${\rm CO}_{2}$ Laser Annealing for USJ Formation in Silicon: Comparison of Simulation and Experiment

In this paper, we compare the experimental and simulation results of the effect of laser annealing on boron implanted silicon using a CO2 source and commercially available process TCAD tools. The main objective of this analysis is the prediction of the evolution of temperature distribution induced into the wafer during the laser irradiation below the melting threshold as well as its effect on boron diffusion and activation kinetics. A series of factors are considered along with the use of advanced heat transfer and dopant diffusion models provided by the TCAD tool to accurately prototype the effect of the CO2 irradiation on temperature and dopant distribution. These include the nonuniformity of the incident laser beam, the strong dependence of heat capacity and thermal conductivity from temperature and, most importantly, the dependency of the absorptivity from temperature and dopant distribution, which requires the solving of dopant and heat transfer equations in a coupled and self-consistent way. Using surface temperature data obtained by pyrometry measurements, it was possible to calibrate and to verify the validity of the results. Experimental boron profiles were then used to compare with simulations in the transient regime where dopant diffusion just starts to occur. Both TCAD and experimental data confirmed previous suggestions that submelt CO2 laser annealing is an efficient tool for profile engineering, allowing diffusionless activation of the plasma doped boron profiles.

[1]  C. Tavernier,et al.  BF{sub 3} PIII modeling: Implantation, amorphisation and diffusion , 2012 .

[2]  S. Aid,et al.  Comparison of boron diffusion in silicon during shallow p+/n junction formation by non‐melt excimer and green laser annealing , 2011 .

[3]  Elena Bruno,et al.  Kinetics of large B clusters in crystalline and preamorphized silicon , 2011 .

[4]  F. Torregrosa,et al.  PULSION® HP: Tunable, High Productivity Plasma Doping , 2011 .

[5]  D. Giubertoni,et al.  Non-melting annealing of silicon by CO2 laser , 2010 .

[6]  C. Grigoropoulos,et al.  Formation of silicon Ultra Shallow Junction by nonmelt excimer laser treatment , 2009, 2009 Proceedings of the European Solid State Device Research Conference.

[7]  Y. Sasaki,et al.  Aiming for the Best Matching Between Ultra-Shallow Doping and Milli-To Femto-Second Activation , 2007, 2007 15th International Conference on Advanced Thermal Processing of Semiconductors.

[8]  N. Cowern,et al.  Deactivation of ultrashallow boron implants in preamorphized silicon after nonmelt laser annealing with multiple scans , 2006 .

[9]  M. Doubenskaia,et al.  Pyrometry in laser surface treatment , 2006 .

[10]  Pooi See Lee,et al.  Dopant activation in subamorphized silicon upon laser annealing , 2006 .

[11]  Y. Wang,et al.  Minimizing Pattern Dependency in Millisecond Annealing , 2006, 2006 International Workshop on Junction Technology.

[12]  N. Cherkashin,et al.  Clusters formation in ultralow-energy high-dose boron-implanted silicon , 2003 .

[13]  A. Claverie,et al.  Prediction of boron transient enhanced diffusion through the atom-by-atom modeling of extended defects , 2003 .

[14]  Dieter Bäuerle,et al.  Laser processing and chemistry: recent developments , 2002 .

[15]  W. Szyszko Melting and diffusion under nanosecond laser pulse , 1995 .

[16]  O. W. Holland,et al.  Pulsed excimer and CO2 laser annealing of ion-implanted silicon , 1985 .

[17]  E. Rauhala,et al.  Substrate heating effects in CO2 laser annealing of ion‐implanted silicon , 1983 .