Intersection of conjugate solvable subgroups in symmetric groups
Anton Baykalov

TL;DR
This paper proves that for certain solvable subgroups within almost simple symmetric groups, there exist conjugates whose intersection is trivial, advancing understanding of subgroup intersections in group theory.
Contribution
It establishes the existence of conjugates of solvable subgroups in almost simple groups with trivial intersection, a novel result in the structure of symmetric groups.
Findings
Existence of conjugates with trivial intersection
Applicable to solvable subgroups in symmetric groups
Enhances understanding of subgroup intersection properties
Abstract
It is shown that for a solvable subgroup of an almost simple group which socle is isomorphic to there are such that
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
**Intersection of conjugate solvable subgroups in symmetric groups
Anton Baykalov
**
Novosibirsk State University
Abstract
It is shown that for a solvable subgroup of an almost simple group which socle is isomorphic to there are such that
Kay words: symmetric group, solvable group, almost simple group.
Introduction
Assume that a finite group acts on a set An element is called a -regular point if , i.e. if the stabilizer of is trivial. Define the action of the group on by the rule
[TABLE]
If acts faithfully and transitively on , then the minimal number such that the set contains a -regular point is called the base size of and is denoted by For a positive integer the number of -regular orbits on is denoted by (this number equals 0 if ). If is a subgroup of and acts by the right multiplication on the set of the right cosets of then acts faithfully and transitively on (Here is the core of .) In this case, we denote and by and respectively. Thus is the minimal number such that there exist
for which
Consider Problem 17.41 from “Kourovka notebook”[1]:
** Problem 1****.**
Let be a solvable subgroup of a finite group and does not contain nontrivial normal solvable subgroups. Are there always exist five subgroups conjugated with such that their intersection is trivial?
The problem is reduced to the case when is almost simple in [2]. Specifically, it is proved that if for each almost simple group and solvable subgroup of condition holds then for each finite nonsolvable group with trivial solvable radical and solvable subgroup of condition holds.
We have proved the following
** Theorem 1****.**
Let be a solvable subgroup of an almost simple group whose socle is isomorphic to , Then In particular
The proof contains calculations in GAP system. Texts of programs can be found here:
1 Preliminary results
Our totation is standard.
By we denote the semidirect product of groups and , where acts on , by the permutation wreath product of groups and , where .
If group acts on the a and is an invariant subset, then denote projection of on by .
** Lemma 1****.**
[2, Lemma 13] * Let be a finite group with trivial solvable radical, be a solvable subgroup of , and be a natural number such that for every solvable subgroup of the following inequality holds: . Then the solvable radical of is trivial and for every solvable subgroup of the following inequality holds: *
** Lemma 2****.**
[5, Theorem 1] * Let be a maximal primitive subgroup of or ; and Then for *
** Lemma 3****.**
Let and . Then
Proof.
If and ) is a regular point of (and also ), then it is easy to see that points
[TABLE]
are regular in distinct -orbits.
If and ) is a regular point of (and also for otherwise ), then it is easy to see that points
[TABLE]
are regular in distinct -orbits.
If and ) is a regular point of (and also for otherwise ), then it is easy to see that points
[TABLE]
are regular in distinct -orbits. ∎
** Lemma 4****.**
Let be a solvable primitive subgroup of , where is isomorphic to or , . Then In particular
Proof.
The group is known to possess a non-trivial solvable primitive subgroup if and only if for a prime and an integer [7, §4, Theorem 9]. The proof is divided into 3 cases:
Case 1 (). Subgroup lies in a maximal primitive subgroup of . Then by Lemma 2 we have , thus
Case 2 (; , ). In view of the known structure of maximal primitive solvable subgroups in [7, §4, Theorem 10] in this case is isomorphic to . Up to conjugation is a semidirect product . Here , and is the stabilizer of point 1 in . It is easy to verify by hand or using GAP, that in this case Furthermore, is generated by an odd permutation , i.e. permutations and are even, thus So, we have
Case 3 (, ). If then is isomorphic to If then is isomorphic to a subgroup of . These two cases can be verified directly by counting orbits using GAP. Note that group contains an odd permutation, so it suffices to prove that ∎
** Lemma 5****.**
Let be a semiregular subgroup of . Then there is such that
Proof.
It suffices to prove the lemma for the case when the group is regular. Indeed, let be non-regularly and is the union of orbits . Then acts regular on each orbit, so , where is the permutation group of . Let be the projection of on . Since is regular, there are such that Thus
Let be regular. If is elementary abelian, then by [8, Theorem 1] we obtain the required. Consider the case when is not elementary abelian.
We show that if and , then is not contained in . Consider possibilities for the structure of the element . Since is regular, every non-trivial element of has no fixed points. There are three cases:
**1. ** Points and are contained in distinct independent cycles, and there are more that two independent cycles, Then fix all points which are not contained in the same cycle as or , i.e. .
