%I #24 Nov 13 2025 10:34:23
%S 42,0,0,18,186,66,0,234,930,750,0,2244,4578,6498,120
%N Number of proper 3-colorings of the generalized chorded cycle graph C_n^{(3)}.
%C The sequence counts the exact number of proper vertex colorings using 3 colors of circular chord graphs C_n^(3), defined as cycle graphs C_n with chords connecting vertices at offset 3 (vertices i and i+3 mod n), and with diametric edges added for even n.
%C Notably, the sequence displays modular phase transitions and recurring zeros for even values of n divisible by 4 (n=8,12,16,...). These zeros occur due to structural constraints from chords and diametric edges preventing any valid 3-colorings.
%C The observed modular non-monotone pattern is unique and does not match known classical graph families, motivating deeper combinational and algebraic investigations.
%C Empirical analysis using the transfer matrix method indicates that the sequence a(n) = P(C_n^(3), 3) satisfies a linear recurrence relation of finite order. Specifically, the number of 3-colorings of C_n^(3) can be represented using adjacency-like matrices encoding local constraints imposed by chords and diametric edges.
%C Formally, let T be the transfer matrix representing transitions of valid colorings between successive vertices or segments of the graph. The count a(n) corresponds to a trace or specific linear combination of powers of T: a(n) = Tr(M * T^n), for some suitable projection matrix M, capturing the graph's cyclical boundary conditions and additional chord and diameter constraints.
%C The minimal polynomial of the transfer matrix T dictates the order of this recurrence. Although computationally validated for initial terms, determining an explicit closed-form solution or exact minimal polynomial and recurrence relation analytically remains an open combinational and algebraic problem.
%D N. L. Biggs, Algebraic Graph Theory. Cambridge University Press, 2nd ed., 1993.
%D D. B. West, Introduction to Graph Theory. Prentice Hall, 2nd ed., 2001.
%D R. J. Wilson, Graph Theory. Longman, 5th impression, 1996.
%H George D. Birkhoff, <a href="https://www.jstor.org/stable/1967597">A determinant formula for the number of ways of coloring a map</a>, Ann. Math., 14:42-46.
%H Rogelio N. Lopez-Bonilla, Julian Allagan, Shawn M. Langley, and Angel J. Clinton, <a href="https://arxiv.org/abs/2509.05845">Golden Ratio Growth and Phase Transitions in Chromatic Counts of Circular Chord Graphs</a>, arXiv:2509.05845 [math.CO], 2025. See pp. 1, 8, 11.
%H Ronald C. Read, <a href="https://doi.org/10.1016/S0021-9800(68)80087-0">An Introduction to Chromatic Polynomials</a>, Journal of Combinatorial Theory, 4(1968), 52-71.
%H Rogelio Lopez-Bonilla, Julian Allagan, Shawn M. Langley, and Angel J. Clinton, <a href="https://doi.org/10.9734/jamcs/2025/v40i102060">Chromatic Polynomials of C_n^(3) Graphs: Lucas Sequences, phi^n Asymptotics and Linear Recurrence Existence</a>, J. Adv. Math. Comp. Sci. (2025) Vol. 40, Issue 10, Art. JAMCS.146917, 79-98. <a href="https://www.researchgate.net/profile/Rogelio-Lopez-Bonilla/publication/397141446_Chromatic_Polynomials_of_C_Graphs_Lucas_Sequences_ph_Asymptotics_and_Linear_Recurrence_Existence">Alternative link</a>. See pp. 80, 95.
%e For n=6, consider the graph C_6^(3), constructed as follows:
%e - Start with a cycle graph (hexagon) having vertices labeled {0,1,2,3,4,5}.
%e - Add chords connecting vertex i with vertex i+3 mod 6, forming edges (0,3), (1,4), (2,5).
%e - Since n is even, include diametric edges connecting opposite vertices: edges (0,3), (1,4), (2,5). (Note these diametric edges coincide with chords for n=6.)
%e The resulting graph is symmetric and moderately dense. Enumerating explicitly all possible vertex-coloring assignments with exactly three colors, we find precisely 42 distinct valid 3-colorings (each satisfying the condition that no two adjacent vertices share the same color).
%e Thus, a(6)=42.
%p with(GraphTheory):
%p Cn3_graph := proc(n)
%p local G, i;
%p G := CycleGraph(n);
%p for i from 0 to n-1 do
%p AddEdge(G, {i, (i+3) mod n});
%p end do;
%p if modp(n, 2) = 0 then
%p for i from 0 to n/2 - 1 do
%p AddEdge(G, {i, (i + n/2) mod n});
%p end do;
%p end if;
%p return G;
%p end proc:
%p a := proc(n) local G;
%p G := Cn3_graph(n);
%p return ChromaticPolynomial(G, 3);
%p end proc:
%p # Compute initial terms from n=6 to n=20:
%p seq(a(n), n=6..20);
%t Cn3Graph[n_] := Module[{g, edges, i},
%t edges = Table[{i, Mod[i + 1, n]}, {i, 0, n - 1}]; (* Cycle edges *)
%t edges = Join[edges, Table[{i, Mod[i + 3, n]}, {i, 0, n - 1}]]; (* Chord edges *)
%t If[EvenQ[n],
%t edges = Join[edges, Table[{i, Mod[i + n/2, n]}, {i, 0, n/2 - 1}]]
%t ];
%t Graph[edges, VertexLabels -> "Name"]
%t ];
%t a[n_] := Length@Select[
%t Tuples[{1, 2, 3}, n],
%t And @@ (#[[#[[1]] + 1]] != #[[#[[2]] + 1]] & /@
%t EdgeList[Cn3Graph[n]] /. {x_, y_} :> {x, y})
%t ] &;
%t (* Generate terms for n from 6 to 20 *)
%t Table[a[n], {n, 6, 20}]
%o (Python)
%o # Illustrative brute-force check for small n using networkx
%o import networkx as nx
%o from itertools import product
%o def Cn_k_graph(n, k):
%o G = nx.cycle_graph(n)
%o for i in range(n):
%o G.add_edge(i, (i+k)%n)
%o if n % 2 == 0:
%o for i in range(n//2):
%o G.add_edge(i, i+n//2)
%o return G
%o def count_colorings(G, colors=3):
%o nodes = list(G.nodes())
%o count = 0
%o for coloring in product(range(colors), repeat=len(nodes)):
%o if all(coloring[u] != coloring[v] for u,v in G.edges()):
%o count += 1
%o return count
%o # Example usage:
%o for n in range(6, 21):
%o G = Cn_k_graph(n, 3)
%o print(f'n={n}, colorings={count_colorings(G)}')
%Y Cf. A000670 (number of preferential arrangements), A001047 (chromatic polynomial of cycles at x=3), A003049 (chromatic polynomial of complete graphs), A129912 (number of 3-colorings of certain circulant graphs).
%Y Related to chromatic polynomial evaluations and modular coloring patterns not captured by standard families. May also be compared to sequences involving nonzero chromatic roots and Beraha numbers.
%K nonn,hard,more
%O 6,1
%A _Rogelio Lopez Bonilla_, May 07 2025