAM are consistent with Mie theory. The experiments also find that WGAM can be applied in laser particle sizing based on large-angle forward scattering method.

Cube-corner prism is widely applied in laser ranging because of its retro-reflective property. The requirements of its precision are different according to applied cases. The far-field diffractive characteristics of cube-corner prism will directly affect long-distance laser ranging. By the analysis of dihedral angle offset and flatness error of cube-corner, the reflective beam wave aberration distribution of cube-corner is presented and the far-field diffractive patterns are computed. The patterns show that divergence of the emerging beam increases with the dihedral angle offset up to completely separate to six beams; the flatness error will result in deterioration of the patterns and debase the quality of the emerging beam. The diffractive characteristics of cube-corner prism provides an evaluation method and approach to design this kind of prism.

The formulae of Gaussian parameters between the relay lens and the(two-mirror) system are derived. They satisfy the condition of exit pupil of the IR system matching the cold stop of the IR sensor. Once the diameter of the cold stop and its distance to the image plane are confirmed, the Gaussian parameters of the relay lens and the two-mirror system must satisfy these formulae, then the exit pupil will coincide with the cold stop. The elimination of aberrations may be done by aspherization of the primary and secondary mirrors and one surface of the relay lens. In order to prove the efficiency of the derived formulae,(6 real) systems are solved. The diameter of the IR system is taken as 250 mm, the diameter of the cold stop is 5 mm, and the distance between the cold stop and the IR image plane is(20 mm.) The radius of curvature of the primary mirror has two values:(-1000 mm) and(-800 mm,) the focal length of the two-mirror system:(2000 mm,)(1500 mm) and(1000 mm.) So,there are 6 systems with different constructive data are calculated.

The relationships of mean output difference and distribution index of ring laser gyroscope with the intensity of random noise are obtained. Based on the relationships, the way to design random noise is produced and exampled. The results show that pure random noise is no good to reduce the error because the introduction of random noise will reduce the measurement scope of ring laser gyroscope sharply or enlarge the output difference of ring laser gyroscope heavily. However, superimposing random noise on periodical dither signal can reduce the error deeply while the measurement scope of ring laser gyroscope is not reduced and increasing sample period moderately can reduce the enlarged output difference.

The electrooptic effect of the optical active crystals is studied in the experimental set of polarized light interference. In the polarized light interferometric experiment, the express equation of the relation between the emitted light intensity and the optical activity(I=A~2_0cos~2β-(π/λ)(n_l-n_r)l) is given. The express equation of the relation between the emitted light intensity and the electrooptic effect of the optical active crystals is also given, that is(I=A~2_0cos~2β-(π/λ)(n_l-n_r)l+(π/λ)(n_2-n_1)l.) According to these relations expressed by the equations, the electrooptic Q switch is made by using La_3Ga_5SiO_(14) crystal which is a typically optical active crystal. The Q switch works normally as those switches which have not optical activity. It may be used in medium output energy lasers to partly take the place of DKAP Q-switches.

As the basis of application of one-dimensional photonic crystals, the investigation of photonic band gap is very important. The properties of one-dimensional photonic crystal′s band gap and the factors affecting it are investigated through transfer matrix method. The importance of relative bandwidth to the design of photonic crystal is described. Based on these discussions, the methods used to extend the band gap of 1-D photonic crystal are discussed. A new method, using some 1-D photonic crystals to extend the band gap at angle space, is proposed. By using this method, some structures composed of 2, 3 or 4 1-D photonic crystals are designed. At the end, the reflection of two structures at optical frequencies are calculated. The relative band width of hybrid structure consisted of four 1-D photonic crystals is up to 57.52%, which strongly indicates that the extension of relative band width of 1-D photonic crystals based on incident angle domain is very effective.

The nanocrystalline TiO_2 thin films plays an important role in the photoelectrochromic devices. The microstructure of nanocrystalline TiO_2 has effects on the properties of absorption of dyes, diffraction of light and transportation of charges. Nanocrystalline TiO_2 thin films were prepared by the electron beam evaporation. The surface morphology, crystal state and composition were analyzed with atomic force microscope(AFM), X-ray diffraction(XRD) and Auger electronic spectroscopy(AES). The photoelectric performances affected by particle size were discussed theoretically. And the blue-shift phenomenon of the absorption peak value was explained by quantum confinement effect.

Spectroscopic ellipsometric measurements in the infrared region is used to characterize the structure of CVD-diamond films. According to the status of surface and interface, different models are developed for diamond films on smooth silicon substrate and rough alumina substrate respectively with theoretical simulations followed. The Bruggeman effective media approximation, which includes the factors such as roughness, void and phase mixture, is applied to simulate the effects of rough surface and complicated interface on the ellipsometric data. It is found that the establishment of appropriate model has the strongest influence on the fit of ellipsometric spectra. The films on silicon substrates are well described by a two-layer film model. The bulk film(first layer) is pure diamond. The second layer(surface) represents a rough surface sublayer about(879 nm) thick, which contains 28.3% voids. But satisfactory agreements between experimental and calculated data cannot be obtained when the two-layer film model is aplied to diamond films on alumina substrates. The best fit is achieved for the samples on alumina substrates with a three-component interface layer(64.13% alumina+23.34% diamond+12.53% void) included by Bruggeman effective media approximation.

In order to make the optical disk having good recording/reading performance and long-term stability, the multilayer structure of the phase-change optical disk must be optimized. By using the design software capable of simulating and analyzing the reading process for phase-change optical disk design, a phase-change optical disk structure for the blue laser(405 nm) was designed optically. The relationships between the(multilayer′s) optical parameters(reflectivity and reflectivity contrast) and the film thickness of the structure and the depth of the groove are studied. The optimum structures of the phase-change optical disk are lower ZnS-SiO_2/AgInTeSb/upper ZnS-SiO_2/Al layered structure of 100 nm/10 nm/25 nm/60 nm for land recording and 140 nm/15 nm/30 nm/60 nm for groove recording. The depth of the groove is 50 nm. The simulation results have great instructional effect on the fabrication of the high density blue-ray phase-change optical disk in the future.