Generalized Alexander (Sawollek) Polynomial
- The generalized Alexander polynomial is a two-variable invariant for knots, extending classical invariants via robust algebraic and skein-theoretic frameworks.
- It is computed using matrix determinants from Alexander biquandle modules and bosonic oscillator constructions, often refined with Gröbner bases.
- The invariant unifies the classical, HOMFLY, and Jones polynomials while effectively detecting nonclassicality and supporting studies in knot concordance and quantum topology.
A generalized Alexander polynomial—often termed the Sawollek polynomial in the context of virtual knots—extends the classical Alexander polynomial by enhancing both its algebraic and skein-theoretic structure, leading to a two-parameter invariant with applications to virtual knot theory, quantum invariants, and concordance obstructions. The invariant interrelates with the HOMFLY and Jones polynomials and provides a unifying framework for extracting and studying a range of combinatorial and quantum knot invariants (Pavlyuk, 2015, Mellor, 2016, Crans et al., 2011, Boden et al., 2014, Kauffman, 2020, Boden et al., 2019).
1. Foundational Definitions and Algebraic Structure
The two-variable generalized Alexander polynomial is typically denoted or for a knot or link . It is most naturally described by either a bosonic oscillator construction (Pavlyuk, 2015), Alexander biquandle module presentations (Crans et al., 2011, Kauffman, 2020), or via a determinant formula involving explicit matrices associated to the knot diagram (Mellor, 2016, Boden et al., 2019). For virtual knots, these presentations are equivalent up to overall normalization and change of variables.
The key algebraic starting points are:
- -numbers: , satisfying the recurrence .
- Alexander Biquandle Module: Given an oriented diagram, arcs are labeled and linear relations in a free module over (or ) are imposed at each classical crossing, encapsulating the knot algebraically (Crans et al., 2011, Kauffman, 2020).
The generalized Alexander polynomial is then defined as the determinant of a (square) presentation matrix obtained from these relations, up to multiplication by a unit:
For virtual knot diagrams, a concise formulation involves associating to each crossing a 0 block matrix depending on the sign, assembling these into a block-diagonal 1 matrix 2, and subtracting a permutation matrix 3 encoding the diagram's arc sequence. The invariant is then:
4
with normalization adjustments according to monomial factors (Mellor, 2016, Boden et al., 2019).
2. Skein Relations and Specializations
The generalized Alexander polynomial admits a two-variable skein relation that recovers, as specializations, the classical Alexander polynomial, the HOMFLY polynomial, and the Jones polynomial (Pavlyuk, 2015):
- Skein Relation: 5, where 6 are the generalized Alexander polynomials of three diagrams differing at a crossing.
- Reduction to Classical Invariants: With 7, one recovers the standard Alexander skein relation. The change of variables 8 translates the generalized Alexander skein relation to that of the HOMFLY polynomial; specializing 9 yields the Jones polynomial (Pavlyuk, 2015).
The relation to the Sawollek polynomial 0 is effected by invertible substitution 1, so that 2, and conversely 3 (Pavlyuk, 2015).
In the virtual knot context, the Sawollek polynomial 4 is characterized by its skein and divisibility by 5, as well as its vanishing on all classical knots—a diagnostic for nonclassicality (Mellor, 2016, Boden et al., 2019).
3. Computation: Presentation Matrices and Gröbner Bases
The computation of the generalized Alexander/Sawollek polynomial is explicit and algorithmic, rooted in linear algebra over polynomial rings.
- Alexander Biquandle Matrix: The presentation matrix 6 reflects the module relations at crossings (as specified above). Elementary ideals (generated by minors of 7) encapsulate additional information; the zeroth ideal determines the principal polynomial, while higher ideals and their Gröbner bases refine the invariant (Crans et al., 2011).
- Examples:
- Virtual trefoil: 8 (Kauffman, 2020).
