Circinus X-1: Young Neutron-Star Binary

• Cir X-1 is a young neutron-star X-ray binary and microquasar characterized by a 16.5-day eccentric orbit and pronounced, phase-dependent accretion behavior.
• Its jets are resolved on scales from milliarcseconds to parsecs, displaying both mildly relativistic ejecta and indications of an unseen ultrarelativistic component.
• The association with a natal supernova remnant provides key insights into post-supernova binary evolution and challenges traditional models of accretion geometry.

Circinus X-1 (Cir X-1) is a confirmed neutron-star X-ray binary and microquasar distinguished by a 16.5–16.6 day eccentric orbit, extreme orbital modulation from X-rays to radio, resolved jets on scales from milliarcseconds to parsecs, and association with a natal supernova remnant that makes it the youngest known X-ray binary (Heinz et al., 2013, Heinz et al., 2015, Rankin et al., 2023). It has also been one of the most structurally ambiguous Galactic accretors: it has exhibited both atoll and Z-state properties, has been modeled at different times with distances from a few to $\sim 10$ kpc, and has supported apparently incompatible inferences about jet speed and inclination, from mildly relativistic radio-emitting ejecta close to the plane of the sky to an unseen ultrarelativistic flow with $\Gamma \gtrsim 22$ (Torok et al., 2010, Miller-Jones et al., 2011, Heinz et al., 2015).

1. System identity and basic parameters

Cir X-1 is identified as a neutron-star system by type I X-ray bursts, and several studies treat it as a weakly magnetized neutron-star X-ray binary with strong orbital-phase-dependent accretion and outflow behavior (Wang et al., 2013, Rankin et al., 2023). Its orbital period is consistently given as about 16.5–16.6 days, while published eccentricities span a broad range: $e \sim 0.45$ is used in recent work on the young supernova remnant and IXPE polarimetry, whereas values up to $e \sim 0.7$–0.9 were invoked in earlier accretion-disk and jet studies (Heinz et al., 2013, Wang et al., 2013, Rankin et al., 2023).

Quantity	Reported value	Source context
Orbital period	$P_{\rm orb} \simeq 16.5$–16.6 d	Binary ephemeris and phase-resolved studies
Eccentricity	$e \sim 0.45$ to $0.9$	Companion, SNR, and accretion interpretations
Distance	$4$–$11$ kpc in literature; $9.4^{+0.8}_{-1.0}$ kpc from light echo	Historical debate and kinematic distance
Age	$\Gamma \gtrsim 22$0 yr	Natal supernova remnant
Compact object	Neutron star	Type I bursts

A major shift in the system’s basic interpretation came from the X-ray light-echo analysis of the 2013 flare, which associated four dust-scattering rings with molecular clouds and derived a kinematic distance $\Gamma \gtrsim 22$1 (Heinz et al., 2015). At that distance, the source is frequently super-Eddington, and previously inferred apparent jet speeds imply extremely strong relativistic constraints if those pattern speeds track the flow directly (Heinz et al., 2015).

The system’s phenomenology extends well beyond the compact binary. Cir X-1 is surrounded by an arcminute-scale radio nebula and exhibits both radio and X-ray jets; later work showed that at least part of the large-scale environment is more naturally interpreted as a supernova remnant shell than as a pure jet-inflated lobe (Sell et al., 2010, Heinz et al., 2013). This dual status—as both an actively accreting microquasar and a very young post-supernova binary—is central to most current interpretations.

2. Binary orbit, companion star, and optical modulation

The orbital cycle organizes much of the observed behavior. Phase $\Gamma \gtrsim 22$2 is periastron and phase $\Gamma \gtrsim 22$3 apastron in the ephemeris used for phase-resolved X-ray spectroscopy (Wang et al., 2013). Near periastron, the system commonly shows X-ray dips, strong flares, enhanced local absorption, and radio activity; between periastron and apastron, the accretion flow evolves rapidly, and after apastron it is often more stable (Wang et al., 2013, Moin et al., 2011, D'Aì et al., 2012).

Optical work places unusually strong constraints on the companion. Gemini observations showed that the optical counterpart varies by 1.2 magnitudes in four days, with a light-curve shape closely resembling that seen in the 1980s despite large secular changes in X-ray and radio brightness (Johnston et al., 2015). The same study found strong, variable H$\Gamma \gtrsim 22$4 emission with multiple components, weak He I emission, and no secure photospheric absorption features from the donor, indicating that the optical continuum is dominated by reprocessed emission rather than a clean stellar spectrum (Johnston et al., 2015).

