The early days of the electron microprobe were characterized by the variety of designs emerging from different laboratories in Europe, the United States and the USSR. Notable amongst these was that of Castaing in 1954, which employed a magnetic lens in combination with an optical microscope for viewing the sample and positioning the electron probe on the desired point for analysis. The X-ray emission was analysed by two high resolution spectrometers having their axes in the same plane as the electron-optical axis, and with their foci accurately set to coincide with the point of impact of the electron probe. This was a design well suited to point analysis by high resolution X-ray spectroscopy and formed the basis of the first Cameca instrument (Fig 1a).

By contrast, work by Duncumb in the Cavendish Laboratory in Cambridge started with the object of scanning the electron probe over the sample, in order to image the surface in terms of its characteristic X-ray emission. This required a strong lens to give a high current into a finely focused electron probe (Fig. 1b). The first element maps were demonstrated in 1956, and led to the design of Cambridge Instruments’ Microscan, intended as a metallurgical instrument, in conjunction with D.A. Melford of Tube Investments Research Laboratories.

Meanwhile, Long, also in Cambridge, was pioneering applications to mineralogy, and built an instrument for studying conventional slide-mounted rock samples, which could be viewed optically while analysis was in progress (Fig. 1c). This made use of a weaker probe-forming lens, with space for an inclined sample to be viewed in transmitted light. The slim design of the lens allowed it to be partially enclosed in the spectrometer, which received X-rays leaving the sample at a high angle to the surface - a benefit carried through into the Cambridge Geoscan.