Next Article in Journal
Antecedents of Symmetry in Physicians’ Prescription Behavior: Evidence from SEM-Based Multivariate Approach
Next Article in Special Issue
Facility Location Problem Approach for Distributed Drones
Previous Article in Journal
F(R,G) Cosmology through Noether Symmetry Approach
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Khovanov Homology of Three-Strand Braid Links

1
Department of Mathematics, Dong-A University, Busan 49315, Korea
2
Faculty of Information Technology, University of Central Punjab, Lahore 54000, Pakistan
3
Department of Mathematics, Division of Science and Technology, University of Education, Lahore 54000, Pakistan
4
Department of Mathematics, University of Gujrat, Gujrat 50700, Pakistan
5
Department of Mathematics and RINS, Gyeongsang National University, Jinju 52828, Korea
6
Center for General Education, China Medical University, Taichung 40402, Taiwan
*
Authors to whom correspondence should be addressed.
Symmetry 2018, 10(12), 720; https://doi.org/10.3390/sym10120720
Submission received: 7 November 2018 / Revised: 14 November 2018 / Accepted: 15 November 2018 / Published: 5 December 2018
(This article belongs to the Special Issue Symmetry in Applied Mathematics)

Abstract

:
Khovanov homology is a categorication of the Jones polynomial. It consists of graded chain complexes which, up to chain homotopy, are link invariants, and whose graded Euler characteristic is equal to the Jones polynomial of the link. In this article we give some Khovanov homology groups of 3-strand braid links Δ 2 k + 1 = x 1 2 k + 2 x 2 x 1 2 x 2 2 x 1 2 x 2 2 x 1 2 x 1 2 , Δ 2 k + 1 x 2 , and Δ 2 k + 1 x 1 , where Δ is the Garside element x 1 x 2 x 1 , and which are three out of all six classes of the general braid x 1 x 2 x 1 x 2 with n factors.
MSC:
57M27; 55N20

1. Introduction

Khovanov homology was introduced by Mikhail Khovanov in 2000 in Reference [1] as a categorification of the Jones polynomial, which was introduced by Jones in [2]. His construction, using geometrical and topological objects instead of polynomials, was so interesting that it offered a completely new approach to tackle problems in low-dimensional topology.
Khovanov homology plays a vital role in developing several important results in the field of knot theory. Soon after the discovery of Khovanov homology, Bar-Natan proved in Reference [3] that Khovanov’s invariant is stronger than the Jones polynomial. He also proved that the graded Euler characteristic of the chain complex of a link L is the un-normalized Jones polynomial of that link. In 2005, Bar-Natan extended the Khovanov homology of links to tangles, cobordisms, and two-knots [4]. In [5] Bar-Natan gave a fast way of computing the Khovanov homology. In 2013, Ozsvath, Rasmussen, and Szabo introduced the odd Khovanov homology by using exterior algebra instead of symmetric algebra [6]. Gorsky, Oblomkov, and Rasmussen gave some results on stable Khovanov homology of torus links in Reference [7]. Putyra introduced a triply graded Khovanov homology and used it to prove that odd Khovanov homology is multiplicative with respect to disjoint unions and connected sums of links Reference [8]. Manion gave rational Khovanov homology of three-strand pretzel links in 2011 [9]. Nizami, Mobeen, and Ammara gave Khovanov homology of some families of braid links in Reference [10]. Nizami, Mobeen, Sohail, and Usman gave Khovanov homology and graded Euler characteristic of 2-strand braid links in [11].
In Reference [12], Marko used a long exact sequence to prove that the Khovanov homology groups of the torus link T ( n ; m ) stabilize as m . A generalization of this result to the context of tangles came in the form of Reference [13], where Lev Rozansky showed that the Khovanov chain complexes for torus braids also stabilize (up to chain homotopy) in a suitable sense to categorify the Jones–Wenzl projectors. At roughly the same time, Benjamin Cooper and Slava Krushkal gave an alternative construction for the categorified projectors in Reference [14]. These results, along with connections between Khovanov homology, HOMFLYPT homology, Khovanov–Rozansky homology, and the representation theory of rational Cherednik algebra (see [15]) have led to conjectures about the structure of stable Khovanov homology groups in limit K h ( T ( n ; 1 ) ) (see [15], and results along these lines in Reference [16]). More recently, in Reference [17], Robert Lipshitz and Sucharit Sarkar introduced the Khovanov homotopy type of a link L. This is a link invariant taking the form of a spectrum whose reduced cohomology is the Khovanov homology of L.
Although computing the Khovanov homology of links is common in the literature, no general formulae have been given for all families of knots and links. In this paper, we give Khovanov homology of the three-strand braid links Δ 2 k + 1 , Δ 2 k + 1 x 2 , and Δ 2 k + 1 x 1 , where Δ is the Garside element x 1 x 2 x 1 . Particularly, we focus on the top homology groups.

