• Tidak ada hasil yang ditemukan

Artin exponent of some rational groups

N/A
N/A
Protected

Academic year: 2024

Membagikan "Artin exponent of some rational groups"

Copied!
3
0
0

Teks penuh

(1)

Extended Abstracts of the 41thIranian International Conference on Mathematics 12-15 September 2010, University of Urmia, Urmia, Iran, pp 212-214

ARTIN EXPONENT OF SOME RATIONAL GROUPS

H. SHARIFI

Abstract. A finite group whose irreducible complex characters are rational valued is called a rational group. We prove that if a rational group G contains a CC-subgroup or G is a direct prod- uct of Frobenius rational groups, then every rationally represented character is a generalized permutation character.

1. Introduction and Preliminaries

Let CharQ(G) and P(G) denote the ring of Z-linear combination of rationally represented characters and permutation characters of a finite group G, respectively. It is easy to see that P(G) a is subring of CharQ(G). By a theorem of Artin, | G | χ ∈ P(G) for all χ ∈ CharQ(G). The minimal number d ∈ N such that dχ ∈ P(G) for all χ ∈ CharQ(G), is called the Artin exponent of G and is denoted by γ(G). Indeed γ(G) is the exponent of CharQ(G)/P(G). The Artin exponent induced from cyclic subgroups of finite groups was studied extensively by T. Y. Lam in [2]. He proved that A(G) = exponent (PChar(G)Q(G)

cyclic)=1 if and only ifGis cyclic. One can show thatγ(G) divides theA(G) and therefore divides|G|. There is a fundamental distinction between γ(G) andA(G). While groups satisfying A(G) = 1 have been

2000Mathematics Subject Classification. Primary 20C15.

Key words and phrases. Rational group, Artin exponent, CC-subgroup, Local splitting.

Speaker.

1

(2)

2 H. SHARIFI

characterized, there is no such characterization for groups satisfying γ(G) = 1. In this paper, we characterize all rational groups having a CC-subgroup, then prove that γ(G) = 1 for such rational groups. We also prove that if G is a direct product of Frobenius rational groups, then γ(G) = 1.

Definition 1.1. Let G be a finite group and pa prime number. Sup- pose that K is a Q-p-class of G and s is a p-regular element of K.

The group G is said to be locally split onK if CG(< s >)p is a direct factor ofNG(< s >)p; that is,NG(< s >)p =CG(< s >)p×V for some V ≤NG(< s >)p . The groupGis locally split at pifG is locally split on every Q-p-class K. The group G is everywhere locally split if G is locally split at all primes.

A finite group G is called a rational group or a Q-group, if all irre- ducible complex characters ofGare rational. For example, the symmet- ric groupSn and the Weyl groups of the classical complex Lie algebras are rational groups (for more details see [5]). A comprehensive descrip- tion of rational groups can be found in [1] and [6]. Throughout the paper, we use the following notations and terminology: H : K stands for the semi-direct product of the groupsH andK. Zndenotes a cyclic group of order n. E(pn) where p is a prime number, denotes the ele- mentary abelian p-group of orderpn and Q8 is employed to denote the quaternion group of order 8.

2. Main results

Theorem 2.1. ( [6], Corollary 212 A) If G is rational group which is everywhere locally split, then γ(G) = 1.

Definition 2.2. A proper subgroup H of G is called a CC-subgroup of G if CG(x)≤H for any x∈H−1.

Theorem 2.3. Let H be a CC-subgroup of odd order in a rational group G. Then H ∼=E(3n) or E(5n). Also

(i) G∼=E(3n) :Z2 , (ii) G∼=E(32m) :Q8

(3)

ARTIN EXPONENT OF SOME GROUPS 3

(iii) G∼=E(52) :Q8

Theorem 2.4. Let G be a rational group with a CC-subgroup, then γ(G) = 1.

Corollary 2.5. LetGis a rational group that having abelian 2-subgroup.

Then γ(G) = 1.

Theorem 2.6. Let G be a finite group that the number of kernels is equal to the number of classes. Then γ(G/O2(G)) = 1.

Remark 2.7. G. Segal, J. Ritter and J. Rasmussen in 1972 indepen- dently showed that γ(G) = 1 for any p-group G by using methods of algebraic K-theory [7]. Ritter and Rasmussen proved the same result using group-theoretic methods and tried to extend the result to nilpo- tent groups. However, J. Serre [3] remarks that γ(Z3×Q8) = 2, thus showed that the extension of Segal’s result to nilpotent groups required some modification. Rasmussen did this by showing that γ(G) = 1 or 2 when Gis nilpotent and providing necessary and sufficient condition for when each case occurs [4].

References

1. Ya. G. Berkovich and E. M. Zhmud, Characters of finite groups Part 1, Transl. Math. Monographs172, Amer. Math. Soc. Providence, Rhode, 1997.

2. T. L. Lam,Artin exponent for finite groups, J. Algebra9, (1968), 94-119.

3. J. P. Serre,Representations lineaires des groupes finis, Hermann, 1971.

4. J. Rasmussen,Rationally represented characters and permutation characters of nilpotent groups, J. Algebra29, (1974), 504- 509.

5. J. H. Conway, R. T. Curtis, S. P. Norton, R. A. Parker and R. A. Wilson,Atlas of finite groups. Oxford University press, 1985.

6. D. Kletzing,Structure and representations ofQ-group, Lecture Notes in Math.

1084, Springer-Verlag, 1984.

7. G. Segal,Permutation representations of finite p-groups, Quart. J. Math. Ox- ford23, 1972, 375-381.

8. B. Huppert,Endliche Gruppen I, New York, 1967.

Department of Mathematics Faculty of Science University of Sha- hed, Tehran, Iran.

E-mail address: [email protected]

Referensi

Dokumen terkait