An ultrashort pulse laser system with precisely controlled output-timing and carrier-envelope phase (CEP) is reported. Recently developed technology Ofl CEP control of a mode-locked laser not only introduced an optical frequency comh in frequency domain hut also gave us a way to generate optical pulses whose oscillating electric field is under a fixed phase relation with the intensity shape. Fortunately, recent advances on optical physics have also showed that sonic types of light-matter interactions become sensitive to the field shape when the pulse approaches a few cycles in duration and has a high peak intensity. Owing to those advances, field-controlled ultrashort pulse generation, based on suh-femtosecond resolution timing-control and sub-radian CEP control of femtosecond lasers, becomes an attractive challenge. Our final goal is to realize a shaped electric field within optical-cycle time scale br researches on light-matter interaction and other future application. CEP control Ofl a mode-locked Ti:sapphire laser is the first step of such a laser system. Trade-off between the accuracy and robustness of the control, and the monitoring technique of CEP br amplilication, will he discussed. Amplification of a CEP-controlled pulse, which is necessary for most of time-domain application, is successfully performed by the CEP monitoring technique. Our chirped-pulse amplifier, that includes a grating-based stretcher/compressor, has a potential to achieve higher-energy amplification of a fixed CEP pulse. Multichannel phase control of spectrally divided ultrashort pulses is applied to dynamic control of pulse-timing and CEP of amplifled pulses. Related results on short-pulse, sub-l3fs, generation by a chirped-pulse Ti:sapphire amplifier, and multicolor phase-coherent pulse sources will be also discussed briefly, showing our on-going efforts to approach the final goal.

High-power solid-state laser programs at China Academy of Engineering Physics have made great progresses in recent years. A three-stage Ti:sapphire laser system, SILEX-I, was completed early in 2004 which could deliver 26-fs pulses at 5TW, 30TW, and 300TW to the corresponding target chambers for diverse applications. SILEX-I has been working very stably since its completion for experiments, demonstrating that it is the most powerful femtosecond Ti:sapphire laser for exploring strong-field phenomena in the world. The SG-III Nd:glass laser facility has been under conceptual design to meet the requirements from laser fusion applications. The SG-III facility is planned to have sixty-four beamlines divided into eight bundles with an output energy more than 100kJ at 0.35μm for 3- to 5-ns pulses. The eight-beamline TIL (Technical Integration Line), the prototype of the SG-III laser facility, has been installed in the new laboratory in Mianyang. The commissioning experiments have been conducted and one of the eight beams has produced 1-ns pulses of 3.0kJ and 1.2kJ at 1.053μm and 0.35μm, respectively. All the eight beamlines will be activated by the end of 2005 and completed in 2006 for operation. Meanwhile, the eight-beam SG-II laser in Shanghai Institute of Optics and Fine Mechanics has been operated for the experiments since 2001 and an additional beam, built in 2004, has been used for plasma backlighting experiments.

Within the last years, modern ultra-short pulse lasers have successfully proven their potential for application in medical tissue treatment in many respects. In dentistry, overheating of the pulp and induction of micro cracks are usually among the most problematic issues which can be solved in this way. An additional benefit can be seen in the possibility of plasma emission spectroscopy as a means of feedback. Up till now it was shown by many authors that the application of picosecond or femtosecond pulses allows to perform ablation with very low damaging potential also fitting to the special physiological requirements. Beside the short interaction time with the irradiated biological matter, lateral scanning of ultra-short pulses following optimized algorithms turned out to be crucial for ablating cavities with the required quality and size, a finding which we also believe to be valid for dental restoration materials. Additionally, out of practical reasons, scanning is necessary to treat larger volumes than just the focal spots typically having dimensions on the order of more than 1 mm³, thereby allowing to realize an "optical drill".

Two independent femtosecond Ti:sapphire lasers are synchronized by using a new passive synchronization design. By enhancing the intracavity cross-phase modulation (XPM), stable synchronization operation of remaining for more than 24 hours with a timing jitter of 0.4fs was demonstrated; the tolerance of cavity length mismatch is larger than 10 micrometers.

We have constructed two kinds of table-top femtosecond terawatt (TW) Ti:sapphire laser systems based on the chirped-pulse amplification (CPA). With a compact design using only two-stage amplifiers, output energies of 36mJ and 640mJ at 10hz repetition rates were obtained with recompressed pulse duration of 25 fs, and 31 fs respectively, corresponding to peak powers of about 1.4 TW and 20 TW. The total pump energy for the last stage is 260 mJ and 2.8 J at wavelength of 532 nm. These results represent a significant efficiency in amplification and a compact configuration in size. By using an adaptive optical system to correct the wave-front distortion of the 20TW laser. we further demonstrated the improvement of beam quality br higher focusable laser intensity.

