Silicon carbide thin films for EUV application deposited

Silicon carbide thin films for EUV application deposited by means of Pulsed Laser Deposition (PLD) Gianni Monaco LUXOR-IFN Laboratory COST meeting Krakow, May 2010 COST-STSM-MP 0601 -05393 Report gmonaco@dei. unipd. it

Outline Si. C Carbides in the EUV; q PLD systems; q Si. C thin films deposited at IOP-WAT Warsaw with Excimer laser; q Analysis of the films; q Some Si. C thin film deposited at LUXOR with Nd-YAG. q Gianni Monaco COST meeting Krakow, 26 may 2010 2/24

List of pubblications on Si. C and PLD materials @LUXOR lab. q P. Nicolosi, D. Garoli, M. G. Pelizzo, V. Rigato, A. Patelli, and F. Rigato, “VUV reflectance measurements and optical constants of Si. C thin films” J. Electron Spectrosc. Relat. Phenom. 144– 147, (2005). q D. Garoli, G. Monaco, F. Frassetto, M. G. Pelizzo, P. Nicolosi, L. Armelao, V. Matterello, V. Rigato “Thin film and multilayer coating development for the extreme ultraviolet spectral region” “Radiation Physics and Chemistry”, 75 (11), p. 1966 -1971, Nov 2006 q G. Monaco, D. Garoli, R. Frison, V. Mattarello, P. Nicolosi, M. G. Pelizzo, V. Rigato, L. Armelao, A. Giglia, S. Nannarone, “Optical constants in the EUV soft x-ray (5÷ 152 nm) spectral range of B 4 C thin films deposited by different deposition techniques”, Proceedings SPIE, 6317, (2006). q D. Garoli, F. Frassetto, G. Monaco, P. Nicolosi, M. -G. Pelizzo, F. Rigato, V. Rigato, A. Giglia, S. Nannarone, “Reflectance measurements and optical constants in the extreme ultraviolet-vacuum ultraviolet regions for Si. C with a different C/Si ratio” Appl. Opt. 45(22) (2006) 5642 -5650. q Gianni Monaco, M. Gastaldi, P. Nicolosi, M. G. Pelizzo, E. Gilioli, S. Rampino, S. Agnoli, G. Granozzi and N. Manuzzato, “Silicon carbide thin films for EUV and soft X-ray application” Eur. Phys. J. ST 169 -1 (2009). q Gianni Monaco, M. Suman, M. G. Pelizzo, P. Nicolosi, “Optical constants of silicon carbide deposited with emerging PVD techniques”, Proc. SPIE Vol. 7360 (2009). Gianni Monaco COST meeting Krakow, 26 may 2010 3/24

Carbides material CVD Si. C Hotpressed B 4 C Mo Al Si Zerodur® Density, ρ (kg m– 3 ) x 10 -3 3. 21 2. 52 10. 3 2. 7 1. 85 2. 55 Coefficient of thermal expansion, α (K– 1 x 10– 6) 2. 4 5. 6 5. 4 25. 0 11. 4 0. 15 Specific heat, C (J kg – 1 K– 1) 700 950 250 899 1880 820 Termic conductivity, κ (W m– 1 K– 1) 200 30 - 42 134 237 216 6. 0 Young's Modulus , E (GPa) 466 450 -470 250 76 303 90 Hardness KH (kg mm-2) 2480 material properties 275 • low density • high melting point • low expansion coefficient • can be polished to lower roughness than metals Gianni Monaco COST meeting Krakow, 26 may 2010 4/24

Reflecting materials for EUV q Presence of a great number of atomic resonance q radiation absorbed on very short distances q complex refractive index: n=1 - +ik (n=1 - ) q Fresnel normal incidence reflectance: R= (1 -n)/(1+n) 2 =( 2 + k 2 )/4 q Below 30 nm (Soft X-ray) , k « 1 =>R < 10 -4 q optics must be used at grazing incidence in order to take advantage of total reflection q Multilayers optics (ML’s) Gianni Monaco COST meeting Krakow, 26 may 2010 5/24

