[Paper Review] Phonons and elasticity in critically coordinated lattices

T. C. Lubensky, C. L. Kane+3 more|arXiv (Cornell University)|Mar 4, 2015

Topological Materials and Phenomena81 references5 citations

TL;DR

This paper investigates vibrational modes and elasticity in critically coordinated lattices—specifically periodic square, kagome, and related lattices at the critical coordination number $ z_c = 2d $—using an index theorem relating zero modes and states of self-stress. It demonstrates that modifying the kagome lattice can induce topologically protected boundary modes, analogous to topological insulators, with robust zero modes at free surfaces and interfaces between topological phases.

ABSTRACT

Much of our understanding of vibrational excitations and elasticity is based upon analysis of frames consisting of sites connected by bonds occupied by central-force springs, the stability of which depends on the average number of neighbors per site $z$. When $z

Motivation & Objective

• To understand the mechanical stability and vibrational properties of lattices at the critical coordination number $ z_c = 2d $, relevant to jammed packings and network glasses.
• To extend Maxwell's rule for mechanical stability to periodic lattices and analyze the interplay between zero modes and states of self-stress.
• To explore how topological invariants in lattice structure lead to protected edge modes in finite systems with free boundaries.
• To establish a connection between topological classification in periodic lattices and the existence of robust surface zero modes.
• To develop a complex analysis framework using the argument principle to count zero modes in the Brillouin zone and at surfaces.

Proposed method

• Applies an index theorem $ N_0 - N_S = dN - N_B $ to relate the number of zero-energy modes ($ N_0 $) and states of self-stress ($ N_S $) to the number of sites ($ N $) and bonds ($ N_B $).
• Uses complex analysis and the argument principle $ \frac{1}{2\pi i} \oint \frac{d}{dz_y} \ln \det \mathbf{C}_{\text{Sym}} = m_y $ to count zeros of the determinant of the symmetric matrix $ \mathbf{C}_{\text{Sym}} $, corresponding to zero modes.
• Constructs surface unit cells via gauge transformations that shift phase factors $ e^{-i\mathbf{q} \cdot \mathbf{R}_L} $, eliminating poles and enabling direct counting of surface zero modes.
• Analyzes the determinant $ \det \mathbf{C}_{\text{Sur}} $ in terms of $ z = e^{iq/2} $, where the number of zeros within the unit circle corresponds to the number of surface modes.
• Demonstrates that topological invariants—encoded in the polarization charge and lattice vector shifts—determine the number of protected boundary modes.
• Compares non-topological and topological lattices by evaluating the topological index $ m_x $ or $ m_y $, which takes values $ -2 $ to $ 2 $, with $ m = 1 $ or $ 2 $ indicating topological protection.

Experimental results

Research questions

• RQ1 How do zero modes and states of self-stress relate in periodic lattices at the critical coordination number $ z_c = 2d $?
• RQ2 What conditions lead to the emergence of topologically protected zero modes at the boundaries of finite lattices?
• RQ3 How can the argument principle in complex analysis be used to count zero modes in the Brillouin zone and at surfaces?
• RQ4 What role do gauge transformations and lattice vector shifts play in identifying topological invariants in mechanical lattices?
• RQ5 How do modifications to the kagome lattice alter the topological classification and the number of robust boundary modes?

Key findings

• At $ z = z_c = 2d $, the index theorem $ N_0 - N_S = dN - N_B $ holds exactly, establishing a balance between zero modes and states of self-stress.
• For the kagome lattice, only trivial translational zero modes exist in the bulk, but topological modifications can generate protected boundary modes.
• The topological invariant $ m_y = 1 $ in a modified kagome lattice indicates two zeros of $ \det \mathbf{C}_{\text{Sym}} $ within the unit circle, corresponding to two protected surface modes.
• In the topological phase, the surface determinant $ \det \mathbf{C}_{\text{Sur}} $ becomes a polynomial with no poles, allowing direct counting of zero modes via the argument principle.
• The number of surface zero modes is determined by the topological index $ m_x $, which takes values $ m_x = 2 $ or $ m_x = 1 $ depending on the surface orientation and lattice structure.
• Transformations to surface unit cells via gauge shifts $ e^{-i\mathbf{q} \cdot \mathbf{R}_L} $ remove unphysical poles and ensure the argument principle correctly counts physical zero modes.