Asml Holding Nv · EURONEXT:ASML
Asml Holding Nv — Conference Presentation (November 2006)
47 slides from Asml Holding Nv's conference presentation, Nov 8, 2006
Slide 1: The only cost effective extendable lithography option: EUV
The only cost effective extendable lithography option: EUV Martin van den Brink Executive Vice President Marketing and Technology Barcelona, October 17, 2006 Slide 1
Slide 2: Content
Content Lithography will remain key driver for shrink Need for shrink continues The only cost effective extendable lithography option: EUV EUV partnering and infrastructure Conclusion UY NS Barcelona, October 2006/ Slide 2 >» ASML
Slide 3: The impact of lithography in 2006
The impact of lithography in 2006 Semiconductor Market >$250B Litho Tools, Masks, Mask Equipment, Resist Materials and RET SW $13B Litho Exposure Tools $6.5B Source: VLSI Research, Barcelona, October 2006/ Slide 3 ZY VA Sy ASML
Slide 4: The impact of lithography in 2011
The impact of lithography in 2011 Semiconductor Market >$350B Litho Tools, Masks, Mask Equipment, Resist Materials and RET SW $19B Litho Exposure Tools $11B Source: VLSI Research, Barcelona, October 2006/ Slide 4 ZY & ASML
Slide 5: Content
Content Lithography will remain key driver for shrink Need for shrink continues The only cost effective extendable lithography option: EUV EUV infrastructure Conclusion YY NKS Barcelona, October 2006/ Slide 5 >» ASML
Slide 6: Customers' appetite for shrink continues unabated
Customers’ appetite for shrink continues unabated 200) 100 Resolution, “Shrink” (nm) 2 @ oOo o b So 00 01 02 O03 04 O05 06 OF O08 09 10 11 12 4p ASML Barcelona, October 2006/ Slide 6
Slide 7: Shrink drives cost per function and market growth
Shrink drives cost per function and market growth 1GB 4GB 8GB 10-20 GB 2-16GB 60-80GB 80-150 GB ate n Projected cross- over E= NAND size, GB i HDD - NAND, GB AH : 1 =—— 25 NAND Revenue, M$ 2000 2001 2002 2003 2004 2005 2006 2007 a 2009 2010 SS = == Source: Gartner Dataquest, iSuppli, ASML a & ASML B
Slide 8: Resolution, CD uniformity & overlay drive shrink
Resolution, CD uniformity & overlay drive shrink Layout 6 transistor SRAM Cell Design Rule & Cell Area [m2] Aggressive cDU cDU cell area 0.24 um? X-section ae Overlay & Overlay metal pitch 130nm through Cell Spacing ee ArF immersion Y. ; Source: IMEC, TI NN ASML jarcelona, October 2006/ Slide 8 “pv
Slide 9: Overlay and resolution (-control) key for device scaling
Overlay and resolution (-control) key for device scaling 100 90 Modeled 80 equal 99.9% = 7 yield plane i= = 6 2, 2 3 2 40 ois 3 20 10 0 100 OL, 50 40 oy 2 30[nm] *3o ty op cD uniformity M, WK Barcelona, October 2006/ Slide 9 vy ~~» ASML
Slide 10: Roadmap scenarios
Roadmap scenarios Resulting k, as function of resolution, wavelength and NA 22 2011 16 2013 half pitch 100 65 45 32 year 2005 2007 2009 4 [nm] NA 248 0.80 193 0.93 | 0.31) | 1.20 040 0.28, 1.35 0.31) 0.22 0.15, 1.55 0.26 0.18 0.25 0.59 0.41 0.35 | 057 0.41 | 0.45 0.53) 0.37 k, = (half pitch) * NA/ w
Slide 11: Highest air-based NA system will support 60-65 nm
Highest air-based NA system will support 60-65 nm half pitch 80 65 45 32 22 16 11 year 2005 2007 2009 2011 2013 2015 4 [nm] NA 248 0.80 | KA 193 0.93 : 1.20 I To: 0.28 | 1.35 | 0.31, 0.22) 0.15, 1.55 0.26 0.18 0.25 I 0.59) 0.44 0.35 | 0.57 0.41, 0.45 0.53 0.37 4 SKS Barcelona, October 2006/ Slide 11
Slide 12: ASML mask and system enhancements extend lithography to the limit of k1
ASML mask and system enhancements extend lithography to the limit of k, @-& DoseMapper for optimum CD Uniformity Flexible off-axis & polarized illumination ©O Application specific - a lens setup In-built wave-front, polarization and pupil metrology Ur SAY Barcelona, October 2006/ Slide 12 yy ASML =”
