Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 3103 https://internationalpubls.com More Results on Complementary Tree Domination Number of Semi Total Point Graph S. Jayalakshmi1 and P. Vidhya2 1Research Scholar (Part Time), School of Mathematics Madurai Kamaraj University, Madurai - 625021, Tamilnadu, India E-mail: jayark83@gmail.com 2Associate Professor, Department of Mathematics, EMG Yadava Women’s College Madurai - 625014, Tamilnadu, India E-mail: vidhyaramman@gmail.com, p.vidhya-mat@emgywomenscollege.ac.in Article History: Received: 02-01-2025 Revised: 25-02-2025 Accepted: 20-03-2025 Abstract: A set D βŠ† V of a graph G = (V, E) is a complementary tree dominating set if the induced subgraph < 𝑉(G) βˆ’ D > is a tree. The complementary tree domination number Ξ³ctd(G) is the minimum cardinality of a complementary tree dominating set (ctd-set) of G. The semi total point graph T2(G) is the graph G whose vertex set is V(G) βˆͺ E(G). Where two vertices are adjacent if and only if (i) they are adjacent vertices of G or (ii) one is a vertex and the other is an edge of G incident with it. In this paper complementary tree domination number of semi total point graph of graphs, its bounds and relation between Ξ³ctd(G) and Ξ³ctd(T2(G)) are obtained. Keywords: Dominating set, Complementary tree dominating set, Semi total point graph. 1. Introduction A Graph 𝐺(𝑉, 𝐸) discussed in this paper be a simple, finite, undirected, connected graph with p vertices and q edges. A set of vertices in a graph G is independent, if no two vertices are adjacent. The largest number of vertices in such a set is called the independence number and is denoted by Ξ²0(𝐺). The corona 𝐺1 ∘ 𝐺2 of two graphs 𝐺1 and 𝐺2 are defined as the graph G obtained by taking one copy of 𝐺1 of order 𝑝1 and 𝑝1 copies of 𝐺2 and then joining the π‘–π‘‘β„Ž vertex of 𝐺1 to every vertex in the π‘–π‘‘β„Ž copy of 𝐺2. The Corona 𝐺1 ∘ 𝐺2 has 𝑝1(1 + 𝑝2) vertices and π‘ž1 + 𝑝1π‘ž2 + 𝑝1𝑝2 edges. The graph 𝐢𝑛 (𝑑) is the one point union of t cycles of length n. A graph G is unicyclic if it contains exactly one cycle. A broom graph 𝐡𝑛,π‘š is a graph of n vertices which have a path π‘ƒπ‘š and 𝑛 βˆ’ π‘š pendant vertices, all of these vertices are adjacent to either the origin u or the terminus v of the path. Any undefined term in this paper may be found in Harary [2]. The concept of domination in graphs was introduced by Ore [5]. A set 𝐷 βŠ† 𝑉 is said to be a dominating set of G, if every vertex in 𝑉 βˆ’ 𝐷 is adjacent to some vertex in D. The minimum cardinality of a dominating set is called the domination number of G and is denoted by Ξ³(𝐺). The complementary tree domination number of a graph was introduced byS. Muthammai, M. Bhanumathi and P. Vidhya [4] have established some results on complementary tree domination