Topologically protected vortex knots and links

The work addresses whether vortex knots or links can be tied so that decay via local reconnections and strand crossings is topologically prohibited by the properties of the medium. Prior observations in fluids and superfluids showed that even small dissipation enables reconnections and crossings that untie vortices, challenging robustness. The authors focus on non-Abelian topological vortices, where elements of the fundamental group do not commute, leading to constraints on crossings and to the formation of connecting vortex cords. By modeling vortex configurations as links in R^3 with order-parameter maps to spaces whose fundamental groups include the quaternion group Q8, they aim to define invariants and rules of evolution that establish and classify topologically protected linked structures relevant to systems such as spin-2 Bose–Einstein condensates and biaxial nematic liquid crystals. The study is significant because it provides a framework in which certain knotted and linked vortex structures are robust against the most common local core-topology changes observed experimentally.

The paper situates its contribution amid a broad literature on knots and links in physical systems, including liquid-crystal disclinations, vortices in classical fluids and superfluids, optical and acoustic fields, DNA, Skyrmion cores, and quantum knots in spinor condensates. Classical ideal-fluid helicity conserves knottedness, but experiments and simulations have shown spontaneous untying via reconnections in realistic settings. Prior topological analyses of vortex configurations considered isotopic evolution without allowing reconnections or strand crossings, limiting physical relevance. Foundational topological work on defect crossings in ordered media shows non-commuting vortices cannot freely cross and instead generate connecting cords. The authors also draw upon mathematical tools such as Wirtinger presentations, colored link diagrams, and link-homotopy classification (Milnor invariants) to build invariants applicable to colored links representing non-Abelian vortex configurations.

• Representation of vortex configurations: Model the medium as R^3 with vortex cores forming a link L, and an order parameter map Ψ: R^3\L → M. Use the induced homomorphism ψ: π1(R^3\L) → π1(M) to encode configurations as π1(M)-colored link diagrams via the Wirtinger presentation.
• Physical target spaces: Consider order-parameter spaces for spin-2 BEC phases (cyclic M_C and biaxial nematic M_BN) and biaxial nematic liquid crystals (M_BNLQ). These have trivial π2, and their π1 contain Q8 (exactly Q8 for M_BNLQ; as a subgroup for M_C and M_BN corresponding to vortices without scalar phase winding). Thus many configurations admit Q8-colored descriptions.
• Allowed core-topology changes: Define topologically allowed local reconnections and strand crossings as those that preserve coloring consistency. Strand crossings are allowed only for commuting elements in π1(M), equivalently when the undercrossing color does not change. Forbid (i) loop annihilation by shrinking to a point, (ii) crossings of non-commuting vortices (which would form energy-costly vortex cords), and (iii) vortex splitting along significant distances (energetically disfavored in targeted systems). Show that if long-distance splitting were allowed, any configuration could decay to an unknot via planarization and amalgamation (demonstrated with Q8-colored examples).
• Q8-colored diagram rules: Provide a graphical bicoloring scheme for strands corresponding to ±i (red), ±j (gray), ±k (blue), and a purple color for −1, with local rules dictating bicoloring flips under crossings and constraints on allowed reconnections/crossings.
• Invariants: Define a Z2-valued linking invariant I as the parity of crossings of one color over another (independent of which color pair is chosen). Define a refined Z4-valued Q-invariant by multiplying quaternion factors encountered at undercrossings of loops of a chosen color, corrected by self-writhe and pairwise linkings; prove independence from color and basepoint, and conservation under allowed reconnections and crossings. Provide formal proofs in Methods using α-invariants, Reidemeister moves adapted to colored diagrams, and case analyses.
• Classification procedure: Reduce any Q8-colored link via allowed operations to at most three non-purple components (ignoring a potential purple loop), then use Milnor’s link-homotopy classification (pairwise linking numbers μ(ij) and triple linking number μ(123) modulo gcd) together with bicoloring constraints and surgery operations that change μ(ij) by even integers. Enumerate resulting classes and compute Q-invariant values for representatives (looped chain of length three with odd pairwise linkings, Borromean rings with μ(ij)=0 and μ(123)=1, and trivial disjoint unions).

