V1674 Her: Record Fast Nova

• V1674 Her is a classical nova noted for its record-fast optical rise and decline, exhibiting a 10-mag increase in just ~5 hours with t2 ~1 day.
• Its multiwavelength observations, from gamma-rays to X-rays, reveal strong shocks and rapid re-establishment of magnetically mediated accretion in an intermediate polar system.
• Detailed spectral and timing analyses uncover bipolar ejecta morphology, early coronal emissions, and a debated white-dwarf mass between 1.09 and 1.36 M☉.

Searching arXiv for recent and foundational papers on V1674 Her to ground the article in the literature. V1674 Her, also designated Nova Her 2021, is a classical nova that erupted in June 2021 and is repeatedly described in the literature as the fastest classical nova on record or the fastest Galactic nova known. Reported optical decline times span $t_2=0.904$ d, $1.0$ d, $1.1$ d, and $1.2$ d in different analyses, reflecting dataset and methodology differences rather than any dispute over its extreme speed (Habtie et al., 2023, Woodward et al., 2021, Sokolovsky et al., 2023, Kumar et al., 6 Feb 2026). The system is now identified as an intermediate polar, with an orbital period near $0.153$ d and a white-dwarf spin period near $501$ s, and it has become a reference object for the study of nova onset, internal shocks, coronal emission, supersoft X-ray pulsations, and the rapid re-establishment of magnetically mediated accretion (Patterson et al., 2022, Luna et al., 2023, Bhargava et al., 2023).

1. Discovery and optical rise

The eruption was discovered on 2021 June 12.5484 UT at visual magnitude $8.4$, and the nova reached $V=6.14$ mag on 2021 June 12.96 UT. Wide-field surveys subsequently reconstructed an even earlier onset: ASAS-SN detected the source on 2021 June 12 at HJD 2459377.6944, while another analysis defined the eruption onset at June 12.1903 UT, when the source had already risen from quiescence to $g\approx16.6$ mag (Woodward et al., 2021, Quimby et al., 2024, Sokolovsky et al., 2023).

The pre-maximum and rising light curve is unusually well sampled. One early study reported a plateau in the pre-maximum light curve at $g\sim14$, approximately $1.0$0 mag below peak, lasting for at least three hours; this was described as unprecedented at such a large offset below maximum (Quimby et al., 2021). A later high-cadence reconstruction resolved the rise into three phases: a “slow rise” from $1.0$1 to $1.0$2 over $1.0$3 d, a “fast rise” from $1.0$4 to $1.0$5 over $1.0$6 d, and a “faster rise” from $1.0$7 to $1.0$8 over $1.0$9 d, with breaks at $1.1$0 d and $1.1$1 d in that paper’s time coordinate (Quimby et al., 2024). Taken together, the optical data record a rise of roughly $1.1$2 mag in about $1.1$3 h, including more than $1.1$4 mag in approximately $1.1$5 d (Hachisu et al., 9 Jul 2025, Quimby et al., 2024).

The decline was comparably abrupt. Published values include $1.1$6 d, $1.1$7 d, and $1.1$8 d, while one optical analysis obtained $1.1$9 d and $1.2$0 d (Woodward et al., 2021, Sokolovsky et al., 2023, Kumar et al., 6 Feb 2026, Habtie et al., 2023). Optical monitoring over longer baselines found a smooth rapid decline without a dust-formation dip, while spectra obtained as late as day 147.66 still showed nebular emission, indicating that the system had not yet returned to quiescence (Habtie et al., 2023, Kumar et al., 6 Feb 2026).

2. Earliest high-energy behavior and shocks

V1674 Her exhibited prompt high-energy emission. Fermi-LAT detected $1.2$1–$1.2$2 GeV $1.2$3-rays beginning about $1.2$4 h after the eruption onset, with an average photon flux $1.2$5 over an $1.2$6 h interval and photon index $1.2$7 (Sokolovsky et al., 2023). A separate analysis reported a stacked $1.2$8 d detection with TS$1.2$9 (about $0.153$0), a rapid decline $0.153$1, and an emission duration of less than one day, described there as the shortest duration known for a $0.153$2-ray nova (Lin et al., 2022).