number of graphs. A set 𝐷 βŠ† 𝑉(𝐺) is said to be complementary tree dominating set (ctd-set) if the Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 3104 https://internationalpubls.com induced subgraph βŸ¨π‘‰(𝐺) βˆ’ 𝐷⟩ is a tree. The minimum cardinality of a ctd-set is called the complementary tree domination number of G and is denoted by γ𝑐𝑑𝑑(𝐺). E. Sampath Kumar and S.B. Chikkodimath[3] introduced the concept of semi total point graphs of a graph. Also B. Basavanagoud, S.M. Hosamani and S.H. Malghan[1] have obtained some results on domination number of semi total point graph. Given a graph G, the semi total point graph 𝑇2(𝐺) of G is the graph whose point set is 𝑉(𝐺) βˆͺ 𝐸(𝐺) where two points are adjacent if and only if (i) they are adjacent points of G or (ii) one is a point of G and the other is a line of G, incident with it. For notation convenience, an edge (𝑒1, 𝑒2) ∈ 𝐸(𝐺) then its corresponding edge vertex is denoted by 𝑒12 β€² in 𝑇2(𝐺). In this paper complementary tree domination number of semi total point graph of graphs, its bounds and relation between γ𝑐𝑑𝑑(𝐺) and γ𝑐𝑑𝑑(𝑇2(𝐺)) are obtained. 2. Prior Results Observation 2.1. [6] (i) For the path 𝑃𝑝, Ξ³ 𝑐𝑑𝑑 (𝑇2(𝑃𝑝)) = 𝑝 βˆ’ 1, where (𝑝 β‰₯ 2). (ii) For the cycle 𝐢𝑝, Ξ³ 𝑐𝑑𝑑 (𝑇2(𝐢𝑝)) = 𝑝 βˆ’ 1, where (𝑝 β‰₯ 3). (iii) For the star graph 𝐾1,π‘βˆ’1, Ξ³ 𝑐𝑑𝑑 (𝑇2(𝐾1,π‘βˆ’1)) = 𝑝 βˆ’ 1 or π‘ž, 𝑝 β‰₯ 2. (iv) For a complete graph 𝐾𝑝, 𝑝 β‰₯ 4 then Ξ³ 𝑐𝑑𝑑 (𝑇2(𝐾𝑝)) = 𝑝2βˆ’5𝑝+12 2 . (v) For a wheel graph π‘Šπ‘, 𝑝 β‰₯ 4Ξ³ 𝑐𝑑𝑑 (𝑇2(π‘Šπ‘)) = 𝑝 where π‘Šπ‘ = πΆπ‘βˆ’1 + 𝐾1 for (𝑝 β‰₯ 4). (vi) For a corona graph 𝑃𝑝 ∘ 𝐾1, 𝑝 β‰₯ 2 then Ξ³ 𝑐𝑑𝑑 (𝑇2(𝑃𝑝 ∘ 𝐾1)) = 2𝑝 βˆ’ 1. (vii) For a corona graph 𝐢𝑝 ∘ 𝐾1, 𝑝 β‰₯ 3 then Ξ³ 𝑐𝑑𝑑 (𝑇2(𝐢𝑝 ∘ 𝐾1)) = 2𝑝. (viii) Ξ³ 𝑐𝑑𝑑 (𝑇2(𝐾1 + 𝑃𝑛)) = 𝑝, 𝑝 β‰₯ 3. Proposition 2.2. [6] Let 𝐺(𝑝, π‘ž) be a connected graph with Ξ΄(𝐺) β‰₯ 2 then atleast one vertex of G is the member of ctd-set of 𝑇2(𝐺). Proposition 2.3. [6] For any connected graph 𝐺(𝑝, π‘ž) with 𝑝 β‰₯ 2,⌈ 𝑝 Ξ”(𝐺)+1 βŒ‰ ≀ γ𝑐𝑑𝑑(𝑇2(𝐺)) ≀ 𝑝 + π‘ž βˆ’ 2. Theorem 2.4. [6] γ𝑐𝑑𝑑(𝑇2(𝐺)) = 1 if and only if 𝐺 β‰… 𝐾2. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 3105 https://internationalpubls.com 3. Characterisation of Complementary Tree Dominating Set in Semi Total Point Graph T2(G) In the following, a necessary and sufficient condition for a ctd-set of a graph G to be a ctd-set of its semi total point graph 𝑇2(𝐺) is found. Theorem 3.1. A ctd-set D of a connected graph 𝐺 = (𝑉, 𝐸) is also a ctd-set of 𝑇2(𝐺) if and only if (i) ⟨𝐷 ⟩ has an isolated vertices. (ii) For each 𝑣 ∈ 𝐷, 𝑁(𝑣) ∩ (𝑉 βˆ’ 𝐷) β‰ οͺ and (iii) βŸ¨π‘‰ βˆ’ 𝐷⟩ β‰… 𝐾1 and if 𝑣 ∈ 𝑉 βˆ’ 𝐷 then 𝑁(𝑣) ∩ 𝐷 β‰  οͺ . Proof. Let D be a ctd-set of both G and 𝑇2(𝐺). (i) Let 𝑣 ∈ 𝐷 be not an isolated vertex in ⟨𝐷 ⟩ then its edge vertex 𝑣′ in 𝑇2(𝐺) is isolated inβŸ¨π‘‰(T2(G))β€ˆ βˆ’ β€ˆDβ€ˆβŸ© which contradicts the ctd-set of 𝑇2(𝐺). Therefore, ⟨𝐷 ⟩ has an isolated vertices. (ii) Let there exists a vertex 𝑣 ∈ 𝐷 such that 𝑁(𝑣) ∩ (𝑉 βˆ’ 𝐷) = οͺ . Then, its edge vertex 𝑣′ is isolated in βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷 ⟩. (iii) If 𝐾2 is an induced subgraph of βŸ¨π‘‰ βˆ’ 𝐷 ⟩. Then βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷 ⟩ contains a cycle. Therefore, βŸ¨π‘‰ βˆ’ 𝐷⟩ β‰… 𝐾1. Let 𝑣 ∈ 𝐾1. Since G is connected so that remaining vertices are adjacent to v. There exists a edge vertices 𝑣𝑖 β€² ∈ (𝑣, 𝑣𝑖) such that βŸ¨π‘£, 𝑣𝑖, 𝑣𝑖 β€²βŸ© β‰… 𝐢3 in 𝑇2(𝐺).Hence 𝑁(𝑣) ∩ 𝐷 β‰  οͺ . Conversely, if (i) is true, D is dominating set of 𝑇2(𝐺). If (ii) holds, then βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷 ⟩ is connected and if (iii) holds, then βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷 ⟩ is acyclic. Therefore, βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷 ⟩ is a tree. Hence, D is also a ctd-set of 𝑇2(𝐺). β–‘ In the following, exact values of complementary tree domination number of semi total point graph of some classes of graphs are given. Proposition 3.2. Let 𝐢𝑝 (𝑑) , 𝑑 β‰₯ 2 be the one point union of t cycles of length p(𝑝 β‰₯ 3) thenγ𝑐𝑑𝑑 (𝑇2(𝐢𝑝 (𝑑) )) = (𝑝 βˆ’ 1)𝑑, 𝑝 β‰₯ 3. Proof. Let 𝐺 = 𝐢𝑝 (𝑑) and u be the point of union of t cycles of length p. Let the vertex set of π‘˜π‘‘β„Ž cycle in 𝐢𝑝 (𝑑) be π‘‰π‘˜ = {𝑒, π‘’π‘˜1, π‘’π‘˜2, … , π‘’π‘˜,π‘βˆ’1}π‘˜ = 1,2, … , 𝑑(𝑑 β‰₯ 2) π‘‰π‘˜ (𝑇2(𝐢𝑝 (𝑑) )) = {𝑒, π‘’π‘˜1, π‘’π‘˜2, … , π‘’π‘˜,π‘βˆ’1} βˆͺ {π‘’π‘˜1 β€² , π‘’π‘˜2 β€² … π‘’π‘˜π‘ β€² } where π‘’π‘˜1 β€² , π‘’π‘˜2 β€² , … , π‘’π‘˜π‘ β€² are the corresponding edge vertices of (𝑒, π‘’π‘˜1), (π‘’π‘˜1, π‘’π‘˜2), … , (𝑒, π‘’π‘˜,π‘βˆ’1). Let π·π‘˜ = {π‘’π‘˜1, π‘’π‘˜3 β€² , … , π‘’π‘˜π‘ β€² }, π‘˜ = 1, 2, … , 𝑑 𝐷 = ⋃ π·π‘˜ 𝑑 π‘˜=1 βŠ† 𝑉(𝑇2(𝐺)). Then βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷 ⟩ is a tree. Hence D is a minimum ctd-set of 𝑇2(𝐺). Therefore, |𝐷| = γ𝑐𝑑𝑑(𝑇2(𝐺)) = (𝑝 βˆ’ 1)𝑑. β–‘ Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 3106 https://internationalpubls.com Proposition 3.3. Let G be a unicyclic graph by attaching a path of length (𝑛 β‰₯ 1) to the 𝑑(≀ 𝑝) consecutive vertices of 𝐢𝑝(𝑝 β‰₯ 3). Then γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑛𝑑 + 𝑝 βˆ’ 1. Proof. In the cycle 𝐢𝑝(𝑝 β‰₯ 3) say 𝑣1, 𝑣2, … , 𝑣𝑝. Consider a path of length π‘ƒπ‘βˆ’1 in 𝐢𝑝 say 𝑣1, 𝑣2, … , π‘£π‘βˆ’1 and attach a path 𝑃𝑛 β€²(𝑛 β‰₯ 1) say 𝑣𝑖 , 𝑒2, … , 𝑒𝑛, 𝑖 = 1, 2, … , 𝑑 to the 𝑑(≀ 𝑝) consecutive vertices of 𝐢𝑝. In 𝑇2(𝐺), the set of all edge vertices of path 𝑃𝑛 β€² of t consecutive vertices of 𝐢𝑝, edge vertices of a path π‘ƒπ‘βˆ’1 and a vertex 𝑣𝑝 forms a minimum ctd-set of 𝑇2(𝐺). Therefore, γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑛𝑑 + 𝑝 βˆ’ 1. β–‘ Corollary 3.4. G be