A passively mode-locked soliton laser is constructed employing semiconductor saturable absorber mirror (SESAM) as mode-locking regime. This laser only has single all-solid Yb-doped photonic bandgap fiber (AS-Yb-PBGF). At 1 μm wavelength region, the fiber has anomalous group velocity dispersion (GVD), thus it provides the laser gain and anomalous dispersion simultaneously. In the cavity, the balanced effect between the nonlinear effect and the anomalous GVD offered by the AS-Yb-PBGF can lead to soliton operation. By means of the split-step Fourier method, the dynamics of single-soliton operation in the laser is numerically simulated. At different initial signals the same stable soliton operation is finally obtained. The single-soliton operation with the pulse duration of 125 fs, the pulse energy up to 135 pJ has been shown, and the time-bandwidth product is 0.33 nearly the transform-limited pulse. The pulse repetition rate is up to 500 MHz. At the same time, the dynamics process of the pulse in the cavity is systemically studied.

The second-order approximate solution of P-doped cascaded Raman fiber lasers were obtained by nonlinear method. The differential equations depicting the distribution of pump and Stokes radiations along Raman fiber were simplified to the algebraic equations, and then the approximate solution under different input pump powers by second-order nonlinear method and the threshold expressions was obtained. The second-order solution agrees well with the numerical simulation. On the one hand, the second-order method was more accurate than the linear method and it could provide better guidance to the experimental design in regard to P-doped cascaded Raman fiber lasers for dual-wavelength outputs. On the other hand, the method effectively improves the numerical computational speed, and is superior to general numerical methods.

A new kind of single-longitudinal-mode, wavelength-switchable fiber laser is proposed and experimentally demonstrated in this paper. Such fiber laser has the Fabry-Perot (F-P) cavity and uses 980 nm pumped erbium-doped fiber (EDF) as gain media. And an umpumped EDF in the cavity is used as the saturable absorber to achieve single-longitudinal-mode operation, meanwhile a 1×N optical fiber switch and N parallel fiber Bragg gratings (FBGs) are used as the wavelength selection module to achieve wavelength-switchable operation. With 90 mW pumping power, the single-longitudinal-mode laser with -0.5 dBm peak power and 3.6 kHz linewidth is generated; And the wavelength can be randomly switched among the wavelengths of 1574.6 nm, 1579.7 nm, 1584.8 nm and 1589.9 nm by applying the suitable electrical signal to fiber switch.

A high-rate and high-power optical parametric oscillator is experimentally demonstrated. 1064 nm Q-switched Nd:YAG laser is pumped by laser diode (LD) laser. The maximal output power is 7.8 W with the pulses repetition rate range of 5～50 kHz. Optical parametric oscillator (OPO) based on periodically-poled lithium niobate (PPMgLN) pumped by the 1064 nm Q-switched Nd:YAG laser. Periodically poled grating (Λ=30.7 μm) was successfully fabricated in 1.0 mm thick MgO:LiNbO3 (doped 5% MgO). The output infrared wavelength will change because of the expansion coefficient of PPLN material when the operating temperatures fluctuate from 40 to 200 ℃. The tuning dimensions of the infrared wavelength are 1570～1676 nm and 2942～3300 nm with the signal maximal power of 613 mW. Several hundred milliwatt of output idler average power was obtained by using short and dual-cavity, the energy of single pulse is 40 μJ, and the conversion efficiency is 3.4%.

Round-way transmission matrix of laser resonator was founded, whose eigenvalue and eigenvector are used for calculating diffraction loss and corresponding mode distribution. The calculated results fit well with the Fox-Li method. The relation of eigenvalue and eigenvector of round-way transmission matrix A2 and one-way transmission matrix A of symmetrical laser resonator, and that of round-way transmission matrix AB and BA of asymmetrical laser resonator, were analyzed. The mathematics expression of transmission matrix relation was founded, which is used for improving calculation efficiency. The calculation discussed the effect of confocal cavity discrete unit amount on the resonator mode. And mathematics expression between optimizing discrete unit amount and radius of resonator mirror was built, and confocal cavity mode with big Fresnel coefficient was calculated.

