Zero Action on Perfect Crystals for Uq(G₂⁽¹⁾)
The actions of 0-Kashiwara operators on the Uq'(GG₂⁽¹⁾)-crystal Bl in [Yamane S., J. Algebra 210 (1998), 440-486] are made explicit by using a similarity technique from that of a Uq'(D₄⁽³⁾)-crystal. It is shown that {Bl}l ≥ 1 forms a coherent family of perfect crystals.
Збережено в:
| Дата: | 2010 |
|---|---|
| Автори: | , , |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Інститут математики НАН України
2010
|
| Назва видання: | Symmetry, Integrability and Geometry: Methods and Applications |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/146307 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Цитувати: | Zero Action on Perfect Crystals for Uq(G₂⁽¹⁾) / K.C. Misra, M. Mohamad, M. Okado // Symmetry, Integrability and Geometry: Methods and Applications. — 2010. — Т. 6. — Бібліогр.: 7 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-146307 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1463072019-02-09T01:23:11Z Zero Action on Perfect Crystals for Uq(G₂⁽¹⁾) Misra, K.C. Mohamad, M. Okado, M. The actions of 0-Kashiwara operators on the Uq'(GG₂⁽¹⁾)-crystal Bl in [Yamane S., J. Algebra 210 (1998), 440-486] are made explicit by using a similarity technique from that of a Uq'(D₄⁽³⁾)-crystal. It is shown that {Bl}l ≥ 1 forms a coherent family of perfect crystals. 2010 Article Zero Action on Perfect Crystals for Uq(G₂⁽¹⁾) / K.C. Misra, M. Mohamad, M. Okado // Symmetry, Integrability and Geometry: Methods and Applications. — 2010. — Т. 6. — Бібліогр.: 7 назв. — англ. 1815-0659 s 2010 Mathematics Subject Classification: 05E99; 17B37; 17B67; 81R10; 81R DOI:10.3842/SIGMA.2010.022 http://dspace.nbuv.gov.ua/handle/123456789/146307 en Symmetry, Integrability and Geometry: Methods and Applications Інститут математики НАН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| description |
The actions of 0-Kashiwara operators on the Uq'(GG₂⁽¹⁾)-crystal Bl in [Yamane S., J. Algebra 210 (1998), 440-486] are made explicit by using a similarity technique from that of a Uq'(D₄⁽³⁾)-crystal. It is shown that {Bl}l ≥ 1 forms a coherent family of perfect crystals. |
| format |
Article |
| author |
Misra, K.C. Mohamad, M. Okado, M. |
| spellingShingle |
Misra, K.C. Mohamad, M. Okado, M. Zero Action on Perfect Crystals for Uq(G₂⁽¹⁾) Symmetry, Integrability and Geometry: Methods and Applications |
| author_facet |
Misra, K.C. Mohamad, M. Okado, M. |
| author_sort |
Misra, K.C. |
| title |
Zero Action on Perfect Crystals for Uq(G₂⁽¹⁾) |
| title_short |
Zero Action on Perfect Crystals for Uq(G₂⁽¹⁾) |
| title_full |
Zero Action on Perfect Crystals for Uq(G₂⁽¹⁾) |
| title_fullStr |
Zero Action on Perfect Crystals for Uq(G₂⁽¹⁾) |
| title_full_unstemmed |
Zero Action on Perfect Crystals for Uq(G₂⁽¹⁾) |
| title_sort |
zero action on perfect crystals for uq(g₂⁽¹⁾) |
| publisher |
Інститут математики НАН України |
| publishDate |
2010 |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/146307 |
| citation_txt |
Zero Action on Perfect Crystals for Uq(G₂⁽¹⁾) / K.C. Misra, M. Mohamad, M. Okado // Symmetry, Integrability and Geometry: Methods and Applications. — 2010. — Т. 6. — Бібліогр.: 7 назв. — англ. |
| series |
Symmetry, Integrability and Geometry: Methods and Applications |
| work_keys_str_mv |
AT misrakc zeroactiononperfectcrystalsforuqg21 AT mohamadm zeroactiononperfectcrystalsforuqg21 AT okadom zeroactiononperfectcrystalsforuqg21 |
| first_indexed |
2025-07-10T23:26:07Z |
| last_indexed |
2025-07-10T23:26:07Z |
| _version_ |
1837304363451678720 |
| fulltext |
Symmetry, Integrability and Geometry: Methods and Applications SIGMA 6 (2010), 022, 12 pages
Zero Action on Perfect Crystals for Uq
(
G
(1)
2
)
?
