Classification of topological insulators and superconductors with multiple order-two point group symmetries

This paper presents a general method for computing the classification groups of topological insulators and superconductors under arbitrary $\mathbb{Z}_2^{\times n}$ point group symmetries, demonstrating that these classifications are fully determined by the number of momentum and real-space variables flipped by individual and paired symmetry generators.

The Gist

Imagine you are an architect trying to build a new kind of "smart" material. These aren't just bricks; they are Topological Insulators and Superconductors. Think of them as materials that act like perfect insulators (blocking electricity) on the inside, but act like perfect conductors (letting electricity flow freely) on their edges, corners, or hinges.

For a long time, scientists knew how to classify these materials if they only had to worry about basic internal rules, like time-reversal symmetry (imagine a movie playing backward). But nature is more complex. Crystals have shapes, and shapes have symmetries. If you rotate a crystal 180 degrees or flip it like a pancake, it might look the same. These are Point Group Symmetries.

This paper by Ken Shiozaki is essentially a master instruction manual for figuring out what kinds of these "smart" materials can exist when you have multiple of these symmetries working at the same time.

Here is the breakdown using simple analogies:

1. The Problem: Too Many Rules, Too Many Confusing Combinations

Imagine you are trying to sort a massive pile of LEGO bricks.

• The Old Way: You only had to sort them by color (Internal Symmetry). Easy.
• The New Challenge: Now, you also have to sort them by shape, and you have to consider that some bricks can be rotated, flipped, or mirrored.
• The Specific Issue: Scientists already knew how to handle one symmetry (like just rotation). But what if you have two symmetries at once? Maybe you have a rotation and a mirror flip? And what if those two symmetries get along (commute) or fight (anticommute)?

When you add more symmetries, the number of possible combinations explodes. It becomes a chaotic mess of "What if I do this, then that?" The paper asks: "Is there a simple way to predict the outcome without building every single possibility?"

2. The Solution: The "Dimensional Elevator"

The author's brilliant insight is using a mathematical tool called Suspension Isomorphism. Let's call this the "Dimensional Elevator."

Imagine you are trying to understand a complex 3D sculpture. Instead of staring at the whole thing, the author says: "Let's take an elevator down to the basement (0 dimensions)."

• The Trick: The paper shows that no matter how many dimensions (space, time, momentum) your material has, the answer to "What kind of topological phase is this?" is mathematically identical to the answer for a single point in zero dimensions, provided you adjust a few "settings."
• The Analogy: It's like realizing that the complexity of a whole city's traffic pattern can be predicted by looking at a single intersection, as long as you know how many cars are entering and leaving. You don't need to map the whole city; you just need the "intersection data."

3. The "Settings" (The Parameters)

The paper reveals that you don't need to know the entire history of the material. You only need to count a few specific things, which act like dials on a control panel:

1. How many variables does Symmetry A flip? (e.g., Does it flip the x-axis? The y-axis?)
2. How many variables does Symmetry B flip?
3. How many variables do they flip together? (The overlap).
4. Do they commute? (If you rotate then flip, is it the same as flipping then rotating?)

The paper proves that only these counts matter. It doesn't matter if you have 3 dimensions or 100 dimensions; if the "flip counts" are the same, the classification is the same.

4. The Result: The "Periodic Table" of Symmetries

Just as Mendeleev created a periodic table for chemical elements, this paper creates a Periodic Table for Topological Materials with Symmetries.

• The Table: The paper provides a giant chart (Table 6 in the text).
• How to use it: You look at your material, count your "flip numbers" (the dials mentioned above), and plug them into the formula.
• The Output: The table instantly tells you the "Classification Group." In plain English, this number tells you how many distinct types of stable, protected states your material can have.
  • Is it just one type? (Group = Z)
  • Is it two types? (Group = Z2)
  • Is it four types? (Group = Z4)
  • Or is it impossible to have a stable state? (Group = 0)

5. Why This Matters

Before this paper, if you wanted to design a new "higher-order" topological insulator (one that conducts electricity only at its corners, not its edges), you had to do incredibly difficult, custom math for every new combination of symmetries.