Slide 13: First super high NA immersion system enables nm
First super high NA immersion system enables 45 nm half pitch 100 65 45 32 year 2005 2007 2009 4 [nm] NA 248 0.80 193 0.93 22 2011 16 2013 11 2015 mm First super hig! 1.20 | aystem enable 135 | 031, 0.22 0.15 1.55 0.26 0.18 0.25 | 0.59) 0.44 0.35 | 0.57 0.44 0.45 | | | 0.53, 0.37 Z NA Barcelona, Oct
Slide 14: Overview dense line XT:1700i imaging results
Overview dense line XT:1700i imaging results 42nm 1.2NA, o=0.89/0.98, Dipole X-35, Y polarization, k1 =0.261 MA 950nm DoF -500nm -300nm -180nm NF +180nm +300nm = +450nm 45nm 1.2NA, o=0.82/0.97, C-Quad-30, XY polarization, k1 = 0.28 -300nm = -240nm_ -120nm +60nm = +120nm +210nm 500nm DoF 50nm 1.2NA,
Slide 15: Roadmap scenarios, the impact of immersion
Roadmap scenarios, the impact of immersion Water-based 193 not sufficient for 32-nm half pitch DPT real option, but costly half pitch 100 65 45 32 22 16 11 year 2005 2007 2009 2011 2013 2015 i [nm] NA 248 0.80 Max NA water-based 193 nm immersion 193 0.93 0.31 requires double patterning to get to 32
Slide 16: New software & algorithms required to split & optimize OPC and stitching
New software & algorithms required to split & optimize OPC and stitching for Double Patterning Memory Logic Challenges _ + Correct decomposition Original * OPC for decomposition layout + Model-based stitching error compensation —h! |, Increasing Difficulty NAND Flash Pattern DRAM split Restricted Lo
Slide 17: Double line patterning; 32-nm half pitch Flash
Double line patterning; 32-nm half pitch Flash Target Poly patterning Min Pitch 64nm SPLIT + OPC Annular 0.8/0.5, X-Y polarized k1 = 0.20 XT:1700i, 193nm - 1.2NA MASK A MASK B liilllN ilfealll a Sell gun My Co work ASML, Imec, Synopsys and Mentor Graphics VA Ss NM Barcelona, October 2006/ Slide 17 Y
Slide 18: Double trench patterning: overlay induced CD change
Double trench patterning: overlay induced CD change A overlay = 0 Exposure 1 | li | | TWINSCAN™ Overlay Learning Exposure 2 25 E 20 Final image > SMM 3 cD rt = £ 40 A overlay = CD/3 = Exposure 1 s ees, CF —— 2 : £ Target 32nm < CD/10 Exposure 2 @ 0 || a) | C] 2000 2002 2004 2006 © «2008 +«©=—-2010 F
Slide 19: 32-nm half pitch with immersion extremely challenging
32-nm half pitch with 193 immersion extremely challenging 22 2011 16 2013 1 2015 half pitch 100 65 45 32 year 2005 2007 2009 2 [nm] NA 248 0.80 193 0.93 § I. NA 1.55 requires new liquid, new glass 4.20 to extent to 32nm 1.35 § | | 2° 0.15 1.55 0.18 025 | E 0.41 035 | 0.57 0.41 0.45 0.53 0.37 Y SKS B
Slide 20: Apertures, field sizes and refractive indices
Apertures, field sizes and refractive indices 6% 15% New fluid n>1.8 New 8 £ lens material E n>1.9 £2 Second gen fluid New resist & n=1.65 n>1.8 3 18 NY 2 lens material { n>1.9 14 1.45 15 1.55 16 15% 6% VN _ 7" ASML Barcelona, October 2006/ Slide 20 L
Slide 21: EUV the only high volume opportunity
EUV the only high volume opportunity 22 2011 half pitch 100 65 45 32 year 2005 2007 2009 A [nm] NA 248 0.80 193 0.93 4.20 EUV required for 32 nm as cost reduction = m for double patterning and more extendable 1.35 technology than non water immersion 1.55 0.25 0.35 0.45 Y WK Barcelona, October 2006/
Slide 22: Arrival at IMEC and Albany Nanotech (Aug.'06)
Arrival at IMEC and Albany Nanotech (Aug.’06) Albany IMEC EUV development systems are being installed at two customer sites My, SKS Barcelona, October 2006/ Slide 22 >» ASML
Slide 23: Likely technology roadmap
Likely technology roadmap half pitch 100 65 45 32 22 16 1 year 2005 2007 2009 2011 2013 2015 A [nm] NA 248 0.80 Pitch relaxation or 193 0.93 — patterning 1.20 1.35 Low k, PE Fluid material 1.55 [ll challenge _n — 6-16 — ah challenge 0.25 0.35 = 0.45 Infrastructure challenge Oh likely {opportunity yy