X-ray observations established a shock-heated component and, later, a supersoft source. At $0.153$3–$0.153$4 d, joint NuSTAR and Swift data were fitted by an optically thin single-temperature plasma with $0.153$5 keV and no additional intrinsic absorption beyond the interstellar column $0.153$6 (Sokolovsky et al., 2023). Using

$0.153$7

the same study inferred $0.153$8 (Sokolovsky et al., 2023). The absence of the expected Fe XXV K$0.153$9 line at $501$0 keV was interpreted as evidence for super-solar CNO abundances and severely sub-solar Fe in the X-ray-emitting shock plasma (Sokolovsky et al., 2023).

Radio observations from $501$1 d to $501$2 d showed an initially inverted spectrum, with brightness-temperature and rise-rate arguments requiring synchrotron emission absorbed by free-free opacity, while thermal free-free emission contributed at very early times at high frequencies (Sokolovsky et al., 2023). By $501$3 d the spectrum was optically thin with spectral index $501$4 (Sokolovsky et al., 2023). These multiwavelength results were taken to show that internal shocks developed inside the ejecta despite the nova’s extraordinarily rapid evolution and the intermediate-polar nature of the binary (Sokolovsky et al., 2023).

The earliest optical rise has also been connected directly to shock and wind physics. One analysis argued that the white-dwarf photosphere was unlikely to have overflowed its Roche lobe prior to the launch of a fast wind, posing a challenge to models in which GeV emission is powered by interaction between a later fast wind and a slow torus stripped from an inflated envelope by the companion (Quimby et al., 2024). This suggests that, in V1674 Her, internal shocks in a super-Eddington wind or very rapidly established ejecta stratification may be more relevant than a prolonged pre-wind Roche-lobe-overflow phase (Quimby et al., 2024).

3. Orbital and spin clocks

The timing phenomenology of V1674 Her established it as an intermediate polar. Optical time-series photometry during eruption revealed periodic signals at $501$5 d and $501$6 s, interpreted as the orbital period and white-dwarf spin period, respectively (Patterson et al., 2022). The same work detected a sideband at the difference frequency,

$501$7

with persistent $501$8 and brief $501$9 signals; this was described as unambiguous confirmation of reprocessed spin radiation by structures orbiting at the binary frequency (Patterson et al., 2022).

The emergence of these clocks was unusually early. The orbital waveform appeared by day 4 and the $8.4$0 s spin pulses by day 12, which was attributed to the extreme speed with which the nova evolved and became transparent to radiation from the inner binary (Patterson et al., 2022). During the first 15 days of outburst the spin period increased by $8.4$1, from the pre-eruption $8.4$2 s to $8.4$3 s, and then began a steady decrease of $8.4$4 ms yr$8.4$5, approximately $8.4$6 faster than usually seen in intermediate polars (Patterson et al., 2022). That behavior was interpreted as an initial response to the sudden loss of high-angular-momentum gas, followed by strong accretion torques at very high mass-transfer rates (Patterson et al., 2022).

Independent measurements across multiple bands corroborated the timing picture. TESS detected the orbital period as $8.4$7 d in Sector 40, with a double-peaked folded light curve whose minima are separated by $8.4$8, and also revealed a transient $8.4$9 d modulation between day 13 and day 17 (Luna et al., 2023). NICER detected a $V=6.14$0 d X-ray period consistent with the optical orbit; after removal of the $V=6.14$1 s rotational X-ray pulsations, the folded X-ray light curve showed a clear double-humped profile with peaks near $V=6.14$2 and $V=6.14$3 and peak-to-peak amplitude of about $V=6.14$4 of the mean count rate (Lin et al., 2022). The same study argued that the profile is consistent with occultation by the companion and/or accretion disk, implying a relatively high inclination, with estimates around $V=6.14$5–$V=6.14$6 (Lin et al., 2022).

