© imec 2003 1
IMEC – ASMLIMEC – ASMLProcess Technology PlatformProcess Technology Platform157nm Lithography Status Update157nm Lithography Status Update
Kurt RonseDirector Lithography Department
IMEC, Leuven (Belgium)
Veldhoven, 13 November 2003
© imec 2003 2
Outline
Introduction
157nm lithography status update
IMEC 157nm program
Summary and conclusions
Acknowledgements
© imec 2003 3
Outline
Introduction� IMEC-ASML process technology platform
157nm lithography status update
IMEC 157nm program
Summary and conclusions
Acknowledgements
© imec 2003 4
IMEC
����Established by state government
(Flanders)Non-profit organization
Initial budget: 62MInitial Staff: ~70
����Largest independent R&D center in EuropeBudget: >145MStaff: >1270Collaboration with >450 partners
© imec 2003 5
Organization
Strategic partnershipsR. De Keersmaecker
Technical Support& Computer Logistics
W. Fluit
FinanceA. Vinck
Design Technology forIntegrated Information &Communication Systems
DESICSR. Lauwereins
Microsystems,Components &
Packaging
MCPR. Mertens
Silicon Process & DeviceTechnology
SPDTL. Van den hove
Industrialization &Training in
Microelectronics
INVOMECH. Maes
Scientific AdvisorH. De Man
President and CEOG. Declerck
Strategic Development UnitJ. Van helleputte
Business DevelopmentL. Deferm
Human ResourcesE. Daenen
© imec 2003 6
(Sub-)45nm Si Technology Research
���
to perform focused research on
�Advanced materials, process steps and modules�Novel device concepts (‘beyond classical CMOS’)
For technologies at least two generations ahead of manufacturing (> “N+2”)
© imec 2003 7
Si Technology Programs (IIAP)Concept
Advanced R&D on Si Technologyis carried out in
Industrial Affiliation Programs (IIAP)
���������������������
Equipment supplier
Material supplier Key IC manufacturers
IMEC focus on process developmenton advanced equipmentwith advanced materials
© imec 2003 8
Today
IMEC-ASML process technology platform
Evolution of IMEC-ASML cooperation� 1990 : first ASML 248nm stepper installed at IMEC (PAS5000/70)
Industrial affiliation program on 28nm lithographyProcess training center for ASML application engineers
� 1999 : first ASML 193nm scanner installed at IMEC (PAS5500/900)Industrial affiliation program on 193nm lithography
� 2001 : ASML imaging team @ IMEC established (8 FTE)Joint projects on
� Optical extension and imaging development(e.g. double dipole, chromeless phase shift)
� (off-line and integrated) metrology development (e.g. electrical CD,…)� Long term exposure tool monitoring� …
� 2003 : first ASML 157nm scanner installed at IMEC (Micrascan VII)Industrial affiliation program on 157nm lithography
� 2004-2007 : plans under discussion…
© imec 2003 9
157nm critical challenges :ISMT stop light chart
���������� �������� ��� ������
Jan
-99
Mar
-99
Jun
-99
No
v-00
Dec
-00
Feb
-01
Jun
-01
Dec
-01
Jun
-02
Sep
-02
Au
g-0
3
Issues/Comments
Mask - Reticle blanks More work on lifetime of attenuators PSM absorbers - Pellicle Workable Hard Pellicle Solution is feasible. Soft Pellicle
lifetime is a major concern. Good progress on hard pellicles
Exposure Tools - Materials CaF2 Yield/quality enhancements continue to show
improvement. Both <111> & <100> crystals are needed. Champion <111> in spec and <100> close to spec. Add 45nm focus and reason for yellow.
Coatings Lifetime concerns versus environmental issues. - Lasers COO a concern due to material damage; linewidth
narrowing needed for 45nm. - Purging/Ctmn Substantial progress to dateResist - Chemistry Asahi resist achieved ~ 1 / � m transparency - Performance Timing is an issue. Concern over PEB Sensitivity,
Poisoning, and Transparancy. LER for 45nm node.Metrology/Inspection - Reticle Inspection through Hard Pellicle
Hard Pellicle provides significant inspection challenge & availability of tool still an issue; not field changeable.