We investigate femtosecond transient absorption in the visible range in a solution sample of poly(9,9'-dioctylfluorene-co-N-(4-butylphenyl)diphenylamine), TFB, using two-photon excitation at 800 nm (1.55 eV). A rapid probe absorbing process taking place within the temporal overlap between the pump and probe pulses was observed in the whole visible spectral range (430 nm to 700 nm), which resulted from the combination of a transition from 1A_g to mA_g via absorbing one pump and one probe photon and a further transition from mA_g to nB_u state due to the pump-induced transient population on mA_g. A long-lived slow process following the rapid one is interpreted as a combination of 1B_u absorption for transitions to a higher-lying state of kA_g and charge absorption. These excitation and absorption channels, as well as the mechanisms of charge separation, were resolved and evaluated quantitatively by the pump-intensity dependence of the transient absorption dynamics.

A novel 40Gbit/s all-optical high-speed format conversion scheme from nonreturn-to-zero (NRZ) format to return-to-zero (RZ) format based on cascaded second-order nonlinear interactions (SHG+DFG) in a periodically poled LiNbO₃ (PPLN) waveguide is presented for the first time, using a nonlinear optical loop mirror (NOLM). The conversion mechanism relies on the combination of amplification and nonlinear phase shift induced on the signal field by SHG+DFG. The SHG+DFG between pump, and co- and counter- propagating signals in the PPLN waveguide are studied, showing that counter-propagating SHG+DFG can be ignored when quasi-phase matching (QPM) for SHG+DFG during co-propagating interactions. The conversion with the influence of walk-off effect is numerically simulated, the effect of nonlinear phase shift on broadening the conversion bandwidth is analyzed, and an approximately 90-nm conversion bandwidth is achieved, providing the possibility of realizing a broadband tunable NRZ-to-RZ format conversion.

The problem of phase-mismatch-compensation for ultra-broadband OPCPA is first studied theoretically by us, and a scheme for phase-mismatch-compensation is proposed, which is based on the matching of both the group velocity and pulse front between signal and idler by the combination of the no collinear-phase-match and pulse-front-tilt that is accomplished by angular dispersion of the interacting rays. By this Scheme, the phase mismatch to first order in frequency shift can be completely compensated, and then an ultra-broadband OPCPA is realized. It is shown that the phase mismatch to both first and second order can be completely compensated simultaneously in some cases, and this leads to an extremely broadband spectral width. Therefore the important criterion for constructing an ultra-broadband OPCPA that both pulse-front and group-velocity between signal and idler are exactly matched simultaneously. Finally, specific numerical calculations and simulations are presented for BBO-OPCPA with type-I noncollinear angularly dispersed geometry.

We report the experimental research on the measurement and controlling of carrier envelope phase offset (CEO) with a home-made femtosecond Ti:sapphire laser, the beat frequency with a signal to noise ratio of as high as 45 dB is obtained with standard self-referencing technique. Locking the beat signal to the TV-Rb frequency standard by a phase-locked loop electronic circuit, a simple compact frequency comb was established. To further control the CEO with the technology of differenced frequency generation, we develop an ultra-broadened bandwidth femtosecond Ti:sapphire laser by balancing the dispersion with chirped mirrors, spectrum of covering from 600 nm to 1050 nm was observed. It will enable us to measure and control the CEO without photonic crystal fiber.

The coupled equations of second harmonic generation of ultra-short pulse through a nonlinear optical crystal CsLiB₆O₁₀ is solved using the Fourier method. The second harmonic pumped by 532 nm, 50 fs fundamental wave through CsLiB₆O₁₀ is simulated numerically. The results point out that the pulse of second harmonic is widened and the harmonic shape is deformed because of group velocity mismatch and group velocity dispersion. The pulse of SHG is widened to 100 fs and the pulse shaping of SHG is delayed to 50~60 fs when the CsLiB₆O₁₀ crystal length is 0.5 mm.

With excellent temporal resolution ranging from nanosecond to sub-picoseconds, a streak camera is widely utilized in measuring ultrafast light phenomena, such as detecting synchrotron radiation, examining inertial confinement fusion target, and making measurements of laser-induced discharge. In combination with appropriate optics or spectroscope, the streak camera delivers intensity vs. position (or wavelength) information on the ultrafast process. The current streak camera is based on a sweep electric pulse and an image converting tube with a wavelength-sensitive photocathode ranging from the x-ray to near infrared region. This kind of streak camera is comparatively costly and complex. This paper describes the design and performance of a new-style streak camera based on an electro-optic crystal with large electro-optic coefficient. Crystal streak camera accomplishes the goal of time resolution by direct photon beam deflection using the electro-optic effect which can replace the current streak camera from the visible to near infrared region. After computer-aided simulation, we design a crystal streak camera which has the potential of time resolution between 1ns and 10ns.Some further improvements in sweep electric circuits, a crystal with a larger electro-optic coefficient, for example LN (γ₃₃=33.6×10^-12m/v) and the optimal optic system may lead to better time resolution less than 1ns.

