EUV multilayer coatings: potentials and limits 2012 International Workshop on EUV Lithography Sergiy Yulin, Torsten Feigl, Viatcheslav Nesterenko, Mark Schürmann, Marco Perske, Hagen Pauer, Tobias Fiedler and Norbert Kaiser Fraunhofer-Institut für Angewandte Optik und Feinmechanik Jena, Germany Thursday, 7 June 2012, Maui, Hawaii
Outline Introduction High-reflective / MLs High-temperature -based MLs Radiation stable / MLs B-based MLs for future generation of EUVL Summary and outlook Seite 2
Some highlights of R&D for EUVL @ Fraunhofer IOF 1999 First R&D work on interface engineering 2002 Transfer of / technology to Schott Lithotec AG 2003 Design and realization of NESSY-1 system for EUVL optics 2004 Development of high-temperature MLs for Cymer Inc. 2005 First high-temperature collector with 320 mm for Cymer Inc. 2006 Development of TiO 2 and Nb 2 O 5 capping layers for Intel Corp. 2007 Development of optics cleaning technologies for Intel Corp. 2008 Development of new capping layer strategy for Intel Corp. 2008 First collector with 660 mm for Cymer Inc. 2010 Design and realization of NESSY-2 system for EUVL optics 2012 Design and realization of NESSY-3 system for R&D Seite 3
Calculated reflectivity, % High-reflective -based multilayer mirrors at 13.5 nm D.G. Stearns and R.S. Rosen: High-performance multilayer mirrors for soft-x-ray projection lithography: Proc. SPIE 1547, 1991 80 70 60 Henke, 1991 AOI = 5 o, s-pol. / Nb/ Ru/ System R 13.5, % FWHM, nm - 75.4 0.58 50 Nb- 74.6 0.53 40 Ru- 72.1 0.69 30 20 R = 75.3% R = 74.6% R = 72.1% / couple was selected due to: 10 0 12,5 13,0 13,5 14,0 14,5 Wavelength, nm 1) Max. theoretical optical performance 2) Minimum structural defectiveness Seite 4
Main imperfections in / multilayer mirrors Interface roughness Diffusion intermixing Surface oxidation / / Cr/Sc ~8 nm x y d S U B S T R A TE 5 nm 5 nm The gap between theoretical and experimental reflectivity is integrated effect of all multilayer defects. Internal and surface defects should be mitigated. Seite 5
Defect mitigation strategy for / mirrors at the IOF Mitigation strategy for multilayer defects includes: Interface-engineering: diffusion barriers (X) capping layer (Y) buffer layer (Z) Optimized deposition process: deposition parameters deposition tools Conventional design S U B S T R A TE d Interface - engineered S U B S T R A TE Y capping layer d X barriers Z buffer layer Transition from conventional (2-materials) to interface-engineered design (at least 5 materials) is general mitigation strategy of multilayer defects used at IOF. Seite 6
Magnetron sputtering systems at the IOF 2 X MRC (1998): R&D system NESSY 1 (2003): 3 x 300mm (with load lock) NESSY 2 (2010): up to 650mm NESSY 3 (2012): R&D system Seite 7
Outline Introduction High-reflective / MLs High-temperature -based MLs Radiation stable / MLs Broadband / narrowband / MLs B-based MLs for future generation of EUVL Summary and outlook Seite 8
