Update on 193nm immersion exposure tool S. Owa, H. Nagasaka, Y. Ishii Nikon Corporation O. Hirakawa and T. Yamamoto Tokyo Electron Kyushu Ltd. January 28, 2004 Litho Forum 1
What is immersion lithography? NA = n sin θ Liquid recovery Projection optics Liquid supply Resolution = k1 λ / NA = k1 λ / (n sin θ) = k1 (λ/n) / sin θ Improvement of Resolution Wafer stage Immersion liquid (water) (index = n) (Scanning motion) Wafer DOF = k2 (λ/n) / 2 (1-cos θ ) = k2 (λ/n) / 4 sin 2 (θ/2) ~ k2 (λ/n) / sin 2 θ = k2 n λ / NA 2 Improvement of DOF Immersion is effective for both resolution and DOF January 28, 2004 Litho Forum Slide 2
Advantage of immersion lithography Effective wavelength (λ/n) (Potential advantage) medium n λ/n ratio ArF dry Air 1.0 193nm 1.00 F2 dry N 2 1.0 157nm 0.81 ArF immersion H 2 O 1.44 134nm 0.69 F2 immersion PFPE 1.37 115nm 0.60 ArF immersion can be considered as 134nm lithography Nikon starts immersion from ArF. January 28, 2004 Litho Forum Slide 3
Nikon Lithography Roadmap CY 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 Technology Node hp130 hp90 hp65 hp45 hp32 hp22 DRAM Half Pitch 130 115 100 90 80 70 65 57 50 45 35 32 25 22 MPU Gate in resist 90 65 53 45 40 35 32 28 25 20 18 15 13 post-etch 65 45 37 32 28 25 22 20 18 14 13 10 9 Critical layer exposure method KrF 248nm ArF 193nm immersion F2 immersion 157nm EPL 100keV EUV 13nm January, 2004 January 28, 2004 Litho Forum Slide 4
Contents 1. Issues 1. Water fill method 2. Edge shot 3. Projection optics design 4. Real time focus sensing 5. Polarization effect 6. Thermal aberration in water 7. Water supply (temperature and bubble) 8. Resist 9. Scanning speed 10. Remaining risks 2. Plan of full-field exposure tool January 28, 2004 Litho Forum Slide 5
193nm Immersion scanner new items for immersion Inline Track Control Software (interface) Reticle Loader Immersion Optics Laser light source Water supply controller Water Purifier wafer Wafer Loader Temperature Flow rate Degas Purity Focus sensor Nozzle (Local fill) Immersion Wafer stage Blue letters: mainly TEL Green letters: mainly Nikon January 28, 2004 Litho Forum Slide 6
Issue 1: Water fill method Local immersion = Local fill Liquid recovery Projection optics Liquid supply Wafer stage Wafer immersion = pool stage Immersion liquid Mirror (Scanning motion) Fluid Inlet Last Lens Element Wafer Tank Fluid Cover Outlet Wafer Fluid Mirror Stage immersion = submarine stage Vacuum Pump Fluid Replenishing Hole Thermal Control Filter Drain (B. J. Lin) Nikon has selected and focused on Local fill since 2002. January 28, 2004 Litho Forum Slide 7
Local fill simulation Water recovery Projection optics Water supply Wafer stage Water t=1mm (Scanning motion) Wafer (simulation in January 2003) January 28, 2004 Litho Forum Slide 8
Local fill experiment (Experiment in June 2003) January 28, 2004 Litho Forum Slide 9
Issue1: Water fill method Status: 1. Local fill is feasible 2. Edge shot becomes an issue by this method. January 28, 2004 Litho Forum Slide 10
Issue 2: Edge shot January 28, 2004 Litho Forum Slide 11
Issue 2: Edge shot Edge shot is necessary for: Uniform CMP condition Chip cost, etc. wafer shot Edge shot water area Near edge shot January 28, 2004 Litho Forum Slide 12
Edge shot demonstration (backside) January 28, 2004 Litho Forum Slide 13
Edge shot demonstration (backside) Edge shot is feasible! January 28, 2004 Litho Forum Slide 14
Issue 2: Edge shot Status: 1. Edge shot is feasible, and successfully demonstrated. January 28, 2004 Litho Forum Slide 15
Issue 3: Immersion projection optics January 28, 2004 Litho Forum Slide 16
Issue 3: Immersion projection optics Nikon has finished designs of: NA = 0.85, 4X, 26mm x 33mm dry imm. NA = 0.92, 4X, 26mm x 33mm dry imm. NA = 1.0, 4X, 26mm x 33mm all refractive NA = 1.2, 4X, 26mm x 33mm catadioptric NA = 1.2 8X, 13mm x 16.5mm all refractive 4X design is most difficult, 5X, 6X, 8X optics are easier. We can keep 4X for 26x33mm 2 & NA > 1.2 January 28, 2004 Litho Forum Slide 17
Lens diameter vs. NA Lens Diameter (A.U.) (4X, 26x33mm field) All Spherical With Aspheres Liquid Immersion 0.6 0.7 0.8 0.9 1.0 1.1 1.2 NA January 28, 2004 Litho Forum Slide 18 SPIE 2003 Feb.28 Santa Clara ML5048-68 Ohmura et. al. Slide 1
