• Aucun résultat trouvé

A characterization of automorphism groups of simple <span class="mathjax-formula">${K}_{3}$</span> -groups

N/A
N/A
Protected

Academic year: 2022

Partager "A characterization of automorphism groups of simple <span class="mathjax-formula">${K}_{3}$</span> -groups"

Copied!
10
0
0

Texte intégral

(1)

A characterization of automorphism groups of simple K

3

-groups

DAPENGYU(*), JINBAOLI(**), GUIYUNCHEN(***), YANHENGCHEN(***)

ABSTRACT- In this paper, a new characterization of automorphism groups of simple K3-groups is presented.

MATHEMATICSSUBJECTCLASSIFICATION(2010). 20D05, 20D45.

KEYWORDS. SimpleK3-groups; conjugacy classes sizes; characterization.

(*) Indirizzo dell'A.: School of Mathematics and Statistics, Southwest Uni- versity, Chongqing 400715, China; also affiliated with Department of Mathematics, Chongqing University of Arts and Sciences, Chongqing 402160, China.

E-mail: yudapeng0@sina.com

Supported by the Scientific and Technological Research Program of Chongqing Municipal Education Commission (Grant No. KJ111207) and the Scientific Research Foundation of Chongqing University of Arts and Sciences (Grant No. Z2013SC10).

(**) Indirizzo dell'A.: Department of Mathematics, Chongqing University of Arts and Sciences, Chongqing 402160, China.

E-mail: leejinbao25@163.com

Supported by the Scientific Research Foundation of Chongqing Municipal Science and Technology Commission (Grant No. cstc2013jcyjA00034), the Scientific and Technological Research Program of Chongqing Municipal Education Commission (Grant No. KJ131204), the Scientific Research Foundation of Yongchuan Science and Technology Commission (Grant No. Ycstc, 2013nc8006).

(***) Indirizzo dell'A.: School of Mathematics and Statistics, Southwest Uni- versity, Chongqing 400715, China.

E-mail: gychen1963@163.com

Corresponding author and supported by the National Natural Science Founda- tion of China (Grant Nos. 11271301, 11171364).

(***) Indirizzo dell'A.: School of Mathematics and Statistics, Chongqing Three Gorges University, Chongqing 404100, China.

E-mail: math yan@126.com

Supported by the Scientific Research Foundation of Chongqing Municipal Science and Technology Commission (Grant No. cstc2014jcyjA00009), the Scientific and Tech- nologicalResearchProgramofChongqingMunicipalEducationCommission(GrantNo.

KJ1401006) and the Fundamental Research Funds for the Central Universities.

(2)

1. Introduction

All groups considered in this paper are finite.

It is known that the setcs(G) of sizes of the conjugacy classes of a finite groupGencodes much information about the structure ofG. Many authors have studied the connection between arithmetical properties ofcs(G) and structural properties ofG. The present paper is a contribution along this line, which is related to Thompson's conjecture (see [11, Problem 12.38]

and [2, 3, 4, 5, 12, 1] for detail).

CONJECTURE1.1. Let Gbe a group with Z(G)ˆ1 andM is a non- abelian simple group satisfying thatcs(G)ˆcs(M), thenG'M.

In this conjecture, M is a non-abelian simple group. Hence it seems interesting to consider the following question.

Let G be a group with Z(G)ˆ1 and M is an almost simple group satisfying that cs(G)ˆcs(M). Then, what can we say about the struc- ture of G?

A groupGis almost simple if there exists a non-abelian simple groupS such that S4G4Aut(S). In [9, 10], the almost sporadic groups, Aut(PSL(2;q)) andPGL(2;p) for some special cases are discussed. In this paper, we investigate the almost simpleK3-groups. A groupGis called aK3- group ifp(G) consists of exactly three distinct primes. Among the known simple groups, there are exactly eight simpleK3-groups (see [7, p. 12]):

A5; A6; L2(7); L2(8); L2(17); L3(3); U3(3); U4(2):

Our main result is as follows.

