Published October 8, 2025 | Version v1

SEVNbenchmark2401

  • 1. University of Barcelona

Description

SEVN BENCHMARK2401

This repository contains the data products of the SEVN runs "Benchmark2014". In particular, the datasets contain the population of binary compact objects (BCOs, Black holes, BHs, and Neutron stars, NSs) obtained simulating 40 milion binaries (divided in the two equivalent, a part for the initial conditions, datasets) with the rapid poulation syntehsis code SEVN (Iorio+23version 2.7.5).  The BCOs stored in the files are solely the one that will merge within an Hubble time (14 Gyr). 

Repository content 

  • README.md: this file is a copy of this same  Zenodo description
  • SEVN_input.tar: this archive contains the scripts used to generate the initial conditions and to estimate the total simulated mass (corrected for incomplete mass sampling; see below). It also includes the run script used to execute the simulations, with all SEVN parameters specified.
  • bcom_scighera2401.tar.gz: archive containing the BCOs of the scighera2401 dataset (results from the run of 20 million binaries).
  • bcom_leonardo2401.tar.gz: archive containing the BCOs of the leonardo2401 dataset (results from the run of 20 million binaries).

Using this dataset 

If you find this dataset useful for your work, you are welcome to use the entire dataset or any part of it.
If you make use of the data in a published work, please include the following citations:

The Dataset

The datasets contains the BCOs obtained running SEVN (Iorio+23version 2.7.5).  to simulate a total of 40 milion binaries split in two datasets:

  • leonardo2401: set of simulations run on the leonardo hpc @ cineca. The initial condition includes 20 milion binaries for a total mass of $`M_\mathrm{sim}=442538939.33`$ Msun.
  • scighera2401: set of simulations run on the scighera @ fiscia.unipd. The initial condition includes20 milion binaries for a total mass of $`M_\mathrm{sim}=442648579.81`$ Msun.
For each simulation set, we generate a single initial condition (IC) file (see below) and evolve the binaries for 15 different metallicities ZZ (mass fraction of metals):
0.0001, 0.0002, 0.0004, 0.0006, 0.0008, 0.001, 0.002, 0.004, 0.006, 0.008, 0.01, 0.014, 0.017, 0.02, and 0.03.
All other SEVN parameters are fixed to their default values for version version 2.7.5) These parameter values can be found in the run script included in the dataset archives or in the SEVN input archive.
 
 
Important Note. The two datasets are compatible and complementary, as they were generated using the same SEVN version and identical SEVN and sampling parameters (although the initial condition samplings are independent).
You can use either dataset individually or combine several (considering the different metallicities) of them (or even all), including subsamples from specific datasets, depending on your needs.
 

Initial Conditions

The initial conditions were generated with the code IC4popsyn(commit version a24c6e60d19f5b37ccab659baa06efb02f261865), using the following prescriptions:

  • The primary mass follows a Kroupa (2001) initial mass function (IMF)
    m−2.35m−2.35 between 5 and 150 M⊙.

  • The mass ratio q=m2/m1q=m2/m1 follows the power law from Sana et al. (2012),
    q−0.1q−0.1, with qq between qmin=min⁡(2.2/m1,0.1)qmin=min(2.2/m1,0.1) and 1.
    With this choice, the secondary mass is always ≥ 2.2 M⊙.

  • The logarithm of the orbital period, P=log⁡(P/days)P=log(P/days), follows a power law from Sana et al. (2012),
    P−0.55P−0.55, with PP between 0.33 and 5.5.

  • The initial eccentricity follows a power law from Sana et al. (2012),
    e−0.42e−0.42, with ee between 0 and 0.9.

  • The eccentricity–period correlation correction from Moe & Di Stefano (2017) (their Eq. 3) has been applied.

The initial conditions can be generated using the script generate_IC.py in SEVN_input.tar

Correction Factors for the Total Mass

Due to the incomplete sampling of the mass function, the total mass of the parent stellar population represented by the simulations is given by

 

where: Mpop = Msim/(fIMF*fbin)

  • Msim is the total mass of the simulated binaries (including both primary and secondary stars);

  • fIMF accounts for the fraction of the IMF actually sampled;

  • fbin represents the fraction of stellar mass in binaries
    (e.g. fbin=0.2 means that 20 % of the total mass is in binary systems).

