Silicon Absorption

Report
Optical absorption in bulk crystalline silicon
as well as in the crystal surfaces
Alexander Khalaidovski1, Jessica Steinlechner2, Roman Schnabel2
1: Institute for Cosmic Ray Research (ICRR)
The University of Tokyo
http://www.icrr.u-tokyo.ac.jp/
Aleksandr Khalaidovski
2: Albert Einstein Institute
Max Planck Institute for Gravitational Physics
Institute for Gravitational Physics of the Leibniz University Hannover
http://www.qi.aei-hannover.de
1
Absorption
bulk crystalline
silicon– and
in the crystal
surfaces
KAGRAinface-2-face
meeting
富山大学
– August
3rd 2013
Outline
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
2
Motivation – Einstein Telescope (ET)
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
3
Motivation – ET Low Frequency Interferometer
 Low frequency interferometer: cryogenic temperature (10 K)
 Conventional fused silica optics no longer usable
 Use crystalline silicon
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
4
Properties of crystalline silicon
 High Q-factor at both room temperature and cryogenic temperatures
Credits: Ronny Nawrodt
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
5
Properties of crystalline silicon
 High Q-factor at both room temperature and cryogenic temperatures
 Available in large diameters (currently about 450mm – 500mm)
Source: http://www.bit-tech.net/hardware/2010/10/20/global-foundries-gtc-2010/4
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
6
Properties of crystalline silicon
 High Q-factor at both room temperature and cryogenic temperatures
 Available in large diameters (currently about 450mm – 500mm)
 Completely opaque at 1064 nm, but ...
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
7
Properties of crystalline silicon
 High Q-factor at both room temperature and cryogenic temperatures
 Available in large diameters (currently about 450mm – 500mm)
 Completely opaque at 1064 nm, but ...
 ... expected to have very low optical absorption at 1550 nm
?
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
8
Properties of crystalline silicon
 High Q-factor at both room temperature and cryogenic temperatures
 Available in large diameters (currently about 450mm – 500mm)
 Completely opaque at 1064 nm, but ...
 ... expected to have very low optical absorption at 1550 nm
 currently chosen as candidate material for ET-LF test masses
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
9
Properties of crystalline silicon
 High Q-factor at both room temperature and cryogenic temperatures
 Available in large diameters (currently about 450mm – 500mm)
 Completely opaque at 1064 nm, but ...
 ... expected to have very low optical absorption at 1550 nm
 currently chosen as candidate material for ET-LF test masses
 we need to confirm low optical absorption at RT and CT
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
10
Optical absorption measurements
at the AEI Hannover
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
11
Photo-thermal self-phase modulation
Lcavity
substrate
Lcavity
Thermal effect
increases with
• Increasing
power
Dr. Jessica
Steinlechner
• Decreasing
scan frequency
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
12
Photo-thermal self-phase modulation
 Absorption leads to a heating of the analyzed substrate and thus (for a
sum of the thermo-refractive index dn/dT and the thermal expansion
coefficient  > 0 ) to a thermally induced optical expansion.
 When the substrate is placed in an optical cavity and the cavity length is
scanned, this thermal expansion affects the detected cavity resonance peaks
in a different way for an increase and a decrease of the cavity length.
 An external increase of the cavity length and the thermally-induced
expansion act in the same direction, resulting in a faster scan over the
resonance and thus in a narrowing of the resonance peak.
 In contrast, an external cavity length decrease and the thermally-induced
expansion partly compensate. As a result, the scan over the resonance is
effectively slower, leading to a broader resonance peak.
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
13
Photo-thermal self-phase modulation
Advantages
 Suitable to measure absorption in bulk and coatings
 High sensitivity (sub-ppm), small error bars
 Does not require high laser power
Drawbacks
 Thermal effect visible not at all laser powers
 Requires a cavity setup around the sample
(can be the sample itself with dielectric coatings)
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
14
More about the method
(Journal: Applied Optics)
(Journal: Applied Optics)
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
15
Silicon absorption at 1550 nm
measurement at a fixed optical power
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
16
Measurement setup
Monolithic Si cavity
 Length 65mm, diameter 100 mm.
 Curved end surfaces, ROC = 1m.
 Specific resistivity 11 kcm (boron)
 Coatings: SiO2/Ta2O5. R = 99.96 %.
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
17
Measurement results are …
α = (264 ± 39) ppm/cm
or 3430 ppm/round trip
Measurement Number
Result of a single Measurement
Aleksandr Khalaidovski
Mean value + error bar
Absorption in bulk crystalline silicon and in the crystal surfaces
18
… much higher than expected
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
19
Measurements by the LMA group
Using beam deflection method
[J. Degallaix, 4th ET symposium, Dec. 2012]
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
20
Silicon absorption at 1550 nm
power-dependent measurements
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
21
Facts about the measurement
 Same monolithic cavity as in previous setup
 Intra-cavity peak intensity: 0.4 W/cm² - 21 kW/cm²
 Impedance-mismatch measurement
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
22
Results
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
23
Discussion
I) Non-linear dependence of absorption on optical intensity
 Results by Degallaix et al. qualitatively confirmed
 Reason: probably two-photon absorption, quantitative analysis in progress
II) Our results are still much higher than the for other groups
 Main differences:
- material purities (difference not too large)
- measurement approach. Our approach is sensitive to absorption in
both the bulk crystal and the surfaces.
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
24
Possible reason
 Surface layer of amorphous silicon
 Literature absorption values: ca. 100/cm – 2000/cm
 High a-Si absorption verified in a different experiment measuring Si/SiO2
dielectric coatings.
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
25
Possible implications
 Absorption contribution of about 800 ppm per surface transmission
 1600 ppm for transmission through input test mass (ITM)
 Absorbed laser power needs to be extracted through the suspensions
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
26
Outlook
 Planned measurements:
- Analysis of samples of different length
- Analysis of samples of different purity, Czochralski and Float Zone
 Analysis of the surfaces in view of a possible layer of amorphous material
 Comparison with other groups, exchange of samples
 Measurements at cryogenic temperatures (Jena)
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
27
Conclusions
 High absorption was found in Si-samples at the AEI
 Such a high absorption contribution is neither expected from the bulk
crystal, nor could it be confirmed by beam deflection measurements
 The absorption probably originates in the crystal surfaces, possibly due to
a layer of amorphous material generated during polishing
 Further measurements are required to clearly separate the bulk and
surface contributions and to evaluate a possible impact on ET
Thank you very much
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
28
Discussion II
(a) Our data
(b) LMA data with added offset of 250 ppm/cm
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
29
Absorption measurement approaches
Power-Measurement
• Power detection before and behind
substrate (photo diode, power
meter,…)
• Simplest absorption measurement
method
• Not very sensitive
Beam-deflection measurement
• Pump beam heats substrate
• Probe beam is deflected by thermal
lens
• Deflection measurement on quadrant
photo diode
• Possible limit: available laser power
Aleksandr Khalaidovski
Absorption in bulk crystalline silicon and in the crystal surfaces
30

similar documents