We report a preliminary experiment result of poling in a femtosecond-laser-treated LiNbO3 (LN) crystal. Because of the precision of the femtosecond laser machine, we can control the width and depth of snicks exactly. Then the high-voltage electric field was operated on +z side of the LN crystal. The visible poling pattern was observed by a microscope and machining parameter influences the electric field distribution in the crystal. This approach does not need expensive masks and the periods of PPLN can be adjusted flexibly. Although the mechanism of core formation is not very clear yet, we believe that it would provide a novel technique of making prototype PPLN and other periodically poled crystals.

A stable passively Q-switched Nd: YVO₄ laser was demonstrated by use of a GaAs absorber grown at a low temperature (LT GaAs absorber) by the Metal Organic Chemical Vapor Deposition (MOCVD) technique, as well as an output coupler. The shortest pulse duration measured was about 12 ns with a single-pulse energy of 4.84 μJ, and the highest average output power is 1.16 W. The repetition rate is 360 KHz, which corresponds to the pump power of 2.8W.

We present a fiber based source of entangled photon-pairs in the 1550 nm telecom band that can be integrated into the existing fiber network and is well suited for quantum information processing. With this source we have demonstrated the generation, storage, and long-distance distribution of polarization entanglement in standard optical fiber. We have also investigated the origin of the large number of accidental coincidences in the experiments, which has been proved to be Raman scattering, and discussed how to suppress the Raman scattering to improve the quality of the fiber source.

CO₂ and YAG lasers are routinely capable of producing cw output powers in the range of sub-W to multi-kW and as such have become the mainstay of the laser cutting and welding industry. Despite numerous advancements in their design, these lasers are still typified by poor wallplug efficiencies (typically 1-10%) and/or relatively poor optical beam qualities. On the other hand, recent developments in laser diode technology, fiber design and beam combining techniques have meant that cladding pumped ytterbium-doped fiber lasers have attracted growing interest as a route to highly efficient (20-40% wallplug efficiencies), high output power, high beam quality (near-diffraction limited) lasers for a vast array of material processing applications. More specifically fiber lasers have a number of distinct advantages over their more conventional alternatives including size, reliability, wavelength selectivity, heat dissipation, wallplug efficiency, and operational cost. Furthermore, they can be operated without the need for active cooling or optical alignment. In this paper we review the recent progress in fiber design that is facilitating the scalability of the output power of fiber-based lasers and amplifiers.

In this paper the photonic crystal fiber (PCF) with square-lattice array in a silica matrix has been put forward. The PCFs studied in this paper have a solid core, obtained by introducing a defect that is by removing two holes at the center of the fiber cross section and with smaller air-holes in the same direction. For the different wavelengths in the range between 700 nm and 1600 nm, the effective index n_eff of the PCF fundamental mode has been obtained by the fast multipole method. The model field, birefringence, and confinement loss of the fibre fundamental mode are simulated. It is found that lower confinement loss and higher birefringence can be realized in the condition of fewer rings of air holes. The simulation results in this paper are important for instructing the fabrication of birefringent photonic crystal fibers.

Tunable photonic bandgap fibers (PBGFs) were theoretically investigated by using the vector plane-wave expansion method and the vector finite element method. The tunable PBGFs are fabricated by filling a high-index material in the air holes of index-guiding photonic crystal fibers. The wavelength dependence of leaky loss and group velocity dispersion (GVD) has been illustrated. We show the leaky loss in the tunable PBGFs can strongly depend on the refractive index of filled material due to the photonic bandgap effect. The tunable attenuator which operates at 1550 nm is designed based on this PBGFs.

Novel highly birefringent photonic bandgap fibers (PBGFs) are obtained by filling high index material in the air holes of total internal reflection birefringent photonic crystal fibers. The effect of filling high-index material on the transmission characteristics has been theoretically investigated. The photonic bandgap has been achieved by using the plane-wave method. Moreover, the polarization mode dispersion (PMD) has been studied by a full-vector finite-element method. Numerical results show that very high PMD with magnitude of order of 10^-10 has been respectively acquired, which is much higher than those of the non-filled fibers. Furthermore, strong coupling between surface modes and the fundamental modes has been found in the bandgap of the birefringent PBGFs.

In this work, 10 GHz over 58 nm 20 dB bandwidth supercontinuum (SC) is generated in 4.5 km dispersion-shifted fiber (DSF) pumped by a 10 GHz regeneratively mode-locked fiber laser (RMLFL). The optimized SC with 51.5 nm 20 dB bandwidth is obtained by properly adjusting the pump power. A 25 nm spectral region from 1535 nm to 1560 nm and a 12nm spectral region of 1567-1579 nm with ±3 dB uniformity are generated. 10GHz, over 30 channels were sliced by an array waveguide grating (AWG) with 100 GHz channel spacing. Timing jitter of each channel is less than 1.25 ps.