2. Braid Links

Definition 1.
A knot is a simple, closed curve in the three-space. More precisely, it is the image of an injective, smooth function from the unit circle to R 3 with a nonvanishing derivative [18]. You can see some knots in Figure 1:
Definition 2.
An m-component link is a collection of m nonintersecting knots [18]. A trivial two-component link and the Hopf link are given in Figure 2:
Definition 3.
Two links L 1 and L 2 are said to be isotopic or equivalent if there is a smooth map F: [ 0 , 1 ] × S 1 R 3 , which confirms that F t is a link for all t [ 0 , 1 ] and that that F 0 = L 1 and F 1 = L 2 . Map F is called isotopy. By the isotopy class of a link L, denoted [ L ] , we mean the collection of all links that are isotopic to L.
Since it is hard to work with links in R 3 , people usually prefer working with their projections on a plane. These projections should be generic, which means that all multiple points are double points with a clear information of over- and undercrossing, as you can see in Figure 3. Such a projection of a link is called the diagram of the link.
Theorem 1. (Reidemeister, [19]).
Let D 1 and D 2 be two diagrams of links L 1 and L 2 . Then, links L 1 and L 2 are isotopic if and only if D 1 is transformed into D 2 by planar isotopies and by a finite sequence of three local moves represented in Figure 4:
Definition 4.
A link invariant is a function that remains constant on all elements in an isotopy class of a link.
Remark 1.
A function to qualify as a link invariant should be invariant under the Reidemeister moves.
Definition 5.
An n-strand braid is a collection of n nonintersecting, smooth curves joining n points on a plane to n points on another parallel plane in an arbitrary order such that any plane parallel to the given planes intersects exactly n number of curves [20]. The smooth curves are called the strands of the braid. You can see a 2-strand braid in Figure 5:
Definition 6.
The product of two n-strand braids α and β, denoted by α β , is defined by putting β below α and then gluing their common endpoints.
Definition 7.
A braid is said to be elementary if it consists of just one crossing. The ith elementary braid, denoted by x i , is given in Figure 6:
Remark 2.
Each braid is a product of elementary braids.
Definition 8.
The closure of a braid β, denoted by β ^ , is defined by connecting its lower endpoints to its corresponding upper endpoints with smooth curves, as you can see in Figure 7.
Remark 3.
1 
All braids are oriented from top to bottom.
2 
From now onward, by braid β we mean its closure β ^ , which is actually a link.
An important result by Alexander, connecting links and braids, is:
Theorem 2. (Alexander [21]).
Each link is a closure of some braid.
Definition 9.
The 0- and 1-smoothings of crossing Symmetry 10 00720 i001 are defined, respectively, by Symmetry 10 00720 i002 and Symmetry 10 00720 i003.
Definition 10.
A collection of disjoint circles obtained by smoothing out all the crossings of a link L is called the Kauffman state of the link [22].

