Does the total energy difference method for modelling core level photoemission fail for bigger molecules?

This study challenges the notion that the $\Delta$SCF method fails for larger molecules by demonstrating through new experimental and computational results on anthrone and a 44-molecule dataset that the method accurately predicts core electron binding energies for systems up to 40 atoms.

The Gist

Imagine you are trying to understand the "soul" of a molecule by listening to the specific notes it sings when you zap it with X-rays. This is what scientists do in X-ray Photoelectron Spectroscopy (XPS). They want to know exactly how much energy it takes to knock an electron out of the deep, inner core of an atom.

For a long time, there was a popular computer method called $\Delta$SCF (Delta-SCF) used to predict these energy notes. Think of this method as a very talented, budget-friendly musician who can play a perfect solo for a small, simple song (a small molecule like methane).

However, a rumor started spreading in the scientific community: "This musician is great for small songs, but if you give them a complex, 25-atom symphony (like the anthrone molecule), they will completely miss the notes."

This paper is the story of how the authors went to investigate that rumor, played the song themselves, and proved the rumor wrong.

The Mystery of the "Big Molecule" Failure

The rumor was based on a specific case study of a molecule called anthrone. Previous studies claimed that when scientists used the $\Delta$SCF method on anthrone, the computer predicted energy levels that were way off compared to real-world experiments. The error was so big (up to 1.5 eV) that people started to believe the method simply "breaks" when molecules get too big or complex.

It was like saying, "This calculator works perfectly for adding 2+2, but if you try to add 1,000 numbers, it will give you the wrong answer."

The Investigation: Re-measuring the Song

The authors of this paper decided to play detective. They didn't just trust the old data; they went back to the lab and the computer to re-examine anthrone.

1. The New Experiment: They went to a massive particle accelerator (a synchrotron) in Sweden to measure the actual energy notes of anthrone gas. They were extremely careful, calibrating their instruments with Argon gas (like using a tuning fork) to ensure their measurements were perfect.
2. The New Calculation: They ran the $\Delta$SCF computer simulation again, but this time using a modern, highly accurate "recipe" for the math (called the SCAN functional).

The Big Reveal

Here is the twist: The old experimental data was wrong.

When they compared their new measurements with the new computer calculations, the two matched almost perfectly. The "failure" of the method wasn't because the method was bad for big molecules; it was because the previous experiment had the wrong reference point.

It turns out the $\Delta$SCF method is like a versatile musician who can play both a simple folk song and a complex symphony just as well. The "big error" was just a bad recording of the symphony, not a bad performance by the musician.

Testing the Method on a Whole Band

To be absolutely sure, the authors didn't just stop at anthrone. They gathered a "band" of 44 different core electron energy measurements from 15 different medium-sized molecules (ranging from 10 to 40 atoms). These included:

• Carborane clusters: Molecules that look like tiny, hollow cages made of boron and carbon.
• Organometallics: Molecules with metals like Iron or Cobalt mixed in.
• Polycyclic aromatics: Flat, ring-shaped molecules like anthracene.

They ran the $\Delta$SCF method on all of them. The result? The average error was incredibly small (0.19 eV). This is comparable to the accuracy seen in tiny molecules.

The "Why" Behind the Magic

You might wonder, "Why does this method work for big things when other methods fail?"

The authors explain it using a concept called localization.

• Imagine a stadium full of people (a big molecule). If one person in the back row stands up (an electron is removed), the people right next to them might react, but the people in the front row don't really care.
• In core-level photoemission, the "hole" left behind is always stuck to one specific atom. It's a local event. The $\Delta$SCF method is smart enough to focus on that specific neighborhood and calculate how the electrons rearrange there, without getting confused by the size of the whole stadium.

The Takeaway

This paper is a "myth-busting" success story. It tells us:

1. Don't judge a book by its cover (or a method by one bad data point): The $\Delta$SCF method does not fail for big molecules.
2. The Rumor was a False Alarm: The previous "failure" with anthrone was due to experimental errors, not a flaw in the theory.
3. The Method is Robust: Scientists can now confidently use this affordable, fast computer method to study complex, large molecules, from organic chemicals to materials used in new technologies.

In short, the "budget-friendly musician" can indeed handle the symphony. The method is ready to be used on all sorts of complex systems, opening the door for better design of new materials and drugs.

Technical Summary

1. Problem Statement

The $\Delta$-Self-Consistent-Field ($\Delta$SCF) method, based on Density Functional Theory (DFT), is a widely used, computationally efficient approach for calculating core electron binding energies (BEs) in X-ray Photoelectron Spectroscopy (XPS). While the method has proven highly accurate for small molecules, previous literature suggested a significant degradation in accuracy for larger, more complex systems.

• Specific Controversy: A notable discrepancy was reported for the anthrone molecule (25 atoms). Previous studies using the PBE0 hybrid functional showed large errors (0.7–1.5 eV) between calculated and experimental C 1s binding energies, leading to the hypothesis that $\Delta$SCF lacks size-extensivity or fails for molecules with fused aromatic rings.
• Theoretical Concern: There was a concern that as system size increases, the $\Delta$SCF result for ionization might simply converge to the highest occupied orbital eigenvalue (a failure mode known for valence ionization in extended systems), rendering it useless for large molecules. However, theoretical arguments suggest core holes are localized and should induce strong screening regardless of system size.

