Survey
* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project
Proposal Thermal and Thermoelastic Noise Research for Advanced LIGO Optics Norio Nakagawa Center for Nondestructive Evaluation Iowa State University in collaboration with E.K. Gustafson and M.M. Fejer Ginzton Laboratory, Stanford University November 29-30, 2001 LIGO-G010415-00-Z Talk at PAC-11, LHO 1 Outline • Objective and Approach • Output of the Prior NSF Support – Laser phase noise formulas for optical resonator and delay line – Coating noise estimation • Proposed Work – Coating noise studies for mirrors of edge geometry – Thermo-elastic noise of coated mirrors – Thermal & thermo-elastic noises of realistic mirror designs • Tasks & Time Lines • Summary • (Broader Impacts) November 29-30, 2001 LIGO-G010415-00-Z Talk at PAC-11, LHO 2 Objective and Approach Objectives • To develop laser-phase-noise formulas – Green’s-function-based – Analytical and computational – two-point laser-phase-fluctuation correlations – complex test mass objects. • To extend the noise estimation method to thermo-elastic noises. • To estimate thermal and thermoelastic noises of coated mirrors for advanced LIGO designs. • To examine merits of interferometer design options for future LIGO. November 29-30, 2001 LIGO-G010415-00-Z Approaches • Calculate phase noise via Green’s function method – Elasticity – Thermo-elasticity & thermal diffusion • Analytical mirror models – Half space – Quarter space – Thin coating layer • Numerical calculations – Realistic mirror shapes – Coating loss – Delay-line vs. Fabry Perot Talk at PAC-11, LHO 3 Output to Date • Publications – N. Nakagawa, Eric Gustafson, P. Beyersdorf and M. M. Fejer “Estimating the off resonance thermal noise in mirrors, Fabry-Perot interferometers and delay-lines: the half infinite mirror with uniform loss,” to appear in Phys. Rev. D – N. Nakagawa, A. M. Gretarsson, E.K. Gustafson, and M. M. Fejer, “Thermal noise in half infinite mirrors with non-uniform loss: a slab of excess loss in a half infinite mirror,” submitted to Phys. Rev. D – N. Nakagawa, E.K. Gustafson, and M. M. Fejer, “Thermal phase noise estimations for fabry-perot and delay-line interferometers using coated mirrors,” in preparation. – D. Crooks, et al., “Excess mechanical loss associated with dielectric mirror coatings on test masses in interferometric gravitational wave detectors,” submitted to Classical and Quantum Gravity. – Gregory M. Harry, et al., “Thermal noise in interferometric gravitational wave detectors due to dielectric optical coatings,” submitted to Classical and Quantum Gravity. November 29-30, 2001 LIGO-G010415-00-Z Talk at PAC-11, LHO 4 Phase-Noise Correlation • Requirement – Compute laser phase noise correlation x1 , t1 x2 , t2 S , x1 , x2 d i t1 t 2 2 x1,t1 e x2 ,t2 x1 x2 E, , E, , Ex. Coating noise model November 29-30, 2001 LIGO-G010415-00-Z Talk at PAC-11, LHO 5 Intrinsic Thermal Noise • Phase Noise Formula 2k T S , r1 , r2 4k 2 B w dS dS r r1 00 r r2 w 00 d x i nj x , r ; c cijkl x k nl x , r ; c 3 V icijkl 1 i cijkl cijkl cijkl S , r1 , r2 ui r u j r r1 , r2 2 2 r r e w 00 w2 November 29-30, 2001 LIGO-G010415-00-Z the two-point laser-beam phasenoise power-spectrum correlation w, k the laser beam spot size (amplitude radius), and wave number the displacement spectral correlation ij elastic Green’s function cijkl [c’ijkl, c”ijkl] elastic constants [dispersive and absorptive parts] () loss function kB , T Boltzmann constant, and temperature The laser beam reflection points (the beam centers) the Gaussian laser-beam profile function Talk at PAC-11, LHO 6 Fluctuation-Dissipation Relation • Surface Force density Strain w F r r1 uij x; r1 F dS i nj x , r ; c 00 r r1 w 00 2k T 1 3 ukl x; r2 S , r1 , r2 4k 2 B d x u x ; r c x ij 1 ijkl 2 V F • Laudau-Lifshitz E mech 2 2 ijkl T0 dV T 2 dV uijijklukl , cijkl V V structural thermo elastic 2 4 k BT 1 S , r , r 4k E mech 2 2 F November 29-30, 2001 LIGO-G010415-00-Z viscous viscous Talk at PAC-11, LHO 7 Various Optical Configurations • Phase noise formulas Computed explicitly for – Single-reflection mirror SSingle S , r , r 1 1 rI 2 2rI S 1 cos 2 SE rI2 SI 2 2 1 rI 1 rI FP N n 1 S ( ) S , rn , rn 2 cos[2( n q) ] SE , rn , rq N DL E n 1 N 1 November 29-30, 2001 LIGO-G010415-00-Z N 1 n 1 S , n , n 2 cos[2( n q) ] SI , n , q – Fabry-Perot resonator – Optical delay line n 2 q 1 I n 1 n 2 q 1 rI the input mirror reflection coefficient the transit time SE , SI the single-reflection phase noises of the input and end-point mirrors. rn the positions of the N-time reflections on the endmirror surface p the positions of the (N-1)-time reflections on the inputmirror surface E, If Half space: Young’s modulus, and Poisson ratio 8k T k 2 1 2 r1 r2 2 S , r1 , r2 B e w E Talk at PAC-11, LHO 2 w2 2 I 0 r1 r2 2 w 2 8 Fabry-Perot vs. Delay Line • Fabry-Perot vs Delay lines 15 – Analytical half-space mirror model – Fabry-Perot interferometer vs several delay lines – Storage time as proposed for LIGO II. – Delay line beam centers R 10 2 3 • evenly spaced • on a circle 4 5 5 • When the spots are not overlapping appreciably, the delay line is less noisy than the Fabry-Perot. 6 0 – Noise levels are similar if • the spot circle radii comparable to the beam spot size • the spots are largely overlapping, and above several hundred Hertz. November 29-30, 2001 LIGO-G010415-00-Z 2w 1 50 150 100 Frequency [Hz] 200 Figure 1: Comparison of the phase noise from a delay-line and a Fabry-Perot interferometer. The solid curve (1) is for a 4 km Fabry-Perot interferometer with an input mirror power reflectivity of RI=0.97 and an end mirror power reflectivity of RE=1.00. Curves 2, 3, 4, 5 and 6 correspond to 4 km delay lines all with 130 spots on the end mirror and laser beam spots of 1/e field radius w in a pattern with their centers on a circles of radius R=w/3 (2), R=2w/3 (3), R=5w/2 (4), R=10w (5) and R=20w (6) where w=3.5 cm, the spot size used for both mirrors of the Fabry-Perot interferometer. The mirror Q is assumed to be 3´108 and the material properties are those of Sapphire E=71.8 GPa and s=0.16 however we are treating sapphire as isotropic for the purpose of this illustration and assuming a single loss function. Talk at PAC-11, LHO 9 Coating Noise; Problem Statement • Coating noise model – Based on half-space mirror model • a lossy layer (thickness d) on a lossy host material – Requirement: Compute laser phase noise correlation – Approach: via analytical Green’s function x1,t1 x2 ,t2 x1 