a unicyclic graph by attaching one pendant vertex to exactly one vertex of 𝐢𝑝 then γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝. Corollary 3.5. G be a unicyclic graph by attaching one pendant vertex to 𝑝 βˆ’ 1 vertices of 𝐢𝑝. Then γ𝑐𝑑𝑑(𝑇2(𝐺)) = 2𝑝 βˆ’ 2. Corollary 3.6. G be a unicyclic graph by attaching a path of length n to exactly one vertex of 𝐢𝑝 thenγ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 + 𝑛 + 1. 4. Bounds and Some Exact Values for the Complementary Tree Domination Number of Semi Total Point Graph of Graphs Theorem 4.1. γ𝑐𝑑𝑑(𝑇2(𝐺)) = 2 if and only if 𝐺 β‰… 𝐾1,2 or 𝐢3. Proof. Let D be a γ𝑐𝑑𝑑-set of 𝑇2(𝐺) such that |𝐷| = 2. Let 𝐷 = {𝑒1, 𝑒2} where 𝑒1, 𝑒2 ∈ 𝑉(𝑇2(𝐺)). Case 1. 𝑒1 and 𝑒2 are vertices in G. Then, D is also a γ𝑐𝑑𝑑-set of G. By Theorem 3.1, it can be seen that 𝐺 β‰… 𝐾1,2. Case 2. 𝑒1, 𝑒2 ∈ 𝑉(𝑇2(𝐺)) βˆ’ 𝑉(𝐺). Let 𝑒1 = 𝑒1 β€² and 𝑒2 = 𝑒2 β€² where (𝑒, 𝑒1) and (𝑒, 𝑒2) ∈ 𝐸(𝐺). Since 𝑒1 β€² and 𝑒2 β€² are edge vertices in 𝑇2(𝐺) and βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷⟩ is connected and acyclic. Hence it can be seen that 𝐺 β‰… 𝐾1,2. Case 3. Let 𝑒1 ∈ 𝑉(𝐺) and 𝑒2 = 𝑒2 β€² ∈ 𝑉(𝑇2(𝐺)) βˆ’ 𝑉(𝐺) Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 3107 https://internationalpubls.com Sub case 3.1. 𝑒2 = 𝑒2 β€² . That is 𝐷 = {𝑒1, 𝑒2 β€² } is a γ𝑐𝑑𝑑-set of 𝑇2(𝐺). Hence 𝐺 β‰… 𝐢3. Sub case 3.2. 𝑒2 β‰  𝑒2 β€² . Let 𝑒2 = 𝑣′ for some (𝑒1, 𝑣) ∈ 𝐸(𝐺) and 𝑣′ β‰  𝑒2 β€² . Then 𝐷 = {𝑒, 𝑣′} is a γ𝑐𝑑𝑑-set of 𝑇2(𝐺). 𝑒2 β€² β‰  𝑣′ implies that βŸ¨π‘’2 β€² , 𝑣, π‘₯ ⟩ form a cycle in βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷 ⟩ forms either cycle or disconnected where π‘₯, 𝑣 ∈ 𝑉(𝐺). Conversely, if 𝐺 β‰… 𝐾1,2 or 𝐢3. Then γ𝑐𝑑𝑑(𝑇2(𝐺)) = 2. β–‘ Theorem 4.2. γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 + π‘ž βˆ’ 2 if and only if 𝐺 β‰… 𝐾2. Proof. Assume γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 + π‘ž βˆ’ 2. Let D be a minimum ctd-set of 𝑇2(𝐺) having 𝑝 + π‘ž βˆ’ 2 vertices. Since βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷⟩ β‰… 𝐾2. Let 𝑉(𝑇2(𝐺)) βˆ’ 𝐷 = {𝑒, 𝑣} either (i) 𝑒, 𝑣 ∈ 𝑉(𝐺) or (ii) 𝑒 ∈ 𝑉(𝐺) and 𝑣 = 𝑒′ ∈ 𝑉(𝑇2(𝐺)) βˆ’ 𝑉(𝐺) where 𝑒′ ∈ (𝑒, 𝑣). Case 1. Let 𝑒′ ∈ 𝐷 where 𝑒′ be the edge vertex of (𝑒, 𝑣) in 𝑇2(𝐺). Therefore, no vertex of 𝑉(𝐺) is an element of D hence 𝑒′ ∈ 𝐷. Therefore 𝐺 β‰… 𝐾2. Case 2. Let 𝑒 ∈ 𝑉(𝐺), 𝑒′ ∈ 𝑉(𝑇2(𝐺)) βˆ’ 𝑉(𝐺). Since βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷⟩ β‰… 𝐾2. 