Asml Euv Use Forecast

5,833 views

Published on

Semiconductor manufacturing roadmaps are contingent upon the lithography tool supplier product. EUV appears to be a viable alternative to DP/HP immersion. ASML details their EUV roadmap, and offers specifications and a look at upgrades for next node product.

0 Comments
4 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
5,833
On SlideShare
0
From Embeds
0
Number of Embeds
29
Actions
Shares
0
Downloads
172
Comments
0
Likes
4
Embeds 0
No embeds

No notes for slide

Asml Euv Use Forecast

  1. 1. EUVL – getting ready for volume introduction SEMICON West 2010 Hans Meiling, July 14, Slide 1 | public 2010
  2. 2. Outline • ASML’s Lithography roadmap to support Moore’s Law • Progress on 0.25NA EUV systems • Progress on 0.32NA EUV systems • Outlook Slide 2 | public
  3. 3. Outline • ASML’s Lithography roadmap to support Moore’s Law • Progress on 0.25NA EUV systems • Progress on 0.32NA EUV systems • Outlook Slide 3 | public
  4. 4. IC & Lithography roadmap towards <10nm 200 Resolution (half pitch) "Shrink" [nm] Logic / SRAM KrF 100 80 ArFi ArF AT:1200 60 XT:1400 6 Transistor 50 SRAM Cell XT:1700i k1 0.40 ~ 0.44 40 DRAM NAND Flash XT:1900i 30 NXT:1950i LOGIC DPT EUV-ADT DPT 20 NXE:3100 NXE:3300B NXE:3300C DRAM EUV DPT² 10 k 0.30 ~ 0.35 1 k1 0.27 ~ 0.30 DPT NAND 8 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 Source: Customers, ASML, 05/10 Year of production start Notes: 1. R&D solution required 1.5~ 2 yrs ahead of Production 2. EUV resolution requires 7nm diffusion length resist 3. DPT = Double Patterning Slide 4 | public
  5. 5. Litho costs back to normal with EUV >100 W/hr 3.5 DPT case, Litho cost EUV case, Litho cost increases 2 ~ 3 times trend returns 3 Litho cost per wafer [a.u.] 2.5 2 1.5 1 0.5 1st Gen. 2st Gen. DTP DTP 0 KrF set 1400 set 1900i set 1900i DPT 1900i DPT EUV EUV 100W/hr 180W/hr Source: Samsung, Prague, oct 2009 Slide 5 | public
  6. 6. EUV can increase the fab capacity 2x Larger footprint required to support Multi Patterning schemes Track NXE NXT Etch CVD Strip Spacer double EUV patterning 1200 m2 clean room 627 m2 clean room OR 130k Wafers / Month 250k Wafers / Month Slide 6 | public
  7. 7. Critical issues EUV 2005-2009 2005 / 32hp 2006 / 32hp 2007 / 22hp 2008 / 22hp 2009 / 22hp 1. Resist resolution, 1. Reliable high power 1. Reliable high power 1. Long-term source sensitivity & LER source & collector source & collector operation with 100 1. MASK met simultaneously module module W at IF and 5MJ/day 2. Defect free masks 2. Resist resolution, 2. Resist resolution, through lifecycle & 2. Collector lifetime sensitivity & LER sensitivity & LER inspection/review 2. SOURCE met simultaneously met simultaneously infrastructure 3. Resist resolution, 3. Availability of 3. Availability of defect 3. Availability of defect defect free mask free mask free mask sensitivity & LER 3. RESIST met simultaneously 4. Reticle protection 4. Reticle protection • Reticle protection • EUVL manufacturing 4. Source power during storage, during storage, during storage, integration handling and use handling and use handling and use  Reticle protection 5. Projection and 5. Projection and • Projection / during storage, illuminator optics illuminator optics illuminator optics handling and use quality & lifetime quality & lifetime and mask lifetime  Projection and illuminator optics quality & lifetime Source: Int’l SEMATECH, EUVL Symposium, Prague (Czech Republic), 2009 Slide 7 | public
  8. 8. Mask infrastructure improvements on blanks & inspection near levels needed for pilot production 350 50 nm@M7360 300 60 nm@M1350H 250 Defect counts 200 1/10 150 100 50 0 Q2’08 Q4’08 Q2’09 Q1’10 ML/QZ ML/ULE Optical inspection able to detect phase defects <3.4 nm x 45.4 nm in size² 1 Source: Hoya, Samsung EUV conference april 2010 2 Source: KLA, EUV symposium Prague, October 2009 Slide 8 | public
