lithography roadmap - coocanaset.la.coocan.jp/event/ifst2005/s2-1_litho-road_imec_no2.pdf · ©...
TRANSCRIPT
© imec 2005
Lithography Roadmapwithout immersion lithography
32 nm45 nm65 nm90 nm130 nm180 nm 22 nm
248nm
193nm
157nm
EUVL
1999 20022001
20042003
20072005
20102007
20132009
3-year cycle:2-year cycle:
20162011
Node
45 nm65 nm90 nm130 nm180 nm250 nm 32 nmHalf pitch
© imec 2005
Lithography Roadmapwith immersion lithography
32 nm45 nm65 nm90 nm130 nm180 nm 22 nm
248nm
193nm
157nm
EUVL
1999 20022001
20042003
20072005
20102007
20132009
3-year cycle:2-year cycle:
20162011
Node
45 nm65 nm90 nm130 nm180 nm250 nm 32 nmHalf pitch
1.1-1.2NA
© imec 2005
Lithography Roadmapwith immersion lithography
32 nm45 nm65 nm90 nm130 nm180 nm 22 nm
248nm
193nm
157nm
EUVL
1999 20022001
20042003
20072005
20102007
20132009
3-year cycle:2-year cycle:
20162011
Node
45 nm65 nm90 nm130 nm180 nm250 nm 32 nmHalf pitch
1.1-1.2NA
~1.2NA
© imec 2005
157nm has no window
Max Resolution with High Index Fluids with sinθ=0.93
30
35
40
45
50
55
60
65
1 1.1 1.2 1.3 1.4 1.5 1.6Refractive Index
Res
olut
ion
@ k
1=0.
3 (n
m) 193nm
157nm
PFPE Fluid high-n
• ArF water immersion replaces F2 dry
Fluid x
needed
• ArF fluid high-n immersion replaces F2 -PFPE
Dry
Water
© imec 2005
SPIE 2005 Fluid highlights
SPIE, M arch 1, 2 005
Testing at 32nm with Dupont IF131 and IF132 High-n Immersion Fluids
(Initial Fluid Screening)
IF131 IF132Similar imaging performance from the two fluids, both very good!
32nm Lines (pitch 64nm)
© imec 2005
Lithography Roadmapwith immersion lithography
32 nm45 nm65 nm90 nm130 nm180 nm 22 nm
248nm
193nm
157nm
EUVL
1999 20022001
20042003
20072005
20102007
20132009
3-year cycle:2-year cycle:
20162011
Node
45 nm65 nm90 nm130 nm180 nm250 nm 32 nmHalf pitch
1.1-1.2NA
~1.2NA
1.3-1.5NA
© imec 2005
Lithography Roadmapwith immersion lithography
32 nm45 nm65 nm90 nm130 nm180 nm 22 nm
248nm
193nm
157nm
EUVL
1999 20022001
20042003
20072005
20102007
20132009
3-year cycle:2-year cycle:
20162011
Node
45 nm65 nm90 nm130 nm180 nm250 nm 32 nmHalf pitch
1.1-1.2NA
1.3-1.5NA
© imec 2005
k1 factor
90 65 45 32 22
0.75 0.35 0.25 0.17 0.12 0.09
0.85 0.40 0.29 0.20 0.14 0.10
0.94 0.44 0.32 0.22 0.16 0.111.05 0.49 0.35 0.24 0.17 0.121.2 0.56 0.40 0.28 0.20 0.14
1.35 0.63 0.45 0.31 0.22 0.15
1.5 0.70 0.51 0.35 0.25 0.171.65 0.77 0.56 0.38 0.27 0.19
1.8 0.84 0.61 0.42 0.30 0.21
Half Pitch Technology Node [nm]N
A [@
193
nm
]NA
kresolution λ.1=
© imec 2005
Lithography Roadmapwith immersion lithography
32 nm45 nm65 nm90 nm130 nm180 nm 22 nm
248nm
193nm
157nm
EUVL
1999 20022001
20042003
20072005
20102007
20132009
3-year cycle:2-year cycle:
20162011
Node
45 nm65 nm90 nm130 nm180 nm250 nm 32 nmHalf pitch
1.1-1.2NA
1.3-1.5NA
© imec 2005
Rayleigh equation
NAkresolution λ.1=
Lord Rayleigh
k1 scaling
Double exposure techniques
© imec 2005
Double dipole imaging
Random logic
DY
DX
© imec 2005
Impact of polarization on SRAM active layer print (110 nm pitch)
DY40 0.96/0.76
Unpolarized Polarized
E=25.8CD=51.5
E=24.6CD=56
E=23.4CD=58
E=22.2CD=60
E=21CD=59
23 % EL
DryNA=0.93
© imec 2005
Double patterning (2x litho + 2x etch)
SiNitride HM
Resist
Int.Exposure 1
& development
Transfer tohardmask
Int.Exposure 2
& development
Transfer tohardmask
Exp.
