Computer memory has increased rapidly over the past few decades but, as scientists struggle to reduce the size of conventional computer chips any further, these advances will sooner or later hit a wall. So can nanotechnology offer a way forward?
It seems inconceivable that the floppy disk was once the flag-bearer
for digital storage devices. Well into the 21st century, computers
regularly retrieved data from these relics of computing antiquity.
Their power was on the wane by 2007, though, as 98% of home
computers sold in the UK that year arrived without floppy disk drives.
That radical shift was a very practical demonstration of the power of
the ideas of nanotechnology.
A 3.5-inch floppy disk could store a measly 1.44 megabytes of data –
not even enough to fit a single mp3 music file. Today, a small one gigabyte (1,000 megabytes) USB flash drive can store about 200 songs. Inside that drive, electrical charges that encode the data are applied to areas across a silicon chip called memory cells, each of which is made from transistors.
The recent explosion in this cheap, high-capacity hardware is down to the ability of computer engineers to reduce the size of the memory cells that make up the devices. In addition, they can pack the cells closer together – in modern storage devices, the cells can often be separated by as little as 25 nanometres. And progress has been rapid: the number of transistors on a chip doubles roughly every 18 months.
The advances have been so rapid that they will soon reach a wall. Pack the cells much closer together and electrical interference between them shoots up, affecting the device’s performance. Again, nanotechnology ideas can come to the rescue – using graphene (a one-atom-thick sheet of carbon) in a layer between the silicon and the memory cells can keep electrical interference low, even past the 25 nanometre barrier, as researchers in California and Australia have shown.
But Michael Kozicki, director of the Center for Applied Nanoionics at
Arizona State University, sees a further roadblock ahead: “Everyone in
the industry agrees that, when we get down to the 11 nanometre level,
we’re done; we won’t be able to travel any further down the charge
To get around this, Kozicki has been working on a technique that stores information by creating tiny nano-sized bridges of copper or silver between two electrodes. In his device, if a bridge is present, it dramatically reduces the electrical resistance between the electrodes.
This technique, called programmable metallization cell memory (PMC),
makes it possible to store one terrabyte of data (1,000 gigabytes) on a single USB stick, enough for 2,000 hours of music. Flash memory USB sticks of this size are already on the market, but retail at a hefty $3,000 (£1,900).
As computer chips continue to get smaller and more powerful, so their prevalence in our everyday lives is set to increase. Kevin Ashton, a British technologist who created a system for tagging and tracking objects using radio frequencies, has predicted a future where everything is connected to the internet via tiny computer chips embedded within, or as he called it, an “internet of things”. A fridge is already available with an on-board computer, allowing it to know its contents, order food when you run out and even suggest suitable recipes, before setting the oven to the right cooking temperature. It is also currently possible to control an entire room – the thermostat, light switch, TV, stereo etc – all from a tablet or smartphone using wirelessly connected chips in each of the controlled devices.
Future developments will also come from probing a scale one hundred
times smaller than the cells in current silicon chips – the realm of quantum computers. “Quantum computing is computation at the level of individual atoms, molecules and photons,” says Artur Ekert, professor of quantum physics at the University of Oxford.
The idea exploits the fact that electrons do not exist in a single fixed state, but can simultaneously exist in many states at once. Each state can be assigned a property, rather like the binary code used to encode information in the “bit” of a standard computer, which is represented either as 0 or 1. The difference between a standard computing bit and a “qubit” (quantum bit) is that the latter can be both a 0 and a 1 at the same time.
This means many calculations can be performed simultaneously – a quantum computer with just 42 qubits would match the fastest supercomputer in Europe, JUGENE, which is capable of one quadrillion operations a second.
Commercial quantum computers are still decades away, but physicists have already built simple versions by trapping atoms using magnetic fields and lasers – the trapped atom corresponds to a qubit. These experimental computers currently fill entire rooms, much like the early electronic computers did in the 1950s and 1960s. Ekert believes nanotechnologists will need to work hard to find ways to make a “convenient commercial interface between quantum technology and the everyday world”.
In its simplest sense, a computer is just a machine capable of performing computations. It doesn’t have to be electronic. Tom Ran, of the Weizmann Institute of Science in Israel, works with computers made out of strands of DNA.
“Working this way, we can get three trillion computers, working in parallel, in a space the size of a water droplet,” he says. The 0s and 1s of conventional computers are replaced with the four DNA bases: A, C, G and T. Operations can be translated into strands of DNA using these bases, and the way the DNA strands interact with each other produces new strands which can be decoded as output values.
The attraction is that these inherently biological computers can interact directly with living cells. One goal is to programme DNA-based computers to work inside the body to combat disease. Take cancer as an example – chemotherapy drugs currently target all rapidly dividing cells, including, for example, hair cells; a DNA computer could be programmed to identify and kill only cancerous cells.
Ongoing work to miniaturise conventional computers on the nanoscale,
as well as the manipulation of individual atoms and strands of DNA to
bring us entirely new types of calculating machines, is pushing the
boundaries of what the computers of tomorrow can achieve. Computing
has come a long way since the days of the floppy disk.