𝑒′ is adjacent to u in βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷 ⟩. That is v is adjacent to a vertex u in G. Therefore, 𝐺 β‰… 𝐾2. Conversely, if 𝐺 β‰… 𝐾2 then γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 + π‘ž βˆ’ 2. β–‘ Remark 4.3. If 𝑝 β‰₯ 3 then γ𝑐𝑑𝑑(𝑇2(𝐺)) ≀ 𝑝 + π‘ž βˆ’ 3. Equality holds if 𝐺 β‰… 𝑃3. Theorem 4.4. Let 𝐺(𝑝, π‘ž) be a complete graph with 4 ≀ 𝑝 ≀ 8 then γ𝑐𝑑𝑑(𝑇2(𝐺)) ≀ ⌈ 𝑝+π‘ž 2 βŒ‰. Proof. We prove induction on p. Let 𝑝 = 4, 𝑒 = (𝑒1, 𝑒2) ∈ 𝐸(𝐺) where 𝑒1, 𝑒2 ∈ 𝑉(𝐺). Let 𝐷 = {𝑒1, 𝑒2, 𝑒12 β€² , 𝑒34 β€² } βŠ† 𝑉(𝑇2(𝐺)) is a ctd-set of 𝑇2(𝐺) where 𝑒12 β€² , 𝑒34 β€² ∈ 𝑉(𝑇2(𝐺)) βˆ’ 𝑉(𝐺) and 𝐷′ = 𝑉(𝑇2(𝐺)) βˆ’ 𝐷 = {𝑒3, 𝑒4, 𝑒13 β€² , 𝑒23 β€² , 𝑒24 β€² , 𝑒41 β€² } since 𝑝 β‰₯ 4 and Ξ΄(𝐺) β‰₯ 2 each vertex in βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷′ ⟩ is adjacent to atleast one vertex in D and βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷⟩ β‰… π‘†π‘š,π‘š,π‘š, π‘š β‰₯ 2 in 𝑇2(𝐺). Therefore, |𝐷| ≀ ⌈ 𝑝+π‘ž 2 βŒ‰. Hence γ𝑐𝑑𝑑(𝑇2(𝐺)) ≀ ⌈ 𝑝+π‘ž 2 βŒ‰. Equality holds if 𝐺 β‰… 𝐾8. β–‘ Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 3108 https://internationalpubls.com Theorem 4.5. Let G be a connected graph such that Ξ΄(𝐺) β‰₯ 2 then γ𝑐𝑑𝑑(𝑇2(𝐺)) ≀ π‘ž βˆ’ Ξ”(𝐺) + 1. Proof. Let v be a vertex of maximum degree in G. Let 𝑆 = {𝑣𝑖 β€²: (𝑣, 𝑣𝑖) ∈ 𝐸(𝐺), 𝑖 = 1, … , Ξ”(𝐺)}. Any set 𝐷 βŠ† 𝑉(𝑇2(𝐺)) such that V(𝑇2(𝐺)) βˆ’ D = {⋃ 𝑣𝑖 π‘βˆ’1 𝑖=1 } βˆͺ 𝑆. Since Ξ΄(𝐺) β‰₯ 2, 𝑑𝑒𝑔(𝑣𝑖) β‰₯ 2, 𝑖 = 1,2, … , 𝑝 βˆ’ 1 and hence 𝑣𝑖 is adjacent to the vertices of G other than v, Then 𝑣𝑖 β€² is adjacent to the vertex 𝑣𝑖, 𝑖 = 1, 2, … , 𝑝 βˆ’ 1 where 𝑣𝑖 is a vertex in 𝑉(𝑇2(𝐺)) βˆ’ 𝐷. Also, v is adjacent to atleast one vertex of G and hence in S. Therefore, D is a dominating set of 𝑉(𝑇2(𝐺)). Moreover βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷⟩ β‰… 𝑇 ∘ 𝐾1 and hence D is a ctd-set of 𝑇2(𝐺). Therefore, γ𝑐𝑑𝑑(𝑇2(𝐺)) ≀ |𝑉(𝑇2(𝐺)) βˆ’ (𝑉(𝐺) βˆ’ 𝑣) βˆ’ 𝑆| = 𝑝 + π‘ž βˆ’ 𝑝 + 1 βˆ’ Ξ”(𝐺) ≀ π‘ž βˆ’ Ξ”(𝐺) + 1 Equality holds if 𝐺 β‰… 𝐾3, 𝐾4&𝐾1 + π‘ƒπ‘›βˆ’1. β–‘ Theorem 4.6. Let 𝑇2(𝐺1(𝑝1, π‘ž1)) and 𝑇2(𝐺2(𝑝2, π‘ž2)) be two connected graphs of order atleast two. Let T be an induced sub-graph of 𝑇(𝐺1) having maximum number of vertices such that T is a tree. If Ξ²0 is the independence number of 𝑇2(𝐺2) and vertices corresponding to the edge joining from copies of 𝑇(𝐺2) to 𝑇(𝐺1) then γ𝑐𝑑𝑑(𝑇2(𝐺1 ∘ 𝐺2)) ≀ 2𝑝1𝑝2 + 𝑝1(1 + π‘ž2) + π‘ž1 βˆ’ (𝑑 βˆ’ 2)Ξ²0 βˆ’ 𝑑. Proof. Let T be an induced sub-graph of 𝑇(𝐺1) having maximum number of vertices such that T is a tree and |𝑇| = 𝑑. Let S be a maximum independent set of 𝑇(𝐺2) and vertices corresponding to the edge joining to the vertices of each copies of 𝑇(𝐺2) to 𝑇(𝐺1) such that |𝑆| = Ξ²0 and 𝐷′ be the set of vertices in S in copies of 𝑇(𝐺2) which are adjacent to the vertices of T then |𝐷′| = (𝑑 βˆ’ 2)Ξ²0 if Ξ΄(𝐺1) β‰₯ 2. Let 𝐷 = (𝑉(𝑇2(𝐺1 ∘ 𝐺2))) βˆ’ (𝑉(𝑇) βˆͺ 𝐷′) then 𝑉(𝑇2(𝐺1 ∘ 𝐺2)) βˆ’ 𝐷 = 𝑉(𝑇) βˆͺ 𝐷′ and (𝑑 βˆ’ 2) vertices of 𝑉(𝑇) are adjacent to (𝑝2 βˆ’ Ξ²0) vertices in a copy of 𝑇2(𝐺2). Also each vertex in 𝐷′ is adjacent to atleast one (𝑝2 βˆ’ Ξ²0) vertices in a copy of 𝑇2(𝐺2). Therefore, D is a dominating set of 𝑇2(𝐺1 ∘ 𝐺2) and βŸ¨π‘‰(𝑇2(𝐺1 ∘ 𝐺2)) βˆ’ 𝐷 ⟩ is a tree. γ𝑐𝑑𝑑(𝑇2(𝐺1 ∘ 𝐺2)) ≀ |𝐷| ≀ |𝑉(𝑇2(𝐺1 ∘ 𝐺2)) βˆ’ (𝑉(𝑇) βˆͺ 𝐷′)| = 2𝑝1𝑝2 + 𝑝1(1 + π‘ž2) + π‘ž1 βˆ’ 𝑑(1 + Ξ²0) βˆ’ 2Ξ²0. β–‘ 5. Relation Between ctd(G) and ctd(T2(G)) Observation 5.1. For any connected graph 𝐺(𝑝, π‘ž), 𝑝 β‰₯ 2 then γ𝑐𝑑𝑑(𝐺) ≀ γ𝑐𝑑𝑑(𝑇2(𝐺)). Equality holds if 𝐺 β‰… 𝐾1,π‘βˆ’1. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 3109 https://internationalpubls.com Theorem 5.2. For any connected graph 𝐺(𝑝, π‘ž), 𝑝 β‰₯ 2 with Ξ΄(𝐺) = 1 then γ𝑐𝑑𝑑(𝑇2(𝐺)) ≀ 𝑝 βˆ’ 1 + γ𝑐𝑑𝑑(𝐺). Proof. Let D be a minimum ctd-set of G and hence |𝐷| = γ𝑐𝑑𝑑(𝐺). Therefore, βŸ¨π‘‰(𝐺) βˆ’ 𝐷 ⟩ is a tree. Now, the set 𝐷′ = 𝐷 βˆͺ (𝑉(𝑇2(𝐺)) βˆ’ 𝑉(𝐺)) is a minimum ctd-set of 𝑇2(𝐺). Hence, γ𝑐𝑑𝑑(𝑇2(𝐺)) ≀ |𝐷′| = γ𝑐𝑑𝑑(𝐺) + 𝑝 βˆ’ 1. β–‘ Theorem 5.3. Given two integers a and b with 2 ≀ π‘Ž ≀ 𝑏 there exists a graph with π‘Ž + 𝑏 + 1 vertices such that Ξ³(𝑇2(𝐺)) = π‘Ž + 1 and γ𝑐𝑑𝑑(𝑇2(𝐺)) = π‘Ž + 𝑏. Also γ𝑐𝑑𝑑(𝑇2(𝐺)) ≀ Ξ³(𝑇2(𝐺)) + 𝑏. Figure 1. Proof. In the cycle πΆπ‘Ž+2(π‘Ž β‰₯ 2) by {𝑣1, 𝑣2, … , π‘£π‘Ž+2} of length π‘Ž + 2. Consider a path of length a. In this path attach (𝑏 βˆ’ π‘Ž) pendant edges at exactly one vertex and attach one pendant edge at each of the remaining (π‘Ž βˆ’ 1) vertices. Let the graph thus obtained be denoted by G and G has π‘Ž + 𝑏 + 1 vertices. In 𝑇2(𝐺), edges of cycle πΆπ‘Ž+2 and edges of pendant vertex of G and the vertices of G are the vertex set of 𝑇2(𝐺). Therefore,|𝑉(𝑇2(𝐺))| = 2π‘Ž + 2𝑏 + 2. The set {𝑣1, 𝑣2, … , π‘£π‘Ž+1} forms a minimum dominating set of 𝑇2(𝐺) and the set consisting of edge vertex of path π‘ƒπ‘Ž+1 a vertex πΆπ‘Ž+2 and all the edge vertex of pendant vertices of G forms a minimum ctd-set of 𝑇2(𝐺). Therefore, γ𝑐𝑑𝑑(𝑇2(𝐺)) = π‘Ž + 𝑏. If a = b then, the equality holds. β–‘ Theorem 5.4. If γ𝑐𝑑𝑑(𝐺) = 1 then γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 βˆ’ 1 where 𝑝 β‰₯ 2 is the number of vertices in G. Proof. Assume γ𝑐𝑑𝑑(𝐺) = 1, then 𝐺 β‰… 𝐾1 + 𝑇 where T is a tree with atleast two vertices. Let 𝑉(𝐾1) = 𝑣 and 𝑉(𝑇) = {𝑣1, 𝑣2, … , π‘£π‘βˆ’1} then 𝑉(𝐺) = {𝑣, 𝑣1, 𝑣2, … π‘£π‘βˆ’1} and 𝑉(𝑇2(𝐺)) = 𝑉(𝐺) βˆͺ 𝑣𝑖 β€² βˆͺ 𝑣𝑖,𝑖+1 β€² where 𝑣𝑖 β€² ∈ (𝑣, 𝑣𝑖) and 𝑣𝑖,𝑖+1 β€² ∈ (𝑣𝑖 , 𝑣𝑖+1) where 𝑖 = 1, 2, … , 𝑝 βˆ’ 1. Let 𝐷 = {𝑣𝑖,𝑖+1 β€² /𝑖 = 1,2, … , 𝑝 βˆ’ 2} βˆͺ {𝑣}. Then 𝐷 βŠ† 𝑉(𝑇2(𝐺)) and βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝐷 ⟩ is a tree. Therefore D is a ctd-set of 𝑇2(𝐺) and hence, γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 βˆ’ 1. β–‘ Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 3110 https://internationalpubls.com Theorem 5.5. Let G be a connected graph, if γ𝑐𝑑𝑑(𝐺) = 2 then 𝛾𝑐𝑑𝑑(𝑇2(𝐺)) = { 𝑝 𝑖𝑓 𝐺 β‰… 𝐺1&𝐺2 𝑝 βˆ’ 1 𝑖𝑓 𝐺 β‰… 𝐺3 . (i) 𝐺1 is the graph obtained from 𝐾1 + 𝑇 with one pendant edge attached at the vertex of 𝐾1, where 𝑇 is any tree with 𝑝 βˆ’ 2 vertices. (ii) 𝐺2 is the graph obtained from a tree T where (|𝑇| = 𝑝 βˆ’ 2) by joining each of the vertices of the tree to the vertices of 𝐾2 such that deg𝐺(𝑣) β‰₯ 2 for all 𝑣 ∈ 𝑉(𝐾2). (iii) 𝐺3 is the graph obtained from a tree by joining each of the vertices of the tree to the vertices of 2𝐾1 such that deg𝐺(𝑣) β‰₯ 1 for all 𝑣 ∈ 𝑉(2𝐾1) and |𝑉(𝐺)| = 𝑝. Proof. Assume γ𝑐𝑑𝑑(𝐺) = 2. Let G be a connected graph with 𝑝 β‰₯ 4 and 𝑆1 = {𝑒1, 𝑒2} is a minimum ctd-set of G then βŸ¨π‘‰(𝐺) βˆ’ 𝑆1⟩ is a tree 𝑇. Hence |𝑇| = |𝑉(𝐺) βˆ’ 𝑆1| = 𝑝 βˆ’ 2. Now construct 𝑇2(𝐺), the vertices of complementary dominating set of G (tree T) and its edge vertex forms a cycle 𝐢3. Let𝑆2 = {𝑣𝑖𝑗 β€² /𝑖 β‰  𝑗, 𝑖 = 1,2, … , 𝑝 βˆ’ 3, 𝑗 = 1,2, … , 𝑝 βˆ’ 2} βŠ† 𝑉(𝑇2(𝐺))where(𝑣𝑖 , 𝑣𝑗) ∈ 𝐸(𝑉(𝐺) βˆ’ 𝑆1). Case 1. 𝑒1 and 𝑒2 are connected then 𝐺 β‰… 𝐺1 or 𝐺2. Let 𝐷 = 𝑆1 βˆͺ 𝑆2 βˆͺ {𝑒12 β€² } is a minimum ctd-set of 𝑇2(𝐺1) or 𝑇2(𝐺2). Hence |𝐷| = |𝑆1| + |𝑆2| + 1 = 𝑝.Therefore, γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 if 𝐺 β‰… 𝐺1 or 𝐺2. Case 2. 𝑒1 and 𝑒2 are not connected. Then 𝐺 β‰… 𝐺3. Let 𝐷 = 𝑆1 βˆͺ 𝑆2 is a minimum ctd-set of 𝑇2(𝐺3). Hence |𝐷| = |𝑆1| + |𝑆2| = 𝑝 βˆ’ 1. Therefore, γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 βˆ’ 1 if 𝐺 β‰… 𝐺3. β–‘ Theorem 5.6. Let G be a connected graph with p vertices (𝑝 β‰₯ 3), 𝑉(𝑇2(𝐺)) = 𝑉(𝐺) βˆͺ 𝑉′(𝐺) then, 𝑉′(𝐺) = 𝑉(𝑇2(𝐺)) βˆ’ 𝑉(𝐺) is a ctd-set of 𝑇2(𝐺) if and only if G is a tree. Proof. Assume 𝑉′(𝐺) is a ctd-set of 𝑇2(𝐺). Then, each vertex in 𝑉(𝑇2(𝐺)) βˆ’ 𝑉(𝐺) is adjacent to atleast one vertex in 𝑉′(𝐺) and βŸ¨π‘‰(𝑇2(𝐺)) βˆ’ 𝑉′(𝐺) ⟩ is a tree. That is βŸ¨π‘‰(𝐺)⟩ is a tree. Conversely, Assume G is a tree.Let𝐷 = 𝑉′(𝐺) that is D contains all the edge vertices of G. Since G is connected each vertex v in 𝑉(𝑇2(𝐺)) βˆ’ 𝐷 = 𝑉(𝐺) forms a cycle 𝐢3 in 𝑇2(𝐺) and 𝑉(𝑇2(𝐺)) βˆ’ 𝐷 = 𝑉(𝐺) is a tree. Hence, D is a ctd-set of G. β–‘ Theorem 5.7. For any connected (𝑝, π‘ž) graph G, γ𝑐𝑑𝑑(𝑇2(𝐺)) + Ξ”(𝐺) = 2𝑝 βˆ’ 2 or 𝑝 + π‘ž βˆ’ 1 if and only if 𝐺 β‰… 𝐾1,π‘βˆ’1β€ˆ(𝑝 β‰₯ 4). Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 3111 https://internationalpubls.com Proof. When 𝐺 β‰… 𝐾1,π‘βˆ’1, γ𝑐𝑑𝑑(𝑇2(𝐺)) + Ξ”(𝐺) = 𝑝 + π‘ž βˆ’ 1. Conversely, γ𝑐𝑑𝑑(𝑇2(𝐺)) + Ξ”(𝐺) = 2𝑝 βˆ’ 2 is possible if γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 βˆ’ 1 and Ξ”(𝐺) = 𝑝 βˆ’ 1 is possible only if G is a star on p vertices. β–‘ Theorem 5.8. For any connected (𝑝, π‘ž) graph G, γ𝑐𝑑𝑑(𝑇2(𝐺)) + Ξ”(𝐺) = 2𝑝 βˆ’ 𝑛 (𝑝 β‰₯ 4) where 𝑛 = π‘‘π‘–π‘Žπ‘š(𝐺), 𝑛 β‰₯ 2 if 𝐺 β‰… broom graph. Proof. For the graphs given in the theorem Ξ”(𝐺) = 𝑝 βˆ’ 𝑛, γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 then γ𝑐𝑑𝑑(𝑇2(𝐺)) + Ξ”(𝐺) = 2𝑝 βˆ’ 𝑛. Conversely, γ𝑐𝑑𝑑(𝑇2(𝐺)) + Ξ”(𝐺) = 2𝑝 βˆ’ 𝑛 only possible if (i) γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 βˆ’ 1 and Ξ”(𝐺) = 𝑝 βˆ’ (𝑛 βˆ’ 1) in this case γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 βˆ’ 1 