Kroupa IMF

Considering the cuts applied in generating the initial conditions, we obtain:

  • fIMF=0.251

  • fbin can be freely chosen.
    If one adopts the binary fraction as a function of primary mass from Moe & Di Stefano (2017),
    this corresponds to an effective fbin=0.569

The mass correction factor can be estimated using the script estimate_mass_correction.py

Chabrier IMF with limit at M=100 Msun 
 
In one of the paper that used the dataset (Marinacci+25), a Chabrier with a limit mass at 100 Msun was assumed.
In order to correct for the different mass range:

- Remove from the bco file all the systems with Mzams_0>100 Msu
- A new correction factor must be included to estimate the popolation total mass. The current total mass of the IC subpopulations with M<100 Msun is 411155226.37 Msun for the leonardo2401 dataset and 411333109.21 Msun for the scighera2401 dataset.
The new correctons factors are:

- fIMF=0.243 
- fbin  can be freely chosen, but if one wants to use the fraction of binary (depending on the mass of the primary star) from Moe&DiStefano17 this is equivalent to an effective fbin=0.628.
 

Data products

The dataset tar archives contain: 
  • BHBHm.csv – CSV file containing all merging binary black holes produced in the simulations across all metallicities.

  • NSNSm.csv – CSV file containing all merging binary neutron stars produced in the simulations across all metallicities.

  • BHNSm.csv – CSV file containing all merging black hole–neutron star binaries produced in the simulations across all metallicities.

  • metadata.yaml – Metadata of the simulations in YAML format.

  • SEVNICxx.info – File summarizing the parameters used to generate the initial conditions (see here).

  • mass_correction_info.txt – File containing the factors required to estimate the total population from the simulated mass, accounting for correction factors related to the IMF and binary fraction.

The properties of the systems are reported at the moment of the formation of the second compact object.
 

Content of the csv files

Column Descriptions

  • ID: [int] Progressive integer ID indicating the position of the binary in the initial conditions file.
    Note: This value alone is not sufficient to uniquely identify a system (see Warning-3).

  • name: [long int] Unique random ID assigned to the binary.
    Note: As with ID, this value alone does not uniquely identify a system (see Warning-3).

  • Z: [float] Metallicity.

  • alpha: [float] Common-envelope efficiency parameter (α<sub>CE</sub>) used in the simulation.

  • BWorldtime: [float, Myr] Age of the system at the time of binary compact object formation (0 = start of the simulation).

  • Mass_0, Mass_1: [float, M⊙] Masses of the compact remnants.

  • Radius_0, Radius_1: [float, R⊙] Radii of the compact remnants.

  • Zams_0, Zams_1: [float, M⊙] ZAMS mass of the last interpolated SEVN track used for the star (not the progenitor’s initial ZAMS mass).

  • Mzams_0, Mzams_1: [float, M⊙] True ZAMS masses of the progenitors.

  • Phase_0, Phase_1: [int] Stellar evolutionary phase (always 7 for compact remnants).

  • RemnantType_0, RemnantType_1: [int] Compact remnant type:

    • –1: massless (not present)

    • 0: not a remnant (not present)

    • 1: HeWD (not present)

    • 2: COWD (not present)

    • 3: ONeWD (not present)

    • 4: ECNS — neutron star formed via electron-capture SN

    • 5: CCNS — neutron star formed via core-collapse SN

    • 6: black hole

  • Xspin_0, Xspin_1: [float] Dimensionless spin parameter of the BH (NaN for NSs).

  • Semimajor: [float, R⊙] Semi-major axis at the moment of binary compact object formation.

  • Eccentricity: [float] Eccentricity at the moment of binary compact object formation.

  • GWtime: [float, Myr] Time to merge via gravitational-wave emission.

  • Semimajor_ini, Eccentricity_ini: [float] Initial orbital parameters (at simulation start).

  • NSt, NSu: [int] Number of stripped and ultra-stripped supernova explosions, respectively.

  • channel: [int] Binary interaction channel:

    • 0 — no interactions (RLO or CE)

    • 1 — stable mass transfer before the first SN, then CE

    • 2 — only stable mass transfer episodes

    • 3 — at least one CE before the first SN; at that moment the system hosts an H-star and a pure-He/naked-CO star

    • 33 — same as 3 but with a single-core CE

    • 4 — at least one CE before the first SN; at that moment both stars are pure-He/naked-CO

    • 44 — same as 4 but with a double-core CE

    • 5 — no interactions before the first SN

  • subchannel: [char] Evolution after the first SN:

    • n — no interactions

    • s — only stable mass transfer episodes

    • c — one CE

    • d — more than one CE

  • collisions: [int] Number of periastron collisions triggered during binary evolution outside RLO phases.

  • CE, CEb, CEa: [int] Number of common-envelope (CE) events in total, before, and after the first SN.

  • SMTa: [int] Number of stable mass transfer episodes after the first SN.

  • Events, EventsSimple, EventsPlus, EventsAll: [str] Encoded sequences of the main evolutionary events of the BCO progenitors.

  • SNtime_0/1, SNMej_0/1, SNvcom_0/1, SNcalpha_0/1, SNtype_0/1: Information on the supernova explosions of each star, including explosion time, ejected mass, post-SN velocity, angular momentum alignment, and SN type.
    SNtype codes:

    • 0 — Unknown

    • 1 — Electron-capture SN

    • 2 — Core-collapse SN

    • 3 — Pulsational pair-instability SN

    • 4 — Pair-instability SN

    • 5 — SN Ia

Warnings

Warning-1: NaN values
NaN values are legitimate in SEVN outputs. In these files, they should appear only in the columns Xspin_0 and Xspin_1.

Warning-2: _0 and _1 suffixes
The suffixes _0 and _1 indicate the position of the star in the initial conditions file.
They do not correspond to the most or least massive component at any stage.
However, by construction, star _0 is always the initially more massive one.

Warning-3: Uniquely identifying a system
All simulations use identical initial conditions with the same random seed; therefore, IDs and names may repeat.
To uniquely identify a system, use the combination of the four columns: ID, name, alpha, and Z.
For this dataset, either ID or name can be used interchangeably, but alpha and Z are still required.

Warning-4: Initial mass of BH/NS progenitors
The progenitor’s true initial ZAMS mass is stored in Mzams_0 and Mzams_1.
Do not use Zams_0 and Zams_1.

Reading the Events

The event columns (Events, EventsSimple, EventsPlus, EventsAll) list, in chronological order, the main evolutionary events of the BCO progenitors.
Each event is separated by a colon (:).
The following describes the possible event keys used in each column.

Column: Events

Keys

  • RB – RLO begins

  • RC – Circularisation at the onset of RLO

  • RE – RLO ends

  • K – Collision at periastron

  • C – Common envelope

  • S – Supernova explosion

 

Column: EventsPlus

Keys

  • RB, RC, RE, K – Same meaning as above

  • Cs – Single-core CE (only one star has a core)

  • Cse – Single-core CE triggered by a pure-He star

  • Cd – Double-core CE (both stars have a core)

  • Cde – Double-core CE triggered by a pure-He star

  • ijSx – Supernova explosion, where:

    • i/j indicate the exploding star and its companion:

      • h — hydrogen star

      • e — pure-He or naked-CO star

      • r — compact remnant

    • x indicates the SN type:

      • t — stripped SN (pure-He progenitor)

      • u — ultra-stripped SN (naked-CO progenitor)

      • (empty) — classical SN

 

Column: EventsAll

Same as EventsPlus, but with additional detail for RLO onset:

  • RBkjc: RLO begins, where

    • k = 0 → star 0 is donor, k = 1 → star 1 is donor

    • j = evolutionary phase of the donor:

      • 1 = main sequence

      • 2 = terminal-age main sequence

      • 3 = shell H-burning

      • 4 = core He-burning

      • 5 = terminal-age core He-burning

      • 6 = shell He-burning

    • c = e if the donor is a pure-He star, empty otherwise

Other keys are the same as in EventsPlus.

Column: EventsSimple

Keys

  • M – Stable mass-transfer episode (begins and ends without CE)

  • C – Common envelope (includes RLOs that directly trigger a CE)

  • K – Collision at periastron directly followed by CE

  • ijS – Supernova explosion, where i/j indicate the exploding star and its companion:

    • h — hydrogen star

    • e — pure-He or naked-CO star

    • r — compact remnant

Files

README.md

Files (316.5 MB)

Name Size
md5:61432641122734b293c03f85f79fd3c9
158.2 MB Download
md5:256ceb871c2db96676fd8fc769c74e43
158.3 MB Download
md5:72233c301c95d83585be7bc867568c40
12.2 kB Preview Download
md5:14e4dd0e6676ffb672a96c1e061f1304
59.9 kB Download

Additional details

Related works

Is cited by
Journal: arXiv:2510.06311 (arXiv)

Funding

European Commission
DEMOBLACK - Demography of black hole binaries in the era of gravitational wave astronomy 770017

Software

Repository URL
https://gitlab.com/sevncodes/sevn/-/releases/V2.7.5
Programming language
C++ , Python
Development Status
Active