ULTRAVIOLET MICKOSCOPY 



use of a Fresnel zone plate whose opaque and their differential absorption is therefore 



bands are made of thin gold and whose open greater in this region (1(5). On the other 



bands are completely transparent to radia- hand, the wavelengths from 10 to 100 A, 



tion between 10 A and 1000 A, because they which are very long x-rays, are considered 



are empty. very short compared with the wavelengths 



A Fresnel zone plate consists of alternately of ultraviolet radiation, and are of potential 



opaque and transparent bands l)ounded by interest in astrophysics. Also, the lines 



circles of radii around 150 A and 300 A emitted by the sun's 



/- , „ o /v^ corona would be of interest as sources for 



r„ = ri\/n; n = 1, 2, 3, • • • (i) . . . 



telescopic miage formation (19). 



Its principal focal length, /, obeys the simple Now consider the focal length, /. In 



relation microscopes we want / to be small, but in a 



j^ ^ ^ 2 (2) telescope a long focal length is an advantage ; 



in fact, a focal length of 100 cm would be 

 where X is the wavelength, and n is the practical and one of 1000 cm, though some- 

 radius of the central circle. what large, is not completely beyond possi- 

 In considering the focusing of x-rays for bility if we could build the right kind of 

 microscopy we see that the product /\ can platform. The reader must bear in mind that 

 be disturbingly small. For example, in an the ultimate platform for telescopes operat- 

 x-ray microscope one might consider / = 1 i^g in the soft x-ray and extreme ultraviolet 

 cm and X = 1 A. This requires n^ = f\ = regions must be a satellite orbiting above 

 10-8 em2 or n = lO"'' cm or one micron. By the earth's atmosphere. 

 Eq. i w^e find that the width of the nth zone is Xq choose figures of the right order of 



sn = (V^m - V^)n, (3) "magnitude, consider / = 100 cm and X = 



100 A. Using n = 25 we get s„ = ri/2\/n = 



which, for large n is approximated by iq-z cm or ten microns. This value is about 



^^ ten times the resolving power of the finest- 



*" ^ 2\/n ' grained photographic films and makes the 



construction of a zone plate for soft x-rays 



If, for example, w = 25 and ri = 10"^ cm, seem feasible. 



then Sn = 10-^/2\/25 cm = 10"^ cm or 0.1 However, a zone plate made by exposing 

 micron. Since the best photographic emul- a photographic film would be useless in the 

 sions (e.g., Eastman 649 spectroscopic region between 10 and 100 A because the 

 plates) have a resolution of about a micron, base on which the emulsion rests, and even 

 a zone plate made to these specifications the emulsion layer itself, are practically 

 looks impossible indeed. The product /X opaque to these wavelengths, 

 needs to be considerably larger before zone Recently we have succeeded in producing 

 plates look feasible. Fortunately, the cur- a zone plate whose opaque elements are thin 

 rent interest in the use of x-rays and euv in concentric bands of gold made self-support- 

 microscopy and astrophysics can lead to ing by the use of thin radial struts (see Fig. 

 larger values of both f and X. 3). The streaks in the photograph are im- 

 Consider X first. Wavelengths much longer perfections in the form of very thin filament- 

 than 1 A are important for different reasons, like pieces of gold. The outer diameter of 

 In x-ray microscopy, for example, there has the zone plate is 0.2596 ± 0.0002 cm. The 

 been a trend towards wavelengths approach- central circle has a diameter of 0.0426 ± 

 ing 100 A because microscopic objects of 0.0002 cm. The bands are bounded by circles 

 biological interest have low atomic numbers whose radii obey the relation r„ = 



554 



