High Power Semiconductor Devices Research Articles

Numerical simulation and optimization of heat dissipation structure for high power insulated gate bipolar transistor (IGBT)

With the increasing integration level of IGBT components, heat dissipation capacity determines whether its working state is stable. The third-generation semiconductor chips are expected to be made of SiC or GaN materials, with lower power loss and higher thermostability. Currently, the widely used backplane heat dissipation schemes cannot meet the heat dissipation requirements of high-power IGBT modules. In this study, an aluminium alloy shell with an AlN column is proposed to replace the traditional plastic shell to further improve the heat dissipation of IGBT devices. 180°C is set as the expected maximum temperature of the module, and finite element modelling and analysis of the heat dissipation structure of the IGBT device is carried out. Four types of design schemes are compared and the heat dissipation structure is optimized. AlN heat-transfer blocks and Ag sintering are introduced to realise the novel design. The results show that traditional single-side heat dissipation can not meet the requirements. The improved double-side dissipation scheme meets the requirement when the heat sink fin height is more than 15mm. And the optimized heat sink shall have 14 fins of 20mm in height. The results of this study can provide a reference for the thermal management and design of high-power third-generation semiconductor devices.

Design strategy of IGCT clamp circuit parameters based on bi-objective nonlinear optimization

The integrated gate commutated thyristor (IGCT) is a high-power semiconductor device with excellent performance, whose operating characteristics are governed by the clamp circuit. In order to further optimize the operating characteristics of IGCT, a bi-objective nonlinear optimization strategy based on IGCT turn-off switching losses and clamp circuit transition time is proposed in this paper, by which the clamp circuit parameters can be obtained quickly and comprehensively. The results show that IGCT turn-off voltage drop and oscillation would be suppressed, and the IGCT turn-off and clamp circuit power losses could also be reduced, which ensures the safe operation of the IGCT and minimizes the impact on the dynamic performance of the power electronics converter based on IGCT in the meantime.

Failure rate calculation method for high power devices in space applications at low earth orbit

In this paper we propose that calculation method to evaluate failure rate of high power semiconductor device against proton flux environment of low earth orbit. Advantage of this method is to expand to evaluate failure rate of any devices under various radiation environment. Possible suggestion to improve high power device reliability was discussed.

Development of Ag-In Alloy Pastes by Mechanical Alloying for Die Attachment of High-Power Semiconductor Devices.

Sintered silver paste is widely used as the die-attachment material for power semiconductors. However, sintered silver joints encounter problems, such as severe coarsening of sintered pores and oxidation issues, in harsh high-temperature environments. These lead to the deterioration of the die-attachment joints. In this paper, a novel method of sintering silver joints is demonstrated, where silver–indium alloy paste is used to improve the reliability of sintered Ag joints. The silver–indium (Ag–In) alloy paste was fabricated through mechanical alloying using the ball-milling technique. The well-bonded sintered Ag–In alloy joints inhibited pore coarsening better than pure sintered Ag joints and significantly enhanced the mechanical properties at high operating temperatures. Lastly, an oxidation mechanism for the sintered joint was proposed, and strategies to prevent such high-temperature oxidation were discussed.

Thermal and oxidation stability of TixW1−x diffusion barriers investigated by soft and hard x-ray photoelectron spectroscopy

The binary alloy of titanium-tungsten (TiW) is an established diffusion barrier in high-power semiconductor devices, owing to its ability to suppress the diffusion of copper from the metallization scheme into the surrounding silicon substructure. However, little is known about the response of TiW to high-temperature events or its behavior when exposed to air. Here, a combined soft and hard x-ray photoelectron spectroscopy (XPS) characterization approach is used to study the influence of post-deposition annealing and titanium concentration on the oxidation behavior of a 300 nm-thick TiW film. The combination of both XPS techniques allows for the assessment of the chemical state and elemental composition across the surface and bulk of the TiW layer. The findings show that in response to high-temperature annealing, titanium segregates out of the mixed metal system and upwardly migrates, accumulating at the TiW/air interface. Titanium shows remarkably rapid diffusion under relatively short annealing timescales, and the extent of titanium surface enrichment is increased through longer annealing periods or by increasing the bulk titanium concentration. Surface titanium enrichment enhances the extent of oxidation both at the surface and in the bulk of the alloy due to the strong gettering ability of titanium. Quantification of the soft x-ray photoelectron spectra highlights the formation of three tungsten oxidation environments, attributed to WO2, WO3, and a WO3 oxide coordinated with a titanium environment. This combinatorial characterization approach provides valuable insights into the thermal and oxidation stability of TiW alloys from two depth perspectives, aiding the development of future device technologies.

