T Coronae Borealis (T CrB) Overview

• T Coronae Borealis is a symbiotic recurrent nova characterized by a massive white dwarf accreting from an evolved M giant through Roche-lobe overflow.
• Detailed orbital and accretion studies reveal a nearly circular binary with prominent pre- and post-eruption phases, impacting mass transfer and disk evolution.
• Multiwavelength and high-energy observations provide diagnostic insights into hydrodynamic interactions, nucleosynthesis, and potential white dwarf mass retention.

T Coronae Borealis (T CrB) is a symbiotic recurrent nova: a long-period interacting binary in which a massive white dwarf accretes from an evolved M giant and undergoes recurrent thermonuclear runaways. It is also the prototype of the recurrent-nova subclass containing a white dwarf and a late-type giant, and it is among the brightest novae known, having reached about $V\simeq 2$ in its secure eruptions of 1866 and 1946 (Hinkle et al., 28 Feb 2025, Pei et al., 22 Dec 2025). Because T CrB is nearby, because its orbital period is about $227.55$–$227.58$ d, and because several recent studies place it in an unusually advanced pre-eruption state, it has become a central testbed for Roche-lobe overflow in symbiotic binaries, long-timescale accretion-disk evolution, recurrent nova triggering, circumstellar shaping, and prospective multi-messenger signatures of an imminent outburst (Planquart et al., 6 Jan 2025, Hinkle et al., 28 Feb 2025, Pei et al., 22 Dec 2025).

1. Historical record and recurrent-nova phenomenology

T CrB is securely known to have erupted in 1866 and 1946, with both eruptions reaching about $V=2.0$ and declining rapidly; a fully Johnson-calibrated reconstruction gives $t_2=3.0$ d, $t_3=5.0$ d, and $t_6=12.0$ d for the primary eruption light-curve template (Schaefer, 2023). The same long-baseline reconstruction, based on 213,730 $B$ and $V$ magnitudes from 1842–2022, also identifies a distinctive cycle morphology: a long high state extending from about $T_{\rm eruption}-10$ yr to $227.55$0 yr, a deep pre-eruption dip, and a delayed “secondary eruption” with about 10 per cent of the energy of the primary outburst (Schaefer, 2023). In that reconstruction, the 1866 and 1946 cycles are described as strikingly similar in the primary eruption, the secondary eruption, and the post-eruption high state, which is why template-based eruption forecasts became influential in the 2015–2025 interval (Schaefer, 2023).

The earlier historical record is more contentious. Schaefer argued that T CrB was observed in eruption near 1787.9 and 1217.8, on the basis of Francis Wollaston’s 1789 catalog entry coincident with T CrB and Abbott Burchard’s 1217 report of a stellar transient in Corona Borealis (Schaefer, 2023). In that reconstruction, the sequence 1217.8, 1787.9, 1866.4, and 1946.1 was taken to support a characteristic recurrence timescale near 80 years, while still allowing cycle-to-cycle scatter (Schaefer, 2023). These earlier events are therefore part of the historical literature on T CrB, but they remain historical inferences rather than modern instrumental detections.

The 20th- and 21st-century light curves also established that T CrB is exceptional even among novae with red-giant donors. In quiescence it is around $227.55$1, making it unusually accessible to dense photometric monitoring, and its pre- and post-eruption high states are not brief precursor flares but multi-year accretion episodes (Schaefer, 2023). A plausible implication is that T CrB’s recurrent-nova clock is tied not only to white-dwarf mass but also to the long-term storage and release of mass in the accretion flow, a point developed explicitly in later accretion-disk studies (Schlindwein et al., 5 Jun 2025, Munari et al., 31 Jul 2025).

2. Binary architecture and orbital solutions

T CrB is consistently modeled as a white dwarf plus an M giant in a nearly circular orbit of about 227.55–227.58 d, with strong ellipsoidal modulation at half the orbital period. Recent multi-parameter solutions differ in detail, but all place the system in the regime of a massive white dwarf and a Roche-distorted red giant (Hinkle et al., 28 Feb 2025, Munari et al., 31 Jul 2025).