  9. 9. EUV resist makes steady progress Extrapolation of 2004-2010 progress matches shrink roadmap NXE:3100 (10mJ/cm2) NA/σ=0.25/0.8 50 NXE:3300 (15mJ/cm2) NA/σ=0.32/0.9 NXE:3300 (15mJ/cm2) NA/σ=0.32/dipole 45 Node Introduction 40 22nm dense lines Resolution 35 Dose 12.8mJ/cm2 4.3nm LWR 0.3NA small field 30 25 20 15 Ja Ja Ja Ja Ja Ja Ja Ja Ja Ja Ja n-0 n-0 n-0 n-0 n-0 n-0 n-1 n-1 n-1 n-1 n-1 4 5 6 7 8 9 0 1 2 3 4 Source: 22nm, Younkin et. al, Intel, 0.3NA MET tool, EUVS Prague, 2009 data scaled to resolution, dose, LWR, optics contrast and 7% LER by KLUP/z-factor scaling Slide 9 | public
  10. 10. EUVL Roadmap supports many generations of shrink 2006 2010 2012 2013 Proto System NXE:3100 NXE:3300B NXE:3300C Resolution 32 nm 27 nm 22 nm 16* nm NA / σ 0.25 / 0.5 0.25 / 0.8 0.32 / 0.2-0.9 0.32 / OAI Overlay (SMO) < 7 nm < 4.5 nm < 3.5 nm < 3 nm Throughput W/hr 4 W/hr 60 W/hr 125 W/hr 150 W/hr Dose, Source 5 mJ/cm2, ~8 W 10 mJ/cm2, >100 W 15 mJ/cm2, >250 W 15 mJ/cm2, >350 W Main improvements Main improvements Platform enhancements 1) New EUV platform: NXE 1) New high NA 6 mirror lens 1) Off-Axis illumination 2) Improved low flare optics 2) New high efficiency illuminator 2) Source power increase 3) New high sigma illuminator 3) Off-axis illumination optional 4) New high power source 4) Source power increase 5) Dual stages 5) Reduced footprint * Requires <7 nm resist diffusion length Slide 10 | public
  11. 11. Outline • ASML’s Lithography roadmap to support Moore’s Law • Progress on 0.25NA EUV systems • Progress on 0.32NA EUV systems • Outlook Slide 11 | public
  12. 12. EUV process viability confirmed by two 0.25NA Systems  13.5 nm NA 0.25 Field size 26 x 33 mm2 Magnification 4x reduction  0.5 • 300mm Single stage • linked to track • Single reticle load • Uses TWINSCAN technology • Sn discharge source Sou rce: um) Uni Belgi vers en, ity o (L euv f Al ban EC y( ur ce : IM Alb any So , NY ) US A Slide 12 | public
  13. 13. Overlay performance supports device integration On-product Overlay Residuals Single Machine Overlay X = 8.0 nm, Y = 7.8 nm X = 2.2 nm, Y = 2.8 nm 300 Frequency 200 100 0 300 Frequency 200 100 0 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 Error X Error Y Source: GlobalFoundries, SPIE 2010 Slide 13 | public
  14. 14. 22 nm (0.079m2) node SRAM after etch process integration SRAM cell Node Half Pitch Cell size Cell size [nm] [nm] [μm2] shrink 45 80 0.314 45 70 0.274 13% 32 62 0.186 32% 22 52 0.079 58% 16 35 0.042 47% SRAM array Source: IMEC, EUV Symposium ’09 (Prague) Slide 14 | public
  15. 15. 24nm champion resolution on 0.25NA/0.5 system From ~32nm half-pitch in 2007 to 24nm in 2010 2.50 no underlayer underlayer 2007 TMAH dev TMAH dev 2.00 2008 blurred NILS 1.50 2009 1.00 2010 24nm HP ~19 mJ/cm2 0.50 0.00 35 30 25 20 15 L/S HP (nm) Slide 15 | public
  16. 16. NXE:3100 integration status, July 2010 Reticle Stage and Handler Reticle Handling and Stage control Vacuum in vacuum system design verified, Reliability testing since Q4/2009 <1hour pump down Source Pilot Source operational at ASML Optics 5 Optics sets delivered, Flare improved to 4% Overlay Wafer Stage and handler Metrology Ongoing Wafer handling and scanning in vacuum Alignment and Leveling demonstrated, Metrology Reliability testing since Q4/2009 position fully functional Imaging First slit exposed Slide 16 | public
  17. 17. NXE:3100 integration: 3 systems completed NXE:3100 #3 System completed EUV source in installation NXE:3100 #2 System completed First wafer exposed NXE:3100 #1 Integration for Imaging System completed Integration for Overlay, S/W, TPT Slide 17 | public PUBLIC