Issue 3: Immersion projection optics Status: 1. We have obtained full field immersion projection optics designs from NA=0.85 to NA=1.2. 2. Nikon can keep 4X reduction ratio for NA > 1.2 and 26mm x 33mm field. January 28, 2004 Litho Forum Slide 19
Issue 4: Real time Focus sensing January 28, 2004 Litho Forum Slide 20
Issue 4: Real time Focus sensing water lens Reflected beam Focus sensing beam Optical, Real-time Focus sensor is adopted as the most reliable method for immersion. January 28, 2004 Litho Forum Slide 21
Focus sensing in dry and immersion θ resist wafer θ water resist wafer Reflectivity at Air/Resist, or, Water/Resist interface Reflectivity 反射率 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 Dry (Air/Resist) 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 R~0.55 0.2 0.1 0.0 70 80 70 80 Incidence angle θ Reflectivity 反射率 Immersion (Water/Resist) Incidence angle θ R~0.55 Equal reflectivity (R~0.55) is possible in both Dry and Immersion. Same accuracy expected. January 28, 2004 Litho Forum Slide 22
Issue 4: Real time Focus sensing Status: 1. Real time optical focus sensor is adopted. 2. Same accuracy with dry sensor is expected. January 28, 2004 Litho Forum Slide 23
Issue 5: Polarization effect January 28, 2004 Litho Forum Slide 24
Issue 5: Polarization effect TE-Polarization (s-polarization) illumination light TM-Polarization (p-polarization) illumination light TE High contrast E-vector E-vector TM Low contrast January 28, 2004 Litho Forum Slide 25
Loss less polarized illumination Polarized illumination without power loss will be available: S307E (NA=0.85), NA=0.92 and immersion tools. Polarized Laser Beam Loss-less polarization controller Dipole + Linear polarization Polarization maintaining illuminator Mask TE (s) polarization enhanced light Annular + Azimuthal polarization January 28, 2004 Litho Forum Slide 26
Polarized illumination: 1D L/S pattern 700 600 500 400 300 200 6% attn-psm ED-tree σ = 0.95 DOF (nm) 800 Dipole + linear polarization Dipole + random polarization dipole r = 0.20σ mask +-2% dose +-3% CD +-10% NA=0.92 dry NA=1.00 imm. NA=1.20 imm. DOF (nm) DOF (nm) 800 800 700 700 600 600 500 500 400 400 300 300 200 100 100 100 0 0 0 40 50 60 70 80 40 50 60 70 80 40 50 60 70 80 Half pitch = CD (nm) Half pitch = CD (nm) Half pitch = CD (nm) 200 January 28, 2004 Litho Forum Slide 27
2D L/S pattern (L/S of any direction) 6% attn-psm Annular + Azimuthal polarization Annular + Random polarization NA=0.92 dry DOF (nm) 800 700 600 500 400 300 200 100 0 40 50 60 70 80 Half pitch = CD (nm) DOF (nm) 800 700 600 500 400 300 200 100 NA=1.00 imm. 0 40 50 60 70 80 Half pitch = CD (nm) DOF (nm) 800 700 600 500 400 300 200 100 0 ED-tree ¾ annular σ = 0.95 dose = +- 3% CD = +- 10% NA=1.20 imm. 40 50 60 70 80 Half pitch = CD (nm) January 28, 2004 Litho Forum Slide 28
1D dense contact holes 6% attn-psm 2xCD 2xCD CD Dipole + linear polarization Dipole + random polarization ED-tree σ = 0.95 dipole r = 0.125σ mask +-2% dose +-3% CD +-10% DOF (nm) 400 350 300 250 200 150 100 50 NA=0.92 dry NA=1.00 imm. DOF (nm) NA=1.20 imm. 0 40 50 60 70 80 90 100 Half pitch = CD (nm) DOF (nm) 400 350 300 250 200 150 100 50 0 40 50 60 70 80 90 100 Half pitch = CD (nm) 400 350 300 250 200 150 100 0 40 50 60 70 80 90 100 January 28, 2004 Litho Forum Slide 29 50 Half pitch = CD (nm)
Mask-side polarization effect Mask becomes polarizing plate. But mask TE-0, TM-0 TE: E-vector is parallel to line TE-1, TM-1 diffraction efficiency 40 35 30 25 20 15 10 5 0 Intensity of TE and TM polarization TM-0 TE-0 TE-1 TM-1 0 20 40 60 80 100 120 Half pitch converted to wafer(nm) (4X reduction ratio) TE (high contrast) is stronger than TM 4X is OK down to 35nm half pitch. It is good for contrast! January 28, 2004 Litho Forum Slide 30
Issue 5: Polarization effect Status: 1. Polarized illumination is effective: 1. Dipole + Linear pol. 1D L/S, 1D dense holes 2. Annular + Azimuthal pol. 2D L/S 2. NA=1.2 ArF-immersion + polarized illumination will enables half-pitch 45nm node. 3. Nikon plans to supply loss-less polarized illumination from NA=0.85-0.92 (dry) exposure tool to immersion tools. 4. Mask-side polarization effect exists. But no problem for 4X reduction ratio. January 28, 2004 Litho Forum Slide 31