This paper gives you a universal calculator.

• For Physicists: It saves years of calculation.
• For Engineers: It helps design materials with specific, exotic properties by simply choosing the right symmetries.
• For the Future: It paves the way for discovering "higher-order" materials that could be used in ultra-efficient electronics or quantum computers.

Summary Metaphor

Think of the universe of these materials as a giant, multi-layered cake.

• Old View: To know what flavor the cake is, you had to taste every single crumb.
• This Paper: The author found a recipe card. He says, "You don't need to taste the whole cake. Just count how many layers of chocolate and vanilla are mixed, and check if the layers are stacked straight or twisted. Based on those two numbers, I can tell you exactly what the flavor is."

It turns a chaotic, infinite problem into a simple, finite counting game.

Technical Summary

1. Problem Statement

The classification of topological phases (insulators and superconductors) is well-established for systems with internal symmetries (Time-Reversal, Particle-Hole, Chiral) and for systems with a single crystalline symmetry (e.g., a single $\mathbb{Z}_2$ point group symmetry). However, a systematic and computationally efficient framework for classifying topological phases protected by multiple simultaneous $\mathbb{Z}_2$ point group symmetries (forming a group $\mathbb{Z}_2^n$) has been lacking.

While general K-theoretic methods exist (e.g., Atiyah-Hirzebruch spectral sequence), they are often computationally heavy. The specific challenge addressed here is to derive explicit classification groups for topological phases in the presence of arbitrary numbers of unitary $\mathbb{Z}_2$ symmetries, accounting for:

• The specific Altland-Zirnbauer (AZ) class (internal symmetries).
• The algebraic relations between the $\mathbb{Z}_2$ generators (commutation/anticommutation).
• The geometric action of these symmetries on momentum ($k$) and real-space ($r$) variables (i.e., which variables are flipped/inverted).

2. Methodology

The author employs K-theory and suspension isomorphisms to reduce the complexity of the classification problem.

A. Parameterization of Symmetries

The system is defined by:

• AZ Class ($s$): Labeled by an integer $s \in \{0, \dots, 7\}$ for real classes (or $0,1$ for complex), encoded in binary to track Time-Reversal (TRS), Particle-Hole (PHS), and Chiral symmetries.
• Additional Symmetries ($U_i$): $n$ unitary $\mathbb{Z}_2$ generators. Each generator $U_i$ is characterized by:
  • Symmetry Type ($t_i$): A binary pair $(t_{i1}, t_{i0})$ determining if $U_i$ commutes/anticommutes with TRS/PHS and if $U_i^2 = \pm 1$.
  • Geometric Action: Defined by bit strings $p_i$ (momentum flips) and $q_i$ (real-space flips).
  • Mutual Relations: Defined by $u_{ij} \in \{0,1\}$, where $U_i U_j = (-1)^{u_{ij}} U_j U_i$.
• Flipping Counts: The method introduces integer counts for variables flipped by single generators ($d_i, D_i$) and pairs of generators ($d_{ij}, D_{ij}$), representing the overlap of symmetry actions in momentum and real space.

B. Suspension Isomorphism

The core theoretical tool is the construction of suspension isomorphisms in K-theory.

• The author constructs a mapping between a non-chiral class ($s_0=0$) and a chiral class ($s_0=1$) by adding an auxiliary dimension (parameter $\theta$) to the Hamiltonian.
• This process allows the reduction of the classification problem from a $(d+D)$-dimensional parameter space (momentum + defect space) to a zero-dimensional (0D) problem.
• Crucially, this reduction reveals that the classification depends only on the difference between the number of flipped momentum variables and real-space variables ($\delta = d - D$) and the corresponding differences for the symmetry generators ($\delta_i = d_i - D_i$).

