EPSRC Engineering Fellowships for Growth: Narrow Band-gap Semiconductors for Integrated Sensing and Communications

Semiconductor materials power much of the current economy, through their use in the ubiquitous computer and much else besides. The most common semiconductor is silicon, and this accounts for about 90% of the world market. There are some other types of semiconductor, however, that provide functions that silicon can't address but which are also very important. Examples of these include: gallium arsenide, which is used in satellite receivers and mobile phones for the communications parts; materials based on indium phosphide, which are used in lasers in CD and DVD players and for long distance communications along optical fibres; and materials based on gallium nitride, which are used to make the white light emitting diodes that are now being used for a range of energy efficient lighting and even in car headlights.
All of these materials belong to a family known as III-V semiconductors, because they contain a mixture of elements from group 3 and group 5 of the periodic table. III-V semiconductors account for most of the remaining 10% of the electronics industry, and are worth approximately £25bn per year worldwide and growing at about 7%p.a. Unlike the silicon industry, the UK has a significant presence in the manufacture of electronic components based on these materials, as well as systems based upon them, and is in a good position to benefit from the rapid growth in the market.
Another member of this III-V semiconductor family in indium antimonide, a compound of indium and antimony, which has the formula InSb. InSb has several interesting properties. Charge carriers can be made to go faster than in any other member of the family and take less voltage to do so. Consequently, this material has the potential to make components that will operate at very high frequencies whilst consuming very little power and so, for example, enable future mobile devices to download massive amounts of data, such as streaming high definition video, without draining the battery or clogging the network. Another application is to enable imaging for detection of illicit explosives or firearms, without use of any harmful radiation. These materials might even find their way into future computers to enable the doubling of computing power to continue every two years, as it has for the last forty years. Other properties of the material mean that we can make infrared sensors for thermal imaging or detection of harmful gases, or photovoltaic devices that would make much more efficient solar energy systems.
A corollary of these properties is that heat can cause the materials to "leak" charge, even at room temperature, so currently the only commercial applications are in high performance thermal imaging systems, where the application can tolerate the cost of having to provide cooling to -200C to make them work. This need to cool was previously assumed to be fundamental, however Ashley and co-workers have shown that this is not necessarily the case, and that uncooled operation is possible in several applications.
This research will put in place the core technology that would enable a range of devices to be made that will work without any cooling. This technology includes being able to make features on the devices that are more than one thousand times smaller than a human hair and still have the devices operating effectively. It includes the addition of "nano-antennas" to the devices to improve their sensitivity to infrared light by orders of magnitude. It also includes work to show that the devices could be integrated with silicon, to benefit from the system cost savings derived from the massive investment in the silicon industry. The successful outcome of this research would be that various industries in the UK are able to quantify the benefits that the technology offers and make decisions to develop it into products. These would include the sensor manufacturers; prospective new companies in the mobile communications field; and renewable energy community.

This proposal includes elements of underpinning concept and technology development, which will ultimately lead to a wide range of novel types of device that will impact across several fields. In order to provide an immediate demonstrator for the research, however, we have also chosen one specific component set of a mid-IR LED and photodetector for use in gas sensing, with the objective being to improve the sensitivity of a sensor by 10x. Therefore the most direct, immediate impact will be with the manufacturers in the gas sensing market, of which there are several operating in the UK (Alphasense; BW Technologies; Crowcon; e2v; Gas Sensing Solutions; Honeywell Analytics; MSA).
We will work closely with one specific manufacturer, Gas Sensing Solutions Ltd. (GSS), to ensure that the important metrics for the components are clearly understood and to provide a rapid route for initial exploitation. GSS estimate a 3x increase in their addressable markets from a successful outcome. Whilst GSS is only an SME, this indicates the overall factor by which the UK's ability to address the gas sensor market may be increased. Private communication from CEO of GSS indicates that the world market for NDIR sensors is £2Bn, and that this is multiplied by a factor of 8 for controllers/ hand held portables incorporating the sensors. The proportion addressable by the UK is 5 to 10%, giving the potential total value to the UK to be of the order £1Bn for components, sensors and portable systems.
The technology developed under this project will also have major impact in other fields. InSb-based materials offer major advantages for ultra-high speed transistors that will operate with very low power consumption for mobile communications. The exponential rise in data-rate has virtually exhausted all techniques to squeeze more from existing frequency bands, so a move to include supplementary bands at higher frequencies is inevitable. The world smartphone market is forecast to be approx. £130bn by 2015, of which the transceiver components represent approx. 5%. Hence if the UK can capture a fraction of the design and component market it would be very valuable. A UK based private equity company, Anglo Scientific, is interested in this prospect and would be a route to establishing a UK based capability in ultra-high frequency transceivers if the technology is promising.
THz components would also have application in THz sensing for security and industrial process control applications. Smiths Detection, in the UK, is active in this field and would be a likely route to market. The technology also offers major benefits for infrared thermal imaging products, where the UK has a major manufacturer in Selex ES, A Finmeccanica Company.
On the world stage, the technology may impact in future microprocessors, and Intel will be involved with the project to guide and monitor progress.
We will work with a UK based semiconductor manufacturer, Compound Semiconductor Technologies Global Ltd. (CST), to ensure that commercialisation proceeds rapidly. CST will be regularly updated on progress and involved in the manufacturability issues. They will, therefore, be able to rapidly adopt the technology and take it to market. CST is, however, not the only potential route to market in the UK. Other, higher volume capabilities exist at Plessey Semiconductors and Compound Photonics. Neither of these companies currently has an activity in the antimonide materials, but both have demonstrated a willingness to take on III-V technology variants if the markets justify the investment. We also have a relationship with IQE, which is a major player in the supply of III-V epitaxial wafers and is establishing a business in the Concentrator PV market that would benefit from the technology.
Societal impact will be through the products derived from the technology, including communications and "internet everywhere"; improved environmental monitoring; enhanced security and primary healthcare.