  18. 18. 3 more NXE:3100 systems in build-up NXE:3100 #5 Currently used for Stage test setup NXE:3100 #6 System in buildup NXE:3100 #4 System almost complete 2 more Cabins used as work centers / test rigs. All 8 cabins can be used for NXE:3300 manufacturing Slide 18 | public PUBLIC
  19. 19. Multiple 3100 lenses manufactured and qualified Wavefront qualified by EUVL interferometer RMS(Z5-Z37) 1.40 RMS(spherical) RMS(coma) 1.20 RMS(ast) RMS(3-foil) 1.00 RMS [nm] 0.80 0.60 0.40 0.20 0.00 ADT 1 2 3 4 Multiple 3100 lenses within flare specifications ~15 8 Flare below 2.0 µm 7 Flare below 2.0 µm (variation over field) 6 design 5 example [%] 4 3 2 • Field size: 26mm 1 • Chief ray at mask: 6° 0 ADT 1 2 3 4 5 • 4x reduction ring field design • Design is extendable to higher NA Slide 19 | public
  20. 20. NXE metrology verified in vacuum Focus and Levelling Mean standard deviation over wafer: 0.9 nm 99.7% value of standard deviations: 1.6 nm Alignment Repeated readouts (drift corrected) (S2N) Static Repro results Smash 3 on fiducial <1nm X 0.9 Y 0.8 Multiple 0.7 0.6 wafer readout [%] 0.5 0.4 3 = 0.6 nm 0.3 0.2 0.1 m+3s 0 1 nm x: 0.5 nm y: 0.6 nm Red Green Nir Fir Slide 20 | public
  21. 21. Reliability testing ongoing on multiple systems Focus on wafer- and reticle exchange functionality 2500 5000 • all wafer- and reticle 2250 exchanges in vacuum 4500 2000 • data summed from 3 4000 different systems Number of Chuckswaps 1750 • all exchanges under 3500 Number of Cycles full SW control 1500 3000 1250 2500 1000 2000 750 1500 500 1000 250 500 0 0 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1011 1010 1003 1004 1005 1006 1009 1028 1029 1007 1008 1026 1027 Reticle Exchanges total Wafer Exchanges total Wafer+reticle exchanges Chuckswaps Slide 21 | public
  22. 22. Sources integrated with systems at ASML First EUV exposures made Source vessel operational and integrated with scanner system First EUV wafer exposed on integrated system CO2 laser operational and integrated with scanner system Slide 22 | public
  23. 23. On-site source performance: current and expectation Performance as installed at ASML • Two sources shipped to ASML, 3rd one in acceptance testing. • Two power upgrades* are planned • Upgrade #1 • Increased CO2 power by increased laser gain length. • Upgrade #2 • Increased CO2-to-EUV conversion efficiency. Source Configuration Raw Power Expose Power Baseline 40 W 20 W Upgrade #1 80 W 40 W Upgrade #2 200 W 100 W • Stable collector performance achieved on proto source. *Ref.: D.C. Brandt (Cymer), SPIE 2010. Slide 23 | public
  24. 24. Upgrade #2: Pre-pulse proof-of-concept being validated 4 With Pre-Pulse Without Pre-Pulse 3 pulse-peak CE, measured 2Q’10 CE, % 2 pulse-average CE, measured 2Q’10 1 0 0 10 20 30 40 50 60 70 80 90 100 Droplet Diameter, um ● The target size and density can be optimized by striking the droplet with a pre-pulse laser. ● The energy of the pre-pulse laser is much less than the main pulse and acts to expand the droplet size and reduce its density. ● Both the energy and timing of the pre-pulse can be adjusted to achieve best performance. Ref.: D.C. Brandt (Cymer), SPIE 2010. Slide 24 | public
  25. 25. Significant source progress required for NXE 3300 Roadmap commitments from multiple suppliers enable NXE productivity 300 Supplier 1 Public roadmaps enable 125 wph Supplier 2 250 at 15 mJ/cm2: ~2x power to go Supplier 3 Expose Power [W] 200 150 100 50 0 Feb 2009 Oct 2009 Apr 2010 NXE:3100 NXE:3300 Source: Cymer, Ushio, Gigaphoton, SPIE 10, Gigaphoton Press release April 2010 published data scaled with dose control and spectral filtering losses Data April 2010: Cymer – 30um droplets, Gigaphoton 60um droplets Slide 25 | public
  26. 26. Outline • ASML’s Lithography roadmap to support Moore’s Law • Progress on 0.25NA EUV systems • Progress on 0.32NA EUV systems • Outlook Slide 26 | public
