1 Introduction

All graphs considered in this paper are simple, finite and undirected, and we follow [2] for terminologies and notations not defined here. Let \(G\) be a graph. We use \(V(G)\), \(E(G)\), \(\Delta (G)\) and \(\delta (G)\) (or simply \(V\), \(E\), \(\Delta \) and \(\delta \)) to denote the vertex set, the edge set, the maximum degree and the minimum degree of \(G\), respectively. For a vertex \(v\in V\), let \(N(v)\) denote the set of vertices adjacent to \(v\) and let \(d(v)=|N(v)|\) denote the degree of \(v\). A \(k\) -vertex, a \(k^{-}\) -vertex or a \(k^{+}\) -vertex is a vertex of degree \(k\), at most \(k\) or at least \(k\), respectively. A \(k\) -cycle is a cycle of length \(k\). We use \((v_{1},v_{2},\ldots ,v_{d})\) to denote a cycle (or a face) whose boundary vertices are \(v_{1},v_{2},\ldots ,v_{d}\) in the clockwise order. Note that all the subscripts in the paper are taken modulo \(d\). We say that two cycles are adjacent (or intersecting) if they share at least one edge (or one vertex, respectively). Let \(C=(v_1,v_2,\ldots ,v_k)(k\ge 4)\) be a cycle. If there is an edge \(v_iv_j\) with \(j\not \equiv i\pm 1\pmod {k}\), then the edge \(v_iv_j\) is called a \(chord\) of \(C\).

A \(k\) -total-coloring of a graph \(G=(V,E)\) is a coloring of \(V\cup {E}\) using \(k\) colors such that no two adjacent or incident elements receive the same color. A graph \(G\) is total- \(k\) -colorable if it admits a \(k\)-total-coloring. The total chromatic number \(\chi ^{\prime }(G)\) of \(G\) is the smallest integer \(k\) such that \(G\) has a \(k\)-total-coloring. Clearly, \(\chi ''(G)\ge \Delta +1\). Behzad [1] and Vizing [11] conjectured independently that \(\chi ''(G)\le \Delta +2\) for each graph \(G\). This conjecture was confirmed for graphs with \(\Delta \le 5\). For planar graphs the only open case is that of \(\Delta =6\) (see [6, 9]). In recent years, the study of total colorings planar graphs has attracted considerable attention. For planar graphs with large maximum degree, it is possible to determine \(\chi ''(G)=\Delta +1\). This first result was given in [3] for \(\Delta \ge 14\), which was finally extended to \(\Delta \ge 9\) in [7]. Zhu [8] proved that if \(G\) is a planar graph with maximum degree 8, and for each vertex \(x\), there is an integer \(k_x \in \{3, 4, 5, 6, 7, 8\}\) such that there is no \(k_x\)-cycle which contains \(x\), then \(\chi ''(G)=9\). Wang et al. [13] extended this result to that there is at most one \(k_x\)-cycle which contains \(x\). Chang [4] proved that for planar graph \(G\) with \(\Delta \ge 7\), if there is an integer \(k_x\in \{3,4,5,6\}\) such that there is no \(k_x\)-cycle which contains \(x\) for each \(x\in V(G)\), then \(\chi ''(G)=\Delta +1\). Wang et al. [12] proved \(\chi ''(G)=\Delta +1\) for some planar graphs with small maximum degree. Hou et al. [5] proved that every planar graphs with \(\Delta \ge 8\) and without 6-cycles are total-9-colorable. Shen and Wang [10] extended this result to planar graphs without chordal 6-cycles. In this paper, we extend this result and get the following theorem.

Theorem 1

Let G be a planar graph with maximum degree \(\Delta \ge 8\). If every \(6\)-cycle of \(G\) contains at most one chord or chordal \(6\)-cycles are not adjacent in \(G\), then \(\chi ''(G)=\Delta +1\).

2 Proof of Theorem 1

First, we introduce additional notations and definitions here for convenience. Let \(G\) be a planar graph having a plane drawing and let \(F\) be the face set of \(G\). For a face \(f\) of \(G\), the degree \(d(f)\) is the number of edges incident with it, where each cut-edge is counted twice. A \(k\) -face, a \(k^{-}\) -face or a \(k^{+}\) -face is a face of degree \(k\), at most \(k\) or at least \(k\), respectively. Denote by \(n_{d}(v)\) the number of \(d\)-vertices adjacent to the vertex \(v\), \(f_{d}(v)\) the number of \(d\)-faces incident with \(v\).

