---
title: 'ZTF SLRN-2020: Planet-Star Engulfment Transient'
url: https://www.emergentmind.com/topics/ztf-slrn-2020
type: topic
---

# ZTF SLRN-2020: Planet-Star Engulfment Transient

ZTF SLRN-2020 is a subluminous red nova-like intermediate-luminosity optical transient associated in the current literature with the engulfment or merger of a planet-mass companion by a main-sequence or near-main-sequence star. It has been treated as the strongest currently known case of a planet-star engulfment transient, and as the third event still compatible with a dynamical-timescale merger between a planet and a main-sequence or near-main-sequence star. Its defining observational features are a low-luminosity red outburst, months of pre-outburst dust formation, and a late-time remnant containing warm circumstellar dust, hot molecular gas, and Brackett-\(\alpha\) emission. These properties place it at the low-energy end of the red nova and ILOT continuum and make it a benchmark object for star-planet interaction studies [2305.04909, 2504.07275, 2507.05365].

## 1. Classification within the red nova and ILOT framework

In the terminology used for ZTF SLRN-2020, an ILOT is an “intermediate luminosity optical transient,” i.e. a broad class of gravitationally powered transients lying between novae and supernovae in energy and luminosity, while “Red Nova” denotes a cool, red merger-like transient within or overlapping that family. ZTF SLRN-2020 is further described as a subluminous red nova: a red nova-like event with much lower radiated energy and shorter duration than canonical stellar mergers, plausibly representing the star-planet end of the merger or interaction continuum. In this interpretation, the event links red novae, stellar coalescence, and planet-star interaction transients [2305.04909, 2504.07275].

Soker groups ZTF SLRN-2020 with ASASSN-15qi and ASASSN-13db as the three surviving cases of star-planet-powered ILOTs or Red Novae. A central point is environmental: ASASSN-15qi and ASASSN-13db occurred in young stellar objects, whereas ZTF SLRN-2020 occurred in an old system. This old-system context reduces the ambiguity that can arise from YSO variability and supports the view that star-planet ILOT phenomenology is not confined to young, still-forming planetary systems. Soker contrasts this class with V838 Mon, whose earlier planet-merger interpretation he regards as having failed on energetic grounds [2305.04909].

The source is also classified against more conventional transient categories. Its outburst properties and follow-up spectra differed from ordinary novae or YSO eruptions; during outburst it showed a red, nearly featureless continuum rather than strong atomic emission lines, and later near-IR spectroscopy revealed cool molecular features such as TiO, VO, and CO, indicating a cool, extended photosphere analogous to red novae. Its optical luminosity, of order \(10^{35}\) erg s\(^{-1}\), and duration \(\lesssim 100\) d place it well below classical stellar-merger red novae and on the extrapolated scale expected for merger or engulfment with a planet-mass companion [2504.07275].

## 2. Observational evolution from progenitor to remnant

The observational timeline extends from archival progenitor imaging to late-time JWST spectroscopy. Archival UKIRT near-IR \(H\) and \(K\) images were obtained at \(-12\) yr relative to the optical peak, while other surveys, including Pan-STARRS1, Gaia, and POSS-II, did not detect the progenitor. Mid-IR photometry beginning at \(t=-244\) d showed gradual brightening and dust formation. De et al.-based values quoted in the later modeling literature give total gas+dust masses of \(\approx 2.8\times10^{-5} M_\odot (d/4\,{\rm kpc})^2\) at \(t=-244\) d and \(\approx 10^{-4} M_\odot (d/4\,{\rm kpc})^2\) at \(t=-44\) d. The optical luminosity then rose over about 10 days to a peak around \(L_{\rm peak}\approx 1.3\times10^{35}\) erg s\(^{-1}\), stayed near peak for about 25 d, and faded by roughly an order of magnitude over \(\sim150\) d; the total radiated energy is quoted as \(E_{\rm rad}\approx 6\times10^{41}\) erg for \(d\approx4\) kpc. Soker, using the same observational basis, quotes \(E_{\rm rad}\simeq 6.5\times10^{41}(d/4~{\rm kpc})^2\) erg, \(L_{\rm p}\simeq 1.1\times10^{35}(d/4~{\rm kpc})^2\) erg s\(^{-1}\), a plateau duration of \(\simeq26\) d, a time to radiate \(90\%\) of the energy of \(103\pm20\) d, and a 3-magnitude decline time of \(\approx200\) d [2305.04909, 2507.05365].