THEOREM 1.2. Let G be a group with Z(G)ˆ1 and S be one of the simple K3-groups. Set MˆAut(S).

(1) If S2 fA5;A6;L2(7);L2(8);L2(17);L3(3);U3(3)gand cs(G)ˆcs(M), then G'M.

(2) If SˆU4(2), cs(G)ˆcs(M) and jGjpˆ jMjp, then G'M, where jGjpdenotes the order of the Sylow p-subgroups of G for p2 f2;3;5g.

2. Preliminaries

Let Gbe a group and construct its prime graph G(G) as follows: the vertices are the primes dividing the order ofG, two verticespandq are

(3)

joined by an edge if and only ifGcontains an element of orderpq(see [13]).

We denote the set of all connected components of the graph G(G) by T(G)ˆ fpi(G)j14i4t(G)g, where t(G) is the number of the connected components of G(G). If the order ofG is even, we always assume that 22p1(G). Forx2G,xGdenotes the conjugacy class inGcontainingxand CG(x) denotes the centralizer ofxinG. A groupGis called a 2-Frobenius groupifGhas a normal series 1HKGsuch thatK andG=Hare Frobenius groups with kernelsHandK=Hrespectively (see, for example [9, Definition 2.1]).

The other notation and terminologies in this paper are standard and the reader is referred to ATLAS [6] and [8] if necessary.

The following lemma is well-known. It follows from [4, Lemma 1.1] or [12, Lemma 3].

LEMMA 2.1. Let G and M be groups satisfying Z(G)ˆZ(M)ˆ1and cs(G)ˆcs(M). Thenp(G)ˆp(M).

LEMMA 2.2[4, Lemma 1.4]. Let G be a group with Z(G)ˆ1and M a group with t(M)>1. Suppose that cs(G)ˆcs(M). ThenjGj ˆ jMj.

LEMMA 2.3 [4, Lemma 1.5]. Let G and M be groups satisfying that jGj ˆ jMjand cs(G)ˆcs(M). Then t(G)ˆt(M)and T(G)ˆT(M).

LEMMA2.4. Suppose that G is a Frobenius group of even order and H, K are the Frobenius kernel and the Frobenius complement of G, respec- tively. Then t(G)ˆ2, T(G)ˆ fp(H);p(K)gand G has one of the following structures:

(i) 22p(H)and all Sylow subgroups of K are cyclic;

(ii) 22p(K), H is an abelian group, K is a solvable group, the Sylow subgroups of K of odd order are cyclic groups and the Sylow 2-subgroups of K are cyclic or generalized quaternion groups;

(iii) 22p(K), H is abelian, and there exists a subgroup K0of K such that

jK:K0j42;K0ˆZSL(2;5);(jZj;235)ˆ1;

and the Sylow subgroups of Z are cyclic.

PROOF. This is Lemma 1.6 in [4]. p

(4)

LEMMA2.5. Let G be a2-Frobenius group of even order. Then t(G)ˆ2 and G has a normal series 1HKG such that p(K=H)ˆp2, p(H)[p(G=K)ˆp1, the order of G=K divides the order of the auto- morphism group of K=H, and both G=K and K=H are cyclic. Especially, jG=Kj<jK=Hjand G is solvable.

PROOF. This is Lemma 1.7 in [4]. p

LEMMA 2.6. Let G be a group with more than one prime graph com- ponent. Then G is one of the following:

(i) a Frobenius or2-Frobenius group;

(ii) G has a normal series1HKG, where H is a nilpotent p1-group, K=H is a non-abelian simple group and G=K is ap1- group such that jG=Kj divides the order of the outer auto- morphism group of K=H. Besides,pi(K=H)ˆpi(G)for i52.