Now, we begin to prove Theorem 1. According to [7], the theorem is true for \(\Delta \ge 9\). So we assume in the following that \(\Delta =8\). Let \(G=(V,E)\) be a minimal counterexample to the planar graph \(G\) with maximum degree \(\Delta =8\), such that \(|V|+|E|\) is minimal and \(G\) has been embedded in the plane. Then every proper subgraph of \(G\) is total-9-colorable. First we give some lemmas for \(G\).

Lemma 1

[3]

  1. (a)

    \(G\) is \(2\)-connected.

  2. (b)

    If \(uv\) is an edge of \(G\) with \(d(u)\le 4\), then \(d(u)+d(v)\ge \Delta +2=10\).

By Lemma 1(b), any two neighbors of a 2-vertex are 8-vertices.

Note that in all figures of the paper, vertices marked \(\bullet \) have no edges of \(G\) incident with them other than those shown and vertices marked \(\circ \) are \(3^+\)-vertices.

Lemma 2

\(G\) has no configurations depicted in Fig. 1, where v denotes the vertex of degree of 7.

Fig. 1
figure 1

Forbidden configurations in G

Proof

The proof of (1), (2), (4) and (6) can be found in [14], (3) can be found in [10], (5) can be found in [7]. \(\square \)

Lemma 3

Suppose \(v\) is a \(d\)-vertex of \(G\) with \(d\ge 5\). Let \(v_{1},\ldots ,v_{d}\) be the neighbor of \(v\) and \(f_{1},f_{2},\ldots ,f_{d}\) be faces incident with \(v\), such that \(f_i\) is incident with \(v_i\) and \(v_{i+1}\), for \(i\in \{1,2,\ldots ,d\}\). Let \(d(v_1)=2\) and \(\{v,u_1\}=N(v_1)\). Then \(G\) does not satisfy one of the following conditions (see Fig. 2).

  1. (1)

    there exists an integer \(k\) \((2\le k\le d-1)\) such that \(d(v_{k+1})=2\), \(d(v_i)=3\) \((2\le i\le k)\) and \(d(f_j)=4\) \((1\le j\le k)\).

  2. (2)

    there exist two integers \(k\) and \(t\) \((2\le k<t\le d-1)\) such that \(d(v_{k})=2\), \(d(v_i)=3\) \((k+1\le i\le t)\), \(d(f_t)=3\) and \(d(f_j)=4\) \((k\le j\le t-1)\).

  3. (3)

    there exist two integers \(k\) and \(t\) \((3\le k\le t\le d-1)\) such that \(d(v_i)=3\) \((k\le i\le t)\), \(d(f_{k-1})=d(f_t)=3\) and \(d(f_j)=4\) \((k\le j\le t-1)\).

Fig. 2
figure 2

Forbidden configurations in G

Proof

Suppose \(G\) satisfies all of the conditions (1)-(3). If \(d(f_i)=4\), then let \(u_i\) be adjacent to \(v_i\) and \(v_{i+1}\). By the minimality of \(G\), \(G'=G-vv_1\) has a \((\Delta +1)\)-total-coloring \(\phi \). Let \(C(x)=\{\phi (xy): y\in N(x)\}\cup \{\phi (x)\}\). First we erase the colors on all \(3^{-}\)-vertices adjacent to \(v\). We have \(\phi (v_1u_1)\not \in C(v)\), for otherwise, the number of the forbidden colors for \(vv_1\) is at most \(\Delta \), so \(vv_1\) can be properly colored and by properly recoloring the erased vertices, we get a \((\Delta +1)\)-total-coloring of \(G\), a contradiction. Without loss of generality, assume that \(C(v)=\{1,2,\ldots ,d\}\) with \(\phi (vv_i)=i\) \((i\in \{2,3,\ldots ,d\})\), \(\phi (v_1u_1)=d+1\) and \(\phi (v)=1\). Thus we have \(d+1\in C(v_i)\) for all \(i\in \{2,3,\ldots ,d\}\), for otherwise, we can recolor \(vv_i\) with \(d+1\) and color \(vv_1\) with \(i\), and by properly recoloring the erased vertices, we get a \((\Delta +1)\)-total-coloring of \(G\), a contradiction, too. In the following we consider (1)-(3) one by one.

  1. (1)

    Since \(d+1\in C(v_i)\) for all \(i\in \{2,3,\ldots ,d\}\), there is a vertex \(u_s\) \((1\le s \le k)\) such that \(d+1\) appears at least twice on \(u_s\), a contradiction to \(\phi \).