2. Let Since , the length of at least one of these cycles is greater than two. Without loss of generality, we can assume that it is the first. Then fixes , i.e. .
3. Let . Since , we have or . Without loss of generality, we can assume that . Then fixes , i.e. .
Therefore, . Since is regular, every non-trivial element of do not have fixed points. Thus, if then is an orbit of and fixes points and , i.e. It means that , is even and In particular, if odd then .
If is even then there is an element of order greater than two in because is not elementary abelian. Up to conjugation in we can assume that maps the point 1 to point 2, then, due to the regularity of , is the only element which maps 1 to 2. Thus , because is not contained in . ∎
2 Case of transitive solvable subgroup
** Lemma 6****.**
Let be a solvable transitive subgroup of , where is isomorphic to or , . Then
Proof.
If is primitive then the lemma follows from Lemma 4. Let be imprimitive. Then is contained in , here and is partitioned into blocks . We denote . Let be the image of in . Then and group is solvable. Thus we may assume that
Case
If then by induction for we have so by Lemma 1 we get , whence .
Assume that If , then so, by Lemma 1, If is not contained in , then there exist an odd permutation in and hence in In this case by the fact that and is normalized by an odd permutation.
Now assume that equals , , or , and equals Since is solvable for , we may assume that , where each is isomorphic to In particular, there is an odd permutation in group normalizing . Hence we may assume that Moreover, up to conjugation in , acts on
Case . Clearly, the intersection stabilizes the partitions and . If an element from the intersection stabilizes a point (Suppose, for definiteness, that stabilizes point 1), then fixes point 2, because it is contained in the same block in the first partition. Thus, because the element fixes point 2, it also fixes point 3, due to the fact that 2 and 3 are contained in the same block in the second partition, etc. So if fixes the first element in a block of the first partition then it also stabilizes the second point of the block and the first point of the next block of the same partition. Thus stabilizes all points, hence i.e. the intersection
[TABLE]
is semiregular and by Lemma 5 we have . So by Lemma 3.
Case . Consider the group
[TABLE]
which stabilizes the partitions
[TABLE]
If an element from this group stabilizes a point (for a definiteness, let stabilizes the point 1) then the point 2 can be moved under the action of only to 2 or 3, because of the first partition. However, points , 1 and 2 compose a block in the third partition, therefore 2 is fixed by , so the point 3 is also fixed by . Since points 2, 3 and 4 compose a block in the second partition, the element has to stabilize the point 4. Thus, repeating the preceding argument, we find that element stabilizes all the points, i.e which implies that the considered group is semiregular.
Let us find the number of orbits of the group
[TABLE]
Note that the points 1, 2 and 3 lie in distinct orbits. Indeed, let maps 1 to 2. In this case, points of the first block of the first partition stay in this block, i.e. maps 3 to 3, or to 1. In the first case 2 moves to 1, because stabilizes the first partition, but then does not stabilize the second partition. In the second case does not stabilize the third partition. Similar arguments show that can not map 1 to 3. Due to the symmetry of entries 1, 2 and 3 in the above considerations, these points lie in distinct orbits. Thus the number of orbits , then Note that the element of order stabilizer all three partitions, so it lies in , from which we obtain .
We have Consider the group
[TABLE]
It have to stabilize the partition , which, as easy to see, is not stabilized by any non-trivial element of , thus
[TABLE]
Therefore, and by Lemma 3 we obtain that
Case . Note that the group
[TABLE]
stabilizes the partition , which yields by the first case that the group
[TABLE]
is semiregular.
Let us find the number of orbits of the group
[TABLE]
which stabilizes the partitions
[TABLE]
Arguments similar to those of the preceding case () shows that
[TABLE]
We have Consider the group
[TABLE]
It have to stabilize the partition , which is not stabilized by the group , and we obtain that
[TABLE]
Consider the points
[TABLE]
where . Arguments similar to those given above, show that all these points are -regular. We will show that they all lie in different orbits.
Assume that and lie in the same orbit, i.e. there is an element , which maps to , then . It is clear that a trivial permutation does not map any of the points (1) – (5) to another, i.e., we can assume that Then the element lies in the group
[TABLE]
therefore , where
[TABLE]
The element maps 1 to , and to , and therefore does not stabilize the partition i.e. can not lie in , so (1) and (2) lie in distinct orbits. The fact that (1) and (3), (1) and (5), (2) and (4), (2) and (5), (3) and (4), (3) and (5 ), (4) and (5) does not lie in the same orbit can be proved by similar arguments.
