Processing math: 100%
Research article

List vertex arboricity of planar graphs without 5-cycles intersecting with 6-cycles

  • Received: 11 January 2021 Accepted: 08 June 2021 Published: 28 June 2021
  • MSC : 05C15

  • The vertex arboricity a(G) of a graph G is the minimum number of colors required to color the vertices of G such that no cycle is monochromatic. The list vertex arboricity al(G) is the list version of this concept. In this paper, we prove that if G is a planar graph without 5-cycles intersecting with 6-cycles, then al(G)2.

    Citation: Yanping Yang, Yang Wang, Juan Liu. List vertex arboricity of planar graphs without 5-cycles intersecting with 6-cycles[J]. AIMS Mathematics, 2021, 6(9): 9757-9769. doi: 10.3934/math.2021567

    Related Papers:

    [1] Hongyu Chen, Li Zhang . A smaller upper bound for the list injective chromatic number of planar graphs. AIMS Mathematics, 2025, 10(1): 289-310. doi: 10.3934/math.2025014
    [2] Yuehua Bu, Hongrui Zheng, Hongguo Zhu . On the list injective coloring of planar graphs without a $ {4^ - } $-cycle intersecting with a $ {5^ - } $-cycle. AIMS Mathematics, 2025, 10(1): 1814-1825. doi: 10.3934/math.2025083
    [3] Xiaoxue Hu, Jiangxu Kong . An improved upper bound for the dynamic list coloring of 1-planar graphs. AIMS Mathematics, 2022, 7(5): 7337-7348. doi: 10.3934/math.2022409
    [4] Yuehua Bu, Qi Jia, Hongguo Zhu, Junlei Zhu . Acyclic edge coloring of planar graphs. AIMS Mathematics, 2022, 7(6): 10828-10841. doi: 10.3934/math.2022605
    [5] Fugang Chao, Donghan Zhang . Neighbor sum distinguishing total choice number of IC-planar graphs with restrictive conditions. AIMS Mathematics, 2023, 8(6): 13637-13646. doi: 10.3934/math.2023692
    [6] Yueying Zhao, Lianying Miao . Planar graphs without $ \{4, 6, 8\} $-cycles are 3-choosable. AIMS Mathematics, 2021, 6(9): 10253-10265. doi: 10.3934/math.2021593
    [7] Yuehua Bu, Qiang Yang, Junlei Zhu, Hongguo Zhu . Injective coloring of planar graphs with girth 5. AIMS Mathematics, 2023, 8(7): 17081-17090. doi: 10.3934/math.2023872
    [8] Xin Xu, Xu Zhang, Jiawei Shao . Planar Turán number of double star $ S_{3, 4} $. AIMS Mathematics, 2025, 10(1): 1628-1644. doi: 10.3934/math.2025075
    [9] Shahbaz Ali, Muhammad Khalid Mahmmod, Raúl M. Falcón . A paradigmatic approach to investigate restricted hyper totient graphs. AIMS Mathematics, 2021, 6(4): 3761-3771. doi: 10.3934/math.2021223
    [10] Zongrong Qin, Dingjun Lou . The k-subconnectedness of planar graphs. AIMS Mathematics, 2021, 6(6): 5762-5771. doi: 10.3934/math.2021340
  • The vertex arboricity a(G) of a graph G is the minimum number of colors required to color the vertices of G such that no cycle is monochromatic. The list vertex arboricity al(G) is the list version of this concept. In this paper, we prove that if G is a planar graph without 5-cycles intersecting with 6-cycles, then al(G)2.



    All graphs considered in this paper are finite, simple and undirected. For a graph G, we use V(G),E(G), Δ(G), and δ(G) to denote its vertex set, edge set, maximum degree, and minimum degree, respectively. A plane graph is a planar graph with a given planar drawing on the Euclidean plane. If G is a plane graph, let F(G) denote the set of faces in G. We say that two cycles (or faces) are adjacent if they share at least one edge. In particular, when they share exactly one edge and two vertices, they are said to be normally adjacent. Two cycles (or faces) are intersecting if they share at least one vertex.

    The vertex arboricity, denoted by a(G), of a graph G is the minimum number of subsets into which V(G) can be partitioned so that each subset induces a forest. Obviously, a(G)=1 if and only if G itself is a forest. In 1968, Chartrand, Kronk and Wall [2] first introduced the vertex arboricity of a graph and proved that a(G)Δ(G)+12 for any graph G and a(G)3 for any planar graph G. It is known that there exist infinitely many planar graphs G such that a(G)=3. In 1989, Hakimi and Schmeichel [6] provided a characterization by showing that a plane graph G has a(G)=2 if and only if G, the dual of G, contains a connected Eulerian spanning subgraph. In 2008, Raspaud and Wang [10] conjectured that every planar graph G without adjacent 3-cycles has a(G)2. To attack this conjecture, Chen, Raspaud and Wang [3] confirmed a weak version, i.e., every planar graph G without intersecting 3-cycles has a(G)2. Some sufficient conditions for a planar graph G to have a(G)2 have been obtained in [5,7,10,13].

    A graph G is said to be L-forested-colorable if for any color list L={L(v)vV(G)}, one can choose a color for each vertex v from its list L(v) so that the subgraph induced by every color class is a forest. The list vertex arboricity al(G) is the minimum number of integer k such that G is L-forested-colorable with |L(v)|k for each vV(G). Obviously, a(G)al(G) for any graph G. In 2009, Borodin and Ivanova [1] proved that al(G)2 if G is a planar graph without 4-cycles adjacent to 3-cycles. This result has been recently extended by Chen, Huang and Wang [4] to a toroidal graph without 4-cycles adjacent to 3-cycles. In 2020, Wang, Huang and Chen [9] proved that every planar graph G without intersecting 5-cycles has al(G)2. The list vertex arboricity of toroidal graphs has also been extensively investigated, see [8,11,14].

