Lanthanide Impact on the Infra-Red Spectra of Nebular Phase Kilonovae

Here is an explanation of the paper "Lanthanide Impact on the Infra-Red Spectra of Nebular Phase Kilonovae," translated into simple, everyday language with creative analogies.

The Big Picture: A Cosmic Firework Show

Imagine two neutron stars (the dense, dead cores of massive stars) crashing into each other. This collision is like a cosmic car crash that creates a massive explosion called a Kilonova. This explosion is a factory that forges heavy elements, like gold and platinum, by smashing atoms together at incredible speeds.

For the first week or so, this explosion is bright and hot, glowing like a normal star. But after about 10 days, the fire dies down, the gas expands, and it becomes a "nebular phase." Think of this as the moment the fireworks stop exploding and you are left looking at the lingering smoke and sparks drifting in the dark.

This paper is about what happens in that "smoke" phase, specifically looking at the infrared light (heat radiation) that our telescopes, like the James Webb Space Telescope (JWST), can see.

The Main Characters: The "Heavy" Elements

In this cosmic smoke, there are two main groups of characters:

The "Light" R-Process Elements: These are the standard heavyweights like Selenium, Tellurium, and Tungsten. They are like the solo singers in a band, making clear, distinct notes.
The Lanthanides: These are a specific group of rare earth elements (like Cerium and Neodymium). The paper calls them the "complex open f-shell" elements.
- The Analogy: If the light elements are solo singers, the Lanthanides are a massive choir with thousands of voices. Because their atomic structure is so complex, they don't just sing one note; they sing a chaotic, overlapping wall of sound. In the language of light, this creates a "forest of lines" that can blur everything together.

The Experiment: Changing the Recipe

The scientists used powerful computer simulations to create 15 different versions of these explosions. They changed two main ingredients in their recipe:

How much stuff was thrown out? (Ejecta Mass: A small pile of debris vs. a huge mountain of debris).
How much "Choir" (Lanthanides) was in the mix? (Some models had almost no Lanthanides, others had a lot).

They watched these simulations evolve from day 10 to day 75 after the crash.

Key Findings: What They Discovered

1. The "Choir" is Loudest Early On and in Dense Crowds

The paper found that the Lanthanide "choir" is most noticeable when the explosion is still dense (early days) and when there is a lot of debris (heavy mass).

The Metaphor: Imagine a crowded room. If the room is packed (high density) and the music just started (early time), the sound of the choir is overwhelming. But as the room empties out (expansion) and time passes, the choir gets quieter and the solo singers (like Tellurium and Selenium) become easier to hear.
The Result: The Lanthanides mostly mess with the light at wavelengths shorter than 4 microns (the Near-Infrared). They act like a fog that blurs the solo singers' voices in that specific range.

2. The Mystery of the "2.1 Micron" Note

Astronomers have seen a bright "note" (emission line) at 2.1 microns in real explosions (like AT2017gfo and AT2023vfi). They thought it was just the element Tellurium singing.

The Paper's Twist: The simulation shows that Tellurium is indeed singing, but it's never singing alone. It's constantly being blended with the "Choir" (Lanthanides like Cerium and Neodymium) and other elements like Zirconium.
The Takeaway: If you see that 2.1 micron note, don't assume it's just Tellurium. It's a team effort, and the Lanthanides are playing a huge part in the harmony.

3. The "Smooth Blackbody" Mystery

One recent explosion (AT2023vfi) looked weird. Instead of seeing distinct notes (lines), the light looked like a smooth, glowing curve (a "Blackbody"), almost like a hot piece of metal. Some scientists thought the Lanthanides were so dense they created a "wall of fog" that smoothed out the light.

The Paper's Verdict: No. The simulations show that even with a huge amount of Lanthanides, the gas is still too thin (optically thin) to create that smooth wall of fog. The "fog" isn't thick enough to hide the individual notes.
The Conclusion: Something else must be creating that smooth glow. Maybe it's dust, or maybe the explosion physics is different than we thought. The Lanthanides alone can't do the job.

4. The "Se" and "W" Debate at 4.5 Microns

In the deeper infrared (4.5 microns), astronomers saw a bright spot and wondered: Is it Selenium (Se) or Tungsten (W)?

The Paper's Verdict: It's almost certainly Selenium. Even in the models with the most Lanthanides, Selenium is just too abundant and sings too loudly. Tungsten is there, but it's a backup singer that gets drowned out.

