Browse Prior Art Database

Modeling the Interrupted Development of TNS Resist With Time-Dependent Dissolution Rate Equations

IP.com Disclosure Number: IPCOM000119895D
Original Publication Date: 1991-Mar-01
Included in the Prior Art Database: 2005-Apr-02
Document File: 5 page(s) / 158K

Publishing Venue

IBM

Related People

Rosenfield, MG: AUTHOR [+2]

Abstract

Interrupted development of TNS resist can be modeled by a set of dissolution rate functions of absorbed energy density. Each rate function describes the dissolution rate for a particular development interval where the resist thickness data were measured.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Modeling the Interrupted Development of TNS Resist With Time-Dependent
Dissolution Rate Equations

      Interrupted development of TNS resist can be modeled by a
set of dissolution rate functions of absorbed energy density.  Each
rate function describes the dissolution rate for a particular
development interval where the resist thickness data were measured.

      The purpose of this disclosure is to demonstrate that a set of
time-dependent dissolution rate functions of absorbed energy density
can be used to simulate the interrupted development of E-beam exposed
TNS resist.  TNS is a novolac-based positive resist which gives high
contrast and process latitude when developed with an interrupted
development process.  This process involves partitioning the total
development time into small intervals with the resist being rinsed
and blown dry during each interruption.  The conventional modeling
technique of fitting a single curve to the development rate versus
absorbed energy data cannot be applied because the dissolution rate
is no longer a single function of absorbed energy alone over the
complete development cycle.  Fig. 1 shows the dissolution rate data
plotted as a function of absorbed energy density for a typical TNS
with interrupted development.  At the high absorbed energy regions (>
400 J/cm3), large variations in the dissolution rate are observed.
If a model is obtained from a best- fitted line through the data set,
the model would underestimate the dissolution rate near the surface
and overestimate the rate near the substrate.  Using the technique
outlined in this disclosure, the data are separated into subsets
which correspond to a specific interrupt development segment. This
results in a greatly improved fit to the data.  This simulation
techniq...