HAL Id: jpa-00223886https://hal.archives-ouvertes.fr/jpa-00223886

Submitted on 1 Jan 1984

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

A NEW LITHOGRAPHIC TECHNIQUE FOR THEMANUFACTURE OF HIGH RESOLUTION ZONE

PLATES FOR SOFT X-RAYSM. Browne, P. Charalambous, R. Burge, P. Duke, A. Michette, M. Simpson

To cite this version:M. Browne, P. Charalambous, R. Burge, P. Duke, A. Michette, et al.. A NEW LITHOGRAPHICTECHNIQUE FOR THE MANUFACTURE OF HIGH RESOLUTION ZONE PLATES FOR SOFTX-RAYS. Journal de Physique Colloques, 1984, 45 (C2), pp.C2-89-C2-92. �10.1051/jphyscol:1984222�.�jpa-00223886�

JOURNAL DE PHYSIQUE

Colloque C2, supplément au n°2, Tome 45, février 1984 page C2-89

A NEW LITHOGRAPHIC TECHNIQUE FOR THE MANUFACTURE OF HIGH RESOLUTION ZONE PLATES FOR SOFT X-RAYS

M.T. Browne, P. Charalambous, R.E. Burge, P.J. Duke*, A.G. Michette and M.J. Simpson

Physios Department, Queen Elizabeth College, University of London, Campden Hill Road, London W8 7 AH, U.K. Daresbury Laboratory, Daresbury, Warrington, WA4 4AD, U.K.

Résumé - Une technique de lithographie utilisant un microscope électronique à ba­layage en transmission est en développement. Cette technique est capable de dessi­ner avec précision sur des petites surfaces, des motifs dont l'épaisseur des lignes est inférieure à 50 nm. On décrit la méthode et on discute de ses applications à la fabrication de composants d'optique pour la diffraction desrayons-X mous.

Abstract - A lithographic technique using a Scanning Transmission Electron Micro­scope is being developed. This technique is capable of drawing accurate patterns, with line widths much less than 50 nm, over small areas. The method is described and its application to the manufacture of diffractive optical components for soft X-rays is discussed.

Introduction

One of the major problems in the use of a Scanning Transmission Electron Microscope (STEM) for the study of biological or other material is the build-up of a contamin­ation layer on the specimen surface, causing degradation of the images. Although very high vacuums (4 10" 7 torr) are used, residual gas analyses of the STEM column atmosphere show the presence of, e.g., water vapour, oxygen, nitrogen, and various heavy hydrocarbon molecules. These hydrocarbons, from pumping oils, are the source of the contamination. Briefly, what happens is as follows. The hydrocarbons form a layer on the surface of the specimen, and are polymerised by the electron beam. Thus they can no longer diffuse, and a build-up of the polymer occurs; if the beam is left in one position a "contamination cone" is formed. Since many electrons pass through the specimen, such cones are formed on both surfaces. Much work has been done to reduce or eliminate this contamination [1], while more recently [2] methods of utilising it have been discussed. In this paper we consider the latter.

Beam Writing

A Vacuum Generators HB5 STEM has been used for the generation of patterns suitable for use as soft X-ray diffractive optical components. For this purpose, the speci­men is replaced by a thin ( v» 10-20 nm)carbon film on a standard electron microscope aperture. The films are prepared on mica and floated onto water. The aperture is moved upwards through this and then baked at w 200°C for a few minutes to remove re­maining water. By Scanning the electron beam across the film in a controlled (by microprocessor) way, contamination patterns may be drawn [3]. Due to the limitations of the electron optics, this may only be done accurately over areas with linear di­mensions of a few microns ("primary fields"); for larger size patterns the specimen stage must be mechanically scanned and the primary fields joined together ("patched") using appropriate registration marks.

As the hydrocarbons are used up, the contamination rate decreases. To prevent this, a specimen holder incorporating a small oil reservoir has been constructed. The oil and the film are together maintained at v> 100°C, leading to a high hydrocarbon vap­our pressure (\A10~^ torr) in the vicinity of the film (the high temperature also increases the mobility of the hydrocarbon molecules). This results in constant

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1984222

C2-90 JOURNAL DE PHYSIQUE

replacement (through adsorption) of molecules immobilized by polymerisation, leading to sustained periods of high contamination rate.

The quality of contamination patterns is closely related to both the beam focus and the length of time for which it irradiates any particular point (pixel) of the film. A defocussed beam would create poorly defined lines, while fixed pixel times would result in patterns of uneven thickness due to variations in the contamination rate. The focus problem has been solved by periodically generating a focus-finding scan line, during the course of which the microprocessor minimises the time required for the line spots to reach a certain thickness by changing the focus in 20 nm steps. The minimum time per pixel corresponds to optimum focus.

The microprocessor is able to detect when each spot has reached the required thick- ness by constantly monitoring the difference between the dark and bright field sig- nals. The bright field signal decreases with spot thickness while the dark field signal increases. At a particular thickness the two signals are equal, this condit- ion being independent of beam current variations and of the rate of contamination. The starting level of each signal can be individually set to allow the contamination thickness to be controlled.

