213 0 Journal of the Royal MicroscopicalSociety, VoZ. 83, Pts. I & 2, June 1964. Pages 213-216 SCANNING ELECTRON MICROSCOPY By W. C. NIXON (Engineering Laboratory, Cambridge University) PLATE 97-98 AND ONE TEXT-FIGURE SYNOPSIS Scanning electron microscopy is a method of microscopy that permits resolution better than that of the optical microscope (about 100 A) while examining one surface of a bulk specimen. The technique depends on electronic application to microscopy and has been developed over many years by electronic engineers rather than physicists or microscopists. The main field of application so far has been nonbiological, but two biological examples are given here. The reference list includes the main papers on scanning electron microscopy written in this laboratory. INTRODUCTION CONVENTIONAL transmission electron microscopy has reached the stage where commercially available instruments will yield routine resolution of 10 A with suitable specimens. However, suitability is sometimes hard to reproduce with thin sections that must be used in transmission. In some cases thin sections or foils may not be prepared at all. A resolution between that of the optical and electron microscope would be acceptable if one polished surface of a solid specimen could be examined as in normal optical metallurgical microscopy. If this could be achieved then the specimen could be heated, cooled, strained, fractured, etc., while under observation, which is not possible with a replica and only to a certain extent with thin foils. The scanning microscope produces this type of micrograph. METHOD The basic features of a scanning electron microscope are shown in text-fig. 1 (K. C. A. Smith (1961).-"Encyclopedia of Microscopy", ed. G. L. Clark. Reinhold Pub. Corp., p. 241). The electron-optical column has the electron gun at the bottom and the electron beam is accelerated upwards by a potential of about 20 kv. Two magnetic electron lenses reduce the size of the electron source in the gun, say 50 p, to a few hundred hgstroms at the specimen surface. With the most modern instruments using three electron lenses the probe size is less than looa (R. F. W. Pease & W. C. Nixon, to be published). The resolution of the instrument is given by this electron-probe size. The electron current in such a probe is less than one micromicro ampere and the aberrations of the present lenses prevent an increase in this current. This electron probe is scanned across the surface of the specimen by the use of beam de%ecting coils within the electron-optical column. The scanning generator is also connected to the scanning coils of the display and photographic-recording cathode ray tubes. Mamcation is achieved by reducing the current to the column coils while leaving the
214 Transactions of the Society Text-fig. 1. same current in the cathode-ray-tube coils. In this way a small raster of a few microns is scanned on the specimen while the cathode-ray-tube face contains a raster about 10 cm square. The magnification is given by the ratio of these two rasters and can be up to 100,000 times. Several hundred lines are scanned in each frame with a frame rate of about one second for visual work and up to 5 minutes for photography. The longer time permits integration of the signal and suppression of the noise due to the particulate nature of the electron beam in the main microscope column. The electron detector is now designed to collect the secondary electrons emitted from the specimen over a large solid angle by applying up to -I- 10 kv to the collector. These accelerated secondary electrons from the specimen strike a plastic scintillator and the light produced reaches a photomultiplier by passage through a perspex light pipe. The electrical signal from the photomultiplier is proportional to the number of secondary electrons emitted from the surface of the specimen as the electron probe is scanned slowly across the surface to build up the picture on the long-persistance cathode-ray-tube screen line by line. This signal modulates the brightness of the cathode-ray-tube screen and the final photograph shows the variation of collected electrons from the specimen surface. The contrast is due mainly to topographic variations as shown in the accompanying photographs. The range of contrast may be very great and so a gamma control is incorporated in the amplifier between the photomultiplier and the cathode-ray-tube control electrode. Contrast may thus be expanded or contracted electronically to suit the specimen. These principles have all been incorporated in the well known scanning electronprobe X-ray microanalyser with the addition of X-ray detection and therefore contrast display due to the variation of X-rays (i.e. the variation of elements) from point to point on the surface. A final comment on the method is that although the micrographs are obtained by a scanning electron beam it is very difeicult to see the scanning lines. The line and frame scanning rates and the electron-spot size in both the electron column and the cathode-ray tube are all adjusted so that the lines will not show on the subsequent micrograph. The
