compact binary mergers as the origins of r-process elements Shinya - PowerPoint PPT Presentation

compact binary mergers as the origins of r-process elements � Shinya Wanajo (Sophia Univ. / RIKEN iTHES) with Y. Sekiguchi (Toho U), N. Nishimura (Keele U), K. Kiuchi (YITP), K. Kyutoku (RIKEN), M. Shibata (YITP), Y. Ishimaru (ICU), T. Ojima (ICU), N. Prantzos (IAP) Nuclear Physics, Compact Stars, and Compact Star Mergers 2016 October 17-November 18, 2016, YITP, Kyoto, Japan

origin of gold (r-process elements) is sWll unknown… www.carWer.jp NPCSM2016 Wanajo 2

who made the r-process elements? neutron-star mergers core-collapse supernovae (since LaZmer+1974; (since Burbidge+1957; Symbalisty+1982) Cameron 1957) v n-rich ejecta from coalescing v n-rich ejecta nearby proto-NS NS-NS or BH-NS v typical SNe appear to make v recent studies show promise only weak r-process nuclei NPCSM2016 Wanajo 3

“universality” of the r-process surviving old stars record a 0 nucleosynthesis memories 2 in the early universe – 4 Relative log –6 –8 v r-process enhanced –10 stars show constant –12 b 1 abundance paaerns log 0 for 50 < A < 80 –1 Individual stellar abundance offsets with respect to Simmerer et al. (2004) c 1 log 0 v the r-process appears Average abundance offsets with respect to Arlandini et al. (1999) ‘‘stellar model’’ –1 to be robust for A ≥ 56 30 40 50 60 70 80 90 Atomic number and to have variaWons CS 22892-052: Sneden et al. (2003) HD 115444: Westin et al. (2000) Sneden+2008 CS 31082-001; www.eso.org for A < 50 and A > 80 BD+17°324817: Cowan et al. (2002) CS 31082-001: Hill et al. (2002) HD 221170: Ivans et al. (2006) HE 1523-0901: Frebel et al. (2007) NPCSM2016 Wanajo 4

supernovae: not such neutron-rich? Wanajo 2013 Nishimura, Takiwaki, Thielemann 2015 10 -2 2 prompt: B11 β 1.00 M = 1.2 1.4 1.6 1.8 2.0 2.2 2.4 M s u n B12 β 0.25 10 -3 solar r-ab u ndance B12 β 1.00 1 B12 β 4.00 10 -4 delayed: B11 β 0.25 abundance, Y ab u ndance 0 10 -5 10 -6 -1 10 -7 -2 10 -8 -3 10 -9 80 120 160 200 240 mass number, A 50 100 150 200 250 mass n u mber v supernova models (ECSN and v magneWcally driven explosions neutrino-driven wind) explain may produce heavy r-process producWon of only weak r- elements (but depending on elements up to A ~ 110 unconstrained free parameters) NPCSM2016 Wanajo 5

NS merger scenario: most promising? v coalescence of binary NSs expected ~ 10 – 100 per Myr in the Galaxy www.mpa-garching.mpg.de v first ~ 0.1 seconds dynamical ejecWon of n-rich maaer up to M ej ~ 10 -2 M ¤ (today’s talk) v next ~ 1 second neutrino or viscously driven wind from the BH accreWon torus up to M ej ~ 10 -2 M ¤ ?? (see the talk by R. Fernandez) NPCSM2016 Wanajo 6

neutron star mergers: too neutron-rich? Goriely+2011 (also similar results by Korobkin+2011; Rosswog+2013) 10 0 1.35–1.35M o NS 1.35-1.35M o NS Solar 10 -1 1.20-1.50M o NS 10 -2 Mass fraction 10 -3 mass fraction 10 -4 10 -5 10 -6 10 -7 0 50 100 150 200 250 A v fission cycle leads to robust 0.015 0.021 0.027 0.033 0.039 0.045 0.051 r-paaern for only A > 120 with Y e too small A < 120 nuclei Wdal (or weakly shocked) ejecWon of “pure” n-maaer with Y e < 0.1 v fission cycle itself is not “the” r-process NPCSM2016 Wanajo 7