Altitude dependent failure rate calculation for high power semiconductor devices in aviation electronics

The electric power usage in aircraft has reached 1 MW. Therefore, use of high power semiconductor devices expected to increase in avionics. Single event burnout failure happens when power devices operating in blocking condition interact with the cosmic radiation. The failure rate in power devices is more in airplane altitude compare to terrestrial operation. In this paper, the failure rate of high power silicon PiN diode is evaluated when operating in airplane altitude due to the interaction of cosmic ray neutrons. The proposed formula has the unique feature of decoupling between failure cross section and cosmic ray neutron flux. This makes it possible to calculate the failure rate under any cosmic radiation environment using the proposed failure rate formulation.

Unified power control strategy for new generation poloidal field power supply

The power level of full-controllable devices has been improved as the fast-growing technology of high-power semiconductor device, which promotes the upgrade of the poloidal field power supply (PFPS). In this paper, a new PFPS is proposed by using full-controllable devices. It is composed of voltage source converter (VSC) and H-bridge converter. Then combining with the dual closed-loop control strategy, the unified power control strategy is presented. To verify the correctness and reliability of the proposed analysis and strategy, a series of simulations about the dynamic response speed and the stability of the intermediate capacitor voltage are carried out. And the results prove that the unified power control strategy effectively improves the performance of the new PFPS.

Development of Ag–In Alloy Pastes by Mechanical Alloying for Die Attachment of High-Power Semiconductor Devices

Development of Ag–In Alloy Pastes by Mechanical Alloying for Die Attachment of High-Power Semiconductor Devices

Application of electric drive fracturing equipment in shale gas reservoir stimulation

Fracturing equipment is the core device in shale gas reservoir stimulation. As domestic shale gas exploration and development steps into deep layers, the development of fracturing technology puts forward higher requirements for relevant equipment. Electric drive is an important development direction of fracturing equipment technology. In this paper, the technical progress of electric drive fracturing equipment in China and abroad was investigated. It is shown that a high-power frequency conversion system is the key technology to determine the performance of electric drive fracturing equipment. After the adaptability of the high-power frequency conversion technology to electric drive fracturing equipment was analyzed, based on the fracturing operation of Model 2500 electric drive fracturing trucks in the Weiyuan Block of the Sichuan Basin for shale gas reservoir stimulation, electric drive and diesel-driven fracturing equipment were comparatively analyzed from the aspects of economic and technical indicators. And the following research results were obtained. First, compared with a diesel-driven fracturing truck of the same power, an electric drive fracturing truck can realize a full power coverage and a continuous adjustment of output displacement, and can better meet the operation requirements of fracturing process for a precise control of the pumping displacement, while reducing the power cost by 68% and the equipment purchase cost by 10–20%. Second, compared with the skid mounted equipment, an electric drive fracturing truck has a better transport performance, being suitable for the fracturing well sites with poor road conditions, such as loess gullies, hills and mountains. It is suggested that the following development direction of the electric drive fracturing equipment should focus on the improvement of single machine power density. In addition, it is recommended to enhance the basic studies on high-pressure and high-power semiconductor devices and make a good plan for the power demand of shale gas platform construction in advance, so as to give a better play to the operating cost advantage of electric drive fracturing equipment.