x2 E, , E, , November 29-30, 2001 LIGO-G010415-00-Z x1 , t1 x2 , t2 Talk at PAC-11, LHO d i t1 t2 2 e x1 x2 S , x1 x2 d i t1 t2 2 e 10 Coating Noise • Static Green’s function for layer-on-substrate sos sos z ~p, z 1 1 pE 1 sos ~ 1 21 pz i 1 G p,0 2 pz 3 cosh pz sos ~ 3 4 1 pz 1 2 G p , 0 pz i sinh pz 3 2 1 2 2 ~p, z 211 2 pz 3G sos ~p,0 21 pz i 1 cosh pz 211 21 pz i 1 G sos ~ p,0 pz 3 1 2 2 sinh where pz 1 cosh 2 pd 411 324 1 2 1 2 2 3 4 sinh 2 pd 1 2 1 2 cosh 2 pd 1 1 2 sinh 2 pd sos ~ 1 2 2 1 G p ,0 81 2 1 2 3 4 2 1 2 3 4 1 2 3 4 sinh pd 2 2 811 2 1 1 3 4 pd 1 41 pd 1 1 3 4 3 2 cosh pd 2 11 1 2 3 4 sinh pd cosh pd 2 11 1 2 sinh pd 4 11 2 1 1 3 4 pd 2 11 EE , 1,2,3 Pauli matrices November 29-30, 2001 LIGO-G010415-00-Z D. M. Burmister, J. Appl. Phys. 16, 89-94, 1945. Talk at PAC-11, LHO 11 Coating Noise • Intrinsic Thermal Phase-Noise Estimation 2k T Scoating , r 4k 2 B 1 w 1 2 r 2 2 w2 I 0 r 2 2 w2 E e 2 2 1 2 1 1 1 1 2 E 2 2 2 d r2 1 E 1 E e w 2 1 1 2 1 1 E Scoating , r The phase noise two-point correlation for a coated half-space mirror; double-sided E, , Young’s modulus, Poisson ratio, and loss function of the substrate material E , , Those of the coating material d The coating thickness I0(z) The 0-th order modified Bessel function of the first kind; I0(0)=1 November 29-30, 2001 LIGO-G010415-00-Z 2f r w2 O d 2 w2 Frequency a relative position vector between the two beam centers on the coating surface; r 0 for a single reflection. k, w kB , T Talk at PAC-11, LHO The laser beam wave number, and spot size (amplitude radius) The Boltzmann constant and the temperature 12 Coating Noise • Resonator coating S ,0 4k 2 2k BT 1 1 2 1 2 1 1 d w E 1 E w • Delay lines 2k T Scoating , r 4k 2 B 1 w 1 2 r 2 2 w2 1 2 1 1 d r 2 2 2 e I 0 r 2 w e E 1 E w 1 r r November 29-30, 2001 LIGO-G010415-00-Z Talk at PAC-11, LHO w2 13 60 Phase Noise [prad/root-Hz] Fabry-Perot resonator w/ coated mirrors (Various beam radii) 40 wo/ coatings w=3.5cm w=3.5cm 2*w 2*w 5*w 5*w 10*w 10*w Substrate=Fused silica E 72.6 GPa σ 0.16 Q 3 107 Coating= average of Al2O3 and Ta2O5 (Crooks et al.) E 260 GPa σ 0.26 Q 1.6 10 4 20 0 50 100 150 200 Frequency [Hz] November 29-30, 2001 LIGO-G010415-00-Z Talk at PAC-11, LHO 14 60 Phase Noise [prad/root-Hz] Optical Delay Lines w/ coated mirrors (w=3.5cm) 40 wo/ coating R=w/3 R=w/3 R=2w/3 R=2w/3 R=5w/2 R=5w/2 R=10w R=10w 20 0 50 100 150 200 Frequency [Hz] November 29-30, 2001 LIGO-G010415-00-Z Talk at PAC-11, LHO 15 Proposed Work • Edge effects on coated mirror noise – Analytical quarter-space model • Extension to thermo-elastic noise – Coated mirrors – Delay-line interferometers – Mirrors at low temperatures • Thermal & thermo-elastic noises – Mirrors of realistic shapes • Numerical Green’s function calculation November 29-30, 2001 LIGO-G010415-00-Z Talk at PAC-11, LHO 16 Quarter-Space Mirror Model • Edge effects on noise calculation – Analytical model November 29-30, 2001 LIGO-G010415-00-Z Talk at PAC-11, LHO 17 Quarter-Space Mirror Model (con’t) • Finite-mirror size effect on thermal noise (scalar model) 15 Phase Noise [prad/sq. root Hz] Beam (power-)radius=6cm Number of Delay-line spots=130 Diameter of D.-L. circle=27cm F.-P. (Mirror dia.=30cm) D-L #1 (Mirror dia.=30cm) 10 D-L #2 (Mirror dia.