Penn State Theory Seminar, March 2, 2018 The Future of Ground-based Gravitational-wave Detectors David Reitze LIGO Laboratory California Institute of Technology LIGO-G1800292-v1 LIGO Hanford Observatory LIGO-G1800292 Outline ● Why Make Bigger and Better Detectors? ● Improving Advanced LIGO: A+ ● Exploiting the Existing LIGO Facility Limits: Voyager ● Future ‘3G’ Facilities: Cosmic Explorer and Einstein Telescope 2 LIGO-G1800292 Some of the Questions That Gravitational Waves Can Answer ● Outstanding Questions in Fundamental Physics Black Hole Merger and Ringdown » Is General Relativity the correct theory of gravity? » How does matter behave under extreme conditions? » No Hair Theorem: Are black holes truly bald? ● Outstanding Questions in Astrophysics, Astronomy, Cosmology Image credit: W. Benger » Do compact binary mergers cause GRBs? » What is the supernova mechanism in core-collapse of massive Neutron Star Formation stars? » How many low mass black holes are there in the universe? » Do intermediate mass black holes exist? » How bumpy are neutron stars? » Can we observe populations of weak gravitational wave Image credit: NASA sources? » Can binary inspirals be used as “standard sirens” to measure GW Upper limit map the local Hubble parameter? » Are LIGO/Virgo’s binary black holes a component of Dark Matter? » Do Cosmic Strings Exist? Credit: LIGO Scientific Collaboration3 LIGO-G1800292 LIGO-G1800292 Observing Plans for the Coming 5 Years NOW Abbott, et al., “Prospects for Observing and Localizing Gravitational-Wave Transients with Advanced LIGO, Advanced Virgo and KAGRA”, https://arxiv.org/abs/1304.0670 5 LIGO-G1800292 Gravitational-wave Science is Sensitivity Driven 1) We want more Binary Neutron Stars events 3 running at O2 sensitivity 10 running at O3 sensitivity 825X running at aLIGO O4 sensitivity no A+ upgrade Nevents = <R>VT A+ upgrade 230X Where: 102 120X <R> -- average 45X astrophysical rate 1 10 V -- volume of the universe probed Volume x Time (in units of O1) à (Range)3 100 2016 2018 2020 2022 2024 2026 Calendar year T -- coincident observing time à t linear LIGO-G1800292 6 Gravitational-wave Science is Sensitivity Driven Binary Neutron Stars 2) Many sources 3 running at O2 sensitivity 10 running at O3 sensitivity 825X require higher SNR running at aLIGO O4 sensitivity no A+ upgrade A+ upgrade 230X to uncover new astrophysics: 102 120X - tidal disruption in BNS 45X mergers - tests of alternative 101 theories of gravity - Black hole ringdowns Volume x Time (in units of O1) - Stochastic background - Isolated neutron stars 100 2016 2018 2020 2022 2024 2026 - Galactic supernova Calendar year - .... LIGO-G1800292 7 Example: Probing the Neutron Star Equation of State A+ 8 LIGO-G1800292 Improving LIGO: A+ 9 LIGO-G1800292 A+: a Mid-Life Enhancement for Advanced LIGO Comoving Ranges: NSNS 1.4/1.4 M- and BHBH 30/30 M- ● Near term: ‘A+’, O2 data: NSNS 96 Mpc, BHBH 983 Mpc aLIGO design: NSNS 173 Mpc, BHBH 1606 Mpc a mid-scale Aplus design: NSNS 325 Mpc, BHBH 2563 Mpc upgrade of Advanced LIGO Hz] 10-22 » Improvements across all bands ● Projected time -23 scale for A+ 10 operation: 2023 - Strain noise h [1/ 2025 10-24 101 102 103 Frequency (Hz) 10 LIGO-G1800292 The Rationale for A+? ● A+ is an incremental upgrade to aLIGO that can happen in the next 5-7 years ● A+ leverages existing technology and infrastructure, with minimal new investment, and moderate risk ● Target improvement: factor of 1.75* increase in range over aLIGO è A factor of ~ 5 greater CBC event rate ● A+ is a stepping stone to 3G detector technology ● Can be observing within 5 years (possibly late 2022) ● “Scientific breakeven” within 1/2 year of operation ● Incremental cost: a small increment of the aLIGO cost *BBH 30/30 M¤: 1.87x Slide inspiration: Mike Zucker, LIGO Laboratory *BNS 1.4/1.4 M¤: 1.7x LIGO-G1800292 11 Key A+ Upgrades Aplus design curve - NSNS (1.4/1.4 M-) 325 Mpc and BHBH (30/30 M-) 2563 Mpc Quantum Seismic Newtonian Suspension Thermal Reduced Radiation Coating Brownian -22 Pressure via Laser 10 Amplitude ‘Squeezing’ Coating Thermo-optic Substrate Brownian Excess Gas Hz] Total noise Improved Mirror Coatings With Lower Reduced Shot Noise -23 Thermal Noise 10 Via Laser Phase Squeezing Strain [1/ 10-24 101 102 103 Frequency [Hz] 12 LIGO-G1800292 Squeezing: Reducing Quantum Noise • Electromagnetic fields are H. P. Yuen, Phys. Rev. A13, 2226 (1976) C. M. Caves, Phys. Rev. D26, 1817 (1982) quantized: Wu, Kimble, Hall, Wu, PRL (1986) ˆ ˆ ˆ E = X1 cosωt +iX2 sinωt 3.5 dB squeezing Quantum fluctuations exist in the TypicalTypical noisenoise without without squeezing squeezing (1.5X reduced noise) • SqueezingSqueezing-enhanced−enahnced sensitivity sensitivity vacuum state: GEO600 Hz] −21 √ 10 Main Laser Coherent State −22 10 FaradayLIGO H1 Strain Sensitivity [1 / Squeezed Vacuum Isolator Squeezed Vacuum Source x State 2 2.1 dB squeezing (1.27X reduced noise) │E │ −23 Photodiode 10 2 x 3 10 1 10 ϕ Frequency (Hz) Aasi, et al., (LIGO Scientific Collaboration), Nature Physics, 7, 962 (2011); Nature Photonics 7 613 (2013). LIGO-G1800292 13 The Best of Both Worlds: Frequency Dependent Squeezing Squeezed−enhanced aLIGO 18 Quantum−enhanced aLIGO Quantum Noise 16 Coating Brownian −22 14 10 Quantum−enhanced aLIGO Total noise aLIGO quantum noise 12 aLIGO total noise 10 8 Pro.local − 6 Hz] 4 √ MacBook 2 − 0 -2 −23 10 -4 ◦ -6 φ =–2.1 ± 2.3 , ∆γ =39± 43 Hz φ =16.2 ± 1.2◦, ∆γ =–56± 23 Hz -8 ◦ φ =36.4 ± 0.5 , ∆γ =49± 11 Hz -10 φ =67.3 ± 0.2◦, ∆γ =401± 12 Hz Noise relative to coherent vacuum [dB] . ± . ◦ ∆ ± 2012 by lisab on lisabs φ =915 0 2 , γ =–97 17 Hz − -12 Model May -14 − Overall sensitivity improvement -16 2 3 4 10 10 Frequency [Hz] 10 Strain Sensitivity [1/ −24 10 created using gwinc_fig.m on 16 1 2 3 10 10 10 Frequency [Hz] Concept:Oelker, et Kimble, al., “Audio-band Levin, Matsko Frequency-dependent, Thorne, Vyatchanin, Phys.Squeezing”, Rev. D (2001)Phys. Rev. Lett. 116, 041102 (2016). 14 LIGO-G1800292 Thermal Noise in Optical Coatings ω0 Y. Levin Phys. Rev. Δx D 57 659 (1998) 2 > ) ω ( x low Q Δ signal < band ∝ 1/Q high Q ω /ω0 ● Simple picture: kT of energy per mechanical mode, viscous damping ● For coating dominated noise and structural damping: coating thickness coating elastic loss −4 φTiO225 :Ta O =2× 10 2kT d Yʹ Y 5 Slide Credit: B ⎛⎞ 4 10− SfTx (,)≈22φ ⎜⎟+φSiO2 =× Marty Fejer, Stanford π fwY ⎝⎠YYʹ beam radius Compare: Bulk Silica φ ~ 10-6-10-8 LIGO-G1800292 Low-frequency losses in amorphous dielectrics ● Loss mechanism: conventionally associated with low energy excitations (LEEs) » conceptualized as two-level systems (TLS); Distribution of TLSs arising