Generally, a passive Q-switching laser has a wider time jitter. In order to reduce the jitter time and the size of the laser, gain switching principle is researched. According to the principle, the pumping power density is controled by the power which generates a step current. A Q-switched laser pumped with a ring laser diodes according to gain switching principle is developed. A 30 mJ output energy and about 16ns pulse duration of this kind of passive Q-switching laser are obtained. There is only one set of pupming source to be used in the experiment. Gain step pulse is directly obtained by controling the waveform of current, and the factors which affect the laser pulse jitter are reduced by this way. Statistic reports of multi-measurement indicate that the laser pulse jittering time is less than 1 μs comparing with the trigger signal. According to the measured stability of pre-pumping and the transient power of the step pulse, the jitter time Δτd is calculated, which agrees well with the experimental result.

In corner-pumped configuration, dichroic mirror of high transparency for pump light and high reflection for laser light can be avoided, and with laser crystal of low doping concentration the pump absorption efficiency is still acceptable, so this pump configuration is suitable for demanding pumping requirements of radiation-balanced laser (RBL). The formula for the laser crystal width is optimized in corner-pumped configuration to avoid the leakage of the pump light from the crystal pumping corner. When the pumping wavelength is 992.8 nm and the laser crystal of RBL is Yb:KGW with 5% doping concentration, ray trace method is applied to simulate corner-pumped configurations both with and without diffusion technique numerically. According to efficiency and uniformity of pumping light absorption, in the design of both configurations the best pump incident angle is 45° and the best pump total reflection number is 1, but the corner-pumped configuration with diffusion technique is about 20% more efficient than the that without it.

Random lasing action of rhodamine 6G (R6G) in ethyl-cryanoethyl cellulose [(E-CE)C]/acrylic acid (AA) cholesteric liquid crystal (LC) solution without doping fine powder was firstly studied. It is found that random laser with coherent feedback occurrs in the LC solutions, while only narrowing spectra phenomenon was observed in isotropic (E-CE)C/AA solution and AA solvent. And the narrowest full-wavelength at half-maximum (FWHM) of laser peak is 0.3 nm. The laser threshold changes from 27 μJ to 20 μJ when the concentration of (E-CE)C/AA solution varies from 35% to 46.5%. This random laser with coherent feedback is related to the multiple scattering resulting from the selected reflection and low diffusion constant in LC solution.

Analysis shows that intensity of the focal spot will be decreased by the systemic optical aberration in a multi-hundred TW ultra-short laser system. In order to increase the intensity of the far-field by avoiding both axial chromatic and spherical aberrations caused by the lens as in the laser system, source of the aberrations was studied, simulation of layout was performed by ray tracing method and air-spaced doublet as the last spatial filter was specially designed to compensate those aberrations. Experimentation proves that the air-spaced doublet can diminish both axial chromatic and spherical aberration of the whole system effectively , the radius of the output focal spot decreases from 8.6 μm to 5.7 μm, and the intensity of the focus increases by eight times.

The time and spatial-resolved emission spectrum of Co atom 389.408 nm spectra line in laser induced Co plasma was measured. The electron density of plasmas was calculated from the measured intensity and Stark broadening of emission spectral line. The time and spatial evolution properties of the electron density were finally discussed. The measured results show that the plasma electron density is varied from 0.02×1017 to 0.73×1017 cm-3 when the time delay is in the range from 100 to 1000 ns. It is also concluded that the plasma electron density is varied from 0.3×1017cm-3 to 0.8×1017 cm-3 along the direction of the laser beam when the distance from target surface is in the range from 0 to 1.8 mm. The distribution of the plasma electron density along the laser beam has well symmetry.

Based on the shaping characteristic of femtosecond in the focal region and the free electron density rate equation, through modifying the electron diffusion rate and cascade ionization rate, the ellipsoidal time and space (ETS) model for femtosecond laser-induced optical breakdown in water was given. Theoretical breakdown thresholds calculated by ETS model with a Runge-Kutta method were presented under the condition of pulse durations of 100 fs and 300 fs. The optical breakdown thresholds and the electron distribution within the breakdown region at different times were given, and the change of free electrons with time at fixed positions within the region was predicted. Results show that the breakdown threshold of water induced by femtosecond laser using the ETS model is better agreement with experimental result than the classical models.