Reliability test for subsea power semiconductors

Following the demand from a new domain of application, a test setup was developed to evaluate the performance of high-power semiconductor device when exposed to a dielectric liquid and high pressure. The specific test conditions were established to emulate the conditions inside a tank place in a deep-sea environment. Those conditions can be characterized by two key parameters: the pressure and the chosen dielectric/thermal liquid. The power semiconductor device is to be tested at high and low temperature to simulate operation under maximum load as well as when turned off, respectively. The reliability tests outlined here serve to support the development and the validation of Subsea technology for semiconductors power modules.

Accurate simulation and modeling of the control system and the power electronics of a 72-pulse VSC-based generalized unified power flow controller (GUPFC)

The rapid growth and development of manufacturing technology for controlling processors and high-power semiconductor devices has led to the enhancement of flexible AC transmission system (FACTS) controllers, thus improving the performance of power systems. The generalized unified power flow controller (GUPFC) is today introduced as the most advanced generation of the FACTS family, which is based on voltage source converter (VSC). It can, hence, control all the five parameters of a power system given above. The paper, for the first time, presents the accurate modeling of different parts of the control system and power electronics of the 72-pulse VSC-based GUPFC in detail to satisfy operation constraints and improve power quality in terms of harmonic level reduction. There is no need to use active or passive filters in the proposed model for reducing harmonic level of the GUPFC. In this regard, the calculation of four main parameters, namely D_alpha, alpha, angular speed, and sigma, as inputs of the pulse-firing system for the 72-pulse VSC is performed dynamically and optimally to meet all expected constraints in terms of operation, power quality, and harmonic reduction of a network with a GUPFC. The presented modeling is implemented on a 230 kV GUPFC-compensated double-circuit transmission line in MATLAB/Simulink environment, and the desired results of the presence of the 72-pulse VSC-based GUPFC are shown to meet the desired constraints of operators.

Automated UV-Epoxy-Based Micro-Optic Assembly for Kilowatt-Class Laser-Diode Arrays and Modules

This paper outlines the numerous interdisciplinary activities required to implement two ultraviolet (UV) epoxy-based automated micro-optic assembly tools within a “smart factory” for high-power semiconductor laser manufacturing. After a brief overview of the motivation and approach, two examples of automated micro-optic assembly tools [e.g., fast-axis collimation (FAC) lens attach and mirror stripe alignment] for kilowattclass laser-diode arrays and modules are given that demonstrate the breadth of automation activities required within a modern factory. For each system, the complete assembly sequence is outlined, and the steps that presented fully automated challenges are highlighted. In contrast, to the previous published results that focused on the details of the automated micro-optic alignment, the largest challenges primarily involved automation and repeatability of the UV-epoxy dispense and attach processes for these high-power semiconductor devices. These were solved by precise and stable mechanical design, customized lighting, and machine vision solutions. Finally, the complete integration of the tools into a “smart-factory” system is outlined.

Microscale pattern etch of 4H–SiC by inductively coupled plasma

SiC is commonly used in the field of high temperature, high frequency, high power and radiation-resistant semiconductor devices and ultraviolet photodetectors due to its excellent photoelectric characteristics. Plasma etching is one of the key processes in the fabrication of SiC devices. However, it is difficult to obtain a smooth and vertical etch profile due to the subtrench effect when the conventional inductively coupled plasma (ICP) is used to etch SiC with small-size patterns (such as 1 μm holes or trenches), thus affecting the subsequent process and device performance. To study microscale pattern etch of 4H–SiC, we used different mask and etch gas systems for ICP etching. The etch profile was characterized using SEM. When the chlorine gas was used, there was no subtrench but a very inclined angle between the SiC bottom and sidewalls, which is hard to improve. When we used photoresist (PR) mask–HBr system, sidewall deposition happened. SiO2 mask–HBr/Ar system could help to etch different angles but has a limitation angle of up to 80°. Compared to above systems, the etching profile was much more vertical when HBr/SF6/O2 was used. After many rounds of process debugging, we obtain a good etching profile with flat and vertical sidewalls, no subtrench in the bottom and no damage of the SiO2 mask. Our work shows a good way of microscale pattern SiC etching by ICP.