  2. (2)

    Since \(d+1\in C(v_i)\) for all \(i\in \{2,3,\ldots ,d\}\), \(\phi (v_ku_k)=\phi (v_{k+1}u_{k+1})=\cdots =\phi (v_{t-1}u_{t-1})=\phi (v_{t}v_{t+1})=d+1\). We also have \(\phi (v_{t}u_{t-1})=t+1\). For otherwise, we can recolor \(v_tv_{t+1}\) with \(t+1\), \(vv_{t+1}\) with \(d+1\) and color \(vv_1\) with \(t+1\). By properly recoloring the erased vertices, we get a \((\Delta +1)\)-total-coloring of \(G\), a contradiction. Similarly, \(\phi (v_{t-1}u_{t-2})=\phi (v_{t-2}u_{t-3})=\cdots =\phi (v_{k+1}u_{k})=t+1\). So we can recolor \(vv_{t+1}\) with \(d+1\), \(v_{t}v_{t+1}\) with \(t+1\), \(v_{t}u_{t-1}\) with \(d+1\), \(v_{t-1}u_{t-1}\) with \(t+1\),\(\ldots \), \(v_{k+1}u_{k+1}\) with \(t+1\), \(v_{k+1}u_{k}\) with \(d+1\), \(v_{k}u_{k}\) with \(t+1\) and color \(vv_1\) with \(t+1\). By properly recoloring the erased vertices, we get a \((\Delta +1)\)-total-coloring of \(G\), also a contradiction.

  3. (3)

    If \(d+1\not \in \{\phi (v_{k-1}v_k)\cup \phi (v_{t}v_{t+1})\}\), then there is a vertex \(u_s\) \((k\le s \le t-1)\) such that \(d+1\) appears at least twice on \(u_s\), a contradiction to \(\phi \). So without loss of generality, assume \(\phi (v_{k-1}v_k)=d+1\). If \(\phi (v_{k+1}u_k)=d+1\), then \(\phi (v_{k+2}u_{k+1})=\phi (v_{k+3}u_{k+2})=\cdots =\phi (v_{t}u_{t-1})=d+1\). By the discussion of (2), we also have \(\phi (v_{k}u_{k})=\phi (v_{k+1}u_{k+1})=\cdots =\phi (v_{t-1}u_{t-1})=\phi (v_{t}v_{t+1})=k-1\). Then we can recolor \(vv_{k-1}\) with \(d+1\), \(v_{k-1}v_k\) with \(k-1\), \(v_{k}u_k\) with \(d+1\), \(v_{k+1}u_k\) with \(k-1\), \(\ldots \), \(v_{t-1}u_{t-1}\) with \(d+1\), \(v_{t}u_{t-1}\) with \(k-1\), \(v_{t}v_{t+1}\) with \(t+1\), \(vv_{t+1}\) with \(k-1\) and color \(vv_1\) with \(t+1\). By properly recoloring the erased vertices, we get a \((\Delta +1)\)-total-coloring of \(G\), a contradiction. If \(\phi (v_{k+1}u_{k+1})=d+1\), then \(\phi (v_{k+2}u_{k+2})=\phi (v_{k+3}u_{k+3})=\cdots =\phi (v_{t-1}u_{t-1})=\phi (v_{t}v_{t+1})=d+1\). Similarly, we have \(\phi (v_{t}u_{t-1})=\phi (v_{t-1}u_{t-2})=\cdots =\phi (v_{k+1}u_{k})=t+1\). Let \(\phi (v_{k}u_{k})=s\). Then we can recolor \(vv_{t+1}\) with \(d+1\), \(v_{t}v_{t+1}\) with \(t+1\), \(v_{t}u_{t-1}\) with \(d+1\), \(v_{t-1}u_{t-1}\) with \(t+1\), \(\ldots \), \(v_{k+1}u_{k+1}\) with \(t+1\), \(v_{k+1}u_{k}\) with \(s\), \(v_{k}u_{k}\) with \(t+1\), and color \(vv_1\) with \(t+1\). By properly recoloring the erased vertices, we get a \((\Delta +1)\)-total-coloring of \(G\), a contradiction, too.\(\square \)

By the Euler’s formula \(|V|-|E|+|F|=2\), we have

$$\begin{aligned} \sum _{v\in V}(2d(v)-6)+\sum _{f\in F}(d(f)-6)=-6(|V|-|E|+|F|)=-12<0 \end{aligned}$$

We define \(ch\) the initial charge that \(ch(x)=2d(x)-6\) for each \(x\in V\) and \(ch(x)=d(x)-6\) for each \(x\in F\). So \(\sum _{x\in V\cup F}ch(x)=-12<0.\) In the following, we will reassign a new charge denoted by \(ch^{\prime }(x)\) to each \(x\in V \cup F\) according to the discharging rules. If we can show that \(ch^{\prime }(x)\ge 0\) for each \(x\in V \cup F\), then we get an obvious contradiction to \(0\le \sum _{x\in V\cup F}ch^{\prime }(x)=\sum _{x\in V\cup F}ch(x)=-12\), which completes our proof. Now we define the discharging rules as follows.