Timing - 65nm half pitch - 2007 Mfg. Start
Legend Requires Invention/Potential Showstopper
Critical Issue/Development Requiredin 2003, node changed to 45nm Solution Known
SubstratePellicle
CaF2 (quality and quantity)Optical material contamination and lifetimeLight source (F2 laser)N2 Purging
Resist transparency and profilesProcess integration aspects
Masks
ExposureTool
Resist
© imec 2003 10
Outline
Introduction
Status of 157nm critical issues� 157nm resist status� 157nm exposure tool status� 157nm pellicle status� 157nm mask substrate status
IMEC 157nm program
Summary and conclusions
Acknowledgements
© imec 2003 11
157nm resists2000-2001 : Absorbance challenge
Development of new chemistries� Siloxane, silsesquioxanes� Fluoropolymers
W. Conley et al, 3th International Symposium on 157nm lithography
Polymer absorbance : from 6 to below 1/µm
Abs
orba
nce
µm-1
Time
4
6
3
5
2
1
0.7 perfluorocarbon*
UV6
I
C
O
O
C(CH3)3F3CCF3
OH
n m
I
C
O
O
C(CH3)3F3CCF3
OH
n m
II
CF3
F3CO CH2 O CH2 CH3
F3CCF3
OH
n m
II
CF3
F3CO CH2 O CH2 CH3
F3CCF3
OH
n m
III
YX
CF3
F3C OHF3CCF3
O
O
III
YX
CF3
F3C OHF3CCF3
O
O
YX
CF3
F3C OHF3CCF3
O
O
n CF2
COO
n CF2 n
CF3
CF2CF2
CF3CF3
OH
nn CF2
COO
n CF2 n
CF3
CF2CF2
CF3CF3
OH
nn CF2
COO
n CF2 n
CF3
CF2CF2
CF3CF3
OH
n
CF2 CF2 n
CF3CF3
OH
nCF2 CF2 n
CF3CF3
OH
nCF2 CF2 n
CF3CF3
OH
n
RS-001
CH2CF2
F
FOH
CF3
HH
F
n
RS-001
CH2CF2
F
FOH
CF3
HH
F
n
OH
( )
OH
( )
OH
( )
OH
CF3F3C
( )
IBM/MITOH
CF3F3C
( )
OH
CF3F3C
( )
OH
CF3F3C
( )
IBM/MIT
248nm chemistry : UTR
© imec 2003 12
157nm resists2001-2002: resist transparancy allows thickness increase
Absorbance
012345678
Jan-01
Apr-01
Jul-01
Nov-01
Feb-02
May-02
Sep-02
Dec-02
Mar-03
Jun-03
Date
Ab
sorb
ance
[1/
um
]
110nm thick
barc
67 nm thick
silicon
110nm thick
silicon
200nm thick
barc
© imec 2003 13
157nm resistsStatus today
State of the art resist Absorbance < 1.0 /µm
< 60 nm resolution in 150nm thickness (NA=0.85, PSM)
F. Houlihan et al, 4th International Symposium on 157nm lithographyExposure : Exitech NA=0.85, Selete
100nm 90nm 80nm
70nm 65nm 60nm
© imec 2003 14
157nm resistsStatus today
Courtesy of Selete and Nissan Chemicals, 4th International Symposium on 157nm lithographyExposure : Exitech NA=0.85, Selete
© imec 2003 15
157nm resistsSummary
Significant resist progress� Absorbance of resist has decreased to below 1/um, allowing
imaging in resist thicknesses > 150nm
� Today’s best resists are capable of 60nm1:1 resolution(at 0.85NA, PSM) in 150nm thick resist
Further improvements needed� Line edge roughness is a challenge
� Etch resistance should be further improved
� Post-exposure delay sensitivity (amine contamination)
� Resist sensitivity
© imec 2003 16
157nm exposure toolCaF2 improvements
0
0,5
1
1,5
2
SD
B r
ms
[nm
/cm
]
157 nm193 nm (111)
materialfor /900
materialfor /1100
status 2001
material 1600
(100)(100)(111)
(111)
1998 2000 2001 2002 2003
system 4system 1-3
S CHOT Tlithotech
© imec 2003 17
157nm exposure toolFirst full field scanner successfully installed at IMEC
ASML MicraScanVII step&scanPlatform specs:Lens reduction ratio: 4XLens NA: variable 0.4 to 0.75Field size: 34 mm x 20 mmWafer size: 200 mmReticle size: 6” x 6” x 0.25”Partial coherence range: 0.3 to 0.85Reticle cleaning unit off-line
Interfaced to TEL Clean Track ACT8
ESI TMB-150 Total Amine Monitor
© imec 2003 18
130nm
100nm 90nm
80nm
S. Dana et al, 4th International Symposium on 157nm lithographyExposure : ASML MSVII NA=0.75, BIM
157nm exposure toolFirst full field scanner successfully installed at IMEC
Binary mask, 0.75 NAConventional illumination
© imec 2003 19
157nm exposure toolFirst full field scanner successfully installed at IMEC
70nm 1:1 L/S
F= + 0.1 µm
F= - 0.1 µm
Best focus
150 nm resist thickness
60nm 1:1.25 L/S
Binary mask, 0.75 NA
Dipole illumination
© imec 2003 20
157nm exposure toolSummary
Significant progress� First full field exposure tool installed in the field and
showing acceptable performance
� Intrinsic birefringence compensation successful(CaF2 <M111> and <100> crystal orientations)
Further learning needed� Long term optical path stability
(hydrocarbon and inorganic contamination risks mitigated ?)