C. Reduction to 0D

Through repeated application of the suspension isomorphism, the K-group for arbitrary dimensions is shown to be isomorphic to a 0D K-group:
$$K_{R+nU}(\dots; d, \dots; D, \dots) \cong K_{R+nU}(s - \delta, \{t_i - \delta_i\}, \{u_{ij} + \dots\})$$
This implies that the classification is determined entirely by the effective AZ class and the effective symmetry parameters in zero dimensions, which depend only on the parity of the variable flips.

3. Key Contributions

1. General Reduction Formula: The paper provides a rigorous derivation showing that the classification of topological phases with $n$ unitary $\mathbb{Z}_2$ symmetries is fully determined by a small set of parameters: the effective AZ class index, the effective symmetry types ($t_i$), the mutual commutation relations ($u_{ij}$), and the dimensionality difference ($\delta$).
2. Independence of Higher-Order Overlaps: A significant theoretical finding is that the classification groups do not depend on the number of variables simultaneously flipped by three or more generators ($d_{ijk}, d_{ijkl}, \dots$). The classification is insensitive to triple or higher-order overlaps of symmetry actions, depending only on single and pairwise overlaps.
3. Explicit Classification Tables for $\mathbb{Z}_2 \times \mathbb{Z}_2$: The author explicitly computes the complete classification tables for the case of two additional $\mathbb{Z}_2$ symmetries ($\mathbb{Z}_2^2$). This covers all combinations of:
   • Real and Complex AZ classes.
   • All possible symmetry types ($t_1, t_2$) and commutation relations ($u_{12}$).
   • All physically relevant geometric configurations (e.g., onsite + reflection, reflection + rotation, etc.) up to three spatial dimensions.
4. Handling Antiunitary Symmetries: The framework is extended to include complex AZ classes with antiunitary additional symmetries, showing they reduce to real AZ classes with fewer unitary symmetries.

4. Results

The primary results are presented in Table 6, which serves as a periodic table for topological insulators and superconductors with $\mathbb{Z}_2 \times \mathbb{Z}_2$ symmetry.

• Input Parameters: The table takes as input the AZ class ($s$), the symmetry types ($t_1, t_2$), the commutation relation ($u_{12}$), and the geometric flipping differences ($\delta, \delta_1, \delta_2, \delta_{12}$).
• Output: The resulting classification group (e.g., $\mathbb{Z}$, $\mathbb{Z}_2$, $2\mathbb{Z}$, or 0).
• Key Observations from the Tables:
  • The classification exhibits an 8-fold periodicity for real classes and 2-fold for complex classes, shifted by the dimensionality parameters.
  • Different geometric configurations (e.g., two reflections vs. reflection and rotation) map to different effective 0D symmetry classes, leading to distinct topological invariants.
  • The tables allow for the immediate identification of higher-order topological phases (corner/hinge states) based on the codimension of the defect.

5. Significance

• Systematic Framework: This work provides a practical, algorithmic method for classifying topological phases in crystals with multiple point group symmetries, bypassing the need for complex spectral sequence calculations for every new material or symmetry combination.
• Higher-Order Topology: It directly facilitates the study of higher-order topological insulators and superconductors (HOTIs/HOTS), which are characterized by gapless states at boundaries of codimension $>1$ (corners, hinges). The classification of these phases relies heavily on the interplay of multiple crystalline symmetries.
• Material Discovery: By providing explicit tables for $\mathbb{Z}_2 \times \mathbb{Z}_2$ (a common symmetry group in many crystal systems, e.g., orthorhombic or tetragonal lattices with specific inversion/reflection combinations), the paper offers a direct tool for predicting and categorizing new topological materials.
• Theoretical Insight: The discovery that triple overlaps of symmetry actions do not affect the classification simplifies the theoretical landscape, suggesting that the topological structure is governed by pairwise interactions of symmetry generators.

In summary, Shiozaki's paper establishes a powerful, parameterized K-theoretic framework that reduces the complex problem of multi-symmetry topological classification to a manageable set of 0D data, culminating in comprehensive classification tables for the $\mathbb{Z}_2 \times \mathbb{Z}_2$ case.