Augmentation of Thermal Performance of Plate Fin Heat Sink

Heat sinks are popularly used in various industrial applications to cool electrical, electronics and automobile components. They are useful in removing the heat from the surfaces at elevated temperatures. The life of such devices depends on their operating temperature. Heat sinks are important parts of thermal management systems of most of these devices eg: Diods, thyristers, high power semiconductor devices such as integrated circuits of inverters, audio amplifier, microprocessors, microcontrollers etc. In many situations where heat transfer is by free convection where convective heat transfer coefficient is low, fins are the best solution because of their less cost and trouble free operation. The weight and size of equipment are the most important parameters of design. Present day demand, the use of compact systems in every application which leads to higher packing density. The failure rate of electronic equipments increase exponentially with the temperature. Also the high thermal stresses in the solder joints of electronic components mounted on circuit boards resulting from temperature variation are major causes of failure. Therefore thermal control has become an important factor in the design and operation of electronic equipment. The most preferred method for cooling these systems is passive cooling because it is cost effective and reliable. This leads to focus on development of effective fin heat sink. To make heat sink effective, geometry and orientation of the heat sink as well as heat transfer augmentation techniques plays important role. This paper highlights the use of heat sinks in electronic cooling applications and review of related literature of improving the heat transfer performance of plate fin heat sinks by surface modifications, interrupting the boundary layer and changing the orientation.

Augmentation of Thermal Performance of Plate Fin Heat Sink

Heat sinks are popularly used in various industrial applications to cool electrical, electronics and automobile components. They are useful in removing the heat from the surfaces at elevated temperatures. The life of such devices depends on their operating temperature. Heat sinks are important parts of thermal management systems of most of these devices eg: Diods, thyristers, high power semiconductor devices such as integrated circuits of inverters, audio amplifier, microprocessors, microcontrollers etc. In many situations where heat transfer is by free convection where convective heat transfer coefficient is low, fins are the best solution because of their less cost and trouble free operation. The weight and size of equipment are the most important parameters of design. Present day demand, the use of compact systems in every application which leads to higher packing density. The failure rate of electronic equipments increase exponentially with the temperature. Also the high thermal stresses in the solder joints of electronic components mounted on circuit boards resulting from temperature variation are major causes of failure. Therefore thermal control has become an important factor in the design and operation of electronic equipment. The most preferred method for cooling these systems is passive cooling because it is cost effective and reliable. This leads to focus on development of effective fin heat sink. To make heat sink effective, geometry and orientation of the heat sink as well as heat transfer augmentation techniques plays important role. This paper highlights the use of heat sinks in electronic cooling applications and review of related literature of improving the heat transfer performance of plate fin heat sinks by surface modifications, interrupting the boundary layer and changing the orientation.

Room Temperature Bonding of GaN on Diamond Wafers by Using Mo/Au Nano-Layer for High-Power Semiconductor Devices

For overcoming the heat dissipation problem of GaN-based high-power semiconductor devices, GaN wafers were bonded on polycrystalline and single-crystalline diamond wafers by using Mo/Au nano-layer at room temperature. Mo/Au double layers (~5 nm/11 nm) were deposited on the surfaces of GaN and diamond wafers. The scanning acoustic microscope (SAM) analyzation and tensile strength testing results show that a low voidage of ≤1.5% can be obtained for Mo/Au nano-layer, along with a bonding strength of 6.8 MPa. In addition, the thermal-impact-resistance performance was evaluated with thermal cycling testing.

Online IGBT Temperature Measurement via Leakage Current in HTRB Test

To evaluate the reliability of endure high voltage when IGBT is turned off and eliminate the defective devices, HTRB (High Temperature Reverse Bias) test is one of the necessary tests for IGBT before it’s produced. According to the IEC standard, DUTs (Device Under Test) are usually blocked, and bear forward bias voltage, which is preferably 80% of VCESmax. During the HTRB test, DUT junction temperature should be set at maximum virtual junction temperature Tvj(max) which is usually between 125°C to 150°C. Due to high voltage, even though the leakage current is very low, high-power semiconductor devices generate dissipated power during HTRB test, which lead to junction temperature increase. And to access the reliability accurately, the rise of temperature cannot be neglected. To insure the test is carry out at standard temperature, this paper monitors DUTs junction temperature by using leakage current as a TSEP (Temperature Sensitive Electrical Parameter),. The experiment proofs that higher the junction temperature is, the more accurate the measure results will be, in the operating temperature range. In this paper, thermal resistance measurement is used to evaluate the validity of this method.