Study	Orbital solution	Masses and geometry
Planquart et al.	$227.55$2 d; $227.55$3; $227.55$4 au	$227.55$5; $227.55$6 (Planquart et al., 6 Jan 2025)
Hinkle et al.	$227.55$7 d; preferred circular orbit; Gaia-distance fit $227.55$8	$227.55$9; $227.58$0 (Hinkle et al., 28 Feb 2025)
Munari et al.	$227.58$1 d; preferred $227.58$2	$227.58$3; $227.58$4 (Munari et al., 31 Jul 2025)

These solutions share several structural features. Hinkle et al. describe the orbit as “circular to a high precision,” with $227.58$5 in an eccentric fit but a circular orbit preferred under the Lucy–Sweeney criterion (Hinkle et al., 28 Feb 2025). Munari et al. likewise adopt a null eccentricity, while their radiative model requires the red giant to fill its Roche lobe completely; they state that even a 3% underfilling would lower the total brightness by 0.10 mag, inconsistent with the observed light curves (Munari et al., 31 Jul 2025). Planquart et al. explicitly treat the giant as Roche-filling and use standard Roche geometry, adopting

$227.58$6

with $227.58$7, to characterize the $227.58$8 geometry relevant for stream-fed accretion (Planquart et al., 6 Jan 2025).

A notable point of disagreement concerns donor rotation. Hinkle et al. measure $227.58$9 and treat tidal synchronization as part of their joint fit (Hinkle et al., 28 Feb 2025). By contrast, Munari et al. report $V=2.0$0 and argue that the giant is rotating much slower than the $V=2.0$1 co-rotation value implied by a Roche-lobe-filling star in the preferred geometry (Munari et al., 31 Jul 2025). In that interpretation, the donor is markedly sub-synchronous, and its envelope angular-momentum state may itself be affected by sustained mass loss through $V=2.0$2 (Munari et al., 31 Jul 2025).

3. Accretion flow, Roche-lobe overflow, and the super-active phase

The most direct kinematic decomposition of T CrB’s accretion flow is the 2011–2023 Doppler-tomography study based on 100 HERMES spectra, 83 obtained during the recent super-active phase (SAP) (Planquart et al., 6 Jan 2025). In that analysis, the ballistic stream launched from $V=2.0$3 and the characteristic disk radii,

$V=2.0$4

were used to interpret emission-line structures in velocity space (Planquart et al., 6 Jan 2025). The central result is that during the SAP, T CrB was transferring mass predominantly by Roche lobe overflow rather than by a generic red-giant wind (Planquart et al., 6 Jan 2025).

The evidence is line-specific and geometric. The $V=2.0$5 8446 Å and $V=2.0$6 6678 Å lines produce an arc-like spot in the Doppler maps at the expected stream–disk impact region, with $V=2.0$7 near the stream intersection with the Keplerian velocity corresponding to the tidal truncation radius and $V=2.0$8 6678 between the velocity circles associated with $V=2.0$9 and $t_2=3.0$0 (Planquart et al., 6 Jan 2025). This is the canonical bright-spot morphology of a stream-fed accretion disk, and the same study concludes that the disk is large, optically thick, fully viscously evolved, and extends close to its maximal radius, “filling almost 70% of the Roche surface at the WD” (Planquart et al., 6 Jan 2025).

The SAP tomography resolves multiple interaction sites simultaneously: the outer-rim bright spot, irradiation of the donor’s WD-facing hemisphere, stream–disk overflow, an accretion-disk wind, a hot WD-centered inner region traced by $t_2=3.0$1 4686 and Bowen $t_2=3.0$2 4640, and a dense expanding nebular component traced by [$t_2=3.0$3] and $t_2=3.0$4. The [$t_2=3.0$5] profile evolution is interpreted as a bipolar jet or blast-wave-like outflow launched during the rise of the SAP; its velocity splitting increased for about the first 1000 days, approximately as $t_2=3.0$6, before saturating near $t_2=3.0$7 (Planquart et al., 6 Jan 2025). The same work therefore portrays T CrB during the SAP as a scaled-up cataclysmic-variable-like accretor embedded in a symbiotic geometry.

