Integrated optics components based on second-order nonlinearities are provided by structures presenting departure from centrosymmetry, both at microscopic and at macroscopic levels. In a non-exhaustive review, we consider optical methods for the evaluation of molecular first hyperpolarizabilities, presenting Hyper-Rayleigh Scattering (HRS) as a more direct method than inference from the standard Electric Field-Induced Second Harmonic Generation (EFISH). From a macro perspective on films, also alternatively to EFISH, we consider determination of second-harmonic susceptibilities from the Maker fringe technique, as well as of electro-optic (EO) coefficients by several methods, using a dc/modulated electric field and light in either transmission or reflection, external or internal. As a case study from our laboratory, with the standard p-nitroaniline as reference, we illustrate application of HRS to the newer 2-(2’-hydroxy-4’-aminophenyl)-6-nitrobenzoxazole (HBO-BO6) molecule. The guest-host systems formed by its incorporation, as well as of commercial chromophores, into silicate sol-gel film matrices, were subjected to Corona poling. Application of the Poling Optical Polarimetry (POP) method to these systems during poling has revealed significant birefringence. For the resulting oriented structures with high-hyperpolarizability ingredients, such as the HBO-BO6 molecule, POP results have indicated the importance of EO characterization methods with explicit consideration of anisotropy, associated with very high EO coefficient materials.

Optical waveguides in lithium fluoride (LiF) crystals have been obtained by He⁺ ion beam irradiation. The waveguides have been characterized by several techniques. In particular, we describe here the application of confocal microscopy to their characterization and show the first results obtained. We have also carried out a preliminary evaluation of the potential of this technique for the assessment of structural and spectroscopic characteristics of the waveguides.

The problem of directed waveguide mode scattering in the irregular planar optical waveguide (PWG) is solved with the help of the theory of perturbations. The solution of the inverse waveguide scattering problem consists in restoring of autocovariance function and determination of irregularities' parameters by the measuring data of scattering diagram in the far zone. The computer simulation allow approving, that our method permits to receive approximate correct solution of the inverse problem with rms error of restoring of the given autocovariance functions no more than 35% in the presence of high real noise (SNR greater than or equal to 1). The statistic parameters of irregularities in this case can be determined with an error less than 15-30 %.

We proposed and demonstrated a micromachined filter with a strain control layer, which gives us novel functions including temperature insensitive operation, thermal wavelength tuning, wavelength trimming and 2-D multi-wavelength integration. In this paper, we present the design and the fabrication of micromachined thermally tunable filters with a low tuning voltage. In our micromachined micromachined filter, an air gap is formed between GaAlAs/GaAs DBRs with an upper DBR mirror freely suspended above the substrate by a cantilever structure. A novelty in our devices is to add a GaAs or GaAlAs thermal strain control layer on the upper DBR. We can freely control the temperature dependence of the proposed MEMS cavity. Either temperature insensitive operation or wide wavelength tuning induced by temperature change can be realized. Also, we fabricated a micromachined thermally tunable filter with a heating element. There are two electrodes integrated on the top p-type doped strain control layer of this filter for heating the cantilever. When a voltage is applied between the two electrodes resulting in heating, the micromachined cantilever moves due to thermal strain. The proposed structure enables thermal wavelength tuning either for red shift or blue shift. The amount of wavelength tuning is controlled by the length and the thermal capacity of the cantilever. We can expect much lower tuning voltage than conventional electrostatic force tuning scheme. We measured the tuning characteristics of fabricated filters with changing an applied voltage between two electrodes. We could obtain blue-shift wavelength tuning of over 50 nm with an applied voltage of 6 V.

Preliminary analysis has shown that quantum dots enable tens of millivolt-range operation of phase-shifters in a semiconductor Mach-Zehnder interferometer modulator. Our methodology based upon the quantum dot experimental work of Hse et al, makes use of his measured exciton line shapes to estimate refractive index changes in a PIN structure in which the intrinsic laser is loaded with self-organizing quantum dots and their associated wetting layers. We consider both forward and reversed bias cases; in the former, the interferometer phase shift sections become DFB lasers, and in the latter, the phase shift is caused by the quantum-confined Stark effect (QCSE). With the latter, we found a trade-off between low operating voltage and modulating bandwidth. For a phase shifter insertion loss of 5 dB, a 250-micron long phase section will yield a pi/2 control voltage of 50 mV at a bandwidth of around 18 GHz. Ifi 90 mV control voltage swing can be tolerated, the modulator bandwidth increases to 30 GHz. If a resonant tunneling diode (RTD) is made part of the assembly, the local E-field is enhanced by a factor of 5 to 10, thereby reducing the drive requirements even further. Similar, though narrower bandwidth results were noted for the DFB laser phase modulator concept.

