Abstract
We examine the p-nilpotency and supersolvablitiy of a finite group under the assumption that certain subgroups of prime power order are \(S_*\)- embedded in the group itself. In particular, we extend and generalize the recent results of Li (Commun. Algebra 50(4):1585–1594, https://doi.org/10.1080/00927872.2021.1986056, 2022) and the related results in the literature.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
All the groups are considered to be finite and G always stands for a finite group. \(G_{p}\) denotes a Sylow p-subgroup of G where p is a prime number. We use conventional notions and notation as in [1]. Recall that a subgroup A of G is said to be S-permutable in G if \(AG_{p}=G_{p}A\) for every Sylow p-subgroup \(G_{p}\) of G [8]. Zhang and Wang [18] called a subgroup A of G S-semipermutable in G if A permutes with every Sylow p-subgroup \(G_{p}\) of G such that \((p,|A|)=1\). Ballester-Bolinches and Pedraza-Aguilera [2] called a subgroup A of G S-permutably embedded in G if all Sylow p-subgroups of A are also Sylow p-subgroups of some S-permutable subgroup F of G. Obviously, the class of all S-permutably embedded subgroups is wider than the class of all S-permutable subgroups. More recently, Li [9] introduced a new embedding property which covers both of S-permutably embedded and S-semipermutable concepts as follows: A subgroup A of G is said to be \(S_*\)-embedded in G if G has an S-permutable subgroup F such that AF is S-permutable in G and \(A \cap F \le A_*\), where \(A_*\) is a subgroup contained in A which is either S-permutably embedded or S-semipermutable in G.
By using this new concept, Li [9] studied the group structure when certain subgroups are \(S_*\)-embedded and obtained new results generalised many classical and recent results in the literature. More precisely, he proved:
Theorem A
([8, Main Theorem]) Assume for each non-cyclic Sylow p-subgroup \(G_{p}\) of G, either of the following two conditions is held:
-
1.
All maximal subgroups of \(G_{p}\) not having a supersolvable supplement in G are \(S_*\)-embedded in G.
-
2.
All cyclic subgroups of \(G_{p}\) of prime order or of order 4, that are without a supersolvable supplement in G, are \(S_*\)-embedded in G.
Then, G is supersolvable.
Our main object in this paper is to go further in studying the influence of \(S_*\)- embedded subgroups on the group structure. In fact, we prove:
Theorem B
Suppose \(A\mathrel {\unlhd } G\) such that G/A is p-nilpotent for some prime divisor p of |G|. If A has a Sylow p-subgroup \(A_{p}\) such that \(N_G(A_{p})\) is p-nilpotent and all maximal subgroups of \(A_{p}\), that are without a p-nilpotent supplement in G, are \(S_*\)-embedded in G, then G is p-nilpotent.
Theorem C
Suppose \(A \mathrel {\unlhd }G\) such that G/A is supersolvable. If all maximal subgroups of each non-cyclic Sylow subgroup of A, that are without a supersolvable supplement in G, are \(S_*\)-embedded in G, then G is supersolvable.
Theorem D
Suppose \(A\mathrel {\unlhd }G\) such that G/A is supersolvable. If all cyclic subgroups of each non-cyclic Sylow subgroup of A of prime order or of order 4, that are without a supersolvable supplement in G, are \(S_*\)-embedded in G, then G is supersolvable.
2 Preliminaries
Lemma 1
([9, Theorem 3.2]) Assume p is some prime divisor of |G| satisfying \((|G|,p-1)=1\), and \(G_{p}\) is a Sylow p-subgroup of G. If all maximal subgroups of \(G_{p}\), that are without a p-nilpotent supplement in G, are \(S_*\)-embedded in G, then G is p-nilpotent.
Lemma 2
([9, Lemma 2.5]) Let A be an \(S_*\)-embedded p-subgroup of G.
-
1.
If \(A \le B \le G\), then A is \(S_*\)-embedded in B.
-
2.
If \(L \mathrel {\unlhd } G\) and \(L \le A\), then A/L is \(S_*\)-embedded in G/L.
-
3.
If \(L \mathrel {\unlhd } G\) and \((|L|,|A|)=1\), then AL/L is \(S_*\)-embedded in G/L.
-
4.
If \(L \mathrel {\unlhd } G\) and \(A \le L\), then G has an S-permutable subgroup F contained in L such that AF is S-permutable in G and \(A \cap F \le A_*\).
Lemma 3
([12, Theorem A]) Suppose that A is an S-permutable p-subgroup of G. Then \(O^p(G)\) is contained in \(N_G(A)\).
Lemma 4
([17, Lemma 2.1(d)]) Suppose that A is a p-subgroup of G and B is a normal p-subgroup of G. If A is either S-permutably embedded or S-semipermutable in G, then \(A \cap B\) is also S-permutable in G.
