1 Japan Update EUVA (Extreme Ultraviolet Lithography System Development Association) Koichi Toyoda SOURCE TWG 2 March, 2005 San Jose
Outline 2 EUVA LPP at Hiratsuka R&D Center GDPP at Gotenba Branch Lab. University LPP research at Osaka University (Leading Project)
3
EUV LPP Source Roadmap 4 Fiscal Year 2003 2004 2005 ~2008 EUV power -- 4W 10W 115W @ intermediate focus Plasma target Xe Xe Xe Droplet Conversion efficiency 0.6% 0.7% 0.8% TBD Laser power 1.5kW 2.5kW 5kW TBD EUV power in 2pisr 2%BW 4.0W 9.1W 40W 420W Available collection solid angle -- -- 4sr 5sr Repetition rate 10kHz 10kHz 10kHz TBD Technology for 10W Nd:YAG Laser, Liquid Xe jet Technology for 115W CO 2 Laser, droplet target Magnetic field mitigation
Concept of LPP source by CO 2 laser 5 Collector Mirror Target Chamber Main Laser: CO 2 Laser pulse width: ns Beam Splitter EUV / 13.5nm Pre-Pulse Laser: Nd:YAG Laser pulse width: Sub-ns Xe Target Conversion Efficiency (CE) of 0.6% has been obtained!
LPP source system CO 2 laser 6 Xenon Recirculation System (XRS) MOPA CO 2 Laser 1500 Target Chamber Power Supply 2000 2000 2000 2500 4000 2000 1500 1500 1600 1700 2000
7 Cost estimation of LPP light source Initial Cost Running Cost 0.55~0.82 Initial Cost 5.6~9.0 5.6~9.0 Mill.$ @ CO 0.55~0.82 Mill.$/year @ CO 110 JYen = 1 USD CO (C.E.=1.0~0.5%) Total Component Initial Cost (M$) Laser System 3.7~7.1 EUV Chamber 1.2 Xe Re-Circulation System 0.5 Collector Mirror 0.2 Estimation based on: 21.7~31.7 Mill. $ @ YAG 2.54~3.63 Mill.$/year @ YAG Mill.$ @ CO 2, 21.7~31.7 @ CO 2, 2.54~3.63 CO 2 Running Cost (M$/year) Initial Cost (M$) YAG (C.E.=1.2~0.8%) Running Cost (M$/year) 5.6~9.0 0.55~0.82 21.7~31.7 2.54~3.63 3.7~7.1 0.32~0.43 20.0~30.0 2.29~3.33 42~84kW, 100kHz 35~53kW, 10kHz 1.2 0.03 1.0 0.02 0.5 0.14 0.5 0.14 0.2 0.06~0.22 0.2 0.09~0.14-115W Source Power at I.F. - 100 units produced in 2016. - 120 wafer/hr throughput 21.3 Billion pulse /year @ 10kHz 213 Billion pulse /year @ 100kHz CO 2 Driver laser system for LPP
Xe droplet technology for 115W source 8 Xenon droplet for 115W source Irradiation interval : > 600µm Driver laser frequency : 100kHz Required droplet speed : > 60m/s High speed droplet High pressure Xe supply v = 2 P ρ V : velocity P:Pressure ρ :Density Bernoulli s theorem 5µs Stationary Moving 200µm Irradiated droplet Droplets Velocity :20m/s Frequency:100kHz Interval :202µm 15µs Irradiation Interval 600µm Required Xe pressure : > 5.4MPa
Fast Ion Mitigation for Xenon Plasma 9 Xe Jet Nozzle B Mirror Laser QCM detector EUV Ion Laser Magnetic Coil Magnetic Field Erosion Rate (nm / M pulse) 200 180 160 140 120 100 80 60 40 20 0 0 50 100 150 200 250 300 350 Laser Energy (mj) Erosion rate dependence on magnetic field and laser energy Mirror Life Estimation Experimental Value 115 W Source Item Nd:YAG CO2 Laser Energy (mj) 100 600 Mirror Distance (mm) 95 100 Laser Frequency (Hz) 10 100000 Mirror Life (M pulse) 188 35 Mirror Life with Magnet (M pulse) 5625 1039 Estimated collector mirror lifetime with magnetic mitigation is > 1B pulses.
DPP source development 10 Discharge tube Insulator Our approach Electrode Discharge gas Pulse generator Electrode EUV --Moderate diameter of of capillary. --Relatively low low current compared to to conventional Z-pinch. --Mitigation of of wall wall damage and and effect of of plasma instability. --Small source size. size. Key issues for for DPP Increase of of EUV EUV power at at intermediate focus. Lifetime matter of of discharge head. Debris mitigation.