In this work we exploit growth as well as linear and nonlinear optical properties of long, parallel single-crystalline hexaphenylene (p-6P) nanofibers grown on mica surfaces. Typical widths and heights of these needle-like structures are a few 10 - 400 nanometers, whereas lengths of up to one millimeter can be achieved. The nanofibers allow us to perform experiments at either densely packed, well-aligned bunches of aggregates or at isolated entities. Linear optical properties are probed by local spectroscopy using a fiber-optic spectrometer and by guiding UV-light through individual fibres and relating the waveguiding efficiency to their morphology using atomic force and fluorescence microscopy. Results are compared with an analytical theory. As nonlinear optical probes we use two-photon luminescence as well as optical second harmonic generation induced by ultrashort laserpulses in the near-infrared spectral range.

Wide-band-gap semiconductor-doped-glasses were obtained by synthesizing SnO₂:SiO₂ nanostructured glassceramics. In this binary system, comprising two chemically compatible oxides, crystalline SnO2 nanoclusters were embedded in a pure silica matrix in a controlled way, by setting appropriate thermochemical parameters, up to 10% of volume fraction of the semiconductor crystalline phase. Measurements of third order non-linearity were carried out by means of z-scan technique at 1064 nm finding a non linear refractive index comparable with that of glasses doped with Cd chalcogenides. Optical spectroscopy, micro-Raman scattering and electron microscopy indicated good optical and nano-structural features, suitable for stable optical applications, both in bulk and film samples.

Tin(IV) doped SiO₂ xerogel and SnO₂/SiO₂ glass-ceramics were synthetised by sol-gel route. Synthesis and physical properties of those materials will be disclosed in this paper. Particularly, the solubility behaviour of tin(IV) in SiO₂ and the oversaturation condition in glass and xerogel intermediate were investigated. Both materials (glass and glass ceramics) are useful in photonics, and different application will be proposed in this work.

With the rapid growth of the telecommunications industry over the last 5 to 10 years has come the need to solve ever more complex electromagnetic problems and to solve them more precisely than ever before. The basic EME (EigenMode Expansion) technique is a powerful method for calculation of electromagnetic propagation which has been well known amongst academic environments and also in microwave fields, representing the electromagnetic fields everywhere in terms of a basis set of local modes. It is at the same time a rigorous solution of Maxwell's Equations and is able to deal with very long structures. We discuss here progress that the authors and others have made recently in applying and extending it to integrated, fibre, and diffractive optics - including development of efficient ways of modelling tapers and other smoothly varying structures, new more efficient boundary conditions and improved mode finders. We outline the advantages it has over other techniques and also its limitations. We illustrate its application with a variety of real life examples, including diffractive elements, directional couplers, tapers, MMI's, bend modelling, periodic structures and others.

We review key algorithms for the numerical solution of waveguide eigenvalue problems and discuss their application to a typical simulation problem in integrated optics - the computation of eigenmodes of a MQW laser structure. Here we focus mainly on a self adaptive realization of the codes supplying solutions with prescribed accuracy in a CPU-time as short as possible. In a brief outlook we show how to extend these principles to solve general time-harmonic and time-dependent scattering problems in an adaptive finite element context.

Design issues such as optical transmission, interference mechanisms and splitting ratio of a compact parabolically tapered Multimode Interference-based 3dB power splitter on an InP-based deeply etched ridge waveguide, using the finite element-based beam propagation method approach, are presented.

We discuss the use of multiple layer air trench and silicon strip structures to realize high efficiency 90° bends for low index contrast waveguides. We use a micro-genetic algorithm (mGA) coupled with a 2-D finite difference time domain method to perform rigorous electromagnetic optimization of multi-layer structures for single mode waveguides. We find that a 3-layer air trench structure can be designed for a 90° waveguide bend that exhibits 97.2% efficiency for TM polarized light at a wavelength of 1.55 μm. We are also able to design five- and six-layer silicon strip bends that have high efficiency for both TE and TM polarizations. For example, simulation results for a six-layer design show 95.2% and 97.2% for TE and TM polarizations, respectively. Moreover, the bend efficiency for each polarization state is greater than 90% over a broad wavelength range (1.5 μm to 1.7 μm).

