Process simulation of dopant diffusion and activation in germanium

Abstract

Germanium with its high carrier mobility is currently being investigated as an alternative material to silicon for advanced MOS devices. We have reviewed the literature on n‐type and p‐type doping of germanium and established a baseline calibration for technology process simulation. Fundamental parameters for germanium point defects have been selected and extended defect evolution has been calibrated. Models and parameters for accurate simulation of ion implantation, diffusion, and activation of the most common dopants (phosphorus and arsenic for n‐type, boron for p‐type) have been defined. We discuss the current accuracy and limitations of the process calibration and specify missing experimental data needed for enhancing and broaden the simulation capabilities for germanium processing.

Stability, bonding, and electronic properties of silicon and germanium arsenides

“…1 Introduction Arsenic has been used as a doping agent in silicon and germanium to produce n-type semiconductors for several decades, and many studies have been done on this subject. Even quite recently, the doping issues of As impurities in silicon and germanium are still research topics indicating the importance of the Si-As and Ge-As systems [1][2][3][4][5][6][7]. The phase diagrams of the silicon-arsenic and germanium-arsenic systems showed two types of compounds, i.e., XAs and XAs 2 (X ¼ Si, Ge) [8,9].…”

Defect engineering for shallow n‐type junctions in germanium: Facts and fiction

“…It has been demonstrated that, both for P ion implantation and for in situ doped molecular beam epitaxy (MBE) multilayers of P, As, or Sb, the presence of C retards the diffusion of the dopants in germanium. DFT calculations have shown that the interaction between C and DV pairs leads to energetically favorable and relatively immobile complexes , whereby the association between V, Sb, and C leads to more stable complexes than with As or P. Recently, a simple C‐DV cluster model has been implemented in a commercial technology computer aided design (TCAD) package (Sentaurus), yielding satisfactory agreement between measured secondary ion mass spectrometry (SIMS) and simulated profiles . The drawback of this is that these complexes are electrically neutral or even compensating acceptors, so that the sheet resistance of the resulting shallower n + junctions may be compromised.…”

Processing of germanium for integrated circuits

“…Zographos and Erlebach reviewed process simulation [115] for Ge focussing on point defects and extended defects; B, P, and As diffusion, activation, and dose loss; and C co-implantation and co-diffusion. From their perspective at Synopsys, more efforts are needed to improve the existing models and to develop currently missing models for N and F co-implantation, strain effects, and alloys such as SiGe with high Ge fraction and GeSn.…”