High-frequency switching power semiconductor devices are at the heart of power electronic converters. To date, these devices have been dominated by the well-established silicon (Si) technology. However, their intrinsic physical limits are becoming a barrier to achieving higher performance power conversion. Wide Bandgap (WBG) semiconductor devices offer the potential for higher efficiency, smaller size, lighter weight, and/or longer lifetime. Applications in power grid electronics as well as in electromobility are on the rise, but a number of technological bottle-necks need to be overcome if applications are to become more widespread - particularly packaging. This book describes the development of advanced multi-chip packaging solutions for novel WBG semiconductors, specifically silicon carbide (SiC) power MOSFETs. Coverage includes an introduction; multi-chip power modules; module design and transfer to SiC technology; electrothermal, thermo-mechanical, statistical and electromagnetic aspects of optimum module design; high temperature capable SiC power modules; validation technologies; degradation monitoring; and emerging packaging technologies. The book is a valuable reference for researchers and experts in academia and industry.
Inspec keywords: modules; reliability; electronics packaging; silicon compounds; wide band gap semiconductors; system-in-package
Other keywords: silicon carbide power module design; system-in-package; chip scale packaging; statistical analysis; ageing; reliability; wide band gap semiconductors; SiC; packaging; electronics packaging; silicon compounds; power MOSFET
Subjects: Product packaging; General electrical engineering topics; Reliability
This chapter proposes a review of SiC power MOSFETs' characteristics and discusses their performance and robustness in real applications. SiC MOSFETs have become an industrial reality. The added benefits they offer over established Si technology for the electrical energy conversion domain have been largely demonstrated and critically analysed. By now, they are making their way into large-volume applications in voltage classes ranging from 650 V to 3.3 kV.
With the excellent performance of today's power semiconductors, it is not uncommon for the packaging to be the main source of performance limits of the whole component or system it is used in. The reasons for a seemingly relatively modest technological progress and advancement of packaging solutions as compared to semiconductor devices lie in the multiplicity and variety of functions that packaging solutions are required to perform. Here, perhaps more than anywhere else, the design task consists in finding an optimum compromise among oftentimes antinomic imperatives: if it is relatively easy to design effective packages from either a thermal or a electrical point of view on the one hand, or to find an inexpensive solution on the other, achieving a merge of the three features with the right degree of trade-off is like the squaring of a circle! To perform its various roles, a power module must make use of a variety of structural elements, constituent materials, and joining technologies. This chapter proposes an introduction to the topic.
Si-IGBTs appeared in the early 1990s and have been constantly improved over time, targeting low loss, improved cost, longer life, whilst replacing GTO and other established power devices, contributing to the realization of more advanced power electronics. Even after 30 years of continuous improvement, Si-IGBTs have still not reached the deadlock of their potential, particularly in terms of power density and efficiency. The driving force behind the development has been 'cost/power', with newer devices still being invented and launched to market to enhance 'cost/power'. SiC offers significant advantages over Si for the power semiconductor devices. This chapter discusses the replacement of Si-IGBTs with eminent wide bandgap devices from the viewpoint of 'value/cost' whilst further progressing 'cost/power'.
In this chapter, a model for SiC power MOSFETs is proposed, which includes the influence of the interface traps and represents a good candidate for implementation in ET simulation tools. Despite its accuracy, the model is simple and associated with a straightforward parameter extraction procedure; more specifically, it is based on the combination of a standard MOSFET for the channel region, and a bias- and temperature-dependent resistance to emulate the lightly-doped drift region.
In this chapter, an innovative approach is proposed, the aim of which is to optimize the trade-off between computational efficiency and accuracy when handling problems with a relatively large amount of heat sources. The proposed strategy relies on a fully circuital representation of the whole device, wherein the equivalent network emulating the power-temperature feedback is obtained from a dynamic compact thermal model (DCTM), in turn automatically derived from an exceptionally accurate finite-element method (FEM) description of the device. A multicellular 4H-SiC power MOSFET operated under dc, short-circuit (SC), and unclamped inductive switching (UIS) conditions is considered as a case study.
This chapter is about the use of innovative statistical methodologies for the design of robust and reliable power modules (PM), based on the inclusion of device characteristics and their inevitable spread among components from the same manufacturing lot, as well as unbalances associated with module layout and interconnections.
