Calculation of hyperfine structure in Tm II

This paper presents new theoretical calculations of magnetic dipole hyperfine structure constants for low-lying levels in singly ionized thulium (Tm II) using the configuration interaction method with random-phase-approximation corrections, which resolve previous discrepancies and show good agreement with recent experimental measurements.

The Gist

Imagine an atom as a tiny, bustling solar system. The nucleus is the sun, and the electrons are planets whizzing around it. Usually, we think of these planets as just orbiting, but they also have a secret superpower: they act like tiny magnets. The nucleus is a magnet too. When these two magnets interact, they create a subtle "wiggle" in the atom's energy levels. Scientists call this the hyperfine structure.

This paper is about a specific atom called Thulium (specifically, a version of it that has lost one electron, making it a positive ion). Thulium is a bit like a complex, crowded dance floor where the electrons are spinning and jumping in very intricate patterns.

Here is the story of what the author, Andrey Bondarev, did:

The Problem: A Mismatched Puzzle

For a long time, scientists had two different ways of figuring out how strong this magnetic "wiggle" is in Thulium:

1. The Experiment: They used lasers to measure the real atom in a lab.
2. The Theory: They used powerful computers to calculate what the atom should do based on physics rules.

For a long time, these two methods didn't agree. It was like having a map and a GPS that pointed to two completely different locations. A previous study in 1989 found big differences, and a newer study in 2024 found that some of the old measurements were actually wrong (like a typo in a recipe). This left scientists with a confusing picture: the new measurements were better, but the computer calculations still didn't quite match them.

The Solution: A Better Computer Model

The author decided to build a better computer model to solve this mystery. He used a method called Configuration Interaction (CI).

The Analogy:
Imagine trying to predict the weather.

• Old Method: You might just look at the temperature and guess.
• This Paper's Method: You set up a massive simulation that accounts for every single cloud, wind current, and temperature shift, letting them all interact with each other.

In the atom, the "weather" is the electrons. The author let the electrons interact in a complex dance, considering how they bump into and influence one another. He also added a special correction called Random-Phase Approximation (RPA). Think of RPA as adding a "noise-canceling" feature to the simulation. It filters out the static interference caused by the inner electrons (the "frozen core") so the outer electrons can be seen more clearly.

The Results: Finally, a Match!

When the author ran his new, more detailed simulation:

• The Good News: For the lower-energy states of the Thulium ion, the computer results finally matched the new, corrected experimental measurements very well. The "noise-canceling" (RPA) was crucial here; without it, the computer was still off-target.
• The "Why": The author explained that for some energy levels, the magnetic forces from different electrons cancel each other out (like two people pulling a rope in opposite directions). This makes the final result very small and very hard to calculate accurately. The new model handled this delicate balance much better than before.
• The Prediction: Since the model works well for the levels we can measure, the author used it to predict the magnetic "wiggles" for other levels of the atom that haven't been measured yet. These are like predictions for the weather in a city where no one has built a weather station yet.

What About the Glitches?

The model wasn't perfect for every single level. For one specific high-energy level, the computer prediction was still a bit off compared to the experiment. The author suggests this is because that specific electron state is getting "crowded" by other nearby states, creating a complex interaction that the current computer model can't fully untangle yet. It's like trying to hear one person speak in a room where three other people are shouting at the exact same time.

The Bottom Line

This paper is a success story of theory catching up to experiment. By refining the computer calculations and adding the right corrections, the author showed that our understanding of how Thulium ions behave is now much more accurate.

Why does this matter (according to the paper)?
The paper mentions that this work is a stepping stone for experiments with radioactive isotopes of Thulium. Scientists are currently trying to measure the properties of unstable, radioactive versions of this element. To do that, they need to know exactly how the stable version behaves first. This paper provides that reliable "blueprint" so future experiments with radioactive atoms can be planned correctly.

In short: The author fixed the computer model, made it agree with the new lab measurements, and used it to predict the behavior of parts of the atom we haven't seen yet.