    In this paper, we prove the following result:

    Theorem 1. If G is a planar graph without 5-cycles intersecting with 6-cycles, then al(G)2.

    We first introduce a few concepts and terminology used in the paper. Let G be a plane graph. For xV(G)F(G), let dG(x), or simply d(x), denote the degree of x in G. A vertex of degree k (resp., at least k, at most k) is called a k-vertex (resp., k+-vertex, k-vertex). Similarly, we can define a k-face, k+-face and k-face. For a k-vertex vV(G), we usually use v0,v1,,vk1 to denote the neighbors of v in G in a clockwise order, and let fi denote the incident face of v that contains vvi,vvi+1 as two boundary edges for i=0,1,,k1, where indices are taken modulo k. For a face fF(G), we use b(f) to denote the boundary walk of f and write f=[v1v2vn] if v1,v2,,vn are the vertices of b(f) in a clockwise order. Let V(f)=V(b(f)). Moreover, the face adjacent to f with e=vivi+1 as a common boundary edge is simply denoted by fvivi+1. For vV(G), let Fi(v) (or Fi+(v)) denote the set of i-faces (or i+-faces) incident to v. Moreover, let mi(v)=|Fi(v)| and mi+(v)=|Fi+(v)|. A k-vertex v is called a (d0,d1,,dk1)- vertex if d(fi)=di for i=0,2,,k1. A cycle C in a plane graph G is called separating if both its interior and exterior contain at least one vertex of G.

    This paper is organized as follows: In section 2 we give the proof of Theorem 1. Initially, we explore structural properties of a minimal counterexample. Then we use the discharging technique to contradict the existence of such a graph. Finally, in Section 3 we conclude with conjectures.

    We prove Theorem 1 by contradiction. Suppose that G is a minimal counterexample to the Theorem 1, i.e., G is a planar graph satisfying the following conditions:

    (ⅰ) without 5-cycles intersecting with 6-cycles;

    (ⅱ) al(G)>2; and

    (ⅲ) having as few as possible vertices.

    Let L be a list assignment of V(G) such that every vV(G) has |L(v)|=2. If G contains a vertex v of degree at most 3, let H=Gv. By the minimality of G, H admits an L-forested-coloring π. Based on π, we may color v with a color in L(v) which appears at most once in its neighbors such that π is extended to G, which is a contradiction. Hence suppose that δ(G)4.

    As given in [9], the following lemma still holds for the current graph G:

    Lemma 1. G contains no a 4-cycle C=x1x2x3x4x1 with a chord x1x3 such that dG(x1)5 and dG(xi)=4 for i=2,3,4, as shown in the configuration C1 of Figure 1.

    Figure 1.  The configurations C1 and C2.

    For SV(G), let G[S] be a subgraph of G induced on S and let π be an L-forested-coloring of G[S]. Note that by definition every color class of π induces a forest in G. We call π a partial L-forested-coloring of G. For a vertex uV(G)S and a color cL(u), we use τL(c,u) to denote the number of times that c appears in the colored neighbors of u in G. Set

    τL(u)=min{τL(c,u)|cL(u)}

    Vertex u is said to be free with respect to (L,π) if π can be extended into a partial L-forested-coloring of G by coloring u with a color cL(u) such that τL(c,u)=τL(u).

    Lemma 2. ([12]) Let uV(G)S be a 4-vertex. If at least one of the following conditions holds, then u is free.

    (1) At least one of neighbors of u is uncolored;

    (2) At least three colors appear in the neighbors of u;

    (3) Some color appears at least three times in the neighbors of u.

    Lemma 3. G does not contain a internal 4-vertex v incident to four 3-faces such that v1,v2 are internal 4-vertices and v0,v3 are internal 5-vertices, as shown in the configuration C2 of Figure 1.

    Proof. Suppose that G contains such a 4-vertex v. By symmetry, we have to discuss the following cases.

    Case 1. v0v2G.

    Let v0,v0 be the neighbors of v0 other than v,v1,v3; v3,v3 be the neighbors of v3 other that v,v0,v2; v1 be the neighbor of v1 other than v,v0,v2; v2 be the neighbor of v2 other than v,v1,v3. Consider the graph H=G{v,v0,v1,v2,v3}. By the minimality of G, H has an L-forested-coloring π. Define a sublist L of colors as follows:

    L(vi)=L(vi){π(vi),π(vi)} for i=0,3.

    It is easy to see that |L(vi)|0 for i=0,3. Moreover, if π(vi)=π(vi), or π(vi)π(vi) and L(vi){π(vi),π(vi)}, then |L(vi)|1. Let S={x{v0,v3} | |L(x)|1}. Obviously, 0|S|2. We are going to extend π to G, which leads to a contradiction.

    Firstly, we give each vi a color, i=0,1,2,3. For i=1,2, since |L(vi)|2 and vi has only one neighbor which is colored, there exists a color aiL(vi)π(vi) that we use to color vi with αi. Suppose that i{0,3}. If viS, then we color the vertex vi with a color biL(vi). Otherwise, color vi with the color π(vi). We have to discuss the following cases.

    Case 1.1 |S|1.