Kailash C. MISRA †, Mahathir MOHAMAD ‡ and Masato OKADO ‡
† Department of Mathematics, North Carolina State University,
Raleigh, North Carolina 27695-8205, USA
E-mail: misra@unity.ncsu.edu
‡ Department of Mathematical Science, Graduate School of Engineering Science,
Osaka University, Toyonaka, Osaka 560-8531, Japan
E-mail: mahathir75@yahoo.com, okado@sigmath.es.osaka-u.ac.jp
Received November 13, 2009, in final form March 03, 2010; Published online March 09, 2010
doi:10.3842/SIGMA.2010.022
Abstract. The actions of 0-Kashiwara operators on the U ′
q
(
G
(1)
2
)
-crystal Bl in [Yamane S.,
J. Algebra 210 (1998), 440–486] are made explicit by using a similarity technique from that
of a U ′
q
(
D
(3)
4
)
-crystal. It is shown that {Bl}l≥1 forms a coherent family of perfect crystals.
Key words: combinatorial representation theory; quantum affine algebra; crystal bases
2010 Mathematics Subject Classification: 05E99; 17B37; 17B67; 81R10; 81R50
1 Introduction
Let g be a symmetrizable Kac–Moody algebra. Let I be its index set for simple roots, P the
weight lattice, αi ∈ P a simple root (i ∈ I), and hi ∈ P ∗(= Hom(P,Z)) a simple coroot (i ∈ I).
To each i ∈ I we associate a positive integer mi and set α̃i = miαi, h̃i = hi/mi. Suppose
(〈h̃i, α̃j〉)i,j∈I is a generalized Cartan matrix for another symmetrizable Kac–Moody algebra g̃.
Then the subset P̃ of P consisting of λ ∈ P such that 〈h̃i, λ〉 is an integer for any i ∈ I can be
considered as the weight lattice of g̃. For a dominant integral weight λ let Bg(λ) be the highest
weight crystal with highest weight λ over Uq(g). Then, in [5] Kashiwara showed the following.
(The theorem in [5] is more general.)
Theorem 1. Let λ be a dominant integral weight in P̃ . Then, there exists a unique injective
map S : Bg̃(λ) → Bg(λ) such that
wtS(b) = wt b, S(eib) = emi
i S(b), S(fib) = fmi
i S(b).
In this paper, we use this theorem to examine the so-called Kirillov–Reshetikhin crystal. Let g
be the affine algebra of type D(3)
4 . The generalized Cartan matrix (〈hi, αj〉)i,j∈I (I = {0, 1, 2})
is given by 2 −1 0
−1 2 −3
0 −1 2
.
Set (m0,m1,m2) = (3, 3, 1). Then, g̃ defined above turns out to be the affine algebra of
type G(1)
2 . Their Dynkin diagrams are depicted as follows
D
(3)
4 :
0 1 2
◦ ◦ ◦_jt G
(1)
2 :
0 1 2
◦ ◦ ◦_ *4
?This paper is a contribution to the Proceedings of the Workshop “Geometric Aspects of Discrete and Ultra-
Discrete Integrable Systems” (March 30 – April 3, 2009, University of Glasgow, UK). The full collection is available
at http://www.emis.de/journals/SIGMA/GADUDIS2009.html
mailto:misra@unity.ncsu.edu
mailto:mahathir75@yahoo.com
mailto:okado@sigmath.es.osaka-u.ac.jp
http://dx.doi.org/10.3842/SIGMA.2010.022
http://www.emis.de/journals/SIGMA/GADUDIS2009.html
2 K.C. Misra, M. Mohamad and M. Okado
For G(1)
2 a family of perfect crystals {Bl}l≥1 was constructed in [7]. However, the crystal elements
there were realized in terms of tableaux given in [2], and it was not easy to calculate the action
of 0-Kashiwara operators on these tableaux. On the other hand, an explicit action of these
operators was given on perfect crystals {B̂l}l≥1 over U ′
q
(
D
(3)
4
)
in [6]. Hence, it is a natural idea
to use Theorem 1 to obtain the explicit action of e0, f0 on Bl from that on B̂l′ with suitable l′.