Manufacture of Linear Diffraction Grating

To perfect the pattern generation method and to test for distortions a (nominally) - 1 1 pm x 30 pm linear grating with a pitch of (nominally) 110 nm (a 1.8 x 104 lines/ nun) was manufactured. An area of 5 um x 5 pm was chosen as the primary field,ad- dressable with an accuracy of 2.5 nm. A row of 16 registration lines (each .A 1.3 pm long) was drawn on the first primary field, each line being accurately registered with respect to the first. The film was then mechanically shifted so that the last line was brought to the position previously occupied by the first, and 15 more lines were drawn, each registered with the new first line. After repeating this process until the row of lines was v 30 pm long, the exact position of each line was deter- mined by registering with the first line of the relevant primary field and the line was then extended by about 5 pm on either side. Two more lines were then drawn in each of the gaps between the lines. After baking at -500°C for a few minutes to re- move excess oil, gold was evaporated onto the film (at an angle to prevent the gaps being filled). The resulting grating is shown in figure 1 - the thick lines perpen- dicular to the grating structure are for support. Subsequent analysis showed the pitch to be - 125 nm ( 4 10% larger than intended) due to the STEM being poorly calibrated. This, and other effects causing the distortions visible in figure 1, such as uneven motion of the mechanical stage giving occasionally misplaced lines, have since been corrected. A rotating anode X-ray generator, with a carbon coated anode, was used to obtain a diffraction pattern using the grating (figure 2 ) . Lines due to carbon K*, , oxygen K.c (presumably from the carbon paint stabiliser1,and sil- icon K6 (from vacuum grease) can be seen.

Fig. 1 (left) STEM gemrated linear diffraction grating .Irr, Fig. 2 (above) Diffraction pattern obtained using the STEM grating. The subscripts indicate the diffraction order

Manufacture of Zone P la tes

The f i r s t zone p l a t e s drawn on the STEM have 100 zones, an outer diameter o f d O um and an outermost zone width o f 4 5 nm. This i s well within the c a p a b i l i t i e s of the technique, and subsequent zone p l a t e s w i l l have much f i n e r zones, and there fore higher reso lu t ions . The zone p l a t e s were drawn on carbon f i lms i n i t i a l l y coatedwith a t h i n l ayer of contamination (giving a t o t a l f i l m thickness of 20 nm) f o r s t a b i l - i t y . The pa t te rn was generated i n 24 equal angular sec tors each containing th ree prlmary f i e l d s . The sec tors were al igned by r e g i s t r a t i o n t o an accurately drawn in- ner c i r c l e . Each a l t e r n a t e zone was drawn a s a s e r i e s of u 20 nm wide a rcs , each reg is te red with a mark drawn a t the inside of t h e primary f i e l d . A v i s u a l check was made on the posi t ioning of the zones a f t e r every t h i r d of an ou te r primary f i e l d , every half of a middle primary f i e l d and every inner primary f i e l d . This ensured t h a t the zones could be positioned with an accuracy of no worse than 5 nm. After each sec tor the s tage was ro ta ted through 15' and the process repeated; thus each sec tor was drawn with the beam i n t h e same region, so t h a t x-y d i s t o r t i o n was mini- mised. Since each sec tor covered an angle of 16", the re was a s l i g h t overlap a t the ends of the a r c s , ensuring a good join. After a l l the sec tors had been drawn, the c e n t r a l zone was f i l l e d i n (hiding the r e g i s t r a t i o n c i r c l e and inner r e g i s t r a t i o n marks). An example of such a zone p l a t e i s shown i n f igure 3 - the contamination thickness i s - 0.15 um and t h e p a t t e r n i s uncoated. Methods of coat ing t h e p a t t e r n s with heavy metal l ayers a r e cur ren t ly being invest igated.

Fig. 3 A STEM generated zone p la te , showing r e g i s t r a t i o n marks

Further papers i n these proceedings w i l l d iscuss the e f f e c t s of manufacturing inac- curacies on the imaging proper t i es of STEM generated zone p l a t e s [4] and proposed t e s t s of such zone p l a t e s i n X-ray microscopy [5] . One i n t e r e s t i n g p o s s i b i l i t y us- ing such manufacturing methods i s t o make "composite" zone p l a t e s , i n which the cen- t r a l zone p l a t e pa t te rn , of f i r s t order focal length f , i s surrounded by f u r t h e r zones whose t h i r d order foca l length is f ( f igure 4 ) . This i s a way of increasing the resolut ion (by a s much a s a f a c t o r of 2 ) and increasing the on-axis focussed in- t e n s i t y ( f igure 5) when the manufacturing method l i m i t s the narrowest Line t h a t can be drawn.

JOURNAL DE PHYSIQUE

- 12.00 i - 100 *one= 0.02

_-- - - 150 zone*

10.00 - \ 200 zones

X ........ 250 zones 2 C - ', ' , I,-.--- : 8.00 - ! i al , ', 350 zones Y . , C ' ! - al . $ 8 , . . . . . - . . . ,.

.. ' 5

! 4 4.00 - - ' \ al ' I , . C !

L2I I - '\ \ .

2.00 -----. x!.

0.03 0.20 0.03 0.20

Radtue

0.00 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20

Fig. 4 A composite zone plate Fig. 5 Intensity distributions for 100 inner zones and various numbers of outer zones

Acknowledgements

We are grateful to the Science and Engineering Research Council (SERC) for financial support. MJS is supported by an SERC grant.

References

[I] See, e.g., REIMER L., and WACHTER M., Ultramicroscopy 2 (1978) 169-174. HARADA V., TOMITA T., and WATABE T., Scanning Electron Microscopy 11 (1979) 103-110.

[2] BROERS A.N., CUOMO J., HARPER J., MOLZEN W., LAIBOWITZ R., and POMERANTS M., Ninth International Congress on Electron Microscopy, Toronto vol. I11 (1978) 343-354.

[ 3 ] BROWNE M.T., CHARALAMBOUS P., and BURGE R.E., Inst. Phys. Conf. Ser. 9 (1982) 43-44.

[4] SIMPSON M.J., BROWNE M.T., BURGE R.E., CHARALAMBOUS P., DUKE P.J. and MICHETTE A.G. These proceedings p.

[5] BURGE R.E., DUKE P.J., MacDOWELL A., MICHETTE A.G., MILLER A., ROSSER R., SIMPSON M., and WEST J.B. These proceedings p.