The color and spectroscopic evolution are central to its red-nova-like classification. During outburst the continuum was red and nearly featureless rather than line-dominated. Subsequent near-IR spectroscopy revealed TiO, VO, and CO, consistent with a cool, extended photosphere. Dust-shell modeling inferred an inner-edge expansion speed of about \(v_{\rm dust}\approx 35\) km s\(^{-1}\). Crude ejecta values used in the 2023 interpretation were \(M_{\rm ej}\approx 3\times10^{-5}M_\odot\) and \(v_{\rm ej}\approx100\) km s\(^{-1}\), while a later \(+320\) d dust-shell fit yielded \(\log(M_{\rm ej}/M_\odot)=-3.89^{+0.29}_{-3.21}\), corresponding to a best-fit ejecta mass of approximately \(1.3\times10^{-4}M_\odot\) [2305.04909, 2504.07275, 2507.05365].

Late-time characterization was enabled by JWST and Gemini. JWST spectroscopy was obtained on 2022 September 5, about \(+830\) d after the optical \(i\)-band peak at MJD 58993. NIRSpec fixed-slit spectroscopy used G395H/F290LP over \(2.9\)–\(5.1\,\mu\)m at \(R\sim2700\), and MIRI low-resolution spectroscopy covered \(5\)–\(12\,\mu\)m at \(R\sim100\). Near-contemporaneous Gemini-N/NIRI adaptive-optics imaging provided \(J\) and \(H\) photometry consistent with near-IR stability over \(\sim4\) months. The combined \(1\)–\(12\,\mu\)m SED rises from roughly 30 to \(200\,\mu\)Jy toward longer wavelengths, with a near-IR peak around \(H\)-band attributed to the remnant stellar photosphere and longer-wavelength emission attributed to circumstellar thermal dust [2504.07275].

## 3. Merger dynamics and outflow scenarios

A detailed physical model developed by Yarza et al. treats the event as the consequence of orbital decay via stellar tidal dissipation followed by a surface-grazing phase, dynamical plunge, and post-merger evolution. In this picture, a giant planet spirals inward around a star close to the main sequence, begins skimming the stellar surface, shocks and ejects material over months, then plunges into the stellar interior on a timescale of order hours. The observed \(\sim100\) d transient is explicitly argued to be inconsistent with a single episode of dynamical mass ejection. Instead, it is attributed to either hydrogen recombination in an outflow, contraction of a small inflated envelope, or both. The same work concludes that tidal heating within the star was likely unobservable in the archival image taken \(12\) yr before the merger, because even if \(\dot E_{\rm tide}\) approached or exceeded \(L_\star\) near contact, the heated layers could not expand or restructure fast enough in the fiducial cases examined [2507.05365].

The surface-interaction phase is modeled as strongly supersonic. With \(z\equiv R_\star-a\), the Mach number is estimated as
\[
\mathcal{M}\approx16\left(\frac{z}{10^{-2}R_\odot}\right)^{-1/2}\left(\frac{R_\star}{R_\odot}\right)^{1/2},
\]
and the drag-induced orbital decay time near the surface as
\[
\tau_{\rm drag}\approx16~{\rm hour}
\]
for representative parameters given in the paper. Comparing the observed pre-merger dust-inferred masses with the shocked-mass estimate,
\[
M_{\rm shocked}\approx \frac{1}{2}M_{\rm p}\frac{z}{R_\star},
\]
Yarza et al. conclude that planets with \(M_{\rm p}\gtrsim5\,M_{\rm J}\) are most consistent with the months-long pre-merger dust formation [2507.05365].

Soker proposes a different but related accretion-and-jets interpretation. In his scenario, a planet of mass roughly \(M_{\rm p}\simeq0.01M_\odot\) orbits a star of mass \(M_1\simeq0.8\)–\(1.5M_\odot\) and radius \(R_1\simeq1\)–\(4R_\odot\), at a characteristic separation of order \(a\sim2R_\odot\). During the months of pre-outburst activity, the planet may accrete mass from the star, form a disk or accretion structure around itself, and launch jets or a bipolar disk wind. As the planet is engulfed, some of the mass previously accreted by the planet may be stripped and deposited around the star, creating an accretion disk around the star that could launch a second jet episode during the main outburst. Soker states explicitly that this disk-formation sequence “must be confirmed by three-dimensional hydrodynamical simulations,” so it is presented as plausible but speculative rather than demonstrated [2305.04909].