We remark that Kirillov–Reshetikhin crystals are parametrized by a node of the Dynkin diagram
except 0 and a positive integer. Both Bl and B̂l correspond to the pair (1, l).
Our strategy to do this is as follows. We define Vl as an appropriate subset of B̂3l that is
closed under the action of êmi
i , f̂mi
i where êi, f̂i stand for the Kashiwara operators on B̂3l. Hence,
we can regard Vl as a U ′
q
(
G
(1)
2
)
-crystal. We next show that as a Uq
(
G
(1)
2
)
{0,1}(= Uq(A2))-crystal
and as a Uq
(
G
(1)
2
)
{1,2}(= Uq(G2))-crystal, Vl has the same decomposition as Bl. Then, we can
conclude from Theorem 6.1 of [6] that Vl is isomorphic to the U ′
q
(
G
(1)
2
)
-crystal Bl constructed
in [7] (Theorem 2).
The paper is organized as follows. In Section 2 we review the U ′
q
(
D
(3)
4
)
-crystal B̂l. We then
construct a U ′
q
(
G
(1)
2
)
-crystal Vl in B̂3l with the aid of Theorem 1 and see it coincides with Bl
given in [7] in Section 3. Minimal elements of Bl are found and {Bl}l≥1 is shown to form
a coherent family of perfect crystals in Section 4. The crystal graphs of B1 and B2 are included
in Section 5.
2 Review on U ′
q
(
D
(3)
4
)
-crystal B̂l
In this section we recall the perfect crystal for U ′
q
(
D
(3)
4
)
constructed in [6]. Since we also
consider U ′
q
(
G
(1)
2
)
-crystals later, we denote it by B̂l. Kashiwara operators ei, fi and εi, ϕi on B̂l
are denoted by êi, f̂i and ε̂i, ϕ̂i. Readers are warned that the coordinates xi, x̄i and steps by
Kashiwara operators in [6] are divided by 3 here, since it is more convenient for our purpose.
As a set
B̂l =
b = (x1, x2, x3, x̄3, x̄2, x̄1) ∈ (Z≥0/3)6
∣∣∣∣∣∣
3x3 ≡ 3x̄3 (mod 2),∑
i=1,2
(xi + x̄i) + (x3 + x̄3)/2 ≤ l/3
.
In order to define the actions of Kashiwara operators êi and f̂i for i = 0, 1, 2, we introduce
some notations and conditions. Set (x)+ = max(x, 0). For b = (x1, x2, x3, x̄3, x̄2, x̄1) ∈ B̂l we set
s(b) = x1 + x2 +
x3 + x̄3
2
+ x̄2 + x̄1, (2.1)
and
z1 = x̄1 − x1, z2 = x̄2 − x̄3, z3 = x3 − x2, z4 = (x̄3 − x3)/2. (2.2)
Now we define conditions (E1)–(E6) and (F1)–(F6) as follows
(F1) z1 + z2 + z3 + 3z4 ≤ 0, z1 + z2 + 3z4 ≤ 0, z1 + z2 ≤ 0, z1 ≤ 0,
(F2) z1 + z2 + z3 + 3z4 ≤ 0, z2 + 3z4 ≤ 0, z2 ≤ 0, z1 > 0,
(F3) z1 + z3 + 3z4 ≤ 0, z3 + 3z4 ≤ 0, z4 ≤ 0, z2 > 0, z1 + z2 > 0,
(F4) z1 + z2 + 3z4 > 0, z2 + 3z4 > 0, z4 > 0, z3 ≤ 0, z1 + z3 ≤ 0,
(F5) z1 + z2 + z3 + 3z4 > 0, z3 + 3z4 > 0, z3 > 0, z1 ≤ 0,
(F6) z1 + z2 + z3 + 3z4 > 0, z1 + z3 + 3z4 > 0, z1 + z3 > 0, z1 > 0.
