杏吧原创

Light work wins Nobel for electronics pioneers

The Nobel prize in physics is shared by the inventors of optical fibres and of the chips that made digital cameras possible

Shine a light, they've won Shine a light, they鈥檝e won

The has been awarded to some of those whose work with light laid the foundations of the modern digital age.

The first half of the prize went to Charles Kao, formerly vice-chancellor of the Chinese University of Hong Kong, who in the 1960s made it possible for the world to talk via the light inside optical fibres.

The second half was awarded to Willard Boyle and George Smith at Bell Labs in Murray Hill, New Jersey, for the invention of the charge-coupled device (CCD) image sensor chip 鈥 a crucial component in today鈥檚 digital cameras.

High fibre

Kao鈥檚 work made communication through fibre optics possible over great distances. Glass has a higher refractive index than air, so most light travelling along a glass rod remains trapped inside 鈥 a phenomenon known as total internal reflection. But in the early 20th century, guiding light by refraction in this way was possible only over short distances 鈥 for instance, it found a role in the instruments used by medics to peer into a patient鈥檚 gastrointestinal tract.

Most people thought it was impossible to send light further through optical fibres because of attenuation 鈥 the loss of light from the fibre. State-of-the-art optical fibres available in the early 1960s lost light at a rate of 1000聽decibels per kilometre. So, communications networks based around optical fibres seemed a remote possibility.

In 1965, Kao was working at Standard Telecommunications Laboratories in Harlow, UK, and suggested that attenuation was due to impurities within the glass of the optical fibre rather than the inherent physical properties of the silica from which the glass was made. With purified glass, he suggested that attenuation rates below 20聽decibels per kilometre would be possible.

Post Office sees the light

Kao spent the next five years trying to sell his dream of fibre-optic communications to other research labs. The British Post Office research station at Dollis Hill 鈥 the UK counterpart of Bell Labs 鈥 was quick to see the advantages of the technology and installed basic optical fibres between telephone switching offices in neighbouring towns.

Kao鈥檚 predictions were vindicated in 1970, when the Corning Glass Works in New York succeeded in making fibres with far lower attenuation. Other labs had tried to purify ordinary glass, but Corning chose to start with a highly purified form of silicon dioxide, called fused silica, which it had developed earlier.

By doping the light-carrying core of a fused silica fibre with titanium, Corning reduced attenuation to 17聽decibels per kilometre. Later, Corning reduced attenuation rates to just 4聽decibels per kilometre with germanium dioxide as a dopant.Today鈥檚 best optical fibres have attenuation below 0.2聽decibels per kilometre.

Forty years on from Kao鈥檚 work, optical fibres lie at the heart of modern telecommunications, including the internet.

Digital heart

The second half of the prize was awarded for the invention of the CCD image-sensor chip.

Like the transistor, for which a Bell Labs team also won a Nobel prize in 1956, the CCD ousted an existing technology that used vacuum tubes.

Whereas the transistor ousted fragile vacuum tube amplifiers and switches, the CCD knocked a fragile, image-smear-prone vacuum-based device containing photoconductors off its perch as the sensor of choice in television cameras.

Later, the CCD would become the heart of consumer camcorders and digital cameras 鈥 and even the technologies that have superseded it have borrowed its principles.

Bubbling up

When Boyle and Smith began work on the CCD, they were thinking of it as a novel memory chip for videophones and computers. It was inspired by 鈥渂ubble鈥 memory, an emerging technology at Bell Labs in the 1960s that recorded data as magnetic domains, dubbed bubbles, in thin semiconductor films.

Bubble memory was eventually destroyed by advances in hard disk recording, but Boyle and Smith wondered if it was possible to do a similar job with electric charge: could small 鈥渂uckets鈥 of electrons be used to represent binary data in a microchip? And if so, how could the data be stored and then moved around to be read out?

The pair developed an ingenious way of doing this. They realised that they could deposit columns of closely spaced capacitors on a chip, each of which could be filled up with electrons when a 鈥渂it鈥 of data was to be saved in it. Each capacitor had a 鈥渃locking鈥 wire attached to it, and a pulse applied via this wire made the contents of each charge bucket tip into the next one. In this way, a series of 鈥渃lock pulses鈥 would allow a whole row of bits to be marched off to the edge of the chip, where they would be read as a sequence of ones or zeroes by a transistor.

As the charges were coupled 鈥 tipping into the next cell along 鈥 the devices became known as charge-coupled devices or CCDs. But because the way they worked was similar to the way low-tech firefighting teams pass buckets of water to each other, the devices also became known as 鈥渂ucket brigade devices鈥.

Memory to light

But the chips鈥 light sensitivity means they didn鈥檛 remain as memories for long. Light focused on a CCD could knock electrons out of their shells, creating a charge in the cell proportional to the intensity of light at that point. In other words, Boyle and Smith had developed a high-resolution analogue image sensor that could shunt the charges created by each row of cells, or pixels, to be read as digital data 鈥 with none of the low sensitivity, fragility and image-smear problems inherent in vacuum tubes.

, the CCD chip was built into a video camera by 1970 and was producing broadcast-quality TV images by 1975. But it has also revolutionised imaging in fields of science such as astronomy and medical imaging.