1.3+1.3 M ¤ neutron star merger with full-GR and neutrino transport (SFHo) simulation by Yuichiro Sekiguchi NPCSM2016 Wanajo 8

NPCSM2016 Wanajo 9

r-process in merger ejecta ( Y e = 0.09) ( n, γ ) and β -decay based on HFB-21 NPCSM2016 Wanajo 10

weak interac?on saves merger scenario Wanajo+2014 nucleosynthesis abundances in the ejecta 10 -2 10 0 Y e distribuWon in the ejecta solar r-abundance this work 10 -3 10 -1 mass fraction 10 -4 abundance 10 -2 10 -5 10 -3 10 -6 10 -7 10 -4 0.0 0.1 0.2 0.3 0.4 0.5 10 -8 Y e 250 0 50 100 150 200 250 mass number v good agreement with full solar v positron capture and neutrino r-process range for A = 90-240 absorpWon on free nucleons (similar result by Goriely+2015 result in less neutron-rich but by Radice+2016) ejecta with Y e ~ 0.1-0.45 NPCSM2016 Wanajo 11

Sekiguchi+2015; 1.35+1.35 M ¤ NSs Recent result with finite-temperature EOS � Multi-EOS study (Thanks to M. Hempel ) � GR ¡approximate ¡ν -rad hydro simulation � Adopted EOS 14.5km � TM1 (Shen EOS) � TMA 13.2km � DD2 � IUFSU 11.8km � SFHo Consistent with � NS radius estimation � Chiral effective theory NPCSM2016 Wanajo 12

dependence on EOSs Wanajo+2016; in prep. solar r-abundance 1 -2 10 solar r-abundance 0 -3 TM1 (1.35+1.35) abundance 10 DD2 (1.35+1.35) SFHo (1.35+1.35) -1 -4 10 abundance -5 -2 TM1 (1.35+1.35) 10 DD2 (1.35+1.35) -6 10 -3 SFHo (1.35+1.35) -7 relative to SFHo 10 1.0 -8 0.5 10 0.0 0 50 100 150 200 250 mass number -0.5 -1.0 40 50 60 70 80 90 atomic number v soqer EOS predicts less heavy r-process products, but v effects of EOSs are mild to r-process (good for universality?) NPCSM2016 Wanajo 13

uniqueness of double NS binaries Kiziltan+2013 NS+WD NS+NS NS+WD NS+NS v binaries have various NS masses (1.2-2.0 M ¤ ), but for v double NS binaries (~ 1.21-1.43 M ¤ , but see MarWnez+2015) NPCSM2016 Wanajo 14

Sekiguchi+2016; 1.35+1.35, 1.30+1.40, 1.25+1.45 M ¤ Unequal mass NS-NS system: SFHo1.25-1.45 � Orbital plane : Tidal effects play a role, ejecta is neutron rich � Meridian plane : shock + neutrinos play roles, ejecta less neutron rich 1.25 + 1.45 M ¤ NPCSM2016 Wanajo 15

dependence on mass ra?os (SFHo) Wanajo+2016; in prep. solar r-abundance 1 -2 10 solar r-abundance 0 -3 SFHo (1.25+1.45) abundance 10 SFHo (1.30+1.40) SFHo (1.35+1.35) -4 -1 10 abundance -5 SFHo (1.25+1.65) -2 10 SFHo (1.25+1.55) SFHo (1.25+1.45) -6 10 -3 SFHo (1.30+1.40) SFHo (1.35+1.35) relative to 1.3+1.4 -7 10 1.0 0.5 -8 10 0.0 0 50 100 150 200 250 mass number -0.5 -1.0 40 50 60 70 80 90 atomic number v small asymmetry predicts small variaWon in light r-process products v uniqueness of the double NSs may be the origin of the universality? NPCSM2016 Wanajo 16