A novel type of distributed fiber optic sensor is presented to detect and locate a time-varying disturbance along the whole fiber length. The sensor consists of a Mach-Zehnder interferometer (MZI) where two directional optical signals are simultaneously traveling. This is achieved by a fiber-loop in the configuration, causing one MZI to operate as two co- and counter- propagating interferometers. The sensor is illuminated by only one continuous-wave laser, which is different from the conventional scheme. The phase shift caused by any disturbance around the fiber is detected and converted to the information on the perturbation position and amplitude, which can be determined by combining two phase signals from the above device. In addition, passive homodyne demodulation with application of a 3×3 fiber optical coupler for recovering a signal of interest from the MZI is also described. This technique can yield a large dynamic range with phase amplitudes for its symmetry. Also, high speed digital processing (DSP) technology is used in the sensor system, which can carry out all operations in real time and promote the resolution of localization. The system has capability for cross-correlation algorithm and fast Fourier transform (FFT) analysis of the detected disturbance signals, for the sake of determining the disturbance position and type. Finally, a positional resolution better than 100 m has been theoretically discussed and experimentally demonstrated in a system over 6.8 km fiber length.

Because an ytterbium-doped fiber amplifier has broad-gain bandwidth, high output power, and excellent power conversion efficiency, it has attracted much attention. Based on two-level model, the gain and propagation characteristics of Yb-doped fiber amplifier at 1064nm by dual-wavelength pumping (910nm and 975nm) were analyzed after introducing the overlapping factors and ignoring the effect of ASE and the fiber loss. In order to compare, other two pumping schemes were discussed, namely, 910nm pumping and 975nm pumping respectively. The differences between the Yb- and Nd-doped fibers for amplifying the light of 1064nm were also discussed.

We report a new fiber-optical solution concentration sensor based on an etched long-period fiber grating with one opening head coated with a silver film, which is written symmetrically by three beams of focused high-frequency CO₂ laser pulses in a standard single-mode fiber (Corning SMF-28). This sensor is tested with two kinds of solution, i.e. sugar and CaCl₂, and the experimental results show that this etched fiber-optical sensor with one opening head is flexible to operate and has very high resolution. The resolution of this sensor increases with approximate linearity, as the concentration of solution increases, and the resolutions for sugar solution at low and high concentrations are 1.65 (g/l) and 1.36 (g/l), respectively. Its average resolution has been improved above 28% after the cladding radius of this sensor is reduced to 61μm by etching with HF acid. The method of etching with HF acid can also be used to tune the resonant wavelengths of a long-period fiber grating.

A novel super-fluorescence fiber source (SFS) in the wavelength range that covers the C+L-band with stable spectrum, high output power and high slope efficiency is demonstrated. The topology consists of two stages: the first stage provides primarily the L-band gain spectrum whilst the second is solely for the C-band. These two stages are combined in series. We merely used the low erbium-doped fiber (EDF) 10.8m long as a gain medium in the second stage, which was pumped backward by a 980nm semiconductor laser diode (LD). As a result, the Amplified Spontaneous Emission (ASE) spectrum had a threshold of 5.1mW and a slope efficiency of 13.48% in the C-band. With a 980 nm LD pumping forward at 11-m-long moderate EDF and 58-m-long low EDF in series, L-band ASE was yielded with 52.3 mW threshold and a slope efficiency of 35.82% in the first stage. The ASE in the L-band was put into the second stage through an isolator and combined with C-band ASE as an output then a broad ASE spectrum of C+L-band was observed. The maximum output power of the C+L-band source was 33mW (15.3dBm) and its power stability was better than ±0.02 dB. Through the optimization of fiber parameters and the adjustment of pumping power in both stages, the flatness of ASE spectrum was improved. The optical bandwidth for the SFS output power, which ripples in 3 dB, was 63.7 nm (1536.92-1600.62 nm) of optical bandwidth without gain flattening filter.

In this paper the parameters of a CW double-clad fiber laser are theoretically analyzed, which is of important references to designing the kindred double-clad fiber lasers. This paper also offers a personalizing design scheme of double-clad fiber lasers, with the influence of some important parameters on the output power clarified.

Based on the analysis of conventional L-band EDFA we demonstrate a novel structure that improves the L-band amplification performances. In conventional L-band EDFA, 35nm (±1dB) flattened amplification bandwidth from 1565 to 1600 nm is obtained. Small signal gain of 25dB is achieved at 1590 nm with input power of -30dBm and the saturated output power reach to 10dBm. The linear relation between pump power and signal power are demonstrated experimentally. By utilizing a fiber Bragg grating (FBG) with center wavelength of 1550 nm, the population inversion of EDFA is intensified, which gives rise to gain improvement. Experimental results show that the signal gain is enhanced by the FBG by more than 8dB. The use of FBG has also shown a better performance in gain clamping. The amplifier gain is clamped at 25dB with a gain variation of less than 0.5dB for input power as high as -15dBm.