=45cm) D-L #3 (Mirror dia.=60cm) D-L #4 (Mirror dia.=300cm) 5 0 50 November 29-30, 2001 LIGO-G010415-00-Z 100 Frequency [Hz] Talk at PAC-11, LHO 150 200 18 Thermo-elasticity & Thermal diffusion • From previous identification 4k T 1 S , r , r 4k 2 B2 E mech 2 F 2 Emech T0 dV T dV uijijklukl V V 2 ukl Emech T0 dV T 2 dV uij cijkl V V • If the adiabatic condition holds 2 2 T T0 TC0 v v l t ull p 2k T Sthel , r1 , r2 4k 2 B November 29-30, 2001 LIGO-G010415-00-Z T w w d S d S r r 00 1 00 r r2 v Cv 2 l d x vt2 2 3 V i l Talk at PAC-11, LHO x , r ; c x , r ; c nl i k nk 19 Numerical elasticity • Noise study: Cylindrical mirrors 2ui jTij 0 2 ik j ijk ik x x P – Numerical Green’s function computation • Betti-Rayleigh-Somigliana formula • Nodal discretization by the boundary element method. – Cylindrical mirror model • Single reflection • Fabry-Perot • Delay-line – Demonstrate the delay-line vs. resonator assertion – Effects of mirror aspect ratios. November 29-30, 2001 LIGO-G010415-00-Z V xP uk xP dS j ui ijk Tij ik S ui x Displacement Tij x stress tensor Tij cijkl k ul ij x the fundamental solution (Green’s function) ijk x cijlm l mk cijkl elastic constants Density V, S V x volume of a region and its boundary surface Talk at PAC-11, LHO the characteristic function of V (=1 inside V, =0 outside) 20 Static Elasticity f • Needs to avoid rigid-body motions n – Mass and moment of inertia M dV , I ij dV x 2 ij xi x j V V – Given surface forces f can be counterbalanced by volume forces 1 a V F g x a b x , M 1 b V I K S g V K where F dS f x , K dS x f x S S Fg dV g x , K g dV x g x V November 29-30, 2001 LIGO-G010415-00-Z F Fg 0, K K g 0 V Talk at PAC-11, LHO O b x bx 21 Tasks & Time Lines T1. Coating noise estimation – spatial de-correlation T2. Numerical noise estimations – coating T6. Anisotropy effects T7. Dielectric loss study T8. Off-beam-axis scatterings T3. Quarter space – Analytical model – Finite size effect November 29-30, 2001 LIGO-G010415-00-Z – delay-line vs. resonator – coating T5. Quarter space II – Realistic mirror shapes – Validation Task T1 T2 T3 T4 T5 T6,T7,T8 T4. Thermo-elastic noise estimation 3/31/2003 Talk at PAC-11, LHO 3/31/2004 3/31/2005 22 Summary Accomplishments • Obtained two-point phase noise correlation formulas – Resonator – Delay line – Delay lines can be quieter • Given coating noise formulas Proposed Activities • Studies of intrinsic noises of coated mirrors – Thermal noise – Thermo-elastic noise – Relative significance of coating noise – Realistic mirror shapes – Relative magnitudes between coating and substrate noises. – Their behaviors against the beam size and other optical parameters November 29-30, 2001 LIGO-G010415-00-Z Talk at PAC-11, LHO 23 Broader Impacts • Education – Student involvement through the collaboration with Stanford Group • Contribution to LSC – The thermal noise estimation methodology – Impact on sorting out advanced optics designs – Impact on mirror material selection • Impacts on other federally funded programs – Through advancement of computational physics methodology • NSF Industry/University Cooperative Research Center program • DOE/NERI project; “On-line NDE for advanced reactors” November 29-30, 2001 LIGO-G010415-00-Z Talk at PAC-11, LHO 24