Slide 24: ASML 300mm Product Roadmap
ASML 300mm Product Roadmap oan aS tcn cnn = rm iT ee 2x Patterning XT:1900G} Immersion XT:875F XT:870F 2 oS SE T:760F ————————_————— x -400F XT-450F| a a & ASML Barcelona,October 2006 / Slide 24
Slide 25: Content
Content Lithography will remain key driver for shrink Need for shrink continues The only cost effective extendable lithography option: EUV EUV partnering and infrastructure Conclusion 4 SKS Barcelona, October 2006/ Slide 25 Yy ASML
Slide 26: Full field, through focus 40-nm lines, contacts
Full field, through focus 40-nm lines, 55 contacts Resist: MET-2D, ~18 mJ/em2 NA=0.25 c= 0.5 (conventional illumination) Gj KS Barcelona, October 2006/ Slide 26 yy ASML
Slide 27: Scanning, full field, EUV overlay nm
Scanning, full field, EUV overlay 7 nm ) o (2) o = tad Frequency r- o = nN ° 0 “8.5 -6.8 -5.1 -3.4 -1.7 0.0 1.7 34 5.1 68 8.5 Overlay [nm] Corrected for X/Y shifts YW SNS
Slide 28: Source power progress has been increasing
Source power progress has been increasing Actual data Planned performance > 1000 = 180 W <»100 W/hr @ 10 mJ/cm? — < S 100 te = £ a aes wu Xe | © 10 = = = = * SnXe o O supplier 1 S 1 - supplier 2 3 fo) . a supplier 3 supplier 4 0.1 2001 2003 2005 2007 2009 2011 Year QZ NAS Barcelona, October 2006/ Sl
Slide 29: Resist shot noise likely not the Line Edge Roughness limitation
Resist shot noise likely not the Line Edge Roughness limitation LER versus Sensitivity for selection of known EUV resists 14 12 10 E 8 i 6 =I 4 2 ) ) 10 20 30 “ad Sensitivity (mJ/cm42) Status: Line Edge Roughness (HVM Spec): < 1.6 nm Line Edge Roughness (Best Current); ~ 2.3 nm Ss Bruno La Fontaine,
Slide 30: Non litho post processing improved LER
Non litho post processing improved LER 0.1000 o 4 8 —Resist LER, ILS=10um" = == Resist LER, ILS=20um" ——PolySi LER, ILS=10um" ~ = ~ PolySi LER, ILS=20um" ° 8 3 Power Spectral Density (nm*him") 0.0001 1 19 Frequency (ym) Source Pawloski, SPIE microlithogrphy '06 Barcelona, October 2006/ Slide 30
Slide 31: Progress in blank defect density reduction
Progress in blank defect density reduction EUV test mask 10.000 Total blank defects = >120 nm 5 @ 1.000 5 >80 nm E SS o 3B ; = 2 0.100 § ae $ Ne 3 Oo} Potential further 2 >60nm eduetion ~ goto L by smoothing 5 Total added def = z mien A /W gee SS 0.003 substrate 43 defects only. >43 nm ol | | Ly Lee
Slide 32: Zero added particles per reticle exchange required
Zero added particles per reticle exchange required o o 10.000 size of scanned area in > the measurement o A (mask active area = & 5 1.000 14x14cm? = 20cm?) Qa 2 Ss particle size: >75 nm = 2 0.100 Edge Area Scanned. = a (near frame & o . contact point) os 0.010 incl. o¢ no — ++ clamp oS i CS <t clamp
Slide 33: No clear customer consensus for nm, EUV preferred for nm
No clear customer consensus for 32 nm, EUV preferred for 22 nm Customer Lithography preferences for 32 nm & 22 nm (2 year roadmap) Litho Technology 32nm node/ 2009 22nm node/ 2011 EUV preferred for Memory EUV preferred Immersion (single or double exposure) preferred for Logic Poll of 14 customers, A
Slide 34: Historic reduction stepper imaging technology changes
Historic reduction stepper imaging technology changes Technology Incubation time [yr] Production Insertion Diffraction | Incremental limit [nm] Improvement % 436 nm/air 1980 109 365 nm/air 1989 91 19 248 nm/air 1995 62 47 193 nm/air 2002 48 28 157 nm/air 39 23 193 nm/water 2006 34 44 193 nm/HI >2010
Slide 35: Increased litho process complexity drives cost
