Vol 48, No 7 (2019)

One of the promising approaches to solve the problem of increasing the efficiency for interelement connections consists in the use of integrated optical switching systems, whose main elements represent injection lasers with functionally integrated optical radiation modulators. Injection lasers make it possible to modulate the laser radiation by subpicosecond controlling pulses at a constant pumping current, as well as making it possible to implement the sources and modulators of optical radiation in a united A^IIIB^V nanoheterostructure with second-type heterojunctions. This work is devoted to the studies concerning the transport of charge carriers within a functionally integrated modulator laser with internal frequency modulation of the generated optical radiation using a proposed two-dimensional diffusion-drift model and a numerical simulation technique. The results of the numerical simulation of charge carrier transport in a modulator laser when the pumping current is switched on, as well as under pulsed variation of the controlling voltage, take the structural features, the transport effects, the mechanisms of stimulated and spontaneous radiative recombination, and the photon lifetime into account. It is shown that the maximum modulation frequency of laser radiation is determined by the subpicosecond time of the controlled spatial relocation of the charge-carrier density maxima in the quantum zones of the modulator laser, as well as by the photon lifetime in the laser resonator, and corresponds to the terahertz range. To increase the maximum modulation frequency, it is necessary to reduce the photon lifetime in the active zone of the modulator laser to values lower than 3 ps by changing the resonator’s parameters in a corresponding manner. The proposed model and the method of numerical simulation make it possible to optimize the parameters of a functionally integrated modulator laser, and to provide the required relationships between the maximum modulation frequency of the optical radiation, the modulation coefficient, and the density of the threshold pumping current.

In order to ensure the operability, durability, and reliability of three-dimensional microelectronic modules, the engineering calculation of the structures, including the simulation of adhesive-bonded joints, as well as a valid choice of materials and constructive decisions, is necessary. The main operational and physical specifications for three-dimensional microelectronic modules and the advantages of the application of adhesive materials in the structures of modules are presented. It is established that the structural strength of three-dimensional microelectronic modules and the operating characteristics of adhesive-bonded joints depend on many factors. The most important factors among them are the properties of the adhesive material, the structure of the joints and the operating conditions. The types of loads, which affect the modules during their operation and their influence on the structural strength are considered. It is revealed that irregular loads, such as peeling and bending, are the most dangerous loading types for the adhesive-bonded joints of modules. It is established that in peeling a high concentration of edge stresses arises, which leads to the destruction of the joints, and in bending there is a concentration of normal and tangential stresses along the length of the glue lines. The types of adhesive materials which are used for assembling the modules are analyzed. The adhesive materials are selected taking the main design and technological constraints and the requirements on adhesive-bonded joints into account. The module’s structures are simulated and the influence of the physical, mechanical, and thermal properties of the adhesive materials on the stresses in the adhesive-bonded joints and the strength of the products under the impact of inertial loads with an acceleration of 1000g and heated by 40°С is determined. Recommendations on the simulation of adhesive-bonded joints of three-dimensional microelectronic modules are made. The obtained results revealed that the stresses depend on the elastic properties of the adhesive-bonded joints, as well as on the nature and value of the load on them, and they are determined by the mechanical strength and rigidity of the adhesive material. In order to reduce the stresses, it is necessary to use more rigid structural materials, and the adhesive materials need to be chosen based on the operating conditions of the microelectronic module.

The junctionless MOS-transistors (junctionless MOSFET) have a number of advantages over conventional transistors in terms of the simplicity of design, manufacturing technology, and reduction of the impact of short-channel effects on the characteristics of the device. However, the well known experimental nanowire junctionless MOSFET have high subthreshold currents due to the appearance of the parasitic bipolar transistor effect in a closed state. The structural model of a planar SOI junctionless MOSFET by the technology standards of 90 nm and the route of mathematical simulation using the Synopsys Sentaurus TCAD environment are developed. The influence of the impurity concentration in a SOI silicon film of a junctionless MOSFET at the threshold voltage, saturation currents, and subthreshold currents is investigated. The research results reveal that at the impurity concentrations in the working channel of the device below 10¹⁷ cm^–3, when the effect of band-to-band tunneling is absent and the parasitic bipolar transistor effect does not arise, the subthreshold currents decrease up to 10^–13 A/μm, which is significantly lower than those of the conventional MOSFET, while maintaining the saturation currents at an acceptable level.

SOI MOSFETs have the worst properties of heat removal from an active region, which negatively affects the reliability and efficiency of integrated circuits. Using TCAD modeling, we investigate the self-heating effect in the following structures of deeply submicron MOSFETs with different configurations of buried oxide: traditional bulk MOSFET, SOI structure, SELBOX structure, partial SOI structure, thin-BOX SOI structure, UTBB SOI structure, and quasi-SOI structure. It is shown that, for a number of new designs, the maximum temperature in the MOSFET structure is significantly reduced as compared to Т_max of the standard SOI MOSFET structure; it approaches the values typical of standard MOSFETs on bulk silicon.