    By symmetry, assume that |L(v0)|1 and a0L(v0). Let π(v1)=a1, π(v2)=a2, and π(v3)=a3.

    Case 1.1.1. a0a3.

    Suppose that {a1,a2}{a0,a3}. Then v is free by Lemma 2 and π can be extended to G, a contradiction. Otherwise, {a1,a2}={a0,a3}. If a1=a0 and a2=a3, then we recolor v2 with a color in L(v2){a2} and color v with a color in L(v){a0}. If a1=a3 and a2=a0, then we color v with a color in L(v).

    Case 1.1.2. a0=a3.

    If a1a2, then v is free by Lemma 2 and π can be extended to G, a contradiction. If a1=a2=a0, then recolor v1 with a color in L(v1){a0}, and color v with a color in L(v){a0}. Otherwise, a1=a2a0, then we recolor v2 with a color in L(v2){a2} and color v with a color in L(v){a0}. Now if G does not contain monochromatic cycle, then we get an L-forested-coloring of G, a contradiction. Otherwise, G contains a monochromatic cycle v2v3v2v2, then |L(v3)|=0, L(v3)={π(v3),π(v3)} and π(v3)π(v3). Recoloring v3 with a color in L(v3){a3}, we extend π to G, a contradiction.

    Case 1.2. |S|=0.

    Without loss of generality, let π(vi)=ai, i=0,1,2,3.

    Case 1.2.1. a0a3.

    Suppose that {a1,a2}{a0,a3}. Then v is free by Lemma 2 and π can be extended to G, a contradiction. Otherwise, {a1,a2}={a0,a3}. If a1=a0 and a2=a3, then we recolor v0 with the color π(v0), and color v with a color in L(v){a3}. If there is no monochromatic cycle in G, then π is extended to G, a contradiction. Otherwise, if G contains a monochromatic cycle v0v3v0v0, then color v3 with the color π(v3). If a1=a3 and a2=a0, then we color v with a color in L(v).

    Case 1.2.2. a0=a3.

    Recolor v0 with the color π(v0), and then the proof can be given as in Case 1.2.1.

    Case 2. v0v2G.

    Let v0 be the neighbors of v0 other than v,v1,v2,v3; v3,v3 be the neighbors of v3 other that v,v0,v2; v1 be the neighbor of v1 other than v,v0,v2.

    Consider the graph H=G{v,v0,v1,v2,v3}. By the minimality of G, H has an L-forested-coloring π. Define a sublist L(v3)=L(v3){π(v3),π(v3)}.

    Firstly, we give each vi a color, i=0,1,2,3. For i=0,1, since |L(vi)|2 and vi has only one neighbor which is colored, there exists a color aiL(vi){π(vi)} that we use to color vi with ai. We color the vertex v2 with a color a2L(v2){a0}. If L(v3)1, then we color the vertex v3 with a color a3L(v3). Otherwise, color v3 with the color π(v3).

    Case 2.1. a0a3.

    Suppose that {a1,a2}{a0,a3}. Then v is free by Lemma 2 and π can be extended to G, a contradiction. Otherwise, {a1,a2}={a0,a3}. Thus, a0=a1 and a2=a3. We recolor v1 with a color in L(v1){a0} and color v with a color in L(v){a3}.

    Case 2.2. a0=a3.

    If a1a2, then v is free by Lemma 2 and π can be extended to G, a contradiction. If a1=a2, then recolor v1 with a color in L(v1){a2}, and color v with a color in L(v){a0}.

    This completes the proof of Lemma 3.

    To obtain a contradiction by applying discharging technique, we define a new graph H from G as follows: If there is no separating 3-cycle in G, let H=G; otherwise, choose a separating 3-cycle T with the least internal vertices, and let H=G[V0(T)V(T)], where V0(T) is the set of internal vertices of T. Let V0(H)=V0(T), which are called the set of internal vertices in H. Let fΔ denote the outer face of H, and let the set of interior faces of H be F0(H)=F(H){fΔ}. For each vertex vV0(H), it holds obviously that dH(v)=dG(v). Since G is connected, so is H. In the following, we will contradict the existence of subgraph H and therefore of G, by using the discharging technique. Using Euler's formula |V(H)||E(H)|+|F(H)|=2, we can deduce the following identity.

    vV(H)(2dH(v)6)+fF(H)(dH(f)6)=12 (2.1)

    Define an initial weight w on vertices and faces of H by w(v)=2dH(v)6 if vV(H) and w(f)=dH(f)6 if fF(H). It follows from (2.1) that the total sum of weights is equal to 12. In what follows, we will design some discharging rules and redistribute weights accordingly. Once the discharging is finished, a new weight function w is produced. However, the total sum of weights is kept fixed during the discharging process. Nevertheless, we can show that the sum of w(x) is at least 11.6 for all xV(H)F(H). This leads to the following obvious contradiction

    11.6xV(H)F(H)w(x)=xV(H)F(H)w(x)=12

    and hence the theorem follows.

    For a 3-face fF(G), we say that it is bad if V(f) contains three internal 4-vertices; weak if V(f) contains two internal 4-vertices and one 5+-vertex, or two internal 4-vertices with |V(f)V(T)|=1; and good otherwise.

    Observation 1. Let f,fF0(H) with dH(f)=3 and dH(f)=4. If f is normally adjacent to f, then b(f)b(f) contains a 5-cycle.

    Let C5C6 denote a configuration that a vertex lies in a 5-cycle and a 6-cycle at the same time.