PROOF. It follows straight forward from Lemmas 1-3 in [13], Lemma

1.5 in [3] and Lemma 7 in [5]. p

LEMMA 2.7 [9, Corollary 5.1]. Let G be a group with Z(G)ˆ1 and MˆAut(L2(q)), where G(M) is not connected. If cs(G)ˆcs(M), then G'M.

LEMMA2.8 [12, Lemma 4]. Suppose that G is a group with Z(G)ˆ1 and p is a prime inp(G)such that p2 does not dividejxGjfor all x in G.

Then a Sylow p-subgroup of G is elementary abelian.

3. Proof of Theorem 1.2

LEMMA3.1. Let G be a group with Z(G)ˆ1and S2 fA5;L2(7);L2(8);L2(17);L3(3);U3(3)g:

Let MˆAut(S)and cs(G)ˆcs(M). Then G'M.

PROOF. By [6], we have thatG(M) is not connected if S2 fA5;L2(7);L2(8);L2(17);L3(3);U3(3)g:

Therefore, by Lemma 2.7,G'Mprovided that S2 fL2(7);L2(8);L2(17)g:

Now, it suffices to discuss the remaining three cases.

(5)

First, we suppose thatSˆA5. Then G(M) is not connected. By the hypothesis and lemma 2.2, we obtain thatjGj ˆ jMj. Since p2(M)ˆ f5g by [6], it follows from Lemma 2.3 thatp2(G)ˆ f5g. Since 5 does not di- vide 23, by Lemma 2.4, we haveGis not a Frobenius group. Suppose that G is a 2-Frobenius group. Then, by Lemma 2.5,G has a normal series 1HKGsuch thatjK=Hj ˆ5. LetPbe a Sylow 3-subgroup ofH.

ThenPis a normal subgroup ofGof order 3. Letx2Gsuch thatjxj ˆ5.

Then xacts trivially on P and soG(G) is connected. This contradiction shows that Gis not a 2-Frobenius group either. Now, by lemma 2.6, G has a normal series 1HKG such that K=H is a non-abelian simple group. It is easy to see that K=His isomorphic toA5. IfH6ˆ1, thenHis of order 2 and soH4Z(G), contrary to our assumption forG.

Hence Hˆ1 and thereforeK'A5. It follows thatG'S5, as desired.

Next, we assume that SˆL3(3). Since G(M)ˆG(Aut(S)) is not con- nected, Lemma 2.2 together with the hypothesis imply thatjGj ˆ jMj. B y [6], we have thatf13gis a component ofG(M). Hence, by Lemma 2.3,f13gis a component ofG(G). Similar to above discussion, we can show thatGis not a Frobenius group. We assert thatGis not a 2-Frobenius group either. If not, then, by Lemma 2.5, G has a normal series 1HKG such that jK=Hj ˆ13. Let P be a Sylow 2-subgroup ofH andVˆV1(Z(P)). Since (13;jGLn(2)j)ˆ1, wherenˆ2;3;4;5, an elementxofGof order 13 act tri- vially onV. ThereforeGhas an element of order 26, a contradiction. ThusGis not a 2-Frobenius group. By Lemma 2.6,Ghas a chief factorK=Hsuch that K=His a simpleK3-group and 132p(K=H). By [6],K=Hmust be isomorphic toL3(3). Similarlyas above,weobtainthatHˆ1 and soG'MˆAut(L3(3)).

Finally, we prove that G'MˆAut(U3(3)). As above, one can show that G is neither a Frobenius group nor a 2-Frobenius group. By [6], p2(M)ˆ f7gand sop2(G)ˆ f7gby Lemma 2.3. It follows from Lemma 2.6 that G has a chief factorK=Hsuch thatK=His a simpleK3-group and 72p(K=H). ThenK=His isomorphic toL2(7),L2(8) orU3(3). LetP be a Sylow 2-subgroup ofH,Qa Sylow 3-subgroup ofHandxan element ofGof order 7. If K=H'L2(7), then jQj ˆ32. Since (7;jAut(Q)j)ˆ1, x acts trivially on Qand so 7 and 3 are joined, a contradiction. IfK=H'L2(8), then P is of order 23 and so G has an element x such that jxGj 7, a contradiction by [6]. Hence K=H'U3(3). It is easy to see that Hˆ1 since otherwise,His contained inZ(G), a contradiction. Thus,K'U3(3)

andG'Aut(U3(3)). p

LEMMA 3.2. Let G be a group with Z(G)ˆ1 and MˆAut(A6). If cs(G)ˆcs(M), then G'M.