if and only if G is a tree on p vertices. But for a star Ξ”(𝐺) = 𝑝 βˆ’ 1 and π‘‘π‘–π‘Žπ‘š(𝐺) = 2. (ii) If G is a broom graph on p vertices with path of length 2, Ξ”(𝐺) = 𝑝 βˆ’ 2 and π‘‘π‘–π‘Žπ‘š(𝐺) = 3. γ𝑐𝑑𝑑(𝑇2(𝐺)) + Ξ”(𝐺) = 𝑝 βˆ’ 1 + 𝑝 βˆ’ 2 = 2𝑝 βˆ’ 3 Therefore G is a broom graph of length 𝑛 βˆ’ 1.i.e., 𝑛 = π‘‘π‘–π‘Žπ‘š(𝐺)(𝑛 β‰₯ 2). β–‘ Theorem 5.9. Let 𝑑1 and 𝑑2 be two trees with order 𝑝1 β‰₯ 2 and 𝑝2 β‰₯ 2 respectively. Then γ𝑐𝑑𝑑(𝑇2(𝑑1 ∘ 𝑑2)) ≀ 2γ𝑐𝑑𝑑(𝑑1 ∘ 𝑑2) + 𝑝1(1 + 𝑝2) βˆ’ 1. Proof. We have γ𝑐𝑑𝑑(𝐺) ≀ 𝑝1(𝑝2 βˆ’ 1)[4]. Let D be the minimum ctd-set of 𝑑1 ∘ 𝑑2. Hence |𝐷| ≀ 𝑝1(𝑝2 βˆ’ 1). Let 𝐷′ be the number of edge vertices of 𝑇2(𝑑1 ∘ 𝑑2). Then 𝐷 βˆͺ 𝐷′ βŠ† 𝑉(𝑇2(𝐺)) is a minimum ctd- set of 𝑇2(𝐺). Therefore, γ𝑐𝑑𝑑(𝑇2(𝐺)) ≀ |𝐷 βˆͺ 𝐷′| ≀ 2𝑝1(𝑝2 βˆ’ 1) + 𝑝1(1 + 𝑝2) βˆ’ 1 ≀ 2γ𝑐𝑑𝑑(𝐺) + 𝑝1(1 + 𝑝2) βˆ’ 1. β–‘ Theorem 5.10. Let G be a (𝑝, π‘ž), 𝑝 β‰₯ 5, graph such that both G and 𝐺 are connected then (i) 8 ≀ γ𝑐𝑑𝑑(𝑇2(𝐺)) + γ𝑐𝑑𝑑 (𝑇2(𝐺)) ≀ 2(𝑝 + π‘ž βˆ’ 4) (ii) 4 ≀ γ𝑐𝑑𝑑(𝑇2(𝐺)). γ𝑐𝑑𝑑 (𝑇2(𝐺)) ≀ (𝑝 + π‘ž βˆ’ 4)2 Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 3112 https://internationalpubls.com Proof. By Theorem 3.4 γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 + π‘ž βˆ’ 2 if and only if G is a graph 𝐾2. But in this case 𝐺 is disconnected. γ𝑐𝑑𝑑(𝑇2(𝐺)) = 𝑝 + π‘ž βˆ’ 3 if and only if G is the graph 𝑃3&𝐢4. But in this case 𝐺 is disconnected. Therefore, γ𝑐𝑑𝑑(𝑇2(𝐺)) ≀ 𝑝 + π‘ž βˆ’ 4. Hence,γ𝑐𝑑𝑑(𝑇2(𝐺)) + γ𝑐𝑑𝑑 (𝑇2(𝐺)) ≀ 2(𝑝 + π‘ž βˆ’ 4). For lower bound γ𝑐𝑑𝑑(𝑇2(𝐺)) = 4 if and only if 𝐺 β‰… 𝑃4 and 𝐢5. In this case 𝐺 is connected. Hence γ𝑐𝑑𝑑(𝐺) + γ𝑐𝑑𝑑(𝐺) β‰₯ 4. (ii) follows similarly. β–‘ Refrences [1] B. Basavanagoud, S.M. Hosamani, and S.H. Malghan. β€œDomination insemi total-point graph”. J. Comp. & Math. Sci., 1(5):598–605, 2010. [2] F. Harary. β€œGraph Theory”, Addison-Wesley, Reading Mass. 1972. [3] S. Muthammai, M. Bhanumathi and P. Vidhya, β€œComplementary tree domination number of a graph”. International Mathematical Forum, 6(26):1273βˆ’1282, 2011. [4] S. Muthammai and P. Vidhya, β€œMore Results on Complementary Tree Domination Number of Graphs”,International Journal of Mathematics And its Applications,Vol. 4, No. 1-D, pp. 17-20, 2016. [5] S. Muthammai and P. Vidhya β€œComplementary Tree Domination in Splitting Graphs of Graphs”,International Journal of Mathematics Trends and Technology Vol. 31 No.2,pp. 53-56, 2016. [6] O. Ore. β€œTheory of Graphs”, Amer. Math Soc. Colloq. Publ., Providence.38, 1962. [7] E. Sampath Kumar and S.B. Chikkodimath. β€œSemi-total graphs of a graph I”. Journal of The Karnatak University-Science, XVIII:274–280, 1973. [8] P. Vidhya, S. Jayalakshmi, and G. Mahalakshmi. β€œComplementary tree domination number of semi total point graph”. The International Journal of Analytical and Experimental Modal Analysis, XII:682–689, 2020.