Lemma 5
Let A be a subgroup of G.
-
1.
[3] If A is S-permutable in G, then \(A/A_G\) is nilpotent.
-
2.
[8] If A is S-permutable in G, then A is subnormal in G.
-
3.
[16] If A is a subnormal p-subgroup of G, then A is contained in \(O_p(G)\).
-
4.
[13] \(A_{SG}\) is an S-permutable subgroup of G, where \(A_{SG}\) is a subgroup of A generated by all S-permutable subgroups of G that contained in A.
-
5.
[12] If A and B are S-permutable subgroups of G, then \(A \cap B\) is also S-permutable in G.
Lemma 6
([10, 18]) Suppose that A is either S-permutably embedded or S-semipermutable in G. If A is a p-subgroup contained in \(O_p(G)\), then A is S-permutable in G.
Lemma 7
([9, Lemma 2.1(3)]) Suppose that p is some prime divisor of |G| satisfying \((|G|,p-1)=1\). If G is p-supersolvable, then G is p-nilpotent.
Lemma 8
([15, Lemma 2.4]) If A is a maximal subgroup of G and B is a normal p-subgroup of G such that \(G=AB\), then \(A\cap B \mathrel {\unlhd } G\).
Lemma 9
([11, Theorem 3.5]) If \(A \mathrel {\unlhd } G\) such that G/A is supersolvable and all maximal subgroups of any Sylow subgroup of A are normal in G, then G is supersolvable.
Lemma 10
([8]) Let A be an S-permutable subgroup of G.
-
1.
If \(A \le B \le G\), then A is S-permutable in B.
-
2.
If \(L \mathrel {\unlhd } G\), then AL/L is S-permutable in G/L.
Lemma 11
([5]) Suppose that A and B are normal supersolvable subgroups of G with \(G=AB\). If the indices |G : A| and |G : B| are relatively prime, then G is supresolvable.
Lemma 12
([16]) If A is a subnormal subgroup of G such that the number |G : A| is not divisible by p, then every Sylow p-subgroup \(G_{p}\) of G is contained in A.
3 Proofs
Theorem 1
Suppose \(G_{p}\) is a Sylow p-subgroup of G for some prime divisor p of |G|. If \(N_{G}(G_{p})\) is p-nilpotent and all maximal subgroups of \(G_{p}\), that are without a p-nilpotent supplement in G, are \(S_*\)-embedded in G, then G is p-nilpotent.
Proof
If \(p=2\), then, by Lemma 1, G is p-nilpotent. So, it can be assumed that \(p>2\). Suppose the result is not true providing G as a counterexample of minimal order. Following the method of the first part of the proof of [6, Theorem 2.3], we conclude that all maximal subgroups of \(G_{p}\) are \(S_*\)-embedded in G. Now, we build up the proof by the following steps:
-
(1)
\(O_{p'}(G)=1\).
Assume that \(O_{p'}(G) \ne 1\). Let \(X/O_{p'}(G)\) be a maximal subgroup of the Sylow p-subgroup \(G_{p}O_{p'}(G)/O_{p'}(G)\) of \(G/O_{p'}(G)\). Then \(G_{p}\) has a maximal subgroup \(X_1\) such that \(X=X_1O_{p'}(G)\). Since \(X_1\) is \(S_*\)-embedded in G, it follows that \(X_1O_{p'}(G)/O_{p'}(G)\) is \(S_*\)-embedded in \(G/O_{p'}(G)\) by Lemma 2(3). Also, we have \(N_{G/O_{p'}(G)} (G_{p}O_{p'}(G)/O_{p'}(G)) = N_{G}(G_{p})O_{p'}(G)/O_{p'}(G)\) is p-nilpotent. Our choice of G implies that \(G/O_{p'}(G)\) is p-nilpotent. Consequently, G is p-nilpotent which is a contradiction.
-
(2)
If \(G_{p} \le A < G\), then A is p-nilpotent.
Clearly, \(N_A(G_{p}) \le N_G(G_{p})\) is p-nilpotent, and by Lemma 2(1) all maximal subgroups of \(G_{p}\) are \(S_*\)-embedded in A. The minimal choice of G implies that A is p-nilpotent.
-
(3)
G is p-solvable.