Enhanced reflectivity with diffusion barriers Conventional design Interface-engineered design Main material requirements: Barriers (X) Minimum absorption at 13.5 nm Low diffusion mobility Continuous film growth d 1 d 2 Currently used barrier materials: Boron carbide (B 4 C) Carbon (C) licon carbide (C) S U B S T R A TE S U B S T R A TE Typical thickness: 0.2 0.5 nm Maximum reflectivity benefit: < 2.0% S. Bait et al. Improved reflectance and stability of / multilayers: Opt. Engin., 41(8), pp. 1797 1804 (2002). Seite 9
Reflectivity, % Reflectivity (13,5 nm), % Enhanced reflectivity of /C//C multilayer mirrors calculation 70 60 50 [/C//C] 60 + without 0.4 nm C on 0.4 nm C on R = 68.8 % = 13.5 nm FWHM = 0.50 nm 76 74 [/C//C] 60 + C on C on C on both -1.2 % 40 30 20 10 R = 69.6 % = 13.36 nm FWHM = 0.52 nm R = 67.4 % = 13.1 nm FWHM = 0.47 nm 72 70 68 [/ 2 // 2 ] 60 + -5.4 % -6.8 % 0 12,6 12,8 13,0 13,2 13,4 13,6 13,8 14,0 Wavelength, nm 0,0 0,2 0,4 0,6 0,8 Thickness of carbon, nm [/] 60 [/C (0.4 nm)/] 60 + + 0.8% Seite 10
Reflectivity, % Reflectivity (13,5 nm), % Enhanced reflectivity of /C//C multilayer mirrors calculation 70 60 50 [/C/] 60 + t C = 0.5 nm 0.0 nm C 0.5 nm C on -on- [/C/] 60 R = 69.1 % = 13.44 nm FWHM = 0.52 nm 76 74 [/C///C] 60 ++O 2 C on -on- C on -on- C on both / -0.8 % 40 30 20 [/] 60 R = 68.8 % = 13.5 nm FWHM = 0.5 nm 72 70-3.8 % -4.5 % 10 0 12,8 13,0 13,2 13,4 13,6 13,8 14,0 Wavelength, nm [/ 2 (1.4 nm)// 2 (0.6nm)] 60 + 68 0,0 0,2 0,4 0,6 0,8 Thickness of C, nm [/] 60 [/C(0.5 nm)/] 60 + + 0.3% Seite 11
Reflectivity (13,5 nm), % Reflectivity (13,5 nm), % Material selection for diffusion barriers (C, B 4 C, 2 C...) calculation 78 76 Carbon or B 4 C? [/C//C] 60 + [/B 4 C//B 4 C] 60 + 76 B 4 C or 2 C? Or...? [/B 4 C//B 4 C] 60 + [/ 2 C// 2 C] 60 + 74 74 72 70 C on C on C on both / B 4 C on B 4 C on B 4 C on both 72 70 / B 4 C on B 4 C on B 4 C on both 2 C on 2 C on 2 C on both 68 [/ 2 // 2 ] 60 + [/ 2 // 2 ] 60 + 0.0 0.2 0.4 0.6 0.8 Barrier thickness, nm 68 0.0 0.2 0.4 0.6 0.8 Barrier thickness, nm B 4 C is more preferable than carbon. 2 C is more preferable than B 4 C. Seite 12
Current state and prospects for reflectivity enhancement The highest reflectivity at 13.5 nm & AOI 5: System R exp. % FWHM, nm Ref. /X//X 70.15 0.54 FOM /B 4 C//B 4 C 70.0 0.53 LBLL /B 4 C//C 69.9 0.54 IWS //C 69.6 0.53 IOF Further prospects: Transition from / to Nb/X//X, Ru/X//X. multilayer mirrors Application of new promising diffusion materials Application of new oxidation resistant capping layers Maximum reflectivity benefit is 1.0-2.0%... Seite 13
Outline Introduction High-reflective / MLs High-temperature -based MLs Radiation stable / MLs B-based MLs for future generation of EUVL Summary and outlook Seite 14
Reflectivcity, % Thermal stability of conventional / multilayer mirrors 70 60 / (M744) N = 50 = 2 h as-deposit T=100 o C T=200 o C / multilayer mirrors: 50 T=300 o C T=400 o C Max. reflectivity: ~ 69.0% 40 Max. temperature: ~ 100 o C 30 Structure damage: ~ 700 o C 20 10 0 12,3 12,6 12,9 13,2 13,5 13,8 14,1 Wavelength, nm 50 nm V.V. Kondratenko et al: Thermal stability of soft X-ray / and 2 / multilayer mirrors, Appl. Opt, 1993. H. Takenaka et al.: Thermal stability of /C//C multilayer soft X-ray mirrors: J. Elect. Spectroscopy, 1996. Seite 15
