Transcript Input Optic requirements and components for High Power lasers

Adv. LIGO
Input optic requirements and
components for high power lasers
ESF Exploratory Workshop
Perugia, Italy
September 21st –23rd, 2005
LIGO-G050471-00-Z
Guido Mueller
University of Florida
For the LIGO Scientific
Collaboration
Table of Content
Input Optic for Advanced LIGO
 Requirements for Adv. LIGO
 Layout
» Modulators
» Mode cleaner
» Isolator
Documents:
LIGO-T020020-00-D
LIGO-T020027-00-D
LIGO-T010075-00-D
LIGO-T020097-0-D
LIGO-G050471-00-Z
IO-Subsystem Design Requirements Document
IO-Subsystem Conceptual Design Document
Advanced LIGO Systems Design
Auxiliary Suspended Optics Displacement …
2
Advanced LIGO
Changes which affect the input optics:
• Detuned Signal-recycling
• Higher Laser Power
• Increased Arm Finesse: T=0.5%
• Decreased Recycling
Cavity Finesse: T=6%
40 kg
SILICA
Iso.
PRM
BS
ITM
ETM
SRM
PD
LIGO-G050471-00-Z
Power Recycling Mirror
Beam Splitter
Input Test Mass
End Test Mass
Signal Recycling Mirror
Photodiode
3
Requirements
 Detuned Signal Recycling
» Creates asymmetric RF-sidebands
– All demodulated signals are sensitive to phase between RFsidebands and carrier (no technical noise suppression)
» For RF-sensing scheme: Modulation phase stability req.:
– ISSB (10 Hz) < -92 dBc/Hz
– ISSB (100 Hz) < -140 dBc/Hz
– ISSB (1 kHz) < -163 dBc/Hz
» Compare with Rb Standard: PRS10 (Stanford Research)
–
–
–
–
ISSB (10 Hz) < -130 dBc/Hz
ISSB (100 Hz) < -145 dBc/Hz
ISSB (1 kHz) < -150 dBc/Hz
But that is for a 10MHz signal not 180MHz!
» Options:
1. Lock to IFO
LIGO-G050471-00-Z
2. Reduce Frequency
4
3. DC-Sensing
Requirements
 Detuned Signal Recycling
» DC-sensing (now baseline):
RF signals are used only for auxiliary d.o.f.s
– Requirements unclear. Complicated function of locking scheme,
cross coupling between channels, noise spectrum, and feedback
bandwidth. But will be less difficult than in RF sensing.
» DC-sensing has additional advantages
– Lower Shot noise
– Less sensitive to laser frequency noise
– Reduced requirements on high-frequency, high-power photo
detectors
– ….
LIGO-G050471-00-Z
5
Requirements
 Higher Laser Power
» Relative Intensity Noise (RIN):
– Generates technical RPN in arm cavities
– Couples to asymmetry in arm cavity build-up
– Only important for Carrier, sideband power noise does not create
RPN!
» Requirement: 2x10-9 RIN/rHz @ 10 Hz on carrier intensity!
– Stabilization will work with main laser beam (carrier + SBs)
– Any change in the modulation index (SB power) will be undetected in
the intensity servo but will change carrier power and generate RIN
» Generates a requirement for the stability of the modulation
index:
 dG < 10-10/G (f/Hz) 1/rHz (includes safety factor of 10)
For G=0.1 dG< 10-8 /rHz @ 10Hz
Experimental tests on their way, but this is non-trivial!
LIGO-G050471-00-Z
6
Requirements
Laser Beam Pointing at PR-mirror:
 Couples to misaligned mirrors
 Trade off between pointing and DC alignment
Measured in terms of 10-amplitude relative to 00-amplitude:
LIGO-G050471-00-Z
7
Optics Express, Vol13(18) pg.7118
Requirements
Spatial Mode quality:
 10-mode ~ misalignment (just discussed)
 BE-more (20+02 mode) ~ mode mismatch
» Depends on thermal lensing in main IFO (TCS-system)
 Content in all other modes should be below < 2%
Power issue, no direct noise coupling expected (calculated)
Additional Requirements:
See LIGO Documents mentioned on 2nd page
LIGO-G050471-00-Z
8
IO Hardware




LIGO-G050471-00-Z
Modulators
Mode Cleaner
Faraday Isolator
Stable Recycling Cavities
9
Modulators
 LIGO I modulators will not handle the increased laser power
(losses and subsequent thermal lensing to high)
 New materials:
» KTP, KTA, RTA, RTP have high damage thresholds and high EOcoefficients.
» RTP has also very low optical and electrical losses. Measurements
at 50 W haven’t shown any measurable thermal lens. Long term
(16d) measurements at ~100W did not show any degradation. Then
laser failed.
» Requires additional long term, high power testing but looks OK.
 Parallel vs. complex Modulation:
» Cross products (SB on SB) generated in serial modulation might
need to be reduced:
– Parallel modulation in Mach-Zehnder
– Complex modulation using additional AM-modulator
LIGO-G050471-00-Z
10
Modulator
Sources:
RTA-Crystals:
• Raicol in Israel
Complete Modulator:
• Self made, need probably 3/IFO +
spares
• Also collaborate with New Focus
to build their modulator around
our RTA crystals
LIGO-G050471-00-Z
11
Modulation Schemes
 Serial Modulation:
Problem: SB on SB modulation
Has same frequency than SB-SB beat
Frequency
 Parallel Modulation
 Complex Modulation
Frequency
(modulate also at SB-SB frequency with opposite sign)
LIGO-G050471-00-Z
12
Mode Cleaner
Requirements:
 Length Stability < 3.6x10-15 (Hz/f) m/rtHz for (f<1kHz) (RPN?)
 Mode cleaning
 Angular Stability:
Current Design:
 Triangular Cavity
» Flat mirrors at Input and Output near MC waist
» Curved mirror at acute angle ROC=26.9m (cold), expect 27.9m (hot)
 L = 33.2m (Roundtrip), FSR = 9 MHz
 Finesse = 2000 (current design)
» Was driven by pointing from laser (overestimated pointing)
» Will probably be reduced (My best guess: 600)
LIGO-G050471-00-Z
13
High-power Faraday isolators
Possible Problems:
 Depolarization reduces isolation
efficiency
 Thermal lensing reduces spatial
mode quality
Depolarization:
 Two novel optical architectures
with two Faraday crystals and
wave plate (b) or Quartz Rotator (c)
 Developed by IAP, Nizhni
Novgorod, Russia
Thermal Lensing:
 Compensated with material with
opposite dn/dT, preferably using
a crystal, not a glass
LIGO-G050471-00-Z
14
High Power Faraday Isolator
l/2
QR
Pt
Pr
H
H
HP Faraday isolator design uses
quartz rotator:
- Developed at IAP, Russia
- 33dB at 180W laser power
Design with thermal compensation
(still with FK51 glass):
- No significant lensing up to 90W
Currently under test at LZH
LIGO-G050471-00-Z
15
Stable Recycling Cavities

Current Baseline: Recycling
Cavities are only marginally
stable
» Essentially flat-flat cavities
» Will increase scatter of
RF-SB and GW-signal into
higher order modes

Option: Stable Recycling
Cavity
» Move mode matching
telescope into Recycling
cavity
» Stabilizes the Recycling
cavities and reduces losses
into higher order modes

Add TCS and we should
have very small problems
with Thermal Deformations
LIGO-G050471-00-Z
16
Summary
Input Optics for Advanced LIGO:
 Faraday Isolator, Modulator expected to be able to handle thermal
noise w/o degrading the beam quality significantly
 Mode Cleaner should be fine, no thermal degradation expected
» Careful with frequency noise driven by technical RPN



Mode matching problems related to thermally distorted IFO eigenmode
(stable recycling cavities might help)
Pointing requirements seem to be within reach
Stability of Modulation phase seem to be OK for DC-sensing
» Likely driven by frequency stabilization servo
My main concern:
 Stability of Modulation index (RIN in carrier field)
 Unknown spatial mode in main IFO (Greg Harry: TCS)
LIGO-G050471-00-Z
17
Summary
Input Optics for Advanced LIGO:
 Faraday Isolator, Modulator expected to be able to handle thermal
noise w/o degrading the beam quality significantly
 Mode Cleaner should be fine, no thermal degradation expected
» Careful with frequency noise driven by technical RPN



Mode matching problems related to thermally distorted IFO eigenmode
(stable recycling cavities might help)
Pointing requirements seem to be within reach
Stability of Modulation phase seem to be OK for DC-sensing
» Likely driven by frequency stabilization servo
My main concern:
 Stability of Modulation index (RIN in carrier field)
 Unknown spatial mode in main IFO (Greg Harry: TCS)
LIGO-G050471-00-Z
18