Increased litho process complexity drives cost ) Hard Mask Develop Inorganic BARC |) Expose Ml organic BARC ~=—— Metrology 1 Resist TB Hard Mask Etch 1) Top Coat MN Strip & Clean ArFi ArFi DPT EUV 45nm&32nm 32nm 32 nm Process complexity / cycle time / Cost Barcelona, October 2006/ Slide 35
Slide 36: Cycle time of multiple exposure strategies increases
Cycle time of multiple exposure strategies increases A Process (Single Exposure) ACycle time B Process (Double Exposure) Cycle time Higashiki, Toshiba, Santa Clara, SPIE march 06 b Uy Barcelona, October 2006/ Slide 36 ys ASML
Slide 37: Content
Content Lithography will remain key driver for shrink Need for shrink continues The only cost effective extendable lithography option: EUV EUV partnering and infrastructure Conclusion Y SKS Barcelona, October 2006/ Slide 37 ~» ASML
Slide 38: EUV at Research Institutes
EUV at Research Institutes In August, ASML shipped EUV Alpha Demo Tools to the College of Nanoscale Science and Engineering (CNSE) of the State University of New York (SUNY) at Albany, N.Y., and to IMEC in Leuven, Belgium. ASML is integrating these tools on-site to prepare them for full-system quali
Slide 39: Continuous progress of EUV infrastructure through partners
Continuous progress of EUV infrastructure through partners Carl Zeiss Cymer Philips Extreme Rohm and Haas Electronic Materials Toppan Photomasks XTREME technologies 4 & ASML Barcelona, October 2006/ Slide 39
Slide 40: EXTUMASK, EXCITE, More Moore consortium partners
EXTUMASK. 2001-2004, EXC IT E. 2002-2005. more Moore: 2004-2006, “push the limits of lithography to enable and exceed the requirements for the 22 nm node” ASML, SIGMA-C SOFTWARE, PHYSTEX, IMAGINE OPTIC, XTREME TECHNOLOGIES, PHILIPS EXTREME UV, FOCUS, EPPRA, XENOCS, SAGEM DEFENSE SECURITE, AMTC, CARL
Slide 41: EUV achievements More Moore consortium
EUV achievements More Moore consortium . Sources Introduction of tin as plasma Introduction of rotating disc electrodes Full size transmission Spectral Purity Filter prototype for suppression of out-of-band EUV radiation - Optics multi layers improvement resulting in 70 % mirror reflectance Demonstr
Slide 42: EUVSources, EXTUMASK, More Moore, EAGLE
EUVSources. 2001-2004, EXTUMASK 2001-2004 more Moore: 2004-2006. EAGLE 2006-2008, “develop the technology for an EUV lithographic platform for high volume manufacturing” ASML, FOM, Carl Zeiss SMT AG, Philips EUV, Xtreme Technologies, Alcatel, Sagem, Barcelona, October 2006/ 2 Media Lario
Slide 43: European Project partners
European Project partners & Sagem Défense Sécurité ENB pepe ome SCHOTT lithotec is >*freescale —- <enocs ang 5 9 limelts “XTREME out, a . ues a eGeim Uru Q PHILIPS Ae Pips Text i CARL 2E155 SAT 4 NES Barcelona, October 2006/ Slide 43 Yy ASML
Slide 44: Content
Content Lithography will remain key driver for shrink Need for shrink continues The only cost effective extendable lithography option: EUV EUV partnering and infrastructure Conclusion 4, NKS Barcelona, October 2006/ Slide 44 »» ASM L
Slide 45: Lithography roadmap
Lithography roadmap Water-based immersion will capture the 40-nm half pitch using 1.35-NA 193-nm lithography. Non water-based immersion needs new lens materials to increase resolution capability significantly: . Without new lens material, new fluid technology full field resolution advantage limited
Slide 46: The promises of EUV
Barcelona, October 2006/ Slide 46 The promises of EUV Process complexity and cycle time will go up and EUV now becomes a cost and cycle time reduction opportunity. Cost reduction to be achieved by single exposure, single mask and low OPC content. EUV mask and process infrastructure will be developed
Slide 47: Commitment
ASML Holding — Conference Presentation (November 2006) on 6ix