🔭 astrophysics

A comparison of three neodymium atomic data sets for kilonova modeling

This study demonstrates that the choice of atomic data sets for neutral neodymium (Nd I) significantly alters kilonova light curves and spectral structures, with peak luminosities varying by nearly a factor of 1.5 and infrared features being highly sensitive to data source and energy calibration.

Original authors: Christopher J. Fontes, Nicholas Vieira, Chris L. Fryer, Adithan Kathirgamaraju, Oleg Korobkin, Marko Ristić, Ryan T. Wollaeger

Published 2026-04-07

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Christopher J. Fontes, Nicholas Vieira, Chris L. Fryer, Adithan Kathirgamaraju, Oleg Korobkin, Marko Ristić, Ryan T. Wollaeger

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a cosmic fireworks display called a kilonova. This happens when two neutron stars crash into each other, creating a massive explosion that scatters heavy elements across the universe. One of the most important elements in this explosion is Neodymium (a rare earth metal used in magnets and lasers).

Scientists try to predict what these explosions look like to our telescopes by building computer models. Think of these models as a recipe for a cake. You need ingredients (the physics of the explosion) and a specific set of instructions (the atomic data) to know how the cake will taste and look.

This paper is essentially a taste test to see what happens if you change the "instruction manual" for Neodymium.

The Three Chefs (The Atomic Data Sets)

The researchers took three different "instruction manuals" (atomic data sets) for Neodymium, each created by a different team of scientists using different computer programs:

The LANL Team: A group from Los Alamos National Laboratory using their own complex software.
The JLG Team: A collaboration between Japan and Lithuania using a program called HULLAC.
The Autostructure Team: Using a program called Autostructure.

They wanted to see: If we keep the explosion exactly the same but swap out the instruction manual for Neodymium, does the final "cake" (the light and color of the kilonova) change?

The Big Surprise: The Manuals Matter A Lot

The answer was a resounding yes. Even though the explosion physics was identical, the three different manuals produced very different results:

Brightness: The "LANL" manual predicted the explosion would be about 50% brighter at its peak than the other two. It's like one chef saying, "This cake will be huge," while the others say, "It'll be medium-sized."
Color: The most dramatic difference was in the color.
- The JLG and Autostructure manuals predicted the explosion would fade into a standard reddish glow.
- The LANL manual predicted the explosion would turn into a deep, glowing infrared (heat) light that we can't see with our eyes but can detect with special cameras.

The Culprit: The "Neutral" Neodymium

Why did the LANL manual produce such a different result? The researchers found the culprit was a specific version of Neodymium called Neutral Neodymium (Nd I).

Think of Neodymium atoms like musical instruments.

When an atom is "ionized" (stripped of some electrons), it plays high-pitched notes (visible light).
When it is "neutral" (has all its electrons), it plays deep, bass notes (infrared light).

The LANL manual had a much bigger list of bass notes (transitions) for the neutral Neodymium than the other manuals. Because the explosion cools down over time, the Neodymium atoms settle into this "neutral" state. The LANL manual said, "Oh, now that they are neutral, they are going to absorb a ton of visible light and re-emit it as deep infrared heat." The other manuals didn't have as many of these "bass notes," so they didn't predict that massive shift to infrared.

The "Calibration" Experiment

The researchers also tried a third experiment. They took the LANL manual and cross-referenced it with a real-world database (NIST) to fix any errors in the energy levels, like a musician tuning their instrument to a perfect pitch.

Result: When they "tuned" the LANL manual, the deep infrared glow disappeared! The explosion looked much more like the predictions from the other two teams.
Lesson: This proves that tiny errors in how we calculate the energy levels of atoms can completely change our prediction of what the universe looks like.

Why Should You Care?

Imagine you are trying to identify a suspect in a lineup based on a blurry photo.

If your "photo manual" (atomic data) is slightly off, you might think the suspect is wearing a red hat when they are actually wearing a blue one.
In astronomy, if our atomic data is wrong, we might misjudge how much heavy stuff (like gold or uranium) was created in the explosion, or we might miss the explosion entirely because we were looking in the wrong color spectrum.

The Bottom Line

This paper is a warning to scientists: We need to be incredibly precise with our atomic data. Just like a chef needs the exact right measurements to bake a perfect cake, astronomers need the exact right "instruction manuals" for atoms to understand the most violent and beautiful events in the universe.

If we get the atomic physics wrong, we might be looking at the universe through the wrong colored glasses.

1. Problem Statement

Kilonovae (KNe), the electromagnetic transients resulting from binary neutron star or neutron star-black hole mergers, are powered by the radioactive decay of r-process elements. Modeling their light curves and spectra requires accurate atomic physics data, particularly for lanthanides like Neodymium (Nd), which significantly influence opacity and the "lanthanide curtain" effect (shifting emission from optical to infrared).

However, there is a lack of consensus on the best theoretical atomic data sets for these heavy elements. Different codes generate varying energy levels, oscillator strengths, and line lists. This paper addresses the critical question: How sensitive are kilonova observables (light curves and spectra) to the choice of atomic data for Neodymium? The authors aim to quantify the uncertainties introduced by different atomic physics codes to improve the interpretation of future observations (e.g., from JWST).