/ multilayers with enhanced thermal stability First request: Cymer Inc. @ EUVL Symposium in Miyazaki (2004) Requirements: R > 60% @ 13.5 nm and T 500 o C Application: Collector EUVL optics (lithium cleaning: Tm = 108 o C) Seite 16
Reflectivity, % Enhanced thermal stability of 2 / mirrors 40 T = 500 C [/ 2 ]60+ q = 1.5 2 / multilayer mirrors: 30 20 As-deposited =1 h =10 h =100 h Temperature: Reflectivity: FWHM: 20-600 o C 41.2%@13.5 nm 0.26 nm 10 0 12,5 13,0 13,5 14,0 Wavelength, nm V.V. Kondratenko et. al: Thermal stability of soft X-ray / and 2 / multilayer mirrors, Appl. Opt, 1993. Seite 17
Reflectivity, % Enhanced thermal stability of /C//C mirrors 50 40 30 [/C(0.8 nm)//c(0.8nm)] 60 60 T = 500 C [/C//C] 60 + q = 1.5 As-deposited =1 h =10 h =100 h As-dep. 500 o C, 100h 20 10 10 nm 10 nm 0 12,5 13,0 13,5 14,0 Wavelength, nm [/C(0.8)//C(0.8)] 60 R = 60% @ 13.5 nm but T < 400 o C Seite 18
Reflectivity, % / multilayers with enhanced thermal stability 60 50 [/X//X] 60 [/X///X] 60 + = 100 hours q = 1.5 As-deposited T = 400 C T = 500 C T = 600 C As-deposited 600 o C, 100h 40 30 20 10 0 12,9 13,2 13,5 13,8 14,1 Wavelength, nm 2 nm 2 nm Combination: R 60.0% @ 13.5 nm & T 600 o C Seite 19
reflectance Reflectance of the first high-temperature collector optics 2005 60% 50% 40% 30% 20% r = 40mm r = 50mm r = 60mm r = 70mm r = 80mm r = 90mm r = 100mm r = 110mm r = 120mm r = 130mm 10% 0% 12.9 13.1 13.3 13.5 13.7 13.9 14.1 wavelenght, nm Combination: R 56.0% @ 13.5 nm & T 600 o C Seite 20
Current LPP collector coating challenges 2012 Rs > 65 % = (13.5 ± 0.03) nm Dd = 0.015 nm = 15 pm Diameter: Lens sag: Tilt: Weight: > 660 mm > 150 mm > 45 deg > 40 kg Seite 21
Current state and prospects for enhanced thermal stability Experimental reflectivity & thermal stability of -based multilayers: Further prospects: Our limits: System T max, o C R max, % FWHM, nm / < 100 69 0.51 2 C/ 350 67 0.48 /C//C 400 60 0.50 2 / 600 42 0.26 /X//X 600 60 0.47 Development of HT multilayers mainly focuses on the improvement of optical performance. Development of HT- multilayers with T max > 600 o C is limited by temperature of crystallization of -layers (T cryst. ~ 650 o C). Seite 22
Outline Introduction High-reflective / MLs High-temperature -based MLs Radiation stable / MLs B-based MLs for future generation of EUVL Summary and outlook Seite 23
Reflectivity, % Degradation of -capped / multilayer mirrors Industry goal (imaging optics ): EUV intensity: I EUV ~ 10 mw/mm 2 70 60 [/] 60 + P H2O = 5x10-7 Torr D EUV = 165J/mm 2 unexp. exposed Pressure: P H2O < 10-7 Torr 50 ~10 hours R p = 68.0% P HC < 10-9 Torr 40 R p = 45.6% Reflectivity loss: D R < 1% (over 30.000 hours) 30 20-22.7 % 10 0 12,8 13,0 13,2 13,4 13,6 13,8 14,0 Damage caused by two main mechanisms: Wavelength, nm potentially irreversible surface oxidation demonstrably reversible surface contamination (mainly carbon growth) Seite 24
