Light fantastic
To celebrate the International Day of Light, James McKenzie reflects on the wonders of optical communications
I hope you don’t need me to remind you, but Monday 16 May 2022 will be the fifth International Day of Light. It marks the anniversary of the first successful operation of the laser on 16 May 1960 by physicist and engineer Theodore Maiman at Hughes Research Laboratories in California. Of all the many “international days” in the calendar, this to me is the most important, given how vital the laser has been for digital communications, medicine, spectroscopy and countless other fields besides.
What’s particularly interesting this time round, however, is that we’re bang in the middle of the 2022 International Year of Glass. Now I know glass is a fascinating material in many different ways, but its most interesting applications stem from the fact that light can pass so easily through it. In particular, the refractive index of glass can be controlled by tweaking the nature and number of additives to enable even more amazing things to happen. Without glass-based fibre optics, the world would be a truly poorer place.
Physics World has a special issue on glass planned for next month, but let me get you in the mood by reflecting on just how far glass-based fibre optics have come since those early days of the laser. That progress was wonderfully illustrated two years ago when researchers from Japan, the US and France sent one petabit (a million gigabits) of data per second down a single-core, multi-mode optical-fibre cable over a distance of 23 km. That smashed the previous record of 0.4 petabits per second, which is remarkable when you think that the first commercial systems in the late 1970s could do barely 6 megabits per second.
Back in time
It’s all a far cry from Maiman’s first laser, which consisted of a cylinder of synthetic ruby 1 cm in diameter and 1.5 cm long, the ends of which were silver-coated to create a partially reflective “Fabry–Perot cavity”. When Maiman pumped the ruby rod with light from photographic flashlamps, he produced a deep-red 694 mm laser pulse. Once those basic principles were demonstrated and understood, scientists could make lasers from a range of materials, including semiconductors.
As for the optical fibre, it was first produced in 1954 by Dutch scientist Abraham van Heel when he covered a bare glass fibre with a transparent coating. This cladding had a lower refractive index than the fibre, preventing light from leaking out. Those primitive fibres had large cores and were used to make “fibrescopes” that could see round corners and look inside the body. However, they lost signal quickly, with attenuations of more than 1000 decibels per metre (db/m) – roughly seven orders of magnitude worse than today’s best fibres.
It was at Standard Telecommunication Laboratories in Harlow, UK, that Charles Kao and co-workers famously showed that the high loss of those early fibre-optic cables was due to impurities in the glass, rather than from any underlying problem of light scattering in the glass itself. In fact, Kao and George Hockham concluded in 1965 that the lower limit for light attenuation in glass was under 20 dB/km. Being less than the value for copper wire, it meant that optical communications was possible – if such a low-loss glass could ever be made.
When Kao first suggested that optical fibre could replace copper for long-distance communications, his ideas were ridiculed. Undeterred, Kao and colleagues pressed ahead and by 1969 he had measured the intrinsic loss of “bulk-fused” silica to be 4 dB/km. It was the first evidence of ultra-transparent glass, opening the door to the modern fibre-optics industry as we know it. Kao, who died in 2018, went on to win the 2009 Nobel Prize for Physics for his achievements.
Meanwhile, laser technology was progressing apace too. The first semiconductor laser chip was developed in 1970 and the first narrow-wavelength distributed feedback (DFB) laser was created two years later. By the mid 1970s, optical-communications systems were able to exploit the low losses of fibres at infrared wavelengths, which were below 0.47 dB/km for light at 1250 nm. An even lower-loss window was discovered in optical fibres at 1550 nm and by 1978 the Japanese firm NTT had created a fibre with a loss of just 0.2 dB/km.
Across the ocean
By 1988 the first transatlantic fibre-optic cable (TAT-8) had been laid, carrying data at a rate of 280 Mbit/s (equivalent to 40,000 telephone circuits) between the US, UK and France (the 8 in its name indicating it was the eighth cable to cross the Atlantic). Another key breakthrough was the development of erbium-doped fibre amplifiers (EDFAs) by teams led by David Payne at the University of Southampton and Emmanuel Desurvire at Bell Labs in 1986 and 1987. Operating at 1550 nm, they promised to make long-distance fibre systems cheaper by reducing – or eliminating entirely – the need for optical-electrical-optical repeaters.
The amount of data that can be sent down fibre was boosted further with the development of wavelength division multiplexing (WDM), which lets fibres simultaneously carry laser light of different colours. By 1996 the TAT-12 transatlantic cable had been installed, which used EDFAs instead of repeaters to provide a transmission speed of 5 GB/s. Two years later, the TAT-13 submarine fibre cables had been upgraded with WDM with three wavelengths, boosting their capacity to 15 Gb/s per cable. The concept of “fibre to the home” swiftly followed and by 2017 Corning had shipped over 1 billion kilometres of optical fibre – enough to go round the Earth 25,000 times.
Optical fibre is now the backbone of telecommunications networks and there seems to be no shortage of demand, with the total amount of data they carry projected to grow to more than 180 zettabytes (180 × 1021 bytes) by 2025. According to a recent Verified Market Research report, the global optical communications business was estimated to be worth $18.7bn in 2020 and could reach almost $38bn by 2028, growing at a compound annual growth rate of more than 5.5% from 2021 to 2028.
Isn’t “glass” fantastic!