    Observation 2. Let f,fF0(H) with dH(f)=3 and dH(f)=5. If f is normally adjacent to f, then b(f)b(f) contains a configuration C5C6.

    Proof. Suppose that f=[x1x2x3x4x5] and f=[x1x2x6]. Since f is a 5-face, x1,x2,,x5 are mutually distinct. Since f and f are normally adjacent, it follows that x6V(f) and hence a 6-cycle x1x6x2x3x4x5x1 is found. Thus, b(f)b(f) has a configuration C5C6 at the vertex x1.

    Claim 1. If vV0(H) is a 5+-vertex, then m3(v)3dH(v)4.

    Proof. Since v is an internal vertex, all the faces incident to v are internal faces. It is easy to check that if v is incident to four consecutive 3-faces, then G will contain a configuration C5C6. This shows that m3(v)3dH(v)4.

    Claim 2. If vV0(H) is incident to two 3-faces fi,fi+1, then dH(fj)4,5 for each j{i1,i+2}.

    Proof. Without loss of generality, assume that i=1.

    dH(f3)=4.

    Suppose that f3=[vv3uv4]. If u,v1,v2 are mutually distinct, then a 5-cycle vv2v3uv4v and a 6-cycle vv1v2v3uv4v are found. Thus, b(f1)b(f2)b(f3) has a configuration C5C6 at the vertex v, a contradiction. If v2=u, then a separating 3-cycle vv2v4v with fewer internal vertices than T appears, contradicting the choice of T. If v1=u, then a separating 3-cycle vv1v3v with fewer internal vertices than T appears, contradicting the choice of T.

    dH(f3)=5.

    Suppose that f3=[vv3u1u2v4]. By Observation 2, f2,f3 are not normally adjacent, thus v2=u1 or v2=u2. If v2=u1, then v3 is a 2-vertex and v is a vertex on T, a contradiction. If v2=u2, then a separating 3-cycle vv2v4v with fewer internal vertices than T appears, contradicting the choice of T.

    Claim 3. Let vV0(H) with dH(v)4. If dH(fi)=dH(fi+1)=dH(fi+2)=3, then dH(fi1),dH(fi+3) 7.

    Proof. Without loss of generality, assume that i=1. By Claim 1, dH(f0)3. By Claim 2, dH(f0)4,5. If dH(f0)=6, a 5-cycle vv1v2v3v4v and a 6-cycle f0 are found. Thus, b(f0)b(f1)b(f2)b(f3) has a configuration C5C6 at the vertex v, a contradiction.

    Claim 4. If vV0(H) is incident to two 3-faces fi,fi+2, then dH(fi+1)4,5.

    Proof. Without loss of generality, assume that i=1.

    dH(f2)=4.

    Suppose that f2=[vv2uv3]. If u,v1,v4 are mutually distinct, then a 5-cycle vv1v2uv3v and a 6-cycle vv1v2uv3v4v are found. Thus, b(f1)b(f2)b(f3) has a configuration C5C6 at the vertex v, a contradiction. If v1=u, then a separating 3-cycle vv1v3v with fewer internal vertices than T appears, contradicting the choice of T. If v4=u, then a separating 3-cycle vv2v4v with fewer internal vertices than T appears, contradicting the choice of T.

    dH(f2)=5.

    Suppose that f2=[vv2u1u2v3]. By Observation 2, f1,f2 are not normally adjacent and f2,f3 are not normally adjacent, thus v1=u1 and v4=u2. In this case, v2 and v3 are both 2-vertices and they are stand on different 3-faces, a contradiction.

    Our discharging rules are defined as follows:

    (R0) Let vV(T). We carry out the following three subrules (R0.1), (R0.2) and (R0.3):

    (R0.1) v sends 0.5 to each incident 4-face, and 0.2 each incident 5-face.

    (R0.2) If dH(v)5, then v sends 2 to each incident internal 3-face. Otherwise, assume that 3dH(v)4. If m3(v)=dH(v), then v sends 1.5 to each incident internal 3-face. If m3(v)dH(v)1, then v sends 2 to each incident internal 3-face.

    (R0.3) If f=[vx1x2x3] is a 4-face, then v sends 0.1 to x2 through the face f.

    (R1) Every 4-face gets 0.5 from each of its incident internal vertices.

    (R2) Every 5-face gets 0.2 from each of its incident internal vertices.

    (R3) Let vV0(H) with m3(v)1.

    (R3.1) Suppose that dH(v)=4.

    Assume that m3(v)=4. If v is incident with a bad 3-face and a good 3-face, then v sends 1 to bad 3-face, 0 to good 3-face and 0.5 to each of its incident weak 3-faces. Otherwise, v sends 0.5 to each of its incident 3-faces.

    Assume that m3(v)=3. If v is incident with a bad 3-face, then v sends 1 to bad 3-face, and 0.5 to each of its other incident 3-faces. If v is incident with at least one good 3-face, then v sends 0.5 to good 3-face, and 0.75 to each of its incident weak 3-faces. Otherwise, v sends 0.5 to each of its incident weak 3-faces.

    If m3(v)2, then v sends 1 to each of its incident 3-faces.

    (R3.2) Suppose that dH(v)=5.

    If dH(fi)=3 for i=0,1,2, and dH(f3),dH(f4)7, then v sends 1.25 to each of f0,f1,f2.

    If dH(fi)=3 for i=0,1,3, and dH(f2),dH(f4)6, then v sends 1.25 to each of f0,f1 and 1.5 to f3.

    For other cases, v sends 1.5 to each of its incident 3-faces.