Deposition of Silicon Carbide thin films for EUV Si atom • CVD-techniques • • monocrystalline β-Si. C with T≈ 1400°C high normal incidence reflectance (R > 40% for λ < 60 nm) good stability C atom 9Å 8. 1 3. 08Å Fernandez-Perea et al. Proc. SPIE 6317, (2006) • Sputtering techniques (Ion beam, or magnetron) • • • worse performances than CVD-Si. C only amorphous Si. C Suitable for multilayer lower temperature lower cost Reflectivity degradation Gianni Monaco COST meeting Krakow, 26 may 2010 Larruquert et al. , Appl. Opt. 39 (2000); J. B. Kortright and D. L. Windt Appl. Opt. 27, 2841– 2846 (1988) 6/24

Si. C deposition techniques/2 q HOW TO OBTAIN HIGH REFLECTIVE Si. C AT LOWER TEMPERATURE THAN CVD PROCESS? q. Plasma Enhanced-CVD (R. A. M. Keski-Kuha, Appl. Opt. 27 (1988). q. With Pulsed Laser Deposition at around 800 °C is possible to obtain a crystalline Si. C. Pelt et al. Thin Solid Films, 371 (2000). Let’s try Pulsed deposition techniques (as PLD)! Gianni Monaco COST meeting Krakow, 26 may 2010 7/24

PLD deposition systems Features: q. Very high heating rate of the target surface (108 K/s ). qdeposition of crystalline film demands a much lower substrate temperature qstoichiometry of the target can be retained Particulate generation q can be connected to two macroscopic processes: exfoliational and hydrodynamical-sputtering. q Related to the laser parameters: wavelength, fluence and pulse duration q The particulate content decreases with the wavelength Gianni Monaco COST meeting Krakow, 26 may 2010 8/24

PLD Deposition facility at the MUT Gianni Monaco COST meeting Krakow, 26 may 2010 9/24

Experimental Set-up q. Silicon Carbide β-Si. C (crystalline) target q. Substrates q Single Crystal Sapphire orientated on the 0001 C-plane (for heteroepytaxial grow) q. Si (111) (for heteroepytaxial grow) q. Si (100) (for further analysis) Gianni Monaco COST meeting Krakow, 26 may 2010 10/24