The dynamics of a two-mode model of the erbium-doped fiber laser is analytically and numerically investigated. Apart from the intrinsic nonlinearity of a two-mode laser, the existence of erbium ion pairs introduces a supplementary nonlinearity with a saturable absorber action. The laser is described through six coupled differential equations. Steady states are analyzed, their dependence on the laser parameters, and the onset of the self-pulsing and chaotic states is pursued.

In this paper, the application in return-zero (RZ) light source based on regeneratively mode-locked fiber laser (RMLFL) is demonstrated. By using Mach-Zehnder modulator and pseudo-random code modulation, the back-to-back eye-diagrams with good performances are obtained and RZ format light source with pulse duration of 5ps based on RMLFL is achieved, which quite meets the requirements of 10 Gbit/s long haul transmission. Furthermore, such a 10 GHz RZ source could be multiplexed into 40 GHz RZ light source simply by 4 fold optical time division multiplexing (OTDM).

This paper begins with a brief description of the nonlinear Schroedinger equation (NLSE). Then split-step Fourier transform method (SSFT) is used to simulate the pulse evolution when high order group dispersion and high order nonlinear effect are considered. Different phenomena induced by high order group dispersion are discussed, where the effect of the polarity of the pulse chirp, the third-order dispersion coefficient, and the effect of the high-order dispersion on self-phase modulation (SPM) are also discussed, respectively. For the high power density in the wavelength diversion multiplexing (WDM) system, some phenomena induced by high-order nonlinearity such as self-steepening are simulated, and qualitative analysis is given at the same time.

Using a plane wave expansion method, we numerically show that the bandgaps of honeycomb photonic bandgap fibers can be effectively tuned by interstitial air holes. It is shown that the number of bands varies with the filling fraction of interstitial air holes, and also that the width of the bands can be substantially increased by interstitial air holes. For a honeycomb photonic bandgap fiber with a large air filling fraction, the widths of the primary and secondary bandgaps are increased twice by interstitial air holes. In the case of a small air filling fraction, the relative sizes of the two gaps can reach up to 11.7% and 6.4% using interstitial air holes. Therefore, using interstitial air holes proves to be an effective way to tune the bandgaps of honeycomb photonic bandgap fibers.

Lithium niobate-based Y junction waveguide multifunction integrated optic chip (M-IOC) is one of core optic components in fiber optic gyro. Its performances have very important influences on fiber optic gyro (FOG) accuracies. Fully understanding its performances in FOG will help us to further improve fiber optic gyro performance (bias drift < 0.01°/h). The paper focuses on the studying of application features of Y-junction integrated optic chip(Y-IOC) in FOG with bias drift less than 0.01 °/h. It also describes the system test methods for checking integrated optic chip performances. Interesting research points are: Y-IOC optic technology and its application in FOG, features of Y-IOC in FOG, the corresponding test system and the matching of Y-IOC with FOG system. Research results show: Y-IOC optic performances, Y-IOC electrical characteristics, as well as the matching between electrical circuits of FOG system and electrical features of Lithium niobate- based Y junction waveguide multifunction integrated optic chip all together play very important roles in achieving high performances of FOG.

A Fiber Bragg Grating (FBG) sensor demodulation system of matched Bragg gating with a simple structure is studied. The cantilever is designed by using acryl glass, and the wavelength's linear tuning of FBG is operated by tuning the cantilever of acryl glass. The sensor signal of Real-time detect is detected by using an oscilloscope after the optical sensor signal enters the photoelectric detector. Demodulation operation of matched-gating is achieved. Wavelength resolution of 2.33pm is obtained. The advantages of this demodulation system is simple structure, electromagnetic interference resistance, high wavelength resolution and high rate of cost performance, so it has a great application prospect.

In this paper, we propose and demonstrate a novel torsion sensor based on a long-period fiber grating (LPFG) induced symmetrically by three beams of focused high-frequency CO₂ laser pulses (20~24 kHz). Experimental results show, when the LPFG is twisted clockwise, the resonant wavelength shifts toward shorter wavelengths, and the peak loss decreases. When the LPFG is twisted anticlockwise, the resonant wavelength shifts toward longer wavelengths, and the peak loss increases. The resonant wavelength and the peak loss are all similarly proportional to the twist angle applied. Based on this LPFG, a new kind of fiber-optic torsion sensors can be made, which can not only directly measure the torsion angle, but also determine the torsion direction simultaneously by means of measuring the shift of the resonant wavelength and/or the peak loss of the LPFG.

