Bulletin Special Supplement on Prof. Charles K. Kao, Former Vice-Chancellor and Nobel Laureate The Love and Labour of a Laureate

24  Chinese University Bulletin Special Supplement 2010  But the paper pointed out that the intrinsic loss due to scattering could be as low as 1 dB /km, wh i ch wou l d hav e a l l owed p r opaga t i on ov e r p r a c t i c a l distances. The culprit is the impurities: mainly ferrous and ferric ions at these wavelengths. Quoting from the paper: ‘It i s fo r e s e eab l e t ha t g l a s s e s with a bulk loss of about 20 dB /km at around 0.6 micron will be obtained, as the iron- impurity concentration may be reduced to 1 part per million’. In layman terms, if one has a sufficiently ‘clean’ type of glass, one should be able to see through a slab as thick as several hundred metres. That key insight opened up the field of optical communications. The paper considered many other issues: The loss can be reduced if the mode is chosen so that most of the energy is actually outside the fiber. The fiber should be surrounded by a cladding of lower index (which became the standard technology). The loss of energy due to bends in the fiber is negligible for bends larger than 1 mm. The losses due to non-uniform cross sections were estimated. The properties of a single-mode fiber (now a key technology especially for long distance and high data rate transmission) were analyzed. It was explained how dispersion limits bandwidth; an example was worked out for a 10 km route — a very bold scenario in 1966. It may be appropriate to quote from the Conclusion of this paper: The realization of a successful fiber waveguide depends, at present, on the availability of suitable low-loss dielectric material. The crucial material problem appears to be one which is difficult but not impossible to solve. Certainly, the required loss figure of around 20 dB/km is much higher than the lower limit of loss figure imposed by fundamental mechanisms. Basically all of the predictions pointed accurately to the paths of developments, and we now have 1/100 of the loss and 10,000 times the bandwidth then forecast — the revolutionary proposal in the 1966 paper was in hindsight too conservative. • • • • •