A distinctive feature of the jet-based interpretation is its velocity argument. For illustrative parameters, Soker adopts \(v_{\rm orb}=310\) km s\(^{-1}\), \(v_{\rm p,es}\simeq200\) km s\(^{-1}\), and \(v_{\rm 1,es}\simeq440\) km s\(^{-1}\), with
\[
v_{\rm jet,m,1}=\sqrt{(\beta v_{\rm p,es})^2+v_{\rm orb}^2}.
\]
Escape from the stellar potential requires \(\beta>v_{\rm orb}/v_{\rm p,es}\simeq1.5\). The physical consequence is that only the fastest jet material barely escapes, so the terminal outflow speed can be much smaller than the stellar escape velocity, consistent with the observed dust speed \(v_{\rm d}\simeq35\) km s\(^{-1}\). Because \(v_{\rm p,es}\) and \(v_{\rm orb}\) are of the same order, the escaping flow is expected to be concentrated around \(\theta_{\rm j}\approx45^\circ\) to the equatorial plane, implying mid-latitude bipolar lobes rather than a purely equatorial or purely polar structure [2305.04909].

## 4. Energetics, timescales, and constraints on the companion

Placement on the energy-time diagram is central to the interpretation of ZTF SLRN-2020 as a planet-powered ILOT. Using the radiated energy and characteristic durations quoted above, Soker adopts a timescale range from 26 to 200 d, emphasizing that this range reflects ambiguity in how duration is defined rather than observational uncertainty. The effective timescale from the plateau luminosity is
\[
\frac{E_{\rm rad}}{L_{\rm p}}\simeq70~{\rm days}.
\]
When kinetic energy is not known, the energy-time diagram often assumes \(E_{\rm tot}\approx10E_{\rm rad}\), yielding \(E\simeq6.5\times10^{42}\) erg for \(d=4\) kpc. Using the crude ejecta estimate \(M_{\rm ej}\approx3\times10^{-5}M_\odot\) and \(v_{\rm ej}\approx100\) km s\(^{-1}\) gives \(E_{\rm ej}\approx3\times10^{42}\) erg and \(E_{\rm ej}+E_{\rm rad}\approx3.6\times10^{42}\) erg. On this basis ZTF SLRN-2020 is placed in the same low-energy, moderate-duration region as ASASSN-15qi and ASASSN-13db, clearly below ordinary stellar-merger ILOTs [2305.04909].

Yarza et al. formulate the energetic constraints directly in terms of orbital energy and ejecta mass. For a close-in planet,
\[
E_{\rm orb}=-\frac{GM_\star M_{\rm p}}{2a},
\]
which is of order \(1.8\times10^{45}\) erg for \(M_{\rm p}=1\,M_{\rm J}\) at \(a\sim R_\star\). The energy required to eject mass \(M_{\rm ej}\) at approximately the stellar escape speed is
\[
E_{\rm ej}\approx3.79\times10^{44}~{\rm erg}
\]
for \(M_{\rm ej}=10^{-4}M_\odot\), \(M_\star=M_\odot\), and \(R_\star=R_\odot\). These scalings imply that the observed \(10^{-4}M_\odot\)-scale ejecta require a giant planet rather than an Earth- or Neptune-mass object. The resulting robust lower limit is \(M_{\rm p}\gtrsim1\,M_{\rm J}\), while the preferred constraint from pre-merger dust and surface interaction is \(M_{\rm p}\gtrsim5\,M_{\rm J}\) [2507.05365].

The long optical duration also constrains the powering mechanism. Yarza et al. argue that a recombination transient from a single dynamical ejection should last only hours to less than a day, because \(t\sim R_{\rm phot}/v_{\rm ej}\) gives \(\sim1\) hour for \(R_{\rm phot}\sim3.5R_\odot\) and \(v_{\rm ej}\sim v_{\rm esc,\star}\), or \(\sim0.8\) day even if the dust-shell speed of \(35\) km s\(^{-1}\) is adopted as an extreme lower limit. The observed \(\sim100\) d emission is therefore attributed to slower, extended energy release. One possibility is hydrogen recombination in an outflow, for which the required ejecta mass is
\[
M_{\rm ej}=\frac{E_{\rm rad}}{E_{\rm H}}\frac{m_{\rm p}}{X}\approx3.4\times10^{-5}M_\odot.
\]
Another is contraction of an inflated envelope with \(M_{\rm env}\approx1.1\times10^{-6}M_\odot\), which can reproduce the late-time decline in a Kelvin-Helmholtz contraction framework. The paper regards both channels as plausible contributors [2507.05365].