3. Homology

Definition 11.
Let V = n V n , be a graded vector space with homogeneous components { V n } of degree n. The graded dimension of V is the power series q dim V : = n q n dim V n .
Definition 12.
The degree of the tensor product of graded vector space V 1 V 2 is the sum of the degrees of the homogeneous components of graded vector spaces V 1 and V 2 .
Remark 4.
In our case, the graded vector space V has the basis < v + , v > with degree p ( v ± ) = ± 1 and the q-dimension q + q 1 .
Definition 13.
The degree shift . { l } operation on a graded vector space V = V n is defined by
V . { l } n = V n l .
Construction of Chain Groups: Let L be a link with n crossings, and let all crossings be labeled from 1 to n. Arrange all its 2 n Kauffman states into columns 1 , 2 , , n so that the rth column contains all states having r number of 1-smoothings in it. To every stat α in the rth column we assign graded vector space V α ( L ) : = V m { r } , where m is the number of circles in α . The rth chain group, denoted by [ [ L ] ] r : = α : r = | α | V α ( L ) , is the direct sum of all vector spaces corresponding to all states in the rth column.
Definition 14.
The chain complex C ¯ of graded vector spaces C r ¯ is defined as:
C ¯ r + 1 d r + 1 C ¯ r d r C ¯ r 1 d r 1
such that d r d r + 1 = 0 for each r.
In a system of converting the chain group into a complex, we use the maps between graded vector spaces to satisfy d d . For this purpose we can label the edges of the cube { 0 , 1 } χ by the sequence ξ ϵ { 0 , 1 , } χ , where ξ contains only one ⋆ at a time. Here, ⋆ indicates that we change a 1-smoothing to a 0-smoothing. The maps on the edges is denoted by d ξ , the height of edges | ξ | . The direct sum of differentials in the cube along the column is
d r : = | ξ | = r ( 1 ) ξ d ξ .
Now, we discuss the reason behind the sign of ( 1 ) ξ . As we want from the differentials to satisfy d d = 0 , the maps d ξ have to anticommute on each of the vertex of the cube. A way to do this is by multiplying edges d ξ by ( 1 ) ξ : = ( 1 ) i < j ξ i , where j is the location of ⋆ in ξ .
For better understanding, please see the n-cube of trefoil knot x 1 3 in Figure 8.
It is useful to note that the ordered basis of V is v + , v and the ordered basis of V V is v + v + , v v + , v + v , v v .
Definition 15.
Linear map m : V V V that merges two circles into a single circle is defined as m ( v + v + ) = v + , m ( v + v ) = v , m ( v v + ) = v and m ( v v ) = 0 .
Map Δ : V V V that divides a circle into two circles is defined as Δ ( v + ) = v + v + v v + and Δ ( v ) = v v ; see Figure 9.
Definition 16.
The homology group associated with the chain complex of a link L is defined as H r ( L ) = ker d r im d r + 1 .
Definition 17.
The kernel of the map d r : V r 1 V r , denoted by ker d r , is the set of all elements of V r 1 that go to the zero element of V r . The elements of the kernel are called cycles, while the elements of im d r + 1 are called boundaries.
Remark 5.
Note that the image of the chain complex of d r + 1 is a subset of kernel d r as, in general, d r d r + 1 = 0 .
Definition 18.
The graded Poincar e ´ polynomial Kh ( L ) in variables q and t of the complex is defined as
Kh ( L ) : = r t r q d i m H r ( L ) .
Theorem 3. (Khovanov [1]).
The graded dimension of homology groups H r ( L ) are link invariants. The graded Poincar e ´ polynomial Kh ( L ) is also a link invariant and Kh ( L ) | t = 1 = J ^ ( L ) .

