Vol 47, No 7 (2018)

The use of silicon in optoelectronic elements and microwave devices is limited by its electrical properties. In this range of application, silicon is replaced by wider-gap materials, e.g., GaN, AlN, InN, and solid solutions based on them. This allows highly efficient devices, including light-emitting diodes and photodetectors operating in a wide radiation spectral range, to be fabricated. In addition, GaN-based materials have been successfully used in designing high-power microwave devices, in particular, transistors with highly mobile electrons operating at high temperatures. We discuss specific features of the technology for fabricating active layers in the AlGaN/GaN/InGaN/GaN heterostructures with the use of metalorganic compounds. The initial substances used are ultra-pure ammonia NH₃ and gallium, aluminum, and indium metalorganic compounds in the trimethyl form. The temperature dependence of the epitaxial GaN layer growth rate is investigated. This dependence is shown to be weak in a wide temperature range. The importance of adsorption processes occurring on the growth surfaces is confirmed. Thermodynamical analysis of the regularities of the investigated process is carried out to simulate the process and establish the conditions for forming GaN-based solid solutions. It is shown that during the formation of Ga_1 – xIn_xN solid solutions in the high-temperature region the attainable InN content is not higher than 0.4 mole fractions. As the growth temperature decreases to 600°С, the conditions for incorporating In in the solid solution are noticeably improved and the InN concentration increases to 0.9 mole fractions. It is demonstrated that growing the GaAlN solid solutions in a wide temperature range makes it possible to obtain the AlN content ranging from 0.1 to 0.9 mole fractions. The experimental investigations are confirmed by the calculations. Therefore, when growing the Ga_1 – xIn_xN quantum well layers in the active heterostructural area for commercial blue light-emitting diode chips with an indium content of х = 0.1–0.15, the growth temperature should be reduced. At low temperatures, however, it is difficult to grow epitaxial layers with a high crystal quality.

The general principles of designing three-dimensional multichip modules are presented. The multichip modules are simulated and the effect of the design variables on the strength and temperature condition of the products is studied. The values of thermomechanical stresses, heat resistance, and overheating temperatures of the materials are determined, and the heat removal efficiency in various structural variations of three-dimensional modules is also found. Methods for providing an intensive heat transfer in the structures of the modules and increasing the strength consistency of the products are developed. Recommendations on the design of multichip modules are given.

The influence of the topological parameters of shunts in the cathode regions of a photothyristor on the dV/dt effect is calculated. The analytical condition that makes it possible to determine, in the first approximation, the region of the onset of the structure triggering caused by the dV/dt effect is obtained. When the forward voltage increases on the thyristor structure in the off state, the bias current flowing through the barrier capacitance of the inversely biased p-base-n-base junction causes a spontaneous triggering of the structure. This is the core of the dV/dt effect in thyristors.

High-speed broadband analog integrated circuits based on a complimentary bipolar technology are widely used in modern equipment. The speed of these circuits is directly related to the boundary frequency of the transistors used. In this paper, we consider some modern approaches to develop complimentary bipolar technologies that ensure a high performance of the integrated circuits and systems on a chip. With the help of TCAD software, the device-technological modeling of the electronic component base is carried out. A technological route is developed for manufacturing the operational amplifier by a high-speed complimentary technology with a self-compliant emitter–base node, based on two layers of polysilicon. To ensure the self-alignment of the base and emitter polysilicon layers when manufacturing n–p–n and p–n–p transistors, an L-shaped nitride spacer is formed. The lateral insulation of the components is made in the form of narrow vertical slits that are lined with oxide and silicon nitride and filled with polysilicon. The manufacturing route developed allows the n–p–n and p–n–p transistors on a single crystal to reach the boundary frequency of 8–10 GHz at the collector–emitter breakdown voltages of more than 10 V. The problem of the complementarity of bipolar transistors is largely solved. In the future, this will make it possible to create a new class of domestic broadband and high-speed analog and digital integrated circuits, as well as ensure the technological independence of the Russian Federation.