- Figure-eight: explicit expansion via skein and combining smaller diagrams yields a symmetric and normalized polynomial in 9 and 0 (Pavlyuk, 2015).
Gröbner basis techniques allow the extraction of invariants beyond principal generators, capturing richer structure such as distinguishing between knots with trivial principal polynomials (Crans et al., 2011).
4. Structural Properties and Symmetries
The generalized Alexander polynomial displays a constellation of structural features and symmetries:
- Normalization: 1.
- Symmetry: 2.
- Mirror: 3.
- Connected Sum: 4.
- Degree Bounds: For an 5-crossing diagram, deg6, etc.
- Divisibility: For virtual knots, 7. For classical knots, the polynomial vanishes identically at the lowest order, reflecting the classical Alexander polynomial's degree shift (Mellor, 2016, Boden et al., 2019).
- Functoriality under Concordance: The polynomial vanishes for knots virtually concordant to homologically trivial knots or virtual slices (Boden et al., 2019).
A table summarizing core symmetries:
| Operation | Formula | Effect |
|---|---|---|
| Mirror | 8 | Parameter inversion |
| Connected Sum | 9 | Multiplicative |
| Specialization | 0, 1 | Recovers classical Alexander |
| Sawollek conversion | 2 | Change of variable |
The invariant is well-defined (up to monomial multiples) under Reidemeister moves (both classical and virtual) (Mellor, 2016, Boden et al., 2019).
5. Relation to Affine Index and Writhe-Type Invariants
The generalized Alexander polynomial for virtual knots encapsulates all index-type (writhe) invariants and can be specialized to recover Kauffman's affine index polynomial (Mellor, 2016, Kauffman, 2020). Specifically, expanding 3 as a series in 4, the first-order term 5 encodes the affine index polynomial, and higher-order terms yield more refined virtual knot invariants (e.g., second-order writhe polynomials) (Mellor, 2016). The underlying divisibility relations and the explicit extraction algorithm have been laid out by Mellor and others.
Furthermore, every term beyond classical Alexander vanishes on classical knots, making the polynomial an effective detector of nonclassicality and providing lower bounds for the virtual crossing number.
6. Applications, Examples, and Concordance
The invariant is instrumental for:
- Virtual Knot Concordance: It detects whether a knot is virtually slice or concordant to homologically trivial representatives—the polynomial vanishes in such scenarios (Boden et al., 2019).
- Distinction of Virtual Connected Sums: Example computations show that pairs of virtual connected sums with equivalent classical Alexander polynomials can be distinguished by their generalized Alexander polynomials (Crans et al., 2011).
- Detection of Nonclassicality: Non-triviality of the polynomial certifies nonclassicality and non-sliceness.
Sample computations for small virtual knots (e.g., 6, 7, 8) reveal factorization patterns and directly connect to topological properties of the knots in question (Boden et al., 2019).
7. Interactions with Quantum Invariants and Open Problems
The 9-number formalism connects the generalized Alexander polynomial to two-parameter quantum algebras and quantized oscillator models (Pavlyuk, 2015). Under suitable reparametrizations, the invariant interpolates among the Alexander, HOMFLY, and Jones polynomials, providing a unified polynomial framework for quantum knot invariants.
Open directions include:
- Twisted Alexander polynomials via representations of virtual knot groups (Boden et al., 2014).
- Categorification attempts and connections to Khovanov homology (Crans et al., 2011).
- Determination of precise conditions for when virtual connected sums factor via the polynomial (Crans et al., 2011).
- Analysis and extraction of further Vassiliev-type invariants from higher minors or expansion terms (Mellor, 2016).
Research into the generalized Alexander (Sawollek) polynomial leverages explicit algebraic constructions, provides deep connections across classical and virtual knot invariants, and offers both computational and structural insights into low-dimensional and quantum topology (Pavlyuk, 2015, Mellor, 2016, Crans et al., 2011, Boden et al., 2014, Kauffman, 2020, Boden et al., 2019).