This project intends to develop a new Integrated Circuit (IC) technology developed specifically to build improved driver circuits for the newly emerging silicon carbide (SiC) based power transistors. These SiC transistors are enabling much more efficient power electronics which is leading to improved energy efficiency in many application areas. However the adoption of these devices is being limited by the absence of suitable driver ICs, ie able to operate at the elevated temperatures at which the power transistors run. The new technology integrates low voltage transistors, built in a 3C SiC epitaxial layer, grown on a silicon wafer. This technology promises to offer good performance, excellent high temperature capability and be much lower cost than possible all SiC alternatives. The innovative steps include the development of the 3C SiC on Si hetero-epitaxy, the development of the 3C SiC IC process and the design and development of a demonstrator driver IC.

The Silicon Carbide Interconnect Optimisation Project (SCIOP) aims to develop a novel form of die interconnect for the power electronics sector. The project objectives are to explore alternative materials, processes and adhesion technologies which can contribute to improved power module performance and simplified production processes. Improved performance and reduced cost in the enabling technologies of power electronics, due to the pervasive nature of the technology (electric motors, power grids, etc.), has the potential to make a significant contribution in helping the UK exploit semiconductor innovation and in a broader context help move to a low-carbon economy.

Anvil’s unique patented 3C-SiC/Si material and devices will shrink power converters in size, weight and cost, whilst significantly increasing efficiency. Conventional SiC devices (4H-SiC) enable many of these improvements but the cost is too prohibitive for many applications. Anvil is currently developing 650V 3C-SiC MOSFETs and diodes using the 3C-SiC material already developed, but 1200V devices require thicker SiC layers (12µm instead of 8µm) and as wafer bow levels scale with growth thickness, it requires additional stress relief techniques to enable fabrication of the 1200V devices required for many applications within the $billion market for PV inverters, electric vehicles, UPS’s and industrial motors. The stress relief is achieved by carefully tuning the growth parameters and starting substrate. This multi-variable growth process has been optimised using experimentation and numerical modelling and has resulted in low bow, high crystal quality wafers for 650V devices. Extending this to 1200V is not just a case of extending the final growth phase but needs a re-optimisation of the whole growth process with significant experimentation with variables, extrapolation of the models, and characterisation of the grown material using electron microscope and X-ray diffraction.

The project’s goal is to make low-cost 3C-Silicon Carbide (SiC) power devices ready for real-world power applications. Power conversion is a major area of inefficiency in all power systems with silicon (Si) based systems reaching their limit. The use of compound semiconductors in power conversion is widely accepted as a route to a significant increase in efficiency and reduction in size/ weight. However, penetration of SiC devices has been limited by their high cost; currently an order higher than Si equivalents. Anvil has developed a unique technology that, by growing 3C-SiC on Si wafers, has the potential to enable the production of SiC components at a similar cost to Si ones, hence eliminating this barrier to their adoption. Anvil has early prototype devices but achieving the maximum efficiency in applications using wide bandgap semiconductors is not simple. It needs high-speed drivers, novel inverter designs, closed coupled layouts and high-temperature packages. This is best done as a system of device and inverter together rather than piece part design. The purpose of this project is to do exactly that; using models and simulations to optimise circuits and feedback optimal device design characteristics and demonstrate the potential user benefits of 3C-SiC power converters.

Awaiting Public Project Summary

Driven by competition, demand and legislation designers of products are striving for increased efficiency, smaller size and weight and lower cost, but they are limited by the efficiency constraints of Si or the cost of today’s SiC devices. Anvil’s unique SiC technology enables the development of devices with the efficiency and size benefits of SiC but at the cost of silicon. However the benefits that can be achieved by changing to SiC are limited by the switching speeds which are in turn limited by the inductances produced by non-close coupling of discrete devices and ancillaries. This project is to develop a low cost hybrid module to enable close coupling of devices and ancillaries, reduce inductances and achieve switching speeds of 100KHz. This significantly increases efficiencies and reduces size and weight by removing ancillary components and heat sinks. The potential applications for such a module are very wide indeed: for example LED lighting, PV converters, general power supplies, electric car charging and EV/HEV.