3.1. Homology of x 1 3

Now, we give the Khovanov homology of link x 1 3 = Symmetry 10 00720 i004:
  • The n-cube: The 3-cube of x 1 3 is given in Figure 10:
  • Chain complex: The chain complex of x 1 3 ^ is
    0 d 4 V 3 d 3 3 V 2 d 2 3 V d 1 V 2 d 0 0 .
  • Ordered basis of the chain complex: The following are the vector spaces of the chain complex along with their ordered bases:
    V V V = v + v + v + , v v + v + , v + v v + , v + v + v , v v v + , v v + v , v + v v , v v v ( V V ) ( V V ) ( V V ) = ( v + v + , 0 , 0 ) , ( 0 , v + v + , 0 ) , ( 0 , 0 , v + v + ) , ( v v + , 0 , 0 ) , ( 0 , v v + , 0 ) , ( 0 , 0 , v v + ) , ( v + v , 0 , 0 ) , ( 0 , v + v , 0 ) , ( 0 , 0 , v + v ) , ( v v , 0 , 0 ) ( 0 , v v , 0 ) , ( 0 , 0 , v v ) V V V = ( v + , 0 , 0 ) , ( 0 , v + , 0 ) , ( 0 , 0 , v + ) , ( v , 0 , 0 ) , ( 0 , v , 0 ) , ( 0 , 0 , v ) V V = v + v + , v v + , v + v , v v
  • Differential maps in matrix form: Differential map d 3 V 1 V 2 V 3 = m ( v 1 v 2 ) v 3 , v 1 m ( v 2 v 3 ) , v 2 m ( v 1 v 3 ) in terms of a matrix is:
    d 3 = 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 1 0 0 0 0 0 0 1 0 1 0 ,
    and map d 2 V 1 V 2 , V 3 V 4 , V 5 V 6 = m ( v 3 v 4 ) m ( v 1 v 2 ) , m ( v 5 v 6 ) m ( v 1 v 2 ) , m ( v 5 v 6 ) m ( v 3 v 4 ) is d 2 = A 0 0 0 0 A A 0 , where A = 1 1 0 1 0 1 0 1 1 . Also, d 1 V 1 , V 2 , V 3 = Δ ( v 1 ) Δ ( v 2 ) + Δ ( v 3 ) is d 1 = 0 0 0 0 0 0 1 1 1 0 0 0 1 1 1 0 0 0 0 0 0 1 1 1 .
  • Khovanov Homology: On solving d 3 x = 0 or
    1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 1 0 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 1 1 0 0 0 0 0 0 1 0 1 0 x 1 x 2 x 3 x 4 x 5 x 6 x 7 x 8 = 0 ,
    we receive x 1 = x 2 = x 3 = x 4 = 0 , x 2 + x 3 = 0 , x 3 + x 4 = 0 , x 2 + x 4 = 0 , x 6 + x 7 = 0 , x 5 + x 6 = 0 , and x 5 + x 7 = 0 . So the kernel of d 3 = 0 0 0 0 0 0 0 1 . Similarly, the image of d 3 is
    1 1 1 0 0 0 0 0 0 0 0 0 , 0 0 0 1 1 0 0 0 1 0 0 0 , 0 0 0 1 0 1 0 1 0 0 0 0 , 0 0 0 0 0 0 1 1 1 0 0 0 , 0 0 0 0 0 0 0 0 0 0 1 1 , 0 0 0 0 0 0 0 0 0 1 1 0 , 0 0 0 0 0 0 0 0 0 1 0 1 .
    Thus,
    H 3 ( x 1 3 ^ ) = ker d 3 im d 4 = Z ( v v v ) 0 = Z ( v v v ) .
    To compute the homology of the next level, we first cancel out the terms that appear in both ker d 2 and im d 3 , and then use a special trick: Note that the last three summands of ker d 2 make up all of Z ( v v ) 3 , where the last three summands of im d 3 span the subspace of Z ( v v ) 3 generated by vectors ( 0 , 1 , 1 ) , ( 1 , 1 , 0 ) and ( 1 , 0 , 1 ) . Now, form a matrix whose columns are these vectors. Since the eigenvalues of this matrix are 1 , 1 , and 2, we can write:
    Z 3 ( 0 , 1 , 1 ) , ( 1 , 1 , 0 ) , ( 1 , 0 , 1 ) = Z 2 Z Z Z 1 Z Z 1 = Z 2 .
    Reducing the remaining matrices of kernel of d 2 and image of d 3 into reduced row echelon form, quotient ker d 2 im d 3 becomes isomorphic to Z . Hence,
    H 2 ( x 1 3 ^ ) = ker d 2 im d 3 = Z Z 2 .
    The range of d 2 is Z ( v + , v + , 0 ) Z ( v + , 0 , v + ) Z ( 0 , v + , v + ) Z ( v , v , 0 ) Z ( v , 0 , v ) Z ( 0 , v , v ) and the kernel of d 1 is Z ( v + , v + , 0 ) Z ( 0 , v + , v + ) Z ( v + , 0 , v + ) Z ( v , v , 0 ) Z ( 0 , v , v ) Z ( v , 0 , v ) . Since ker d 1 = im d 2 ,
    H 1 ( x 1 3 ^ ) = 0 .
    It is clear from the chain complex that the kernel of d 0 is the full space V V .
    H 0 ( x 1 3 ^ ) = Z ( v + v + ) Z ( v v + ) Z ( v + v ) Z ( v v ) Z ( v v + + v + v ) Z ( v v ) = Z ( v + v + ) Z .