  • R1. Each 2-vertex receives 1 from each of its neighbors.

  • R2. Let \(f\) be a \(3\)-face. If \(f\) is incident with a \(3^-\)-vertex, then it receives \(\frac{3}{2}\) from each of its two incident \(7^+\)-vertices. If \(f\) is incident with a \(4\)-vertex, then it receives \(\frac{5}{4}\) from each of its two incident \(6^+\)-vertices. If \(f\) is not incident with any \(4^-\)-vertex, then it receives 1 from each of its two incident \(5^+\)-vertices.

  • R3. Let \(f\) be a \(4\)-face. If \(f\) is incident with two \(3^-\)-vertices, then it receives 1 from each of its two incident \(7^+\)-vertices. If \(f\) is incident with only one \(3^-\)-vertex, then it receives \(\frac{3}{4}\) from each of its two incident \(7^+\)-vertices; and \(\frac{1}{2}\) from the left incident \(4^+\)-vertex. If \(f\) is not incident with any \(3^-\)-vertex, then it receives \(\frac{1}{2}\) from each of its incident \(4^+\)-vertices.

  • R4. Each \(5\)-face receives \(\frac{1}{3}\) from each of its incident \(4^+\)-vertices.

Next, we show that \(ch^{'}(x)\ge 0\) for all \(x\in V \cup F\). It is easy to check that \(ch^{'}(f)\ge 0\) for all \(f\in F\) and \(ch^{'}(v)\ge 0\) for all 2-vertices \(v\in V\) by the above discharging rules. If \(d(v)=3\), then \(ch^{'}(v)=ch(v)=0\). If \(d(v)=4\), then \(ch^{'}(v)\ge ch(v)-\frac{1}{2}\times 4=0\) by R2 and R3. For \(d(v)\ge 5\), we need the following structural lemma.

Lemma 4

  1. (1)

    Suppose that every \(6\)-cycle of \(G\) contains at most one chord. Then we have the following results.

    1. (a)

      \(G\) has no configurations depicted in Fig. 3(1), Fig. 3(2) and Fig. 3(3);

    2. (b)

      Suppose \(G\) has a subgraph isomorphic to Fig. 3(4). Then \(d(f_1)\ge 4\) and \(d(f_2)\ne 4\). More over if \(d(f_1)=4\), then \(d(f_2)\ge 5\);

    3. (c)

      If \(G\) has a subgraph isomorphic to Fig. 3(5), then \(d(f_1)\ge 5\) and \(d(f_2)\ge 5\).

  2. (2)

    Suppose that all chordal \(6\)-cycles are not adjacent. Then we have the following results.

    1. (d)

      If \(G\) has a subgraph isomorphic to Fig. 3(5), then \(\max \{d(f_1),~d(f_2)\}\ge 4\);

    2. (e)

      \(G\) has no configurations depicted in Fig. 3(6) and Fig. 3(7).

Fig. 3
figure 3

Forbidden configurations in G

Suppose \(d(v)=5\). Then \(f_3(v)\le 4\) by Lemma 4. If \(f_3(v)=4\), then \(f_{6^+}(v)\ge 1\), so \(ch^{'}(v)\ge ch(v)-1\times 4=0\). If \(f_3(v)\le 3\), then \(ch'(v)\ge ch(v)-1\times f_3(v)-\frac{1}{2}\times (5-f_3(v))=\frac{3-f_3(v)}{2}\ge 0\). Suppose \(d(v)=6\). Then \(f_3(v)\le 4\) and \(ch^{'}(v)\ge ch(v)-\frac{5}{4}\times f_3(v)-\frac{1}{2}\times (6-f_3(v))=\frac{3(4-f_3(v))}{4}\ge 0\). Suppose \(d(v)=7\). Then \(f_3(v)\le 5\). By Lemma 2(1), \(v\) is incident with at most two \(3\)-faces are incident with a \(3^-\)-vertex, that is, \(v\) sends \(\frac{3}{2}\) to each of the two \(3\)-faces and at most \(\frac{5}{4}\) to each other \(3\)-face. If \(f_3(v)=5\), then \(f_{5^+}(v)\ge 1\), and \(ch'(v)\ge ch(v)-\frac{3}{2}\times 2-\frac{5}{4}\times 3-\frac{3}{4}\times 1-\frac{1}{3}\times 1=\frac{1}{6}>0\). If \(2\le f_3(v)\le 4\), then \(ch'(v)\ge ch(v)-\frac{3}{2}\times 2-\frac{5}{4}\times (f_3(v)-2)-1\times (5-f_3(v))-\frac{3}{4}\times 2=\frac{4-f_3(v)}{4}\ge 0\). If \(f_3(v)\le 2\), then \(ch^{'}(v)\ge ch(v)-\frac{3}{2}\times f_3(v)-1\times (7-f_3(v))=\frac{2-f_3(v)}{2}>0\).

