Poll: What Will Succeed Silicon?

-Organics will most likely never be close in performance to silicon, so they are an unlikely successor

-Graphene based devices yet have to demonstrate proper on/off ratio and any remotely viable route to manufacturability. They are only good for mobility and university papers so far.

-CNTs -> No clear manufacturing solution for large scale integration as transistors

-Single electron? This is rather a random buzzword than a technology.

OK Phonon, so it's silicon for ever and ever until the end of time, is it?

Silicon has 20 - 50 more years. Most technologies which have such a large infrastructure investment last for a very long time e.g. internal combustion engine. It costs too much to displace them.

The rate of technical progress is about to fall dramatically and stay low for a long time. Things have moved much faster than the historical average for a few decades driven by cheap money but that has gone now and austerity means less research and development and slower progress.

Carbon nanotubes are far from reality due to lack of a controllable high yield manufacturing technology, but you never know what surprises future (>40 years) may have under its sleeves. Graphene is the best of 5 in the near future.

One could argue carbon nanotubes rely on chemistry while graphene uses lithography.

Molecular electronics? Nice for paper maniacs.

Single Electron? What if it leakS?

My God, Tom, I hope you're wrong. If silicon goes on for 50 more years becoming intrinsically more leaky year by year so requiring more and more complicated process steps to make it useful then we're in the most fearful trouble.

I totally disagree with Tom about prediction of decline in technical progress for semi based on the example of the internal combustion engine progress history.

Last 10 years we have seen more advances than in the 90' (turbo technologies, Atkinson cycle, diesel cycle management improvements etc...) and the future is not that dark for them: variable compression rate technology, diesotto cycles etc. which will significantly improves their efficiency. That's jobs for mechanical engineers but also for electronic ones.

I think graphene is the most likely successor.

As to the inertia of existing technology, it depends on how the end result is put to the consumer. It didn't take all that long for the ipod to displace the cassette.

Many apparent marvels have fallen by the wayside for want of an imaginative use; is there a genius out there who can design and make a product using graphene technology that we will all want to rush out and buy?

Excellent point Peter, my concern is not how silicon is presented to consumers but how it's presented to OEMs. If shrinks don't deliver significant advantages any more then OEMs won't feel compelled to move to new generations of ICs and the semi industry won't get the money to invest in the next generation.

Great point, Djonne, so there's life for analogue engineers, life for mechanical engineers but digital silicon engineers are going up a creek without a paddle.

Who said 1-2 years, Grumpy? The question was what will replace silicon - sometime. After all these years of development it's probably safe to say GaAs, SiGe, GaN never will. But if nothing does, the future of the electronics industry is going to be very different, and much less lucrative, than the last 60 years.

I take your point, David, but I think the best way to present any new technology to OEMs would be to show them how much the consumer would want to buy the end product.

For example, memory metal is amazing, but it hasn't exactly set the world on fire; and that is the danger OEMs face. I don't think they can afford to take up a new technology without being certain of the returns.

It needs a transistor radio, a personal computer, or an ipod to generate sufficient sales.