2. Methodology

The authors employed a controlled numerical experiment to isolate the effects of atomic data while keeping all other physical parameters constant.

Radiative Transfer Code: They used SuperNu, a multi-dimensional, multi-frequency Implicit Monte Carlo (IMC) code with Discrete Diffusion Monte Carlo (DDMC) optimization.
Simulation Setup:
- A simple, semi-analytic 1D spherically symmetric kilonova model.
- Composition: 100% Neodymium (Nd) atmosphere to isolate the element's effects.
- Parameters: Mass $0.014 M_\odot$ , mean velocity $0.125c$ , and a thermalization efficiency of 0.25.
- Physics Assumption: Local Thermodynamic Equilibrium (LTE) was assumed for ionization and excitation populations (a known caveat for late-time modeling).
Atomic Data Sets Compared: Three distinct data sets covering the first four ion stages (Nd I–IV) were generated using different codes:
1. LANL: The Los Alamos National Laboratory suite (specifically the CATS code using the semi-relativistic Hartree-Fock method).
2. JLG (HULLAC): Data from the Hebrew University Lawrence Livermore Atomic Code (parametric potential, Dirac equation), previously used by Tanaka et al. (2020).
3. Autostructure: Based on the SUPERSTRUCTURE code, using kappa-averaged relativistic wavefunctions and Thomas-Fermi-Dirac-Amaldi potentials.
Calibration Study: A specific subset of the LANL data was calibrated by replacing ab initio energy levels with values from the NIST Atomic Spectra Database (ASD) to test the impact of energy accuracy on spectral structure.

3. Key Contributions

Systematic Comparison: This is one of the first studies to directly compare the impact of three major atomic physics codes on kilonova observables using a fixed radiative transfer framework.
Identification of Nd I Dominance: The study isolates the neutral stage (Nd I) as the primary driver of discrepancies in late-time (post-merger) kilonova spectra.
Energy Calibration Protocol: The authors developed and applied a specific calibration technique to map LANL-calculated energy levels to NIST ASD values, demonstrating how this correction alters late-time infrared predictions.
Configuration Sensitivity: They investigated the sensitivity of results to the inclusion of specific electron configurations (e.g., $4f^3 5d^2 6s$ ) in the neutral stage, showing that small changes in configuration lists significantly alter spectral features.

4. Key Results

Light Curve Discrepancies:
- The choice of atomic data significantly alters the bolometric luminosity. The peak luminosities of the HULLAC/Autostructure models differed from the LANL model by a ratio of nearly 1.5.
- The LANL data set produced a steeper decline in optical light curves (r and z bands) compared to the other two sets.
Spectral Evolution:
- Early Time (< 5 days): All three data sets produced similar temperatures and spectra, with bulk emission agreeing well.
- Late Time (> 5 days): Significant divergence occurred as Nd I became the dominant ion species. The LANL data set produced significantly redder spectra with strong, structured features in the mid-infrared (mid-IR) range.
- The JLG and Autostructure sets produced structurally similar light curves but differed by $\sim$ 2 AB magnitudes in optical bands at late times.
Role of Neutral Nd (Nd I):
- When the authors artificially removed Nd I lines from the LANL data (creating "NN-LANL"), the resulting bolometric luminosity and light curve shapes aligned much more closely with the JLG and Autostructure results.
- This confirms that the neutral stage (Nd I) is responsible for the unique mid-IR line structure and the "red" shift observed in the LANL simulations.
Impact of Calibration:
- Calibrating LANL energies to NIST values significantly reduced the mid-IR line structure at late times, making the spectra dimmer and less structured.
- The calibration of Nd I specifically was found to be the dominant factor in late-time spectral changes, while calibration of higher ionization stages (Nd II–IV) had negligible effects at late times due to low population fractions.
Data Metrics:
- The LANL data set contained the highest number of levels and lines for Nd I (approx. 18,000 levels vs. ~2,000 for JLG).
- The LANL data set featured unique, strong mid-IR resonance lines between the $4f^4$ ground configuration and excited states, which were less prominent or absent in the other data sets.

5. Significance and Conclusions

Atomic Physics Uncertainty: The study highlights that uncertainties in atomic physics data for a single lanthanide element (Nd) can lead to order-of-magnitude differences in predicted kilonova observables. This underscores the need for spectroscopically accurate, experimentally validated data for lanthanides and actinides.
Modeling Implications: The "lanthanide curtain" effect and late-time infrared emission are highly sensitive to the treatment of neutral atoms. Current models relying on ab initio calculations without calibration may overestimate mid-IR opacity and structure.
Future Directions: The authors conclude that to accurately interpret kilonova observations (from AT2017gfo to future events), the community must:
1. Prioritize the calibration of neutral lanthanide energy levels to experimental databases (like NIST).
2. Develop non-LTE (non-Local Thermodynamic Equilibrium) models, as LTE assumptions may obscure the true impact of atomic data uncertainties at late times.
3. Systematically evaluate other uncertainties (composition, morphology) alongside atomic data improvements.

In summary, this paper demonstrates that the choice of atomic data is not merely a technical detail but a fundamental source of uncertainty in kilonova modeling, with the neutral stage of Neodymium playing a decisive role in shaping the infrared signature of these events.