Suppose \(d(v)=8\). Then \(ch(v)=10\). Let \(v_{1},\ldots ,v_{8}\) be neighbors of \(v\) in the clockwise order and \(f_{1},f_{2},\ldots ,f_{8}\) be faces incident with \(v\), such that \(f_i\) is incident with \(v_i\) and \(v_{i+1}\), for \(i\in \{1,2,\ldots ,8\}\), and \(f_{9}=f_1\).

Suppose \(n_2(v)=0\). Then \(f_3(v)\le 6\). If \(f_3(v)=6\), then \(f_{5^+}(v)\ge 2\), so \(ch'(v)\ge 10-\frac{3}{2}\times 6-\frac{1}{3}\times 2=\frac{1}{3}>0\). If \(f_3(v)=5\), then \(f_{5^+}(v)\ge 1\), so \(ch'(v)\ge 10-\frac{3}{2}\times 5-1\times 2-\frac{1}{3}\times 1=\frac{1}{6}>0\). If \(f_3(v)\le 4\), then \(ch'(v)\ge 10-\frac{3}{2}\times f_3(v)-1\times (8-f_3(v))\ge 0\).

Suppose \(n_2(v)=1\). Without loss of generality, assume \(d(v_1)=2\).

Suppose \(v_1\) is incident with a 3-cycle \(f_1\).

By Lemma 4, \(f_3(v)\le 6\) and all \(3\)-faces incident with no \(3^-\)-vertex except \(f_1\) by Lemma 2(6). If \(f_3(v)=6\), then \(f_{5^+}(v)\ge 2\), so \(ch'(v)\ge 10-1-\frac{3}{2}\times 1-\frac{5}{4}\times 5-\frac{1}{3}\times 2=\frac{7}{12}>0\). If \(4\le f_3(v)\le 5\), then \(ch'(v)\ge 10-1-\frac{3}{2}\times 1-\frac{5}{4}\times (f_3(v)-1)-1\times (6-f_3(v))-\frac{3}{4}\times 2=\frac{5-f_3(v)}{4}\ge 0\). If \(1\le f_3(v)\le 3\), then \(ch'(v)\ge 10-1-\frac{3}{2}\times 1-\frac{5}{4}\times (f_3(v)-1)-1\times (8-f_3(v))=\frac{3-f_3(v)}{4}\ge 0\).

Suppose \(v_1\) is not incident with a \(3\)-cycle.

Suppose every \(6\)-cycle of \(G\) contains at most one chord. Then \(f_3(v)\le 5\) by Lemma 2(2)–(4). If \(4\le f_3(v)\le 5\), then \(f_{5^+}(v)\ge 2\), so \(ch'(v)\ge 10-1-\frac{3}{2}\times (f_3(v)-1)-1\times 1-1\times (6-f_3(v))-\frac{1}{3}\times 2=\frac{17-3f_3(v)}{6}>0\). If \(f_3(v)=3\), then \(f_{5^+}(v)\ge 1\), so \(ch'(v)\ge 10-1-\frac{3}{2}\times 3-1\times 4-\frac{1}{3}\times 1=\frac{1}{6}>0\). If \(f_3(v)=2\), then \(ch'(v)\ge 10-1-\frac{3}{2}\times 2-1\times 6=0\). If \(f_3(v)=1\), then without loss of generality, \(d(f_2)=3\), i.e. \(d(v_3)=3\) and \(d(v_2)\ge 7\), so \(ch'(v)\ge 10-1-\frac{3}{2}\times 1-1\times 6-\frac{3}{4}\times 1=\frac{3}{4}>0\). If \(f_3(v)=0\), then \(ch'(v)\ge 10-1-1\times 8=1>0\).

Suppose any two chordal \(6\)-cycles are not adjacent. Then \(f_3(v)\le 5\) by Lemma 2(2)–(4). If \(f_3(v)\ge 4\), then \(ch'(v)\ge 10-1-\frac{3}{2}\times 2-\frac{5}{4}\times (f_3(v))-\frac{3}{4}\times (8-f_3(v))=\frac{5-f_3(v)}{2}\ge 0\). If \(f_3(v)=3\), then \(ch'(v)\ge 10-1-\frac{3}{2}\times 3-\frac{3}{4}\times 5=\frac{3}{4}>0\). If \(1\le f_3(v)\le 2\), then \(ch'(v)\ge 10-1-\frac{3}{2}\times f_3(v)-1\times (6-2f_3(v))-\frac{3}{4}\times (2+f_3(v))=\frac{6-f_3(v)}{4}>0\). If \(f_3(v)=0\), then \(ch'(v)\ge 10-1-1\times 8=1>0\).