Supersoft X-ray timing further strengthened the spin identification. NICER monitored the source from 2021 July 10 to August 31 and found a coherent pulsation at $V=6.14$7 s with a large pulsed fraction of $V=6.14$8–$V=6.14$9, detectable throughout the supersoft phase (Orio et al., 2022). AstroSat detected highly significant $g\approx16.6$0–$g\approx16.6$1 keV periodicities of $g\approx16.6$2 s, $g\approx16.6$3 s, and $g\approx16.6$4 s on days 37, 38, and 54, respectively, with pulse morphology that evolved from quasi-sinusoidal to skewed and then to a broad plateau with hints of a secondary notch (Bhargava et al., 2023). A later TESS analysis confirmed $g\approx16.6$5 s and $g\approx16.6$6 d, and argued that the agreement between optical and X-ray spin periods implies that the white-dwarf atmosphere during the supersoft phase was not thermally homogeneous (Luna et al., 9 Nov 2025).

Not all early searches found modulation. Second-timescale near-infrared photometry with Palomar Gattini-IR, obtained on three nights between 3 and 6 days after discovery at a temporal resolution of about $g\approx16.6$7 s, found no periodic variability down to a $g\approx16.6$8 upper limit of $g\approx16.6$9 mag (Hansen et al., 2021). That non-detection was interpreted as implying that the periodic variability later reported in the optical was lower by at least a factor of about $g\sim14$0 during the first week of eruption (Hansen et al., 2021).

4. Spectroscopy, ionization structure, and ejecta morphology

Early optical and near-infrared spectra show a rapid transition from low-ionization permitted-line spectra to nebular and coronal conditions. In the first days, broad P Cygni H and Fe II lines dominated, with FWHM values around $g\sim14$1–$g\sim14$2, FWZI up to about $g\sim14$3, and corrugated or multi-peaked profiles indicating clumpy ejecta (Woodward et al., 2021, Kumar et al., 6 Feb 2026). Optical spectroscopy classified the initial stages as pre-maximum and early decline, while noting a rare Fe II $g\sim14$4 He/N $g\sim14$5 Ne hybrid transition (Habtie et al., 2023).

Nebular and coronal phases set in exceptionally early. One optical study reported the appearance of [Ne III] $g\sim14$6 on day 10.00, calling this the earliest observed nebular onset for a classical nova (Habtie et al., 2023). Near-infrared spectroscopy detected coronal lines already on day 11.51, including [Si VI] $g\sim14$7, [Si VII] $g\sim14$8, and [Al VI] $g\sim14$9, and identified this as the earliest onset yet observed for coronal emission in any classical nova (Woodward et al., 2021). Optical coronal lines such as [Fe X] $1.0$00 and [Fe XIV] $1.0$01 appeared later in the optical data, from day 19.87 or day 22.89 depending on the specific dataset and line set considered (Kumar et al., 6 Feb 2026, Habtie et al., 2023).

A central issue has been the origin of the coronal spectrum. Near-infrared analysis argued that on day 11.5 there was no obvious hard-X-ray continuum above about $1.0$02 eV and that the supersoft phase began only around day 18, so photoionization could not account for the earliest high-ionization lines (Woodward et al., 2021). Using the relative strengths of adjacent Si coronal lines and collisional-ionization equilibrium, that study inferred

$1.0$03

and concluded that shocks between multi-velocity clumps with $1.0$04 dominated the initial coronal emission (Woodward et al., 2021). The same work proposed that the temporal sequence—shock-driven coronal lines first, supersoft X-rays later—helps resolve the long-standing collisional-versus-photoionization problem by indicating that both mechanisms operate, but at different times (Woodward et al., 2021).

Photoionization modeling of the optical spectra with \textsc{cloudy} adopted a spherically expanding shell with density profile $1.0$05, filling factor $1.0$06, and two density components, and derived central-source temperatures of $1.0$07–$1.0$08 K and luminosities of $1.0$09–$1.0$10 across days 10–28 (Habtie et al., 2023). The same modeling found He, O, N, and Ne overabundant relative to solar values in both nebular and coronal phases, with Fe abundance decreasing toward solar as Ne increased, again linked to the hybrid spectroscopic evolution (Habtie et al., 2023). A later optical study emphasized the appearance of [Ne III] and [Ne V] lines on day 19.87 as evidence for an ONe white dwarf, and also found that Lyman-$1.0$11 fluorescence dominates the excitation of neutral oxygen, based on the strong O I $1.0$12 Å line and related diagnostics (Kumar et al., 6 Feb 2026).

Morpho-kinematic reconstruction with \textsc{shape} consistently favors an axisymmetric ejecta geometry. One study obtained bipolar lobes with an equatorial ring and inclination $1.0$13, with position angle about $1.0$14 (Habtie et al., 2023). Another, modeling H$1.0$15 profiles at days 25.68 and 65.86, found a bipolar shell with an equatorial ring and polar blobs, inclination $1.0$16, and a Hubble-flow expansion law

$1.0$17

again linking the central peaks to the equatorial ring and the broad wings to polar structures (Kumar et al., 6 Feb 2026). The inclination inferred from these line-profile models is consistent with the high-inclination interpretation of the double-humped X-ray orbital modulation (Lin et al., 2022).

5. Distance estimates, ejected mass, and theoretical modeling

Published distance estimates are not uniform. Near-infrared work combined $1.0$18, an MMRD absolute magnitude of $1.0$19, and a 3D extinction–distance curve to obtain $1.0$20 kpc and $1.0$21 mag (Woodward et al., 2021). An optical photoionization study, using $1.0$22 mag and the distance-modulus relation,

$1.0$23

derived $1.0$24 kpc (Habtie et al., 2023). By contrast, a full theoretical light-curve model based on time-stretching against LV Vul, KT Eri, and V339 Del obtained $1.0$25 and, for $1.0$26, inferred $1.0$27 kpc (Kato et al., 5 Jun 2025). The 2026 quiescent X-ray study used $1.0$28 kpc when converting flux to luminosity (Olbemo et al., 7 May 2026). These differences remain part of the system’s parameter uncertainty budget.

Ejected-mass estimates also vary by method. From dereddened Pa$1.0$29 and Br$1.0$30 fluxes on day 11.51, Case B recombination, and an adopted electron density $1.0$31, the near-infrared analysis obtained an upper limit

$1.0$32

for the hydrogen ejecta mass (Woodward et al., 2021). Optical \textsc{cloudy} modeling instead found $1.0$33–$1.0$34 across five epochs (Habtie et al., 2023). The multiwavelength shock study inferred $1.0$35 from the lack of intrinsic X-ray absorption in a spherical Hubble-flow model, while the early radio emission required a thermal lower limit of at least $1.0$36 (Sokolovsky et al., 2023). The spread reflects the fact that different diagnostics probe different gas phases and geometrical assumptions.

Several theoretical frameworks have been applied to the early light curve. A comprehensive time-dependent nova-cycle model used a $1.0$37 white dwarf, $1.0$38, and a $1.0$39 companion in a $1.0$40 h binary with separation $1.0$41 and white-dwarf Roche-lobe radius $1.0$42 (Kato et al., 5 Jun 2025). In that model the X-ray flash lasts about $1.0$43 h, the photosphere expands to $1.0$44 at $1.0$45 h, the companion is engulfed at $1.0$46 h and re-emerges at $1.0$47 d, optically thick winds begin when $1.0$48, and the wind ends at about day 26 (Kato et al., 5 Jun 2025). The optical decay is approximated by free-free emission,

$1.0$49

and the model reproduces the very steep rise and fast decline over the first $1.0$50 d (Kato et al., 5 Jun 2025).

