Unit:
Κατεύθυνση Τεχνολογίας Ολοκληρωμένων ΚυκλωμάτωνΠληροφορική
Author:
Terzopoulos Trifon
Supervisors info:
Γεώργιος Κόκκορης, Συνεργαζόμενος Ερευνητής, Ινστιτούτο Νανοεπιστήμης και Νανοτεχνολογίας, ΕΚΕΦΕ Δημόκριτος
Original Title:
Προσομοίωση διεργασίας ιονοβολής επιφανειών με δέσμη ιόντων υπό ταυτόχρονη απόθεση: Διερεύνηση των μηχανισμών δημιουργίας οργανωμένων μορφολογιών
Translated title:
Simulation of ion beam sputtering of surfaces under simultaneous deposition of etch inhibitors: Investigation of the mechanisms producing organized nanopatterns
Summary:
Organized (periodic) nanopatterns on surfaces has been found very useful in various
applications such as magnetic storage and chemical reaction catalysis. Ion beam
sputtering (IBS) of surfaces has been used for the formation of organized nanopatterns
on the surfaces. Within the last decade, IBS has been used with co-deposition of
metals, also called etch-inhibitors, to produce anisotropic organized nanopatterns.
This thesis is engaged with the mathematical modeling and simulation of simultaneous
IBS and deposition processes on initially flat surfaces. The aim is to understand the
formation of anisotropic organized nanopatterns produced by Argon (Ar) ion beam
sputtering (with ion energy equal to 1 keV) on Silicon (Si) surfaces under the
simultaneous deposition of Iron (Fe) atoms. Simultaneous deposition is achieved by
placing of Fe targets around the Si substrate. The targets are placed on specific
locations in order to control the organization of the nanopatterns being formed on the Si
surface.
A pre-existing C++ Monte Carlo (MC) code for surface evolution was modified and
further developed, in order to achieve the aim. A cellular representation of the substrate
is utilized in code. The direction distributions of ions and Fe atoms arriving on the
bombarded surface were calculated by the commercial code COMSOL. SRIM (Stopping
and Range of Ions in Matter) code was used to calculate the sticking probabilities of
ions and Fe atoms and the etching yields of Si and Fe (atoms removed pre incident ion)
as a function of angle of ion incidence. The latter functions were integrated in the C++
MC code and the sampling was implemented with the acceptance - rejection method.
The morphology of the simulated surfaces was characterized using the Root Mean
Square (RMS) roughness, the correlation length, the surface coverage fraction by Fe
atoms and the 2D Power Spectrum Density (PSD). The quantification of the
organization of the nanopatterns was done by extracting the period and the parameter
of organization from the 2D PSD.
The results of the simulations showed that: a) The statistical parameters characterizing
the morphology were independent of the surface dimensions and were dependent on
the dimensions of the unit cell of the surface. b) The use of the functions for the sticking
probabilities and the etching yields coming from SRIM code did not produce organized
nanopatterns on the surface. c) Further investigation on the values of the parameters
and on the mechanisms of particle-surface interaction showed that organization was
favored when the sticking probability of ions was equal to 1 (meaning that there is no
ion reflection), when the fraction of Fe atoms on the flux arriving on the surface was
increased, and when the etching selectivity of Fe atoms over the Si atoms approached
1.
Main subject category:
Technology - Computer science
Keywords:
roughness, ion bean sputtering, plasma etching, simulation, Monte Carlo, organized nanopatterns