Note that next Lemma 5 is also true for general planar graphs if we just use the above discharging rules.

Lemma 5

Suppose \(d(v)=8\) and \(2\le n_{2}(v)\le 8\). Then \(ch'(v)\ge 0\).

Proof

Since \(d(v)=8\), then \(ch(v)=10\). First we give a Claim for convenience.

Claim

Suppose that \(d(v_{i})=d(v_{i+k+1})=2\) and \(d(v_{j})\ge 3\) for \(i+1\le j\le i+k\). Then \(v\) sends at most \(\phi \) (in total) to \(f_{i}\) and \(f_{i+1}\), \(f_{i+2}\), \(\ldots \), \(f_{i+k}\), where \(\phi =\frac{5k+1}{4}\) \((k=1,2,3,4,5)\), see Fig. 4.

Fig. 4
figure 4

Claim

By Lemma 2, \(d(f_i)\ge 4\) and \(d(f_{i+k})\ge 4\).

Case 1. \(k=1\) By Lemma 3(1), we have \(d(v_{i+1})\ge 4\) or \(\max \{d(f_i),~d(f_{i+1})\}\ge 5\), so \(\phi \le \max \{\frac{3}{4}\times 2,~1+\frac{1}{3}\}=\frac{3}{2}\) by R3.

Case 2. \(k=2\) If \(d(f_{i+1})=3\), then \(\min \{d(v_{i+1}),~d(v_{i+2})\}\ge 4\) or \(\max \{d(f_i),~d(f_{i+2})\}\ge 5\) by Lemma 3(2), and it follows that \(\phi \le \max \{\frac{3}{4}+\frac{5}{4}+\frac{3}{4},~\frac{1}{3}+\frac{3}{2}+\frac{3}{4}\}=\frac{11}{4}\). Otherwise, \(d(f_{i+1})\ge 4\), then \(\min \{d(v_{i+1}),~d(v_{i+2})\}\ge 4\) or \(\max \{d(f_i),~d(f_{i+1}),~d(f_{i+2})\}\ge 5\) by Lemma 3(1), and it follows that \(\phi \le \max \{1+\frac{3}{4}\times 2,~1\times 2+\frac{1}{3}\}=\frac{5}{2}<\frac{11}{4}\).

Case 3. \(k=3\) Suppose \(d(f_{i+1})=d(f_{i+2})=3\). Then \(d(v_{i+2})\ge 4\). If \(d(v_{i+1})=d(v_{i+3})=3\), then \(d(f_{i})\ge 5\) and \(d(f_{i+3})\ge 5\), so \(\phi \le \frac{3}{2}\times 2+\frac{1}{3}\times 2=\frac{11}{3}\). If \(\min \{d(v_{i+1}),~d(v_{i+3})\}\ge 4\), then \(\phi \le \frac{5}{4}\times 2+\frac{3}{4}\times 2=4\). Suppose \(d(f_{i+1})=3\) and \(d(f_{i+2})\ge 4\). If \(d(v_{i+1})=3\), then \(d(v_{i+2})\ge 7\) and \(d(f_{i})\ge 5\), so \(\phi \le \frac{1}{3}+\frac{3}{2}+\frac{3}{4}+1=\frac{43}{12}\). If \(d(v_{i+2})=3\), then \(d(v_{i+1})\ge 7\) and \(d(v_{i+3})\ge 4\), so \(\phi \le \frac{3}{4}+\frac{3}{2}+\frac{3}{4}+\frac{3}{4}=\frac{15}{4}\). If \(\min \{d(v_{i+1}),~d(v_{i+2})\}\ge 4\), \(\phi \le \frac{3}{4}+\frac{5}{4}+\frac{3}{4}+1=\frac{15}{4}\). It is similar with \(d(f_{i+2})=3\) and \(d(f_{i+1})\ge 4\). Suppose \(\min \{d(f_{i+1}),~d(f_{i+2})\}\ge 4\). Then \(\max \{d(v_{i+1}),~d(v_{i+2}),~d(v_{i+3})\}\ge 4\) or \(\max \{d(f_i),~d(f_{i+1}),~d(f_{i+2}),~d(f_{i+3})\}\ge 5\), so \(\phi \le \max \{1\times 2+\frac{3}{4}\times 2,~1\times 3+\frac{1}{3}\}=\frac{7}{2}\). So \(\phi \le \max \{\frac{11}{3},~4,~\frac{43}{12},~\frac{15}{4},~\frac{7}{2}\}=4\).