(2.3)
Zero Action on Perfect Crystals for Uq
(
G
(1)
2
)
3
The conditions (F1)–(F6) are disjoint and they exhaust all cases. (Ei) (1 ≤ i ≤ 6) is defined
from (Fi) by replacing > (resp. ≤) with ≥ (resp. <). We also define
A = (0, z1, z1 + z2, z1 + z2 + 3z4, z1 + z2 + z3 + 3z4, 2z1 + z2 + z3 + 3z4). (2.4)
Then, for b = (x1, x2, x3, x̄3, x̄2, x̄1) ∈ B̂l, êib, f̂ib, ε̂i(b), ϕ̂i(b) are given as follows
ê0b =
(x1 − 1/3, . . .) if (E1),
(. . . , x3 − 1/3, x̄3 − 1/3, . . . , x̄1 + 1/3) if (E2),
(. . . , x3 − 2/3, . . . , x̄2 + 1/3, . . .) if (E3),
(. . . , x2 − 1/3, . . . , x̄3 + 2/3, . . .) if (E4),
(x1 − 1/3, . . . , x3 + 1/3, x̄3 + 1/3, . . .) if (E5),
(. . . , x̄1 + 1/3) if (E6),
f̂0b =
(x1 + 1/3, . . .) if (F1),
(. . . , x3 + 1/3, x̄3 + 1/3, . . . , x̄1 − 1/3) if (F2),
(. . . , x3 + 2/3, . . . , x̄2 − 1/3, . . .) if (F3),
(. . . , x2 + 1/3, . . . , x̄3 − 2/3, . . .) if (F4),
(x1 + 1/3, . . . , x3 − 1/3, x̄3 − 1/3, . . .) if (F5),
(. . . , x̄1 − 1/3) if (F6).
ê1b =
(. . . , x̄2 + 1/3, x̄1 − 1/3) if z2 ≥ (−z3)+,
(. . . , x3 + 1/3, x̄3 − 1/3, . . .) if z2 < 0 ≤ z3,
(x1 + 1/3, x2 − 1/3, . . .) if (z2)+ < (−z3),
f̂1b =
(x1 − 1/3, x2 + 1/3, . . .) if (z2)+ ≤ (−z3),
(. . . , x3 − 1/3, x̄3 + 1/3, . . .) if z2 ≤ 0 < z3,
(. . . , x̄2 − 1/3, x̄1 + 1/3) if z2 > (−z3)+,
ê2b =
{
(. . . , x̄3 + 2/3, x̄2 − 1/3, . . .) if z4 ≥ 0,
(. . . , x2 + 1/3, x3 − 2/3, . . .) if z4 < 0,
f̂2b =
{
(. . . , x2 − 1/3, x3 + 2/3, . . .) if z4 ≤ 0,
(. . . , x̄3 − 2/3, x̄2 + 1/3, . . .) if z4 > 0,
ε̂0(b) = l − 3s(b) + 3 maxA− 3(2z1 + z2 + z3 + 3z4),
ϕ̂0(b) = l − 3s(b) + 3 maxA,
ε̂1(b) = 3x̄1 + 3(x̄3 − x̄2 + (x2 − x3)+)+,
ϕ̂1(b) = 3x1 + 3(x3 − x2 + (x̄2 − x̄3)+)+,
ε̂2(b) = 3x̄2 + 3
2(x3 − x̄3)+, ϕ̂2(b) = 3x2 + 3
2(x̄3 − x3)+.
(2.5)
If êib or f̂ib does not belong to B̂l, namely, if xj or x̄j for some j becomes negative or s(b)
exceeds l/3, we should understand it to be 0. Forgetting the 0-arrows,
B̂l '
l⊕
j=0
BG†
2(jΛ1),
where BG†
2(λ) is the highest weight Uq(G
†
2)-crystal of highest weight λ and G†
2 stands for the
simple Lie algebra G2 with the reverse labeling of the indices of the simple roots (α1 is the short
4 K.C. Misra, M. Mohamad and M. Okado
root). Forgetting 2-arrows,
B̂l '
b l
2
c⊕
i=0
⊕
i≤j0,j1≤l−i
j0,j1≡l−i (mod 3)
BA2(j0Λ0 + j1Λ1),
where BA2(λ) is the highest weight Uq(A2)-crystal (with indices {0, 1}) of highest weight λ.
3 U ′
q
(
G
(1)
2
)
-crystal
In this section we define a subset Vl of B̂3l and see it is isomorphic to the U ′
q
(
G
(1)
2
)
-crystal Bl.