    (R3.3) If dH(v)6, then v sends 1.5 to each of its incident 3-faces.

    (R4) Assume that vV0(H) is a (3,4,5,4)-vertex with f1=[vv1x1v2] and f3=[vv3x2v0], as shown in the configuration G1 of Figure 2. Then each of x1 and x2 send 0.1 to v.

    Figure 2.  The configurations G1 and G2.

    (R5) Assume that vV0(H) is an internal (3,3,3,3)-vertex and v0 is an internal (3,3,7+,3,7+)-vertex. Let f, other than f3, denote the incident face of the edge v0v3. Such configuration G2 is depicted Figure 2.

    (1) If v1,v2 are internal 4-vertices and v3 is an internal 6+-vertex, then f sends 0.25 to f3.

    (2) If v3 is an internal 4-vertex, then f send 0.25 to f3.

    If v is a k-vertex incident with m 3-faces, then v is called a km-vertex. If a k-vertex v sends the weight a to an incident 3-face f, then v is called a ka-vertex for f. A kam-vertex for a 3-face f is a km-vertex that sends the weight a to f.

    For x,yV(H)F(H), let τ(xy) denote the amount of weights transferred from x to y according to the rules (R0)-(R5).

    Observation 3. Let v be an internal 4+-vertex incident to an internal 3-face f. Then

    (1) τ(vf){1,0.75,0.5,0} if dH(v)=4;

    (2) τ(vf){1.25,1.5} if dH(v)=5;

    (3) τ(vf)=1.5 if dH(v)6.

    Claim 5. Suppose that f=[x1x2x3] is an internal 3-face. Let s denote the total number of 40.75-vertices, 40.5-vertices and 40-vertices which are adjacent to f. Then s1.

    Proof. Suppose that s2. By the discharging rules, if xi is a 40.75-vertex, 40.5-vertex or 40-vertex, then xi is an internal 43-vertex or 44-vertex. We consider the following possibilities by symmetry.

    At least one of x1, x2 is a 43-vertex in V0(H).

    If fx1x2 and fx1x3 are 3-faces, then neither x2 nor x3 is an internal 43-vertex or 44-vertex, for otherwise H will contain a separating 3-cycle with fewer internal vertices than T or a configuration C5C6 at the vertex x1, a contradiction. The case in which fx1x2=[x1x2w] and fx1w=[x1wy] are 3-faces can be similarly discussed, where y and w are the neighbors of x1 other than x2 and x3.

    Both x1 and x2 are internal 44-vertices.

    In this case, H contains a separating 3-cycle with fewer internal vertices than T or a configuration C5C6 at the vertex x1, also a contradiction.

    The above discussion implies that s1.

    Lemma 4. w(f)0 for each fF0(H).

    Proof. Let fF0(H). Clearly, dH(f)3.

    Case 1. Suppose that dH(f)=3 and f=[x1x2x3].

    Then w(f)=3. Since f is an internal face of H, we can derive that |V(f)V(T)|2.

    Case 1.1. |V(f)V(T)|=2.

    Then f is a good 3-face. By (R0), f receives 1.5 or 2 from each of its two vertices of T. Hence w(f)3+1.5×2=0.

    Case 1.2. |V(f)V(T)|=1.

    Without loss of generality, let x1V(T). We distinguish cases depending on whether x2 and x3 are internal 4-vertices or internal 5+-vertices. Assume that both x2 and x3 are internal 4-vertices. Then f is a weak 3-face. Rule (R3.1) assures that 4-vertices always send non-zero weight to weak faces. Hence by Observation 3, x2 and x3 send either 0.5 or 0.75 or 1 to f. Now, by Claim 5, at most one of x2 and x3 is an 40.5-vertex or 40.75-vertex. Equivalently at least one of them is a 41-vertex. Thus, w(f)3+1.5+1+0.5=0 by (R0) and (R3). Assume that x2 is a internal 4-vertex and x3 is a internal 5+-vertex. If x2 is not a 40-vertex, then w(f)3+1.5+1.25+0.5=0.25 by (R0) and (R3). Otherwise, x2 is a 40-vertex. By (R3.1), x2 is a 44-vertex. If dH(x1)=3, then |V(f)V(T)|=2, a contradiction. If dH(x1)4, then fx1x3 is a 4+-face and m3(x1)dH(x1)1, otherwise it will produce a 3-vertex in v0(H), or a separating 3-cycle with fewer internal vertices than T or a configuration C5C6 at the vertex x1, a contradiction. Thus f gets 2 from x1. Therefore, w(f)3+2+1.25=0.25 by (R0) and (R3). Assume that both x2 and x3 are 5+-vertices, then w(f)3+1.5+1.25×2=1 by (R0) and (R3).

    Case 1.3. |V(f)V(T)|=0.

    Let dH(x1)dH(x2)dH(x3).

    Assume that dH(x1)5, then f gets at least 1.25 from each of x1,x2,x3 by Observation 3. Thus w(f)3+1.25×3=0.75.

