ELSEVIER Microelectronic Engineering 30 (1996) 305-308 MICROELECTRONIC ENGINEERING Electron beam lithography using chemically-amplified resist: resolution and profile control V.A.Kudryashov a, V.V.Krasnov a, S.E.Huq b, P.D.Prewett b and T.J.Hall c a Institute of Microelectronics Technology, Russian Academy of Sciences, 142432 Chernogolovka, Moscow District, Russia b Central Microstructure Facility, Rutherford Appleton Laboratory, Chilton, Didcot, Oxon OXll 0QX, UK Department of Electrical Engineering, King's College London, Strand, London, WC2R 2I_S, UK The effect of post exposure bake and softbake conditions on the sensitivity of AZPN114 has been investigated experimentally. A bilayer system for undercut structures has been achieved using two layers of AZPN114 with different softbake temperatures. A single layer of AZPN1 i4 has also been used to produce undercut and tailored resist profffles by two different multiple exposure strategies at different beam energies. 1. INTRODUCTION Recent advances in electron beam resists have included the development of novolac based systems using chemical amplification of a latent image. These resists exhibit high sensitivity (typically <10ptC cm'Z), high contrast and resolution in the sub-micron regime. Additionally, their resistance to plasma etching makes them attractive for pattern transfer using dry-etch processes. The negative tone commercial chemically amplified resist AZPNll4 (Hoechst) is therefore of considerable interest for a range of processes. 2. POST EXPOSURE BAKE CONDITIONS The radiation sensitive resist component in a chemically amplified resist (CAR) is decomposed under electron irradiation producing an acid. A post-exposure bake (PEB), typically of several minutes duration on a hot plate at - 100°C, is required to activate the acid. This then serves as a catalyst for the cross-linking or scission mechanism. Each acid molecule is responsible for a large number of cross-link or scission events through the mechanism called chemical amplification. Since changes in the average molecular weight of the resist take place as a result of a catalysed chemical reaction during heating, it is obvious that this change has to depend on the PEB time and temperature. For PEB with greater temperature and/or time, a smaller number of acid molecules will be sufficient to produce the same changes in the average molecular weight distribution and so a smaller e-beam exposure is required, corresponding to a higher sensitivity. [1] The dependence of dose upon PEB temperature in the range 95 - 115°C and time between 0.5 and 4.0 min for AZPNll4 is shown in Fig. 1. 0.5 1 2 4 iiiiiiiiiiiiiiiiiiiiiiii!iii!iiiiiiiiiiiiiiiiiiiiiiiiiiiiiil liiiiiiiii iiiiiiiilililiiiiiiiiiiiiiil 1~15 110 Figure 1 Dose dependence on PEB parameters for AZPN114 [x axis - time (min); y axis - temperature (°C); z axis - exposure dose (J.tC cm-2). 30keV beam energy] It is obvious that the same resist sensitivity could be obtained with different combinations of PEB 0167-9317/96/$15.00 © 1996 - Elsevier Science B.V. All rights reserved. SSD I 0167-9317(95)00251-0