Development of oxidation resistant capping layers 1999-2005 2005-2010 Ru TiO 2 /Nb 2 O 5 [/] 59 [/] 59 Substrate Substrate ~ 10 3 ~ 10 4 Three capping layer conceptions: 1) Low-oxidation materials: (Ru, Au, Pt ) 2) Stable oxides: (TiO 2, Nb 2 O 5, ZrO 2 ) 3) Multilayer conception: more than two layers 10 nm Thickness < 2.0 nm Good optical properties Chemically inert to oxygen Amorphous structure Compatible with / stack Seite 25
Reflectivity, % Normalized reflectivity (R/Ro) Reflectivity and temporal stability of TiO 2 -capped mirrors 70 PTB [//B 4 C] 60 +TiO 2 without barriers with barriers 60 t B4 ~ 0.4 nm C t TiO2 ~ 1.5 nm R p = 66.9% ( = 13.44 nm) 50 FWHM = 0.48 nm 40 + 1.6 % R p = 68.50 % ( = 13.42 nm) FWHM = 0.51 nm 1,000 0,995 0,990 0,985 O2 TiO2 Ru 30 20 10 0 12,8 13,0 13,2 13,4 13,6 13,8 14,0 Wavelength, nm 0,980 0,975 0,970 : - 0.2% TiO 2 : - 0.3% Ru: - 2.6% 0,965 0 20 40 60 80 100 120 140 160 180 Time after deposition, days R = 68.5 % @ 13.42 nm Reflectivity loss of 0.3% within 6 months Seite 26
Reflectivity/% Enhanced radiation stability of TiO 2 - and Nb 2 O 5 -capped mirrors Synchrotron exposure set-up: 68.0 67.0 Synchrotron: SAGA-LS BL18 Water pressure: 7*10-6 mbar EUV intensity: ~ 8.0 W/cm 2 Total dose: 150 kj /cm 2 66.0 65.0 64.0 63.0 62.0 61.0 60.0 P H2O = 7*10-6 mbar TiO2 Nb2O5 Ru 0 50000 100000 150000 Nb 2 O 5 and TiO 2 : Ru: Dose/(J/cm 2 ) no degradation degradation by surface oxidation S. Yulin et al., EUVL Symposium, Prague, 2009 Seite 27
New Exposure Test Stand (ETS) & Xe-gas discharge ETS requirements: Source: Repetition rate: In-band power: DPP / Xe 4 khz 50W/2 Background pressure 10-8 Torr Contaminant inlet 10-2 Torr Pressure control in-situ EUV-intensity > 25 mw/cm 2 Irradiation area > 15x15 mm 2 EUV intensity 5% Exposure Chamber (P < 10-8 Torr) 200 mm R > 67% Condenser Chamber (P ~ 10-2 Torr ) multaneous exposure up to 4 samples is possible with similar dose ( 5%) EUV-source Seite 28
Further prospects for enhanced radiation stability Currently developed capping layers (for imaging optics): Material Rmax, % Oxidation resistance Carbon cleaning by EUV Synchrotron Pulsed O 2 H 2 Ru 68.7 TiO 2 68.5 Nb 2 O 5 68.5 Further studies/prospects: Our limits: Seite 29 Developed capping layer materials should be tested for EUVL optics Application of new multilayer capping layer systems Considerable reflectivity losses of EUVL optics due to application of oxide capping layer with d >> 2.0 nm (collector optics) Capabilities for lifetime studies with high-radiation power are still limited
Outline Introduction High-reflective / MLs High-temperature -based MLs Radiation stable / MLs B-based multilayers for future generation of EUVL Summary and outlook Seite 30
Maximum reflectivity, % Next generation of lithography (NGL) Wavelength selection 80 70 /Be / Reflectivity requirements: (R > 70%) 60 Cr/Sc Co/C La/B /Sr / 12.4 nm 50 40 /Be 11.4 nm 30 20 10 La/B 6.67 nm 0 3 4 5 6 7 8 9 10 11 12 13 14 Wavelength, nm Seite 31
