On the maximal degree of the Khovanov homology
Keiji Tagami

TL;DR
This paper demonstrates that the maximal homological degree of the Khovanov homology for a knot's cabling provides a lower bound for the knot's minimal positive crossing number, extending previous bounds.
Contribution
It introduces a new lower bound for the minimal positive crossing number based on the Khovanov homology of a knot's cabling, expanding the applicability of homological invariants.
Findings
Maximal homological degree of Khovanov homology bounds crossing number
Cabling preserves the lower bound property
Extends previous bounds to cabling knots
Abstract
It is known that the maximal homological degree of the Khovanov homology of a knot gives a lower bound of the minimal positive crossing number of the knot. In this paper, we show that the maximal homological degree of the Khovanov homology of a cabling of a knot gives a lower bound of the minimal positive crossing number of the knot.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsGeometric and Algebraic Topology
On the maximal degree of the Khovanov homology
Keiji Tagami
Department of Mathematics, Faculty of Science and Technology, Tokyo University of Science, Noda, Chiba, 278-8510, Japan
Abstract.
It is known that the maximal homological degree of the Khovanov homology of a knot gives a lower bound of the minimal positive crossing number of the knot. In this paper, we show that the maximal homological degree of the Khovanov homology of a cabling of a knot gives a lower bound of the minimal positive crossing number of the knot.
1. Introduction
In [4], for each oriented link , Khovanov defined a graded chain complex whose graded Euler characteristic is equal to the Jones polynomial of . Its homology is a link invariant and called the Khovanov homology. Throughout this paper, we denote the homological degree term of the rational Khovanov homology of a link by .
Let be an oriented link. By , we denote the maximal homological degree of the Khovanov homology of , and by , we denote the minimal number of the positive crossings of diagrams of . Note that is not negative. In fact, any link has nonzero Khovanov homology in degree zero because the Lee homology [6] is not trivial in homological degree zero. Then, it is known that gives a lower bound of (Proposition 2.1). From this fact, it seems that the Khovanov homology estimates the positivity of links.
Stošić [9] showed that is , where is the positive -torus link. By using the same method as Stošić, the author [10] proved that is . The author also computed the maximal degree for a cabling of any knot.
In this paper, we give some properties of the maximal degree of the Khovanov homology. In particular, we show that the maximal homological degree of the Khovanov homology of a cabling of a knot gives a lower bound of the minimal positive crossing number of the knot as follows:
Theorem 1.1**.**
Let be an oriented knot. Denote the -cabling of by . For any positive integers and , we assume that each component of has the orientation induced by , that is, each component of is homologous to in the tubular neighborhood of . Then, if , we have the following inequality:
[TABLE]
In particular, we obtain
[TABLE]
In many cases, is not greater than the minimal positive crossing number of -cabling of . Hence the inequality of Theorem 1.1 is possibly stronger than that of Proposition 2.1.
Note that there are some works on the crossing numbers of cable links (for example, see [3], [5, Problem 1.68] and [8]).
This paper is organized as follows: In Section 2, we give the proof of Theorem 1.1. In Appendix, we introduce some properties of the maximal degree of the Khovanov homology.
We refer some informations (knot names, values of knot invariants and so on) in [7]. Throughout this paper, we use the same definition and notation of the Khovanov homology as in [10, p.2848-p.2850]. In particular, for a link diagram of a link , we denote the unnormalized Khovanov homology by , that is,
,
where is the number of the negative crossings of .
2. The positivity of knots and the maximal degree of the Khovanov homology
In this section, we give some estimates on the minimal positive crossing numbers of knots. In particular, we prove Theorem 1.1.
For any oriented link , define
[TABLE]
where () is the number of the positive (negative) crossings of . A link is positive (negative) if (). The following is an immediate consequence of the definition of the Khovanov homology.
Proposition 2.1** (cf. [12, Proposition 2.2]).**
For any oriented link , we have and .
Unfortunately, Proposition 2.1 is not sufficient to determine whether a given link is negative (or positive) (for example, see Example 2.3). In [10], the author computed for a cable link. By the computation, we obtain new estimates of the minimal positive crossing numbers of knots (Theorem 1.1).
Proof of Theorem 1.1.
Let be a diagram of with positive crossings. Let be the number of the negative crossings of .
Suppose that . By [10, Lemma ] (with and ), we have
for and . 111In [10, Lemma ], we put and . In our setting, . Here, the diagram is introduced in [10, Definitions 3.1, 3.9 and 4.1, and Figures and ]. Moreover is equal to , where is a diagram of introduced in [10, Definition 4.1 and Figures and 9]. Since the diagram has negative crossings, by putting , we obtain
[TABLE]
for and . In particular, we have . By using the negative first Reidemeister move repeatedly, we can take as large a as we want. Hence, for , we have
[TABLE]
Suppose that . By [10, Lemma ] and the same discussion as above, we obtain
[TABLE]
for . ∎
Corollary 2.2**.**
If is a negative knot, then is zero for any positive integer .