Issue 6: Thermal aberration in water January 28, 2004 Litho Forum Slide 32
Issue 6: Thermal aberration in water Temperature change by exposure Water recovery Projection optics Water supply Wafer stage Water t=1mm (Scanning motion) Wafer Maximum temperature increase (long time scale) by exposure ~ 0.01K simulation boundary (not actual structure) January 28, 2004 Litho Forum Slide 33
Thermal aberration in the water Temperature distribution caused by exposure light: Maximum T =0.011K Wavefront aberration 1.7mλ Distortion 3 nm Max Focus shift 7 nm Max Small enough Lens Water January 28, 2004 Litho Forum Slide 34
Issue 6: Thermal aberration in water Status: 1. Thermal aberration by exposure light is small enough. 2. Water supply temperature control of T<+-0.01K should be necessary. January 28, 2004 Litho Forum Slide 35
Issue 7: Water supply January 28, 2004 Litho Forum Slide 36
Issue 7: Water supply Requirements: Temperature stability (+- 0.01K) Bubble prevention January 28, 2004 Litho Forum Slide 37
Temperature stability Data of water supply being developed for NA=0.85 tool. 23.10 32.0 23.05 Output water temperature 23.02 +- 0.01 C 30.0 23.00 28.0 22.95 Input water 26.0 22.90 24.0 22.85 60sec Room temperature 22.0 22.80 20.0 Temperature stability of +- 0.01C attained January 28, 2004 Litho Forum Slide 38
Issue 7: Water supply Status: 1. Temperature stability of +-0.01K attained. 2. Bubbles and particles (> 0.1um) are eliminated. January 28, 2004 Litho Forum Slide 39
Issue 8: Resist January 28, 2004 Litho Forum Slide 40
Issue 8: Resist Nikon is collaborating resist suppliers by using a 2-beam interferometer. 193nm laser 2-beam interferometer TE = s TM = p half wave plate Prism wafer Prism Water (0.3mm) 193nm light s p θ p s n 193nm light Resist Wafer Polarization can be changed. January 28, 2004 Litho Forum Slide 41
Resist availability Immersion exposure at λ=193nm 65nm L/S 50nm L/S 45nm L/S Collaboration with TOK January 28, 2004 Litho Forum Slide 42
Resist availability Immersion exposure at λ=193nm 55nm L/S 55nm L/S 45nm L/S Collaboration with JSR January 28, 2004 Litho Forum Slide 43
Resist availability Immersion exposure at λ=193nm 43.7 nm 47.3 nm 50nm half pitch 45nm half pitch Collaboration with Clariant January 28, 2004 Litho Forum Slide 44
Polarization effect TE-polarization (s-polarization) contrast 1.0 TE&TM-mixture (s&p-mixture) contrast 0.5 TM-polarization (p-polarization) 55nm p-polarization contrast 0.47 50nm 0.35 45nm 0.20 In collaboration with TOK January 28, 2004 Litho Forum Slide 45
Issue 8: Resist Status: 1. As results of collaboration with resist suppliers, we have found resists that can image 50 45nm half pitch. 2. Polarization effect was demonstrated by a 2-beam interferometer. January 28, 2004 Litho Forum Slide 46
Issue 9: Scanning speed limit January 28, 2004 Litho Forum Slide 47
Issue 9: Scanning speed Status: 1. 500mm/s scanning was demonstrated. 2. Same scan speed with dry system seems to be possible. January 28, 2004 Litho Forum Slide 48
Remaining risks 1. Hydrodynamic vibration 2. Bottom lens protection 3. Defect by nano bubbles 4. Contamination in water These issues will be investigated by constructing NA=0.85, full-field scanner (2004/3Q). January 28, 2004 Litho Forum Slide 49
Plan of immersion exposure tools 2004/3Q Engineering Evaluation Tool (EET) NA = 0.85 (after immersion) (S307E base, being settled in Nikon Factory) 2005 NA = 0.92 production model 2006 NA > 1.0 production model January 28, 2004 Litho Forum Slide 50
Summary 1. ArF immersion lithography will cover half-pitch 65nm node and half-pitch 45nm node. 2. Local fill and Edge shot are feasible and successfully demonstrated. Scan speed of 500mm/s was attained. 3. Projection optics design of NA=1.2, 4X, 26x33mm 2 field has been obtained. Nikon can keep 4X for NA>1.2. 4. Polarized illumination is effective. Loss-less polarized illuminator will be adopted from NA=0.85-0.92 dry tool to immersion tools. 5. Bubble (size > 0.1um) has been successfully eliminated. 6. NA=0.85, ArF immersion full field scanner will be complete in 2004/3Q. January 28, 2004 Litho Forum Slide 51