(6)

PROOF. We proceed the proof by several steps.

(1) By [6],cs(G) consists ofn1ˆ1,n2ˆ325,n3ˆ245,n4ˆ2325, n5ˆ2432,n6ˆ235,n7ˆ2435,n8ˆ2232,n9ˆ22325.

(2) By Lemma 2.1, we have thatp(G)ˆp(M)ˆ f2;3;5g.

(3) The Sylow 5-subgroups ofGare of order 5.

LetPbe a Sylow 5-subgroup ofG. Then, by Lemma 2.8,Pis elementary abelian since 52does not divide any element incs(G). We assert thatPis of order 5. Assume that 52 divide the order of G. Then, the centralizer of every element ofGcontains an element of order 5. Let y2Gsuch that jyGj ˆ2435 andxan element of CG(y) of order 5. Since the Sylow 5- subgroup ofGis elementary abelian, we have that 5 does not dividejxGj and consequently xGˆ2432 or 2232. If 5 does not divide jyj, then CG(xy)ˆCG(x)\CG(y), from which we can deduce that 26335 divide (xy)G, which is impossible. If 5 dividesjyj, thenyˆy1y2withjy1j ˆ5 and (jy1j;jy2j)ˆ1. It follows thatjyG1jshould dividejyGj, a contradiction. Hence the Sylow 5-subgroup ofGis of order 5.

(4)O220(G)ˆO2(G).

WriteKˆO2(G) and GˆG=K. Suppose that the statement is false.

Then there is r2 f3;5g such that Or(G)6ˆ1. If PˆO5(G)6ˆ1, then jPj ˆ5. Let Qbe a Sylow 3-subgroup ofG andx an element ofZ(Q) of order 3. ThenjxGj ˆn3ˆ245 and so 5 does not dividejCG(x)jby Step 1.

Since (3;5 1)ˆ1, we see thatxacts trivially onP. Thus 5 dividesCG(x).

Since (3;jKj)ˆ1, we have thatCG(x)ˆCG(x)K=Kand so 5 dividesjCG(x)j, a contradiction. Now, we assume that O3(G)6ˆ1. Put V ˆV1(Z(O3(G))).

ThenV is a nontrivial normal subgroup ofG. Letybe an element ofGof order 5. ThenV ˆ[V; y]CV(y). By Step 1,jV :CV(y)jis at most 32. It follows that [[V;y];y]ˆ1, which implies that [V;y]ˆ1. Therefore V ˆCV(y). LetQbe a Sylow 3-subgroup ofG. ThenV\Z(Q)6ˆ1. Letzbe an element ofV\Q of order 3. Then y zˆz y. Since (3;jKj)ˆ1, there exists a preimagezofzinGsuch thatzis contained in the center a Sylow 3- subgroup ofGand so 5 dividesjCG(z)j. But, by Step 1, this is impossible.

(5)G'Aut(A6).

SetKˆO2(G) andGˆG=K. ThenGis insoluble by Step 4. Further- more, we have that M4G4Aut(M), where MˆS1S2 Sk is a direct product of non-abelian simplesS1,S2,. . .,Sk. Sincep(G)ˆ f2;3;5g and the order of the Sylow 5-subgroups ofGis 5, we obtain thatkˆ1, that is,Mis a non-abelian simpleK3-group. HenceMis isomorphic toA5,A6or U4(2).

