Semiconductor process and device modeling

Semiconductor process and device modeling

Lado Filipovic

I develop physical models that connect atomistic mechanisms and reactor conditions to the shape and electrical behavior of semiconductor devices, and I share them as open-source software.

I am an Associate Professor at the Institute for Microelectronics, TU Wien, and director of the Christian Doppler Laboratory for Multi-Scale Process Modeling of Semiconductor Devices and Sensors. My group works closely with industrial and academic partners, and I welcome new collaborations and consulting requests.

Portrait of Lado Filipovic

Research

My research follows semiconductor devices from their fabrication to their operation and reliability.

Map of simulation methods by length and time scale, from DFT, NEGF and molecular dynamics through kinetic Monte Carlo, Monte Carlo and drift-diffusion to compact models, circuits, process, die, wafer and reactor
Length and time scales of process (red) and device (blue) simulation, and the quantities passed between the models, from DFT and molecular dynamics to the reactor and the circuit.

Multi-scale process TCAD

We simulate etching, deposition, implantation and annealing for logic, memory and power devices, and feed the feature-scale models with reactor-scale plasma conditions and atomistic surface data.

Hot carriers and non-equilibrium transport

Ensemble Monte Carlo simulations of carriers and phonons far from equilibrium show how hot carriers cool and how their energy can be extracted in solar cell absorbers and two-dimensional materials.

Two-dimensional semiconductor sensors

We study how gas adsorption, dopants and charge trapping shape the response of MoS2 transistors, combining ab initio calculations, device simulation and our own measurements.

Interconnect reliability

We model electromigration, thermo-migration and stress migration in scaled copper interconnects and in platinum microheaters for gas sensors.

Open-source software

My group develops ViennaPS, an open-source simulator for etching and deposition, together with the ViennaTools libraries it builds on for level sets and ray tracing. For carrier transport we develop the ensemble Monte Carlo code ViennaEMC.

ViennaPS runs on CPUs and GPUs, can be scripted in Python, and installs with pip install ViennaPS.

Diagram of ViennaPS and the ViennaTools libraries ViennaLS, ViennaCS, ViennaRay, ViennaHRLE and ViennaCore, with the external libraries VTK, Embree and OptiX

Consulting and collaboration

Companies and research groups working on semiconductor processes and devices are welcome to contact me with their questions. Typical questions concern etch and deposition profiles, the coupling of reactor and feature-scale models, the calibration of simulations against measurements, and the physics of new materials and devices. Some are settled in a single discussion, and others grow into joint research projects at TU Wien.

Selected recent work

ViennaPS: A Flexible Framework for Semiconductor Process Simulation
T. Reiter and L. Filipovic, SoftwareX 32, 102453 (2025)

Non-equilibrium phonon effects on high-field electron transport in β-Ga2O3
L. Filipovic, Journal of Applied Physics 140, 135701 (2026) – Editor’s Pick

The influence of implantation conditions on dopant activation in Al-implanted 4H-SiC: a MD study applying an Al potential fitted to DFT barriers
S. Leroch, R. Stella, A. Hössinger and L. Filipovic, Journal of Materials Chemistry C 14(34), 15620–15635 (2026)

From Atoms to Reactors: Multi-Scale Modeling for Semiconductor Fabrication
L. Filipovic, International Conference on Simulation of Semiconductor Processes and Devices (SISPAD), pp. 1–4 (2025)

Full publication list