Tuning between a fractional topological insulator and competing phases at $ν_\mathrm{T}=2/3$

This paper investigates a spinful, time-reversal symmetric flatband model at filling fraction $\nu_T=2/3$, revealing a rich phase diagram that includes a fractional topological insulator, phase separation, a spin-polarized fractional quantum Hall state, and a partially particle-hole transformed Halperin (111) state, with the specific phase determined by the parity of short-range Haldane pseudopotentials.

The Gist

Imagine you are a conductor trying to orchestrate a very special, high-stakes dance party. The dancers are electrons, and the dance floor is a tiny, flat, two-dimensional world made of special materials (like twisted layers of atoms).

The goal of this paper is to figure out how to get these electrons to perform a specific, magical dance move called a Fractional Topological Insulator (FTI). This is a "holy grail" state of matter where electrons act like they have a fraction of their usual charge and move in a way that is protected from getting messed up by noise or impurities. It's like a dance that is so perfectly synchronized that if one dancer stumbles, the whole group automatically corrects itself.

However, getting the electrons to do this specific dance is incredibly difficult. They are stubborn, and they have a tendency to break into other, less interesting dance styles.

Here is a breakdown of what the researchers did, using simple analogies:

1. The Setup: Two Opposite Teams

In this experiment, the electrons come in two "flavors" (spin-up and spin-down). Think of them as two teams: Team Red and Team Blue.

• The Twist: Team Red is dancing in a clockwise circle, while Team Blue is dancing in a counter-clockwise circle. They are mirror images of each other.
• The Problem: Because they are spinning in opposite directions, they naturally want to repel each other or form a chaotic crowd. The researchers wanted to see if they could force them to dance together in a perfect, fractional harmony (the FTI).

2. The Knobs: Turning the Volume on "Push" and "Pull"

The researchers couldn't just change the temperature or the magnetic field easily. Instead, they used a theoretical tool called Haldane Pseudopotentials.

• The Analogy: Imagine the electrons are people at a party. The researchers have a remote control with two main dials:
  • Dial $v_0$ (The "On-Site" Dial): This controls how much the dancers hate standing on top of each other. If you turn this up, they desperately try to avoid being in the same spot.
  • Dial $v_m$ (The "Distance" Dial): This controls how they interact based on how far apart they are. The researchers tested different "distances" (labeled $m=2, 3, 4, etc.$).

3. The Discovery: The "Even-Odd" Surprise

As they turned these dials, they discovered a strange pattern based on whether the distance dial ($m$) was an even or odd number.

• The Even Numbers ($m=2, 4, 6...$): These acted like a "Team Red vs. Team Blue" rule. If the interaction was strong, the two teams would separate completely. Team Red would take over the whole floor, and Team Blue would vanish (or vice versa). This is called Spin Polarization. It's like the two teams refusing to mix and one team kicking the other out of the party.
• The Odd Numbers ($m=3, 5, 7...$): These acted differently. They encouraged the teams to mix, but sometimes they led to a different kind of chaos where the dancers would clump together in weird patterns (Phase Separation) or form a super-conductor-like pair.

4. The Competing Dance Styles

The researchers mapped out a "Phase Diagram," which is like a map of the party showing which dance style wins under which conditions. They found four main outcomes:

1. The Goal (FTI): The perfect, fractional dance. This only happens in a very specific, narrow zone where the "hate for standing on top of each other" ($v_0$) is tuned just right—not too high, not too low. It's a delicate balance.
2. The "Split Party" (Phase Separation): The dancers get angry and split into distinct groups, forming stripes or clumps. The floor is no longer a unified dance floor; it's a messy crowd.
3. The "One-Team Takeover" (Spin Polarized FCI): One team (Red or Blue) dominates the floor, and the other is pushed out. They dance a simpler, integer-based version of the dance, but it's not the special fractional one they wanted.
4. The "Mirror Trick" (PH(111)): A weird state that looks like a superconductor. It's a different kind of order that happens when the interactions are very strong and attractive.

5. The Big Takeaway

The main lesson from this paper is that stabilizing the "Fractional Topological Insulator" is like balancing a pencil on its tip.