With the excellent performance of today's power semiconductors, it is common for the packaging to be the main limiting factor for the component as a whole. This may seem surprising at first: why has nearly a totally passive element like packaging, which uses technologies that seem less advanced than microelectronics, not made as much progress as the chips? The answer lies in the multiplicity of packaging functions. Here, perhaps more than anywhere else, the design work consists of finding a compromise between often antinomic imperatives: if it is relatively easy to design an efficient packaging from a thermal or electrical point of view, or to find an inexpensive solution, achieving the three objectives at once is a squaring of the circle! To perform its various roles, a packaging must use a multitude of various elements, materials, and techniques. That is what we are presenting to you in this chapter.
In this chapter, the previous research on the methodologies for the thermal fatigue evaluation of power modules is reviewed based on a mechanical engineering viewpoint such as fracture mechanics and strength of materials. In addition, the research policy for securing the structural integrity of power modules in the design phase is proposed.
Ag sinter paste is becoming an important interconnection technology for die attach in power electronics. It exhibits superior process ability, high-temperature resistance and long-time durability to traditional connection methods such as solder joining or conductive adhesive joining. Massive works have demonstrated Ag sinter paste is capable of achieving a robust and reliable die attach with the Ag finished substrate under a mild sintering condition (pressure-less, low-temperature and atmospheric sintering). However, Ag sinter paste only on the Ag surface metallization is inadequate since there are industrial demands of different surface metallization layers for specific applications or for cutting fabrication costs. Currently, few works have realized robust die attach on different surface metallization layers via Ag sinter paste under mild sintering conditions, and there is still a lack of systematic investigation on the bonding quality and thermal aging reliability of die attach with Ag sinter paste on different surface metallization layers, such as on Ag, Au, Cu and Al. In this chapter, the bonding quality, thermal aging reliability and bonding mechanism of Ag sinter paste on different metal interfaces were introduced for a comprehensive understanding of SiC power modules by Ag sinter paste joining in high-temperature applications.
In this chapter, advanced die-attach validation technologies for power modules are introduced. Especially, sintered silver film as a new bonding material is introduced, and its mechanical characteristics obtained through recent mechanical testing technologies are described. In addition, a new validation technique for the mechanical joint is introduced together with self-propagating exothermic material used as a local heat source for soldering.
The ability to constantly keep track of the actual level of degradation of the modules allows the adoption of preventive maintenance interventions, reducing or even completely eliminating the occurrence of faults during operation, significantly improving the availability of power devices. This capability is particularly interesting in relation to SiC technology, given the higher production costs compared to Si. The structure function is a good engineering tool for monitoring the extent and location of degradation in power modules. Based on this, this chapter presents the development of a test and monitoring methodology, which does not require the addition of power components, and which can be performed during periods of inactivity of the equipment to be controlled. The only requirement in terms of additional circuitry is a measuring circuit, which can be easily integrated, for example, into the gate driver board. A two-level three-phase inverter is considered a case study. The experimental results confirm the possibility of implementing the proposed solution as an on-board health monitoring system for periodic control of the state of deterioration.
This chapter describes an advanced cooling approach, in which the design parameters are adapted to the instantaneous load and ambient temperature conditions, to minimise the temperature variations of the power module, thus reducing its degradation in operation.
The power module for high-power electronics is designed to accommodate semiconductor dies and provide the electrical connection and isolation to and from other components, the extraction and transfer of generated heat in the dies, and protection from environmental conditions (e.g., dust and humidity). The illustration of a conventional power module cross section is presented in Figure 13.1 in which various structure components are highlighted. The structure comprises different materials, such as Al for bond wire, Cu for electrical terminals, and AlN for ceramic-based direct-bonded Cu (DBC) substrate. This multilayer, multimaterial-based structure has limited heat extraction capabilities. Furthermore, certain layers of the structure shown in Figure 13.1 are subject to high mechanical stress due to different coefficients of thermal expansion (CTE) between layers during module power and environmental cycling, thus leading to limited lifetime and early failures caused by thermal stress. This chapter discusses some of the emerging technologies for wide bandgap power modules-including high-performance heat spreaders and substrates-and 3D printed heat sinks.