Maximum reflectivity, % Theoretical reflectivity at 6.7 nm Reflectivity of La/B vs / multilayers 80 Henke, 1991 = 5 deg [La/B] 300 [/] 60 / La/B Issue 60 R = 79.4%@ 6.7 nm FWHM = 0.058 nm R,% 76.1 79.5 ~ 40 R = 76.2%@ 13.5 nm FWHM = 0.6 nm d, nm 6.95 3.4 x 2 N 60 200 x 3 20 D, nm 0.60 0.07 x10 0 6 7 12 13 14 15 Wavelength, nm La/B multilayer mirrors are extremely narrowband (FWHM = 0.07 nm) Higher reflectivity losses can be predicted due to interface defects (d = 3.4 nm) Seite 32
Experimental reflectivity at 6.7 nm (2012) Champion data is 49.8% @ 6.66 nm (Rigaku, Japan ) Reflectivity level > 45 % was shown by few other multilayer suppliers Y. Platonov et al: Status of multilayer coatings for EUVL: EUVL Workshop 2011, Maui, Hawaii. Seite 33
La and B are not the best materials for multilayer mirrors 1) Polymorphismus in La: α- La = 2.46 g/cm 3 - La = 2.35 g/cm 3 - La = 2.52 g/cm 3 La-layer is not stable 2) Three lanthanum boride: LaB 4, LaB 6 LaB 9 Considerable diffuse intermixing 3) La and B have extreme oxidation Considerable surface oxidation Currently only /B 4 C (not La/B) are commercially used for XRF!!! Seite 34
Maximum reflectivity, % Replacement of boron by boron carbide (B 4 C) 80 70 60 Henke, 1991 = 6.7 nm = 5 deg La/B La/B 4 C La/BN System R, % FWHM, nm La/B 79.5 0.064 50 40 R = 79.5%@ 0.064 nm R = 74.0%@ 0.058 nm R = 57.2%@ 0.046 nm La/B 4 C 74.1 0.058 30 La/BN 57.2 0.046 20 10 0 6,55 6,60 6,65 6,70 6,75 6,80 Wavelength, nm Transition from La/B to La/B 4 C results in reflectivity loss of 5.4 % @ 6.7 nm. Seite 35
Reflectivity, % Reflectivity, % B 4 C-based multilayers with La-, Ru- and -absorber layers Theory Experiment 80 70 60 Henke, 1991 = 6.7 nm = 5 deg N = 300 La/B 4 C /B 4 C Ru/B 4 C 80 70 60 PTB, Berlin H = 3.38 nm N = 200 = 10 deg La/B 4 C /B 4 C Ru/B 4 C 50 40 30 R = 74,1% FWHM = 0.058 nm R = 58.7% FWHM = 0.055 nm R = 54.1% FWHM = 0.047 nm 50 40 30 R = 32,1% @ 6.67 nm FWHM = 0.036 nm R = 26.0% @ 6.68 nm FWHM = 0.039 nm R = 18.0% @ 6.70 nm FWHM = 0.039 nm 20 20 10 10 0 6,55 6,60 6,65 6,70 6,75 6,80 Wavelength, nm Target of this study: interface performance temporal stability thermal stability Seite 36 0 6,55 6,60 6,65 6,70 6,75 6,80 Wavelength, nm La/B 4 C: R = 32.1% @ 6.67 nm R exp /R th. ~ 0.42 0.44 ( ~ 0.8 nm) S. Yulin et al. Reflective optics for next generation lithography, EUVL Symposium, Miami (2011)
Summary and outlook Remarkable progress has been made in the field of / multilayer mirrors for EUVL application Interface engineering (barriers, capping layers ) was successfully used for: improvement of optical properties: //C: //C: enhanced thermal stability: R = 69.6% @ 13.5 nm R = 69.1% @ 13.4 nm /C//C R = 60.0% & T 400 C /X//X R = 60.0% & T 600 C enhanced radiation stability: //B 4 C +TiO 2 //B 4 C + Nb 2 O 5 R = 68.5% @ 13.5 nm R = 68.5% @ 13.5 nm Lifetime of EUVL optics is still actual issue for multilayer community Seite 37
Acknowledgements Support: EUV: TEM: IOF: Cymer Inc. and Intel Cor. Ch. Laubis, Ch. Buchholz, J. Puls, A. Barboutis, Ch. Stadelhoff, M. Biel, B. Dubrau and F. Scholze (PTB Berlin) U. Kaiser (University, Ulm) M. Scheler, T. Müller, St. Schulze (IOF) Seite 38
Thanks for your attention! Seite 39