3.2. Homology of Δ 2 k + 1

We now compute the homology of braid link Δ 2 k + 1 , where Δ = x 1 x 2 x 1 . The canonical form of this braid is Δ 2 k + 1 = x 1 2 k + 2 x 2 x 1 2 x 2 2 x 1 2 x 2 2 x 1 2 x 1 2 , having 2 k + 2 factors; you can see Δ 3 in Figure 11.
The co-chain complex of the link Δ 2 k + 1 is 0 d 1 V 3 d 0 6 k + 3 V 2 d 1 2 k + 1 1 4 k + 2 1 V 1 2 k + 1 1 + 4 k + 2 2 V 3 d 3 2 k + 1 1 4 k + 2 2 + 2 k + 1 2 4 k + 2 1 V 1 2 k + 1 1 + 4 k + 2 2 V 3 d 4 d 6 k + 1 4 k + 2 1 V 2 k + 1 2 k + 1 1 V 2 k + 3 d 6 k + 2 V 2 k + 2 d 6 k + 3 0 .
We now represent the differential maps in terms of matrices. The matrix representing differential d 0 has order 24 k + 12 × 8 and is
d 0 = A 0 0 0 0 0 0 0 0 A B 0 0 0 0 0 0 0 C A 0 0 0 0 0 0 0 0 C A B 0 .
Here, each matrix A , B , and C has a ( 6 k + 3 ) × 1 order:
A = 1 1 1 1 1 1 1 1 t B = 0 1 0 0 1 0 0 1 0 t C = 1 0 1 1 0 1 1 0 1 t
Since ker d 0 = Z v v v Z v + v v v v + v + v + v v and im d 1 = 0 , the homology at this level is
H 0 ( Δ 2 k + 1 ) = Z v v v Z v + v v v v + v + v + v v .
Now, we go for differential map d 1 . The matrix that represents it has an order of 20 ( 6 k 2 + 3 ) × 4 ( 6 k + 3 ) and is
d 1 = R 1 0 0 0 0 R 1 R 1 0 0 0 0 0 0 0 0 0 0 0 0 0 R 2 0 0 0 R 2 0 0 0 0 R 2 0 0 0 0 R 2 0 0 0 0 R 2 0 R 3 R 4 0 0 0 R n 1 R n .
The order of each of the matrix R i is ( 12 k + 6 ) × ( 6 k + 3 ) :
R 1 = 1 1 0 0 0 0 0 1 0 0 0 1 0 0 1 0 0 0 0 0 1 0 1 1 0 0 0 0 0 1 0 1 0 0 0 0 1 0 0 0 1 0 0 1 0 0 0 0 1 0 0 0 0 0 1 1 ,
R 2 = 1 0 1 0 0 0 0 1 0 1 0 0 0 0 1 0 0 0 1 0 0 1 0 0 0 0 1 0 1 0 0 0 0 0 1 0 1 0 0 1 0 0 0 1 0 0 1 0 0 0 0 0 0 1 0 1 ,
R 3 = 1 0 1 0 0 0 0 1 0 1 0 0 0 0 1 0 0 0 1 0 0 1 0 0 0 0 1 0 1 0 0 0 0 0 1 0 0 1 1 0 0 0 0 0 1 0 0 1 0 0 0 1 0 1 0 0 0 0 1 0 0 0 1 0 0 0 0 1 0 1 ,
R 4 = 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 1 0 0 1 0 0 0 0 0 0 1 0 1 ,
and, at the end, all rows of matrix R n are zero except for the last row, which is
0 0 0 1 0 1 .
Here, ker d 1 = Z ( v + v + + v + v + + v + v + + v + v + + v + v + + v + v + + v + v + + v + v + ) Z ( v + v + v + v + v + v + v + v + v + v + v + v + v + v + v + v ) Z ( v v + + v v + v + v v + v v + v ) Z ( v + v + v + v + v + v + v v + + v v + + v v + + v v + + v v + ) Z ( v v + v v ) Z ( v v + v v + v v + v v + v v ) and im d 0 = Z ( v + v + ) Z ( v + v ) Z ( v v + ) Z ( v v ) Z ( v + v + ) Z ( v + v ) Z ( v v + ) .
Since the number of Z spaces appear in the kernel of d 1 , it is exactly the same as the image of d 0 , H 1 ( Δ 2 k + 1 ) = 0 .
The image of d 1 is obvious. We just need the kernel of d 2 . The matrix that represents d 2 has an order of ( 2 6 k + 3 + 2 2 k + 2 ) ( 6 k + 5 ) × 20 ( 6 k 2 + 3 ) and is
S 1 S 2 S 3 S 4 S 5 S 6 S 7 S 8 S 9 S 20 S 21 S 21 S 22 S 23 S 24 S 25 S 26 S 27 S 28 S 40 S n 19 S n 18 S n 17 S n 16 S n 15 S n 14 S n 13 S n 12 S n 11 S n .
Here, the order of each S i is ( 4 k 2 + 3 ) × ( 6 k 2 + 3 ) , and is:
S 1 = 0 1 0 0 0 0 0 1 0 0 0 0 0 1 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 , S 2 = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 ,
S 3 = 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 , S 4 = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 ,
S n 2 = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ,
S n 1 = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 ,
S n = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 .
Thus, H 2 ( Δ 2 k + 1 ) = Z Z . Differential d 6 k + 2 of order ( 2 2 k + 2 ) × ( 2 k + 1 ) ( 2 2 k + 2 + 2 2 k + 3 ) is
d 6 k + 2 = Y 1 Y 2 Y 3 Y 4 Y 5 Y 6 k + 3 ,
where Y i are matrices, each having an order of 2 2 k + 2 × 2 2 k + 2 :
Y 1 = 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 0 1 0 1 1 0 1 0 1 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 ,
Y 2 = 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1 1 0 1 0 0 1 0 1 1 0 0 1 1 0 0 0 0 1 0 1 1 0 ,
Y 3 = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 1 1 0 0 0 0 0 0 1 1 0 1 0 0 1 1 ,
Y 4 = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 1 0 1 1 0 0 0 1 0 1 1 0 0 0 0 0 0 0 0 0 1 ,
