Explicit Formulas for the Alexander Polynomial of Pretzel Knots

This paper presents explicit formulas for the Alexander polynomial of pretzel knots, characterizes those with trivial polynomials, and utilizes these results to construct a new family of knots that are topologically slice but not smoothly slice.

The Gist

Imagine you are a detective trying to solve a mystery about knots. But not just any knots—these are mathematical knots made of twisted rubber bands. In the world of math, these are called Pretzel Knots.

Why "Pretzel"? Because if you take a few strips of dough, twist them, and connect the ends, they look exactly like a pretzel. The "twists" in the dough are the numbers $q_1, q_2, \dots$ in the paper. Some twists go left, some go right, and some are just straight.

For a long time, mathematicians knew how to describe these knots, but they didn't have a simple "recipe" (a formula) to calculate a specific fingerprint for every single type of pretzel knot. This fingerprint is called the Alexander Polynomial. Think of the Alexander Polynomial as a unique barcode or a DNA sequence for a knot. If two knots have different barcodes, they are definitely different knots. If they have the same barcode, they might be the same, but it's a strong hint.

Here is what Yury Belousov did in this paper, explained in plain English:

1. The Missing Recipe

Before this paper, if you wanted to find the barcode for a complex pretzel knot, you had to do a lot of complicated, messy math, or you were stuck if the knot didn't fit a special category. It was like trying to bake a cake without a recipe, guessing the ingredients every time.

Belousov wrote down the exact recipe (Theorem 1). He gave us clear formulas for three different types of pretzel knots:

• Type A: The knot has an even number of twists in one spot and odd numbers everywhere else.
• Type B: The knot has an even number of twists in one spot, but the total number of twist sections is even.
• Type C: Every single twist section is an odd number.

Now, anyone can plug in the numbers of twists and instantly get the knot's barcode.

2. The "Invisible" Knots (Trivial Alexander Polynomial)

Some knots are so special that their barcode looks like "1" (or is essentially empty). In math, we call this a "trivial" polynomial.

• The Mystery: If a knot has a trivial barcode, it usually means the knot is "topologically slice."
• The Analogy: Imagine a knot tied in a piece of string. If you could push that string through a fourth dimension (like a ghost passing through a wall) and untie it without cutting the string, it is "topologically slice."
• The Discovery: Belousov used his new formulas to find exactly which pretzel knots have this "invisible" barcode. He found a specific set of rules (a system of equations) that these knots must follow.

3. The "Smooth vs. Rough" Twist

Here is where it gets really cool. The paper finds a new family of knots that are Topologically Slice but Not Smoothly Slice.

Let's use a metaphor:

• Topologically Slice: Imagine you have a knotted rubber band. If you are allowed to stretch it, squish it, and pass it through itself like a ghost (ignoring the fact that rubber bands usually can't pass through each other), you can untie it.
• Smoothly Slice: Now, imagine you are a strict sculptor. You can stretch and squish the rubber band, but you cannot pass it through itself. It has to move smoothly.

Belousov found knots that are "ghost-untieable" (Topologically Slice) but "sculptor-untieable" (Not Smoothly Slice). This is a huge deal in math because it proves that the "ghost" world and the "real" world of knots behave differently. It's like finding a door that opens if you are a ghost, but is locked tight if you are a human.

4. The Hunt for the "Perfect" Numbers

To find these special knots, Belousov had to solve a giant puzzle. He needed to find a list of odd numbers (the twists) that fit a very specific equation.

• For knots with 3 twists, we already knew the answer.
• For knots with 5 twists, he used a computer to search through millions of combinations. He found 38 special groups of numbers that create these "ghost-untieable" knots.
• For knots with 7 twists, he searched through billions of combinations and found nothing.

This led him to a guess (Conjecture): There are infinitely many of these special 5-twist knots, but if you try to make them with 7, 9, or more twists, they simply don't exist.

Why Should You Care?

You might think, "Who cares about math knots?"
Well, this paper is like finding a new key.

1. It solves a 90-year-old puzzle: It gives a complete list of how to calculate the "DNA" of these knots.
2. It reveals a hidden world: It proves that there are shapes in our universe that behave one way if you ignore the rules of physics (smoothness) and another way if you follow them.
3. It creates new tools: Mathematicians can now use these formulas to build even more complex knots and study the fabric of space and time.

In short, Belousov took a messy, confusing pile of twisted rubber bands, wrote down the instructions to understand them perfectly, and found a hidden door that only opens for "ghosts."

Technical Summary

Here is a detailed technical summary of the paper "Explicit Formulas for the Alexander Polynomial of Pretzel Knots" by Yury Belousov.

1. Problem Statement

The paper addresses a gap in knot theory literature regarding the Alexander polynomial ($\Delta_K(t)$) of pretzel knots.

• Context: A pretzel link $P(q_1, q_2, \dots, q_n)$ consists of $n$ twisted bands with $q_i$ half-twists. While specific cases (e.g., those with at least one even parameter or specific odd configurations) had been studied, no general explicit formula existed for the Alexander polynomial of arbitrary pretzel knots in the literature.
• Goal: To derive closed-form expressions for $\Delta_K(t)$ for all valid pretzel knots and utilize these formulas to characterize knots with trivial Alexander polynomials (which are topologically slice).