Scanning Electron Microscopy 215 universal acceptance of visible lines on domestic television receivers obscures the interpretation of scanning images of a higher standard. SCANNING ELECTRON MICROGRAPHS A selection of scanning electron micrographs is presented to show the main features of this type of result common to all specimens. The specimen is inclined at an angle to the electron-beam direction as shown in text-fig. 1. As a result the micrographs are foreshortened as if one were viewing the surface of the specimen at a shallow angle. The collector is set to one side and so the micrograph appears to be lit from one direction. These two features are shown in P1. 97, fig. A, where the specimen is a square gridlying on a solid surface. The square is now a rectangle and the highlights are on the edgesof the grid. Once these features are recognized the unknown specimens may be interpreted correctly. The magnification varies within the micrograph depending on the direction of measurement due to this foreshortening. This is shown in the result by printing a small ellipse on the micrograph as shown in P1. 97, fig. B. The length of the major axis is written within the ellipse and the eccentricity of the ellipse is chosen to match the degree of foreshortening so that the minor axis and in fact any line through the centre of the ellipse will also represent this length on a parallel line on the micrograph. This is a logical extension of t,he customary micron mark on transmission electron micrographs. CONCLUSIONS The scanning electron microscope may be applied to selected biological specimens if care is taken in interpreting the result. The first commercial instrument has been demonstrated at the Physical Society Exhibition by the Cambridge Instrument Company. Wider use of these instruments should show which fields of application, including the biological, may be usefully explored by this relatively new method of non-conventional electron microscopy. ACKNOWLEDGMENTS Text-fig. 1 has been reproduced from K. C. A. Smith (1961), Encyclopedia of Microscopy, ed. G. L. Clark, Reinhold Pub. Corp., p. 241; P1. 97, fig. B, C, D, and PI. 98, fig. G, H, are reproduced from K. C. A. Smith and C. W. Oatley (1955), Brit. J. appl. Phys., 6, 391-399; P1. 97, fig. A, and P1. 98, fig. E, F, were taken originally by K. C. A. Smith at the Engineering Laboratory, Cambridge University. REFERENCES BOYDE, A. & STEWART, A. D. G. (1962).-A study of the etching of dental tissues with argon ion beams. J. Ultraatruct. Res., 7, 159-172. EVERHART, T. E., SMITH, K. C. A., WELLS, 0. C. & OATLEY, C. W. (1958).-Recent developments in scanning electron microscopy. Fourth Internat. Conf. Electron Micr., Berlin, pp. 269-273. Springer-Verlag, Berlin. -& THORNLEY, R. F. M. (1960).-Wide-band detector for micro-microampere low-energy electron currents. J. sci. Instrum., 37, 24G-248. -, WELLS, 0. C. & OATLEY, C. W. (1959).--Factors affecting contrast and resolution in the scanning electron microscope. J. Electronics Control, 7, 97-1 11. MCMULLAN, D. (1953).-An improved scanning electron microscope for opaque specimens. Pror. Inst. Elec. Eng. (Lond.), 100, Pt. 11, 245-259. OATLEY, C. W. & EVERHART, T. E. (1957).-The examination ofp-n junctions mith the scanning electron microscope. J. Electronics, 2, 568-570.
21 6 Transactiorns of the Society SXITH, K. C. A. & OATLEY, C. W. (1955).-The scanning electron microscope and ita fields of application. Brit. J. awl. Phycr., 6, 391-399. STEWART, A. D. G. & BOYDE, A. (1962).-1on etching of dental tissues in a scanning electron microsmp. Ndure, Lond., 196,8142. THORNLEY, R. F. M. (1960).--Recent developments in scanning electron microscopy. Proc. Europ. Reg. Cod. Electron Micr., Delft, I, 173-176. WELLS, 0. C. (1959).-Examination of nylon spinneret holes by scanning electron microscopy. J. Electronics Cdrol. 7, 313-376. DESCRIPTION OF PLATES PLATE 91 Fig. A.-Scanning electron micrograph of a square grid lying on a solid surface. Fig. B.Surface of a meal worm grub, Tenebrio mlitor, silver coated. x 476. Fig. C.-As in Fig. B but at twenty times higher magnification showing a single bristle on the surface. x 9600. More recently Thornley (1900) has been able to operate a scanning electron microscope at 1.5 kv and avoid the charging of insulating specimens. One application of biological intereat is the use of frozen but not freeze-dried material directly in the microscope. Fig. D.-Orlon fibre, gold-palladium coated. x 1800. PLATE 98 Fig. E.-Etched aluminium, specimen angle 26". x 1800. Fig. F.-As in Fig. E but with a specimen angle of 45". Fig. G.-Tungsten x 1800. point in contact with germanium surface, before discharge of a Condenser through the point of contact. x 600. Fig. H.-As in Fig. G but after condenser discharge, showing the ploughing up of the germanium surface. x 000. This last pair of micrographs demonstrates the use of the scanning electron microscope to observe dynamic changes in a specimen by viewing the same field during all stages of the experiment.
W. C. Nixon JOURN. R. MICR. SOC., Vol. 83, PI. 97 A B [To face p. 216
JOURN. R. MICR. SOC., Vol. 83, PI. 98 W. C. Nixon E H