    Assume that dH(x1)=4,dH(x2)5, then f gets at least 1.25 from each of x2,x3 by (R3.2) and (R3.3). If x1 is not a 40-vertex, then f gets at least 0.5 from x1 by (R3.1). Hence, w(f)3+0.5+1.25×2=0. Otherwise, x1 is a 40-vertex, then x1 is a internal 44-vertex and x1 is incident with a bad 3-face by (R3.1). By Lemma 1 and Lemma 3, x2 or x3 is a internal 6+-vertex, say x2, and f gets 1.5 from x2 by (R3.3). If x3 is a internal 6+-vertex, or a internal 5-vertex with m3(x3)=2, then f gets 1.5 from x3 by (R3.2) and (R3.3). Thus, w(f)3+1.5×2=0. Otherwise, x3 is a internal 5-vertex with m3(x3)=3. If x3 is a internal (3,3,3,6+,6+)-vertex, a (3,3,6,3,6+)-vertex, or a (3,3,6+,3,6)-vertex, it will contain a configuration C5C6 at the vertex x3, a contradiction. If x3 is a internal (3,3,7+,3,7+)-vertex, it is the construction G2 and f gets 0.25 from the 7+-face fx2x3 by (R5). Thus, w(f)3+1.5+1.25+0.25=0 by (R3.2) and (R3.3).

    Assume that dH(x1)=dH(x2)=4 and dH(x3)5. By definition, f is a weak 3-face and f gets at least 0.5 from each of x1,x2 by (R3.1). By Claim 5, at least one of x1 and x2 is a 41-vertex. Without loss of generality, assume that x2 is a 41-vertex. If x1 is a 41-vertex or a 40.75-vertex, then w(f)3+1+0.75+1.25=0 by (R3). Otherwise, x1 is a 40.5-vertex. If x3 is a internal 6+-vertex or a 51.5-vertex, then w(f)3+1+1.5+0.5=0 by (R3). Therefore, x3 is a internal 5-vertex with m3(x3)=3. By Lemma 1 and the definition of good 3-face, we can derive that x1 is incident with a good 3-face. If x1 is a internal 43-vertex, then f gets 0.75 from x1 by (R3.1), a contradiction. Otherwise, x1 is a internal 44-vertex. If x3 is a (3,3,3,6+,6+)-vertex, (3,3,6,3,6+)-vertex or (3,3,6+,3,6)-vertex, it will contain a configuration C5C6 at the vertex x3, which is a contradiction. If x3 is a (3,3,7+,3,7+)-vertex, it is the construction G2 and f gets 0.25 from the 7+-face fx2x3 by (R5). Thus, w(f)3+1+0.5+1.25+0.25=0 by (R3).

    Assume that dH(x1)=dH(x2)=dH(x3)=4.

    By definition, f is a bad 3-face and f gets 1 from each of x1,x2,x3 by (R3). Thus, w(f)3+1×3=0.

    Case 2. 4dH(f)6.

    (1) If dH(f)=4, then w(f)=2+0.5×4=0 by (R0) and (R1).

    (2) If dH(f)=5, then w(f)=1+0.2×5=0 by (R0) and (R2).

    (3) If dH(f)=6, then w(f)=0.

    Next, if f=[v1v2v3vn] is a 7+-face, vivi+1 is in the configuration G2, [vivi+1x] is a 3-face and x is a 44-vertex, then we say that f is adjacent to a configuration G2 by vivi+1. Let t denote the number of the configuration G2 which f is adjacent to.

    Claim 6. Suppose that f is a 7+-face in H, then t2dH(f)3.

    Proof. Let f=[v1v2v3vn] be a 7+-face in H.

    If f is adjacent to a configuration G2 by vivi+1, then vi or vi+1 is a internal (3,3,7+,3,7+)-vertex. By symmetry, let vi be a internal (3,3,7+,3,7+)-vertex. If f is adjacent to another G2 by edge vi1vi, then vi would be a (3,3,3,3,7+)-vertex, contradicting the definition of G2. Hence, if f is adjacent to a G2 by edge vivi+1 at least one of vi1vi or vi+1vi+2 is not adjacent to G2. It means that the three consecutive edges on f are adjacent to at most two G2 configurations. Consequently, t2dH(f)3.

    Case 3. dH(f)7.

    Suppose that dH(f) = 7. By Claim 6, t2dH(f)3=4, then w(f)760.25×4=0 by (R5). Suppose that dH(f)8, then w(f)dH(f)60.25×2dH(f)356dH(f)60 by (R5).

    This completes the proof of Lemma 4.

    Lemma 5. wH(v)0 for each vV0(H).

    Proof. Since v is a vertex of G and δ(G)4, it follows that dH(v)=dG(v)4. We have six cases to consider.

    Case 1. dH(v)=4.

    Case 1.1. Suppose that v is x1 (or x2) in the configuration G1.

    In the configuration G1, we can derive that m+5(v)2 and m3(v)1, otherwise it will produce a separating 3-cycle with fewer internal vertices than T or a configuration C5C6, a contradiction. Hence, w(v)20.50.10.2×21=0.

    Case 1.2. Suppose that v is not x1 (or x2) in the construction G1.

    Assume that m3(v)=0, then w(v)20.5×4=0 by (R1) and (R2).

    Assume that m3(v)=1. If m4(v)+m5(v)2, then w(v)210.5×2=0 by (R1), (R2) and (R3.1). Suppose that m4(v)+m5(v)=3. If m5(v)2, then w(v)210.50.2×2=0.1 by (R1), (R2) and (R3.1). Suppose that m5(v)=1 and m4(v)=2. If v is a (4,3,5,4)-vertex, then it will produce a separating 3-cycle with fewer internal vertices than T or a configuration C5C6 at the vertex v, a contradiction. If v is a (4,3,4,5)-vertex, then it is the configuration G1. Thus, w(v)210.5×20.2+0.1×2=0 by (R1), (R2), (R3.1) and (R4). If m4(v)=3, then it will produce a separating 3-cycle or a configuration C5C6, a contradiction.

