Beruflich Dokumente
Kultur Dokumente
4, 187 - 195
1 Introduction
The notion of a Γ-semigroup has been introduced by M. K. Sen in [6]
the year 1981. Many classical notions of semigroup have been extended to Γ-
semigroup (see [6], [7], [1], [2]). Green’s relations for semigroups, first studied
by J. A. Green [4], have played a fundamental role in the development to
semigroup theory (see [5]). In [8], G. Thierrin has introduced a reductive
semigroup. A. Fattahi and H. R. E. Vishki have given a characterization
for regular reductive semigroups in [3]. In this paper, we consider Green’s
relations for Γ-semigroups and reductive Γ-semigroups. Moreover, we give a
characterization for regular reductive Γ-semigroups.
2 Preliminaries
Let S and Γ be nonempty sets. If there exists a mapping S × Γ × S → S,
written (a, γ, b) by aγb, S is called a Γ-semigroup if S satisfies the identities
1
This research was supported by the Commission on Higher Education and a grant of
Thailand Research Fund (TRF).
188 R. Chinram and P. Siammai
3 Main results
The Green’s equivalence relations L, R, H and D on a Γ-semigroup S are
defined by the following rules :
(i) aLb if and only if S 1 Γa = S 1 Γb where S 1 Γa = SΓa ∪ {a}.
(ii) aRb if and only if aΓS 1 = bΓS 1 where aΓS 1 = aΓS ∪ {a}.
(iii) H = L ∩ R.
(iv) D = L ◦ R.
Remark We have
(i) aLb if and only if a = b or there exist x, y ∈ S and α, β ∈ Γ such that
a = xαb and b = yβa.
(ii) aRb if and only if if a = b or there exist x, y ∈ S and α, β ∈ Γ such
that a = bαx and b = aβy.
(iii) aHb if and only if aLb and aRb.
(iv) aDb if and only if there exists c ∈ S such that aLc and cRb.
Theorem 3.1 L ◦ R = R ◦ L.
Proof. Let (a, b) ∈ L ◦ R. Then there exists c ∈ S such that aLc and cRb.
Case 1: a = c. Then aRb. Since aLa and aRb, (a, b) ∈ L ◦ R.
Case 2: b = c. Then aLb. Since aLb and bRb, (a, b) ∈ L ◦ R.
On Green’s relations for Γ-semigroups and reductive Γ-semigroups 189
aηu = yμcηu = d
and
yμb = yμcηu = d
and
The L-class (resp. R-class, H-class, D-class) containing the element a will
be written La (resp. Ra , Ha , Da ).
Theorem 3.4 In a regular D-class, each L-class and each R-class contains
at least one idempotent.
and
and
a∗ βaαa∗ = (f γxβe)βaα(f γxβe) = f γxβ(eβa)αf γxβe = f γxβ(aαf )γxβe =
f γxβ(aγx)βe = f γxβeβe = f γxβe = a∗ .
Then a∗ is an (α, β)-inverse of a. Moreover
Proof. (i) Let a, b ∈ S such that aLb. Then a = b or there exist x, y ∈ S and
α, β ∈ Γ such that a = xαb and b = yβa.
Case 1 : a = b. Then aρ = bρ.
Case 2 : a = xαb and b = yβa. Then aρ = (xαb)ρ = (xρ)α(bρ) and
bρ = (yβa)ρ = (yρ)β(aρ). Therefore aρLbρ.
Conversely, let a, b ∈ S. Assume aρLbρ. Then aρ = bρ or there exist
x, y ∈ S and α, β ∈ Γ such that aρ = (xρ)α(bρ) and bρ = (yρ)β(aρ).
Case 1 : aρ = bρ. Then (a, b) ∈ ρ. Since ρ ⊆ L, (a, b) ∈ L. So aLb.
Case 2 : aρ = (xρ)α(bρ) and bρ = (yρ)β(aρ). Then aρ = (xαb)ρ and
bρ = (yβa)ρ. Then (a, xαb) ∈ ρ and (b, yβa) ∈ ρ. Since ρ ⊆ L, (a, xαb) ∈ L
and (b, yβa) ∈ L. Then a ∈ S 1 Γ(xαb) and b ∈ S 1 Γ(yβa). Thus S 1 Γa = S 1 Γb.
Hence aLb.
(ii) It is similar to (i).
(iii) It follows from (i) and (ii).
[resp. left] reductive congruence. In other words, S is called right [resp. left]
reductive if for each a, b ∈ S, aγt = bγt [resp. tγa = tγb] for all t ∈ S and
γ ∈ Γ implies a = b. A Γ-semigroup is called reductive if it is both right and
left reductive.
Proof. (i) Let ρ be a right reductive congruence. Let aρ, bρ ∈ S/ρ such that
(aρ)γ(tρ) = (bρ)γ(tρ) for all t ∈ S and γ ∈ Γ. Then (aγt, bγt) ∈ ρ for all t ∈ S
and γ ∈ Γ. Since ρ is right reductive, (a, b) ∈ ρ. Hence aρ = bρ.
Conversely, suppose S/ρ is a right reductive Γ-semigroup. Let a, b ∈ S
such that (aγt, bγt) ∈ ρ for all t ∈ S and γ ∈ Γ. Then (aγt)ρ = (bγt)ρ for all
t ∈ S and γ ∈ Γ. Thus (aρ)γ(tρ) = (bρ)γ(tρ) for all t ∈ S and γ ∈ Γ. Since
S/ρ is a right reductive Γ-semigroup, aρ = bρ. Therefore (a, b) ∈ ρ.
(ii) It is similar to (i).
(iii) It follows by (i) and (ii).
Proof. (i) Let (a, b) ∈ ρr . Then aγt = bγt for all t ∈ S and for all γ ∈ Γ. So
aΓS = bΓS. Since a ∈ aΓS and b ∈ bΓS because S is regular, aΓS 1 = bΓS 1 .
Therefore (a, b) ∈ R. Thus ρr ⊆ R.
(ii) It is similar to (i).
Proof. (i) Let a, b ∈ S. Assume that (aγt, bγt) ∈ ρr for all t ∈ S and γ ∈ Γ.
Then aγtβt = bγtβt for all t, t ∈ S and γ, β ∈ Γ. Thus aαt = bαt for all
t ∈ S and α ∈ Γ because S is regular. So (a, b) ∈ ρr . Therefore ρr is a right
reductive congruence on S.
Next, let ρ be any right reductive congruence on S. Let (a, b) ∈ ρr . Then
aγt = bγt for all t ∈ S and γ ∈ Γ. Since ρ is reflexive, (aγt, bγt) ∈ ρ. Therefore
(a, b) ∈ ρ because ρ is right reductive.
(ii) It is similar to (i).
References
[1] S. Chattopadhyay, Right inverse Γ-semigroup, Bull. Cal. Math. Soc.
93(2001), 435-442.
[3] A. Fattahi and H.R.E. Vishki, A note on reductive semigroups, SEA Bull.
Math. 28(2004), 261-264.