The purpose of the project is to develop high efficiency, low cost GaN based LEDs on Anvil’s silicon carbide on silicon wafers (3C-SiC/Si). Anvil has recently completed an Innovate UK funded feasibility study with the University of Cambridge that demonstrated its 3C-SiC layers, developed for low cost high efficiency power devices, have an exciting application in LEDs by providing a cubic substrate that enables the growth of single phase cubic GaN on large diameter silicon wafers. The ability to produce cubic GaN on large diameter silicon wafers is clearly recognised as a key enabler for increasing the efficiency and reducing the cost of LED lighting. Plessey have started to commercialise LEDs produced in conventional (Hexagonal) GaN grown on large diameter silicon wafers using IP originally developed at The University of Cambridge. Anvil’s IP manages the inevitable stresses when growing SiC on silicon wafers. The project brings these three technologies together, to produce high efficiency, low cost LEDs. Such a cost/performance improvement would have a disruptive effect on the LED market advancing the replacement of incandescent lights and CFLs with solid state lighting.

A key challenge facing the UK Distribution Network Operators (DNOs) today is the increasing demand for power being placed on residential networks e.g. by the proliferation of electrical vehicles (EVs) and the move to electro-heat. Also, the increase in distributed generation (DG) is now resulting in unacceptable local voltage rises. This project follows on from a TSB Feasibility Study which showed that a cost effective solution to these problems can be achieved on the existing infrastructure by increasing the local network phase voltage to 400 V (existing cable is rated at 600V). To step the voltage back down to 230 V at each house, DNO-owned, low-cost, 99% efficient power electronic converters (PECs) will need to be installed in the meter-box. Our previous study showed that the 99% efficiency was essential to avoid over-heating in the meter-box and hence new, low-cost 3C SiC devices were mandatory. This system will not only increase network capacity, but also provide optimised connections for emerging EV charging, DG and energy storage - the “smart-grid”. The project will develop a PEC prototype which will be deployed by Western Power Distribution in a small-scale demonstration of the project.

Anvil Semiconductors has developed technology that enables the growth of device quality 3C-SiC wafers on large diameter Si substrates for use in power devices, delivering the performance and efficiency benefits of SiC at the price of silicon. The IP resolves the problem of the stress which is inevitable when growing SiC on Si. This feasibility study determines whether the IP has applicability to LEDs with 3C-SiC providing an effective buffer layer for GaN growth on large diameter Si wafers. The stress relief technology may also enable the growth of GaN with reduced dislocation density, leading to low cost, high intensity LEDs. Such a cost/performance improvement would have a disruptive effect on the LED market making replacement of incandescent lights and compact fluorescents with solid state lighting commercially viable, potentially reducing the world energy consumption by some 11%.

One of the key challenges facing the UK electrical Distribution Network Operators (DNOs) today is the increasing demand for power being placed on residential networks e.g. general load growth, the proliferation of electrical vehicles (EVs) and electrified heat. At the same time, there are problems associated with an increase in the connection of distributed generation (DG) and energy storage devices. The aim of this project is to provide a significant increase in capacity over the existing infrastructure by increasing the local network voltage and/or distributing DC. This will be achieved by the deployment of DNO-owned, low-cost, high-efficiency power electronic converters (PECs) installed at individual properties and a local substation converter for DC distribution. The performance that these PECs need can only be achieved using new Silicon Carbide (SiC) semiconductor switching devices. However, the cost of existing commercially available SiC is very high, which conflicts with the need for low-cost. Therefore this project will consider a new, highly innovative fabrication technique for SiC power devices. The proposed system will not only provide an increase in the capacity of the existing network, but will also provide flexible, optimised connections for emerging EV charging, DG and energy storage devices..

Anvil Semiconductors is developing the concept of a new, very power efficient device structure in SiC power electronics and requires the services of a patent lawyer to secure the IP around this idea.