Starting from Maxwell’s equations we derive a new orthogonality relation and a recirpocity theorem for photonic crystal waveguide modes. The orthogonality relation implies an integration over a plain nonparallel to the waveguide direction. This is in contrast to the common orthogonality relations where an integration over the whole photonic crystal is needed. By means of this new orthogonaly relation coupling coefficients between different photonic crystal waveguides as well as photonic crystal waveguides and conventional wavguides are deduced. The considerations are valid for 2D as well as 3D geometries. In the case of monomode waveguides simple quasi-analytical approximations for the reflection and transmission coefficients are obtained, which generalize the well known formulas for coupling between conventional waveguides. The reciprocity theorem can be used for the efficient simulation or even analytical description of photonic crystal waveguides in the presence of arbitrary perturbations. Depending on the kind of perturbation and the number of modes a set of strongly coupled discrete equations for the field amplitudes in the photonic crystal unit cells results. In the frame of this paper we study two problems by means of this theory. First we investigate the transission - reflection - problem of two consecutive photonic crystal waveguides analytically. Furthermore we study the influence of the dispersion of the dielectric function of the photonic crystal material on the band structure. Especially a polariton lead to strong deformation or even a splitting of the band structure close to an optical resonance.

Recent advances within the realization of silica-on-silicon planar waveguide circuitry are presented. This ranges from the production methods for planar waveguides, including a novel method based on the utilization of focused UV-laser beams for direct waveguide imprinting, to the functionalities that are embedded into the glass materials and waveguide circuitry. The latter includes e.g. optical functions that are realized utilizing phase and amplitude of and resonance conditions for light in waveguide circuitry. We also describe the realization and use of gratings in waveguides. Furthermore, we discuss doping of glass materials used to obtain e.g. amplifiers, lasers, and the pursuit to obtain highly non-linear materials in order to realize purely glass-based switches, modulators and wavelength converters.

Silicon-On-Insulator integrated optics boasts low loss waveguides and tight optical confinement necessary for the design of nanophotonic devices. In addition, the processing is fully compatible with capabilities of standard silicon foundries. Because of crystal symmetry, silicon does not possess 2nd order nonlinear optical effects. However, the combination of nanoscale geometries with the high refractive index contrast creates high optical intensities where 3rd order effects may become important, and in fact, may be exploited. In this context, we study the two main nonlinear processes that can occur in silicon waveguides, namely Stimulated Raman Scattering (SRS) from zone-center optical phonons and Two-Photon Absorption (TPA). Because of the single crystal structure, the Raman gain coefficient in silicon is several orders of magnitude larger than that in the (amorphous) glass fiber while its bandwidth is limited to approximately 100GHz. To achieve Raman gain in the 1550nm region requires the pump to be centered at around 1427nm. We discuss the Raman selection rules in a silicon waveguide and present the design of an SOI Raman amplifier. We show that by causing pump depletion, TPA can limit the amount of achievable Raman gain. TPA also limits the maximum optical SNR of the silicon amplifier.

Thermo-optical silicon-on-insulator (SOI) waveguide switch has been fabricated and characterized. The switch is based on a 2x2 Mach-Zehnder interferometer and 9 microns thick ridge waveguides. The extinction ratio of the switch is 17 dB with ultra-slow modulation and it is limited by the unoptimized directional coupler lengths. Thermo-optical switching with conventional on/off modulation was demonstrated up to 10 kHz. The average power consumption was 150 mW and the extinction ratio was 15 dB in 10 kHz square wave modulation. By using a novel modulation principle the maximum frequency was rised up to 167 kHz, while still maintaining the 15 dB extinction ratio in square wave modulation. With random binary modulation at 167 kHz frequency (3 μs per bit) the extinction ratio remained above 13 dB and the average power consumption was 590 mW. The obtained frequency limits for square wave modulation correspond to a maximum of 1% deviation from the attainable extinction ratio limits. With less strict extinction ratio requirements the maximum frequencies can be much higher. The new modulation method can be used to radically speed up interferometric switches with a tolerable increase in the power consumption.