The sub-100-nm CMOS process with a high-κ gate dielectric is one of the key technologies for the fabrication of digital, analog, and RF VLSI circuits and on-chip systems. The influence of ionizing radiation on 45-nm MOS transistors with a high-κ dielectric fabricated using the bulk-silicon and SOI technologies is simulated. Effects induced by the substitution of SiO₂ with a high-κ dielectric are noted. The processes of selection and tuning of physical models for the simulation of high-κ MOS transistors in the Sentaurus TCAD are outlined. A set of new physical semiempirical models introducing the dependence of radiation-sensitive parameters (carrier lifetime, carrier mobility, charge density in the bulk of SiO₂ and HfO₂ and at the HfO₂/Si interfaces) on the irradiation dose is developed. Nanoscale bulk and SOI MOS transistors with a high-κ dielectric are simulated. It is demonstrated that an increase in the drain current after irradiation in nanoscale SOI structures is induced by the charge accumulation in the side oxide. An acceptable fit between the simulation results and the experimental data is obtained. The simulation results confirm that the leakage current is suppressed (compared to common MOS transistors with SiO₂) in sub-100-nm MOS transistors with a high-κ dielectric. However, the other important parameters of sub-100-nm MOS transistors with a high-κ dielectric are more sensitive to ionizing radiation.

A formalized technique for the automated design of bandgap voltage reference (BVR) is proposed. The distinctive feature of the technique is the automation of the synthesis of the circuit engineering conceptions by the given parameters of the source, which considerably reduces the design time. Based on the developed technique by the 180-nm SOI technology, the circuit of the voltage reference source was designed. The obtained schematic solution has the following desired characteristics: the power supply suppression ratio at the frequencies of 1 kHz in the worst case is –60 dB and at the frequencies of 1 MHz is –44 dB, the maximum consumption current is 25 μA, the temperature coefficient is 24 ppm/°C in the temperature range from –70 to 150°C, and the ratio of changing the output voltage to changing the supply voltage is 354 μV/V in the range of supply voltage from 2.4 to 7 V.

Voltage doublers on switched capacitors are commonly used impulse voltage converters. The modern approach to designing electronic circuits involves the use of specialized circuit design programs, allowing us to take into account many factors and get a result as close as possible to the reality. A mathematical description of transients in the circuit of a voltage doubler is given. Using the mathematical modeling of transients, the formulas for designing the circuit of a voltage doubler based on switched capacitors are derived. The formulas take into account the requirements for setting the voltage and ripple levels and the relaxation time of the output voltage. The validity of analytical calculations is verified by the results of circuit simulation. The practical recommendations for choosing the scheme parameters are given. The work may be of interest to developers of analog circuits. The obtained formulas make it possible to quickly compute the circuit parameters and obtain the necessary characteristics.

An event-driven simulation algorithm is the main algorithm for the simulation of digital circuits used in digital simulators. However, the implicit choice of the firing sequence of simultaneously switching signals can lead to differences in the simulation results. A new algorithm for the event-driven logic simulation of digital integrated circuits based on the use of the modified mathematical apparatus of Petri nets is proposed. The Petri nets apparatus makes it possible to simulate parallel constructs using sequential instructions. Therefore, there is no need to separate events with a δ-delay. The described approach makes it possible to eliminate the ambiguity of switching signals that occur at the same time due to not using the δ-delay. The results of the algorithm’s work are presented using the gate-level simulation of a number of combinational and sequential circuits as an example. The obtained timing diagrams, as well as the simulation time, show that the proposed algorithm is not inferior to the existing simulation tools in terms of reliability and performance.

The influence of continuous microwave fields with the average power level on the characteristics of a square-pulse RC-generator with a multichannel buffer decoupling network and matching circuit is investigated. The RC-generator is made on inverters of the 74HC14 series with a built-in Schmitt trigger. It is shown that increasing the power of the acting electromagnetic radiation reduces the pulse duration and, hence, increases the frequency of the output signal of the RC-generator. Moreover, the sensitivity of the output signal of the RC-generator to the impact of the microwave signal decreases with an increase in either the supply voltage or the microwave signal power. It is stated that the output characteristics of digital radio-electronic systems are the most sensitive to the impact of a microwave signal when using the semiconductor integrated microcircuits as the active elements and when choosing the operation modes of microcircuits by DC voltage, which greatly differs from the rated voltage.