Refined DFT recipe and renormalisation of band-edge parameters for electrons in monolayer MoS$_2$ informed by the measured spin-orbit splitting

By employing a refined DFT+U+V framework to account for orbital mixing, the authors successfully reconciled a significant discrepancy between measured and theoretical conduction band spin-orbit splitting in monolayer $\text{MoS}_2$, while simultaneously improving the accuracy of valence-band splitting and the quasiparticle band gap.

The Gist

The Mystery of the "Missing" Spin: A Tale of Two Recipes

Imagine you are a master chef trying to recreate a legendary, secret soup recipe. You follow the instructions perfectly—the exact amount of salt, the precise temperature, the specific brand of broth—but when you taste it, it’s completely wrong. It’s bland, whereas the original was spicy and bold.

This is exactly what scientists have been facing with monolayer MoS2 (a paper-thin material used in next-generation electronics). They’ve been using a standard "recipe" called DFT (Density Functional Theory) to predict how electrons behave in this material. But there was a massive problem: the recipe predicted the electrons would have almost no "spin-orbit splitting" (a specific way electrons spin and move), while real-world experiments showed they had ten times more than expected.

Here is how the researchers solved this mystery.

---

1. The "Crowded Room" Effect (Many-Body Interactions)

First, the researchers realized that the standard recipe assumes electrons are like polite guests in a large, empty ballroom, each moving independently.

In reality, the electrons in MoS2 are more like people in a packed, crowded subway car. Because they are so close together, they don't just move; they push, pull, and influence each other through electrical forces. This "crowding" (which scientists call exchange enhancement) actually makes the electrons' spin properties appear stronger.

The team used complex math to account for this "crowd effect." It explained about half of the discrepancy, but there was still a huge gap left. The recipe was still failing.

2. The "Hidden Ingredient" (The Orbital Mix)

To find the rest of the missing "spice," the researchers had to look at the microscopic structure of the material itself.

Think of the atoms in MoS2 as building blocks. The electrons live in specific "rooms" called orbitals. The researchers discovered that the standard recipe was miscalculating how much the "Molybdenum rooms" and the "Sulphur rooms" were mixing together.

In the old recipe, these rooms were treated as mostly separate. But in reality, they are heavily interconnected. It’s like a house where the kitchen and the living room have a massive, open doorway. Because the recipe didn't realize how much these rooms were "leaking" into each other, it couldn't see the full strength of the electron's spin.

3. The New Recipe: DFT+U+V

To fix this, the team developed a "refined recipe" called DFT+U+V.

• The "U" (The On-Site Rule): This tells the computer, "Hey, electrons in the same room really dislike being near each other; give them more space!"
• The "V" (The Neighbor Rule): This tells the computer, "And don't forget, electrons in neighboring rooms also feel each other's presence!"

By adding these two "tuning knobs" (the U and the V), the researchers finally hit the jackpot. Their new simulation didn't just fix the spin problem; it also correctly predicted the material's "band gap" (the energy needed to make it conduct electricity) and its "valence band" properties.

Why does this matter?

We are currently in the middle of a revolution in technology, moving from traditional electronics to Spintronics—devices that use the spin of an electron rather than just its charge. This could lead to computers that are faster, smaller, and use much less power.

However, to build these devices, we need a perfect "blueprint" of how materials behave. By fixing this broken recipe, these scientists have provided a much more accurate map, helping us navigate the future of ultra-fast, tiny, and efficient technology.

Technical Summary

Technical Summary: Refined DFT Recipe and Renormalisation of Band-Edge Parameters for Electrons in Monolayer $\text{MoS}_2$

1. Problem Statement

A significant discrepancy exists between theoretical predictions and experimental measurements regarding the conduction band (CB) spin-orbit splitting (SOS) in monolayer molybdenum disulphide ($\text{MoS}_2$).