Modeling the optical light curve as X-ray irradiation of a Roche-lobe-filling companion favors a moderate eccentricity, $\Gamma \gtrsim 22$5, rather than the larger values often assumed in older jet and disk work (Johnston et al., 2015). The optical modulation can be reproduced if the companion is relatively cool and irradiated by an X-ray source with $\Gamma \gtrsim 22$6, whereas hotter-donor solutions require much larger luminosities (Johnston et al., 2015). The same analysis derived extinction estimates of $\Gamma \gtrsim 22$7 mag from diffuse interstellar bands and $\Gamma \gtrsim 22$8 mag from the Balmer decrement (Johnston et al., 2015).

The companion remains problematic in ordinary stellar-evolution terms. Combining the distance, extinction, orbital period, and the requirement that the donor fill its Roche lobe at periastron gives a mean density $\Gamma \gtrsim 22$9, yet the optical magnitude implies a star too faint to be a normal supergiant of that density and too young to be a standard low-mass evolved donor (Johnston et al., 2015). The proposed resolution is that the companion is over-luminous and under-dense because it is still thermally disturbed by the supernova that occurred less than $e \sim 0.45$0 yr ago (Johnston et al., 2015, Heinz et al., 2013). This interpretation is consistent with the independently inferred youth of the system.

3. Inner accretion flow and X-ray diagnostics

Phase-resolved X-ray spectroscopy has shown that the inner flow is strongly non-stationary over the eccentric orbit. RXTE analysis with an Eastern-model continuum found clear orbital modulation of the inferred multicolor-disk inner radius $e \sim 0.45$1: an abrupt increase at periastron, a rapid decrease from phase $e \sim 0.45$2 to $e \sim 0.45$3, and relative stability from apastron back to periastron (Wang et al., 2013). In the same work, the disk luminosity deviated from the constant-radius relation $e \sim 0.45$4, instead following approximately $e \sim 0.45$5, which was attributed to a slow increase in $e \sim 0.45$6 combined with a rapid decrease in $e \sim 0.45$7 as the neutron star moved away from the companion (Wang et al., 2013). The authors interpreted this as the response of the disk to strong tidal perturbation and radiation pressure in a highly eccentric system.

The 2010 May–June outbursts provided unusually complete broad-band coverage of the source over about four orbital cycles. RXTE/PCA, Swift/XRT, and a Chandra/HETGS snapshot showed that the X-ray continuum could be fitted throughout by thermal Comptonization with variable neutral local absorption, without any statistically required additive component (D'Aì et al., 2012). The major May outburst decayed linearly until an “ankle” at $e \sim 0.45$8, where the source underwent a clear transition from optically thick Comptonization to an optically thin state (D'Aì et al., 2012). At the same time, near-periastron intervals required very large partial-covering columns, $e \sim 0.45$9, demonstrating a dense and highly structured local absorber (D'Aì et al., 2012).

The source also displays unusually low-frequency twin kHz QPOs. Reanalysis of the Boutloukos et al. detections with the relativistic precession model in Kerr spacetime showed that the mass cannot be determined independently of the spin: acceptable fits lie along

$e \sim 0.7$0

with feasible masses extending up to $e \sim 0.7$1 for $e \sim 0.7$2 (Torok et al., 2010). The same study emphasized that Cir X-1’s large frequency ratios $e \sim 0.7$3–4.5 place the QPO production radius relatively far from the ISCO, making the mass–spin degeneracy robust within that model (Torok et al., 2010).

IXPE added polarization as a direct probe of accretion geometry. Two 2023 orbital segments yielded polarization degrees of $e \sim 0.7$4 and $e \sim 0.7$5, but with polarization angles of $e \sim 0.7$6 and $e \sim 0.7$7, corresponding to a rotation of $e \sim 0.7$8 along the orbit (Rankin et al., 2023). When binned by hardness ratio, the polarization angle differed by $e \sim 0.7$9 between the lowest and highest hardness states (Rankin et al., 2023). The proposed interpretation is a misalignment between the symmetry axes of the accretion disk and the Comptonizing region, plausibly reflecting misalignment of the neutron-star spin with the orbital angular momentum in an exceptionally young system (Rankin et al., 2023).