Deposited samples Sample Substrate Temp (°C) RF etching (min) 1 Si RT 2 Si RT Freq (Hz) Laser energy 1 133. 5 m. J Distance (mm) Atmosphere Time (min) 15 Base pressure Fluence (torr) (J/cm 2) -5 2. 07 x 10 1. 3 - Vacuum 30 15 2. 07 x 10 -5 1 - Vacuum 30 1. 3 133. 5 m. J 15 2. 9 Si 2 1 200 60 rate, q 3 Frequency: even. RTif an higher repetition rate would have resulted in -an higher Vacuum deposition 15 2. 9 x 10 4 we chose Sap a rate. RT 2 Our goal 1 was 200 of 1 Hz for all the deposition. to get a -crystalline, Vacuum hence 60 15 3 x 10 5 Si RT 3 1 120 74 (ca) Vacuum 90 organized, structure and this could be better accomplished if the atoms on the substrate 15 3 x 10 6 RT 1 120 74 (ca) Vacuum 90 surface. Sihave longer time. No intervals in order to 31. 3 organize themselves. 1 x 10 7 538 1 137 74 (ca) Vacuum 90 No 1 x 10 q 8 Laser fluence: laser was chosen very low. The threshold of. Vacuum Silicon Carbide Sapp 538 fluence 1. 3 1 deposition 137 74 (ca) 90 2 10 min 1 x 10 9 with excimer Si 800 @192 nm is 1 J/cm and we 1. 3 choose 1 to be 135. 6 45 laser around 80 that value. Vacuum to get less (10 mbar Ar) particulate and 900 give raise to a slower process. The two deposition carried 75 at 3 30 min 1 x 10 crystallization 10 Si 1. 3(ca) 1 148 80 Vacuum 2 (10 mbar Ar)used to locate the plume position and direction inside the J/cm (sample 5 and 6) where 30 min 4. 5 x 10 -1 x 10 11 chamber. Sapp 900 1. 3(ca) 1 148 80 Vacuum 75 (10 mbar Ar) min 4. 5 x 10 were -1 x 10 q 12 Substrates: Sapphire used for 1 two reasons: they have low lattice 120 Sapp Silicon 930 (111)30 and 1. 3 138 80 Vacuum (10 mbar Ar) mismatch with 3 C-Si. C and can be suitable for 1. 3 heteroepitaxial growth (3 C-Si. C has 4. 36 Å, 120 30 min 3. 5 x 10 -8 x 10 13 Sapp 930 1 138 80 Vacuum Sapphire 4. 75 Å on its face [0001], Si (111) has 9. 23 Å) , while Silicon (100) (cubic, lattice (10 mbar Ar) 30 min 3. 5 x 10 14 constant Si(111) 930 1. 77 for successive 1 185 characterization 80 Vacuum 5, 43 Å) is mainly used as a test-8 x 10 sample such as 120 film (10 mbar Ar) thickness and composition. 30 min 3. 5 x 10 -8 x 10 15 Sapp 930 1. 77 1 185 80 Vacuum 120 (10 mbar Ar) q 16 Temperature: the temperature is another crucial parameter in our process. As. Vacuum said in the 30 min 3. 5 x 10 -8 x 10 Sapp 930 1. 77 1 185 80 120 previous document in which the project has been exposed, the crystalline CVD silicon carbide (10 mbar Ar) 30 min 3. 5 x 10 17 is obtained Si(111) at a 930 148 we are 80 trying to. Vacuum 90 temperature as high as-6 x 10 1400 1. 45 °C, but 1 with PLD demonstrate (10 mbar Ar) that it is possible 930 to obtain structure at lower temperature. We planned to keep 30 minthe same 3. 5 x 10 -6 x 10 18 Sapp 1. 45 1 148 80 Vacuum 90 (10 mbararound Ar) the deposition temperature 900° C for all the samples to help the crystalline growth. 30 min 3. 5 x 10 -6 x 10 930 148 80 q 19 For the. Sapp last three samples 20, Ar)21, 22 we tried 1. 45 to help 1 film crystallization by use. Vacuum of Ar 90 (10 mbar min 5 x 10 keeping a 30 mild temperature of 1. 7650° C. 1 20 bombardment Sapp ca 650 185 43 Ar (6 x 10 -3 120 x 10 -5 -5 -4 -4 -4 -2 -5 -4 -5 -5 -5 -5 -2 -2 -2 -5 (10 -2 mbar Ar) 21 Sapp ca 650 30 min (10 -2 mbar Ar) 5 x 10 -5 22 Si(111) ca 650 30 min 5 x 10 -5 -2 Gianni Monaco COST (10 mbar Ar) 1. 7 1 185 meeting Krakow, 26 may 2010 43 11/24 43 torr) Ar (6 x 10 -3 torr) 120

AFM analysis of the deposited samples Sapphire substrate Sample n° 8 Sample n° 4 Sample n° 5 Sample n° 15 Gianni Monaco COST meeting Krakow, 26 may 2010 Sample n° 18 12/24

Samples thicknesses q Silicon Carbide has an absorption dip centered at 795 cm− 1 that could be ascribed to TO-phonon mode of Si. C in its cubic or hexagonal phase 880 cm-1 825 cm-1 Gianni Monaco COST meeting Krakow, 26 may 2010 13/24

Samples thicknesses Samples number 5 and 6 (high fluence=of 3 J/cm 2 90 min @RT) Gianni Monaco COST meeting Krakow, 26 may 2010 14/24