Serious Problem in Chem. Evol. with NSMs Long merger timescale ( ～ 100 Myr) could cause the “delayed” appearance of Eu !? Matteucci+2013 � Argast+2004 � t NSM = 100 Myr � t NSM : 1 10 100 Myr � 2 1 [Eu/Fe] 0 -1 -5 -4 -3 -2 -1 0 [Fe/H] Komiya+2014 � Tsujimoto & Shigeyama 2014 � t NSM ∼ 10 Myr � <-2.5 2.5 2 t NSM = 10 Myr � NSM7 NSM7s 2 1 1.5 1 [Eu/Fe] 0 [Ba/Fe] 0.5 0 -1 -0.5 -1 -2 -1.5 <-2.5 -4 -3 -2 -1 0 [Fe/H] -4 -3 -2 -1 [Fe/H] NSMs with long merger time cannot explain observed scatters in metal poor stars?? � NPCSM2016 Wanajo 17

Forma?on Scenario of Sub-halos One of the most plausible formation scenarios of dwarf galaxies: As stars are formed, the ISM is ejected from a galaxy by SNe because of shallow grav. potential. The key parameters are and Star Formation Gas Outflow Rate (SFR) Rate (OFR) Basic chemical evolution suggests <[Fe/H]> ∝ SFR OFR if IMF is universal. (e.g., Pagel 1991, Prantzos 2008) NPCSM2016 Wanajo 18

rareness of mergers in low-mass sub-halos the total number of NSMs in low mass sub-halos must be extremely small!! Ishimaru, Wanajo, Prantzos 2015 10 8 10 7 10 6 10 5 10 4 In case of 10 4 M ¤ sub-halos, the average number of NSMs is ~0.1 It means only one sub-halo out of ten experiences a NSM event and stars in such sub-halo should show strong enhancement in Eu NPCSM2016 Wanajo 19 k k � �

NPCSM2016 Wanajo 20

StochasWc Chemical EvoluWon of sub-halos with NSMs Ojima, Ishimaru, Wanajo, & Prantzos, in prep. Based on this scenario, we examine enrichment of each sub-halo by NSMs, using a Monte-Carlo method. According to the sub-halo mass function; dN/dM * ∝ M * - 1 .7 , total number of model sub-halos which form the Galactic halo are given as follows: stellar 10 8 mass 10 4 - 10 5 10 5 - 10 6 10 6 - 10 7 10 7 - 10 8 - 2x10 8 [M ¤ ] Num. of 741 147 29 6 1 sub-halos Mean Num. of 0.174 1.75 19.1 184 694 NSMs/SH NPCSM2016 Wanajo 21

the best fit model ( t NSM = 100 Myr) SFR / M gas ∝ ( M * ) + 0.2 OFR / M gas ∝ ( M * ) - 0.1 5.5 5 2 2 4 [Eu/Fe] 1 [Ba/Fe] 1 log 10 (n * ) [Eu/Fe] [Ba/Fe] 0 0 -1 -1 1 -2 -2 0 -4 -3 -2 -1 0 -4 -3 -2 -1 0 [Fe/H] [Fe/H] If the Galactic halo was formed from sub-halos with mass-dependent SFR & OFR, NSMs with long coalescence time (~100 Myr) can well explain observed [r/Fe] in metal-poor stars NPCSM2016 Wanajo 22

ultra-faint dwarf (UfD) : Re?cullum II 10 4 - 1 0 4.1 M ¤ sub halos ● , ● : Reticulum II (Roederer+16, Ji+16) NSMs occur in 9 out of 138 SHs M sub = 10 4 M sol (a) (a) (a) (b) (b) (c) -2 -1 0 1 2 -2 -1 0 1 2 -2 -1 0 1 2 [r/Fe] Sr Sr Sr Ba Ba Eu -4 -3 -2 -1 0 -4 -3 -2 -1 0 -4 -3 -2 -1 0 -4 -3 -2 -1 0 -4 -3 -2 -1 0 -4 -3 -2 -1 0 [Fe/H] [Fe/H] [Fe/H] 0 1 2 4 5 5.5 log 10 (n * ) In particular, this scenario predicts 1 out of 10 UfDs (~10 4 M ¤ ) shows extremely high [r/Fe], which is consistent with observational data of UfDs! NPCSM2016 Wanajo 23