Cr2+:ZnSe having a wide range of absorption and emission spectra, is an excellent tuning laser material in mid-IR wavelength range. In this paper, lasing abilities of Cr2+:ZnSe had been studied by analyzing the absorption spectrum, emission spectrum and tuning ability. The Cr2+:ZnSe poly-crystal laser material is prepared by physical vapour transport (PVT) technique. The sample is cut and polished in thickness of 1.7 mm and with diameter of 10 mm. Pumped by a 2.05 μm Tm-ion doped silica fiber laser, delivering maximum output power of 8 W, maximum average output power of 1.034 W around 2.367 μm was demonstrated in a plate-concave resonate. The laser bandwidth was measured as 10 nm. At the same time, the laser can be tuned in 100 nm range by angle-tuning method in research of tuning ability of the laser material.

Because of the complicated optic path and many optical components, relative shift of the optic mechanical parts can cause easily optic-axis parallelism departure the normal direction in high power laser launch equipment and multi-axial photoelectric tracking system. Therefore a real time auto-alignment system is proposed. Based on the alignment principle, up and down datum beam is built and paralleled with the axes of the system and the output laser separately by using the path of emitting laser. By using the CCD auto-collimator to measure the angle error of up and down path, signal is controlled to drive the fast steering mirror until the angle error of the two axes is zero. The alignment between emitting laser and the first beam expanding system and the fine alignment between the first beam expanding system and tracking and pointing system can be realized. Eperimental results show that optical-axis parallelism is 5.21″, and the accuracy of close-loop of fast steering mirror is smaller than 5＂.

Based on coupled nonlinear Schrdinger equation (NLSE) and the split-step Fourier method (SSFM), a numerical investigation of the propagation and the switching of fundamental solitons in two-core fiber nonlinear directional coupler (NLDC) with high order coupling dispersion coefficient is present. In this model, the first coupling dispersion is similar to the group velocity mismatch. The second coupling dispersion is similar to the group velocity dispersion (GVD), and it leads to mutual modulation for the other channel. Optical pulses propagate periodically inside fiber coupler and NLDC have sharp switching characteristics when the first order coupling dispersion coefficient is smaller, but the periodicity of the coupling transmission of optical solitons and the sharpness of switching were destroyed by the increase of the first order coupling dispersion coefficient, so optical pulses breakup in the propagation. The second order coupling dispersion coefficient reduce the coupling length, and increase the switching threshold power, as well as make switching characteristics become sharper.

The wavelength demodulation of fiber Bragg grating laser sensor (FBGLS) system with 3×3 couplers depend on the physical characteristics of the 3×3 couplers. The demodulated results will distort to some extent due to the instability of the parameters in the three outputs of 3×3 couplers including the direct current (DC) component, the fringe visibility of interferometer and the phase difference. A method of measuring the above parameters is given and applied in the computer programming so as to measure the parameters in real-time under large signals, at the same time corresponding demodulation method is also introduced. The computer simulation shows that with the proposed method the appearance of harmonics caused by the instability of the parameters in the three outputs of 3×3 couplers is eliminated. The comparison of the experimentally demodulated results also shows that with the proposed method the harmonics have been reduced and the demodulated results have high correlation coefficients with the reference sensor. Besides, high resolution, dynamic range and linearity of the demodulated results are given in the experiment by using the proposed method.

By combining the nano-technology with the fiber technology, a novel optical fiber doped with InP semiconductor nano-particles was fabricated by using the modified chemical vapor deposition (MCVD) method. Through experiment, the mass fraction of InP in fiber is approximately 0.1%, and the fiber has an excellent waveguide characteristic. By using the scanning electron microscope (SEM), a stereo-scan photograph of the fiber was got, and based on the graph, the effective core area Aeff≈10.01 μm2 was calculated by using the finite element method (FEM). So the nonlinear coefficient of the fiber γ=10.53 W-1·km-1 was got. The results show that this novel optical fiber has a higher nonlinear coefficient than the normal fibers.

