The Royal Academy of Engineering has announced the shortlist for the MacRoberts Award, the UK's biggest innovation prize. The winning team, which gets a cheque for £50,000, will be announced in three weeks' time.

Chairman of the judging panel Dr Robin Paul said it will be very hard to chose between the four finalists, as the companies represent the "very best of British ingenuity and inventiveness". Among the nominees are a self-cleaning window, 3D glasses for use in airport security checks and IBMs Websphere MQ product set.

1. Delphi Diesel Systems, nominated for its E3 electronic unit injector advanced fuel system. By controlling the fuel injection very precisely, the two-valve E3 system lowers emissions of soot, exhaust gases and nitrous oxides. This will allow diesel engines to meet a raft of strict emission guidelines, and means lower fuel consumption. Delphi sold 50,000 units in 2003, and expect to double that in 2004.
2. IBM UK, nominated for its WebSphere MQ middleware infrastructure component. WebSphere MQ can link application on over 40 platforms, making it easier to build interconnected systems without custom coding everything.
3. Pilkington PLC, for its self-cleaning glass. Pilkington developed a micro-crystalline titanium dioxide coating for windows that actually breaks dirt down on its own. It also makes water spread over the surface of the glass more easily. The coating is applied when the glass is at 650°C, and must be only 15nm thick, with variations of less than 1nm, or the glass will become distorted.
4. Sharp Laboratories, for switchable 2D and 3D displays. The 3D view is created using the Parallax Barrier effect to direct discrete images on an LCD screen towards each eye. The user's brain recombines these images as a 3D picture. This can be switched off, leaving a normal 2D image. The technology could find a home in airport security, allowing staff at the X-Ray machine to examine luggage in 3D. ®

Related stories

IBM bangs drum for client middleware
DNA-based nanobot takes a stroll
Medical imaging research awarded £4.5m