A separate composite model of the earliest rise represented the total optical output as

$1.0$51

with the pre-wind phase dominated by the hot white dwarf, irradiated accretion disk, and irradiated companion, and the post-wind phase dominated by free-free emission from optically thin ejecta (Hachisu et al., 9 Jul 2025). In that interpretation, the ASAS-SN points at $1.0$52, $1.0$53, and $1.0$54 mag at $1.0$55–$1.0$56 d correspond to the X-ray flash phase of a $1.0$57 white dwarf, making them the first optical detection of a nova’s X-ray flash phase, while optically thick winds turn on at $1.0$58 d and the break at $1.0$59 reflects the rapid increase in $1.0$60 associated with the Fe opacity peak near $1.0$61 (Hachisu et al., 9 Jul 2025).

A third model addressed the nova’s “superbrightness.” It proposed that a strong reverse shock forms at $1.0$62 d, producing an optically thick shocked shell with velocity about $1.0$63. In that scenario the free-free optical luminosity peaks first at $1.0$64 on day $1.0$65, but an optically thick shell reaches $1.0$66 on day $1.0$67 when $1.0$68, while the GeV $1.0$69-ray flux peaks earlier on day $1.0$70 (Hachisu et al., 23 Dec 2025). This was presented as an explanation for the roughly $1.0$71 mag excess above the free-free peak in a very fast nova classified there as superbright (Hachisu et al., 23 Dec 2025).

6. White-dwarf mass, magnetic field, and current interpretation

The white-dwarf mass of V1674 Her is an active point of contention. Light-curve and evolutionary models that emphasize the extreme speed of the eruption favor a very high mass. The full nova-cycle and early-rise composite models both require $1.0$72 to reproduce the rise and subsequent evolution (Kato et al., 5 Jun 2025, Hachisu et al., 9 Jul 2025). An optical calibration based on $1.0$73 yielded $1.0$74 and a radius of about $1.0$75 (Habtie et al., 2023). In these approaches, the short flash timescale, rapid rise, rapid decline, and early reappearance of orbital and spin modulations are all treated as consequences of a highly massive white dwarf (Kato et al., 5 Jun 2025, Habtie et al., 2023).

Direct quiescent X-ray constraints lead to a different conclusion. Broadband XMM-Newton plus NuSTAR spectroscopy, modeled with a post-shock accretion-column calculation including the temperature gradient and reflection, gave

$1.0$76

under the assumption that the accretion disk is truncated at the co-rotation radius (Olbemo et al., 7 May 2026). An independent timing-based constraint from a power-spectrum break at $1.0$77 mHz yielded

$1.0$78

consistent with the spectral fit (Olbemo et al., 7 May 2026). The same study derived an accretion rate $1.0$79 and a surface magnetic field

$1.0$80

placing the system at the high end of the magnetic-field distribution for intermediate polars (Olbemo et al., 7 May 2026).

This disagreement has broader implications. The X-ray study explicitly argued that empirical $1.0$81 decline-time relations overestimate white-dwarf masses in extreme fast novae, because parameters beyond $1.0$82—including magnetic field strength, accretion geometry, envelope mass, and composition—can strongly influence the optical timescale (Olbemo et al., 7 May 2026). That conclusion is consistent with the broader observational picture: V1674 Her combines an ultra-fast optical evolution, clear intermediate-polar timing signatures, high-energy shocks, early coronal emission, and a high-inclination bipolar ejecta geometry, but the correspondence between “fastest decline” and “near-Chandrasekhar mass” is no longer secure for this object (Patterson et al., 2022, Sokolovsky et al., 2023, Habtie et al., 2023, Olbemo et al., 7 May 2026).

Within nova studies, V1674 Her therefore occupies a distinctive position. Its dense sub-peak optical coverage constrains the onset of thermonuclear runaway and wind launching; its early $1.0$83-ray, X-ray, radio, and coronal-line detections provide a compact record of shock formation across the spectrum; and its persistent spin and orbital signals expose the recovery of a magnetic cataclysmic variable almost in real time (Quimby et al., 2024, Sokolovsky et al., 2023, Orio et al., 2022, Bhargava et al., 2023). A plausible implication is that V1674 Her is less a single-parameter extreme than a system in which eruption speed, magnetic truncation, ejecta structure, and radiative transfer are all unusually compressed in time.