Spectral characteristics of three-layered cascaded long-period fiber gratings (LPFGs) were analyzed by the coupled-mode theory together with the transfer matrix method. The effects of the length, position of the fiber that connects the LPFGs on the transmission spectrum of the cascaded LPFG were discussed, and the transmission spectra between the cascaded LPFG and the phase-shifted LPFG were compared. The results show that their spectra have great consistency when the length of the cascading fiber is small or the cascading position is near the end of the cascaded LPFG. But when the cascading fiber is large or the cascading position is in the middle of the cascaded LPFG, the spectra of the cascaded LPFG and the phase-shifted LPFG are different. These results agree well with the two-layered cascaded LPFG. The changes of the film parameters can adjust flexibly the attenuation peak shifts of the spectra and the peak values of the spectra.

The principle of optical rotation due to the transfer of spin angular momentum from light to particles was discussed by analyzing the interaction between the beam of light and birefringent crystal particles. Through MATLAB simulation, the dependence of the rotation frequency of the trapped particle on the laser power was analyzed, it was obtained that the rotation frequency was proportional to the laser power. The birefringent different particles were rotated by the optical tweezers, the wavelength of He-Ne laser was 632.8 nm, at the same time the rotation frequency was measured, and the maximum speed could reach 5 r/s. The dependence of the rotation frequency of the trapped particle on the laser-power was experimentally studied for several birefringent particles with different sizes in the optical tweezers. When experimental results were compared with theoretical results, experimental results and theoretical analysis were agreeing very well, and the reasons for the error were analyzed, which the bottom friction affected the most.

Based on the general Huygens-Fresnel diffraction integral formula, the focal switch effect of Hermite-cosine-Gaussian beams passing through an astigmatic lens is investigated in detail. The results of numerical calculations show that this kind of focal switch is dependent of the geometric parameter of the optical system δ, and the focal switch effect occurs regardless of whether the δ is 1.0 or not, i.e. the focal switches could occur not only at geometric focus but also at the right or the left of the geometric focus. The influence of beam parameter on the occurrence and the position of focal switches is discussed. The dependence of the relative focal shift on the astigmatic parameter is examined at different geometric parameters of the optical system δ for a particular beam parameter and Fresnel number. The result shows that the occurrence and the position of focal switches can be controlled by changing the beam parameter and astigmatic parameter.

To explore the feasibility of fabricating nasal prosthesis with selective laser sintering method, a digital model of nasal prostheses for patients was produced through the integration of structural light scanning and CAD technique. A wax model of nasal prostheses was fabricated with selected sintering method under special manufacturing parameters. The result of accuracy evaluation and clinical try-in showed a satisfying shape of wax model and a comparatively high precision. Then the wax pattern was processed into the definitive nasal prosthesis and a clinically satisfying effect was achieved. Comparing with conventional manual methods, the new fabricating method shortens time and simplifies manufacturing steps. This digital and model-free manufacture provides a new method for the automatic manufacture of maxillofacial prostheses, and has potential for the clinical application in the future.

Two-color holographic image storage was performed experimentally in LiNbO3:Ce:Cu crystal (w(Ce): 0.085%, w(Cu): 0.011%) with gating light at 365 nm and recording light at 633 nm, and the image fidelity of the hologram was analyzed. The loss of signal-to-noise ratio (SNR) (LSNR) was used in comparing the noise performance of holograms before and after fixing. 50 angle-multiplexed holograms with almost equal diffraction efficiencies were multiplexed in a LiNbO3:Ce:Cu crystal with equivalent exposure schedule. And their average LSNR is 1.14, and the average efficiency reaches 7.64×10-6.

A apodization method with Tukey-window function for digital hologram of particle-field was presented, whose energy transmission is relatively high and the sidelobe peak is relatively low as well as the sidelobe roll-off rate. The principles of in-focus position measurement and digital hologram apodization using Tukey-window for particle were analyzed. The digital holograms of little monolayer particle-field attached on glass surface were recorded, and the apodized holograms adopting Tukey window were reconstructed. The experimental results show that the reconstructed images after digital holograms apodization do not have black-and-white interference fringe, increase contrast and have more clear particle image. The reconstructed particle image located in the fringe field is nearly submerged by the apodized aperture. With increasing the apodization parameter of the window function, the more accurate position of the particles in-focus can be gained, and the measurement precision of particle in-focus position is improved.

