DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference Metrology Roadmap Metrology Roadmap 4 - 05 4 - 05 Europe Europe Rien Stoup (PAN Analytical) Rien Stoup (PAN Analytical) Dick Verkley (FEI- Philips) Dick Verkley (FEI- Philips) Mauro Vasconi (ST) Mauro Vasconi (ST) Japan Japan Atsuko Yamaguchi (Hitachi) Atsuko Yamaguchi (Hitachi) Korea Korea Chin Soobok (Samsung) Chin Soobok (Samsung) Mann-Ho Cho (KRISS) Mann-Ho Cho (KRISS) Taiwan Taiwan J.H. Sheih (TSMC) J.H. Sheih (TSMC) US US Jack Martinez (NIST) Jack Martinez (NIST) John Allgair (Freescale/SEMATECH) John Allgair (Freescale/SEMATECH) Steve Knight (NIST)* Steve Knight (NIST)* Alain Diebold (Int. SEMATECH)* Alain Diebold (Int. SEMATECH)*
28
Embed
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference Metrology Roadmap 4 - 05 EuropeRien Stoup (PAN Analytical) Dick.
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
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
KoreaKorea Chin Soobok (Samsung)Chin Soobok (Samsung)Mann-Ho Cho (KRISS)Mann-Ho Cho (KRISS)
Taiwan Taiwan J.H. Sheih (TSMC)J.H. Sheih (TSMC)
USUS Jack Martinez (NIST)Jack Martinez (NIST)John Allgair (Freescale/SEMATECH)John Allgair (Freescale/SEMATECH)Steve Knight (NIST)*Steve Knight (NIST)*Alain Diebold (Int. SEMATECH)*Alain Diebold (Int. SEMATECH)*
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
AGENDAAGENDA
• 2005 ITRS Changes and Activities
• Lithography Metrology
• FEP Metrology
• Interconnect Metrology
• Key Challenges
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
Transistor EvolutionTransistor Evolution
Today90 nm Node
Strain Enhanced Mobility
Tomorrow
New Materials
CMOSpMOS FINFET
Future15 years
Non-classical CMOS
Beyond CMOS
source
Gate
drain
source
Gate
drain
Molecular Switches ? Nanowire Transistor ?
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
Nanotransistors – The FutureNanotransistors – The Future
Nano TransistorsIdsat ~ W Cox (VG – VT) sat
Long Channel Behavior
Idsat ( 1/Lg ) ( Carrier Mobility ) ( 1/EOT ) ?
= Cload VDD / IdsatC dependenceA = Lg x WDopant Conc.
Short
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
New BusinessNew Business
• Established link with Emerging Research Materials group of ERD– ERM to report requirements– Metrology Reports Methods and Discussion– Provided update on aberration corrected TEM
and 3D TEM developments
• Awaiting changes in Process Tolerance for Gate CD
Void Size for 1% Voiding in Cu Lines 8.5 6.7 4.8 3.4 2.4 1.9 MPU
Detection of Killer Pores at (nm) size 8.5 6.7 4.8 3.4 2.4 1.9 MPU
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
Changes to Litho CD MetrologyChanges to Litho CD Metrology
45 nm contact Holes16 nm Lines -176 nm Pitch
ITRS July Presentation by Dan Wack – KLA-TencorScatterometry to at least the 45 nm NodeBryan Rice (Intel), SPIE, 2004CD-SEM and Scatterometry can reach 32 nm Node w/improvement – impact of SOI not tested
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
0
2
4
6
8
10
12
14
16
10 100 1000 10000
Experiments
Simulation
2m
LWR increase becomes negligible at L=2m
Experiments & Simulation Confirmed by ...