Y i = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 1 1 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ,
Y 6 k + 3 = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 .
Here ker d 6 k + 3 is the full space V 2 k + 1 and the im d 6 k + 2 is Z ( v + v + v + v + ) Z ( v + v v + v + + v v + v + v + ) Z ( v + v + v v + + v + v v + v + ) Z ( v + v v + v + ) Z ( v + v v + v + v v + v + v ) Z ( v + v + v v + v + v v + v ) Z ( v + v + v + v ) Z ( v + v v v + + v v + v v + ) Z ( v + v v + v + v v + v + v ) Z ( v + v v v + ) Z ( v v v + v + ) Z ( v + v + v v + ) Z ( v v + v v + + v v v + v + ) Z ( v + v v v + v v + v v ) Z ( v + v v v ) Z ( v v v + v ) Z ( v v v v + ) Z ( v + v v + v ) Z ( v v v v ) Z ( v + v + v v ) Z ( v v + v + v ) Z ( v v + v v + ) Z ( v v + v v ) .
Thus, H 6 k + 3 ( Δ 3 ) = 0 , and we finally obtain the result:
Theorem 4.
The Khovanov homology of the link Δ 2 k + 1 is
H i ( Δ 2 k + 1 ) = 0 6 k i 3 Z Z i = 2 0 i = 1 Z Z i = 0
The following result gives some homology groups of Δ 2 k + 1 x 2 = x 1 2 k + 3 x 2 x 1 2 x 2 2 x 1 2 x 2 2 x 1 2 x 1 2 .
Theorem 5.
H i ( Δ 2 k + 1 x 2 ) = Z Z i = 0 0 i = 1 0 i = 6 k + 4
Proof. 
The cochain complex of link Δ 2 k + 1 x 2 is
0 d 1 V 3 d 0 6 k + 4 V 2 d 1 2 k + 2 1 4 k + 2 1 V 1 2 k + 2 2 + 4 k + 2 2 V 3 d 2 2 k + 2 1 4 k + 2 2 + 2 k + 2 2 4 k + 2 1 V 2 2 k + 2 1 + 4 k + 2 1 V 4 d 3 d 6 k + 2 2 k + 2 1 V 2 k 4 k + 2 1 V 2 k + 2 d 6 k + 3 V 2 k + 1 d 6 k + 4 0
Differential d 0 having an order of 24 k + 16 × 8 is
d 0 = A 0 0 0 0 0 0 0 0 A B 0 0 0 0 0 0 0 C A 0 0 0 0 0 0 0 0 C A B 0 ,
where A , B , and C, each having an order of ( 6 k + 4 ) × 1 , are:
A = 1 1 1 1 1 1 1 1 t B = 0 1 0 0 1 0 0 0 1 t C = 1 0 1 1 0 1 1 1 0 t
Since im d 1 = 0 and ker d 0 = Z ( v + v v v v + v + v v v + ) Z ( v v v ) , H 0 ( Δ 2 k + 1 x 2 ) = Z ( v + v v v v + v + v v v + ) Z ( v v v ) .
Now, differential d 1 of an order of 18 ( 6 k 2 + 6 ) × 4 ( 6 k + 4 ) is
d 1 = M 1 M 1 0 0 M 1 0 M 1 0 M 1 0 0 M 1 M 2 M 2 0 0 M 2 0 M 2 0 M 3 M 3 0 0 M 3 0 M 3 0 0 M 4 M 4 0 0 M 4 0 M 4 0 0 M n 1 M n ,
where the order of each M i is ( 16 k + 2 ) × ( 6 k + 4 ) and is M 1 = 1 0 0 0 1 0 0 0 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 1 0 0 0 ,   M 2 = 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 1 0 0 0 0 0 1 1 0 0 0 0 1 0 0 0 , M 3 = 0 1 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 , M 4 = 1 0 0 0 1 0 0 1 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 0 , and M n = 0 1 0 1 0 1 .
In this case, the kernel of d 1 and image of d 0 contain the same number of Z spaces. So, H 1 ( Δ 2 k + 1 x 2 ) = 0 .
Finally, the differential of d 6 k + 4 of an order of 2 2 k + 1 × ( 2 k + 3 ) ( 2 2 k ) ( 2 2 k + 1 ) is
d 6 k + 4 = Y 1 Y 2 Y 3 Y 4 Y i ,
where each Y i has an order of 2 2 k + 1 × 6 k + 4 and is Y 1 = 1 1 1 0 0 1 0 0 1 1 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 , Y 2 = 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 1 0 0 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 , Y 3 = 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 1 0 0 1 1 1 1 0 1 1 0 0 0 0 0 0 0 0 0 0 , Y 4 = 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 1 0 0 1 1 0 0 0 1 1 0 1 1 0 0 , a n d Y i = 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 1 0 0 .
It is evident that ker d 6 k + 4 is full space V 2 k + 1 . Moreover, im d 6 k + 3 is also V 2 k + 1 .
We also get the Khovanov homology of braid link Δ 2 k + 1 x 1 :
Theorem 6.
H i ( Δ 2 k + 1 x 1 ) = Z Z i = 0 0 i = 1 0 i = 6 k + 1
Proof. 
The proof is similar to the proof of Theorem 5: Obtain all states, organized them in columns, assign a graded vector space to each state, form chain groups as a direct sum of all vector spaces along a column, and form the chain complex. Then, write the differential maps in terms of matrices using the ordered bases of the chain groups, and compute their kernels and images. Finally, find the Khovanov homology groups using the relation H r ( L ) = ker d r im d r + 1 . ☐