Optical component coupling using self-written waveguides is one of the promising methods to improve the existing alignment control difficulties in optical component coupling. Using conventional UV curable resin as a waveguide material, waveguide spontaneously forms by UV light exposure from an edge of an optical fiber. When UV light is exposed from two faced optical fiber edges, self-written waveguide forms and connect the two optical fibers with decreasing its coupling loss between them. As one interesting feature, optical coupling of the fibers can be attained even significant gap and offset would exist between the fibers. The observed coupling loss after waveguide formation is significantly lower than that before the waveguide formation. Another remarkable feature is that the coupling technique requires no edge treatment of the fibers to be coupled because of similarity of refractive indexes between UV resin and the fibers. Buffer-coated optical fibers are thus coupled using this technique simply after cutting by conventional scissors. The coupling using all solid type self-written waveguides, which have stabilized cladding material can also be achieved though further investigation is necessary to attain low loss waveguides materials. Using this interesting coupling method, conditions required for conventional optical couplings, such as severe alignment control and edge treatment would be significantly relaxed in a optical component coupling procedure.

Tunable add-drop multiplexers are essential components for reconfigurable optical networks. Components that are based on electro-optic technology offer, in addition to add-drop capability, the desired features of rapid tuning and polarization-independent operation. A comparative overview of various technologies used in making such components, as well as the fabrication and results of a guided-wave electro-optic tunable add-drop multiplexer produced in LiNbO₃ at the 1550 nm wavelength regime are presented.

These last years, the growth of data traffic has increased the interest for broadband integrated optic devices. Their applications include, for example, the fiber communications on a single fiber by adding the transmission capacity of two optical telecommunication windows for Local Area Networks (LAN) and Wide Area Networks (WAN) or by combining pump and signal wavelenghts in rare earth doped intergrated optical amplifiers. A promising technology to realize those devices is ion-exchange on glass. Indeed, it allows the integration of different functions in a glass substrate with efficient results and a better compatibility in fiber systems with a low cost. We propose in this paper an original broadband duplexer based on a leaky structure. First, the physical principle of the component is explained. The core of the structure is a leaky zone which involves a non-resonant coupling and ensures a broadband spectral behavior to the component. Then, the broadband duplexer is presented and the focus is specially made on the improvement of the outputs crosstalk through the suppression of parasitical back reflections. Theoretical optimization and validation by simulations are presented. Finally, perspectives of this work are proposed.

Electrooptic polymer waveguide modulators have demonstrated excellent characterizations including high-speed (with a bandwidth up to 100 GHz) and lower drive voltages. One key factor to achieve efficient high-speed modulation in polymer devices is to optimize the microstrip line electrodes. Theoretical simulations of microstrip electrode designs have recently been performed. Results show that an electrode 3-dB bandwidth could be significantly improved by increasing the polymer cladding layer thickness between electrodes, reducing the electrode length, or by using copper or silver instead of commonly used gold. Electrode surface flatness is also of great importance for device performance. Even a slight roughness on the electrode surface induced in processing could cause significant reduction on both electrical and optical response bandwidths.

A method to increase the multiplex capacity of an arrayed waveguide grating (AWG) based interrogator is proposed that makes use of the spectral recurring feature of AWG due to its free spectral range. Using a single AWG, interrogation for two different wavelength regions of 1310nm and 1550nm was performed with a high performance of the standard deviation of the level of sub-pico-meter for 2nm dynamic range. The proposed method is very helpful to reduce the cost of the AWG interrogator.

We have devised a hybridization scheme that, given suitable Fabri-Perot (F-P) gain medium, allows us to fabricate small, mechanically robust single frequency lasers in a wide spectral range, limited only by the transparency of the SiON material. In this report we discuss device fabrication and present characteristics of a laser emitting in red spectral range. The laser operates at or near room temperature under continuous wave excitation and emits 5 mW of power in single mode with 40 dB side mode suppression.

This paper reports the development of the first Liquid Crystal Adaptive Lens (LCAL) with circular electrodes for imaging application. LCAL is an electro-optical device using a set of electrodes to grade the refractive index across its aperture. The principle of liquid crystal adaptive lens is briefly discussed. The special advantages and challenges in using circular electrodes are addressed. Numerical simulation is performed to predict the imaging performance. A prototype of LCAL was designed according to the requirements for application in a microscope imaging system with a diameter of 7.86 mm and a focal length adjustable from 0.2m to infinity. The structure of the LCAL with circular electrodes includes the top ITO ring electrodes and the connecting wires, insulating layer, ITO plugs, and the bottom ITO conducting wires. This prototype was fabricated on a glass substrate by using micro-fabrication process. The focusing performance of the LCAL with circular electrodes was presented. The experimental results agree well with the simulation results. The imaging experiment of LCAL is performed when LCAL focal length is 1m, 0.38m and 0.2m under incoherent source. The resolution of the images formed by LCAL combined with a fixed lens is indistinguishable from that of the image formed only by a fixed lens, while the contrast is lower.