• The Challenge: In nature, the "hate" between electrons (Coulomb repulsion) is usually too strong. It pushes the system into the "Split Party" or "One-Team Takeover" modes.
• The Solution: To get the FTI, you need to suppress that initial "hate" (reduce $v_0$). The paper suggests that by using special materials (like dielectric engineering or specific substrates), scientists can "dampen" the repulsion between electrons.
• The Even-Odd Rule: The researchers found that the "distance" at which electrons interact matters immensely. If you tune the interaction to an "even" distance, the teams separate. If you tune it to an "odd" distance, they might mix but form a different pattern.

In Summary

This paper is a recipe book for a very difficult electronic dish. The chefs (physicists) found that to cook the "Fractional Topological Insulator" (the perfect dish), you have to:

1. Use a very specific type of ingredient (twisted materials).
2. Turn down the heat on the "repulsion" between the ingredients.
3. Be careful not to turn the "distance" knobs to the wrong setting (even vs. odd), or the dish will turn into a messy pile of separated ingredients instead of a perfect meal.

This work is crucial because if we can figure out how to stabilize this state, we could build future computers that are incredibly fast and immune to errors, using the unique properties of these "fractional" electrons.

Technical Summary

1. Problem Statement

The paper addresses the challenge of stabilizing the Fractional Topological Insulator (FTI) in spinful, time-reversal (TR) symmetric systems at a total filling fraction of $\nu_T = 2/3$.

• Context: Moiré materials, specifically twisted transition metal dichalcogenides (like MoTe$_2$), offer a platform where spin-valley locking creates bands with opposite Chern numbers ($C_\uparrow = 1, C_\downarrow = -1$) and opposite spins. This mimics a TR-symmetric quantum Hall bilayer.
• The Challenge: While Fractional Chern Insulators (FCIs) have been experimentally realized in these systems, the FTI (a state exhibiting both topological order and fractionalization while preserving TR symmetry) remains elusive.
• Specific Obstacle: Previous studies suggest that the FTI is highly sensitive to interaction details. In the limit of decoupled FCIs (two independent Laughlin states $|1/3\rangle \otimes |\overline{1/3}\rangle$), the ground state is an FTI. However, realistic inter-spin interactions and long-range Coulomb forces often destabilize this state, leading to competing phases like phase separation or spin-polarized ferromagnetism. The precise role of different interaction components in stabilizing or destroying the FTI was not fully mapped.

2. Methodology

The authors employ exact diagonalization (ED) on a torus geometry to study a spinful Lowest Landau Level (LLL) model.

• Model Hamiltonian: The system is modeled using Haldane pseudopotentials ($v_m$), which represent the interaction energy between two particles with relative angular momentum $m$. The Hamiltonian is constructed as:
  $$H_{int} = \sum_{m, \sigma} v_m P_{\sigma\sigma}^m + \frac{1}{2} \sum_{m, \sigma, \bar{\sigma}} v_m P_{\sigma\bar{\sigma}}^m$$
  where $P$ are projectors onto relative angular momentum states.
  • Key Distinction: Due to fermionic statistics and opposite Chern numbers, intra-spin interactions ($\sigma, \sigma$) only involve odd $m$, while inter-spin interactions ($\sigma, \bar{\sigma}$) involve all $m$.
• System Parameters:
  • Total filling $\nu_T = 2/3$ (specifically $\nu_\uparrow = \nu_\downarrow = 1/3$).
  • Flux quanta $N_\Phi = 15$, Electrons $N_e = 10$.
  • Geometry: Torus with aspect ratio 1.
• Phase Identification:
  • Ground State Degeneracy (GSD): Used to identify topological order. The FTI is characterized by a 9-fold degeneracy (3-fold from each spin sector $\times$ 3-fold center-of-mass degeneracy).
  • Quality Parameter: Defined as $s_9/\Delta_9$, measuring the spread of the ground state energy levels relative to the spectral gap. A value near 0 indicates a high-quality FTI.
  • Symmetry Analysis: The authors analyze ground state momentum sectors and symmetry eigenvalues (rotational $C_4$, mirror $M_y$) to distinguish between topological phases and symmetry-broken phases (like phase separation).
• Parameter Space: The study maps phase diagrams by varying the on-site interaction $v_0$ against a single long-range pseudopotential $v_m$ (for $m \in \{2, \dots, 14\}$), while keeping $v_1$ fixed.