DPP performance roadmap 11 Metrics Feb. 2004 Jun. 2004 Oct. 2004 Feb. 2005 1Q-2008 EUV emitter Xe Xe Xe Xe Sn TBD EUV power at IF 2.7W 4.8W 8.4W 19W 47W * > 50W EUV emission from primary source 12.6W / 1.45sr 39.7W / 2.1sr 59.3W / 3.1sr 93W / 3.1sr 186W / pi sr ** 200W / 3sr Repetition rate 2kHz 7kHz 7kHz 7kHz (7kHz) 7-10kHz Energy dose stability (1 ) Angular distribution stability (1 ) Mirror lifetime (10% loss) 1.1% 4.9% 1.3% 1.3% 2.4% < 0.1 % 8.2% 7.9% 4.8% 4.8% --- < 5% --- --- --- > 1x10 7 shot --- > 0.5x10 6 sec * Pi sr collector optics is assumed ** Estimated value from low rep.-rate measurement results
Flying Circus EUVA source assessment 12 Flying Circus visit led to to upward adjustment of of results. Energy Energy up up to to 8.8 8.8 mj/sr mj/sr (2%BW) (2%BW)@5 @5 Hz Hz Source Source size size @200 @200 sccm: sccm: φ0.17 φ0.17 x x L0.8 L0.8 mm mm // 80% 80% area area φ0.24 φ0.24 x x L0.82 L0.82 mm mm FOM 10.0 10.0 9.0 9.0 8.0 8.0 in-band EUV (mj / 2 sr) 7.0 6.0 5.0 4.0 3.0 in-band EUV (mj / 2 sr) 7.0 6.0 5.0 4.0 3.0 2.0 2.0 1.0 1.0 0.0 8 9 10 11 12 13 14 15 16 17 High Voltage (kv) 0.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 stored capacitor energy (J) Santi Alonso vd Westen, Caspar Bruineman, Fred Bijkerk, Vivek Bakshi
Performance of primary source 13 Instantaneous EUV power (ESA, 2%BW) [W] 120 100 80 60 40 20 0 moving average (50 pulses) 1 51 101 151 Pulse # Relative intensity 1.4 1.2 1 0.8 0.6 0.4 0.2 0 0 10 20 30 40 50 60 70 80 Angle (deg.) 7kHz 7kHzoperation :: Effective Effective solid solid angle angle = 3.1 3.1 sr sr Average Average EUV EUV power power (2%BW) (2%BW) = 93W 93W // 3.1sr 3.1sr (189W (189W/ / 2 pi) pi) Integrated Integrated energy energy stability stability (50 (50 pulses, pulses, sigma) sigma) = 1.3% 1.3% Angular distribution
Usable power at IF 14 Primary source (d= 0.5, L= 1.56) 28% of source power is collected to I.F. 1mm Source EUV image Collector mirrors (Grazing-incidence) Focus Fabricated by MEDIA LARIO s.r.l Primary source : 93W Collection efficiency : 28% Transmission (gas, shield) : 72% Focus image Grazing-incidence collector with cooling line I.F. power = 19W
Mirror lifetime: reflectivity monitoring 15 Test piece mirror 20 deg. Aperture 30 deg. Photodiode-1 20 deg. Discharge head (Debris shield) Photodiode-2 Reflectivity degradation monitor for grazing-incidence mirror Normalized reflectivity 1.20 1.00 0.80 0.60 0.40 0.20 0.00 without debris shield with debris shield 0.0E+00 2.0E+06 4.0E+06 6.0E+06 8.0E+06 1.0E+07 Number of pulses Mirror lifetime: >1 10 7 shots demonstrated.
Emission and absorption of LPP Sn plasma are well characterized using J- to kj-class lasers. opacity measurement EUV and n e profiles probe Thermal radiation (T R = 50 ev) opacity sample target surface 1.2 low-density SnO 2 1.0 Transmission 0.8 0.6 0.4 T e ~ 30 ev 0.2 0.0 8 10 12 14 Wavelength (nm) T e ~ 0 ev 16 18
We have theoretically investigated optimum conditions of laser wavelength, intensity and pulse duration to lead to the maximum conversion efficiency for tin, xenon and lithium. Sn, 1.2 ns Xe, 5 ns Te (ev) 80 ion density (cm 50-3 ) ion density (cm -3 ) Te [ev] 30 Li, 20 ns 20 10 10 17 10 18 10 19 10 20 ion density [cm -3 ]
18
19 Summary Achieved performance LPP by YAG laser - In-band Power 4.0 W (2%BW) at IF <Estimate> - Conversion Efficiency 0.85 % @ 10Hz (2%BW, 2π2 sr) by CO 2 laser - Conversion Efficiency 0.6 % @ 10Hz (2%BW, 2π2 sr) - Short Pulse 6kW CO 2 laser is under development DPP by Xe target - In-band power ~ autumn, 2005 - Mirror lifetime >1 10 10 7 shot by Sn target - In-band power 186 W (2%BW, 47 W (2%BW) at IF <Estimate> 93 W (2%BW, 3.1 sr) at Primary Source 19 W (2%BW) at IF shot (10% loss) (2%BW, sr) at Primary Source <Estimate> Main target for 2008 : 115 W (intermediate focus)
Acknowledgements 20 This work was performed under the management of the Extreme Ultraviolet Lithography System Development Association, a research and development program of NEDO.