  27. 27. NXE:3300B 1st shipment: H1 2012 2nd generation of NXE platform Specifications • NA = 0.32 • Resolution 22 nm; 18/16nm with OAI • Overlay 3.5 nm • Productivity 125 wph 15 mJ/cm2 resist Slide 27 | public
  28. 28. Six-mirror lens design is extendable to 0.32 NA Resolution improves from 27 to 22 nm • Field size 26 mm • Chief ray at mask 6° • Design complexity/cost increases • Larger mirrors • Steeper aspheric mirrors • High angles of incidence design examples 0.25 NA 0.32 NA NXE:3100 NXE:3300 Slide 28 | public
  29. 29. Further resolution improvement with off-axis illumination With dipole illumination resolution improves to below 16 nm Conventional 3.0  2.5 Dipole Quasar Image contrast (NILS) Annular 2.0 Target NILS = 2 1.5  CD  k1   1.0 NA 0.5 11 nm 16 nm 22 nm 32 nm 0.0 0.25 0.35 0.45 0.55 0.65 0.75 k1 half pitch 32 nm 28 24 22 nm 20 18 16 nm 13 11 nm k1 0.76 0.66 0.57 0.52 0.47 0.43 0.38 0.31 0.26 Conventional Annular Quasar Dipole Slide 29 | public
  30. 30. ASML c-lithography roadmap supports EUVL Support of ASML EUV scanners through Brion products 2006 2010 2012 2013 Proto NXE:3100 NXE:3300B NXE:3300C NXE models OPC & LMC SMO LithoTuner 2009.09 release 2010.06 release 2011.06 release 2012.06 release 1. Beta release 1. Production release 1. Models: NXE:3100, 1. Models: NXE:3100, 2. Modeling, correction 2. Models: NXE:3100 NXE:3300B NXE:3300B, 3300C and verification for 3. Optimized Process 2. Source Mask 2. LithoTuner flare, mask Correction (OPC) Optimization (SMO) shadowing and low 4. OPC Verification k1 proximity effect (LMC) LMC = Lithography Manufacturability Check SMO = source-mask optimization public Slide 30 |
  31. 31. NXE:3300 footprint target is <50% of NXE:3100 Incl. shared service area, for multiple systems in fab. Service Area Sub fab Area NXE:3300 Footprint NXE:3100 NXE:3100 NXE:3300 Exposure Unit footprint: 1 0.8 Subfab footprint (excl. prepumps, abatement) 1 0.4 Total footprint (incl. service area) 1 0.4 (all area’s normalized to 3100) Slide 31 | public
  32. 32. NXE:3300 mirrors are in production at Zeiss Slide 32 | public
  33. 33. Construction of new EUV facilities has started Planned NXE production capacity increases ~3x Existing EUV offices & manufacturing, 8 cabins. New EUV offices & manufacturing,15 cabins. Slide 33 | public
  34. 34. Outline • ASML’s Lithography roadmap to support Moore’s Law • Progress on 0.25NA EUV systems • Progress on 0.32NA EUV systems • Outlook Slide 34 | public
  35. 35. EUV extendibility possible beyond 10 nm resolution Through increase of the aperture up to 0.7 Aperture 0.25 0.32 0.5 0.7 6 mirror 6 mirror 8 mirror 6 mirror 8 mirror design Unobscured Central obscuration examples Reference: W.Kaiser et al, SPIE 2008 6924-4 Slide 35 | public
  36. 36. Extendibility of EUV down to sub 5 nm possible Increasing apertures up to 0.7, wavelength reduction down to 6.8 nm using 13 nm compatible optics with depth of focus as the major challenge 40 0.8 k-factor@13 nm Aperture@13 nm 35 0.7 Resolution, Depth of focus [nm] k-factor@6.8 nm 30 0.6 Aperture@6.8 nm k-factor, Aperture 25 0.5 Resolution DOF@13 nm 20 0.4 DOF@6.8 nm 15 0.3 10 0.2 5 0.1 0 0 10 11 12 13 14 15 16 18 22 17 20 21 23 20 20 20 20 20 20 20 20 20 20 20 20 20 Year Slide 36 | public
  37. 37. Summary • 6 NXE:3100 systems have been ordered by customers, in all market segments, worldwide. • 1st HVM source for NXE:3100 is operational at ASML. • performance supports system integration, and needs upgrades for 60 W/hr. • NXE:3100 in final integration phase for shipment H2 2010. • first wafer exposed, reliability testing ongoing. • NXE:3300B with 0.32 NA optics is planned for 1H 2012. • 3 source suppliers committed to meet productivity target. • optics manufacturing has started. • EUVL is extendible for multiple nodes through NA and wavelength changes. Slide 37 | public
  38. 38. public

×