• The Discrepancy: Conventional Density Functional Theory (DFT) predicts a negligible CB SOS (only a few meV), whereas experimental magnetotransport measurements indicate a much larger splitting ($\geq 10$ meV).
• The Complexity: The CB SOS is determined by a delicate balance between the contribution of sulphur $p_x, p_y$ orbitals and the second-order perturbative mixing of molybdenum $d_{z^2}$ orbitals with $d_{xz}, d_{yz}$ orbitals. In standard DFT, these two opposing contributions tend to cancel each other out, leading to the underestimated values.
• The Goal: To reconcile these values by accounting for many-body electron-electron interactions and refining the DFT approach to better capture the orbital hybridization.

2. Methodology

The authors employ a multi-scale modeling approach combining experimental magnetotransport data with advanced theoretical frameworks:

• Experimental Analysis (Magnetotransport): The study uses Shubnikov–de Haas oscillations (SdHO) from dual-gated $\text{MoS}_2$ devices to determine:
  • The threshold density ($n^*$): The density at which the upper spin-split band begins to populate.
  • The effective mass ($m_l$): The band mass of electrons in the lower spin-split subband.
• Many-Body Renormalization (Mesoscale): To extract the "bare" (single-particle) SOS ($\Delta_0$) from the measured $n^*$, the authors account for:
  • Exchange Enhancement: The SOS is enhanced by electron-electron exchange interactions in a finite-density electron gas.
  • Mass Renormalization: The electron effective mass is increased due to correlation effects.
  • They utilize a screened-exchange electron self-energy calculation using the Keldysh potential to model 2D dielectric screening.
• Refined DFT (Microscopic): To resolve the remaining gap between the "bare" SOS and DFT results, the authors implement a DFT+$U$+$V$ framework. This includes:
  • On-site Hubbard $U$: To correct the local electron correlation on the metal site.
  • Inter-site Hubbard $V$: To account for the interaction between the metal (Mo) and the chalcogen (S) sites, thereby fine-tuning the Mo-S hybridization.

3. Key Contributions

• Identification of the "Accidental Cancellation": The paper demonstrates that the small SOS in standard DFT is not due to a lack of spin-orbit coupling, but due to an accidental cancellation of two competing orbital mechanisms.
• Development of a "DFT+$U$+$V$ Recipe": The authors provide a specific computational recipe (optimizing $U$ and $V$ parameters) that corrects the orbital composition of the band-edge states.
• Unified Parameter Set: They propose a single set of Hubbard parameters that simultaneously corrects the conduction band SOS, the valence band SOS, and the quasi-particle band gap.

4. Results

• Extraction of Bare SOS: By accounting for exchange and mass renormalization, the authors determine that the true bare SOS ($\Delta_0$) is approximately $8.1$ meV, significantly higher than previous DFT estimates.
• Optimal Hubbard Parameters: The optimized parameters were found to be $U^* = 0.59$ eV and $V^* = 0.5$ eV.
• Improved Accuracy: Using the $U^*+V^*$ recipe, the model yields:
  • CB SOS ($\Delta_0$): $8.1$ meV (matching experimental extraction).
  • Valence Band SOS ($\Delta_v$): $155$ meV (consistent with prior experiments).
  • Quasiparticle Band Gap ($E_g$): $1.90$ eV (improving upon the $\approx 1.7$ eV predicted by standard DFT and sitting between optical and quasiparticle gap bounds).
• Mass Consistency: The calculated effective masses remained stable and consistent with experimental observations across the analyzed parameter range.

5. Significance

This work provides a crucial bridge between microscopic electronic structure theory and macroscopic experimental observations in 2D semiconductors.

• For Spintronics: Accurate knowledge of the CB SOS is vital for predicting the competition between bright and dark excitons and for designing spin-valleytronic devices.
• For Computational Physics: The study offers a computationally efficient alternative to much more expensive methods (like $GW$ calculations). It demonstrates that incorporating inter-site correlations ($V$) is essential for accurately modeling the electronic properties of transition metal dichalcogenides (TMDs), where metal-chalcogen hybridization dictates the physics.