Case 4. \(k=4\) Suppose \(d(f_{i+1})=d(f_{i+2})=d(f_{i+3})=3\). Then \(\min \{d(v_{i+2}),~d(v_{i+3})\}\ge 4\). If \(d(v_{i+1})=d(v_{i+4})=3\), then \(d(f_{i})\ge 5\) and \(d(f_{i+4})\ge 5\), so \(\phi \le \frac{3}{2}\times 2+1\times 1+\frac{1}{3}\times 2=\frac{14}{3}\). If \(\min \{d(v_{i+1}),~d(v_{i+4})\}\ge 4\), then \(\phi \le \frac{5}{4}\times 3+\frac{3}{4}\times 2=\frac{21}{4}\). Suppose \(d(f_{i+1})=d(f_{i+2})=3\), \(d(f_{i+3})\ge 4\). Then \(d(v_{i+2})\ge 4\). If \(d(v_{i+1})=d(v_{i+3})=3\), then \(d(v_{i+4})\ge 4\) and \(d(f_{i})\ge 5\), so \(\phi \le \frac{3}{2}\times 2+\frac{3}{4}\times 2+\frac{1}{3}\times 1=\frac{29}{6}\). If \(\min \{d(v_{i+1}),~d(v_{i+3})\}\ge 4\), then \(\phi \le \frac{5}{4}\times 2+1\times 1+\frac{3}{4}\times 2=5\). Similar with \(d(f_{i+2})=d(f_{i+3})=3\), \(d(f_{i+1})\ge 4\). Suppose \(d(f_{i+1})=d(f_{i+3})=3\), \(d(f_{i+2})\ge 4\). Then \(\max \{d(v_{i+2}),~d(v_{i+3})\}\ge 4\) by Lemma 3(3), so \(\phi \le \frac{3}{2}\times 1+\frac{5}{4}\times 1+\frac{3}{4}\times 3=5\). Suppose \(d(f_{i+1})=3\), \(d(f_{i+2})\ge 4\) and \(d(f_{i+3})\ge 4\). If \(d(v_{i+1})=3\), then \(d(v_{i+2})\ge 7\) and \(d(f_{i})\ge 5\), so \(\phi \le \frac{3}{2}+1\times 2+\frac{3}{4}\times 1+\frac{1}{3}\times 1=\frac{55}{12}\). If \(d(v_{i+2})=3\), then \(d(v_{i+1})\ge 7\) and \(\max \{d(v_{i+3}),~d(v_{i+4})\}\ge 4\), so \(\phi \le \frac{3}{2}\times 1+1\times 1+\frac{3}{4}\times 3=\frac{19}{4}\). Otherwise, \(\phi \le \frac{5}{4}\times 1+1\times 2+\frac{3}{4}\times 2=\frac{19}{4}\). It is similar with \(d(f_{i+3})=3\), \(d(f_{i+1})\ge 4\) and \(d(f_{i+2})\ge 4\). Suppose \(d(f_{i+2})=3\), \(d(f_{i+1})\ge 4\) and \(d(f_{i+3})\ge 4\). If \(d(v_{i+2})=3\) or \(d(v_{i+3})=3\), then \(\phi \le \frac{3}{2}\times 1+1\times 1+\frac{3}{4}\times 3=\frac{19}{4}\). Otherwise, \(\phi \le \frac{5}{4}\times 1+1\times 2+\frac{3}{4}\times 2=\frac{19}{4}\). Suppose \(\min \{d(f_{i+1}),~d(f_{i+2}),~d(f_{i+3})\}\ge 4\). Then \(\max \{d(v_{i+1}),~d(v_{i+2}),~d(v_{i+3}),~d(v_{i+4})\}\ge 4\) or \(\max \{d(f_i),~d(f_{i+1}),~d(f_{i+2}),\) \(d(f_{i+3}),~d(f_{i+4})\}\ge 5\), so \(\phi \le \max \{1\times 3+\frac{3}{4}\times 2,~1\times 4+\frac{1}{3}\}=\frac{9}{2}\). So \(\phi \le \max \{\frac{14}{3},~\frac{21}{4},~\frac{29}{6},~5,~\frac{55}{12},~\frac{19}{4},~\frac{9}{2}\}=\frac{21}{4}\).

Case 5. \(k=5\) If \(k=5\), then \(\phi \le \frac{13}{2}\). It is similar to prove (e), we omit it here.

Next, we prove the Lemma.

If \(n_2(v)=8\), then all faces incident with \(v\) are \(6^+\)-faces by Lemma 2(2)–(4), that is, \(f_{6^+}(v)=8\), so \(ch^{'}(v)= 10-1\times 8=2>0\). If \(n_2(v)=7\), then \(f_{6^{+}}(v)\ge 6\) and \(f_3(v)=0\), so \(ch^{'}(v)\ge 10-1\times 7-\frac{3}{2}=\frac{3}{2}>0\) by Claim (a).

