GRMHD simulations of remnant accretion disks from neutron star mergers

Transcription

1 4-color Process 0% Cyan 72% Magenta logo can also be rendered in black, grey (60% black), Pantone 280, or Pantone 286; on a darker color background, the logo can be rendered in Pantone 290, 291, or 284, depending on which color works best with the overall design of your product, the media in which it will be reproduced, and its intended use. GRMHD simulations of remnant accretion disks from neutron star mergers White or Pantone 290 (Columbia Blue) Background: Pantone 286 Daniel M. Siegel NASA Einstein Fellow For photographs, use the logo in white against a darker area, positioning it either at top left/right or bottom left/right. Center for Theoretical Physics, Columbia Astrophysics Laboratory, Columbia University INT workshop 2017, Univ. Washington, July 31- Aug 4,

2 (M ) (M ) distribution Prospects for Observing and Localizing GW Transients with aligo and AdV Isotropic Aligned Isotropic Aligned Isotropic Aligned PyCBC GstLAL LIGO: NS mergers PyCBC GstLAL PyCBC GstLAL ,600 3,2 2,530 1,850 2,300 1,280 3,270 2,820 2,490 1,660 2,800 1,270 Table 2. Sensitive space-time volume hv T i and 90% confidence upper limit R90% for NSBH systems with isotropic and aligned spin distributions. The NS spin magnitudes are in the range [0, 0.04] and the BH spin magnitudes are in the range [0, 1]. Values BNS ranges for the various stages of aligo and AdV expected provided in hv T i is calculated using a FAR threshold of 0.01 yr 1. The rate upper are shown for bothevolution the pycbcare andalso gstlal pipelines. Figure 1. limit is calculated using a uniform prior on L = RhV T i and an 18% uncertainty in hv T i from calibration errors. Advanced LIGO Dominik et al. pop syn 1/2 Early ( , Mpc) Mid ( , Mpc) Late ( , Mpc) Design (2019, 200 Mpc) BNS-optimized (215 Mpc) Strain noise amplitude/hz Strain noise amplitude/hz Advanced Virgo 21 1/2 de Mink & Belczynski pop syn Vangioni et al. r-process Early ( , 20 60O3 Mpc) O2 Mid ( , Mpc) Late ( , Mpc) Design (2021, 130 Mpc) BNS-optimized (145 Mpc) O Frequency/Hz 3 O1 Vangioni et al. r-process Jin et al. kilonova Petrillo et al. GRB Jin et al. kilonova Coward et al. GRB Petrillo et al. GRB Siellez et al. GRB Coward et al. GRB Fong et al. GRB Fong et al. GRB Kim et al. pulsar O2 de Mink & Belczynski pop syn aligo 20 rate compendium aligo 20 rate compendium O3 Dominik et al. pop syn Frequency/Hz BNS Rate (Gpc 3yr 1) 4 Figure 1: aligo (left) and AdV (right) target strain Figure sensitivity a functionofofthe frequency. 6. Aascomparison O1 90% The upperbinary limit on the neutron-star (BNS)Fig.: range, the averageofdistance these signals could be detected, is given Sensitivity LIGO totowhich BNS mergers (left) and sensitivity vs. predicted BNS merger rate to other rates discussed in the textin (Abadie megaparsec. Current notions of the progression of sensitivity are given for early, mid and late commissioning et al. 20; Kim et al. 2015; Fong et al. 2015; Siellez et al. phases, as well as the final design sensitivity target and the BNS-optimized sensitivity. While both dates 2014; Coward et al. 2012; Petrillo et al. 2013; Jin et al. 2015; and sensitivity curves are subject to change, the overall progression represents our best current estimates. Vangioni et al. 2016; de Mink and Belczynski 2015; Dominik et al. 2015). The region excluded by the low-spin BNS The commissioning of aligo is well under way. called for three identical in O2 ratethe limitoriginal is shadedplan in blue. Continued non-detection 4-km interferometers, two at Hanford (H1 and H2) and oneand at O3 Livingston 2011, theand LIGO (slash) (dot) with(l1). higherin sensitivities longer oplab and IndIGO consortium in India proposed installing one of the aligo Hanford detectors (H2) eration time would imply stronger upper limits. The O2 and at a new observatory in India (LIGO-India) [64]. As early 2015, Laboratory O3ofBNS ranges areligo assumed to be 1-1.9has andplaced times the H2 interferometer in long-term storage for possible use in India. Funding for the Indian portion larger than O1. The operation times are assumed to be 6 and months (Aasi et government. al. 2016) with a duty cycle equal to that of of LIGO-India is in the final stages of consideration9by the Indian O1data ( 40%). Advanced LIGO detectors began taking sensitive in August 2015 in preparation for the NSBH Rate (Gpc yr ) 3 Figure 7. 1 A comparison of the O1 90% upper limit on the NS merger rates (right) 2016c NSBH merger rate to other rates discussed Abbott+ in the text (Abadie et al. 20; Fong et al. 2015; Coward et al. 2012; Petrillo et al. 2013; Jin et al. 2015; Vangioni et al. 2016; de Mink and Belczynski 2015; Dominik et al. 2015). The dark blue region assumes a NSBH population with masses M and the light blue region assumes a NSBH population with masses 1.4 M. Both assume an isotropic spin distribution. Continued non-detection in O2 (slash) and O3 (dot) with higher sensitivities and longer operation time would imply stronger upper limits (shown for 1.4 M NSBH systems). The O2 and O3 ranges are assumed to be and times larger than O1. The operation times are assumed to be 6 and 9 months (Aasi et al. 2016) with a duty cycle equal to that of O1 ( 40%). LIGO will probe deeply into the predicted NS merger rate distributions by 2018 (O3) exciting discoveries expected soon first observing run. O1 formally began 18 September 2015 and ended 12 January It involved the H1 and L1 detectors; the detectors were not at full design sensitivity. We aimed for a BNS Weboth additionally includewere somerunning more recent range of Mpc for both instruments (see Figure20). 1), and instruments withestimates a from population synthesis for both NSBH and BNS (Dominik Mpc range. Subsequent observing runs will have increasing duration and sensitivity. We aim et al. 2015; Belczynski et al. 2016; de Mink and BelczynGRBs can predict the rate of progenitor mergers (Coward for a BNS range of Mpc over , with observing runs of several months. Assuming 2015) and binary pulsar BNS (Kim et al. 2012; Petrillo et al. 2013; Siellez et al. 2014; Fong et al. that no unexpected obstacles are encountered, theskialigo detectors are observations expected toforachieve a et al. 2015). runs, circa 2020, it might be desirable to 2015). For NSBH, systems with small BH masses are consid200 Mpc BNS range circa After the first observing Daniel Siegel GRMHD simulations of remnant accretion disks from neutron star mergers 1/13