Why This Matters for the Future

This paper gives astronomers a roadmap for using the James Webb Space Telescope (JWST):

Look at the Near-Infrared (1–4 microns): This is where the Lanthanides are hiding. If you want to know how much "Choir" (Lanthanides) is in an explosion, look here, especially in the first few weeks.
Look at the Mid-Infrared (4–30 microns): This is the "safe zone." Even if the explosion is full of Lanthanides, they mostly stay silent here. This is where you can clearly hear the "solo singers" (Selenium, Nickel, etc.) and figure out what the explosion is made of without the Lanthanide fog getting in the way.

Summary in One Sentence

This study uses computer models to show that while rare earth elements (Lanthanides) create a chaotic "fog" that blurs the light in the early, near-infrared phase of a neutron star collision, they aren't thick enough to create smooth glowing curves, and they mostly stay silent in the deeper infrared, allowing us to clearly hear the other heavy elements singing.

Here is a detailed technical summary of the paper "Lanthanide Impact on the Infra-Red Spectra of Nebular Phase Kilonovae" by Pognan et al. (2026).

1. Problem Statement

Binary neutron star (BNS) mergers produce kilonovae (KNe) powered by the radioactive decay of r-process elements. While the photospheric phase (early times) is dominated by lanthanide opacity causing red colors, the nebular phase (typically $t \gtrsim 10$ days) is characterized by optically thin, forbidden emission lines.

The Gap: Previous studies of nebular phase KNe have largely focused on lanthanide-free ejecta or used semi-analytic methods that fix temperature and ionization structures. There is a lack of comprehensive, self-consistent Non-Local Thermodynamic Equilibrium (NLTE) radiative transfer simulations that explore the specific impact of lanthanide elements (with their complex open f-shell structures) on the Infra-Red (IR) spectra, particularly in the Near-IR (NIR) and Mid-IR (MIR) ranges observed by JWST.
Observational Context: Recent observations of AT2017gfo and AT2023vfi show distinct features (e.g., a 2.1 $\mu$ m feature attributed to Te III, and a 4.5 $\mu$ m feature attributed to Se III and/or W III). AT2023vfi also exhibits a smooth, cool blackbody-like continuum, the origin of which (dust vs. line opacity) is debated.

2. Methodology

The authors employ a rigorous numerical approach to simulate the IR emission of KNe:

Hydrodynamic Models: Ejecta profiles and compositions are based on the DD2-135 model from a symmetric BNS merger simulation (using the DD2 equation of state). The ejecta are separated into dynamical (high lanthanide fraction, $X_{La} \approx 0.054$ ) and post-merger (low lanthanide fraction, $X_{La} \approx 0.002$ ) components.
Parameter Space: The study varies:
- Total Ejecta Mass ( $M_{ej}$ ): 0.005, 0.01, and 0.05 $M_\odot$ .
- Dynamical Fraction ( $f_{dyn}$ ): 0, 0.01, 0.05, 0.1, and 0.5. This creates a grid of 15 models ranging from pure post-merger (lanthanide-free) to lanthanide-rich scenarios.
Radiative Transfer Code: The sumo code (NLTE spectral synthesis) is used in its modified KN version.
- Physics: Solves time-dependent NLTE rate equations for temperature, ionization, and excitation structures using Monte Carlo photon propagation.
- Atomic Data: Significant updates include using r-matrix collision strengths (from OPEN-ADAS) where available, calibrating low-lying energy levels to NIST data, and scaling Einstein A-coefficients for shifted levels.
- Composition: Includes 30 elements (10 light, 10 r-process peaks, 10 lanthanides). Lanthanides are forced to a minimum abundance of $10^{-4}$ even if the hydrodynamic model suggests lower values, to ensure their spectral impact is tested.
Simulation Scope: Spectra are synthesized from 10 to 75 days post-merger over a wavelength range of 1.2 – 30 $\mu$ m.

3. Key Contributions

First Comprehensive NLTE Grid: This is the first study to systematically map the impact of lanthanide mass fractions and total ejecta mass on the IR spectra of nebular phase KNe using time-dependent NLTE simulations.
Updated Atomic Data: The integration of high-precision r-matrix collision strengths and NIST-calibrated energy levels for key species (Te, Se, W, Lanthanides) improves the reliability of predicted line positions and strengths.
Continuum Origin Analysis: A quantitative test of whether line opacity alone (from lanthanides/actinides) can produce the optically thick, cool blackbody continuum observed in AT2023vfi.
Observational Predictions: Provides specific predictions for JWST NIRCam and MIRI filters, identifying which elements dominate specific bands and how they evolve over time.

