Process Simulation

Typically an Integrated Circuit (IC) is fabricated by hundreds of sequential process steps, including lithography, etching, deposition, oxidation, ion implantation, diffusion, etc. TCAD (process and device simulation) provides deep insight into the device performance and information of how certain production processes influence the device behaviour, which is very important for IC production.

 

With emerging device architecture such as FinFETs, and requirement of complex channel profiles in bulk device to adjust the threshold voltage beyond the 50 nm gate length, conventional 1D or 2D TCAD simulations are not applicable. For reliable process simulation of FinFETs, state-of-the-art 3D simulation techniques are needed for critical process steps including ion implantation, diffusion, strain engineering etc. We are developing reliable, accurate and concise process simulation tools, most importantly for doping profiles and strain distribution within the device, forming the bases for accurate prediction of device performance and structure and process optimization.

 

n18psn2_net.pngfinfet_20nm_lp3cmp.png

 

 

Strain calculations

Process-induced strain has become a major performance booster for present and future generations of nano-CMOS devices. However, as critical dimensions reduce, strain becomes an important but difficult to control variable due to complex layout and imprecise lithography at nanometre scales. Compressive strain, typically inserted by SiGe epitaxial grown or carbon implanted source/drain regions enhances the hole transport and the performance of the p-channel transistors. While tensile strained Si3N4 contact etch stopper (stressor) layers are typically used to introduce tensile strain in n-channel transistors enhancing the electron transport and the device performance.

 

Modern semiconductor devices such as FinFETs are complex entities comprised of many different materials with different lattice constants and mechanical properties and it is necessary to take into account these differences on the nano-scale in order to accurately obtain the strain distribution, within the active channel region. Strain is calculated from Navier's strain-displacement equation for an inhomogenous material system:

equ1.jpg

In order to determine the effect of strain on transport the critical piece of information required is the strain tensor, this may be related to the ε to the vector displacement field u, by:

   equ2.jpg

Strain_Tensor_pic.jpg

People involved in the project

Liping Wang

Related Publications

L. Wang, A. R. Brown, M. Nedjalkov, C. Alexaner, B. Cheng, C. Millar and A. Asenov, "Impact of Self-Heating on the Statistical Variability in Bulk and SOI FinFETs," IEEE Transactions on Electron Devices, Vol. 62, No. 7, pp. 2106–2112, July 2015.

L. Wang, A. R. Brown, M. Nedjalkov, C. Alexander, B. Cheng, C. Millar and A. Asenov, "3D Electro-Thermal Simulations of Bulk FinFETs with Statistical Variations," inProc. 20th International Conference on Simulation of Semiconductor Processes and Devices (SISPAD), Washington DC, Sept. 9-11, 2015, pp. 112–115.

L. Wang, T. Sadi, M. Nedjalkov, A. R. Brown, C. Alexander, B. Cheng, C. Millar and A. Asenov, "An Advanced Electro-Thermal Simulation Methodology For Nanoscale Device," in Proc. 18th International Workshop on Computational Electronics, West Lafeyette, USA, Sept. 2-4, 2015, pp. 1–4.

L. Wang, A. Brown, B. Cheng and A. Asenov, "Simulation of 3D Doping Profile by Ion Implantations in FinFET," 19th International Conference on Ion Implantation Technology (IIT2012), Valladolid, Spain, June 25-29, 2012 [accepted]

See the Publications page for more.