3 (
nm)
Inspection-area height L (nm)
CD variation
Average LWR
Japan ITRS TWG Input LWR as a Function of L Japan ITRS TWG Input LWR as a Function of L L above 2 L above 2 m is really CD variation
L value works for many resists as well as ADI, resist trim, and after etchL value works for many resists as well as ADI, resist trim, and after etch
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
0
1
2
3
4
5
0 100 200 300 400 500
Wire width(nm)
Res
isti
vity
(μΩ
cm)
p=0(complete diffuse scattering)
p=0.5
Measured Cu resistivitywithout Barrier Metal
ρ(Al):2.74μΩcm
p=0.3
Updated(May2004)
Trend: Sidewall Control will become more Trend: Sidewall Control will become more ImportantImportantLine Edge Roughness impacts Interconnect Resistance and Line Width Line Edge Roughness impacts Interconnect Resistance and Line Width Roughness impacts Transistor Leakage CurrentRoughness impacts Transistor Leakage Current
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
• Measure LER and LWR = for reliability all spatial wavelengths between 1 micron to 20 nm taken over a 2 micron length with a spacing of 10 nm.
• LER: For monitoring Interconnect specular scattering – measure from 34 nm to twice the smallest interval using the smallest possible interval. Measure sidewall after barrier metal dep.
LER and LWRLER and LWRreported as RMS values integrated over the frequency range reported as RMS values integrated over the frequency range specified belowspecified below
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
Overlay MetrologyOverlay Metrology
• Issues are with the Overlay TargetsIssues are with the Overlay Targets• Need better correlation between overlay target Need better correlation between overlay target
and device performanceand device performance– Current roadmap suggests that overlay is in good shape Current roadmap suggests that overlay is in good shape
because precision for box – in – box targets meets ITRS because precision for box – in – box targets meets ITRS guidelinesguidelines
– Step and Scan Litho Exposure tools contribute errors Step and Scan Litho Exposure tools contribute errors not measured by scribe line targetsnot measured by scribe line targets
– Across die overlay issues are not captured by box –in – Across die overlay issues are not captured by box –in – box target data analysisbox target data analysis
• 2005 rewrite will capture these issues2005 rewrite will capture these issues
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
FEP Metrology:FEP Metrology:Increasing Emphasis on Areas beyond High kIncreasing Emphasis on Areas beyond High k
• Increase in Mobility by using local stressing of transistors –
Call for stress metrology by FEP
• Metal Gates increasingly important
• New transistor designs are already past R and into D.
Example: FIN-FETs require metrology
• Strained SOI, GeOI considered longer term substrates
• Crystal Defect mapping including on patterned wafers
metrology requirement for new substrates
• Mobility measurements are key
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
Stress Measurement from Stress Measurement from transistor to wafertransistor to wafer
Wafer Bow
Die Level Stress
Transistor Level Stress
Micro-Area Level Stress
Nano-Raman and CBED
Micro-Raman, XRD
Die level flatness Laser InterferometryNew Interferometry Method
Laser Interferometry, New Interferometry
Trend : Use Modeling to connect what you can Trend : Use Modeling to connect what you can measure with what you need to know measure with what you need to know Example: Metrology of Strained Channel DevicesExample: Metrology of Strained Channel Devices
MD Giles, et. al., VLSI Symposium 2004
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
• Porous low k is being overshadowed by the need to lower k value of barrier layer and etch stops for 45/32 nm node.
• Long term solution for interconnect is ambiguous.
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
New Comments on New Comments on Interconnect Metrology Interconnect Metrology
• 3D CD especially Sidewall Roughness for trench/via
are critical needs
• In Process Development, LER is an important figure of merit for sidewall and AFM can give sidewall/via roughness vs depth as well as trench CD and profiles which provide important control for etch processes
• Need metrology for thickness and uniformity for hard mask above low after CMP – how is erosion pattern dependent
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference
ERD and ERMERD and ERMPremise: Aberration corrected TEM/STEM Premise: Aberration corrected TEM/STEM needed for nano-electronicsneeded for nano-electronics
• Latest TEM and STEM results show improved imaging for sub 0.1 nm beams.
• We have not reached the limit where improving beam diameter does not improve imaging
• Nanowires and nanotubes serve as ideal systems to test theoretical and experimental understanding of beam propagation and image formation. - reduced stresses and amorphous films on the surface
DRAFT – Work In Progress - NOT FOR PUBLICATION 12-13 April 2005 – ITRS Public Conference