IfM'A5, thenG=K'A5orAut(A5) and so 32does not dividejGj, a contradiction.

(7)

Suppose that M'A6. Assume first that G'A6 or S6. Let x be an element of G of order 5 such that n8ˆ jxGj ˆ2232. Then jxGj divides 2232. By [6], we have thatjxGj ˆ1, which implies that x4Z(G). There- forexˆ1 for thatZ(G) is trivial and sox2K, a contradiction.

Assume that G'PGL2(9). Pick x2G with jxj ˆ5 such that jxGj ˆ2232. Then, by [6],jxGj ˆ1 or 2232. As above, it is impossible that jxGj ˆ1. IfjxGj ˆ2232, thenxcentralizesK. IfK6ˆ1, thenxcentralizes an element ofGwhich lies in the center of a Sylow 2-subgroup ofG, con- trary to Step 1. IfK ˆ1, thenG'PGL2(9), butcs(PGL2(9))6ˆcs(G) by [6], a contradiction.

Suppose thatG'M10. Letxbe an element ofGof order 5 such that jxGj ˆ2432. ThenjxGj ˆ1 or 2432by [6]. Similar to the above case, we can derive a contradiction.

IfG'Aut(A6), then, similar to the forgoing argument, one can show thatKˆ1 and soG'Aut(A6), as desired.

If G'U4(2), then G has an element x such that jxGj ˆ27325. It follows thatjxGjis divisible by 27325, which is impossible by Step 1.

Thus, the proof is complete. p

LEMMA 3.3. Let G be a group with Z(G)ˆ1 and MˆAut(U4(2)).

Suppose that cs(G)ˆcs(M)andjGjpˆ jMjp, wherejGjpdenotes the order of the Sylow p-subgroups of G for p2 f2;3;5g. Then G'M.

PROOF. We proceed the proof by the following steps.

(1) cs(G) consists of n1ˆ1, n2ˆ325, n3ˆ2335, n4ˆ245, n5ˆ2435, n6ˆ2535, n7ˆ22335, n8 ˆ23345, n9 ˆ2634, n10 ˆ24325, n11 ˆ25325, n12 ˆ24335, n13 ˆ27325, n14 ˆ 25335,n15ˆ2232,n16 ˆ22345,n17ˆ24345.

This follows from [6].

(2)p(G)ˆp(M)ˆ f2;3;5g.

This follows from [6] and Lemma 2.1.

(3)O220(G)ˆO2(G).

Write KˆO2(G) and GˆG=K. Suppose that the assertion is false.

Then Or(G)6ˆ1, where r2 f3;5g. Assume first that O5(G)6ˆ1. Then, jO5(G)j ˆ5. Letx2Z(Q) withjxj>1, whereQis a Sylow 3-subgroup ofG.

Then jxGj ˆ245 by (1). Sincex acts trivially onO5(G), we have that 5 divides the order ofCG(x), by which we conclude that 5 divides the order of CG(x). This is impossible by (1). Now assume that O3(G)6ˆ1 and set V ˆV1(Z(O3(G))). Letybe an element ofGof order 5. ThenjyGj ˆ2634 or 2232. Since V ˆ[V;y]CV(y), we have that jV :CV(y)j434. If

(8)

jV :CV(y)j433, then [[V;y];y]ˆ1 and so [V;y]ˆ1. Discussing as in Lemma 3.2, we see that this is impossible. If jV :CV(y)j ˆ34, then the Sylow 3-subgroups of G is elementary abelian, which contradicts that jxGj ˆn13ˆ27325 for some 3-elementx2G.

(4)G'M.

