Quantum Material Seminars - "Attoscience and lithography : how to connect them using ab initio"

17/11/2020

Attoscience and lithography: how to connect them
using ab initio


M. J. van Setten1


1Principal member technical staff, IMEC, Kapeldreef 75, 3001 Leuven (Belgium)

 


If you google lithography, you will be directed to various descriptions of the process of
printing images on paper using stone. This may seem quite far removed from atto science
and ab initio methods. The method however did give birth to what is more correctly called
photolithography, the core technique of the presently 1.e12 semiconductor industry.
In the semiconductor industry we talk about technology nodes. At present we are
transferring from the 5 to the 3 nm node. Roughly this means that the smallest dimension
of the interconnect wires on the next generation of microchips will be around 3 nm. Moving
to these dimensions has two main consequences. Firstly, the fraction of atoms in such
structures that is not bulk-like anymore starts to dominate. Consequently, many non-ab
initio modeling methods start to fail. Secondly, visible, and even moderate UV light, is too
‘big’ to print patterns like this. To solve this the industry is moving to Extreme UV light
sources with a photon energy around 92 eV. This light interacts with the so-called
photoresist to cause chemical reactions that form the desired pattern. At 92 eV the
interaction starts at semi core electrons, starting a new and yet largely unknown chemistry.
Understanding this chemistry is pivotal to develop the next generations of semiconductor
devices.
Enabling research into this EUV chemistry has been one of the main driving forces of the
development of the atto lab at imec. The core of atto lab are two pump probe laser beams,
one fixed at 92 and one tunable from 20 to 120. Pulses with a delay of 300 as to 300 ps
are available in this setup. Both vibrational and electronic spectroscopy techniques are
being installed.
In this seminar I will introduce the features of the atto lab in more detail and discuss
computational techniques we are applying to interpret the time resolved spectra. At
present the IR and ARPES machineries are in place. Related to core level electronic
excitations I will discuss the status of GW for molecular systems.


Dr Bertrand Dupé