    Assume that m3(v)=2. If dH(f0)=dH(f1)=3, then dH(f2)6 and dH(f3)6 by Claim 2. Thus, w(v)21×2=0 by (R3.1). If dH(f0)=dH(f2)=3, then dH(f1)6 and dH(f3)6 by Claim 4. Thus, w(v)21×2=0 by (R3.1).

    Assume that m3(v)=3. Suppose that dH(f0)=dH(f1)=dH(f2)=3, then dH(f3)7 by Claim 3. By Lemma 1, at least one of v0,v1,v2,v3 is an internal 5+-vertex or belongs to V(T). If v0 or v3 is an internal 5+-vertex or belongs to V(T), say v0, then one of v1,v2,v3 is an internal 5+-vertex or belongs to V(T) by Lemma 1 and v is incident with at most one bad 3-face. If v1 or v2 is an internal 5+-vertex or belongs to V(T), then v is incident with at most one bad 3-face. Thus, w(v)210.5×2=0, or w(v)20.50.75×2=0, or w(v)20.5×3=0.5 by (R3.1).

    Assume that m3(v)=4. If v is not incident with any bad 3-face, then w(v)20.5×4=0 (R3.1). Otherwise, v is incident with bad 3-faces. By Lemma 1, at least two of v0,v1,v2,v3 are internal 5+-vertices or belong to V(T), it means that v is incident with one bad 3-face, one good 3-face and two weak 3-faces. Thus, w(v)210.5×2=0 by (R3.1).

    Case 2. dH(v)=5.

    Note that w(v)=4. By Claim 1, m3(v)34dH(v)=3. If m6+(v)3, then w(v)41.5×2=0. Assume that m6+(v)2. If m3(v)=0, then w(v)40.5×50.1×5=1 by (R1)-(R4). If m3(v)=1, then w(v)41.50.5×40.1×4=0.1 by (R1)-(R4). Suppose that m3(v)=2. If dH(f0)=dH(f1)=3, then dH(f2)6 and dH(f4)6 by Claim 2. Thus, w(v)41.5×20.50.1=0.4 by (R1)-(R4). If dH(f0)=dH(f2)=3, then dH(f1)6 by Claim 4. If dH(f3)=dH(f4)=4, it will produce a separating 3-cycle with fewer internal vertices than T or a configuration C5C6 at the vertex v, a contradiction. Thus w(v)41.5×20.50.10.2=0.2 by (R1)-(R4).

    Suppose that m3(v)=3. If dH(f0)=dH(f1)=dH(f2)=3, then d(f3)7 and d(f4)7 by Claim 3. Thus w(v)41.25×3=0.25 by (R3.2). If dH(f0)=dH(f1)=dH(f3)=3, then dH(f2)6 and dH(f4)6 by Claim 2. Thus, w(v)41.25×21.5=0 by (R3.2).

    Case 3. dH(v)=6.

    Note that w(v)=6. By Claim 1, m3(v)34dH(v)=4. If m6+(v)2, then w(v)61.5×4=0 by (R3.3).

    Suppose that m6+(v)1. If m3(v)2, then w(v)61.5×20.5×40.1×4=0.6 by (R1)-(R4). Assume that m3(v)=3. If dH(f0)=dH(f1)=dH(f2)=3, then dH(f3)7 and dH(f5)7 by Claim 3, a contradiction. If dH(f0)=dH(f1)=dH(f3)=3, then dH(f2)6 and dH(f5)6 by Claim 2, a contradiction. If dH(f0)=dH(f2)=dH(f4)=3, then dH(f1)6, dH(f3)6 and dH(f5)6 by Claim 4. Assume that m3(v)=4. If dH(f0)=dH(f1)=dH(f2)=dH(f4)=3, then dH(f3)7 and dH(f5)7 by Claim 3, a contradiction. If dH(f0)=dH(f1)=dH(f3)=dH(f4)=3, then dH(f2)6 and dH(f5)6 by Claim 2, a contradiction.

    Case 4. dH(v)=7.

    Note that w(v)=8. By Claim 1, m3(v)34dH(v)=5. If m6+(v)2, then w(v)81.5×5=0.5 by (R3.3). Suppose that m6+(v)1. If m3(v)4, then w(v)81.5×40.5×30.1×3=0.2 by (R1)-(R4). Assume that m3(v)=5. If dH(f0)=dH(f1)=dH(f2)=3, then dH(f3)7 and dH(f6)7 by Claim 3, a contradiction. If dH(f0)=dH(f1)=3, then dH(f2)6 and dH(f6)6 by Claim 2, a contradiction.

    Case 5. dH(v)=8.

    Note that w(v)=10. By Claim 1, m3(v)34dH(v)=6. If m6+(v)2, then w(v)101.5×6=1 by (R3.3). Suppose that m6+(v)1. If m3(v)5, then w(v)101.5×50.5×30.1×3=0.7 by (R1)-(R4). Assume that m3(v)=6. If dH(f0)=dH(f1)=dH(f2)=3, then dH(f3)7 and dH(f7)7 by Claim 3, a contradiction. If dH(f0)=dH(f1)=3, then dH(f2)6 and dH(f7)6 by Claim 2, a contradiction.

    Case 6. dH(v)9.

    We have the following estimation:

    w(v)2dH(v)61.5m3(v)0.5(dH(v)m3(v))0.1(dH(v)m3(v))1.4dH(v)0.9m3(v)61.4dH(v)0.9×3dH(v)4629dH(v)406>0

    This completes the proof of Lemma 5.

