Silicon carbide thin films for EUV application deposited by means of Pulsed Laser Deposition (PLD)

1
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-MP0601-05393 Report gmonaco_at_dei.unipd.i
  t

2
Outline

• SiC Carbides in the EUV
• PLD systems
• SiC thin films deposited at IOP-WAT Warsaw with
  Excimer laser
• Analysis of the films
• Some SiC thin film deposited at LUXOR with
  Nd-YAG.

3
List of pubblications on SiC and PLD materials
_at_LUXOR lab.

• P. Nicolosi, D. Garoli, M. G. Pelizzo, V. Rigato,
  A. Patelli, and F. Rigato, VUV reflectance
  measurements and optical constants of SiC thin
  films J. Electron Spectrosc. Relat. Phenom.
  144147, (2005).
• 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
• 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 (5152 nm)
  spectral range of B4C thin films deposited by
  different deposition techniques, Proceedings
  SPIE, 6317, (2006).
• 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 SiC
  with a different C/Si ratio Appl. Opt. 45(22)
  (2006) 5642-5650.
• 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).
• Gianni Monaco, M. Suman, M. G. Pelizzo, P.
  Nicolosi, Optical constants of silicon carbide
  deposited with emerging PVD techniques, Proc.
  SPIE Vol. 7360 (2009).

4
Carbides material
material properties CVD SiC Hot-pressed B4C Mo Al Si Zerodur
Density, ? (kg m3 ) x 10-3 3.21 2.52 10.3 2.7 1.85 2.55
Coefficient of thermal expansion, a (K1x106) 2.4 5.6 5.4 25.0 11.4 0.15
Specific heat, C (J kg 1 K1) 700 950 250 899 1880 820
Termic conductivity, ? (W m1 K1) 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 275

• low density
• high melting point
• low expansion coefficient
• can be polished to lower roughness than metals

5
Reflecting materials for EUV

• Presence of a great number of atomic resonance
• ?radiation absorbed on very short distances
• complex refractive index n1-?ik (n1- ?)
• Fresnel normal incidence reflectance
  R?(1-n)/(1n)?2 (?2 k 2 )/4
• Below 30 nm (Soft X-ray) ?, k 1 gtR lt 10-4
• ? optics must be used at grazing incidence in
  order to take advantage of total reflection
• ? Multilayers optics (MLs)

6
Deposition of Silicon Carbide thin films for EUV

• CVD-techniques
• monocrystalline ß-SiC with T 1400C
• high normal incidence reflectance
• (R gt 40 for ? lt 60nm)
• good stability
• Sputtering techniques(Ion beam, or magnetron)
• worse performances than CVD-SiC
• only amorphous SiC
• Suitable for multilayer
• lower temperature
• lower cost
• Reflectivity degradation

1.89Å
3.08Å
Fernandez-Perea et al. Proc. SPIE 6317, (2006)
Larruquert et al.,Appl. Opt. 39 (2000) J. B.
Kortright and D. L. Windt Appl. Opt. 27,
28412846 (1988)
7
SiC deposition techniques/2

• HOW TO OBTAIN HIGH REFLECTIVE SiC AT LOWER
  TEMPERATURE THAN CVD PROCESS?
• Plasma Enhanced-CVD (R.A.M. Keski-Kuha, Appl.
  Opt. 27 (1988).
• With Pulsed Laser Deposition at around 800 C is
  possible to obtain a crystalline SiC. Pelt et al.
  Thin Solid Films, 371 (2000).
• Lets try
  Pulsed deposition
  techniques (as PLD)!

8
PLD deposition systems

• Features
• Very high heating rate of the target surface (108
  K/s ).
• deposition of crystalline film demands a much
  lower substrate temperature
• stoichiometry of the target can be retained
• Particulate generation
• can be connected to two macroscopic processes
  exfoliational and hydrodynamical-sputtering.
• Related to the laser parameters wavelength,
  fluence and pulse duration
• The particulate content decreases with the
  wavelength

9
PLD Deposition facility at the MUT
10
Experimental Set-up

• Silicon Carbide ß-SiC (crystalline) target
• Substrates
• Single Crystal Sapphire orientated on the 0001
  C-plane (for heteroepytaxial grow)
• Si (111) (for for heteroepytaxial grow)
• Si (100) (for further analysis)

11
Deposited samples
Sample Substrate Temp (C) RF etching (min) Base pressure(torr) Fluence (J/cm2) Freq (Hz) Laser energy Distance (mm) Atmosphere Time (min)
1 Si RT 15 2.07x10-5 1.3 1 133.5 mJ - Vacuum 30
2 Si RT 15 2.07x10-5 1.3 1 133.5 mJ - Vacuum 30
3 Si RT 15 2.9 x10-5 2 1 200 - Vacuum 60
4 Sap RT 15 2.9 x10-5 2 1 200 - Vacuum 60
5 Si RT 15 3 x10-5 3 1 120 74 (ca) Vacuum 90
6 Si RT 15 3 x10-5 3 1 120 74 (ca) Vacuum 90
7 Si 538 No 1x10-4 1.3 1 137 74 (ca) Vacuum 90
8 Sapp 538 No 1x10-4 1.3 1 137 74 (ca) Vacuum 90
9 Si 800 10 min (10-2 mbar Ar) 1x10-4 1.3 1 135.6 80 Vacuum 45
10 Si 900 30 min (10-2 mbar Ar) 1x10-4 1.3(ca) 1 148 80 Vacuum 75
11 Sapp 900 30 min (10-2 mbar Ar) 4.5x10-5-1x10-4 1.3(ca) 1 148 80 Vacuum 75
12 Sapp 930 30 min (10-2 mbar Ar) 4.5x10-5-1x10-4 1.3 1 138 80 Vacuum 120
13 Sapp 930 30 min (10-2 mbar Ar) 3.5x10-5-8x10-5 1.3 1 138 80 Vacuum 120
14 Si(111) 930 30 min (10-2 mbar Ar) 3.5x10-5-8x10-5 1.77 1 185 80 Vacuum 120
15 Sapp 930 30 min (10-2 mbar Ar) 3.5x10-5-8x10-5 1.77 1 185 80 Vacuum 120
16 Sapp 930 30 min (10-2 mbar Ar) 3.5x10-5-8x10-5 1.77 1 185 80 Vacuum 120
17 Si(111) 930 30 min (10-2 mbar Ar) 3.5x10-5-6x10-5 1.45 1 148 80 Vacuum 90
18 Sapp 930 30 min (10-2 mbar Ar) 3.5x10-5-6x10-5 1.45 1 148 80 Vacuum 90
19 Sapp 930 30 min (10-2 mbar Ar) 3.5x10-5-6x10-5 1.45 1 148 80 Vacuum 90
20 Sapp ca 650 30 min (10-2 mbar Ar) 5x10-5 1.7 1 185 43 Ar (6x10-3 torr) 120
21 Sapp ca 650 30 min (10-2 mbar Ar) 5x10-5 1.7 1 185 43 Ar (6x10-3 torr) 120
22 Si(111) ca 650 30 min (10-2 mbar Ar) 5x10-5 1.7 1 185 43 Ar (6x10-3 torr) 120

• Frequency even if an higher repetition rate
  would have resulted in an higher deposition rate,
  we chose a rate of 1 Hz for all the deposition.
  Our goal was to get a crystalline, hence
  organized, structure and this could be better
  accomplished if the atoms on the substrate
  surface have longer time intervals in order to
  organize themselves.
• Laser fluence laser fluence was chosen very low.
  The deposition threshold of Silicon Carbide with
  excimer laser _at_192 nm is 1 J/cm2 and we choose to
  be around that value to get less particulate and
  give raise to a slower crystallization process.
  The two deposition carried at 3 J/cm2 (sample 5
  and 6) where used to locate the plume position
  and direction inside the chamber.
• Substrates Silicon (111) and Sapphire were used
  for two reasons they have low lattice mismatch
  with 3C-SiC and can be suitable for
  heteroepitaxial growth (3C-SiC has 4.36 Å,
  Sapphire 4.75 Å on its face 0001, Si (111) has
  9.23 Å) , while Silicon (100) (cubic, lattice
  constant 5,43 Å) is mainly used as a test sample
  for successive characterization such as film
  thickness and composition.
• Temperature the temperature is another crucial
  parameter in our process. As said in the previous
  document in which the project has been exposed,
  the crystalline CVD silicon carbide is obtained
  at a temperature as high as 1400 C, but with PLD
  we are trying to demonstrate that it is possible
  to obtain the same structure at lower
  temperature. We planned to keep the deposition
  temperature around 900 C for all the samples to
  help the crystalline growth.
• For the last three samples 20, 21, 22 we tried to
  help film crystallization by use of Ar
  bombardment keeping a mild temperature of 650 C.

12
AFM analysis of the deposited samples
Sapphire substrate Sample
n4 Sample n5
Sample n8
Sample n15 Sample n18
13
Samples thicknesses

• Silicon Carbide has an absorption dip centered at
  795 cm-1 that could be ascribed to TO-phonon mode
  of SiC in its cubic or hexagonal phase

880 cm-1
825 cm-1
14
Samples thicknesses
Samples number 5 and 6 (high fluenceof 3 J/cm2
90 min _at_RT)
15
SEM images
Sample n8
Sample n19
16
XRD spectra of the samples
Sapphire 006
Sapphire 1 1 -2 0
Sapphire 003

• Peak _at_ 43 21 and 38 are due to the Sapphire
  substrate.
• 3 spectra show different features SiC8, SiC12
  and SiC19 (Sapphire peaks disappear)The feature
  of these spectra, as retrieved in the Instrument
  database (ICCD-JCPDS), cannot be attributed to
  any of the SiC crystalline structure.

17
EUV Reflectance measurements

• 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)

18
EUV Reflectance
19
EUV Reflectance /2
20
Conclusions of the STSM

• Hard to find evidences of 3C-Silicon Carbide!
• 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 n5
  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.
• 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).
• 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
  SiC.
• 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 the 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.

21
Appendix further deposition_at_ LUXOR

• Laser NdYAG (? 1064 nm) with variable
  repetition rate
• and 6ns
• Incidence angle of 45 onthe target
• P 8.7 x 10-7 mbar
• Magnetic field intensity on target 100 200
  Gauss
• Variable target-permanent magnet distance
• 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)

Permanent magnet
2
1
3
4
5
substrates
22
Appendix further deposition_at_ LUXOR/2
Si (111) Si(100)
Sapphire Fluence 1.4 J/cm2, T650C, 10 Hz or 2
Hz Repetition rate
23
Appendix further deposition_at_ LUXOR/3
24
Acknowledgments

• Institute of Optoelectronics
• Prof Henryk Fiedorowicz , Dr. Waldemar Mróz,
  Artur Prokopiuk, Michael L. Korwin-Pawlowski and
  Sylwia Burdynska, BoguslawBudner
• LUXOR-INF Laboratory
• Prof. Piergiorgio Nicolosi, Dr. Suman Michele,
  Dr. Maria G. Pelizzo, Dr. Zuppella Paola
• Dr. Garoli Denis and Dr. Natali Marco for SEM
  and XRD measurements
• COST project
• Thank you for your attention!