from disordered structure » Simple physical mechanism: bond flopping in amorphous materials, e.g. SiO2 E V 2 E1 Δ Single TLS Distribution of TLSs 2 γ2 ωτ Δ −12γ ωτ ⎛⎞Δ Q−12sech⎛⎞ g ( V ) f ( ) dV d QN= 22sech ⎜⎟ =22 ⎜⎟ ΔΔ YkT1+ ωτ ⎝⎠ kT YkT∫∫1+ ωτ ⎝⎠ kT Figures: B.S. Lunin LIGO-G1800292 Routes to Improved Coatings with Low Thermal Noise ● Goal: 2X – 4X reduction in coating thermal noise ● Ultrastable glasses ● Internal friction of films deposited at high » Glasses prepared by physical vapor deposition show temperatures Ts extraordinary stability » very different from film annealed at same – slower cooling liquid reaches lower energy states temperature » simulations suggest surface liquid layer has orders of » φ ~ 2x10-6 vs ~10-4 magnitude higher mobility than caged particles in solids X. Liu, F. Hellman, et al, Phys. Rev. Lett. liquid annealed 350°C 113, 025503 (2014) deposition temperatures, Ts Challenge: extendable to amorphous oxides? Scalable to commercial coating methods? S. Singh, et al. Nature Mater. 12, 139 (2013); Parisi, Nature Mater. 12, 94 (2013) LIGO-G1800292 17 Routes to Improved Coatings with Low Thermal Noise ● Goal: 2X – 4X reduction in coating thermal noise ● Crystalline coatings » molecular beam epitaxy to generate crystalline GaAs/AlGaAs multilayer » mirror disc lithography and etch, substrate removal, and direct bonding on substrate Challenge: i) Scalability Current state-of-the-art: 75 mm aperture. LIGO needs: > 300 mm (Cost!) ii) Coating uniformity: 10-4 (IBS) vs 10-2 (crystalline) G. D. Cole, W. Zhang, M. J. Martin, J. Ye, and M. Aspelmeyer, Nature Photonics (2013) 18 LIGO-G1800292 Preliminary A+ Schedule ● Possible project start date in early 2020 (paced by funding) ● Installation start date early 2022 Start Finish 2017 2018 2019 2020 2021 2022 2023 2024 LIGO Laboratory 19 LIGO-G1800292 Exploiting the LIGO Observatory Facility Limits: LIGO Voyager 20 LIGO-G1800292 LIGO Voyager Concept ● Voyager Key Technologies » Silicon Mirrors: 200 kg, 45 cm dia., mCZ process » Mirror Coatings: �-Si/SiO2 (�-Si: ~lossless thin film) » Cryogenics: 123 K (zero CTE), radiative (non-contact) cooling » Lasers (2000 nm): P~ 180 W, PARM ~ 2800 kW » Wavefront Compensation: thermally adjustable lenses only (no actuation of test mass) » Photodiode Quantum Efficiency: 80 -> 99% for 2 micron (No change to Advanced LIGO seismic isolation system!) LIGO Laboratory 21 LIGO-G1800292 LIGO Voyager Conceptual Design Sensitivity 22 LIGO-G1800292 Voyager Reach for Compact Binary Mergers ObservedBinaries LIGO-G1800292 Technology: 2 Micron Lasers EU Funded 2 micron laser study ● 2 µm Tm:YAG, Ho:YAG commercial lasers exist » low power/low noise, or high power/ high noise ● Challenge is to make high power/low noise ● Development programs underway by several gravitational wave groups » Decade long program envisioned http://isla-project.eu/ Slide inspiration: Rana Adhikari, LIGO Laboratory LIGO Laboratory 24 LIGO-G1800292 Technology: 200 kg Silicon Mirror ● Main challenges are i) mirror 30 cm size, ii) absorption » Goal: absorbed Power < 3 W; » 3 ppm/cm (FZ): max diameter ~ 20 cm » mCZ from SEH can get » samples acquired, absorption measurements done (< 4 ppm) » SEH Japan will make 45 cm diameter mCZ » how to sequence all of the annealing? Different processes for substrates, coatings.