3. Key Contributions

1. Comprehensive Phase Diagrams: The authors provide detailed 2D phase diagrams in the $(v_0, v_m)$ plane, revealing the competition between the FTI and various other phases.
2. Identification of Competing Phases: They characterize four distinct families of phases:
   • FTI: The target TR-symmetric state.
   • PH(111): A partially particle-hole transformed Halperin (111) state, adiabatically connected to an inter-spin paired superconductor.
   • Spin-Polarized FCI: A state where one spin species is fully polarized ($\nu_\uparrow=2/3, \nu_\downarrow=0$).
   • Phase Separated (PS) States: States exhibiting spontaneous spatial symmetry breaking and charge ordering.
3. Even-Odd Effect in $m$: The paper uncovers a fundamental parity dependence on the relative angular momentum $m$:
   • Even $m$: Purely inter-spin interactions. Repulsive even-$m$ interactions favor spin polarization to avoid energy penalties.
   • Odd $m$: Involve both intra- and inter-spin interactions. Repulsive odd-$m$ interactions can destabilize the FTI due to dominant intra-spin repulsion.
4. Stabilization Conditions: The study quantifies the specific interaction requirements needed to stabilize the FTI, noting that the ideal Coulomb ratio ($v_0/v_1 \approx 2$) is too high; the on-site interaction must be suppressed (e.g., to $v_0/v_1 \sim 1.25$) to access the FTI phase.

4. Key Results

• FTI Stability: The FTI (GSD=9) exists in a specific region of the phase diagram where $v_0$ is moderately suppressed. It is most stable when competing with full spin polarization, which surprisingly appears beneficial for the FTI quality parameter near the phase boundary.
• The PH(111) State: At large negative $v_0$ (strong attraction), the system enters a state with GSD=1. This is identified as a partially particle-hole transformed Halperin (111) state. It is adiabatically connected to an inter-spin $s$-wave superconductor.
• Phase Separation (PS): Several phase-separated states (PS I, II, III) emerge, characterized by GSDs that are multiples of 4 (e.g., 4, 8, 56). These states break rotational symmetry and exhibit charge ordering. The 56-fold degenerate state (PS III) is a complex mixture of 8 distinct charge-ordered states.
• Impact of $v_m$:
  • Large repulsive $v_{2n}$ drives the system toward spin polarization.
  • Large repulsive $v_{2n+1}$ generally destroys the FTI, favoring phase separation or the PH(111) state.
  • The "ideal" $v_0/v_1$ ratio for the FTI shifts depending on the magnitude of higher-order $v_m$.
• Experimental Relevance: For pure Coulomb interactions, the FTI is unstable. However, the results suggest that dielectric engineering (using substrates with high dielectric constants or metallic gates) to suppress the short-range $v_0$ interaction is a viable pathway to stabilize the FTI in twisted MoTe$_2$.

5. Significance

• Theoretical Guidance: This work provides a roadmap for experimentalists attempting to realize the FTI. It moves beyond the idealized "decoupled" limit to show how realistic interaction tuning (specifically screening to reduce $v_0$) can stabilize the phase.
• Understanding Competing Orders: By mapping the competition between FTI, superconductivity (via PH(111)), and charge order, the paper clarifies the delicate balance required in moiré systems.
• Parity Effects: The discovery of the even-odd effect in pseudopotentials offers a new theoretical lens for understanding multi-component quantum Hall systems with opposite Chern numbers, distinguishing them from standard multi-component systems with equal Chern numbers.
• Material Design: The findings directly inform the design of experiments in twisted transition metal dichalcogenides, suggesting that manipulating the dielectric environment is crucial for accessing the fractional topological insulator regime.