1
Exp.
2
Pitch = 100nmCD=50nmk1=0.19
100nm pitch
50 nm L&S with 0.75NA
© imec 2005
Double patterning complex pattern
1st exposuredevelop and etch
2nd exposuredevelop
+
etch
=
150-nm pitchNA = 0.75193 nm k1 = 0.29
© imec 2005
Double exposure technique
CoO considerations
Requires 2 critical masksReduces throughput (~ factor 2)Adds cost of hard mask and etchCritical overlay requirementImpacts cycle time (additional photo, etch, …)
© imec 2005
Outline
Introduction
193nm immersion lithography
EUV lithography
Global collaboration
Conclusions
Status, ChallengesEUV lithography
© imec 2005
EUV lithographyMain technology challenges
Multilayer coated optics
High powersource
Thin resistHighly sensitive
Low LER
MaskDefect Free
Multilayer coated
Pellicle free mask
CxHy, OopticsSource induced
contamination
Contaminationcontrol
© imec 2005
EUV tool integration progress
servo, vacuum, water,electronics cabinets
reticle stage
wafer stage
reticle handler
wafer handler
source-collectormodule baseframe
servic
e
corr id
or
0.25 NAfull field
© imec 2005
EUV tool integration progress
servo, vacuum, water,electronics cabinets
reticle stage
wafer stage
reticle handler
wafer handler
source-collectormodule baseframe
servic
e
corr id
or
© imec 2005
Multilayer coated optics
High powersource
Thin resistHighly sensitive
Low LER
MaskDefect Free
Multilayer coated
Contaminationcontrol
EUV lithographyMain technology challenges
© imec 2005
Multilayer coated optics
High powersource
Thin resistHighly sensitive
Low LER
MaskDefect Free
Multilayer coated
Contaminationcontrol
EUV lithographyMain technology challenges
EUV mask cross section(TEM courtesy of AMD)
Absorber
Buffer
Low thermal expansion substrate
Absorber
Buffer
Multilayer
Substrate
Absorber
Buffer
MultilayerM
ask
blan
kM
ask
mak
ing
no pellicle during exposure
© imec 2005
Multilayer coated optics
High powersource
Thin resistHighly sensitive
Low LER
MaskDefect Free
Multilayer coated
Contaminationcontrol
EUV lithographyMain technology challenges
SEMATECH MBDC EUV Blank Defect Progress
0.001
0.01
0.1
1
10
100
Sep-03 Dec-03 Mar-04 May-04 Aug-04 Nov-04
Defe
ct D
ensi
ty (d
/cm
^2 @
80
nm)
ML Adders + DecorationML AddersTotal DefectsPMsGoals
Q104 goal: 0.3 @ 90nmQ204 goal: 0.12 @ 80nm
Q404 goal: 0.08 @ 80nm
Significant progress in defect reduction
© imec 2005
Multilayer coated optics
High powersource
Thin resistHighly sensitive
Low LER
MaskDefect Free
Multilayer coated
Contaminationcontrol
EUV lithographyMain technology challenges
0.1
1.0
10.0
100.0
1000.0
Jul-0
1
Jan-
02
Jul-0
2
Jan-
03
Jul-0
3
Jan-
04
Jul-0
4
Jan-
05
Jul-0
5
Jan-
06
Date
EUV
Pow
er a
t Int
erm
edia
te F
ocus
[W] 115 W production requirement
Exponential fit to dataAverage of reported
Data from Sematech EUV Source Workshops
Fast progress on system productivity (source power)
© imec 2005
Multilayer coated optics
High powersource
Thin resistHighly sensitive
Low LER
MaskDefect Free
Multilayer coated
Contaminationcontrol
EUV lithographyMain technology challenges
Optics substrate
Multilayer coated optics
(Images courtesy of VNL / EUVLLC)
© imec 2005
Multilayer coated optics
High powersource
Thin resistHighly sensitive
Low LER
MaskDefect Free
Multilayer coated
Contaminationcontrol
EUV lithographyMain technology challenges
Edge variation over certain line length isLine Edge Roughness
increasing exposure dose
Better edge definition
Resist Sensitivity vs.Line Edge Roughness(trade off)
© imec 2005
Multilayer coated optics
High powersource
Thin resistHighly sensitive
Low LER
MaskDefect Free
Multilayer coated
Contaminationcontrol
EUV lithographyMain technology challenges
Resist Sensitivity vs.Line Edge Roughness(trade off)
SensitivityResolution
Line Edge Roughness