We discuss an electro-optical device that acts as a multifunction logical gate based on a BSO photorefractive crystal. It is an easily reconfigurable device which can perform different logic functions such as AND, OR, NOT, NOR using the same configuration and changing only the controlling parameters.

A key attribute emerging in the optoelectronic component supply industry is the ability to deliver 'solution level' products rather than discrete optical components to equipment manufacturers. This approach is primarily aimed at reducing cost for the equipment manufacturer both in engineering and assembly. Such 'solutions' must be designed to be cost effective - offering costs substantially below discrete components - and must be compatible with subcontract board manufacture without the traditional and expensive skills of fibre handling, splicing and management. Examples of 'solutions' in this context may be the core of a multifunctional OADM or a DWDM laser transmitter subsystem, with modulation, wavelength and power management all included in a simple to use module. Essential to the cost effective production of such solutions is a high degree of optical/optoelectronic integration. Co-packaging of discrete components and electronics into modules will not deliver the cost reduction demanded. At Bookham Technology we have brought together what we believe to be the three key integration technologies - InP for monolithic tunable sources, GaAs for high performance integrated modulation and ASOC for smart passives and hybrid platforms - which can deliver this cost reduction, together with performance enhancement, over a wide range of applications. In the paper we will demonstrate and compare our above integration approaches with the competing alternatives and seek to show how the power of integration is finally being harnessed in optoelectronics, delivering radical cost reduction as well as enabling system concepts virtually impossible to achieve with discrete components. In the paper we will demonstrate and compare our above integration approaches with the competing alternatives and seek to show how the power of integration is finally being harnessed in optoelectronics, delivering radical cost reduction as well as enabling system concepts virtually impossible to achieve with discrete components.

An integrated scheme is proposed for low-loss coupling between a semiconductor laser diode (LD) and a single-mode fiber (SMF) by using a diffractive optical element (DOE). The DOE is designed based on diffractive optics principle, which phase distribution is optimization result of iterative phase retrieval algorithm. A new far-field amplitude constraint is introduced into the iteration to provide very high mode conversion quality. The scheme is applicable to realize high efficient coupling to SMF for any semiconductor LD. Coupling losses lower than 0.02dB have been reached for all the discussed LDs with aspect ratios of the elliptical fields from 1 to 9. The requirements on axial displacement and rotation angle have been removed. The tolerance for 1-dB loss increment for lateral misalignment is 0.9μm. And the coupling loss is insensitive to tilt angle.

In this paper, an original scale-reduction rule without sensitivity loss in integrated optic pressure sensors based on the elasto-optic effect is described. The sensor has a rectangular diaphragm as a pressure-sensitive mechanical structure and a sensing waveguide on the diaphragm. In this type of sensor, sensitivity is theoretically known to be strongly dependent on the dimensions of the diaphragm. According to the theoretical results, the sensitivity can be kept constant even if the diaphragm dimensions are reduced as long as both the side length ratio and the characteristic length remain constant. Here, the characteristic length is introduced as the sube of the shorter side length of the diaphragm divided by the square of the thickness. Such a scale-reduction rule would be very significant in the miniaturizing of a sensor without reducing sensitivity, but it has not been experimentally confirmed. In this study, the scale-reduction rule was experimentally examined using three fabricated sensors, which had the same side length ratio and the same characteristic length. The exact dimensions of the sensors were 2.0 mm x 10mmx35 μm, 2.5 mmx12.5 mmx49 μm and 3.0 mmx15 mmx64 μm. The measured sensitivities of the three sensors were quite similar to each other as theoretically predicted.