With the reduction of the technological dimensions of transistors, the influence of the variations of circuit and technological parameters on the values of the delay of the elements of combinational CMOS-circuits has grown significantly. Due to the spread of the values of these parameters, the uncertainty of delays appears, which leads to the necessity to define the ranges of possible delay values. The peak current in power lines when switching the inputs of gates is another factor increasingly influencing the design process of CMOS circuits in the transition to nanometer technologies. The value of the peak current is used to estimate the voltage drop in the power lines, which in turn is necessary to calculate the width of the power lines of CMOS circuits and switch off the transistors in the method of reducing the static power. The methods of full circuit simulation do not comprehensively analyze the circuits with a large number of inputs and the methods at the logical level of designing CMOS circuits do not provide the desired accuracy of the evaluation of the values of the delays and the peak current in the circuit. The problem of increasing the accuracy and the reliability of the analysis of the performance and peak currents of CMOS combinational circuits, taking into account the simultaneous switching of the inputs, as well as the analysis of logical correlations of the signals, is considered. The proposed method is based on using the cubic approximation of the correction difference of delays, taking into account the simultaneous switching of inputs. It is shown that the developed techniques of the analysis of the performance and peak currents of CMOS combinational circuits improve the accuracy of the upper estimates of the analyzed parameters by up to 3% compared with accurate simulation and make it possible to reduce the pessimistic upper estimate by factors of 2 to 3 compared with the estimate of the worst case. The developed methods of improving the accuracy of simulating delays and peak currents of combinational CMOS-circuits can be used as an addition to the existing CADS tools for VLSI for noise-immunity analysis, the analysis of the peak currents, characterization of complex functional units, and improving the accuracy of classical static analysis. To improving the accuracy of the interval estimates of the minimum delays and maximum peak current, the simulation methods taking into account the simultaneous switching of the inputs of the logical element are developed. In relation to the circuit simulation, the error of these methods does not exceed 3%. Compared with the results obtained without taking the simultaneous switching into account, reducing the minimum delay by up to 50% and the pessimistic estimate of the peak current of combinational circuits is reduced on average by 50–55%.

The study of the characteristics and errors of sensors is as difficult as designing and manufacturing them. The operation of sensors needs to be accurately studied in order to correct the design of sensor models accurately carry out the initial calibration of the ready-made sensors. The stand for investigating the sensitive elements of the gas flow rate sensors, in which the hot-wire anemometric and the calorimetric measurement modes are used, is described. It is shown that the facilities of the automatic control of the gas flow rate and dynamic recording of the measurement results make it possible to process and record the results in millisecond time intervals, investigate the operation of sensitive elements in the pulse mode, and simulate the operation of a microprocessor-based signal processing circuit at the hardware and software levels. The proposed stand makes it possible to take measurements at different gas temperatures and ambient temperatures ranging from –40 to 60°С and control the atmospheric pressure, and it makes it possible to specify the gas flow rate within 0–119 L/min. The considered stand is designed to develop compact sensors of the natural gas flow rate, which could replace the diaphragm flowmeters of the G1.6, G2.5, and G4 series. The stand makes it possible to carry out the comprehensive measurements of sensors in one place and considerably reduces the time of testing the developed converters.

Operation with control systems of moving objects requires taking into account the longitudinal and transverse components of the air flow velocity. For this purpose, the air flow velocity meters utilizing the acoustic method are used. These meters have a short response and no mechanically moving elements. However, their large size and the presence of construction units subjected to considerable deformation under the mechanical impact make it difficult to implement these meters as parts of moving objects. In this work, the operation algorithm of a device measuring the airflow velocity using the ultrasonic transducers is analyzed. We show that the reduction in the size of such devices leads to a decrease in the measurement accuracy without the improvement of the methods for determining the propagation time of an ultrasonic pulse between transducers and without periodic calibration of the device. The block diagram is developed and the main elements are considered for the computing device, which makes it possible to determine the longitudinal and transverse components of the airflow velocity using four ultrasonic transducers. The laboratory tests in an air tunnel with a distance between sensors of 60 mm showed that this device makes it possible to determine the airflow velocity with the maximum absolute measurement error of 2.5 m/s. This error is considerably affected by the structural elements of the device that obscure the measurement area but yield the required strength characteristics. The proposed error compensation algorithm based on the approximation method by trigonometric functions significantly reduces the maximum measurement error to 1.2 m/s. Further improvement of the device’s characteristics requires additional studies to find the shape of the sensor block and refinement of the error compensation algorithms.

The parameter homogeneity of dielectrics imposes increasingly stringent requirements, especially when used in micro- and nanoelectronics products. In this paper, we present an automated method for monitoring the electrophysical parameters of planar dielectrics with increased requirements on accuracy and resolution. A brief review and comparative analysis of the known methods for monitoring the parameters of solid dielectrics are performed here. The method of microwave control of the heterogeneities of the parameters of planar dielectrics by the wave characteristics of a scanning microstrip line (MSL) using a Radon transform is described. The phase response of the MSL is shown to be most sensitive to parameter heterogeneities. A mathematical model for processing the MSL wave characteristics and for constructing a map of the parameter heterogeneity distribution is presented. The parameter realizability and the applicability limits of the mathematical model are estimated. The technical implementation of the method is considered using, as an example, the setup for monitoring the parameter homogeneity of planar dielectrics and the results of modeling the measuring part of the setup in the AWR Microwave Office CAD system. The methodological errors of measuring the electrophysical parameters of dielectrics by the described method are estimated, and they are on average ±1 × 10^–4 in the case of measuring the permittivity and 0.2 mm in the case of measuring the spatial resolution according to the preliminary calculations. The described control method is designed for use in the production of electronic products using dielectric materials with a high level of parameter repeatability.