Technical Summary

Technical Summary: Calculation of Hyperfine Structure in Tm II

Problem Statement
Recent experimental measurements of the magnetic dipole hyperfine structure (HFS) constants ($A$) in singly ionized thulium (Tm II) have revealed significant discrepancies with previous theoretical calculations. Specifically, earlier work by Mansour et al. [13] reported $A$ values for eight levels that conflicted with Multiconfiguration Dirac-Hartree-Fock (MCDHF) predictions, particularly for the $(4f^{13}(^2F_{5/2})6s_{1/2})^o_2$ level. While a subsequent study by Kebapcı et al. [15] expanded the dataset and corrected two erroneous values from the earlier work, the remaining discrepancies necessitate a renewed theoretical investigation. Additionally, ongoing experimental campaigns at CERN and TU Darmstadt aim to measure nuclear moments in radioactive thulium isotopes, requiring accurate theoretical groundwork on the stable isotope $^{169}$Tm.

Methodology
The authors employ the Configuration Interaction (CI) method to calculate magnetic dipole HFS constants ($A$) and Landé $g$ factors for low-lying levels of the $4f^{13}6s$ and $4f^{13}5d$ configurations in Tm II.

• Electronic Structure: A 14-electron CI approach is used, treating the 54 electrons of the Xe-like core as frozen. The basis set consists of one-electron functions with orbital angular momentum $l \leq 4$ and principal quantum number $n \leq 10$ (designated 10spdfg).
• Correlation Effects: Single and double excitations are allowed from the $4f^{13}6s$ and $4f^{13}5d$ configurations to the full basis set. Higher-rank excitations are excluded due to computational limits, though the authors note their impact on neutral Tm is negligible.
• Core Polarization: To account for the interaction with the frozen core, Random-Phase Approximation (RPA) corrections are applied. These are calculated using an extended basis set (35spdfg) and added to the radial one-electron matrix elements of the hyperfine operator.
• Formalism: The calculations utilize the density matrix formalism to evaluate reduced matrix elements of the hyperfine operator. The Bohr-Weisskopf effect and QED corrections are explicitly excluded, as the former is deemed smaller than electron-correlation uncertainties, and the latter is beyond the scope of this specific study.

Key Contributions and Results
The paper presents a comprehensive set of theoretical $A$ constants and compares them with experimental data from Mansour et al. [13] and Kebapcı et al. [15], as well as previous MCDHF results.

• Agreement with Experiment: The CI+RPA results show good agreement with the new experimental data. For the $4f^{13}6s$ configuration, the inclusion of RPA corrections significantly reduces the discrepancy between theory and experiment. Notably, the RPA corrections yield the correct sign for the $A$ constant of the third level ($J=3$), which was previously calculated with the wrong sign and magnitude.
• Analysis of Discrepancies: The authors analyze the origin of remaining deviations. For levels where the $f$-hole and $s$-electron angular momenta are antiparallel (leading to small $A$ values due to cancellation), calculations are more sensitive. The large discrepancy for the $4f^{13}5d$ level at 21713.74 cm$^{-1}$ is attributed to incomplete capture of interactions with other $J=3$ levels (specifically the one at 23934.73 cm$^{-1}$) within the CI framework, resulting in a less accurate wave function for that specific state.
• Predictions: The study provides reliable predictions for $A$ constants and Landé $g$ factors for 11 experimentally known levels of the $4f^{13}5d$ configuration, many of which lack experimental $A$ data. The calculated $g$ factors generally agree very well with experimental values, serving as a validation of the wave functions used.

Significance and Claims
The authors claim that their work provides a necessary theoretical update motivated by corrected experimental data. The primary significance lies in demonstrating that the CI method, when augmented with RPA corrections, can accurately reproduce the magnetic dipole HFS constants for Tm II, resolving previous inconsistencies.

• The results serve as a foundation for ongoing high-precision laser spectroscopy experiments on stable and radioactive thulium isotopes.
• The authors note that while their current focus is on magnetic dipole interactions, the same approach can be extended to calculate electric field gradients for unstable isotopes with nuclear spin $I \geq 1$, which is crucial for planning future experiments with radioactive beams.
• The paper modestly concludes that the accuracy of the predictions for states with good $g$-factor agreement is expected to be comparable to the well-matched levels, offering a robust dataset for states where measurements are not yet available.