4. Conclusions

Although computing the Khovanov homology of links is common in the literature, no general formulae have been given for all families of knots and links. In this paper, we considered a general three-strand braid x 1 x 2 x 1 x 2 , which, depending on the powers of Garside element Δ = x 1 x 2 x 1 , is divided into six subclasses, and gave the Khovanov homology of Δ 2 k + 1 , Δ 2 k + 1 x 2 , and Δ 2 k + 1 x 1 (To learn more about these classes, see Reference [23,24,25,26].) The results particularly cover the 0th, 1st, and top homology groups of these classes, and all homology groups, in general, of link Δ 2 k + 1 . We hope the results will help classifying links, and in studying the important properties of these links.

Author Contributions

Formal analysis, Y.C.K.; writing—original draft, A.R.N., Z.I., D.A., and S.M.K.; writing—review and editing, M.M.

Funding

This work was supported by the Dong-A University research fund.

Acknowledgments

We are very thankful to the reviewers for their valuable suggestions to improve the quality of this paper.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

References

  1. Khovanov, M. A Categorification of the Jones Polynomial. Duke Math. J. 2000, 101, 359–426. [Google Scholar] [CrossRef]
  2. Jones, V.F.R. A Polynomial Invariant for Knots via Von Neumann Algebras. Bull. Am. Math. Soc. 1985, 12, 103–111. [Google Scholar] [CrossRef]
  3. Bar-Natan, D. On Khovanov’s categorification of the Jones polynomial. Algeb. Geom. Topol. 2002, 2, 337–370. [Google Scholar] [CrossRef] [Green Version]
  4. Bar-Natan, D. Khovanov Homology for Tangles and Cobordisms. Geom. Topol. 2005, 9, 1443–1499. [Google Scholar] [CrossRef]
  5. Bar-Natan, D. Fast Khovanov Homology Computations. J. Knot Theory Ramif. 2007, 16, 243–255. [Google Scholar] [CrossRef]
  6. Ozsváth, P.; Rasmussen, J.; Szabó, Z. Odd Khovanov Homology. Algebr. Geom. Topol. 2013, 13, 1465–1488. [Google Scholar] [CrossRef]
  7. Gorsky, E.; Oblomkov, A.; Rasmussen, J. On Stable Khovanov Homology of Torus Knots. Exp. Math. 2013, 22, 265–281. [Google Scholar] [CrossRef]
  8. Putyra, K.K. On Triply Graded Khovanov Homology. arXiv, 2015; arXiv:1501.05293v1. [Google Scholar] [CrossRef]
  9. Manion, A. The Rational Khovanov Homology of 3-Strand Pretzel Links. arXiv, 2011; arXiv:1110.2239. [Google Scholar] [CrossRef]
  10. Nizami, A.R.; Munir, M.; Usman, A. Khovanov Homology of Braid Links. Rev. UMA 2016, 57, 95–118. [Google Scholar]
  11. Nizami, A.R.; Munir, M.; Sohail, T.; Usman, A. On the Khovanov Homology of 2- and 3-Strand Braid Links. Adv. Pure Math. 2016, 6, 481–491. [Google Scholar] [CrossRef]
  12. Stosic, M. Properties of Khovanov Homology for Positive Braid Knots. arXiv, 2006; arXiv:math/0511529. [Google Scholar]
  13. Rozansky, L. An Infinite Torus Braid Yields a Categorified Jones-Wenzl Projector. Fundam. Math. 2014, 225, 305–326. [Google Scholar] [CrossRef]
  14. Cooper, B.; Krushkal, V. Categorification of the Jones-Wenzl projectors. Quantum Topol. 2012, 3, 139–180. [Google Scholar] [CrossRef] [PubMed]