By Thompson’s result [14, Corollary], \(G_{p}\) has a non-trivial characteristic subgroup R such that \(N_{G}(R)\) is not p-nilpotent. Let L be any characteristic subgroup of \(G_{p}\) such that \(O_p(G)< L\le G_{p}\). Since L char \(G_{p} \mathrel {\unlhd } N_G(G_{p})\), we have \(L \mathrel {\unlhd }N_G(G_{p})\). Hence \(G_{p} \le N_G(G_{p}) \le N_G(L)<G\) which implies that \(N_G(L)\) is p-nilpotent by step (2). If \(O_p(G)=1\), then \(1<R\le G_{p}\) and hence \(N_G(R)\) is p-nilpotent; a contradiction. Thus, we may assume that \(O_p(G) \ne 1\). Clearly, \(N_{G/O_p(G)}(G_{p}/O_p(G)) = N_G(G_{p})/O_p(G)\) is p-nilpotent. By using Lemma 2(2), it is easy to see that \(G/O_p(G)\) satisfies the hypothesis of the theorem. Our choice of G implies that \(G/O_p(G)\) is p-nilpotent, and thereby G is p-solvable.
-
(4)
There exists a unique minimal normal subgroup L of G such that G/L is p-nilpotent and \(G=L \rtimes X\), where X is a maximal subgroup of G. Moreover, \(L= C_G(L)=F(G)=O_p(G)\).
Let L be a minimal normal subgroup of G. Then L is an elementary abelian p-group by steps (1) and (3). This implies \(L \le O_p(G) \le G_{p}\). If \(L=G_{p}\), then clearly, G/L is p-nilpotent. Thus, we may assume that \(L<G_{p}\). In view of Lemma 2(2), we can see that the hypothesis still holds for G/L. Our choice of G implies that G/L is also p-nilpotent. Since the class of all p-nilpotent groups is a saturated formation, it follows that L is the unique minimal normal subgroup of G and \(L \nleq \Phi (G)\). Hence, there exists a maximal subgroup X of G such that \(G= L \rtimes X\), \(L= C_G(L)=F(G)=O_p(G)\).
-
(5)
Finishing the proof.
Let \(X_p\) be a Sylow p-subgroup of X such that \(G_{p}=LX_p\), and \(P_{1}\) be a maximal subgroup of \(G_{p}\) containing \(X_p\). Then \(G_{p}=LX_p=LP_1\). We may assume that \(L \ne G_{p}\) (Otherwise, \(G=N_{G}(L)=N_{G}(G_{p})\) is p-nilpotent; a contradiction). Since \(P_1\) is \(S_*\)-embedded in G, then there exists an S-permutable subgroup F of G such that \(P_1F\) is S-permutable in G and \((P_1 \cap F)\le (P_1)_*\). If \(F=1\), then \(P_1\) is S-permutable in G which implies that \(P_1 \mathrel {\unlhd } G_{p}O^p(G)=G\) by using Lemma 3 and since \(P_{1} \ne 1\), we have \(L \le P_{1}\) by step (4) which means \(P_1=G_{p}\); a contradiction. Thus, we may assume that \(F \ne 1\). Assume that \(F_G \ne 1\). Then \(L\le F_G\le F\) which implies that \(P_1\cap L \le P_1\cap F\le (P_1)_*\cap L\) and so \(P_1\cap L=(P_1)_*\cap L\). Lemma 4 yields \(P_1\cap L\) is S-permutable in G. If \(P_1\cap L \ne 1\), then \(L \le (P_1 \cap L)^{O^p(G)G_{p}} \le (P_1)^{G_{p}}=P_1\) by Lemma 3; a contradiction. So, assume \(P_1\cap L=1\). This implies that \(|L|=p\). Since \(X \cong G/L=G/C_G(L) \lesssim Aut(L) \), we have \(|X|\ |\ (p-1)\). It follows that L is a Sylow p-subgroup of G. Hence \(L=G_{p}\); a contradiction. Thus, \(F_G=1\). In view of Lemma 5(1), we have F is a nilpotent group. Then from step (1), F is a p-subgroup. By Lemma 5(2) and Lemma 5(3), we have \((P_1)_*\le P_1\le P_1F\le O_p(G)\) which implies that \((P_1)_*\le (P_1)_{sG}\) by using Lemma 6 and Lemma 5(4). Now, we have \(P_1 \cap F\le (P_1)_*\le (P_1)_{sG}\) which implies \(P_1\cap F \le (P_1)_{sG}\cap F\). Hence, \(P_1\cap F= (P_1)_{sG}\cap F\). In view of Lemma 5(5), we have \(P_1 \cap F\) is S-permutable in G. If \(P_1 \cap F=1\), then \(|F|=p\). It follows that \(P_1F\) is a Sylow p-subgroup of G. Hence, \(P_1F=G_{p}=O_p(G)=L\); a contradiction. Thus, it can be assumed that \(P_1\cap F\ne 1\). Then, by using Lemma 3 and step (4), we have \(L\le (P_1\cap F)^G =(P_1\cap F)^{O^p(G)G_{p}} \le (P_1)^{G_{p}}=P_1\). Hence \(G_{p}=LP_1=P_1\); a final contradiction.\(\square \)
Proof of Theorem B
Assume the result is not true providing G as a counterexample of minimal order. Lemma 2(1) with Theorem 1, imply A is p-nilpotent. Let \(A_{p'}\) be the normal p-complement of A. Since \(A_{p'}\) char \(A \mathrel {\unlhd } G\), then \(A_{p'} \mathrel {\unlhd } G\). Assume that \(|A_{p'}|>1\). Clearly, \(A/A_{p'} \mathrel {\unlhd } G/A_{p'}\) and \((G/A_{p'})/(A/A_{p'}) \cong G/A \) is p- nilpotent. Let \(X/A_{p'}\) be a maximal subgroup of the Sylow p-subgroup \(A_{p}A_{p'}/A_{p'}\) of \(A/A_{p'}\). Then \(A_{p}\) has a maximal subgroup \(X_1\) such that \(X=X_1A_{p'}\). If \(X_1\) possesses a p-nilpotent supplement D in G, then \(DA_{p'}/A_{p'}\) is a p-nilpotent supplement of \(X_1A_{p'}/A_{p'}\) in \(G/A_{p'}\). If \(X_1\) is \(S_*\)-embedded in G, it follows that \(X_1A_{p'}/A_{p'}\) is \(S_*\)-embedded in \(G/A_{p'}\) by using Lemma 2(3). Our choice of G implies that \(G/A_{p'}\) is p-nilpotent. Consequently, G is p-nilpotent which is a contradiction. Thus, it can be assumed \(A_{p'}=1\). Then \(A=A_{p}\), which implies \(G=N_G(A)=N_G(A_{p})\) is p-nilpotent; a final contradiction. \(\square \)
Remark 1
The condition \(N_G(G_{p})\) is p-nilpotent in Theorem 1 and Theorem B is necessary. For example, consider \(G=A_5\) and \(p=3\). Then all maximal subgroups of any Sylow 3-subgroup \(G_{3}\) of G are \(S_*\)-embedded in G, but G is not 3-nilpotent.
We work toward the proof of Theorem C:
Theorem 2
Suppose that p is some prime divisor of |G| satisfies \((|G|,p-1)=1\), and \(A \mathrel {\unlhd } G\) such that G/A is p-nilpotent. If A has a Sylow p-subgroup \(A_{p}\) such that all maximal subgroups of \(A_{p}\), that are without a p-nilpotent supplement in G, are \(S_*\)-embedded in G, then G is p-nilpotent.
Proof
Assume the result is not true providing G as a counterexample of minimal order. By Lemma 2(1) and Lemma 1, we have that A is p-nilpotent. Let \(A_{p^{'}}\) be the normal p-complement of A. Since \(A_{p^{'}}\) char \(A \mathrel {\unlhd } G\), then \({A_{p^{'}}} \mathrel {\unlhd } G\). Assume that \(|{A_{p^{'}}}|>1\). Clearly, \({A/A_{p^{'}}} \mathrel {\unlhd } {G/A_{p^{'}}}\) and \(({G/A_{p^{'}}})/({A/A_{p^{'}}}) \cong G/A\) is p-nilpotent. Let \({X/A_{p^{'}}}\) be a maximal subgroup of the Sylow p-subgroup \({A_{p}A_{p^{'}}}/{A_{p^{'}}}\) of \({A/A_{p^{'}}}\). Then \(A_{p}\) has a maximal subgroup \(X_1\) such that \(X=X_1{A_{p^{'}}}\). If \(X_1\) possesses a p-nilpotent supplement D in G, then \(D{A_{p^{'}}}/{A_{p^{'}}}\) is a p-nilpotent supplement of \(X_1{A_{p^{'}}}/{A_{p^{'}}}\) in \(G/{A_{p^{'}}}\). If \(X_1\) is \(S_*\)-embedded in G, then \(X_1{A_{p^{'}}}/{A_{p^{'}}}\) is \(S_*\)-embedded in \(G/{A_{p^{'}}}\) by Lemma 2(3). So \(G/{A_{p^{'}}}\) is p-nilpotent due to the minimal choice of G. It follows, G is p-nilpotent; a contradiction. Thus, it can be assumed \({A_{p^{'}}}=1\) which yields \(A=A_{p}\) is a p-group. Let \(L/A_{p}\) be the normal p-complement of \(G/A_{p}\). Schur-Zassenhaus Theorem implies that L has a Hall \(p^{'}\)-subgroup \(L_{p^{'}}\) such that \(L=A_{p} \rtimes {L_{p^{'}}} \). Since L is p-nilpotent by Lemma 2(1) and Lemma 1, it follows that \(L=A_{p} \times {L_{p^{'}}} \). Hence, we have \(L_{p^{'}}\) is the normal p-complement of G, and thereby G is p-nilpotent; a contradiction. \(\square \)
Lemma 7 and Theorem 2 lead to the following corollary:
Corollary 1
Suppose \(A \mathrel {\unlhd } G\) provided G/A is p-nilpotent, where p is the smallest prime divisor of |G|. If A has a Sylow p-subgroup \(A_{p}\) such that all maximal subgroups of \(A_{p}\) that are without a p-supersolvable supplement in G are \(S_*\)-embedded in G, then G is p-nilpotent.
Proof
Clearly, \((|G|,p-1)=1\) as p is the smallest prime divisor of |G|. Lemma 7 suggests all maximal subgroups of \(A_{p}\) that are without a p-nilpotent supplement in G are \(S_*\)-embedded in G. Using Theorem 2, gives G is p-nilpotent. \(\square \)
Now we can prove Theorem C:
Proof of Theorem C
Assume the result is not true providing G as a counterexample of minimal order. In view of Theorem A(1), we have A is supersolvable. Let \(G_{p}\) be a Sylow p-subgroup of G, where p is the largest prime divisor of |G|. We distinguish two cases.
-
Case 1.
\(G_{p} \le A\).
Then \(G_{p} \mathrel {\unlhd } A\) as A is supersolvable. Since \(G_{p}\) char \(A \mathrel {\unlhd }G\), it follows that \(G_{p} \mathrel {\unlhd } G\). Now, we show that \(G/G_{p}\) is supersolvable. Clearly, \((A/G_{p}) \mathrel {\unlhd } (G/G_{p})\) and \((G/G_{p})/(A/G_{p}) \cong (G/A)\) is supersolvable. Let \(X/G_{p}\) be a maximal subgroup of the Sylow q-subgroup \(A_{q}G_{p}/G_{p}\) of \(A/G_{p}\). Then \(A_{q}\) has a maximal subgroup \(X_1\) such that \(X=X_1G_{p}\). If \(X_1\) has a supersolvable supplement B in G, then \(BG_{p}/G_{p}\) is a supersolvable supplement of \(X_1G_{p}/G_{p}\) in \(G/G_{p}\). If \(X_1\) is \(S_*\)-embedded in G, then \(X_1G_{p}/G_{p}\) is \(S_*\)-embedded in \(G/G_{p}\) by using Lemma 2(3). The minimal choice of G yields \(G/G_{p}\) is supersolvable and \(G_{p}\) is not cyclic. Let N be a minimal normal subgroup of G contained in \(G_{p}\). It is also easy to see that G/N is supersolvable. Further, since the class of all supersolvable groups is a saturated formation, it follows that N is the unique minimal normal subgroup of G contained in \(G_{p}\) and \(N \nleq \Phi (G)\). Hence G possesses a maximal subgroup X such that \(G=NX\) and \(N \cap X=1\). Since \(G_{p} \cap X\) is normalized by G by using Lemma 8, it follows that \(N=G_{p}\) which yields \(G_{p}\) is an elementray abelian p-group. Now, Let \(N_1\) be a maximal subgroup of N. If \(N_1\) has a supersolvable supplement B in G, then \(G=N_1B=NB\) and \(N=N \cap N_1B=N_1(N\cap B)\), which implies that \(N \cap B \ne 1\). Since \(N \cap B \mathrel {\unlhd } G\) and N is a minimal normal subgroup of G, we have \(N \cap B=N\). Consequently, \(N \le B\) which implies that \(G=B\) is supersolvable which is a contradiction. Thus, we can assume that \(N_1\) is \(S_*\)-embedded in G. In view of Lemma 2(4), G possesses an S-permutable subgroup F contained in \(G_{p}\) such that \(N_1F\) is S-permutable in G and \(N_1 \cap F \le (N_1)_*\). If \(F=1\), then \(N_1\) is S-permutable in G and \(N_1 \mathrel {\unlhd } G_{p}O^p(G)=G\) by using Lemma 3 and so \(|G_{p}|=p\); a contradiction. Thus, \(F \ne 1\). Since \(G_{p}\) is an elemntary abelian p-group, then \(F \mathrel {\unlhd } G_{p}\). Applying Lemma 3 again, we get \(F \mathrel {\unlhd } G_{p}O^p(G)=G\). This implies \(F=G_{p}=N\). Hence \(N_1 \cap F=N_1=(N_1)_*\), is S-permutable in G by Lemma 6; again a contradiction as above.
-
Case 2.
\(G_{p} \nleq A\).
In this case we distinguish the following two subcases.
-
Subcase (i).
\(G_{p}A<G\).
Clearly, \(A \mathrel {\unlhd } G_{p}A\) and \(G_{p}A/A \cong G_{p}/G_{p}\cap A\) is supersolvable. By Lemma 2(1), all maximal subgroups of any non-cyclic Sylow subgroup of A not having a supersolvable supplement in G are \(S_*\)-embedded in \(G_{p}A\). Therefore, \(G_{p}A\) is supersolvable due to the minimal choice of G. Since \(G_{p}A/A\) is a Sylow p-subgroup of G/A, where p is the largest prime divisor of |G| and G/A is supersolvable, it follows that \(G_{p}A/A \mathrel {\unlhd } G/A\) and so \(G_{p}A \mathrel {\unlhd } G\). Since \(G_{p}\) char \(G_{p}A \mathrel {\unlhd } G\), then \(G_{p} \mathrel {\unlhd } G\). Hence \(G_{p} \cap A \mathrel {\unlhd } G\), where \(G_{p} \cap A\) is a Sylow p-subgroup of A. By using the same arguments as in Case 1, we have \(G/G_{p}\cap A\) is supersolvable and \(G_{p} \cap A\) is a minimal normal subgroup of G. Set \(N=G_{p} \cap A\). By our choice of G together with Lemma 9, N has a maximal subgroup \(N_1\) such that \(N_1 \ntrianglelefteq G\). If \(N_1\) possesses a supersolvable supplement B in G, we have G is supersolvable as in Case 1; a contradiction. Thus, we can assume that \(N_1\) is \(S_*\)-embedded in G. In view of Lemma 2(4), there exists an S-permutable subgroup F of G contained in N such that \(N_1F\) is S-permutable in G and \(N_1 \cap F \le (N_1)_*\). By the maximality of \(N_1\) in N, we have either \(N_1F=N_1\) or \(N_1F=N\). If the former holds, then \(N_1\) is S-permutable in G. So, \(N_1 \mathrel {\unlhd } NO^p(A)=A\) and hence \(A \le N_G(N_1)\) by using Lemma 10(1) and Lemma 3. Applying Lemma 3 again, we get \(O^p(G) \le N_G(N_1)<G\). Hence G possesses a maximal subgroup X such that \(O^p(G) \le N_G(N_1) \le X<G\) with \(|G:X|=p\). Since \(X/O^p(G) \mathrel {\unlhd } G/O^p(G)\), then \(X \mathrel {\unlhd } G\). Due to the minimal choice of G, we have X is supersolvable. Therefore, \(G=G_{p}X\) and consequently G is supersolvable by using Lemma 11; a contradiction. Thus, we can assume that \(N_1F=N\). If \(F \mathrel {\unlhd } G\), then \(F=N\) which implies \(N_1 \cap F=N_1=(N_1)_*\) is S-permutable in G by Lemma 6; a contradiction as above. Thus, \(F \ntrianglelefteq G\). Since N is an elementary abelian p-group, then \(F \mathrel {\unlhd } N\). Hence \(F \mathrel {\unlhd } NO^p(A)=A\) by Lemma 10(1) and Lemma 3 and so \(A \le N_G(F)\). Applying Lemma 3 again, there exists a maximal normal subgroup X of G such that \(O^p(G) \le N_G(F) \le X <G\) with \(|G:X|=p\). It follows that X is supersolvable by our choice of G. So, \(G=G_{p}X\) and consequently G is supersolvable again by Lemma 11; a contradiction.
-
Subcase (ii).
\(G=G_{p}A\).
Lemma 12 and Corollary 1 yield A contains all Sylow q-subgroups of G with \(q \ne p\) and G is a \(q_r\)-nilpotent, where \(q_r\) is the smallest prime divisor of |G|. This implies that G has a Sylow tower group of supersolvable type. Therefore, \(G_{p}\mathrel {\unlhd } G\) as p is the largest prime divisor of |G|. Applying Lemma 11, G is supersolvable; a final contradiction.\(\square \)
-
Subcase (i).
In order to show Theorem D, we need the following useful theorem:
Theorem 3
Suppose A is a normal p-subgroup of G such that G/A is supersolvable. If all cyclic subgroups of A of order p or of order 4 not having a supersolvable supplement in G are \(S_*\)-embedded in G, then G is supersolvable.
Proof
Assume the result is not true providing G as a counterexample of minimal order. It is easy to see that all proper subgroups of G are supersolvable by using Lemma 2(1). Hence G is a minimal non-supersolvable group. By Doerk’s result [4] G has a Sylow p-subgroup \(G_{p}\) such that \(G_{p} \mathrel {\unlhd } G\) for a prime divisor p of |G|, \(G_{p}/\Phi (G_{p})\) is a minimal normal subgroup of \(G/\Phi (G_{p})\), and the exponent of \(G_{p}\) is either p or 4. Clearly, \(A\le G_{p}\) (Otherwise, \(G\cong G/G_{p}\cap A \lesssim G/G_{p} \times G/A\) is supersolvable; a contradiction). We build up the proof by the following two steps.
-
(1)
\(A=G_{p}\).
Since \(A\Phi (G_{p})/\Phi (G_{p}) \mathrel {\unlhd } G/\Phi (G_{p})\), we have either \(A\Phi (G_{p})=G_{p}\) or \(A\le \Phi (G_{p})\). If the latter holds, then \(G/\Phi (G_{p})\) is supersolvable. It follows that, from \(\Phi (G_{p}) \le \Phi (G)\), \(G/\Phi (G)\) is supersolvable and so G is also supersolvable by a well-known result of Huppert [7, p.713]; a contradiction. Thus, \(A\Phi (G_{p})=G_{p}\) and thereby \(A=G_{p}\) as required.
-
(2)
Finishing the proof.
Assume that \(|A/\Phi (A)|=p\). Then there exists x in A such that \(A/ \Phi (A) = <x\Phi (A)>\) which implies that A is cyclic and consequently G is supersolvable which is a contradiction. So, \(|A/\Phi (A)|=p^n\), \(n>1\) and \(A/\Phi (A)=<x_1\Phi (A),x_2\Phi (A),...,x_n\Phi (A)>\) as \(A/ \Phi (A)\) is an elementary abelian p-group. Hence, we have \(A=<x_1,x_2,...,x_n>\). Set \(A_i=<x_i>\) for all \(i=1,2,...,n\). So, we have \(|A_i|=p \) or 4. Now, the hypothesis of the theorem assures that \(A_i\) either has a supersolvable supplement in G say B or \(A_i\) is \(S_*\)-embedded in G. If \(A_i\) is not \(S_*\)-embedded in G, then \(G=A_iB\) and so \(A=A\cap G=A\cap A_iB=A_i(A\cap B)\). Obviously, \({(A\cap B)\Phi (A)}/\Phi (A) \mathrel {\unlhd } G/\Phi (A)\) as \(A/\Phi (A)\) is abelian. In view of step (1), \(A/\Phi (A)\) is a minimal normal subgroup of \( G/\Phi (A)\) which implies that either \((A\cap B)\Phi (A)=A\) or \((A\cap B)\le \Phi (A)\). If the latter holds, then \(A=A_i\) is cyclic and so G is supersolvable; a contradiction. Hence \((A\cap B)\Phi (A)=A\) and so \(A\cap B=A\) which implies that \(G=B\) is supersolvable; contradicts our choice of G. Thus, we can assume that \(A_i\) is \(S_*\)-embedded in G. In view of Lemma 2(4), then G possesses an S-permutable subgroup F contained in A such that \(A_iF\) is also S-permutable in G and \(A_i\cap F\le (A_i)_*\). By Lemma 10(2) and the fact that \(A/\Phi (A)\) is abelian, it is easy to see that \({F\Phi (A)}/\Phi (A)\) is S-permutable in \(G/\Phi (A)\) and \({F\Phi (A)}/\Phi (A) \mathrel {\unlhd } A/\Phi (A) \). Applying Lemma 3, we have \({F\Phi (A)}/\Phi (A) \mathrel {\unlhd } (A/\Phi (A))(O^p(G/\Phi (A)) = G/\Phi (A)\). Again, the minimal normality of \(A/\Phi (A)\) in \(G/\Phi (A)\) implies that either \(F\Phi (A)=A\) or \(F\Phi (A)\le \Phi (A)\). If the latter holds, \({A_i\Phi (A)}/\Phi (A)= {A_iF\Phi (A)}/\Phi (A)\) is S-permutable in \(G/\Phi (A)\) by using Lemma 10(2). If \(F\Phi (A)=A\), then we have \(F=A\). Therefore, \(A_i\cap F=A_i=(A_i)_*\) is S-permutable in G by Lemma 6, and so \({A_i\Phi (A)}/\Phi (A)\) is S-permutable in \(G/\Phi (A)\) by Lemma 10(2). By [13, Lemma 2.11], there exists a maximal subgroup \({X\Phi (A)}/\Phi (A)\) of \(A/\Phi (A)\) such that \({X\Phi (A)}/\Phi (A) \mathrel {\unlhd } G/\Phi (A)\); a final contradiction.\(\square \)
Proof of Theorem D
Assume the result is not true providing G as a counterexample of minimal order. If the order of A is of prime power, then G is supersolvable by Theorem 3; a contradiction. Thus, we can assume that the order of A is divisible by at least two distinct primes. By Lemma 2(1) and Theorem A(2), we have A is supersolvable. Hence A possesses a normal Sylow p-subgroup \(A_{p}\), where p is the largest prime divisor of |A|. Since \(A_{p}\) char \(A \mathrel {\unlhd } G\), we have \(A_{p} \mathrel {\unlhd } G\). Let \(U/A_{p}\) be a cyclic subgroup of the Sylow q-subgroup \(A_{q}A_{p}/A_{p}\) of \(A/A_{p}\) such that \(|U/A_{p}|=q\) or 4. Then \(A_{q}\) has a cyclic subgroup R such that \(U=RA_{p}\) and \(|R|=q\) or 4. If R has a supersolvable supplement B in G, then \(BA_{p}/A_{p}\) is a supersolvable supplement of \(RA_{p}/ A_{p}\) in \(G/A_{p}\). If R is \(S_*\)-embedded in G, then \(RA_{p}/A_{p}\) is \(S_*\)-embedded in \(G/A_{p}\) by using Lemma 2(3). Our choice of G yields \(G/A_{p}\) is supersolvable. Applying Theorem 3, we get G is supersolvable; a contradiction. \(\square \)
References
Ballester-Bolinches, A., Esteban-Romero, R., Asaad, M.: Products of Finite Groups. Expositions in Mathematics. Vol. 53, De Gruyter (2010)
Ballester-Bolinches, A., Pedraza-Aguilera, M.C.: Sufficient conditions for supersolvability of finite groups. J. Pure Appl. Algebra. 127(2), 113–118 (1998). https://doi.org/10.1016/S0022-4049(96)00172-7
Deskins, W.E.: On quasinormal subgroups of finite groups. Math. Z. 82(2), 125–132 (1963). https://doi.org/10.1007/BF01111801
Doerk, K.: Minimal nicht uberauflosbare, endliche Gruppen. Math. Z. 91(3), 198–205 (1966). https://doi.org/10.1007/BF01312426
Friesen, D.K.: Products of normal supersolvable subgroups. Proc. Am. Math. Soc. 30(1), 46–48 (1971). https://doi.org/10.2307/2038217
Guo, W., Lu, Y., Niu, W.: S-embedded subgroups of finite groups. Algebra Log. 49(4), 293–304 (2010). https://doi.org/10.1007/s10469-010-9097-2
Huppert, B.: Endiche Gruppen. Springer-Verlag, Berlin (1967)
Kegel, O.H.: Sylow Gruppen und subnormalteiler endlicher Gruppen. Math. Z. 78(1), 205–221 (1962). https://doi.org/10.1007/BF01195169
Li, C.: On \(S_*\)-embedded subgroups of finite groups. Commun. Algebra. 50(4), 1585–1594 (2022). https://doi.org/10.1080/00927872.2021.1986056
Li, Y., Wang, Y., Wei, H.: On \(p\)-nilpotency of finite groups with some subgroups \(\pi \)-quasinormally embedded. Acta Math. Hungar. 108(4), 283–298 (2005). https://doi.org/10.1007/s10474-005-0225-8
Ramadan, M.: Influence of normality on maximal subgroups of Sylow subgroups of a finite group. Acta Math. Hungar. 59, 107–110 (1992). https://doi.org/10.1007/BF00052096
Schmid, P.: Subgroups permutable with all Sylow subgroups. J. Algebra. 207(1), 285–293 (1998). https://doi.org/10.1006/jabr.1998.7429
Skiba, A.N.: On weakly S-permutable subgroups of finite groups. J. Algebra. 315(1), 192–209 (2007). https://doi.org/10.1016/j.jalgebra.2007.04.025
Thompson, J.G.: Normal \(p\)-complements for finite groups. J. Algebra. 1(1), 43–46 (1964). https://doi.org/10.1016/0021-8693(64)90006-7
Wei, H., Wang, Y.: On c*-normality and its properties. J. Group Theory. 10, 211–223 (2007). https://doi.org/10.1515/JGT.2007.017
Wielandt, H.: Subnormal subgroups and permutation groups. Ohio State University, Lectures Note (1971)
Yu, H.: On S-semipermutable or S-permutably embedded subgroups of finite groups. Acta Math. Hungar. 151(1), 173–180 (2017). https://doi.org/10.1007/s10474-016-0674-2
Zhang, Q., Wang, L.: The influence of s-semipermutable subgroups on the structure of a finite group. Acta Math. Sinica (Chin. Ser.). 48, 81–88 (2005)
Acknowledgements
The Deanship of Scientific Research (DSR) at King Abdulaziz University (KAU), Jeddah, Saudi Arabia has funded this project, under grant no. (KEP-PhD: 20-130-1443).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interest
On behave of all authors, the corresponding author states that there is no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Heliel, A.A., Al-Shomrani, M.M., Almestady, M.O. et al. Finite groups with specific \(S_*\)-embedded subgroups. Ricerche mat (2023). https://doi.org/10.1007/s11587-023-00776-y
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11587-023-00776-y