Based on charge-coupled device (CCD) technique for high temperature measurement, laser molten pool in laser cladding for Ni-base alloy powder was measured. Laser cladding experiments of powder feeding method and pre-fabricated powder method were carried out. Shape, size and temperature field distribution of laser molten pool were measured. It is shown that metal powder melt in part with rough molten pool shape when laser power was low than 1100 W. At 1300 W, the x,y dimensions of laser molten pool were 2.8 mm, 2.7 mm respectively with approximately ellipse shape, the average temperature was 1800 K. Its average temperature increased with small increase of shape dimension when laser power increased. Laser cladding parameters were optimized by the measurement results.

The uniform Ni-P-Al203 eletroless plating was produced on Fe-C alloy, and treated by high power continuous-wave (CW) CO2 laser. Transmission electron microscope (TEM) was applied to observe the morphology of the Al2O3 particles. Optical microscope (OM) and scanning electron microscope (SEM) were used to observe the morphologies of the coating. X-ray diffraction (XRD) and X-ray energy dispersive spectroscopy (EDS) were employed to analyze the phases and elements of the coating. The changes of micro-hardness before and after laser processing were tested respectively. The results reveal that the strengthened coating has good performances with smooth surface, uniform components and fined microstructure. The phases in strengthened layer obviously changes from amorphous state to crystalline state and the main phase is Ni3P with some non-equilibrium phases. Al2O3 particles are homogeneously dispersed in the grain boundaries, which can greatly improve the material properties. The whole coating is divided into four regions from surface to inner as follows: laser strengthed region, transition zone, heat affected zone (HAZ) and substrate. The microhardness of strengthened region is about 3 times of substrate owing to the dispersion of the Al2O3 particles and the phosphide as strengthening phase.

Aiming at the disadvantages of the present techniques for repairing the damages of aluminum-alloy aircraft structure, laser welding technique was proposed for repairing the damages of 2A12 alloy structure. Laser welding neck of Al-5Mg-0.3Sc alloy and 2A12 alloy was investigated. Through metallographic microscope and scanning electronic microscope, the properties such as metallurgical structure, tensile strength of the welding seam and the matrix alloy, tensile strength contrast to rivet specimen were studied. The experimental results show that the grain of the welded joint between 2A12 and Al-5Mg-0.3Sc is obviously refined, and tensile strengths of welded joint and riveted structure of 2A12 are equivalent. Laser welding neck structure of Al-5Mg-0.3Sc alloy and 2A12 can replace riveted repair structure of 2A12.

Raman spectra of real and counterfeit RMB 100 banknotes excited by an infrared laser with wavelength of 785 nm were obtained in the portrait watermark, security thread, latent image for the face value and OVI printing features. Difference between the spectra of real and counterfeit RMB banknotes was found. Meanwhile, difference of counterfeit banknotes from different counterfeiters was also observed. The study shows that Raman spectroscopy could be a potential tool not only for the identification of counterfeit and altered RMB, but also for the sources of the counterfeit RMB.

The existing slicing algorithms for metal laser solid forming (MLSF) were analyzed. A rapid slicing algorithm based on dynamic topological reconstruction for STL (stereo lithography) model was proposed. According to geometrical information of the triangular facets and the slicing thickness, the times of traversing the triangular facets are reduced by using grouping matrix method. The times of intersecting between the slicing plane and the triangular facets are reduced by using the local dynamic topological construction algorithm. The fact that the topology of triangles between two slicing layers is similar or the same, a dynamic topological reconstruction algorithm is built. The time of searching adjoining relation among triangles is reduced, the efficiency of the slicing algorithm was increased effectively. On the basis of the algorithm, the software with independent intellectual property rights was developed using Visual C++ and OpenGL.

A new method based on laser ultrasonic was proposed to survey welding stress distributions on the material surface according to the acoustoelastic principle, The high frequency Rayleigh waves excited by the Nd:YAG pulsed laser were used to measure the welding stress and picked up by a nonlinear interferometer. A serial of ultrasonic pulsed signals can be detected by the interferometer through changing the distance between fixed detecting point and ultrasonic source point by the scanning of the laser source. The waveform cross-correlation technique was applied to compute the propagation time delay of the adjacent Rayleigh waves, and the propagation velocity of the Rayleigh waves was obtained, by which the stress value was figured out via acoustoelastic relation. The residual stress distribution can be measured around welded joint while the laser source scans around the welded joint. This experimental installation was utilized to test the welding stress on the surface of a welded aluminum plate and the stress distribution was measured. The experimental results indicate that this method can scan over the specimen surface and measure the welding stress quickly, which make it have potential in the non-destructive evaluation of the surface welding stress distribution.