The physical origin of the SAP remains debated. The tomography paper favors a chronology in which the departure from quiescence started in the disk and was likely triggered by a thermal-viscous disk instability analogous to a dwarf-nova outburst, after which irradiation of the giant reinforced Roche-lobe overflow (Planquart et al., 6 Jan 2025). A different 2025 time-dependent disk model instead reproduces the long bright state as an enhanced mass-transfer event from the donor, with a viscosity parameter $t_2=3.0$8, an event duration $t_2=3.0$9 yr, quiescent and high-state transfer rates of $t_3=5.0$0 and $t_3=5.0$1, and a pre-eruption dip produced by expansion of the inner disk radius at an average speed of $t_3=5.0$2 (Schlindwein et al., 5 Jun 2025). That model explicitly rejects a standard thermal-viscous DIM origin for the full 15-year bright state and also rejects steady nuclear burning as too luminous for the observed continuum (Schlindwein et al., 5 Jun 2025).

A third synthesis by Munari et al. interprets the 2015–2023 SAP as an inside-out collapse of a disk that had accumulated mass over most of the inter-eruption interval; in that account, the mean accretion rate onto the white dwarf during SAP was about 28 times larger than in quiescence, and renewed mass flow at disk inner radii from May 2024 may compensate for the fainter recent SAP relative to the historical pre-1946 episode (Munari et al., 31 Jul 2025). The coexistence of these interpretations is itself significant: T CrB is no longer treated as a purely wind-fed symbiotic, but the relative roles of donor-driven mass-transfer enhancement and disk-intrinsic instability remain unresolved.

4. Multiwavelength state changes from radio to X-rays

The 2014–2015 transition into the recent high state was accompanied by strong wavelength-dependent brightening and an X-ray anti-correlation. A 20-year Swift+UVOT+AAVSO study finds that the average brightening relative to quiescence was about 0.9 mag in $t_3=5.0$3, 0.38 mag in $t_3=5.0$4, 0.22 mag in $t_3=5.0$5, and 0.12 mag in $t_3=5.0$6, while the UV brightened by 3.01 mag in UVW1 and 4.60 mag in UVM2; over the same interval the XRT 0.3–10 keV count rate fell from about $t_3=5.0$7 ct s$t_3=5.0$8 to $t_3=5.0$9 ct s$t_6=12.0$0, and the system became much bluer, with $t_6=12.0$1 shifting from $t_6=12.0$2 to $t_6=12.0$3 (Pei et al., 22 Dec 2025). The same study identifies a primary pre-eruption dip beginning around MJD 60050 in April 2023, with fading amplitudes after orbital correction of 1.364 mag in $t_6=12.0$4, 0.738 mag in $t_6=12.0$5, 0.293 mag in $t_6=12.0$6, 0.148 mag in $t_6=12.0$7, and 0.074 mag in $t_6=12.0$8, accompanied by simultaneous brightening in both soft and hard X-rays, and also reports a second, lower-amplitude dip from roughly September 2024 to February 2025 (Pei et al., 22 Dec 2025).

That multiwavelength pattern is interpreted in boundary-layer terms. The high state corresponds to a higher accretion rate and an optically thick boundary layer that suppresses hard X-rays while enhancing optical/UV luminosity; the dip corresponds to a reduction in $t_6=12.0$9 and a boundary layer becoming optically thinner, thereby strengthening the X-ray output while the hot accretion-powered optical/UV component fades (Pei et al., 22 Dec 2025). A later HST/Swift/NuSTAR/XMM-Newton analysis quantified this transition as a highly significant $B$0 increase in the luminosity of the optically thin cooling flow $B$1 and a marginal $B$2 decrease in the optically thick boundary-layer luminosity $B$3 as T CrB entered the faint state, and found no significant periodicities once autoregressive red-noise modeling was applied (Luna et al., 22 Jan 2026). That study explicitly argues against dust obscuration, citing the lack of infrared excess and the stability of the 2175 Å feature (Luna et al., 22 Jan 2026).