Shot noisestatistics
DiffusionLength
© imec 2005
HO O
O
O
OHO
O
O
t-Boc
t-Boc
t-Boc
t-Boc
t-Boc
t-Boc
New Resist Approaches
10.0mJ/cm2 Bright Field (Courtesy of Prof. Chris Ober, Cornell University)
Cornell University
Positive-tone Molecular Glass Resist
Shorter polymers may provide a possible solution
40 nm (2:1)40 nm (2:1)
60605050454540403535
2:12:1
© imec 2005
hp 32nm options
NA k1 λ
NAkresolution λ.1=
ArF Immersion1.8 NA(k1=0.3)
Single exposure
Lens size and cost
Liquid (nf>1.8)
New lens material (nf>1.8)
ArF Immersion1.35 – 1.5 NA
(k1=0.22)
Doubleexposure
Cost of mask (2x)
Throughput reduction
Cost of additional process steps (hard mask dep/etch)
Overlay requirement
EUVL0.25 NA(k1=0.6)
Single exposure
Defect free mask- Mask blank defect density- No pellicle during exposure
CoO- Source power- Optics lifetime- Resist sensitivity
Critical challenges:
© imec 2005
Outline
Introduction
193nm immersion lithography
EUV lithography
Global collaboration
Conclusions
Global collaboration
© imec 2005
Research challenges require…
World-wide Research Partnershipsbetween
ResearchOrganization
ICManufacturers
EquipmentSuppliers
© imec 2005
imec 193nm Immersion programThe world’s largest 193nm immersion lithography effort
LamRESEARCH
STMicr oelectronicsSTMicr oelectronics
True global partnership
© imec 2005
International EUV Initiative…
IEUVI International
EUVInitiative
IEUVI Chairman:Paolo Gargini :
Europe:• MEDEA+• LETI• IMEC
Japan:• ASET• EUVA• MIRAI
US:EUV LLCSEMATECHVNLSRC
IEUVI Technical Working Groups (TWG)
Mask TWGChair: Phil Seidel SEMATECHOrganizer: Shinij Okazaki EUVA
Optics TWGChair: Ginger Edwards SEMATECH / FreescaleCo-Chair: Yasuaki Fukuda EUVA / CanonOrganzier: Giang Dao SEMATECH / Intel
Source TWGChair: Koichi Toyoda EUVA Co-Chair: Stefan Wurm SEMATECH / InfineonOrganizer: Dieter Goltz MEDEA
Resist TWGChair: Kim Dean SEMATECH Co-Chair: Wolf-Dieter Domke InfineonOrganizer: Serge Tedesco CEA / LETI
…an International Consortium Network
e.g. Mask handling solutions
© imec 2005
Outline
Introduction
193nm immersion lithography
EUV lithography
Global collaboration
ConclusionsConclusions
© imec 2005
Conclusion
ArF immersion has eliminated 157nm lithography
Phenomenal progress has been obtained for ArF immersion lithography in less than 24 months
End 2003: ASML/IMEC demonstrated first data on immersion feasibility prototype scanner (0.75NA AT1150i)End 2004: Two development scanners (0.85NA) have been installed in the field
Key challenges for immersion have been identified; solutions have been proposed/demonstrated for all issues
In order to cope with the extremely short introduction time, world-wide partnerships and collaboration have proven to be very effective
ArF immersion is clearly usable down to at least hp 45nm
hp 32nm is unlikely to happen with immersion lithography and is believed to be the introduction node for EUVL
Significant progress has been made on the most important critical issues for EUVL (mask defects, source power, …)
© imec 2005
Acknowledgement
Many people have contributed to this presentation
IMEC lithography team (K. Ronse, G. Vandenberghe, …)
ASML (M. van den Brink)
ZeissSematech (S. Wurm)
© imec 2005
© imec 2005
Immersion Lithography
However, early 2004:
Feasibility study of IMMERSION LITHOGRAPHY successful
Several advantages (over 157nm) :
Build further on “mature” 193nm resistsMajority of optical material remains fused silica (not CaF2)Same reticle materials continue to be used
No thick Qz pellicles neededNo modified Qz blanks needed
© imec 2005
Resolution outlook
303234363840424446
1.5 1.6 1.7 1.8 1.9 2index [-]
half
pitc
h [u
m] k1=0.3
k1=0.27
38-nm with new fluid andNew lens
32-nm requires
very high index