Based on the artificial neural network (ANN), a model is established to describe the laser cladding parameters and the characteristic and performance of laser cladding layers. The characteristic and performance of laser cladding layers are predicted with the model in which the input parameters consist of laser power, scanning velocity, laser spot diameter, and coating proportion and the output parameters include the clad hardness, the clad width, and the clad height. The results show that the mean error is small, and the model has good verifying precision and excellent ability of predicting. The model can basically forecast the characteristic and performance of laser cladding layers.

A 355 nm Nd:YVO4 laser with output power of 8 W and a 1064 nm Nd:YAG laser with output power of 50 W are used for etching the copper clad laminate (CCL) and flexible printed circuit (FPC). The effects of laser process parameters, such as laser power density, repetition rate, laser scanning speed and single pulse energy, on the etching quality of microstructure are studied in detail. The experimental results demonstrate that the lasers within ultraviolet (UV) range needs less power to etch the copper layer entirely and causes less thermal effect, as copper and polymer have higher absorptivity to UV. On the other hand, the infrared laser is suitable for the ablation of copper layer due to its little damage on polymer. The 355 nm UV laser is suitable for the cutting process of the pinted circut board (PCB) because it can make the facile separation of thick polymer substrates quickly.

We theorically discussed the relation between the refractive index matching and the residual reflectance of the laser glass and cladding glass. The experimental results show that the better the refractive index matching and the higher product of the absorption coefficient and the cladding glass thickness, the lower the residual reflection, which result in the better restraint of the parasitic oscillation in amplified spontaneons emission（ASE ） and the higher gain from the laser system.

Based on the dual-hyperbolic two-temperature and double-step heat conduction model ,numerical simulation for the electron and lattice temperatures field of gold film is performed by ultra-short laser pulses, using the finite-difference method with viscosities and adaptive time steps. The temperature distributions of the electron and lattice at the front surface with different laser intensities and pulse widths were discussed. Simultaneously, it is also analyzed that the effect of the electron- lattice coupling coefficient on the change of electron-lattice temperature and the time of electron-lattice coupling to thermal equilibrium. Numerical results show that the laser intensity and pulse width impact significantly on the peak of the electron temperature, and electron lattice temperature rising rate and coupling time are determined by the coupling factor. Eectron temperature and its gradient quickly reach the maximum in a small region near the irradiated surface, corresponding with high hot-electron blast force, which could be a dominating mechanism for mechanical damage at the early stage of ultra-short laser ablation.

The basic principle and methods of rate control in e-beam evaporated optical coatings are introduced. Rate control experiments are performed based on materials of SiO2 and HfO2, whose evaporation characteristics are hard to control. Using the proportion integral differential(PID) closed loop control, the original control parameters were set by Ziegler-Nichols experimental formula. Then the original parameters were adjused and the integral and derivative roles were delt with in subarea based on the experiment. The rate control experiment results show that good performance can be achieved by this tuning way combining with improved process flow. The problem existing in rate control are analyzed and the improving methods are proposed. Combining the rate control with e-gun sweep can futher improve the rate control performance.

Ta2O5 films were prepared on BK7 substrates using electron-beam evaporation (EBE) and ion-beam sputtering (IBS), and the films deposited by EBE were also annealed. The optical properties, laser-induced damage threshold (LIDT), absorption, scattering and root-mean square (RMS) roughness, microdefect density and impurity content of the prepared films were investigated. The results show that the optical properties of the annealed Ta2O5 films deposited by EBE can be improved near to that by IBS. The LIDT of Ta2O5 films deposited by EBE is lower than that by IBS, which is due to the larger absorption, microdefect density and impurity content and has nothing with the scattering and RMS roughness. The films prepared by EBE after being annealed show lower absorption and microdefect density, and improved LIDT. The LIDT after being annealed is still lower than that by IBS, which can be attributed to that the annealing cannot decrease the impurity content of the films. Therefore, the higher LIDT would be achieved by using high purity starting material and reducing the contaminant in the evaporation course.