At longer wavelengths, the radio and millimeter data trace the ionization state of the giant wind. During the 2016–2017 super-active state, VLA observations showed a rising spectrum with $B$4 to $B$5 and mean $B$6, consistent with optically thick thermal bremsstrahlung; the inferred radio photosphere diameter of 5.6–13.2 AU was far larger than the $B$7 AU binary separation, implying that the radio-emitting region lay in an enhanced, photoionized circumstellar wind well outside the orbit (Linford et al., 2019). By contrast, ALMA observations in late 2024 found T CrB to be a faint mm/sub-mm source with a phase-$B$8 continuum described by $B$9, corresponding to about 0.1 mJy at 44 GHz and 0.4 mJy at 400 GHz; the source was unresolved, no significant line emission was detected, and the spectrum was about a factor of 5 fainter near 35 GHz than the 2016/17 high-state radio spectrum (Petry et al., 5 Sep 2025). The ALMA study concludes that the 2024 wind was far from fully ionized, in contrast to the more highly ionized circumstellar state inferred during the super-active episode (Petry et al., 5 Sep 2025).

5. Circumbinary environment, remnant structure, and predicted outburst hydrodynamics

T CrB is now known to sit in an extended structured environment on both parsec and au scales. Deep narrowband imaging with the Condor Array Telescope revealed a faint, roughly $V$0-scale nebula centered on T CrB, detected in H$V$1, [N II], and [S II] but not in [O III] or continuum; at the Gaia distance used in that study, the physical diameter is about 31.9 pc (Shara et al., 2024). The nebula is interpreted as a nova super-remnant generated by repeated recurrent-nova eruptions over at least $V$2 yr, with a shell mass of about $V$3 to $V$4, dominated by swept-up ISM rather than direct nova ejecta (Shara et al., 2024). The same analysis argues that the upcoming nova flash is very unlikely to produce detectable fluorescent optical echoes, because both the recent ejecta and the super-remnant are too optically thin, although dust-scattered continuum echoes on $V$5–$V$6 scales remain possible and were identified as an HST/JWST opportunity (Shara et al., 2024).

Closer to the binary, 3-D hydrodynamic simulations of the forthcoming eruption adopt a circumbinary medium consisting of a spherical wind plus a torus-like equatorial density enhancement (EDE), along with the red giant and a large accretion disk (Orlando et al., 27 Jul 2025). In those models the CBM is significantly less dense than in RS Oph or V745 Sco, with a preferred mass-loss rate $V$7 for a $V$8 wind, but the blast is still collimated along the poles by the accretion disk and EDE, producing a bipolar shock (Orlando et al., 27 Jul 2025). The red giant partially shields the ejecta, creating a bow shock and a hot wake, and the predicted X-ray evolution proceeds through three phases: an early phase dominated by shocked disk material in the first few hours, an intermediate phase from about one week to one month driven by reverse-shocked ejecta, and a late phase dominated by shocked EDE (Orlando et al., 27 Jul 2025). The peak X-ray luminosity is about $V$9 erg s$T_{\rm eruption}-10$0, with soft X-rays tracing shocked ejecta, hard X-rays arising mainly from shocked ambient gas, and synthetic line profiles showing asymmetric blueshifts due to absorption in expanding ejecta (Orlando et al., 27 Jul 2025).

A complementary 1-D NOVA study of the next eruption emphasizes the white-dwarf thermonuclear physics rather than the 3-D blast interaction. For $T_{\rm eruption}-10$1 white dwarfs accreting at rates designed to reproduce an $T_{\rm eruption}-10$2-yr recurrence interval, it predicts peak effective temperatures above $T_{\rm eruption}-10$3 K in all radiative cases, implying a short early X-ray phase and a UV flash capable of ionizing the red-giant wind and producing narrow lines superposed on broad nova profiles (Starrfield et al., 15 Feb 2025). Taken together, the environmental and hydrodynamic papers imply that the next T CrB eruption should be both intrinsically luminous and unusually diagnostic of disk, wind, and EDE structure from the first hours onward.