## 5. Late-time remnant, circumstellar gas, and dust components

The late-time remnant was characterized most extensively by JWST at \(+830\) d. NIRSpec revealed Brackett-\(\alpha\) emission at \(4.05\,\mu\)m and the \(^{12}\)CO fundamental \(\nu=1\!-\!0\) band in emission at \(\sim4.7\,\mu\)m, as well as a tentative broad emission bump at \(\sim4.3\,\mu\)m identified as possible PH\(_3\). Br\(\alpha\) is detected at \(\sim10\sigma\), with integrated flux \((2.5\pm0.4)\times10^{-17}\) erg s\(^{-1}\) cm\(^{-2}\), FWHM \(0.0016\pm0.0001\,\mu\)m, and a luminosity \(\mathrm{L}_{\mathrm{Br}\alpha}\sim10^{-5}\,L_\odot\) at 4 kpc. Using the empirical accretion-line relation quoted in the paper gives \(\mathrm{L}_{\rm acc}\sim4\times10^{-3}L_\odot\) and, for \(M_*/R_*\sim M_\odot/R_\odot\), \(\dot M_{\rm acc}\sim10^{-10}M_\odot\,{\rm yr}^{-1}\). The authors interpret Br\(\alpha\) as evidence for hot circumstellar gas and suggest that it may arise from accreting fallback material around the remnant star [2504.07275].

The CO band was modeled with slab-fitting tools. The inferred parameters are \(T_{\rm CO}=1340^{+130}_{-110}\) K, \(\log(\Omega_{\rm CO}/{\rm sr})=-19.64^{+0.11}_{-0.12}\), and \(\log(N_{\rm CO}/{\rm cm}^{-2})=18.02^{+0.21}_{-0.16}\). Assuming a circular emitting region at 4 kpc gives \(R_{\rm CO}=15.2^{+2.0}_{-2.0}R_\odot\), \(\log(M_{\rm CO}/M_\odot)\approx-13.1^{+0.3}_{-0.3}\), and, for fully molecular gas with solar-like composition and \(N({\rm H_2}/^{12}{\rm CO})=10^4\), \(\log(M_{\rm gas}/M_\odot)\approx-10.2^{+0.3}_{-0.3}\). The interpretation is hot, close-in molecular gas in circumstellar or accretion-flow material rather than distant ejecta [2504.07275].

The PH\(_3\) identification is explicitly tentative. A 300 K iSLAT slab model gives the closest qualitative match to the \(\sim4.3\,\mu\)m feature, whereas 1000 K PH\(_3\) and CO\(_2\) models do not. Assuming an emitting radius of 22 au, the paper estimates \(N_{{\rm PH}_3}\sim2\times10^{15}\) cm\(^{-2}\), a PH\(_3\) mass of \(\sim2\times10^{-11}M_\odot\), and a PH\(_3\)/H\(_2\) abundance of \(\sim10^{-8}\). However, the second strongest PH\(_3\) band at \(8\)–\(11\,\mu\)m is not clearly seen, a nearby feature at \(4.35\,\mu\)m is unidentified, and a narrow apparent absorption is probably a bad pixel artifact, so the identification remains uncertain [2504.07275].

Radiative-transfer modeling of the SED with DUSTY yields two distinct dust components. At \(+830\) d, the preferred DUSTY model has \(T_*=3500\) K, \(L_*=0.29^{+0.03}_{-0.06}L_\odot\), \(T_d=720^{+80}_{-50}\) K, \(\tau_V=0.7^{+0.3}_{-0.1}\), \(R_{\rm in}=50.5^{+9.5}_{-11.0}R_\odot\), and \(\log(M_d/M_\odot)=-10.61^{+0.08}_{-0.16}\); with gas-to-dust \(=100\), the total warm circumstellar mass is \(\log(M_{\rm CSM}/M_\odot)=-8.61^{+0.08}_{-0.16}\). Reanalysis of the \(+320\) d SED gives \(L_*=13.06^{+5.61}_{-10.76}L_\odot\), \(T_d=280^{+450}_{-20}\) K, \(\tau_V=4.2^{+0.4}_{-3.6}\), \(T_*=5800^{+800}_{-2800}\) K, \(R_{\rm in}=4730^{+1600}_{-4420}R_\odot\), and \(\log(M_d/M_\odot)=-5.89^{+0.29}_{-3.21}\), corresponding to \(\log(M_{\rm ej}/M_\odot)=-3.89^{+0.29}_{-3.21}\) for gas-to-dust \(=100\). The warm compact component is interpreted as fallback from the ejecta, whereas the earlier cool, extended, more massive component is associated with dust condensed in the ejecta [2504.07275].

