Novel GaN Power Devices and Packaging Technologies for 300 degC Ambient Operation

Lead Research Organisation: University of Glasgow
Department Name: School of Engineering

Abstract

This proposal straddles two key topics, High Temperature (HT) Electronics and Power Electronics. Present electronics is silicon-based and therefore limited to maximum operating junction temperatures of less than 150 degC, which gives a maximum ambient ceiling of around 105 degC. Commercially available components (rated for operation at elevated temperature) are in the range of 210-225 degC maximum. Therefore electronics for the automotive sector, especially for the emerging electric vehicles, and the aerospace sector is kept as far from the engine as possible to minimise the cooling requirements. Similarly, oil and gas engineers, attempting to harvest the fossil fuels (which we are still highly dependent on), face exactly the same problem with the electronics that are driving the drilling tool motor. Power electronic devices delivering hundreds of Watts of power to the motor must do so in an ambient that can exceed 225 degC, operating 10 km or deeper under the ground with only slurry pumped from the surface to cool the devices (temperature and time restrictions apply). The potential benefit for having electronics operating in these environments without cooling is huge, leading to greater efficiency, reliability, saving space, weight and importantly cost.

Power Electronics plays a very important role in the electrical power conversion and is widely used in transportation, renewable energy and utility applications. By 2020, 80% of electrical power will go through power electronics converters somewhere between generation, transmission, distribution and consumption. So high-efficiency, high-power-density and high-reliability are very important for power electronics converters. The conventional Si-based power electronics devices have, however, reached the limit of their potential (after almost 40 years of development). The emergence of wide-bandgap material such as silicon carbide (SiC) and gallium nitride (GaN) based devices has brought in clear opportunities enabling compact, more efficient power converters, operating at higher voltages, frequencies and powers, and harsh environments (e.g. 300 degC ambients) and so can meet the increasing demand by a range of existing and emerging applications. Advances in GaN device structure and in process technology to significantly improve performance are pushing the adoption of these new power devices for very high voltage (>600 V), high temperature (>125 degC) and high power (mainly 6-40kW) applications. This trend is set to continue as the technology evolves. For 600V operation, a threshold voltage +3V would be desirable (well above the +1.6V maximum now achievable) for improved noise immunity. Also, presently, the device architecture compromises converter performance, e.g. in a half-bridge power converter module the current through the top switch transistor is modulated by its floating substrate potential. When this deficiency can be solved, the two transistors of the basic building block of all power electronic systems can be manufactured as a single integrated circuit reducing switching path inductance thus allowing faster switching and smaller cheaper passive components, increasing switch yield per wafer for the small devices targeted and reducing packaging costs.

Reliable packaging methods for the new devices and ICs are indispensable for the required testing during development, and for the eventual exploitation in industrial HPHT applications. The required materials and joining methods at >300 degC ambient environments are completely different from those of conventional electronics, and need to be developed. These challenges with HT electronics and GaN switches/packaging form the main motivations for this project.

The project brings together the UK's key academic and industrial expertise to work in synergy to investigate HT packaging and GaN power devices to realise a robust and high performance High Power High Temperature (HPHT) technology.

Planned Impact

Many industries are calling for electronics that can operate at extremely high temperatures. The oldest user of high temperature (HT) electronics is the down-hole oil and gas industry. Temperatures in these hostile wells can exceed 200 degC (approx.25 degC/km depth), with pressures >25 kpsi. Active cooling is not efficient in this harsh environment, and passive cooling techniques are not effective when the heating is not confined to the electronics. As such, system reliability is of utmost importance, as the cost of downtime due to equipment failure can be quite severe. For example, by some estimates a failed electronics assembly on a drill operating miles underground can take more than a day to retrieve and replace; and the rate for operating a complex deep-water offshore rig can be as high as £1M per day, which calls for electronics amenable to this environment such as GaN.

In the aerospace sector, engine monitoring and control is traditionally a centralized architecture managed by a computer/controller system. There are many sensors and actuators in close proximity to the hot engine, all of which must be hard-wired to the controller with long, heavy wire harnesses with hundreds of conductors and multiple connector interfaces. Moving to a distributed control scheme places the engine controls closer to the engine, reducing the complexity of the interconnections (factor of 10), saving hundreds of pounds of aircraft weight and increasing the reliability of the system. This, however, requires that the sensors and actuators are connected to smart nodes in close proximity to the engine, thus fuelling the demand for HT electronics. Although electronics can be cooled in this application, it is undesirable for two reasons: cooling adds cost and weight to the aircraft, and, most importantly, failure of the cooling system could lead to failure of the electronics that control critical systems. In addition to engine monitoring, there is a desire to put sensors and instrumenation on other hot areas of the aircraft, such as landing gear and braking systems. Also, the trend toward more electric aircraft calls for a replacement of hydraulic, pneumatic and mechanical systems with electronics. These electronics may need to be in high ambient temperature sections of the craft or require densely packed power electronics. If the power electronics could run without cooling it would significantly simplify the environmental control system, improve reliability and decrease weight. The trend is similar in the automotive industry, with the industry migrating from purely mechanical and hydraulic systems to electromechanical or mechatronic systems, which requires locating sensors, signal conditioning, and control electronics closer to heat sources. For example, the mechatronic system consisting of a transmission controller co-located on the mechanical transmission assembly must operate at 150 degC ambient. Other locations on the vehicle such as under the hood, on wheel sensors, and in the exhaust would also use HT electronics.

Therefore, aiding the implementation of GaN power devices for HT electronics, which is the aim of this proposal, would result in major economic benefit for the UK from this emerging technology either directly through epitaxial wafer supply (e.g. IQE), GaN manufacturing (Memsstar, INEX & Optocap) or indirectly through its application in the oil and gas industries, e.g. BP, and in the automotive and aerospace sectors (companies such as Rolls-Royce, Jaguar LandRover, etc.). These companies need to understand and de-risk any new technology before it can be implemented cost-effectively. A successful demonstration of the HPHT technology will therefore impact key sectors in society, the oil and gas industry, and the automotive and aerospace sectors with attendant benefits including economic, energy savings, etc. to society.

Publications

10 25 50
 
Description A new technique to keep devices cool under operating conditions was developed. It also leads to high performance devices. It uses a so-called distributed architecture in which the device comprises active and inactive regions, implemented is a planar distributed architecture.
Exploitation Route Our technique helps with the realisation of more efficient devices and is applicable across the device family under investigation, and so is likely to be used by others
Sectors Aerospace, Defence and Marine,Digital/Communication/Information Technologies (including Software),Education,Electronics,Energy