It is well known that PCFs, compared with traditional fibers, have many special characteristics. Because of these, PCFs have been used in sensors. Based on the principles of traditional fiber sensors, this paper analyzes the probability of applications of PCFs in gas or liquid sensors. Due to the multi-hole properties of the PCFs, we can make some gas or liquid sensors by putting different gases or liquids into its air-hole. In this paper, the sensitivity of PCF sensors was simulated by the multipole method. By changing the core radius of PCFs or the gas fraction in the hole of PCFs, we can improve their sensitivity. Due to the superiority in their configuration, the research on PCF sensing characteristics will surely promote the development and application of the new-type sensors and other photoelectron devices.

An efficient shooting algorithm based on the simple-shooting method and the modified Newton-method for fiber Raman amplifier design is proposed. By introducing the Broyden's rank-one method, the time-consuming calculation of the Jacobian matrix is dramatically relieved. Numerical simulation results show that the simulation efficiency of the proposed method has been improved more than 70 percent compared with the conventional shooting method.

A new numerical optimization method assisted by functional model and improved Newton's iterations is proposed. The method would be applied to the pump power configure optimal design for the gain flatness optimization in the laser-diode discrete Fiber Raman amplifier (FRA). Compared with other optimization algorithms, the proposed improved Newton's iteration has fast convergence speed and good stability. So the method can exploit better solutions and greatly shorten the amount of run time. Two samples show the feasibility of the method with different initial variables. The comparison with Genetic algorithm (GA) is obtained.

In this paper, a numerical method is presented to design a biconical waveguide. We use this method to calculate the transmitted and radiated power of the biconical waveguide in an integrated acousto-optic modulator, then we plot the normalized power loss curves versus the taper length and get the optimum design.

In this letter, a S-band discrete Raman fiber amplifier(RFA) with the maximum net gain of 18.3dB was obtained. Using the gain flattening filter based on a cascaded long-period fiber grating (LPFG), the gain profile of this RFA was flattened to within ±1dB in the wavelength range from 1 480 nm to 1 515 nm, and an average gain of 11.3dB was achieved.

A broadband source (1520-1620nm) with output power of more than 200mW is achieved using only a single-wavelength pumping scheme dependent on the combined effects of stimulated Raman scattering, parametric process, and Erbium-doped fiber amplification.

Photonic-crystal fibers provide efficient nonlinear-optical transformations of femtosecond Cr: forsterite laser pulses, delivering linearly chirped frequency-shifted broadband light pulses, optimized for pump-probe nonlinear absorption spectroscopy of molecular aggregates. The blue-shifted output of a photonic-crystal fiber with a spectrum stretching from 530 to 680 nm is used to probe one- and two-exciton bands of thiacarbocyanine J aggregates in a polymer film excited by femtosecond second-harmonic pulses of the Cr: forsterite laser.

Using an efficient, full-vectorial numerical simulation method, we analyze the existent conditions of the photonic bandgap (PBG), which is further demonstrated through experimental research. We utilize the method of transmission spectrum to measure the hollow-core microstructure fibers (HC-MSFs) in the visible and near infrared regions. The signal is obtained by detecting light from the end of the fiber. The experimental results indicate that there are several strong transmission bands in the near infrared region, but hardly any bandgaps in the visible region. Furthermore the attenuation in the visible wavelength is very considerable. The parameters of the HC-MSFs structure used in the measurement are the distance of nearest air holes pitch Λ (2.65μm), the diameter of air holes in the cladding d (2.10μm), and the central air core diameter (8.37μm). The spectrum positions of the bandgap in the spectrogram are 2297nm, 2406nm, and 2525nm, respectively. The repetition of the experimental results is fine.

The Stacking-capillary Method is widely used in the fabrication of the Micro-structure Fiber (MSF). We describe an improved stacking-capillary method, which can fabricate the MSF without interstitial holes. The method includes several steps. Firstly, the MSF preform is made by the stacking-capillary method; secondly, the MSF preform is put into the high temperature furnace to heat at 1600°, then the positive pressure is produced into the capillaries by adding air, every three adjacent capillary holes are expanded in the preform, and the interstitial holes are eliminated, all the capillaries are fused together, although the round capillary have changed to hexagon, the size of the prefrom does not change, and still keeps very good structure. The step of eliminating the interstitial hole can help to keep the MSF structure during drawing fiber. We get well results by the Improved Stacking-capillary Method.

The light is the ultimate energy source for the life on the earth and also it is with the help of light that we see and understand the world surrounding us. The energy quanta are called photons. In addition to illuminating, imaging, and spectroscopy, a typical example of the recent uses of photons in our daily life is optical communication. A more and more important new role that photons are playing is manufacturing, particularly in three-dimensional micro-nanofabrication on the basis of nonlinear photon-matter interactions. This article reports our recent progress in this research direction: the challenge to the spatial resolution of fabrication, and the utilization of the technology on nanophotonic and micro-nanomechanical devices.