B y (3),Gis insoluble. Hence we have thatM4G4Aut(M), whereM is a direct product of non-abelian simple groups. Since, by (2) p(G)ˆ p(M)ˆ f2;3;5g, we know thatMis simpleK3-groups by (3). Thus,Mis isomorphic to A5, A6 or U4(2). If M is isomorphic to A5 or A6, then 33 does not divide the order of G, a contradiction by (1). Suppose that M'U4(2). Then G'U4(2) or U4(2):2. If G'U4(2), then Z(G) is non- trivial, a contradiction. If G'U4(2):2, then G is isomorphic to Aut(U4(2)), as desired.

Proof of Theorem 1.2.

It follows from Lemmas 3.1-3.3. p

Acknowledgments. The authors would like to express their sincere thanks for the referee's helpful comments.

REFERENCES

[1] N. AHANJIDEH,On Thompson's conjecture for some finite simple groups, J.

Algerba,344(2011), pp. 205-228.

[2] G. Y. CHEN, On Thompson's Conjecture, PhD thesis, Sichuan University, Department of Mathematics, 1994.

[3] G. Y. CHEN, On Thompson's cconjecture for sporadic simple groups, Proc.

China Assoc. Sci. and Tech. First Academic Annual Meeting of Youths, Chinese Sci. and Tech. Press, Beijing 1992, pp. 1-6. (in Chinese)

[4] G. Y. CHEN,On Thompson's conjecture, J. Algebra,185(1996), pp. 184-193.

[5] G. Y. CHEN,Further reflections on Thompson's conjecture, J. Algebra,218 (1999), pp. 276-285.

[6] J. H. CONWAY- R. T. CURTIS- S. P. NORTON- R. A. PARKER- R. A. WILSON, Atlas of Finite Groups, Clarendon Press, Oxford, 1985.

[7] D. GORENSTEIN,Finite simple groups, Plenum Press, New York, 1982.

[8] W. B. GUO,The Theory of Classes of Groups, Science Press-Kluwer Academic Publishers, Beijing-New York-Dordrecht-Boston-London, 2000.

[9] A. KHOSRAVI - B . KHOSRAVI, A new characterization of almost sporadic groups, J. Algebra and Its Applications,3(2002), pp. 267-279.

[10] B. KHOSRAVI - M. KHATAMI, A new characterization of PGL(2;p) by its noncommuting graph, Bull. Malays. Math. Sci. Soc. (2), 34(3) (2011), pp.

665-674.

(9)

[11] V. D. MAZUROV- E. I. KHUKHRO,Unsolved Problems in Group Theory: The Kourovka Notebook, 17th edition, Sobolev Institute of Mathematics, Novosi- birsk, 2010.

[12] A. V. VASIL0EV,On Thompson's conjecture, Siberian Electronic Mathematical Reports,6(2009), pp. 457-464.

[13] J. S. WILLIAMS, Prime graph components of finite groups, J. Algebra, 69 (1981), pp. 487-513.

Manoscritto pervenuto in redazione il 2 Settembre 2013.

(10)

Références

Documents relatifs

All the proofs in this paper in some part depend on using another zeta function introduced to classify the unclassifiable by Grunewald, Segal and Smith [21]. One can also define a

The aim of this paper is to show that if one reformulates the problem in terms of derivations, as we do in Section 2, then the general case of an arbitrary p-group of maximal class

The idea is to replace k by a field F(k) of characteristic zero, and whose motivic cohomology and algebraic K-the- ory groups are naturally isomorphic to the same groups for k...

L’accès aux archives de la revue « Rendiconti del Seminario Matematico della Università di Padova » ( http://rendiconti.math.unipd.it/ ) implique l’accord avec les

Kenku, Rational torsion points on elliptic curves defined over quadratic fields,

This follows from Lemma 2.3 since both sides have the same image under Â’ by Definition 2.4 and the first property of Â’.. 0 Using the formalism of formal Dirichlet

The universal ring U is the polynomial ring generated by the 03BBn(u), where u is the canonical element.. and the induction hypothesis. From now on we.. therefore assume

the algebraic local ring of a normal singular point on a rational surface,.. and attempt to classify all such rings in a given