One thing leads to another in the search for more efficient LEDs


A small consultancy project enabled by Cambridge Enterprise has blossomed over a short space of time into a major new initiative to develop and commercialise greener and more efficient light emitting diodes (LEDs).

The new technology could even lead to LED light bulbs that mimic natural light, changing wavelengths over the course of the day and reducing the impact on our body clocks of living under artificial light. 

‘My first response when Anvil Semiconductors contacted us out of the blue and asked if we could grow some cubic Gallium Nitride for them was, “It’ll never work”‘, says Professor David Wallis of the Cambridge Centre for Gallium Nitride, leader of the research group that worked on the initial project and continues to develop the technology.  

Gallium Nitride, GaN, is the key component of LED light bulbs. It is used in crystalline form, with the crystals predominantly grown on to sapphire substrates, but the resulting hexagonal crystal structure imposes limitations on the ultimate performance of the LEDs. ‘The hexagonal crystal phase of GaN currently used in LEDs is problematic since it creates internal electric fields, which limit energy efficiency, particularly as the wavelength of light produced is extended to green wavelengths,’ says Wallis. 

A more efficient form of GaN, the cubic phase, was known, and in theory would be the better choice for LEDs. However, cubic crystals had proved difficult to grow in the past and the easier-to-produce hexagonal phase became the industry norm. 

Anvil Semiconductors, which produces crystals for electronic devices with cubic silicon carbide-on-silicon technology, saw that if they could successfully use these as substrates for the cubic form of GaN, it would open up new possibilities – and potentially new markets – for their products. They came to Cambridge searching for the experts, and after his initial reaction, Wallis decided that the idea was worth pursuing. 

‘We were happy to have a go, and Cambridge Enterprise, through Cambridge University Technical Services, made it possible for us to do the work, funded by a small grant from Anvil Semiconductors.’ 

Work by other researchers had suggested that the chosen growth method, metalorganic chemical vapour deposition (MOCVD), would not be ideal for cubic GaN. But despite the conventional wisdom, Wallis and his colleagues persisted, and eventually managed to produce some of the best quality cubic GaN that had ever been grown. 

The initial project was so successful that  Anvil Semiconductors and the team in the Cambridge Centre for Gallium Nitride went on to secure an Innovate UK project to collaborate with Plessey Semiconductors to develop the technology further. A second Innovate UK project followed, and more recently, the team was awarded a three-year Engineering and Physical Sciences Research Council (EPSRC) grant to explore the fundamental physics of the cubic phase of GaN with a view to better understanding its properties and how to use them. 

‘People have been trying to solve the problems created by the internal electric fields and wavelengths in LEDs for for more than a decade,’ Wallis says. ‘These cubic GaN crystals could be the answer.’ 

Wallis and his team have already demonstrated that the cubic form of GaN does not generate internal electric fields, so it could be used to develop more energy-efficient LEDs. Now they are working on the challenge of demonstrating working devices. 

‘The ideal is to develop an efficient LED light bulb that generates white light from red, green and blue wavelengths. Currently, this is not possible due to the very low efficiency of commercial green LEDs. The solution at present is to add a phosphor compound to blue LEDs that produces yellow light, but the resulting white light can seem a bit harsh to the human eye. Cubic GaN LEDs should be able to generate green light, and so could be used alongside blue and red LEDs to give LED light bulbs that can produce more natural light, or even be tuned throughout the day to mimic the natural variations of sunlight.’ 

In addition to the research, a spin-out company, Kubos Semiconductors, was launched in November 2017 to commercialise the new technology. Kubos has already reached a technical milestone and been awarded a grant from the Energy Entrepreneurs Fund managed by the UK Government’s Department of Business, Energy and Industrial Strategy.   

In just three years, what was a small consultancy project has attracted additional funding and grown into a major initiative to create new LEDs that are not only better for their users, but also better for the environment. 

Image caption: GaN LED wafer on probe station  Image credit: Cambridge Centre for Gallium Nitride

To read more information, click here.

Cambridge Enterprise exists to help University of Cambridge inventors, innovators and entrepreneurs make their ideas and concepts more commercially successful for the benefit of society, the UK economy, the inventors and the University.

Cambridge Enterprise, University of Cambridge