Pwh. De Jager et al., Experimental results of the stochastic Coulomb interaction in ion projection lithography, J VAC SCI B, 17(6), 1999, pp. 3098-3106
Throughput and resolution are connected in ion and electron projection lith
ography (IPL and EPL) because of the space charge and Coulomb interaction b
etween the particles in the beam. Due to the lack of experimental data it w
as not possible Co estimate this effect accurately. Therefore an experiment
setup has been developed which has the most significant parameters close t
o planned IPL exposure tools. These parameters are the linear particle dens
ity and the crossover shape and size. The stochastic Coulomb interaction bl
ur, depending on the total beam current, has been measured in about 100 set
tings of the beam current, beam energy, and crossover shape. The results sh
ow that the stochastic Coulomb interaction blur scales to the power of 0.58
7+/-0.101 (1 sigma) of the Linear particle density in a system with a unifo
rm crossover of 400 mu m. To decrease the current density in crossover IPL
systems can have an aberrated crossover. In case of this type of crossover
of 670 mu m the current dependency is 0.820+/-0.072 (1 sigma). With an aber
rated crossover the stochastic Coulomb interaction is not uniform over the
exposure held; it decreases towards the edges of the exposure field and pla
teaus near its center. This reflects the shape of the crossover. The experi
mental results can be used to make predictions of the allowable current in
planned IPL exposure tools. The process development tool can have 1.00+/-0.
25 (1 sigma) mu A He+ ion current at 100 nm resolution. The allowable H+ io
n current in the beta tool at 100 nm resolution is 9.09+/-2.14 (1 sigma) mu
A and 3.69+/-0.81 (1 sigma) mu A at 50 nm. This corresponds to a throughpu
t of 40.1+/-0.4 (1 sigma) wafers per hour (300 mm) at 100 nm and 26.1+/-2.0
(1 sigma) wafers per hour (300 nm) at 50 nm. (C) 1999 American Vacuum Soci
ety. [S0734-211X(98)17606-6].