  15. Gorsky, E.; Oblomkov, A.; Rasmussen, J.; Shende, V. Torus Knots and the Rational DAHA. Duke Math. J. 2014, 163, 2709–2794. [Google Scholar] [CrossRef]
  16. Hogancamp, M. Categorified Young Symmetrizers and Stable Homology of Torus Links. Geom. Topol. 2018, 22, 2943–3002. [Google Scholar] [CrossRef]
  17. Lipshitz, R.; Sarkar, S. A Khovanov Stable Homotopy Type. J. Am. Math. Soc. 2014, 27, 983–1042. [Google Scholar] [CrossRef]
  18. Manturov, V. Knot Theory; Chapman and Hall/CRC: Boca Raton, FL, USA, 2004. [Google Scholar]
  19. Reidemeister, K. Elementary Begründung der Knotentheorie. Abh. Math. Sem. Univ. Hambg. 1927, 5, 24–32. [Google Scholar] [CrossRef]
  20. Artin, E. Theory of Braids. Ann. Math. 1947, 48, 101–126. [Google Scholar] [CrossRef]
  21. Alexander, J. Topological invariants of knots and links. Trans. Am. Math. Soc. 1928, 20, 275–306. [Google Scholar] [CrossRef]
  22. Kauffman, L.H. State Models and the Jones Polynomial. Topology 1987, 26, 395–407. [Google Scholar] [CrossRef]
  23. Berceanu, B.; Nizami, A.R. A recurrence relation for the Jones polynomial. J. Korean Math. Soc. 2014, 51, 443–462. [Google Scholar] [CrossRef]
  24. Khovanov, M. Patterns in Knot Cohomology I. Exp. Math. 2003, 12, 365–374. [Google Scholar] [CrossRef]
  25. Lawson, T.; Lipshitz, R.; Sarkar, S. The Cube and the Burnside Category. arXiv, 2015; arXiv:1505.00512. [Google Scholar]
  26. Reidemeister, K. Knot Theory; Chelsea Publ. and Co.: New York, NY, USA, 1948. [Google Scholar]
Figure 1. Knots.
Figure 1. Knots.
Symmetry 10 00720 g001
Figure 2. Links.
Figure 2. Links.
Symmetry 10 00720 g002
Figure 3. Crossing.
Figure 3. Crossing.
Symmetry 10 00720 g003
Figure 4. Reidemeister moves.
Figure 4. Reidemeister moves.
Symmetry 10 00720 g004
Figure 5. 2-strand braid.
Figure 5. 2-strand braid.
Symmetry 10 00720 g005
Figure 6. Elementary braid x i .
Figure 6. Elementary braid x i .
Symmetry 10 00720 g006
Figure 7. Braid closure.
Figure 7. Braid closure.
Symmetry 10 00720 g007
Figure 8. n-cube of x 1 3 .
Figure 8. n-cube of x 1 3 .
Symmetry 10 00720 g008
Figure 9. m and Δ maps.
Figure 9. m and Δ maps.
Symmetry 10 00720 g009
Figure 10. The 3-cube of x 1 3 .
Figure 10. The 3-cube of x 1 3 .
Symmetry 10 00720 g010
Figure 11. Δ3.
Figure 11. Δ3.
Symmetry 10 00720 g011

Share and Cite

MDPI and ACS Style

Kwun, Y.C.; Nizami, A.R.; Munir, M.; Iqbal, Z.; Arshad, D.; Min Kang, S. Khovanov Homology of Three-Strand Braid Links. Symmetry 2018, 10, 720. https://doi.org/10.3390/sym10120720

AMA Style

Kwun YC, Nizami AR, Munir M, Iqbal Z, Arshad D, Min Kang S. Khovanov Homology of Three-Strand Braid Links. Symmetry. 2018; 10(12):720. https://doi.org/10.3390/sym10120720

Chicago/Turabian Style

Kwun, Young Chel, Abdul Rauf Nizami, Mobeen Munir, Zaffar Iqbal, Dishya Arshad, and Shin Min Kang. 2018. "Khovanov Homology of Three-Strand Braid Links" Symmetry 10, no. 12: 720. https://doi.org/10.3390/sym10120720

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop