At the time of writing there are at least 55 published progenitor theories for FRBs. Models for FRB progenitors can be grouped along several lines: repeating or non-repeating, long-lived or cataclysmic source, nearby or cosmological, rotationally or magnetically powered, etc. Many progenitor theories involve compact objects, the processes involved in their birth, or the medium surrounding them. Here we explore the models in more detail, grouped by the primary source involved, and in some cases splitting the category up further by looking at isolated or interacting/colliding mechanisms to generate the radio pulse. A tabular summary of existing FRB theories is maintained on the FRB Theory Catalogue.
9.1. Neutron star progenitors
The majority of current FRB progenitor theories involve neutron stars. Their large rotational energies and strong magnetic fields, as well as the often turbulent environments they occupy, make them plausible candidates for the progenitors of FRBs and some characteristics of FRB emission appear similar to radio pulsars (see also Section 8). Here we discuss the FRB progenitor theories that predict bright radio pulses from extragalactic neutron stars – grouping by models that invoke isolated neutron stars (Section 9.1.1), neutron stars interacting with other bodies or their environment (Section 9.1.2), and neutron stars colliding with other compact objects (Section 9.1.3).
9.1.1. Isolated neutron star models
A number of theories argue that FRBs can be generated by isolated neutron stars, either via beamed radio emission from their magnetosphere, during the collapse of a supramassive neutron star due to its own gravity, or by relativistic shocks in the surrounding medium.
Both Cordes and Wasserman (2016) and Connor et al. (2016b) theorize that some rotationally powered pulsars can produce FRBs as part of their normal emission process, from supergiant pulses from young neutron stars in the case of Connor et al., and from nano-shot giant pulses in the case of Cordes and Wasserman. Lyutikov et al. (2016) have proposed that young rotationally powered neutron stars with millisecond rotation periods could also produce FRBs from the open magnetic field lines at the poles that generate the normal radio emission. Additionally, Katz (2017) has suggested that FRBs may originate from radio pulsars with unstable rotational axes that result in ‘wandering beams’ on the sky. Other theories have argued that FRBs are generated from the magnetically powered neutron stars with ultra-strong magnetic fields known as magnetars. Popov and Postnov (2010) proposed that an FRB might be generated during a magnetar hyperflare and Wang et al. (2018) theorized that FRBs are generated in starquakes on the surface of a magnetar. Lieu (2017) predicts a single bright radio pulse generated seconds after the birth of a magnetar with a millisecond rotation period, whereas Metzger et al. (2017) predict repeating pulses from a stably emitting young millisecond magnetar in a dense supernova remnant. Metzger et al. (2019) theorize that FRBs are produced through maser emission in the ultra-relativsitic shocks through the ionized medium surrounding a young magnetar; this model also predicts a significant RM contribution from propagation through the highly magnetized outer layers of the mangetar wind nebula.
Cataclysmic models involving isolated neutron stars include the ‘blitzar’ model, where an FRB is produced by a supramassive neutron star as it collapses to form a black hole decades or centuries after its creation in a supernova explosion (Falcke and Rezzolla, 2014). Similarly, Zhang (2014) proposed a comparable collapse mechanism, but happening in the seconds or minutes after the supramassive neutron star or magnetar is formed in a binary neutron star merger, coincident with a short GRB. Fuller and Ott (2015) have proposed that FRBs are generated by isolated neutron stars whose collapse is triggered by dark matter capture in the neutron star core.
In almost all cases, the neutron star is not associated with any other observable stable body. In the case of a flare or collapse after birth in a supernova or binary neutron star merger, the FRB might be associated with multi-wavelength emission either in the form of an X-ray flare from a magnetar as is observed in our own Galaxy (Kaspi and Beloborodov, 2017), the multi-wavelength emission from a supernova such as an optical or radio afterglow (Metzger et al., 2017), or the prompt emission from a binary neutron star merger such as a short GRB (Zhang, 2014). For a young magnetar ejecta model, the supernova that created the magnetar may also produce an X-ray or γ-ray afterglow (Metzger et al., 2019).
9.1.2. Interacting neutron star models
Additionally, several models explaining FRBs invoke the interaction between a neutron star and its environment or a less massive orbiting body. In these cases, the FRB emission is generated in the neutron star magnetosphere or through a triggered reaction from the interaction of the two bodies.
Similar to the theories involving isolated neutron stars, many such theories involve relatively normal rotationally powered neutron stars in other galaxies. Egorov and Postnov (2009) propose that FRBs are generated by magnetic reconnection of the neutron star after being struck by an energetic supernova shock and Zhang (2017a) invoke a ‘cosmic comb’ of fast-moving plasma hitting the magnetosphere of a neutron star, which triggers radio emission. Close approach between a neutron star and a supermassive black hole (Zhang, 2017b) or a pair of neutron stars in central stellar clusters of galactic nuclei (Dokuchaev and Eroshenko, 2017) have also been proposed.
In several models, FRBs are produced as the result of accretion onto a neutron star. van Waerbeke and Zhitnitsky (2018) invoke magnetic reconnection after a magnetar accretes dark matter. Istomin (2018) proposed that FRBs are created as a neutron star accretes ionized plasma blown off of another body in a close approach, and Gu et al. (2016) propose that FRBs are generated as a neutron star accretes material from a white dwarf companion that has overflowed its Roche lobe. Neutron stars interacting with small bodies such as comets or asteroids are also a common theme: e.g., neutron stars traveling through asteroid belts (Dai et al., 2016), asteroids or comets impacting the surface of the neutron star (Geng and Huang, 2015) or rocky bodies orbiting a neutron star within the magnetosphere (Mottez and Zarka, 2014). Finally, Lyubarsky (2014) proposed that a magnetically powered hyperflare from a magnetar is released and then interacts with the surrounding medium to produce an FRB in the forward shock.
9.1.3. Colliding neutron star models
Lastly, a few neutron star theories predict that an FRB pulse is generated at the time of collision between a neutron star and another compact object. Lyutikov (2013) predicts an FRB from the precursor wind of a binary neutron star merger, Totani (2013) predicts an FRB from the magnetic braking associated with the same event, and Yamasaki et al. (2018) predicts FRBs from a neutron star produced in the merger. An FRB from coherent curvature radiation in a binary neutron star merger has also been predicted by Wang et al. (2016). Dokuchaev and Eroshenko (2017) take this argument one step further and predict FRBs from binary neutron star collisions only in or near the center of densely packed stellar clusters in galactic nuclei, and Iwazaki (2015) theorizes that FRBs are generated in the collision between a neutron star and a dense axion star. Alternatively, Liu (2017) proposes that FRBs are produced in neutron star – white dwarf collisions.
9.2. Black hole progenitors
Although not as numerous as theories involving neutron stars, several theories have also been put forward proposing black holes as the engines of FRB production. Even before the identification of FRBs as a source class, Rees (1977) predicted observable millisecond-duration radio pulses from evaporating black holes both in the Galaxy and from other galaxies.
Black holes interacting with their surrounding environment have also been proposed. Vieyro et al. (2017) predict FRBs from the interaction between the jet of an accreting active galactic nucleus and the surrounding turbulent medium. Similarly, Das Gupta and Saini (2017) propose a model where a Kerr black hole produced from the collapse of a supramassive neutron star interacts with the surrounding environment to produce multiple repeating FRBs. Stellar mass black holes in binaries have been proposed to produce FRBs by Yi et al. (2018) through collisions of clumps in the jet produced during accretion.
Collisional progenitor theories involving black holes are limited since binary black hole mergers are thought to produce little or no emission in the electromagnetic spectrum. However, Zhang (2016) proposes that a binary black hole merger where one or both of the black holes carries charge could produce an FRB pulse at the time of coalescence. Additionally, Abramowicz et al. (2017) predict the production of an FRB through magnetic reconnection in the event of collisions between primordial black holes and neutron stars in galaxy dark matter halos, and Mingarelli et al. (2015) predict double-peaked FRBs as a precursor to some black hole–neutron star mergers. Li et al. (2018) predict FRBs from the accretion disk produced after a black hole – white dwarf collision.
In the progenitor models above, which all invoke black hole engines for FRB emission, no additional observable emission is predicted either in the radio band or in other parts of the electromagnetic spectrum. Even black hole mergers are expected to be electromagnetically weak and in the progenitor theories included here the radio pulse is the only observable electromagnetic emission predicted from the interaction or merger.
9.3. White dwarf progenitors
Only two models currently exist for the production of an FRB from one or more white dwarfs. The model of Gu et al. (2016) mentioned in Section 9.1.2 predicts an FRB from the accretion of material from a Roche-lobe-filling white dwarf onto a neutron star. White dwarfs alone have difficulty accounting for the energy budget required to generate a bright millisecond radio pulse visible at Mpc or Gpc distances. Moriya (2016) has predicted an FRB from the accretion-induced collapse of a white dwarf where the burst is produced in the strong shock from the explosion ejecta colliding with the circum-stellar medium. Kashiyama et al. (2013) also predict that a single FRB could be produced at the polar cap of a massive white dwarf formed in a binary white dwarf merger.
In the cases mentioned above, the FRB might also be associated with optical or radio synchrotron emission produced in the expanding ejecta from the stellar collapse or merger. However, these signatures may be too faint to detect in other galaxies as the energy budget for white dwarfs is much lower than that of typical neutron stars.
9.4. Exotic progenitors
There are a number of models for FRBs that do not neatly fall into the categories listed above. The only Galactic model currently proposed is that FRBs originate in activity from Galactic flare stars (Loeb et al., 2014) and that the excess DM from the FRB is accrued in the ionized stellar corona. All other theories propose an extragalactic origin and invoke rare or exotic phenomena to generate FRB pulses.
Some of these exotic models still feature dense compact objects and theorize that, for example, an FRB is generated when a primordial black hole explodes back out as a white hole (Barrau et al., 2014) or that the interaction between a strange star (a star made of strange quarks) and a turbulent wind might produce FRBs (Zhang et al., 2018). Others have proposed that the collapse of a strange star to form a black hole could generate an FRB similar to the model for a neutron star by Falcke and Rezzolla (2014), or that an isolated neutron star collapsing to form a quark star in a ‘quark nova’ could produce a millisecond radio pulse (Shand et al., 2016).
Still other models are arguably even more exotic, theorizing that FRBs come from superconducting cosmic strings (Vachaspati, 2008, Ye et al., 2017, Cao and Yu, 2018), the decay of cosmic string cusps (Zadorozhna, 2015, Brandenberger et al., 2017), superconducting dipoles either in isolation or orbiting around supermassive black holes (Thompson, 2017), or the decay of axion miniclusters in the interstellar media of distant galaxies (Tkachev, 2015). Both Romero et al. (2016) and Houde et al. (2018) theorize that clusters of molecules in other galaxies could produce FRBs: from cavitons in a turbulent plasma excited by a jet (Romero et al., 2016) or through maser-like emission known as Dicke superradiance (Houde et al., 2018). It has even been proposed that FRBs are the signatures of beamed emission powering light sails of distant spacecraft (Lingam and Loeb, 2017).
9.5. Differentiating between progenitor models
A much larger sample of FRBs, with well characterized burst properties and robustly identified hosts, is needed to differentiate between the dozens of proposed progenitor theories described above.
CHIME and other wide-field FRB discovery machines will provide a large sample in the coming years, but it is also important to have detailed characterization of bursts – e.g. full polarimetric information and time resolution that is not limited by instrumental smearing. The shortest-possible timescale for FRB emission is currently poorly constrained. It is also important to explore the detectability and properties of FRBs across the full possible range of radio frequencies and to continue to search for prompt multi-wavelength and multi-messenger counterparts. Repeating FRBs provide a practical advantage for detailed characterisation via follow-up observations, but detailed characterization of the properties of apparently non-repeating FRBs is also required. This means that real-time voltage buffers are highly valuable.
The statistics provided by a sample of hundreds to thousands of FRBs can better quantify how common repeaters are, and their range of activity level. Through sheer statistics, it may be possible to convincingly show that there are distinct populations of repeaters and non-repeaters – as opposed to a wide spectrum of activity levels from a population of FRBs that are all capable of repeating, in principle. The distribution of dispersion measures will go some way towards quantifying the spatial distribution of FRBs, but this is still complicated by the unknown host contribution.
ASKAP and other precision-localisation machines will deliver a much larger sample of FRBs with unambiguous host galaxy associations. The local environment and host galaxy type are powerful diagnostics, and precision localisations also enable deep searches for associated persistent emission from radio to high-energies.
As the distributions of FRB properties become better known, this will better inform observational strategies that optimize discovery rate, and it may even lead to the discovery of new FRB-like signals by exploring different areas of parameter space.