We report on an experimental investigation into the recording and readout of 3D high-density optical data storage in transparent optical glasses using a femtosecond laser. A compact Ti:sapphire ultrafast laser amplifier as writing laser was developed to output 100mW 2.1ps at 1KHz with 1W pump laser. Laser pulses modulation is realized by modulating two circuits of trigger pulses signal which are used to control laser pulses trapping and switching out from cavity, respectively. A formatted and coded writing was realized in fused silica and doped polymethylmethacrylate(PMMA) bulk. A phase-contrast and a reflection confocal reading system with software are developed for optical readout of multi-layer data bits in parallel.

A new laser plasma technique for non-vacuum deposition of thin films has been developed recently. This technique was successfully applied to deposit hard carbon coatings on steel substrates. The obtained diamond-like carbon (DLC) coatings with amorphous structure demonstrated high nanohardness and good adhesion to the substrate. Here we present new experimental data concerning parameters of the produced carbon plasma and features of the plasma-substrate interaction, which are necessary for further progress of this technique. Spectroscopic study of the plasma emission was performed for picosecond and nanosecond laser pulses at different wavelengths (λ=1078 nm, 539 nm and 248 nm). The laser plasma composition was studied and conditions for the occurrence of low-threshold air breakdown near the target surface were determined. The ranges of the substrate-target distances, which ensure carbon film deposition or plasma etching of the substrate, were determined. The revealed influence of the laser pulse energy, pulse duration and wavelength on the plasma-substrate interaction is reported.

Dynamics of self-pumping photorefractive double phase-conjugate mirrors is considered. It is shown that depending on experimental conditions as phase-conjugation with efficiency up to 60-80 % as formation of unstable structures of soliton-like filaments can be realized.

New porous silicon preparation technique has been suggested and realized with using vapor etching of silicon in iodine and HF contained vapors. It has been shown that vapor etching allows the preparation of luminescent porous layers on heavy doped (n⁺⁺ and p⁺⁺ type) silicon. Comparison of Raman and CW excitation PL measurements of vapor etched porous layer with typical anodized luminescent porous silicon indicated that they have in general similar structural and PL properties. Time resolved photoluminescence measurements reveal however that excitation recombination for iodine contained vapor etched samples is two times faster with higher photoluminescence efficiency, which can be interpreted as increasing of radiative recombination rate for luminescence centers in new nanocrystalline silicon.

Pulsed Laser Deposition of indium tin oxide in the femtosecond regime has been performed in our laboratory. Plume diagnostics has been carried out by means of a fast Intensified Coupled Charge Device (ICCD) camera. Optical emission spectroscopy has been applied to characterize the transient species produced in the femtosecond regime. The time evolution of emission lines has been discussed and compared with that obtained for nanosecond regime. The films, deposited on glass substrates, has been analysed by scanning electron microscopy, energy dispersive x ray analysis, x ray photoelectron spectroscopy and x ray diffraction. The results obtained from femtosecond ablation show that the processes in the short pulse regimes are very different from the nanosecond one. In particular the plume angular distribution shows a characteristic high cosine exponent and the composition of the deposits is completely stoichiometric and independent from the laser fluence, even if to obtain crystalline films a substrate temperature of 600 °C is needed. Preliminary data indicate that the deposited films are suitable for gas sensor applications.

It is investigated that the duration of stimulated Brillouin scattering (SBS) amplification of Stokes pulses is smaller than the acoustics lifetime and the oscillation period of hypersound with the nanosecond pulse as pump light. In the experiment, the Fluorinert (FC-72) is used as the Brillouin amplification medium, the Stokes pulse duration less than 1/4 of oscillation period of hypersound of FC-72 as seed pulse is achieved by two-stage SBS compressor to compress 8ns laser pulse. The experiment and simulation show that, in this case, their energy amplification is efficient, but the duration of amplified pulses is longer than the initial seed pulse. Simulation results of higher energy than the maximum energy in the experiment show that at sufficient high pump or initial seed pulse power intensity, the duration of amplified pulses can be no larger than the initial seed pulse duration., even less than the seed pulse duration.

A novel cascaded χ⁽²⁾ wavelength conversion of picosecond pulses based on sum frequency generation and difference frequency generation (SFG+DFG) is proposed and experimentally demonstrated in LiNbO₃ waveguides. The signal pulse with 40-GHz repetition rate and 1.57-ps pulse width is adopted. First of all, high conversion efficiency about -18.93dB can be achieved with low power level required for both two pump lights, which is greatly enhanced approximately 8dB compared with the conventional cascaded second-order nonlinear interactions (SHG+DFG) with a single and much higher power pump. Secondly, the wavelength of the converted idler wave can be tuned from 1527.4 to 1540.5nm when the signal wavelength is changed from 1561.9 to 1548.4nm, and about 13.1nm converted idler bandwidth is achieved with the conversion efficiency higher than -31dB. Thirdly, two pump wavelengths can be separated as large as 17.3nm. Meanwhile, when one pump wavelength is fixed at 1549.1nm, the other can be tuned within a wide wavelength range about 7.6nm with the conversion efficiency higher than -34dB, which is much larger than that in the SHG+DFG situation. Finally, the temporal waveform of the converted idler pulse is observed with rather clear appearance achieved, and no obvious changes of the pulse shape and width are found compared with its corresponding original injected signal, showing that our proposed scheme exhibits a very good conversion performance.