• Existence of topological protection: For non-Abelian vortex systems describable by Q8, there exist linked configurations that cannot be unlinked by any sequence of topologically allowed local reconnections and strand crossings. Protection is detected by conserved invariants.
• Invariants: The Z2 linking invariant I is conserved and indicates nontriviality when I=1. A stronger Z4-valued Q-invariant is defined and shown to be conserved; it reduces to I modulo 2. The Q-invariant is independent of color choice and basepoints.
• Classification: Up to allowed reconnections and strand crossings, Q8-colored links fall into a finite set of classes:
  • Six nontrivial classes: two for each Q ∈ {[1],[2],[3]} in Z4, with and without an additional purple loop; representatives include a Q8-colored looped chain of length three (Q=[1] or [3]) and a Q8-colored Borromean rings configuration (Q=[2]).
  • Sixteen trivial classes with Q=[0], corresponding to disjoint unions of any subset of the four color types (including the empty link).
• No protected knots for Q8: Systems with π1 given by Q8 do not support topologically protected single-loop knots under the allowed operations; only links (multi-component) are protected.
• Relation to Milnor invariants: For links with at most one component of each non-purple color and I=0, Q=[2 μ(123)] in Z4, connecting the Q-invariant to Milnor’s triple linking number.
• Limits of protection: If long-range vortex splitting is allowed, any Q8-colored link can be reduced to an unknot, showing that protection depends on forbidding such energetically costly processes.
• Broader example: A tricolored trefoil knot colored by S3 elements provides an example of a topologically protected knot in principle (outside Q8), though a corresponding physical realization was not identified here.

The results demonstrate that non-Abelian topology can endow vortex links with robustness against the dominant local core-topology changes—reconnections and strand crossings—observed in experiments. The invariants (particularly the Q-invariant) provide practical markers of protected configurations and yield a remarkably simple finite classification, in stark contrast to the infinite variety of ordinary link types under isotopy. This suggests that in realistic media with non-Abelian vortex groups (e.g., Q8), the relevant long-time topological states are few and characterizable. The work bridges mathematical link invariants and physical defect dynamics, offering a framework extendable to other groups and systems. It also clarifies the energetic and topological conditions under which protection holds, highlighting the role of suppressed non-commuting crossings and vortex splitting. The identification of potential systems (spin-2 BEC phases and biaxial nematic liquid crystals) underscores experimental relevance, and the S3 trefoil example points to the possibility of protected knots for other non-Abelian groups.

The paper introduces a framework for topologically protected vortex links in media with non-Abelian fundamental groups, formulating the dynamics via Q8-colored link diagrams and defining conserved invariants (linking parity and a refined Z4-valued Q-invariant). It delivers a finite classification of Q8-colored links up to allowed reconnections and strand crossings, identifies explicit protected linked configurations, and shows that single-loop knots are not protected for Q8, while protected knots can exist for other groups (e.g., S3). Future work includes numerical simulations to validate dynamical stability and protocols for preparation and observation in candidate systems, extension of the invariant and classification program to other physically relevant groups (e.g., tetrahedral group), and experimental realization in spin-2 Bose–Einstein condensates and nematic liquid crystals.

• Model assumptions restrict allowed core-topology changes to topologically consistent local reconnections and strand crossings of commuting elements; crossings of non-commuting vortices (which form vortex cords) and long-distance vortex splitting are excluded based on energetic arguments. Protection does not survive if significant vortex splitting is allowed.
• Vortex loop annihilation by shrinking is forbidden; although realistic in some systems (e.g., spin-1 polar BECs favor vortex loops over point defects), this may not hold universally.
• The analysis relies on systems with trivial π2 and on configurations describable within the Q8 subgroup (no scalar-phase winding), limiting generality.
• Physical verification of the assumptions and stability requires experiments or detailed simulations, which are left for future work.
• While an S3-colored protected trefoil is proposed conceptually, a corresponding physically accessible system is not identified, and a rigorous proof of its protection is deferred.