3 NS mergers remnants SMNS / HMNS long-lived NS NS NS prompt collapse BH - torus BH NS BH - torus sim. & vis.: W. Kastaun BH 2/13

4 NS mergers remnants SMNS / HMNS long-lived NS NS NS prompt collapse BH - torus BH NS sim. & vis.: W. Kastaun BH BH - torus This talk: post-merger accretion disks and r-process nucleosynthesis 2/13

5 3/13 NS post-merger accretion disks: formation Movie: long-term evolution of post-merger accretion disk, MBH=3Msun (spin: 0.8), Mdisk=0.02Msun Movie: BNS merger and formation of post-merger accretion disk courtesy D. Radice

6 NS post-merger accretion disks: numerical setup First self-consistent simulations modeling r-process nucleosynthesis from disk outflows from first principles: GRMHD: magnetic instabilities (MRI) mediating turbulence (transport of angular momentum) in the disk weak interactions in GRMHD approximate neutrino transport (leakage scheme) realistic EOS (Helmholtz EOS) valid at low temperatures and densities, capturing nuclear binding energy release from alpha-particle formation full r-process network calculations on disk outflows using 4 tracer particles (SkyNet; Lippuner & Roberts 2015) Previous Newtonian alpha-disk simulations: Fig.: disk properties; contours: optical depth for electron neutrinos Fernandez & Metzger 2013 Metzger & Fernandez 2014 Fernandez Fernandez Just /13

7 4/13 GRHydro: part of the Einstein Toolkit einsteintoolkit.org

8 NS post-merger accretion disks: numerical setup First self-consistent simulations modeling r-process nucleosynthesis from disk outflows from first principles: GRMHD: magnetic instabilities (MRI) mediating turbulence (transport of angular momentum) in the disk weak interactions in GRMHD approximate neutrino transport (leakage scheme) realistic EOS (Helmholtz EOS) valid at low temperatures and densities, capturing nuclear binding energy release from alpha-particle formation full r-process network calculations on disk outflows using 4 tracer particles (SkyNet; Lippuner & Roberts 2015) Previous Newtonian alpha-disk simulations: Fig.: disk properties; contours: optical depth for electron neutrinos Fernandez & Metzger 2013 Metzger & Fernandez 2014 Fernandez Fernandez Just /13

9 Onset of MHD turbulence Fig.: disk properties; contours: optical depth for electron neutrinos Fig.: Onset of the magnetorotational instability (MRI), showing exponential growth of the magnetic field 5/13

10 Onset of MHD turbulence MRI Fig.: disk properties; contours: optical depth for electron neutrinos Fig.: Onset of the magnetorotational instability (MRI), showing exponential growth of the magnetic field 5/13