2. Methodology

The authors employed a combined experimental and computational approach to re-evaluate the validity of $\Delta$SCF for medium-to-large molecules.

Computational Methods:

• Software: All-electron DFT calculations were performed using FHI-aims.
• Functional: The SCAN (Strongly Constrained and Appropriately Normed) meta-GGA functional was used, implemented via dfauto.
• $\Delta$SCF Protocol:
  • Binding energies were calculated as the total energy difference between the $N$-electron ground state and the $(N-1)$-electron final state with a core hole.
  • The core hole was simulated by fixing the occupancy of a specific Kohn-Sham state to zero in the spin-up channel.
  • Basis Sets: Standard "tight" basis sets were used for geometry optimization. For the atom with the core hole, a larger, core-hole-adapted basis set was employed to facilitate electron relaxation.
• Valence Spectrum: A $G_0W_0$ calculation (one-shot) based on a PBE0 starting point was performed for the valence band to compare with experimental valence spectra.
• Dataset: The study evaluated 44 core electron binding energies across 15 medium-sized molecules (10–40 atoms), including organometallics, carborane clusters, and polyaromatic systems.

Experimental Methods:

• Facility: Experiments were conducted at the FinEstBeAMS beamline of the MAX IV synchrotron (Lund, Sweden).
• Sample: Gas-phase anthrone was introduced via an effusion cell heated to 50°C.
• Calibration: Absolute binding energy scales were calibrated using the Ar 2p$_{3/2}$ line (248.63 eV) for core levels and the H$_2$O 1b$_1$ peak (12.62 eV) for valence levels.
• Measurements: High-resolution XPS spectra were recorded for C 1s, O 1s, and valence regions using a Scienta R4000 hemispherical analyzer.

3. Key Contributions

1. Re-measurement of Anthrone: The authors performed new, high-precision gas-phase XPS measurements on anthrone, correcting previously published experimental values.
2. Benchmarking SCAN: They demonstrated that the $\Delta$SCF method using the SCAN functional yields high accuracy for medium-sized molecules, challenging the notion that the method fails for larger systems.
3. Dataset Expansion: The study provides a rigorous benchmark dataset of 44 core BEs for 15 diverse molecules, moving beyond simple small molecules to complex fused-ring and organometallic systems.
4. Resolution of Discrepancy: The paper attributes the previously reported large errors in anthrone not to a failure of the $\Delta$SCF method, but to inaccuracies in the reference experimental data used in prior studies.

4. Results

• Anthrone C 1s Spectrum:
  • New Experimental Values: The authors measured C 1s peaks at 290.3 eV (main peak) and 292.7 eV (shoulder). These differ significantly from previous literature values (290.72 eV and 293.6 eV).
  • Theoretical Agreement: The $\Delta$SCF/SCAN calculation predicted peaks at 290.11 eV and 292.28 eV. The agreement is excellent, with a mean absolute error (MAE) of 0.19 eV for the anthrone dataset.
  • Spectral Features: The calculation correctly resolved three features: sp$^2$ carbons, the sp$^3$ carbon (C10), and the carbon adjacent to oxygen (C9).
• General Dataset Performance:
  • Across the 44 data points (15 molecules), the Mean Absolute Error (MAE) was 0.19 eV.
  • The Mean Error (ME) was -0.05 eV, indicating no systematic bias.
  • Outlier Analysis: One data point (C1 in 1,5-C$_2$B$_3$(C$_2$H$_5$)$_5$) showed a large deviation (1.6 eV), likely due to an anomalous experimental value (287.2 eV is unusually low for a gas-phase C 1s). Excluding this outlier reduced the MAE to 0.15 eV.
  • Carborane Clusters: The method successfully reproduced binding energies for complex carborane clusters, including cases where atom assignments were swapped to improve agreement with theory.
• Valence Band: The calculated valence spectrum matched the experimental shape well, though the theoretical splitting of peaks near 9–10 eV was slightly underestimated compared to the resolved experimental features.

5. Significance

• Validation of $\Delta$SCF for Large Systems: The study conclusively demonstrates that the $\Delta$SCF method does not fail for larger molecules (up to 40 atoms). The accuracy remains comparable to that achieved for small molecules, provided high-quality experimental references and appropriate functionals (like SCAN) are used.
• Correction of Literature: It corrects the historical record regarding anthrone, showing that the perceived failure of the method was actually a failure of previous experimental calibration or reference data.
• Theoretical Justification: The results support the theoretical argument that core holes are localized and induce strong local screening, meaning the method remains size-extensive for core-level excitations even in extended systems.
• Practical Impact: This validates the use of $\Delta$SCF as a reliable, low-cost tool for interpreting XPS spectra of complex organic materials, organometallics, and extended systems, reducing the need for more expensive methods like GW for routine core-level analysis.