6. Composition, high-energy messengers, and the debate over long-term fate

The anticipated eruption is expected to be chemically diagnostic. A Monte Carlo nucleosynthesis study based on recent hydrodynamic T CrB models finds that both CO and ONe white-dwarf compositions robustly overproduce characteristic CNO isotopes such as $T_{\rm eruption}-10$4C, $T_{\rm eruption}-10$5N, and $T_{\rm eruption}-10$6N, while the strongest composition differences emerge for elements with $T_{\rm eruption}-10$7 (Wallace et al., 3 Sep 2025). Sulfur is identified as the cleanest observational discriminator because the ONe/CO abundance difference is about 30 while the reaction-rate uncertainty is $T_{\rm eruption}-10$8; neon, silicon, and phosphorus differ by factors of about 150–250 and are also useful tracers, whereas chlorine, argon, and potassium are less robust because the $T_{\rm eruption}-10$9 abundance ranges overlap (Wallace et al., 3 Sep 2025). A direct implication is that early optical, UV, and IR abundance measurements in the next eruption could materially constrain whether the accretor is CO or ONe.

High-energy and multi-messenger forecasts are equally active. One model-based study predicts that, in an external-shock scenario where nova ejecta collide with the red-giant wind, gamma rays from T CrB should be detectable with Fermi-LAT, MAGIC, H.E.S.S., MACE, LHAASO, and HERD, while neutrino prospects are poor except in the uppermost part of the external-shock parameter space (Sarmah et al., 26 Dec 2025). The same work considers a magnetic-reconnection scenario near the white dwarf, in which gamma rays are fully absorbed locally but neutrinos escape; in that case IceCube and KM3NeT have substantially better detection prospects, and a distinctive temporal signature is predicted in which neutrinos could precede external-shock photons or neutrinos by about 9–10 hours (Sarmah et al., 26 Dec 2025). IceCube has already prepared real-time follow-up using both GRECO Astronomy and FRA, with a principal search window of $227.55$00 days around optical peak; on the basis of optical scaling from the 1866 and 1946 eruptions, that study argues that T CrB should produce a signal several times stronger than RS Oph, at a declination where IceCube sensitivity is another factor of a few better (Thwaites et al., 9 Jul 2025). A separate 3-D hydrodynamic plus DSA study similarly predicts bright GeV emission for weeks, detectable TeV emission for days to about a week, early high-energy flux strongly boosted by the accretion disk, and neutrino detectability primarily in high-explosion-energy, high-density models (Petruk et al., 16 Mar 2026).

The system’s secular evolution remains explicitly disputed. Several papers treat T CrB as a near-Chandrasekhar recurrent nova and therefore an important laboratory for recurrent ignition, mass retention, and possible Type Ia supernova or accretion-induced collapse pathways (Hinkle et al., 28 Feb 2025, Starrfield et al., 15 Feb 2025). By contrast, a 2025 orbital-period analysis claims that the 1946 eruption caused a period increase of $227.55$01 d, implying $227.55$02, about 540 times the accreted mass, and therefore concludes that the white dwarf is losing mass and that T CrB can never become a Type Ia supernova (Schaefer, 2 Oct 2025). The literature thus contains both near-Chandrasekhar growth scenarios and a directly contrary period-change argument that interprets the white dwarf as eroding over cycles.

Forecasting the exact eruption date is likewise unsettled. Template-based optical analyses proposed dates such as $227.55$03 and, after the onset of the pre-eruption dip, $227.55$04 (Schaefer, 2023, Schaefer, 2023). Later multiwavelength work is more cautious: it treats the modern dip sequence as consistent with an imminent nova and describes T CrB as being in an unusually advanced pre-eruption state, but stops short of treating the dip as a precise eruption clock (Pei et al., 22 Dec 2025). What is not disputed is the system’s present status as a rare pre-outburst recurrent nova for which orbital structure, accretion physics, circumstellar environment, nucleosynthetic yields, high-energy signatures, and historical recurrence are all being constrained in advance of the next event.