Knyazev M.A., Dubonos S.V., Svintsov A.A., Zaitsev S.I.
Institute of Microelectronics Technology, Russian
Academy of Sciences,
A new method for fast definition of an electron resist contrast is presented. The development velocity of an electron resist can be written :
where is the development rate of the electron resist exposed to dose , is the development rate of the electron resist exposed to dose (resist sensitivity), and is the resist contrast. Using (1), a test structure (dose wedge) for contrast definition can be designed. The dose wedge is a set of small rectangles closely following each other (Fig.1). During exposure, the dose increase for each sequential rectangle. The dose for the first (left edge) rectangle is zero, reaching the maximum for the last (right edge) one, . If the number of rectangles is N, then the exposure dose for the n-th rectangle is:
Three different profiles can, generally, be revealed after
the development of the exposed dose wedge (Fig. 1). As shown in Fig. 2, when
higher than the real resist contrast exp
the profile is concave, when
the profile is convex, and when
are equal the profile is plane. Therefore, when
the change in the residual resist thickness is linear. The real test structures
consist of several dose wedges. The exposure dose for each wedge was calculated
(from 1.5 to 4.5) using (2). Fig 3 shows the result of test structure development
in a MMA copolymer resist exposed by a JEOL 840A under NanoMaker hardware/software
system  control at 25 KV and developed in IPA-H2O 8:1 solution.
The exposure data were prepared using the software part of the NanoMaker according
to (2). Fig 3. is an optical microscope photo obtained using a red light filter.
Generally, the dose wedge looks as shown in Fig. 3 in an optical microscope
with any light filter. When the residual resist thickness grows linearly, the
distance between neighboring intensity minima or maxima is constant. In the
the distance decreases with a thickness increment and, vice versa, if
the distance between minima or maxima increases with a thickness increment.
For the test structure in Fig. 3, the value of the resist contrast is between
3 and 3.5. The considered methods is similar to approach developed for measurements
of proximity function parameters .
The partial support of the EC-funded project NaPa (Contract no. NMP4-CT-2003-500120) is gratefully acknowledged. The content of this work is the sole responsibility of the authors.
 A.R. Neureuther, D. F. Kyser and C. H. Ting. IEEE Trans. Electron
Dev. ED-26, 686, 1979.
|Copyright © 2002-2017 Interface Ltd. & IMT RAS|