Fig. 5
figure 5

\(n_{2}(v)\le 6\)

Suppose \(n_2(v)\le 6\). The possible distributions of 2-vertices adjacent to \(v\) are illustrated in Fig. 5. For Fig. 5(1), we have \(f_{6^+}(v)\ge 5\) and \(ch'(v)\ge 10-1\times 6-\frac{11}{4}=\frac{5}{4}>0\) by Claim (b).

For Fig. 5(2)–(4), we have \(f_{6^+}(v)\ge 4\) and \(ch'(v)\ge 10-1\times 6-\frac{3}{2}\times 2=1>0\). For Fig. 5(5), we have \(f_{6^+}(v)\ge 4\) and \(ch'(v)\ge 10-1\times 5-4=1>0\) by Claim (c). For Fig. 5(6)–(7), we have \(f_{6^+}(v)\ge 3\) and \(ch'(v)\ge 10-1\times 5-\frac{3}{2}-\frac{11}{4}=\frac{3}{4}>0\). For Fig. 5(8)–(9), we have \(f_{6^+}(v)\ge 2\) and \(ch'(v)\ge 10-1\times 5-\frac{3}{2}\times 3=\frac{1}{2}>0\). For Fig. 5(10), we have \(f_{6^+}(v)\ge 3\) and \(ch'(v)\ge 10-1\times 4-\frac{21}{4}=\frac{3}{4}>0\) by Claim (d). For Fig. 5(11) and (13), we have \(f_{6^+}(v)\ge 2\) and \(ch'(v)\ge 10-1\times 4-\frac{3}{2}-4=\frac{1}{2}>0\). For Fig. 5(12) and (16), we have \(f_{6^+}(v)\ge 2\) and \(ch'(v)\ge 10-1\times 4-\frac{11}{4}\times 2=\frac{1}{2}>0\). For Fig. 5(14) and (15), we have \(f_{6^+}(v)\ge 1\) and \(ch'(v)\ge 10-1\times 4-\frac{3}{2}\times 2-\frac{11}{4}=\frac{1}{4}>0\). For Fig. 5 (17), we have \(ch'(v)\ge 10-1\times 4-\frac{3}{2}\times 4=0\). For Fig. 5(18), we have \(f_{6^+}(v)\ge 2\) and \(ch'(v)\ge 10-1\times 3-\frac{13}{2}=\frac{1}{2}>0\) by Claim (e). For Fig. 5(19), we have \(f_{6^+}(v)\ge 1\) and \(ch'(v)\ge 10-1\times 3-\frac{3}{2}-\frac{21}{4}=\frac{1}{4}>0\). For Fig. 5(20), we have \(f_{6^+}(v)\ge 1\) and \(ch'(v)\ge 10-1\times 3-\frac{11}{4}-4=\frac{1}{4}>0\). For Fig. 5(21), we have \(ch'(v)\ge 10-1\times 3-\frac{3}{2}\times 2-4=0\). For Fig. 5(22), we have \(ch'(v)\ge 10-1\times 3-\frac{3}{2}-\frac{11}{4}\times 2=0\). For Fig. 5(23), we have \(f_{6^+}(v)\ge 1\). Suppose \(d(f_{2})=d(f_{3})=d(f_{4})=d(f_5)=d(f_6)=3\). Then \(\min \{d(v_{3}),~d(v_{4}),~d(v_{5}),~d(v_{6})\}\ge 4\). If \(d(v_{2})=d(v_{6})=3\), then \(d(f_{1})\ge 5\) and \(d(f_{7})\ge 5\) by Lemma 3, so \(ch'(v)\ge 10-1\times 2-\frac{3}{2}\times 2-\frac{5}{4}\times 2-1\times 1-\frac{1}{3}\times 2=\frac{5}{6}>0\). If \(f_{2}\), \(f_{3}\), \(f_{4}\), \(f_{5}\) and \(f_{6}\) are incident with no \(3^-\)-vertex, then \(ch'(v)\ge 10-1\times 2-\frac{5}{4}\times 5-\frac{3}{4}\times 2=\frac{1}{4}>0\). For Fig. 5(24), we have \(ch'(v)\ge 10-1\times 2-\frac{3}{2}-\frac{13}{2}=0\). For Fig. 5(25), we have \(ch'(v)\ge 10-1\times 2-\frac{11}{4}-\frac{21}{4}=0\). For Fig. 5(26), we have \(ch'(v)\ge 10-1\times 2-4\times 2=0\). \(\square \)

Hence we complete the proof of the theorem.