The remnant luminosity is a major diagnostic. From \(L_*=0.29^{+0.03}_{-0.06}L_\odot\) and the mass-luminosity scaling \((L/L_\odot)\approx(M/M_\odot)^4\), the paper infers a host-star mass of about \(0.7M_\odot\), consistent with a K-type main-sequence star. Even allowing for obscuration by an undetected outer dust shell, the corrected luminosity is estimated as only \(0.4\)–\(0.5L_\odot\) for the fiducial cases considered, still consistent with a K star of mass \(\lesssim0.85M_\odot\). A larger distance, up to the previously allowed 7 kpc, would weaken this inference by raising the luminosity to \(\sim0.9L_\odot\), but the preferred 4 kpc solution implies that post-main-sequence stellar expansion is unlikely to have triggered the engulfment [2504.07275].

## 6. Relation to other candidate events and unresolved questions

The nearest comparators are ASASSN-15qi and ASASSN-13db, which have also been interpreted as star-planet interaction transients. ZTF SLRN-2020 occupies the same low-energy, moderate-duration region of the ILOT energy-time plane as those objects, but differs in occurring in an old system rather than a YSO or pre-main-sequence environment. ASASSN-13db is quoted as having a plateau or high-emission duration of \(\approx800\) d, a decline duration of \(\approx55\) d, radiated energy \(E_{\rm rad}\approx2\times10^{41}\) erg, and total energy \(E\sim10^{42}\)–\(2\times10^{42}\) erg. ZTF SLRN-2020 is somewhat more energetic but is treated as belonging to the same broad class. The old-system environment is especially significant because it weakens the claim that the transient could be “just” an accretion burst from youth-related disk processes [2305.04909].

Several alternative interpretations are discussed and generally disfavored. The late-time JWST spectrum is argued to be incompatible with common contaminant classes such as YSO or Herbig Ae/Be systems, because the outburst spectra lacked strong atomic lines, the CO temperature is hotter than typical T Tauri disk values, the spectrum lacks common ice absorptions of embedded YSOs, and MIRI does not show PAHs expected in Herbig Ae/Be environments. An ordinary nova interpretation is also inconsistent with the cool red continuum and red-nova-like molecular evolution. Yarza et al. further argue against a planetary tidal-disruption/accretion-disk transient in the form that would produce a super-Eddington wind luminosity \(L\approx10^{37}\,{\rm erg\,s^{-1}}(M_p/M_J)\) and later hot disk emission, because the observed temperatures stayed \(\lesssim10^4\) K and the luminosity was only \(\sim10^{35}\) erg s\(^{-1}\) [2504.07275, 2507.05365].

At the same time, several physical issues remain open. One is the distinction between true engulfment and tidal disruption. For a K dwarf, the mean stellar density may exceed that of many inflated hot Jupiters, and the literature cited in the 2025 JWST study notes that low-density planets may be tidally disrupted near or above the stellar surface rather than plunging intact into the star. A denser hot Jupiter or hot Neptune could instead be fully engulfed. Another open issue is the driving mechanism of the outflow: Soker’s jet-mediated bipolar picture, with two possible jet episodes and a predicted bipolar expanding nebula, differs from Yarza et al.’s emphasis on surface-grazing shocks, dynamical plunge, recombination outflow, and contraction of a bound inflated envelope. The jet geometry and disk sequence are explicitly model-based and await three-dimensional hydrodynamical confirmation, while the fallback interpretation for the warm dust is physically plausible but not directly proven [2305.04909, 2504.07275, 2507.05365].

Taken together, the current literature converges on a consistent core picture. ZTF SLRN-2020 is treated as a low-energy red nova-like transient produced by the interaction of a near-main-sequence star with a giant planet, with the strongest quantitative constraints favoring a companion at least as massive as Jupiter and probably several Jupiter masses. The main empirical pillars are the pre-merger dust formation over months, the low radiated energy relative to stellar mergers, the \(320\) d ejecta mass near \(1.3\times10^{-4}M_\odot\), the structured late-time remnant revealed by JWST, and the likely low-mass main-sequence nature of the surviving star. The remaining controversy is not whether the event belongs in the star-planet merger class, but how exactly the merger, ejection, fallback, and late-time circumstellar structure were mediated.

Source: https://www.emergentmind.com/topics/ztf-slrn-2020