    Lemma 6. w(fΔ)+ΣxV(T)w(x)11.6.

    Proof. By (R0), w(fΔ)36=3. Let vV(T). First assume that dH(v)=2. Note that v is not be incident with any 3-face or 4-face for otherwise we can find another separating 3-cycle with fewer internal vertices than T, which is impossible. Consequently, w(v)2×260.2=2.2 by (R0). Next assume that dH(v)=3. If m3(v)=3, then w(v)2×361.5×2=3 by (R0). Otherwise, w(v)2×3620.6=2.6 by (R0). Now assume that dH(v)=4. If m3(v)=4, then w(v)2×461.5×3=2.5 by (R0). Otherwise, w(v)2×462×20.6=2.6 by (R0). Finally assume that dH(v)5. It follows that m3(v)dH(v)1, otherwise a configuration C5C6 at the vertex v will be found. Thus, w(v)2×dH(v)62×(dH(v)2)0.6=2.6 by (R0).

    If fΔ is incident with a vertex that is not a 33-vertex, then w(fΔ)+ΣxV(T)w(x)3332.6=11.6. Otherwise there is an internal 3-vertex, which is impossible.

    Recall that the sum of the initial weights of vertices and faces of H equals 12. From Lemma 4-6, the corresponding sum of the new weights satisfies ΣxV0(H)F0(H)w(x)+w(fΔ)+ΣxV(T)w(x)11.6. Since the total weight of H has been preserved during the discharging process, this is a contradiction to our initial assumptions that G, the minimal counterexample to Theorem 1, exists. This completes the proof of Theorem 1.

    In this paper, we show that the list vertex arboricity of every planar graph without a 5-cycle intersecting with a 6-cycle is at most 2. This implies directly that if G is a planar graph without 5-cycles or without 6-cycles, then al(G)2. We like to conclude this paper by raising the following conjectures:

    Conjecture 1. Every planar graph G without intersecting 4-cycles has al(G)2.

    Conjecture 2. If G is a planar graph without a vertex lying on all cycles of length from 3 to 7, then al(G)2.

    The authors would like to thank the anonymous referees for their valuable comments and suggestions, which led to considerable improvement of the article.

    The research is supported by the Natural Science Foundation of China (Grant No. 12031018).

    The authors declare that they have no competing interests.



    [1] O. V. Borodin, A. O. Ivanova, Planar graphs without 4-cycles adjacent to 3-cycles are list vertex 2-arborable, J. Graph Theory, 62 (2009), 234–240. doi: 10.1002/jgt.20394
    [2] G. Chartrand, H. V. Kronk, C. E. Wall, The point-arboricity of a graph, Israel J. Math., 6 (1968), 169–175. doi: 10.1007/BF02760181
    [3] M. Chen, A. Raspaud, W. F. Wang, Vertex-arboricity of planar graphs without intersecting triangles, Eur. J. Combin., 33 (2012), 905–923. doi: 10.1016/j.ejc.2011.09.017
    [4] M. Chen, L. Huang, W. F. Wang, List vertex-arboricity of toroidal graphs without 4-cycles adjacent to 3-cycles, Discrete Math., 339 (2016), 2526–2535. doi: 10.1016/j.disc.2016.04.010
    [5] I. Choi, H. H. Zhang, Vertex arboricity of toroidal graphs with a forbidden cycle, Discrete Math., 333 (2014), 101–105. doi: 10.1016/j.disc.2014.06.011
    [6] S. L. Hakimi, E. F. Schmeichel, A note on the vertex arboricity of a graph, SIAM J. Discrete Math., 2 (1989), 64–67. doi: 10.1137/0402007
    [7] D. J. Huang, W. C. Shiu, W. F. Wang, On the vertex-arboricity of planar graphs without 7-cycles, Discrete Math., 312 (2012), 2304–2315. doi: 10.1016/j.disc.2012.03.035
    [8] L. Huang, M. Chen, W. F. Wang, Toroidal graphs without 3-cycles adjacent to 5-cycles have list vertex-arboricity at most 2, Int. J. Math. Stat., 16 (2015), 97–105.
    [9] W. F. Wang, L. Huang, M. Chen, List vertex-arboricity of planar graphs without intersecting 5-cycles, Acta Math. Appl. Sin. Engl. Ser., 36 (2020), 439–447. doi: 10.1007/s10255-020-0936-1
    [10] A. Raspaud, W. F. Wang, On the vertex-arboricty of planar graphs, Eur. J. Combin., 29 (2008), 1064–1075. doi: 10.1016/j.ejc.2007.11.022
    [11] Y. Q. Wang, M. Chen, W. F. Wang, A note on the list vertex arboricity of toroidal graphs, Discrete Math., 341 (2018), 3344–3347. doi: 10.1016/j.disc.2018.08.021
    [12] Y. P. Yang, Y. Q. Wang, P. Wang, W. F. Wang, List vertex arboricity of planar graphs of diameter two, Adv. Math. (China), 50 (2021), 335–344.
    [13] A. F. Yang, J. J. Yuan, On the vertex arboricity of planar graphs of diameter two, Discrete Math., 307 (2007), 2438–2447. doi: 10.1016/j.disc.2006.10.017
    [14] H. Zhang, On list vertex 2-arboricity of toroidal graphs without cycles of specific length, Bull. Iranian Math. Soc., 42 (2016), 1293–1303.
  • Reader Comments
  • © 2021 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(2770) PDF downloads(102) Cited by(0)

Other Articles By Authors

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog