Scattering Loss Hiro Yamamoto - Caltech LIGO

Transcription

1 Scattering Loss Hiro Yamamoto - Caltech LIGO LIGO I mirror loss estimation» surface figure, micro roughness and larger angle scattering LIGO I mirror surface qualities» polished surface data by CSIRO and coated surface data by LIGO Does not include new issues of LMA mirrors» bubbles, AR coating, etc Advanced LIGO loss requirement» implication of the LIGO I mirror quality Diffractive loss in stable Michelson cavity» Loss due to the effective long distance propagation in the stable Michelson cavity 1

2 Recycling gain, visibility, etc are consistent with 140 ppm loss per arm known loss per arm Crude estimation of loss budgets in a LIGO I arm» surface figure (λ>0.5cm) : 10 ppm / mirror x 2» ETM transmission : 7 ppm» absorption : 4ppm / mirror x 2» diffractive loss : 2ppm 140ppm total loss - known loss = 100 ppm / arm or 50 ppm / mirror power recycling gain Recycling gain as a function of extra loss per mirror in FFT simulation H2 50ppm? H1 L extra loss [ppm] 2

3 Polished (CSIRO) vs coated (LIGO) surface data CSIRO LADI A 300-mm aperture phase shifting fizeau interferometer CSIRO TOPO three dimensional noncontact optical profiler surface figure loss CSIRO LADI Caltech Phasemap CSIRO TOPO Integrating sphere Y 7 x f X Bill K Valery F FFT Simulation X 3

4 surface figure CSIRO vs Phasemap PSD ITM SUB X SUB Y POL X POL Y CSIRO PSD 10 1 CSIRO and CIT phase data 1D psd tilt and power subtracted CSIRO - uncoated surface CIT - coated surface mm nm^ CIT ETM03 6x CIT ETM02 CSIRO ETM 02 CSIRO ETM02 2.5x / mm /mm 4

5 Surface figure loss of AdvLIGO loss per arm : rms = 1nm vs 0.5nm ppm rms = 0.5nm loss per mirror population 0.5 nm 1.0 nm based on 80 simulated surfaces each LIGO I mirror phasemap in advligo arm rms = 0.44nm rms = 1nm loss : ppm / mirror

6 Loss in the least investigated region : λ ~ 1mm intermediate region loss by integral of CSIRO fit in 2~100 cm -1 = 26ppm low end : LIGO < CSIRO high end : LIGO > CSIRO Caltech Phasemap CSIRO LADI CSIRO TOPO Integrating sphere Y 7 x f X Bill K Valery F FFT Simulation X

7 Orange peel - psd shows bumps at ~0.3 cm - Caltech phasemap loss ~ 2.4 ppm new FFT grid size = 0.14 ~ 0.28cm loss ~ 1.3 ppm x 2 old FFT : loss(35cm/256) - loss(35cm/128) CSIRO PSD Bill K estimated ~10ppm CSIRO LADI Caltech Phasemap CSIRO TOPO Integrating sphere Y 7 x f X Bill K Valery F 7 X

8 Large angle scattering loss Scatterometer port Sr ITM Main arm beam large angle scattering loss In situ, large angle measurement : PD CSIRO LADI Bill Kells H1 ITMy June 2003 Valera L1 ITMxy&ETMxy October 2007 Caltech Phasemap CSIRO TOPO Integrating sphere Caltech OTF : integrating sphere Liyuan FFT Simulation Y 7 x f X Bill K Valery F X 8

9 Large angle scattering LLO ITM/ETM by Valery Frolov 10!23 loss : 1 ~ 50 degree ITMX ITMY ETMX ETMY CSIRO PSD FIT FIT < 5 degree FIT all 10!24 loss : 22ppm 1D PSD (m 3 ) 10!25 loss : 3.6ppm loss : 20ppm 10!26 10! LSC-Virgo Hannover k (1/cm) on October 24,

10 Integrating sphere data Liyuan, Herena (T070233) 1.5 1cm x 1cm 0.3mm beam size 0.1mm step ppm ppm ppm ppm ppm ppm ppm 10

11 Advanced LIGO total arm loss budget = 70ppm known loss : ~ 50 ppm / arm» diffraction : 0.4 ppm» absorption : 0.5 ppm x 2» surface figure : 20ppm x 2» ETM transmission : 7 ppm 11 ppm / mirror for micro roughness, large angle and all other losses LIGO I mirror : 50 ppm 11

12 Diffraction effect in Stable Michelson cavity RM ROC = m w = 2.2 mm R = m R = 15.7 m m m MMT2 ROC = m w = 3.7 mm R = m R = 0.96 m MS-RM R = 1440m Baffle 21.4cm 24.9cm m 4.5 m m w = 6.1 cm R = 1457 m w = 6.0 cm R = 1432 m ROC = m Aperture = 37 cm ROC = 2076 m Aperture = 26 cm thickness = 6 cm Aperture = 34 cm MMT3 LSC-Virgo SB Hannover on October 24, 2007 ITMt ETMt 12

13 Beam profile on MMT3 coming from MMT2 RM MMT2 r=13cm log(p) vs r 2 diffraction tail 26cm 230ppm 65ppm log(p) = -r 2 / w 2 MMT3 ITM 40ppm r 2 (m 2 ) 13

14 Summary LIGO I mirror loss estimation» Scattering loss per mirror in LIGO I arm (Power Recycling gain, etc) : 50ppm» Loss(λ>5mm) ~ 10 ppm/mirror» Loss(λ~1mm) ~ not well understood ( 3 ~ 30ppm/mirror?? )» Loss(λ<0.1mm) ~ 10 ~ 30ppm LIGO I mirror surface quality - some inconsistencies» λ>5mm : PSD(coated surface) ~0.1 PSD(polished surface)» λ~1mm : not well understood» λ<0.1mm : mesured loss ~ 10 x estimation by polished surface data Advanced LIGO loss requirement < 35ppm / mirror» Loss(λ>5mm) ~ 20~25 ppm/mirror with RMS < 0.7nm» Need to understand LIGO I mirror losses and to suppress losses or change the AdvLIGO specification to be more tolerant to extra loss Loss in stable Michelson cavity» Due to far field propagation in the cavity, ~500 ppm loss by diffraction with w(itm)=6cm.» With asymmetric arm configuration (w(itm)=5.5cm,w(etm)=6.2cm), this is suppressed to ~50ppm. 14