11 Onset of MHD turbulence Fig.: disk properties; contours: optical depth for electron neutrinos Fig.: magnetic fields in the disk; contours: rest-mass density magnetic properties very similar to Ciolfi /13

12 B-field effect on EOS and neutrino emission rates assume magnetic field effects become important when cyclotron frequency of electrons and Fermi energy become comparable: ~!c &1 EF 5/13

13 Onset of MHD turbulence average radially for space-time diagram Fig.: disk properties; contours: optical depth for electron neutrinos Fig.: magnetic fields in the disk; contours: rest-mass density magnetic properties very similar to Ciolfi /13

14 Accretion disk dynamo: butterfly diagram magnetic energy is generated in the mid-plane migrates to higher latitudes dissipates into heat off the mid-plane hot corona hot corona launches thermal outflows (neutron-rich wind) NS post-merger accretion disk are cooled from the mid-plane by neutrinos (rather than from the EM photosphere)! 6/13

15 NS post-merger accretion disks: self-regulation neutrinos Neutrino-cooled accretion disks self-regulate themselves to mild degeneracy (low Ye matter): Beloborodov 2003, Chen & Beloborodov 2007, Metzger viscous heating via magnetic turbulence neutrino cooling charged-current processes: e + p! n + e e+ + n! p + e pair annihilation: e + e+! e + e e + e+! µ, + µ, plasmon decay:! e + e! µ, + µ, Fig.: disk properties; contours: rest-mass density 7/13

16 NS post-merger accretion disks: self-regulation neutrinos Neutrino-cooled accretion disks self-regulate themselves to mild degeneracy (low Ye matter): Beloborodov 2003, Chen & Beloborodov 2007, Metzger viscous heating via magnetic turbulence neutrino cooling balance with feedback mechanism: higher degeneracy µe /kt fewer e-, e+ (lower Ye) less neutrino emission, i.e., cooling higher temperatures lower degeneracy µe /kt Fig.: disk properties; contours: rest-mass density direct evidence of self-regulation 7/13

17 The origin of heavy nuclei: r-process nucleosynthesis Movie: r-process nucleosynthesis from NS merger remnant disks 8/13

18 NS post-merger accr. disks: r-process nucleosynthesis 1st peak 2nd peak 3rd peak rare-earth peak 2nd peak 3rd peak rare-earth peak under-production of light nuclei absence of high-ye tail robust 2nd and 3rd peak r-process low outflow velocities 5 9/13

19 /13 Influence of neutrino absorption 1st peak 2nd peak 3rd peak rare-earth peak with neutrino absorption: also good fit to 1st 2nd peak elements robust 2nd and 3rd peak r-process Fig.: total disk neutrino luminosity (top) and neutrino emissivity averaged neutrino temperature at emission production of all r-process elements

20 11/13 BH accretion vs. disk outflows steady wind! accreted mass converged Fig.: accretion rate onto the BH Fig.: number of tracer particles outside a given radius By end of simulation: accreted mass converged but still steady outflows remaining disk mass likely unbound

21 12/13 NS post-merger accretion disks: r-process nucleosynthesis GRMHD disk simulations: significantly higher ejected mass than in previous Newtonian hydro simulations unbound outflows >0.2 Mdisk, likely ~0.4Mdisk for disk in present simulation: > Msun, likely ~ 0.01 Msun disk outflows alone are consistent with constraints on r-process enrichment observations log(required ejecta mass [M sol ]) aligo nsbh rate compendium nsns Bauswein+ 13 nsns Rosswog 13 nsns Hotokezaka+ 13 LIGO nsbh O2 LIGO nsbh O3 LIGO nsns O3 LIGO nsbh O1 pop. synth. nsns Chruslinska+16 sgrb rate Petrillo + 16 LIGO nsns O2 LIGO nsns O1 aligo nsns rate compendium log ( R [yr -1 gal. -1 ] ) [yr -1 Gpc -3 ] ALL r-process r-process A > 80 r-process A> 130 nsbh Foucart+ 14 nsbh Kyutoku+ 13 Rosswog Fig.: constraints on r-process enrichment rates vs. ejected mass NS post-merger disk outflows are promising site for the r-process

22 Conclusions Simulations of NS post-merger accretion disks Siegel & Metzger 2017b, i. prep. GRMHD with weak interactions and approx. neutrino transport first fully self-consistent study of its kind evidence for hot coronae that launch thermal outflows first identification of self-regulation in neutrino-cooled accretion disks suggest NS post-merger systems are robust site of the r-process can produce all r-process elements underproduction of light elements compensated by high-ye material from long-lived NS Wu+ 2016, Just low velocity outflows enable narrow-line spectroscopy to identify composition of r-process matter electromagnetic signature (kilonovae) potentially most promising EM counterpart to GWs JWST GRMHD simulations Short gamma-ray of remnant burstsaccretion in the time-reversal disks from neutron scenario star mergers 13/13