4. Jets and outflows across scales

Cir X-1 is one of the few neutron-star systems with resolved jets from AU to parsec scales. On milliarcsecond scales, Australian LBA imaging at 8.4 GHz produced the first resolved images of the inner jets: a symmetric structure extending from $P_{\rm orb} \simeq 16.5$0 mas to $P_{\rm orb} \simeq 16.5$1–20 mas, corresponding to $P_{\rm orb} \simeq 16.5$2 au at 7.8 kpc, aligned along PA $P_{\rm orb} \simeq 16.5$3 east of north and unresolved across the minor axis, implying an opening angle $P_{\rm orb} \simeq 16.5$4 (Miller-Jones et al., 2011). Structural evolution during the observation implied combined proper motions of $P_{\rm orb} \simeq 16.5$5–35 mas d$P_{\rm orb} \simeq 16.5$6, which the authors interpreted as only mildly relativistic radio-bright ejecta (Miller-Jones et al., 2011).

A companion ATCA millimeter study at 33–35 GHz resolved sub-arcsecond jet structure and detected a north-west component $P_{\rm orb} \simeq 16.5$7 arcsec from the core (Calvelo et al., 2011). That component could not be associated with the same orbit’s periastron flare; if linked to a previous orbit, the implied proper motion is of order $P_{\rm orb} \simeq 16.5$8–60 mas d$P_{\rm orb} \simeq 16.5$9, broadly consistent with mildly relativistic flow rather than the extreme values inferred in earlier timing arguments (Calvelo et al., 2011). The paper also argued that jet orientation may vary with distance from the core, or that the source’s precession parameters changed (Calvelo et al., 2011).

Arcsecond- and arcminute-scale radio imaging later showed the jet axis twisting with scale. ATCA imaging from 2.1 to 35 GHz was fitted with a Hjellming–Johnston precession model whose best-fit parameters were $e \sim 0.45$0, $e \sim 0.45$1, precession cone half-opening angle $e \sim 0.45$2, and period $e \sim 0.45$3 d, if precession is the correct interpretation (Coriat et al., 2019). That model explains the observed oscillatory change in projected position angle from the inner jets to the larger-scale lobes, although environmental bending within the surrounding remnant remains an alternative or additional effect (Coriat et al., 2019).

On still larger scales, deep Chandra imaging revealed parsec-scale bipolar X-ray shocks, or “caps,” coincident with the radio jets (Sell et al., 2010). Their spectra were interpreted as cooled synchrotron emission, with a cooling age of approximately $e \sim 0.45$4 yr and a jet-power estimate between $e \sim 0.45$5 and $e \sim 0.45$6 (Sell et al., 2010). These features were taken to be terminal shocks where the jets run into the surrounding medium, making Cir X-1 the only known microquasar in that study to show stationary large-scale X-ray emission (Sell et al., 2010).

The radio phenomenology is equally distinctive. A complete 5.5/9 GHz orbit during a historically faint epoch showed a stable pre-flare baseline of $e \sim 0.45$7 mJy at 5.5 GHz and $e \sim 0.45$8 mJy at 9 GHz, followed by a periastron flare with measured peak decline fluxes of $e \sim 0.45$9 mJy at 5.5 GHz and $0.9$0 mJy at 9 GHz (Calvelo et al., 2011). KAT-7 monitoring later documented the return to Jansky-level flares, re-establishing that the source can recover the radio-bright behavior seen in the 1970s (Armstrong et al., 2013).

5. Natal supernova remnant and system youth

The identification of the surrounding shell as Cir X-1’s natal supernova remnant fundamentally changed the evolutionary picture. Deep Chandra and ATCA imaging showed a roughly circular diffuse X-ray halo and an edge-brightened, asymmetric radio shell centered on the binary (Heinz et al., 2013). The outer X-ray spectrum is thermal rather than synchrotron-dominated: a power law was strongly ruled out, while non-equilibrium ionization models such as PSHOCK, NEI, and SEDOV fit well, with prominent Mg, Si, and S lines and $0.9$1 (Heinz et al., 2013).

