4/17/2002 1 Fundamental Beam Studies of Radical Enhanced Atomic Layer Deposition (RE-ALD) SFR Workshop & Review April 17, 2002 Frank Greer, David Fraser, John Coburn, David Graves Berkeley, CA 2002 GOALS: 1). to investigate and model the adsorption of TiCl 4 on industrially relevant surfaces. 2). to investigate fundamental surface reactions mechanisms during N/H radical exposures.
12
Embed
Fundamental Beam Studies of Radical Enhanced Atomic …
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
4/17/2002
1
Fundamental Beam Studies of Radical Enhanced Atomic Layer Deposition
(RE-ALD)SFR Workshop & Review
April 17, 2002
Frank Greer, David Fraser, John Coburn, David GravesBerkeley, CA
2002 GOALS: 1). to investigate and model the adsorption of TiCl4on industrially relevant surfaces. 2). to investigate fundamental surface reactions mechanisms during N/H radical exposures.
4/17/2002
2
RE-ALD for TiN Diffusion Barriers• Future generation microelectronic devices require highly conformal, ultra-
thin TiN Diffusion Barriers • Atomic layer deposition (ALD) and Radical Enhanced ALD
– Uniform deposition over large area– Uniform deposition for arbitrary topography– Precise growth rate control
• Many different approaches and films reported1,2,3,4
– TiCl4 + NH3à TiN + HCl – TiCl4 + N + H à TiN + HCl– TaCl5 + H à Ta + HCl– Good resistivity reported– With Plasmas/Radicals, low impurity content at low substrate temperatures
• Can achieve Cl <0.3%, Resistivity ~1000 µΩ.cm• High Cl% may lead to corrosion of metallic films
1. Satta et al., MRS 2000 Spring Mtg. (D6.5)2. Ritala et al., JES 147, 2000
3. Greer et al., J.Vac.Sci.Tech. B. submitted4. Kim et al., MRS 2002 Spring Mtg. (B8.5)
4/17/2002
3
Adsorption Issues in RE-ALD of TiN• RE-ALD1 uses a volatile metallic precursor and a radical source to deposit a film
• TiCl4/H/N Reactants introduced separately to achieve self-limiting growth
– Purpose is to grow film 1 monolayer at a time
• Film thickness controlled by repeating TiCl4/H/N cycles
• Nucleation regime observed by many authors
• Growth Rate typically less than one monolayer/cycle even after nucleation regime
• Most studies done at higher temperatures > 200oC due kinetic constraints
1A. Sherman U.S. Patent 1999.
TiCl4Plasma (N/H)Ar
Time
Rea
ct. C
onc.
Film Thickness
Cycle Number
NucleationRegime ?
4/17/2002
4
Surface Reaction Products
Atom Source
Schematic of the Beam Apparatus in Cross-section (Top View)Quadrupole Mass
Spectrometer (QMS)
Main Chamber
Analysis SectionRotatable Carousel
+Experimental Diagnostics1. QCM
- Measures mass change of film
2. QMS- Measures products
formed on film- Characterize beams
Si/Cu/W/Porous SiO2(low-k)-coatedQuartz Crystal
Microbalance (QCM)
Ion Source
Tuning forkchopper
X
TiCl4
Precursor Doser
X = N or D
Load Lock
In-vacuoAuger
Analysis
+
Atom Source
NH3
4/17/2002
5
TiCl4 Precursor Adsorption
• Two distinct ads. regimes (Initial rapid followed by significantly slower)• Desorption of precursor à deposition rate/cycle lower at higher temperatures• Adsorption appears more uniform at higher temperatures • Adsorption likely submonolayer in all cases due to steric hinderance
*TiClx
TiCl4
Adsorption
*
TiCl4
0 2000 4000 6000 8000 10000
0.00E+000
2.00E+014
4.00E+014
6.00E+014
8.00E+014
1.00E+015
# of
TiC
l 2 Mol
ecul
es A
dsor
bed/
cm2
TiCl4 Exposure (1015 molecules/cm2)
Temp300K Temp350K Temp410K Temp445K
Submonolayer Adsorption
TiClx
Si QCM
4/17/2002
6
BET Predictions for Multilayer Adsorption• BET Isotherm is an equilibrium
calculation• Assumes surface interactions of
multilayers similar to liquid phase • Adsorption inconsistent with