Heat treatment with oxygen, laser conditioning and ion post-treatment were post processed respectively on ZrO2 single layers deposited by electron beam evaporation (EBE). Optical properties and laser-induce damage threshold (LIDT) of the films were studied. It is found that heat treatment can get rid of the hydrosphere adsorbed in the coating, offset the lost of oxygen in the deposition, which leads to a short shift of spectra curve, a decrease of absorption and an increase of LIDT. Laser pretreatment could wipe off the defects at a certain extent and increase the LIDT, but has no use in spectrum and absorption. And ion post-treatment could increase the stack density, lessen the defects, reduce the absorption and improve the LIDT. Each method has its own treat mechanism, and the most suitable method should be chosen according to the characteristics of the coatings.

A method for obtaining the value of light ring of the plane optical element is presented. Three tested zones and searching zones in the interferogram are determined to search the average curvature of the interference fringes and the average fringe spacing, and the optimum estimation value of light ring of the plane optical element is obtained by using the weight coefficients of the tested zones. By measuring the standard optical elements and comparing with the visual interpretation results of interference fringes, the maximum relative error of 5% is obtained. Experimental results show that the method can be utilized to improve the adaptation capability of processing interferograms and substitute the visual interpretation of interference fringes.

To study the influence of the spatial irregular distribution of the fundamental beams on the measurement of pulse duration of the second-order autocorrelator, the principle of single-shot second-order autocorrelator was analyzed. The influences of periodical modulation and local dim /bright region on the measurement of pulse duration were discussed numerically. The results show that when the intensity was modulated periodically, at certain extent the higher the modulation and the sparser the strips, the bigger the autocorrelator measurement warp will be. When the intensity was modulated by local dim /bright region and the dim region is not large, the wider the dim region or the narrower the bright region, as well as the higher the modulation, the bigger the autocorrelator measurement warp will be.

To overcome some disadvantages of traditional methods to measure the oxygen concentration, the tunable diode laser absorption spectroscopy combined with the phase-lock amplifier is studied. By tuning the drive current of the laser diode (LD) to change the output wavelength, one or more absorption peaks of the oxygen will be scanned. And then the phase lock technique will be used to analyze the harmonic signal of the absorption spectroscopy. This method combines the high power, good monochrome, wavelength easily tuned with the drive current and the advantage of phase-lock amplifier to improve the signal and noise ratio and the measure precision. Through the measurements of some different concentration oxygens, the correlation coefficient of the result is 0.99 and the detection limit is 1142.86 mg/m3, moreover the method has the good stability and anti-interference feature.

Using a distributed-feedback (DFB) diode laser as light source, a high sensitivity system of continuous wave (CW) cavity ring down spectroscopy was built. With the diode laser’s feature of current modulation and tuning, the ring-down transient and spectrum of cavity loss can be observed easily, and the repeated measurement error of system is about 0.65×10-6. The etaloning effects found in experiment were theoretically analyzed and experimentally researched in detail, then the external weak reflection of cavity was confirmed as the source of the effects. And the main reflection face was found through analyzing. Ways of software and hardware to eliminate the effects were put forward and testified. The results show that the fitting error of software is about 0.4×10-6, and after hardware improving, the standard error of cavity loss spectrum’s fluctuation decreases from 9.2068×10-6 to 0.4561×10-6.

The gas concentration calculated from second harmonic signal would be affected by the variety of linewidth induced by different gas parameters during the measurement by wavelength modulation absorption spectrometry using tunable diode laser. The effect of linewidth variety on second harmonic signal for Lorentz lineshape was analyzed. A correction method using ratio of peak to valley from second harmonic signal was also given here. The experiment on CO2 concentration at 1578.22 nm was carried out in a 10 cm absorption cell with different modulation amplitudes, gas pressures and concentrations to validate the effect of linewidth variey on second harmonic signal. The results show that the relative errors of gas concentration would be decreased from 34.3% to 1.8% by correction method when the pressure rises from 1.0×105 Pa to 1.5×105 Pa, and decreased from 12.8% to 1.8% by correction method when the gas concentration changes from 20% to 100%. The correction method can eliminate the effect of linewidth variety on gas measurement by second harmonic signal effectively.