Assume that and lie in the same orbit, i.e. there is an element , which maps to . As above we obtain The element maps 1 to . If then, as above, maps 4 to and can not lie in . Let , then maps 4 to 8. Then, because the element lies in and stabilize the corresponding partition, it is necessary that . Therefore we obtain that and it is a contradiction for all cases except the case , for which the statement of the Lemma is known. The fact that (2) and (3) does not lie in the same orbit can be proved by similar arguments.
Therefore, , as wanted. ∎
3 Proof of the main theorem
** Lemma 7****.**
Let be a solvable subgroup of , where is isomorphic to or , . Then
Proof.
If is transitive then the lemma follows from Lemma 6. Let be intransitive, i.e. If for all then by Lemma 6 there are , such that , here is the projection of the group on hence
[TABLE]
thus
Further proof is by induction on the number of orbits. The base of induction is the cases when there are only the orbits of sizes 2, 3 and 4. It suffices to consider the cases , , , , , . The validity of the statement for these groups can be verified directly with the help of GAP.
Assume that , Now , up to conjugation in , acts on , while acts on Let be the projection of on . As earlier we may assume that the projection of on equals since is solvable for . By induction, there exist such that
[TABLE]
Denote .
Case . Consider the intersection , and let lies in the intersection. Since lies in , it cannot map to or . On the other hand, as an element of , can map only to . So fixes . Thus it fixes and hence it fixes , i.e. . We obtain that is contained in
Consider elements such that and points and lie in distinct -orbits. Then, as we prove above,
[TABLE]
and points
[TABLE]
lie in the same -orbit if and only if there is such that
[TABLE]
Suppose that such exists. Then since , we obtain that
[TABLE]
As shown above, in this case , in particular map to , but this is impossible, since the points are from distinct orbits. Hence
Case . Consider the group
[TABLE]
Let be an element of this group. Then stabilizes the point 3, because lies in . Since , can map 2 only to 2 or 1, but so 2 can not be moved to 1, hence stabilizes 2, thus also stabilizes 1 and 4. Now since , the element stabilizes 5, and thus lies in . Therefore
Consider the elements such that and points lie in distinct -orbits. Then, by what was proved above we obtain
[TABLE]
and points
[TABLE]
lie in the same -orbit if and only if there is such that
[TABLE]
Suppose that such exists then we obtain
[TABLE]
since . As it is shown above, in this case , then maps to , which contradicts the assumption, since the points are from distinct orbits. Thus
Case . Consider the group
[TABLE]
Let be an element of this group. As easy to see, stabilize the point 4. Since , it stabilize the point 3. Hence, because , we obtain that 2 and 5 also fixed by . Thus also fixes 1 and 6 since . Therefore also stabilizes 7. Thus , i.e.
Consider the elements such that and points and lie in distinct -orbits. Then, by what was proved above, we obtain that
[TABLE]
and points
[TABLE]
lie in the same -orbit if and only if there is such that
[TABLE]
Suppose that such exists then we obtain that
[TABLE]
since . As it is shown above, in this case , then maps to , which contradicts the assumption since the points are from distinct orbits. Thus
Note that in these cases the group is normalized by , from this property and the fact that we obtain that .
∎
The main theorem for the case of an almost simple group with socle isomorphic to , , follows from Lemma 7. The main theorem for the case of an almost simple group with socle isomorphic to follows from [4, Theorem 1.1], if subgroup does not lie in Aschbacher class ,and from [3, Lemma 8], if subgroup lies in Aschbacher class .
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[1] Victor D. Mazurov and Evgeny I. Khukhro Unsolved problems in group theory. The Kourovka notebook; No 18. URL: http://arxiv.org/abs/1401.0300
- 2[2] E. P. Vdovin , On the base size of a transitive group with solvable point stabilizer; Journal of Algebra and Application, v. 11 (2012), N 1, 1250015 (14 pages)
- 3[3] E. P. Vdovin , On Intersections of Solvable Hall Subgroups in Finite Simple Exceptional Groups of Lie Type, Proceedings of the Steklov Institute of Mathematics, 2014, Vol. 285, Suppl. 1, pp. S 183–S 190.
- 4[4] Timothy C. Burness , On base sizes for actions of finite classical groups, London Math. Soc. (2) 75 (2007) 545–562.
- 5[5] Timothy C. Burness, Robert M. Guralnick and Jan Saxl , On base sizes for symmetric groups, Bull. London Math. Soc. 43 (2011), 386–391.
- 6[6] A. Seress , The minimal base size of primitive solvable permutation groups, J. London Math. Soc. 53 (1996), 243–255.
- 7[7] D. A. Suprunenko , Matrix Groups, Nauka, (1972).
- 8[8] V. I. Zenkov , Intersection of Abelian subgroups in finite groups, Matematicheskie Zametki, Vol. 56, No. 2, pp. 150 152, August, 1994.
