1987
DOI: 10.1109/t-ed.1987.22965
|View full text |Cite
|
Sign up to set email alerts
|

Source—Drain contact resistance in CMOS with self-aligned TiSi2

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

1
8
0

Year Published

1989
1989
2006
2006

Publication Types

Select...
8
1

Relationship

0
9

Authors

Journals

citations
Cited by 56 publications
(10 citation statements)
references
References 10 publications
1
8
0
Order By: Relevance
“…SUPREM III predicted a surface concentration greater than 102~ cm-3; however, the silicon consumption during the titanium silicide formation ceuld modify the interfacial doping concentration of the n § region (10). Taur et al (8) reported a specific contact resistivity of 1.8 • 10 -6 ~-cm 2 for an interracial doping concentration of 3 • 10 ~9 cm -3, which is similar to the observed specific contact resistivity to n + diffusion in Table II. This suggests that the interfacial doping concentration of the n + diffusion is in the low 10 ~9 cm -~ range due to the segregation and/or redistribution of As ions in the titanium silicide layer during the silicide deposition (10).…”
Section: Resultssupporting
confidence: 52%
“…SUPREM III predicted a surface concentration greater than 102~ cm-3; however, the silicon consumption during the titanium silicide formation ceuld modify the interfacial doping concentration of the n § region (10). Taur et al (8) reported a specific contact resistivity of 1.8 • 10 -6 ~-cm 2 for an interracial doping concentration of 3 • 10 ~9 cm -3, which is similar to the observed specific contact resistivity to n + diffusion in Table II. This suggests that the interfacial doping concentration of the n + diffusion is in the low 10 ~9 cm -~ range due to the segregation and/or redistribution of As ions in the titanium silicide layer during the silicide deposition (10).…”
Section: Resultssupporting
confidence: 52%
“…13,14) In addition, the creep-up phenomenon during the formation of TiSi 2 may cause a bridge to form between the gate and the source/drain regions in the device structure, leading to device failure. 15) On the other hand, as the device feature size is scaled down to gate lengths below 40 nm and junction depths below 100 nm, the CoSi 2 process is faced with a difficult challenge on narrow polylines, 16) and high Si consumption during the formation of CoSi 2 silicide becomes a major drawback in forming shallow junctions. 17) Moreover, the formation of CoSi 2 -contacted shallow junctions is sensitive to oxygen in the processing environment, making it necessary to have a more complex silicidation process, such as one including the capping of passivation film during silicidation.…”
Section: Introductionmentioning
confidence: 99%
“…2 ' 3 In the conventional way of preparing the silicide the required for the silicon reaction is consumed from the substrate and the profile is shifted toward the interface as discussed by previous authors. 2 The latter alters not only the junction depth but also the surface concentration of the dopant. This is more important in heavily doped shallow junctions where the profile is flattened near the surface, while it falls abruptly with depth.…”
Section: Resultsmentioning
confidence: 99%
“…However, in the conventional way of forming the silicide, that is by deposition of a titanium layer and further annealing, silicon is consumed from the substrate and consequently, the junction dopant profile is altered. 2 Thus, the junction may fail when it is very shallow (0.1 pLm). Attempts to overcome this problem such as dopant drive-out schemes from the TiSi 2 have not produced satisfactory junction results.…”
Section: Introductionmentioning
confidence: 99%