In this paper, the relationship between sensitivity and diaphragm dimensions in a glass-based integrated optic pressure sensor is described. The sensor has a rectangular diaphragm as a pressure-sensitive structure and a straight sensing waveguide across the diaphragm. The sensor operation is based on the phenonemon that a phase difference between two orthogonal guilded modes is induced by the elasto-optic effect in the presence of applied pressure. The sensitivity of the sensor is theoretically known to be dependent on the thickness and side length of the diaphragm. Such dependencies are worth investigating to obtain helpful design rules for miniaturization of the sensor, but have not been examined experimentally in detail. In this study, to examine the relationship between sensitivity and thickness, two sensors were fabricated with 10 mm x 10 mm x 0.3 mm (sensor #1) and 10 mm x 10 mm x 0.22 mm (sensor #2) diaphragms. The sensitivity of sensor #2 was larger than that of sensor #1 by a factor of 1.72, which closely agreed with the theoretical factor of 1.86. Moreover, to determine the relationship between sensitivity and side length, two more sensors, besides sensor #2, with 7mm-square (sensor #3) and 14mm-square (sensor #4) diaphragms were fabricated with a diaphragm thickness of 0.22 mm. The measured sensitivities agree approximately with the theoretical ones although there was a slight difference in sensor #4.

A numerically stable and systematic implementation of the rigorous coupled-wave analysis (RCWA) for the general multilayered grating structures is presented for both TE and TM modes. Numerical results of the approach are shown for the diffraction-based optical device as an example and are compared with the scalar diffraction method to illustrate the limited applicability of the scalar analysis.

Arsenic sulfide (As-S) and arsenic selenide (As-Se) glass optical fibers typically possess extrinsic absorption bands in the infrared wavelength region associated with residual hydrogen and oxygen related impurities, despite using purified precursors. We report a purification process based on the addition of 0.1 wt%tellurium tetrachloride (TeCl₄) to the glass. During melting, the chlorine from TeCl₄ reacts with the hydrogen impurities to produce volatile products (e.g. HCl) that can be removed by subsequent dynamic distillation. The processing conditions have been modified accordingly to give low H-S (1.5 dB/m) and low H-Se (0.2 dB/m) impurity content.

We describe the design, simulation and testing of an optoelectronic interconnection system that uses fiber image guides (FIGs) to transfer optical data packets among network nodes having both optical and electrical input-output (I/O) ports. FIGs are a tightly packed array of thousands of optical fibers, and are capable of 2D parallel image transmission with more flexible alignment and packaging than free-space alternatives. We have designed printed circuit boards (PCBs) for the system demonstration. The PCB has optoelectronic components such as: vertical cavity surface emitting lasers (VCSELs) and metal-semiconductor-metal (MSM) detector arrays for optical I/O, and transimpedance amplifier receiver (TIAR) arrays for converting photodetected current signals into 2.5V CMOS compatible signals. Using FIGs our system demonstrates high data transmission rates over 16 channels with low crosstalk. We discuss various techniques for coupling the FIGs to the optical I/O array and optimizing the coupling distance for low crosstalk and insertion loss. We also present 3D waveguide propagation simulations of FIGs based on the beam propagation method and compare the results with experiments. Both experiment and simulation show that the coupling distance and alignment are critical to achieve the best output optical power profile.

Micro Optical Electro Mechanical Systems (MOEMS) gain more and more importance in technical applications. The combination of optical actuators and micromachined silicon technology arise possibilities to realize equipment in high volumes for reasonable prices, that have formerly been expensive laboratory equipment. This paper reports on the performance and applications of a spectrometer in MOEMS technology. It is based on a scanning mirror chip with a grating structure on top. Thus a spectrometer with selectable wavelength range was realized. The resolution is not limited by line width of a multisensor detector, as it is the case for state of the art low cost spectrometers. A single detector is used, cheaper than arrays and available for all wavelength ranges. The setup can be small and light, the grating withstands shocks and vibration much better than a classical spectrometer. The grating moves with a frequency of 500 Hz respectively 1000 Hz, a whole spectrum is acquired within milliseconds. The resolution is given by the grating line density and the spectrometer dimensions, as it is valid for every single detector spectrometer. Depending on the number of systems built, a price of the system can be expected significantly below those of low cost systems with fixed grating. Many possibilities arise in every application where light is analyzed spectroscopically.

New bismuth zinc tellurite glasses were prepared. The as-quenched glasses were found to contain crystallites confirmed via XRD and TEM studies. The various structural units present in the glasses were probed by Raman and IR spectroscopies. The glasses were poled by conventional thermal poling and also using a new method of two stage poling. The second harmonic generation of the poled glasses were measured using the Maker-fringes technique and it was found that the two stage poling enhanced the SHG efficiency when compared to the conventionally poled samples.