The remnant radius is $0.9$2, corresponding to $0.9$3 pc, and the Sedov fit gives a shock temperature $0.9$4 with a robust $0.9$5 lower limit $0.9$6 (Heinz et al., 2013). That lower limit implies a conservative age constraint

$0.9$7

while the formal best fit is $0.9$8 (Heinz et al., 2013). Cir X-1 was therefore identified as the youngest known X-ray binary and the only firmly established Galactic accreting neutron-star binary still inside its natal remnant (Heinz et al., 2013).

This youth has several consequences. Because the system produces type I bursts and lacks pulsations, the neutron star’s surface field must be far below the $0.9$9 scale of standard young high-field pulsars (Heinz et al., 2013). The remnant age then implies either low-field birth or very rapid accretion-driven field burial (Heinz et al., 2013). The same work argued that the arcminute-scale radio nebula should be reinterpreted as synchrotron emission from the supernova-remnant forward shock rather than as a pure jet-inflated bubble, with the present jets embedded within the remnant interior (Heinz et al., 2013).

A later morphological re-interpretation went further. Using MeerKAT images, a 2025 study proposed that the opposite rings and the large southern blowout in the “Africa Nebula” were created not by the present binary jets but by multiple jet episodes during the core-collapse explosion itself, within the jittering jets explosion mechanism (Soker et al., 14 Aug 2025). That work treated the ring interpretation explicitly as a proposal, supported by three-dimensional hydrodynamical simulations of late explosion jets inside an expanding shell (Soker et al., 14 Aug 2025). It therefore belongs to the class of current morphological hypotheses rather than settled system parameters.

6. Distance, geometry, and unresolved controversies

Cir X-1 remains a system in which different diagnostics emphasize different pieces of the flow. The best-developed example is the jet-speed controversy. Direct radio imaging on AU scales and structural evolution over hours to days favor mildly relativistic radio-emitting components, often with relatively symmetric brightness on the two sides of the core (Miller-Jones et al., 2011, Calvelo et al., 2011). In contrast, earlier timing arguments tied to downstream brightenings implied a much faster unseen flow, and when combined with the kinematic distance $4$0 kpc they yield $4$1 and $4$2 (Heinz et al., 2015). Objective synthesis therefore requires distinguishing between the radio-bright ejecta and a potentially dark ultrarelativistic energy-carrying flow, a distinction explicitly considered in both the LBA and e-VLBI analyses (Miller-Jones et al., 2011, Moin et al., 2011).

High-resolution e-VLBI monitoring over a full orbit provides a related constraint. Compact 1.4–1.7 GHz emission was detected only after periastron, at phases following periastron passage, and was absent at other orbital phases down to sub-mJy levels (Moin et al., 2011). This ruled out a persistent compact radio core at the sensitivity of that campaign and strengthened the interpretation that AU-scale compact emission is an orbital-phase-dependent flare product rather than a constant feature (Moin et al., 2011).

A second major controversy concerns system orientation. Optical and X-ray work, including dips, winds, and irradiation modeling, tends to favor a relatively high-inclination accretion geometry and a moderate eccentricity $4$3 (Johnston et al., 2015, Rankin et al., 2023). Some radio imaging likewise favors mildly relativistic ejecta close to the plane of the sky (Miller-Jones et al., 2011, Coriat et al., 2019). Yet the light-echo distance paper sharpened the case that, if the historically inferred arcsecond-scale pattern speeds are interpreted literally, the fastest jet channel must lie within a few degrees of the line of sight (Heinz et al., 2015). The most economical way to reconcile these statements is not to force a single axis or a single outflow speed onto all observables, but to allow stratified flow, changing jet direction, and misalignment between orbital, disk, neutron-star, and jet axes; several of the recent studies argue in precisely that direction (Coriat et al., 2019, Rankin et al., 2023).

The present research picture is therefore not one of a tidy canonical neutron-star binary, but of a strongly non-axisymmetric, very young, eccentric accretor whose observable properties depend on which scale and which tracer is examined. Its accretion disk expands and contracts with orbital phase, its optical flux is dominated by irradiation, its radio behavior alternates between faint and Jansky-level flaring epochs, its jets are resolved from AU to parsec scales, and its large-scale environment still preserves the imprint of the supernova that created it (Wang et al., 2013, Armstrong et al., 2013, Heinz et al., 2013). That combination makes Cir X-1 both exceptionally difficult to parameterize and unusually valuable as a test case for post-supernova binary evolution, neutron-star accretion geometry, and the coupling of relativistic outflows to a very young ambient medium.