2. Methodology

The author employs a combination of inductive reasoning, skein relations, and combinatorial algebra.

• Skein Relation: The core of the proof relies on the symmetrized Alexander polynomial's skein relation:
  $$\Delta_{L_+}(t) - \Delta_{L_-}(t) = (t^{1/2} - t^{-1/2})\Delta_{L_0}(t)$$
  where $L_+, L_-, L_0$ differ by a local crossing change.
• Induction Strategy:
  • The proof proceeds by induction on the twist parameters $q_i$.
  • Base Cases: The author utilizes known results for torus links (specifically $(2, m)$-torus links) and connected sums of torus knots.
  • Recursive Step: By applying the skein relation to a pretzel knot $P(q_1 \pm 2, q_2, \dots, q_n)$, the author relates the polynomial of the modified knot to the polynomials of knots with fewer or simpler parameters.
• Symmetric Polynomials: The formulas are expressed using elementary symmetric polynomials ($\sigma_k$) of the twist parameters $q_i$. This allows for a compact representation of the coefficients of the Alexander polynomial.
• Computational Search: For the application section, the author uses a C++ program to search for integer solutions to specific systems of equations derived from the condition $\Delta_K(t) \doteq 1$.

3. Key Contributions: Explicit Formulas (Theorem 1)

The paper provides three distinct formulas based on the parity of the number of strands ($n$) and the twist parameters ($q_i$). Let $K = P(q_1, \dots, q_n)$.

Case A: $n$ is odd, one $q_i$ is even (assume $q_1$ even)

$$\Delta_K(t) \doteq \left( \prod_{i=2}^n \frac{1+tq_i}{1+t} \right) \left( 1 + \frac{q_1}{2}(t^{-1}-t) \left( \sum_{i=2}^n \frac{t^{q_i}}{1+tq_i} - \frac{n-1}{2} \right) \right)$$

Case B: $n$ is even, one $q_i$ is even (assume $q_1$ even)

$$\Delta_K(t) \doteq \left( \prod_{i=2}^n \frac{1+tq_i}{1+t} \right) \left( t^{q_1/2} + (t^{q_1/2} - t^{-q_1/2}) \left( \sum_{i=2}^n \frac{t^{q_i}}{1+tq_i} - \frac{n}{2} \right) \right)$$

Case C: $n$ is odd and all $q_i$ are odd

$$\Delta_K(t) \doteq \frac{1}{2^{n-1}} \sum_{k=0}^{(n-1)/2} (t-1)^{2k}(t+1)^{n-1-2k} \sigma_{2k}(q_1, \dots, q_n)$$
Note: The symbol $\doteq$ indicates equality up to multiplication by $\pm t^k$.

Additionally, the paper provides a Lemma for two-component pretzel links (where $n$ is even and all $q_i$ are odd), distinguishing between co-directed and oppositely directed strands.

4. Key Results and Corollaries

A. Knot Determinant

The author derives a unified formula for the determinant of any pretzel knot $K$:
$$\det(K) = |\sigma_{n-1}(q_1, \dots, q_n)| = \left| q_1 q_2 \dots q_n \left( \sum_{i=1}^n \frac{1}{q_i} \right) \right|$$
This reproduces a result previously obtained by Kolay but is derived here directly from the new Alexander polynomial formulas.

B. Degree and Triviality

• Degree: For pretzel knots with all odd parameters, the degree of the Alexander polynomial is $n-1$ if and only if $\sum \sigma_{2k} \neq 0$.
• Trivial Alexander Polynomial: The paper characterizes when $\Delta_K(t) \doteq 1$. This occurs if and only if the symmetric polynomials satisfy a specific system of linear equations:
  $$\sigma_{2k}(q_1, \dots, q_n) = (-1)^k \binom{(n-1)/2}{k}$$
  for $0 \le k \le (n-1)/2$.

C. Topologically Slice but Not Smoothly Slice Knots

• Freedman's Theorem: Any knot with a trivial Alexander polynomial is topologically slice.
• Smooth Slice Obstruction: The author identifies "irreducible" solutions (where $|q_i| > 1$) to the triviality system for $n=5$.
  • A computational search found 38 irreducible solutions for $n=5$.
  • Citing Brylinski [Bry17], the author notes these knots are not smoothly slice.
  • Result: This constructs a new infinite family of knots that are topologically slice but not smoothly slice, distinguishing the smooth and topological categories of knot concordance.
• Conjecture: The author conjectures that while there are infinitely many irreducible solutions for $n=5$, there are no irreducible solutions for $n > 5$. (Searches for $n=7$ yielded no solutions).

5. Significance

1. Completeness: The paper fills a significant gap in the literature by providing the first general explicit formulas for the Alexander polynomial of pretzel knots across all parity cases.
2. Computational Utility: The formulas allow for the direct calculation of invariants (like the determinant and degree) without needing to perform complex diagrammatic reductions.
3. Concordance Theory: The discovery of new families of topologically slice knots that are not smoothly slice advances the understanding of the difference between topological and smooth 4-manifold structures.
4. Methodological Clarity: The proof demonstrates a robust application of the skein relation combined with symmetric polynomial identities, offering a template for analyzing other families of links.

In summary, Belousov's work provides a definitive algebraic characterization of pretzel knot invariants and leverages this characterization to generate new examples in low-dimensional topology.