Theoretical analysis on phase matched second-harmonic generation (SHG) in a more general manner is presented from the viewpoint of modulation of the nonlinear coefficient. The conventional quasi-phase matching (CQPM) is firstly analyzed, showing a noticeable result that the intensity of second-harmonic wave grows most rapidly at odd times of half Lc, i.e., Lc/2, 3Lc/2, 5Lc/2, etc., but most slowly at even times of half Lc, i.e., 0, Lc, 2Lc, 3Lc, etc. Based on which, a new scheme to realize phase matched SHG by modulating the nonlinear coefficient using alternating "+", "0", "-", "0" every Lc/2 is proposed with the analytical expressions for conversion efficiency (η) and conversion bandwidth (Δ_FWHM) derived. Compared with CQPM, it is found that the same Δ_FWHM approximately 5.566/L, and at the same time, a much more effective η can be achieved when ignoring partial waveguide in which the nonlinear coefficient is "0", however, when considering it there is a 3dB penalty for η. Furthermore, a much more general scheme to realize phase matched SHG is proposed, also using alternating "+", "0", "-", "0" but with variable duty cycle. We present the analytical expressions for η and Δ_FWHM , with the result showing that Δ_FWHM keeps constant, i.e., 5.566/L. In addition, the fabrication tolerances are discussed, and our proposed scheme exhibits a better performance.

A new method to reduce the residual pulse peak in stimulated Brillouin scattering (SBS) optical limiting by an injected seed is presented. The pulse shapes of the transmitted pump beam are studied for various delay times and injected seed powers theoretically. The numerical simulated results show that the height of the residual pulse peak can be controlled by changing the delay time and injected seed intensity. Experimentally, various optical limiting pulse shapes with different heights of the residual peak are observed by changing the delay time and intensity of the seed injected into the SBS optical limiter. The experimental and theoretical results show that the residual pulse peak in SBS optical limiting can be controlled by the delay time and power of an injected seed.

Based on the multi-axial theory of stimulated Brillouin scattering (SBS), a group of coupled wave equations without limitation to the mode separation of a laser is deduced when the random initial phases of the axial modes are introduced for SBS pumped by a free operation laser. The coupling coefficient between two axial modes is obtained, and the dependences of the threshold and reflectivity of broadband SBS on the bandwidth, the mode number, the mode separation of the laser, are simulated numerically. The SBS threshold and reflectivity are dependent not only on the laser bandwidth, but also on the mode number and the mode separation of the laser.

The issues of modeling the semiconductor laser with circuit elements have been addressed by several authors. Previously reported laser models are based upon rate equations and many relations between electron and photon. This needs the deep understanding of the clear physical meaning of all variables and the electrical and optical properties of the active region of a semiconductor laser diode. The relations and the formulas are very complex and the model's parameters are hard to get. In the previous models, node voltages are used as circuit variables. In practice the LD is powered by a constant or pulse current source. The node voltages are only measurable variables and decided by the input current for a ready-made LD. So taking the input current as the circuit variable is very simply and easy to understand and use. In this paper a circuit model of DHLD has been developed from rate equations and the well known models. Simulink module is built according to the rate equations. The input current is the only variable for the module. The transient response of the LD model is plotted. The simulating results such as transient and frequency responds agree well with the reported data.

Characteristics of transmitted temporal profile of active stimulated Brillouin scattering (SBS) optical limiting are studied. Methods of impairing transmitted residual peak power and shortening the delay time of SBS are proposed. Based on coupled wave equations of SBS process, physical models of optical limiting based on active SBS are constructed. Using an implicit finite differencing in time and a downward differencing scheme in space, theoretical calculation models of active SBS limiting for numerical simulations are built up. Three parameters, i.e. residual peak power, delay time and limited power are defined renewably to characterize active SBS optical limiting. The relationship of transmitted temporal profile to seed peak power is obtained by numerical calculations. The influence of the injected seed power on transmitted residual peak power is analyzed in detail. It is shown that the residual peak power is weakened with strengthening of the seed power. By selecting suitable power of seed, residual peak is removed and consequently the ideal flat pulse output is achieved. On the basis of theoretical analysis, active SBS optical limiting temporal characteristics are examined experimentally with a 20 ns Nd:YLF laser and one nonlinear medium (CCl₄). The experimental results are compared with analytical and numerical calculations.