© imec 2003 21
157nm pelliclethe issue
First option : Soft pellicle : 1 µm thick polymer :� Transparency� Radiation hardness
Pellicle: to protect pattern from defects
© imec 2003 22
157nm pellicleSoft pellicle
Soft pellicle lifetime (initial target : 200J/cm2 at 70% T)� No more progress since mid 2002, focus more on understanding
Transmission ratio from Initial
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15 20
Dose E (J/cm2)
Tra
nsm
itta
nce
(A
.U)
Data presented at157nm Symposium,Antwerp Belgium, Sept., 2002
Data presented at157 nm SymposiumDana Point CA, May, 2001
© imec 2003 23
157nm pelliclesoft pellicle
Pellicle: to protect pattern from defects
First option : Soft pellicle : 1 µm thick polymer :� Transparency� Radiation hardness
© imec 2003 24
Alternative : Hard or ‘thick’ pellicle : 800µm thick quartz� Expensive� Optical element (aberrations, distortions)
157nm pelliclehard pellicle
Pellicle: to protect pattern from defects
First option : Soft pellicle : 1 µm thick polymer :� Transparency� Radiation hardness
© imec 2003 25
157nm pellicleSummary
Significant hard pellicle progress
� All hard pellicle imaging effects are understood andcan be compensated using lens manipulators
Further improvements needed
� Hard pellicle mounting needs to reduce local tilt
� Soft pellicle polymer development will improve CoO
© imec 2003 26
157nm mask substratethe issue
Modified SiO2
Surface organiccontaminationreduces transmission
© imec 2003 27
157nm mask substrateVUV cleaning can restore transmission
Principle VUV cleaning� Expectation that the transmission loss of F-doped quartz by CR-
airborne contamination is 1-2%(ref. T. Bloomstein et al., ISMT 157nm reticle handling meeting,San Diego, 2000)
� 172nm light in presence of oxygen generates dissociated oxygenradicals that break atomic bonds
ASML off-line UVO cleaner
© imec 2003 28
157nm mask substrateVUV cleaning can restore transmission
75
77
79
81
8385
87
89
91
93
150 155 160 165 170
Wavelength (nm)
Tra
nsm
issi
on
2% to 3% transmission recovery
© imec 2003 29
157nm mask substrateSummary
Significant progress
� VUV cleaning has been proven to restore mask substratetransmission
Further improvements needed
� Handling of reticles in fab, in mask shop, and in transportneeds to be checked for irreversible transmissiondegradation on the longer term
© imec 2003 30
Outline
Introduction
Status of 157nm critical issues
IMEC 157nm program� Objectives� Partners� Projects
Summary and conclusions
Acknowledgements
© imec 2003 31
IMEC 157nm Lithography ProgramProgram objectives
IMEC157nm
ProcessProgram
Resist
benchmarkingO
ptical pathm
onitoring
Hard pellicle
printing157n
m r
etic
le
hand
ling
Resist
integration on
critical layers
© imec 2003 32
Int’l Sematech
AMDHPIBM
InfineonIntel
MotorolaPhilips
TITSMC
IMEC 157nm Lithography ProgramProgram partners
Non-Sematech membersMicron
SamsungST Microelectronics
IMEC157nm
ProcessProgram
Merch. mask shopsPhotronics
Dai Nippon Printing(Toppan Printing)
(DuPont)
Equipment suppliersASMLTELKLA-TencorLam Research
Resist suppliers
Arch/FujiFilmClariant
JSRTOK
BrewerNissan Chemical
Shin Etsu
SoftwareSynopsys
Mentor GraphicsASML Masktools
© imec 2003 33
Outline
Introduction
Status of 157nm critical issues
IMEC 157nm program
Summary and conclusions
Acknowledgements
© imec 2003 34
Summary and conclusions
157nm critical challenges update� 157nm resists
Steady progress (transmittance and lithographic performance)Significant further progress needed for 45nm node (LER, integration)
� Exposure tool (Micrascan VII)The world’s first 157nm full field scanner operational at IMECLong term optical path stability (contamination) to be investigated
� Hard pellicle printingHard pellicle imaging contributions understood and compensatedPellicle-flatness specs for mounting needs to be proven
� 157nm VUV reticle cleaningOrganic contamination effect limited (2-3%) and reversibleReticle handling procedures in wafer fab and mask shop to beinvestigated (long term reticle usage)
Summary� A lot of technical progress� Still a lot of challenges since target has moved to 45nm node
© imec 2003 35
Acknowledgements
Thanks for contributions from� IMEC SPDT/LITHO� ASML 157 nm development teams (Wilton, Veldhoven)� Infineon Technologies (providing the alternating PSM)� Resist vendors (Clariant, TOK, …)� ISMT (K. Turnquest, G. Feit, K. Dean)� Selete (T. Itani)
Thanks for funding :� Part of this work is sponsored through International Sematech� European Commision : IST–2000-30175 UV2Litho� Medea+ : T401 FLUOR