Guest Editorial: Special Section on Transient Behaviors of Multi-Timescale Megawatt Power Electronics Systems

With the rapid development of a new generation of power systems and high power traction, megawatt power electronics receives increasingly more global attention for enhanced capability and reliability. Megawatt power electronics apparatuses and systems are composed of control units, high-power semiconductor devices, passive elements, and generalized connecting parts, in which effective transformation and transmission of electric power are achieved via the control of information flow over energy flow. When considering multiple time-scales, the megawatt power electronics systems turn into a combination of linearity and nonlinearity, and also the one of discrete and continuous quantities, which results in complicated transient phenomena. In addition, scalability and modularity become growing trends for the megawatt power converters, which are preferred by the manufacturers.

Novel Solution for High-Temperature Dielectric Application to Encapsulate High-Voltage Power Semiconductor Devices

Traditional semiconductor packaging techniques and materials have been working well for conventional Si devices, which usually operate at temperatures up to 175 °C. As the operating temperature increases, these techniques exhibit failures such as bulk flows, volume shrinkage, brittleness and subsequent cracking, and deterioration of dielectric strength. For the new wide-bandgap power devices, which work at voltages as high as 1200 V and junction temperatures as high as 250 °C, there is currently no known dielectric material to encapsulate and protect the active devices. This paper summarizes novel solutions for high-temperature dielectric materials for encapsulating high-power semiconductor devices without any dielectric breakdown and also without introducing excessive thermal, electrical, and mechanical stresses to the encapsulated devices. Polymer dielectric candidates investigated in this paper include cyanate ester-based resin and silicone-based resin. In addition, conformal coating approaches that include alumina deposited by atomic layer deposition (ALD) technique and tetrahedral amorphous carbon (ta-c) deposited by filtered cathodic vacuum arc technique were evaluated in this paper. Dielectric strength performance of the material combinations with respect to temperatures was evaluated. Among the polymer encapsulants, a silicone resin with silica fillers was determined to be the prospective candidate. Breakdown voltage and leakage current of the silicone-based encapsulant with and without the conformal coating was measured by breakdown tests. It was determined that the ta-c conformal coating deteriorates the dielectric performance of the encapsulant, while the alumina thin film deposited by ALD approach reduces the leakage current of the encapsulation material and also increases the breakdown voltage of the silicone encapsulant. Thus, the combination of alumina thin film deposited by ALD approach along with the silicone encapsulant is recommended for this application, involving high temperature and high voltage.

Novel copper particle paste with self-reduction and self-protection characteristics for die attachment of power semiconductor under a nitrogen atmosphere

A novel die-attach material, Cu particle paste with self-reduction and self-protection characteristics, was designed by simply adding ascorbic acid (AA) into Cu paste. The self-reduction characteristic is due to the addition of AA into the paste, as it reduces the Cu oxide layer on the metal Cu, even at room temperature. The self-protection characteristic is due to the decomposition of AA preventing further oxidation during sintering process. These characteristics are beneficial for the sintering of Cu particles and for enhancing the bondability. The increase to bonding strength is the direct result of the concentration of AA in the Cu paste. The bonding strength of Cu joints sintered from Cu paste with 1.3 wt% AA addition was as high as 24.8 MPa, thus making it attractive as a die-attach material. On the other hand, the bonding strength of those sintered from Cu paste without AA addition was only 9.7 MPa. These results strongly suggest that the self-reduction and self-protection characteristics resulting from AA addition are necessary and crucial for sintered joints fabricated from readily oxidized Cu particle paste. This novel Cu paste is an efficient, safe, and environmentally-friendly die attachment material, especially well-suited for applications in high power semiconductor devices.