SEM images Sample n° 8 Gianni Monaco COST meeting Krakow, 26 may 2010 Sample n° 19 15/24

Sapphire 006 Sapphire 1 1 -2 0 Sapphire 003 XRD spectra of the samples q. Peak @ 43° 21° and 38° are due to the Sapphire substrate. q 3 spectra show different features: Si. C 8, Si. C 12 and Si. C 19 (Sapphire peaks disappear) The feature of these spectra, as retrieved in the Instrument database (ICCD-JCPDS), cannot be attributed to any of the Si. C crystalline structure. Gianni Monaco COST meeting Krakow, 26 may 2010 16/24

EUV Reflectance measurements q q q Source: hollow cathode or spectral lamps (40 -500 nm) Monochromator: Johnson Onaka – normal incidence Detector: Channel Electron Multiplier or photomultiplier Sample and detector on manual stages Polarization factor known (from 121. 6 to 40 nm) Gianni Monaco COST meeting Krakow, 26 may 2010 17/24

EUV Reflectance Gianni Monaco COST meeting Krakow, 26 may 2010 18/24

EUV Reflectance /2 Gianni Monaco COST meeting Krakow, 26 may 2010 19/24

Conclusions of the STSM q Hard to find evidences of 3 C-Silicon Carbide! I. The films are crystalline but are simply too thin to be revealed with the utilized techniques. This could sound strange if we think that we placed the substrates in the position of the sample n° 5 which was demonstrated to have a thickness of 60 nm. Since we have kept the same target-substrate distance, the explanation surely lies in the laser fluence which is three times higher compared to the other samples. II. The samples are not crystalline, films are too thin and we cannot see any film by the utilized techniques (such as IR which is not sensitive to amorphous structure). III. Contamination of the surface, due to the residual atmosphere and to the clamp (made in stainless steel) prevails with respect to deposited films. Hence, it is difficult to see the IR absorption and the XRD spectra since the crystalline structure has not been formed, or some other structure have been formed instead crystalline Si. C. q A TEM analysis could probably help to solve the first and the second uncertainties, while an XPS could be helpful for the third uncertainties. Nevertheless, with the results we obtain we can exclude third supposition since the reflectance it is not affected by the presence of absorbing elements, such as Oxygen, that would lower the Reflectance yield compared to the substrate. Gianni Monaco COST meeting Krakow, 26 may 2010 20/24

Appendix: further deposition@ LUXOR Bz • Laser: Nd: YAG (λ = 1064 nm) with variable repetition rate Permanent magnet and 6 ns • Incidence angle of 45° on the target 2 1 • P ≈ 8. 7 x 10 -7 mbar 3 4 • Magnetic field intensity 5 on target: 100 – 200 Gauss • Variable target-permanent magnet distance substrates • Can guest more than one target • Ceramic heater (up to 1500 °C depending by the vacuum) • Five samples deposited (different position relative to the target) Gianni Monaco COST meeting Krakow, 26 may 2010 21/24

Appendix: further deposition@ LUXOR/2 Si (111) Si(100) Fluence 1. 4 J/cm 2, T=650°C, 10 Hz or 2 Hz Repetition rate Gianni Monaco COST meeting Krakow, 26 may 2010 22/24 Sapphire

Appendix: further deposition@ LUXOR/3 Gianni Monaco COST meeting Krakow, 26 may 2010 23/24

Acknowledgments q Institute of Optoelectronics Prof Henryk Fiedorowicz , Dr. Waldemar Mróz, Artur Prokopiuk, Michael L. Korwin-Pawlowski and Sylwia Burdyńska, Boguslaw. Budner q LUXOR-INF Laboratory Prof. Piergio Nicolosi, Dr. Suman Michele, Dr. Maria G. Pelizzo, Dr. Zuppella Paola q Dr. Garoli Denis and Dr. Natali Marco for SEM and XRD measurements q COST project q Thank you for your attention! Gianni Monaco COST meeting Krakow, 26 may 2010 24/24