Home – Home – Jornals – Avtometriya 2012 number 4

A new version of the method of direct multibeam laser writing of diffractive optical elements (DOE) is proposed and investigated. A writing area in the form of an array of focused light spots is formed by splitting the writing laser beam into multiple beams by using a Dammann grating and focusing of these beams in the plane of a moving carrier with a photosensitive material. Adjustment of the radial pitch of writing and correction of the uniformity of the beam intensity is carried out by tilting the Dammann grating and displacing it in the dispersion direction. In writing DOEs, the radial pitch of discrete displacement of the writing area with respect to the plane of the DOE is set equal to or multiple of the average radial distance between the radial projections of the centers of the focused light spots. This version provides improved performance and accuracy due to high-quality paralleling of the writing beam and the averaging effect in superimposed writing.

The quality of the image formed by a MIOL-Akkord bifocal intraocular lens (diffractiverefractive lens) implanted into various known eye models is studied analytically. As the pupil radius increases, the Strehl ratio characterizing the resolution is demonstrated to decrease appreciably (down to 0.05-0.09) for models where the absolute value of asphericity of the frontal surface of the cornea is smaller than or equal to 0.26 (far vision) and 0.18 (near vision). The effect of the aspherical shape of the rear surface of the cornea on the image quality is considered. It is noted that the best results for both far and near vision are provided by the model with Lotmar's cornea for which the diffractive component of the MIOL-Akkord eye lens is calculated.

The configuration of a two-beam interferometer with fixed inclined mirrors for recording holographic gratings is described. The convergence angle of the partial beams is adjusted in the range from 0 to 180º by changing the angle of incidence of the primary beam on the entry edge of the beam-splitting cube without additional adjustment of the mirrors. The path lengths of the interfering beams are automatically aligned by the mirror symmetry of the system. The optimal kinematics of adjusting the convergence angle is found that allows working with beams of sufficiently large diameter in the whole range of adjustment. The interferometer is highly resistive to vibration due to lack of adjustment devices.

A 3D image in the form of a linear combination of two two-dimensional images located at different depths along the line of vision of the eye is considered. The loss of contrast and nonlinearity of accommodation over the depth with a high resolution acuity of 3D images are demonstrated in computer experiments with different eye models.

Numerical simulation was used to demonstrate the possibility of designing a two-mirror multibeam reflective diffraction interferometer using a single-mode fiber with a cladding. The method is based on the calculation of the multimode and multibeam interference of the cladding modes in the fiber describing the diffraction on a diffuser made in the form of an insert of a different fiber as part of one of the interferometer mirrors. It is shown based on bulk optics that using certain optimization, one can obtain a reflection-response function profile of this interferometer similar to that of the transmission-response function of a Fabry-Perot interferometer.

The dependence of the diffraction efficiency of polarization holographic gratings on the temperature of their formation is presented. The temperature range corresponding to the maximum diffraction efficiency of such gratings is determined. The possibility of thermo-optical switching of recorded gratings is demonstrated.

The architecture of silicon multiplexers for multielement IR photodetectors is considered. A review is given of industrially oriented developments of linear (1 × 32, 1 × 288, 1 × 576) and matrix (32 × 32, 128 × 128, 160 × 128, 320 × 256) silicon multiplexers for mid- and far-IR photodetectors based on cadmium-mercury-tellurium, lead-tin-tellurium, and multilayer structures with quantum wells. An analysis is made of the temperature resolution of matrix IR photodetectors based on multiplexers with row and frame integration of photo signals using micrometer to deep submicron CMOS technologies. Typical parameters of the multiplexer and some of the infrared photodetectors designed at the Rzhanov Institute of Semiconductor Physics, SB RAS are given.

A device for studying the relaxation dynamics of nonequilibrium capacitance and charge in the mode of nonequilibrium depletion of MIS structures is designed. Methods of measurements were developed that provide a sensitivity of 0.03 pF and 0.02 pC along the capacitance and charge measurement channels with the acceleration time 1.5 µs and 0.7 µs, respectively. The results are obtained with a double amplitude of the test signal of capacitance measurement of ~15 mV and frequency 5 MHz. A mode of measuring the capacitance-voltage characteristics with a scanning duration up to 100 s is implemented.

The dependence of the fluorescence intensity of Cs vapor irradiated by elliptically polarized light resonant with the F = 4 → F = 5 transition on the scanned magnetic field has a complex shape - a narrow dip against the background of a broad peak. It is found experimentally that the dip is due to the small component of the magnetic field orthogonal to the direction of the laser beam. A two-level atom model with degenerate angular-momentum states predicts an order of magnitude lower contrast of this dip. The experimentally observed contrast of the dip can be explained by assuming that due to the low buffer-gas pressure in the optical cell, part of the resonant atoms is able to return to the laser beam without thermalized collisions with the wall of the cell, increasing the contrast. In addition, accounting for all optical transitions of the D₂-line and the optical pumping of the hyperfine-structure components of the ground state is a fundamental point for the interpretation of the experimental results.

Spectra of the cross section of inelastic scattering of electrons (product of the mean free path of inelastic scattering and its differential cross section) are obtained for SiO₂/Si(100) layered structures from experimental spectra of energy losses of reflected electrons with different energies of primary electrons. Computer simulations of the spectra of the cross section of inelastic scattering of reflected electrons for these layered structures are performed with the use of the dielectric function of the film and substrate materials. It is found that the SiO2 layer thickness determined through comparisons of experimental and model spectra agrees with results of ellipsometric measurements.

A possibility of using electric field energy for accelerating solids to significant velocities (1-10 km/s) is considered. A microelectromechanical electrostatic energy converter is used as a capacitive linear actuator, where energy conversion is performed in nanometer gaps, which allows the specific energy density of ~(3-10) J/m² and more to be reached in a single act of energy conversion at clock frequencies of 1 MHz and more. Such parameters allow pumping of electric field energy up to 10-30 MJ with the 1 m² area of the energy converter, with subsequent conversion of this energy to the kinetic energy of object motion. Specific features of the structure of the high-energy-intensity electrostatic energy converter and its operation at high clock frequencies in the regime of acceleration of a slider with a mass up to 10 kg are considered.

The structure of a quasi-optimal device for joint detection and estimation of the position of a moving spatial object observed in an applicative mixture with a random background is studied by statistical synthesis. It is shown that existing algorithms for detecting stationary objects provide only boundary estimates of the detection probability.

Asymptotic properties of a nonparametric estimate of the separating surface equation defined in the space of independent attributes of classified objects are studied. Based on this analysis, the importance of the influence of a priori information about independence of random variables on the approximation properties of the nonparametric decision function in a two-alternative problem of pattern recognition is established.

An algorithm for selecting the optimal sampling frequency of continuous signals based on the information optimality of the time series of samples is proposed. The effectiveness of the algorithm is shown on the example of solving the problem of determining the relative time delay of signals in multichannel transmission based on an algorithm of digital filtering of harmonic filling.