Harold Hopkins (physicist)

Following a request in the late 1940s from the BBC, who wanted a single lens to replace the classic "turret' of different focal length lenses, he produced the familiar zoom lens. Although there had been earlier attempts to produce a lens which could achieve continuously varying magnification without re-focusing, none of them could provide a good quality image throughout their zooming and aperture ranges. The design of a zoom lens is enormously more complicated and difficult than that of a fixed focal length. The performance of the Hopkins designed zoom lens revolutionised television images, especially outdoors-broadcasts and opened the way to the ubiquitous use of zooming in modern visual media. It was even more remarkable for being produced pre-computer, the ray-tracing calculations being performed on large desk top electro-mechanical machines such as the Marchant Calculator. Even so, the early zoom lenses still fell short of the fixed lenses. The application of computer design-programs based on his Wave Theory of Aberrations in conjunction with new types of glass, coatings and manufacturing techniques has transformed the performance of all types of lenses. Whilst zoom lenses can never out-perform fixed focal lengths, the differences are no longer significant in most applications.

The ancient Romans knew how to heat and draw-out glass into fibres of such small diameter that they became flexible. They also observed that light falling on one end was transmitted to the other. (due to successive reflections from the internal surface of the fibre.) These multiple reflections mix the light beams together thereby preventing an image from being transmitted by a single fibre – (more accurately, the different path-lengths experienced by individual light-rays alter their relative phases rendering the beam incoherent and thus unable to reconstitute the image.) The end result is that the light emerging from a single fibre will be an average of the intensity and colour of the light falling on the incident end.

If a bundle of fibres could be arranged such that their ends were in matching positions at either end, then focusing an image on one end of the bundle would produce a 'pixel-ated' version at the far end which could be viewed via an eyepiece or captured by a camera. A German medical student, Heinrich Lamm produced a crude coherent bundle in the 1930s of perhaps 400 fibres. Many of the fibres were misaligned and it lacked proper imaging optics. It also suffered from leakage where adjacent fibres touched; which further degraded the image. To produce a useful image, the bundle would need to contain not a few hundred but tens of thousands of fibres all correctly aligned. In the early 1950s, Hopkins devised a way to accomplish this. He proposed winding a single continuous length of fibre in a figure-of-eight around a pair of drums. Then, when sufficient turns had been added, a short section could be sealed in resin, cut through and the whole straightened out to produce the required coherent bundle. Having polished the ends, he was then able to add the optics he had designed to provide an objective and eyepiece. Once enclosed in a protective flexible jacket the 'fibroscope' (now more commonly called a fiberscope) was born. Details of this invention were published in papers by Hopkins in Nature in 1954 and Optica Acta in 1955. However, the bare fibres still suffered from light leakage where they touched. At the same time a Dutchman, Abraham van Heel was also trying to produce coherent bundles and had been researching the idea of cladding each fibre to reduce this 'cross-talk'. He published details of his work in the very same issue of Nature. Eventually a system for cladding fibres with a layer of glass of lower refractive index was developed by Larry Curtis et al., which reduced the leakage to such an extent that the full potential of the fiberscope was realised.

Fibroscopes have proved extremely useful both medically and industrially (where the term borescope is commonly used). Other innovations included the use of additional fibres to channel light to the objective end from a powerful external source (typically a xenon arc lamp) thereby achieving the high level of full spectrum illumination needed for detailed viewing and good quality colour photography. At the same time this allowed the fibroscope to remain cool, which was especially important in medical applications. (The prior use of a small filament lamp at the tip of the endoscope had left the choice of either viewing in a very dim red light or increasing the light output at the risk of burning the inside of the patient.) In the medical application, alongside the improvement to the optics, came the ability to 'steer' the tip via controls in the endoscopist's hands and innovations in remotely operated surgical instruments contained within the body of the endoscope itself. It was the beginning of key-hole surgery as we know it. These advances were equally useful in industry.

There are physical limits to the image quality of a fibroscope. In modern terminology, a bundle of say 50,000 fibres gives effectively only a 50,000 pixel image – in addition to which, the continued flexing in use, breaks fibres and progressively loses pixels. Eventually so many are lost that the whole bundle must be replaced (at considerable expense). Hopkins realised that any further optical improvement would require a different approach. Previous rigid endoscopes suffered from very low light transmittance and extremely poor image quality. The surgical requirement of passing surgical tools as well as the illumination system inside the endoscope tube – which itself is restricted by the dimensions of the human body – left very little room for the imaging optics. The tiny lenses of a conventional system required supporting rings that obscured the bulk of the lens area. They were incredibly difficult to manufacture and assemble – and optically minimally useful. The elegant solution that Hopkins devised in the 1960s was to use glass rods to fill the air-spaces between the 'little lenses', which could then be dispensed with altogether. These rods fitted exactly the endoscope's tube – making them self-aligning and requiring of no other support. They were much easier to handle and utilised the maximum possible diameter. As with the fibroscopes, a bundle of glass-fibers would relay the illumination from a powerful external source. With the appropriate curvature and coatings to the rod ends and optimal choices of glass-types, all calculated and specified by Hopkins, the image quality was transformed – light levels were increased by as much as eightyfold with no heat; resolution of fine detail was finally achieved; colours became true; and diameters as small as a few millimetres were possible. With a high quality 'telescope' of such small diameter, the tools and illumination system could be comfortably housed within an outer tube.

Hopkins patented his lens system in 1959. Seeing promise in this system, Karl Storz GmbH bought the patent and in 1967 began to produce endoscopic instruments with a tremendously brilliant image and superb illumination.[6] Thus began a long and productive partnership between Hopkins and Storz. Whilst there are regions of the body that will always require flexible endoscopes (principally the gastrointestinal tract), the rigid rod-lens endoscopes have such exceptional performance that they remain the instrument of choice and have become the enabling factor in modern key-hole surgery.

Previous to Hopkins' work, the resolution of an optical system was mainly assessed using 3-bar resolution charts, with the limit of resolution being the main criterion. But Harold studied at the University of Besançon with Duffieux, who had already begun to lay the foundations of Fourier optics. The seminal paper,[7] which he presented in 1962 when he delivered the Thomas Young Oration of the Institute of Physics, was one of the first to establish the modulation transfer function (MTF) – sometimes called the contrast transfer function (CTF) – as the leading measure of image quality in image-forming optical systems. Briefly, the contrast of the image of a sinusoidal object is defined as the difference in intensities between the peaks and troughs, divided by the sum. The spatial frequency is the reciprocal of the period of the pattern in this image, normally measured in cycles/mm. The contrast, normalised to make the contrast at zero spatial frequency equal to unity, expressed as a function of spatial frequency, is the definition of the modulation transfer function. MTF is still used by optical designers as the principal criterion of image quality, although its measurement in production is less widespread than it used to be. It is calculated from the lens data using software such as OSLO, Zemax and Code V.

Originally an analogue video play-back system, the Philips laserdisc format was adapted to digital in the late 1970s and was the forerunner of the CD and DVD. The digital data is encoded as a series of depressions in a reflective disc. They are arranged along a spiral path that a laser can read sequentially (similar to a stylus following the groove on a vinyl record). The laser must be focused onto, and track this path and the reflected beam must be collected, diverted and measured. The prototype optics to achieve this was an expensive glass-lens arrangement. Hopkins was able to show by a thorough mathematical analysis of the system, that with a carefully calculated geometry, it was possible to use a single piece of transparent moulded-plastic instead. This continues to be a major factor in the low cost of laser disc-readers (such as CD players).