Example 2.3**.**
We see that (see [7]). Hence, the inequality in Proposition 2.1 cannot determine whether these knots are negative. However, by “The Mathematica Package KnotTheory” [2], we have
[TABLE]
By Theorem 1.1, these knots are not negative.
Remark 2.4**.**
Let be an oriented knot. Then, if , and satisfy one of the following conditions (see also Figure 1):
- (1)
and , 2. (2)
for some , and , 3. (3)
for some , and .
Condition (1) follows from Theorem 1.1, and (2) and (3) follow from [10, Lemma ].
Question 2.5**.**
*For any non-negative knot , are there some and such that ? *
Question 2.6**.**
For any knot , does the following hold?
[TABLE]
Note that the last inequality holds by Theorem 1.1.
Appendix A Appendix: Other properties of the Khovanov homology
In this section, we introduce other properties of the maximal degree of the Khovanov homology.
A.1. versus
By the following result, the accuracy of the inequality in Proposition 2.1 does not depend on the choice of orientations of links.
Proposition A.1.1**.**
For any oriented link , does not depend on the orientation of .
Proof.
Let be a component of . Let be the link obtained from by reversing the orientation of . Khovanov [4, Proposition ] proved that
[TABLE]
Hence we have . On the other hand, we have . These imply this proposition. ∎
Example A.1.2**.**
Let be the positive -torus link. For positive integers and , we have . Indeed, the standard diagram of has positive crossings. Moreover, each 2-component sublink of is and has at least positive crossings. Since the link has 2-component sublinks, we obtain . On the other hand, Stošić [9, Theorem ] showed that . Hence, we have .
Let be the link obtained from by reversing the orientations of exactly components. Similarly, we have and . Hence we have .
Corollary A.1.3**.**
For any positive integer , there exists some oriented link such that .
Proof.
By Example A.1.2, we have . ∎
A.2. Additivity of
We prove the additivity of the maximal degree of the Khovanov homology.
Proposition A.2.1**.**
For any oriented knots and , we have
[TABLE]
where is the disjoint union of the knots and is the connected sum of the knots.
Proof.
By [4, Proposition ], we obtain
[TABLE]
Note that our coefficient ring is the rational number field . Hence we have .
In [4, Proposition ], the following exact sequence was introduced:
[TABLE]
Here, we forget the quantum grading of the Khovanov homology. By the first line, we have if . Hence we obtain .
Put . Then . By the second and third lines, we have
[TABLE]
By repeating this process, for any , we obtain
[TABLE]
Since for sufficiently large , we have
[TABLE]
for any . This implies that . Hence we obtain . ∎
A.3. Almost positive knots
A link diagram is almost positive (negative) if it has exactly one negative (positive) crossing. A link is almost positive (negative) if it is not a positive (negative) knot and has an almost positive (negative) diagram. Then we have the following.
Corollary A.3.1**.**
For any almost negative link , we have .
Proof.
By Proposition 2.1, we have or . For contradiction, assume that . Then, any almost negative diagram of satisfies . In [4, Proposition ], Khovanov proved that if and only if the diagram is adequate (for the definition of adequate diagrams, for example, see [4, Definition ]). However, any reduced almost negative link diagram is not adequate. Hence, . ∎
Remark A.3.2**.**
The Khovanov homology and the Rasmussen invariant for an almost positive link are studied in [1] and [11].
**Acknowledgements: ** The author would like to thank the referees for their helpful comments.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[1] T. Abe and K. Tagami, Characterization of positive links and the s 𝑠 s -invariant for links , to appear in Canad. J. Math.
- 2[2] D. Bar-Natan, The Knot Atlas , http://www.math.toronto.edu/drorbn/K Atlas/.
- 3[3] M. H. Freedman and Z. He, Divergence-free fields: energy and asymptotic crossing number , Ann. of Math. (2) 134 (1991), no. 1, 189–229. MR 1114611 (93a:58040)
- 4[4] M. Khovanov, A categorification of the Jones polynomial , Duke Math. J. 101 (2000), no. 3, 359–426. MR 1740682 (2002 j:57025)
- 5[5] R. Kirby, Problems in low-dimensional topology , Geometric topology (Athens, GA, 1993), AMS/IP Stud. Adv. Math., vol. 2, Amer. Math. Soc., Providence, RI, 1997, pp. 35–473. MR 1470751
- 6[6] E. S. Lee, An endomorphism of the Khovanov invariant , Adv. Math. 197 (2005), no. 2, 554–586. MR 2173845 (2006 g:57024)
- 7[7] C. Livingston and J. Cha, Knot Info , http://www.indiana.edu/%7eknotinfo/.
- 8[8] A. Stoimenow, On the satellite crossing number conjecture , J. Topol. Anal. 3 (2011), no. 2, 109–143. MR 2819190 (2012 g:57017)