4. Key Results

A. Lanthanide Impact on Spectra

Wavelength Dependence: Lanthanides primarily affect the spectrum at $\lambda \lesssim 4 \mu$ m. Their impact is strongest at earlier epochs ( $t \sim 10$ days) and for higher ejecta masses (higher densities).
Dominant Species:
- Ce III and Nd II are the most significant lanthanide contributors to spectral formation in the NIR.
- Te III at 2.1 $\mu$ m is a major feature but is always blended with Zr II, Ce III, and Nd II. It rarely dominates the flux alone (contributing 18–72% depending on the model).
- Se III produces a robust double-peaked feature at 4.5 and 5.7 $\mu$ m, dominating the MIR in most models.
- W III at 4.5 $\mu$ m is consistently subdominant to Se III due to lower elemental abundance, even in lanthanide-rich models.
Time Evolution: As time progresses ( $t > 30$ days), lanthanide emission diminishes significantly due to inefficient collisional excitation and photoexcitation of their specific low-lying levels. The spectra become dominated by first-peak elements (Se, Ni, Zr, Y).

B. The Blackbody Continuum in AT2023vfi

Line Opacity Insufficiency: The authors tested whether lanthanide line opacity could create an optically thick photosphere to explain the smooth continuum in AT2023vfi.
Result: Even in the most lanthanide-rich model ( $X_{La} \approx 0.026$ ) with a dense inner boundary ( $v_{in} = 0.02c$ ), the ejecta remain optically thin in the IR at 30 days.
Conclusion: Line opacity from r-process elements alone is insufficient to produce the observed cool ( $T \sim 660$ K) blackbody continuum. Alternative mechanisms (e.g., dust, different opacity sources, or a different progenitor scenario) must be considered.

C. Comparison with Observations

AT2017gfo: Models generally reproduce the 2.1 $\mu$ m and 4.5 $\mu$ m features but struggle to match the exact flux distribution without fine-tuning composition. The 2.1 $\mu$ m feature is confirmed as a blend, not pure Te III.
AT2023vfi:
- The models fail to reproduce the smooth continuum.
- The 4.6 $\mu$ m feature is well-reproduced by Se III (and potentially W III in specific low- $Y_e$ scenarios).
- The 2.2 $\mu$ m feature is not perfectly matched; models show excess emission from Br I/II and lanthanides that is not seen in the data.
- The lack of Br I/II emission in observations suggests the ejecta of AT2023vfi may be deficient in first-peak elements, favoring a scenario with very low electron fraction ( $Y_e \lesssim 0.2$ ), potentially inconsistent with long-lived remnant BNS mergers.

D. Photometric Evolution

Brightest Bands: The F444W (4.4 $\mu$ m) and F550W (5.5 $\mu$ m) JWST bands are typically the brightest due to Se III emission.
NIR Colors: Contrary to the intuition that lanthanides always make ejecta "redder," lanthanide-rich models can appear bluer in specific NIR color indices (e.g., F440W - F200W) at early times ( $t \sim 10-20$ days) because lanthanide emission is strong in the 1–3 $\mu$ m range, boosting the F200W flux.

5. Significance

Interpretation of JWST Data: The paper provides a critical framework for interpreting upcoming JWST observations of KNe. It establishes that MIR spectroscopy ( $\lambda > 4 \mu$ m) is the most reliable way to probe non-lanthanide r-process elements (like Se and Ni), as these regions are less affected by lanthanide blending.
Composition Constraints: The results suggest that distinguishing between mildly lanthanide-rich and lanthanide-poor ejecta in the NIR is difficult at late times, but early-time NIR observations ( $t \sim 10$ days) can constrain the lanthanide fraction.
Progenitor Physics: The inability of standard BNS merger models (with long-lived remnants) to reproduce the specific spectral features of AT2023vfi (particularly the lack of Br and the presence of a continuum) hints that AT2023vfi might originate from a different merger scenario (e.g., short-lived remnant or NS-BH merger) or requires new physics (e.g., dust formation).
Atomic Data Needs: The study highlights the urgent need for complete atomic data (recombination rates, collision strengths) for all r-process ions to reduce uncertainties in NLTE modeling.

In summary, this work demonstrates that while lanthanides significantly shape the early NIR spectra of KNe, their influence fades rapidly in the MIR, where first-peak elements dominate. It also definitively rules out line opacity as the sole source of the cool continuum in AT2023vfi, pointing toward the need for alternative physical explanations.