Doublings and halvings. These words are in the news all the time. Computers run at double the speed that they did 18 months ago; DNA sequencing costs only half as much this year as last.
The doubling of computer performance is referred to as Moore’s Law, named after Gordon Moore, the founder of Intel Corporation, in honour of his observation in 1965 that the number of transistors in silicon integrated circuits (the brains inside every computer and smartphone) doubles every two years.
This trend has been sustained for nearly 50 years. Over that time, computers have become more than 10 million times more powerful.
Individual technologies come and go but so far Moore’s law shows no signs of being broken. Every day the bubbling cauldron of innovation fills with the creative juices of scientists and engineers, popping and crackling more excitedly in some countries than in others; but overall it stays on the boil.
Calling it a “law”, of course, is a euphemism. Moore’s Law just reflects what Moore observed. It’s not the same as Newton’s Law of Gravitation, which will be true forever. I’ve no doubt that if I did a Rip Van Winkle and slept for 100 years I’d find apples fall from trees with exactly the same acceleration as today. In all likelihood Moore’s law will, sooner or later, be broken.
Nevertheless, I doff my hat to Moore and his “law” not simply because it has lasted for nearly five decades but because it is a measure of the relentless progress of human ingenuity.
In a parallel success story, over the past few years the race to read the human genome has led to DNA-sequencing machines doubling in performance every nine or 10 months, a stunning testament to ingenuity and innovation.
So it is perhaps not unreasonable for some to wonder why we are not seeing a similar pace of technological innovation in an area that is in dire need – solar panel efficiency. After all, these are a silicon-based technology, just like integrated circuits.
The efficiency of silicon solar panels will not double from today’s level.
Sadly, Moore’s law does not apply to solar panels because it gets trumped by a real law: the law of Conservation of Energy.
This law of physics says that energy can neither be created nor destroyed. No matter how ingenious we are, it is not possible, ever, to generate more than 1 kilowatt of electricity from 1 kilowatt of sunlight. This limitation has implications for the efficiency of solar panels.
By 1970, silicon solar panels had reached 10% efficiency. If Moore’s Law of doubling every two years had applied during the 40 years from 1970 to 2010, silicon solar panel efficiency would have reached 10 million per cent. A single array of panels the size of a football field would have been sufficient to meet the electrical energy needs of the whole of Australia.
This would be literally impossible. No panel to convert sunlight into electricity can ever, ever exceed 100% efficiency. Modern commercial solar panels are approximately 20% efficient. One might hope, then, for a couple more doublings in efficiency. Unfortunately, because silicon atoms can only capture and optimally use light of a particular wavelength, the maximum efficiency for silicon panels to convert sunlight into electricity is a little over 30%. This means that the efficiency of silicon solar panels will not double from today’s level. But even today’s benchmark of 20% efficiency is mighty impressive. Consider that most plants harvest sunlight with an average year-round efficiency of around 0.02%. Sugarcane, the top performer, reaches only 1%.
By articulating his law, Gordon Moore highlighted the limitless nature of human ingenuity and innovation. But even Moore’s law loses out to the laws of physics.