The set Vl is defined as a subset of B̂3l satisfying the following conditions:
x1, x̄1, x2 − x3, x̄3 − x̄2 ∈ Z. (3.1)
For an element b = (x1, x2, x3, x̄3, x̄2, x̄1) of Vl we define s(b) as in (2.1). From (3.1) we see that
s(b) ∈ {0, 1, . . . , l}.
Lemma 1. For 0 ≤ k ≤ l
]{b ∈ Vl | s(b) = k} =
1
120
(k + 1)(k + 2)(2k + 3)(3k + 4)(3k + 5).
Proof. We first count the number of elements (x2, x3, x̄3, x̄2) satisfying the conditions of co-
ordinates as an element of Vl and x2 + (x3 + x̄3)/2 + x̄2 = m (m = 0, 1, . . . , k). According to
(a, b, c, d) (a, d ∈ {0, 1/3, 2/3}, b, c ∈ {0, 1/3, 2/3, 1, 4/3, 5/3}) such that x2 ∈ Z + a, x3 ∈ 2Z + b,
x̄3 ∈ 2Z + c, x̄2 ∈ Z + d, we divide the cases into the following 18:
(i) (0, 0, 0, 0), (ii) (0, 0, 2/3, 2/3), (iii) (0, 0, 4/3, 1/3),
(iv) (0, 1, 1/3, 1/3), (v) (0, 1, 1, 0), (vi) (0, 1, 5/3, 2/3),
(vii) (1/3, 1/3, 1/3, 1/3), (viii) (1/3, 1/3, 1, 0), (ix) (1/3, 1/3, 5/3, 2/3),
(x) (1/3, 4/3, 0, 0), (xi) (1/3, 4/3, 2/3, 2/3), (xii) (1/3, 4/3, 4/3, 1/3),
(xiii) (2/3, 2/3, 0, 0), (xiv) (2/3, 2/3, 2/3, 2/3), (xv) (2/3, 2/3, 4/3, 1/3),
(xvi) (2/3, 5/3, 1/3, 1/3), (xvii) (2/3, 5/3, 1, 0), (xviii) (2/3, 5/3, 5/3, 2/3).
The number of elements (x2, x3, x̄3, x̄2) in a case among the above such that a+(b+c)/2+d = e
(e = 0, 1, 2, 3) is given by f(e) =
(
m−e+3
3
)
. Since there is one case with e = 0 (i) and e = 3 (xviii)
and 8 cases with e = 1 and e = 2, the number of (x2, x3, x̄3, x̄2) such that x2+(x3+x̄3)/2+x̄2 = m
is given by
f(0) + 8f(1) + 8f(2) + f(3) =
1
2
(2m+ 1)(3m2 + 3m+ 2).
For each (x2, x3, x̄3, x̄2) such that x2+(x3+x̄3)/2+x̄2 = m (m = 0, 1, . . . , k) there are (k−m+1)
cases for (x1, x̄1), so the number of b ∈ Vl such that s(b) = k is given by
k∑
m=0
1
2
(2m+ 1)(3m2 + 3m+ 2)(k −m+ 1).
A direct calculation leads to the desired result. �
Zero Action on Perfect Crystals for Uq
(
G
(1)
2
)
5
We define the action of operators ei, fi (i = 0, 1, 2) on Vl as follows.
e0b =
(x1 − 1, . . .) if (E1),
(. . . , x3 − 1, x̄3 − 1, . . . , x̄1 + 1) if (E2),(
. . . , x2 − 2
3 , x3 − 2
3 , x̄3 + 4
3 , x̄2 + 1
3 , . . .
)
if (E3) and z4 = −1
3 ,(
. . . , x2 − 1
3 , x3 − 4
3 , x̄3 + 2
3 , x̄2 + 2
3 , . . .
)
if (E3) and z4 = −2
3 ,
(. . . , x3 − 2, . . . , x̄2 + 1, . . .) if (E3) and z4 6= −1
3 ,−
2
3 ,
(. . . , x2 − 1, . . . , x̄3 + 2, . . .) if (E4),
(x1 − 1, . . . , x3 + 1, x̄3 + 1, . . .) if (E5),
(. . . , x̄1 + 1) if (E6),
f0b =
(x1 + 1, . . .) if (F1),
(. . . , x3 + 1, x̄3 + 1, . . . , x̄1 − 1) if (F2),
(. . . , x3 + 2, . . . , x̄2 − 1, . . .) if (F3),(
. . . , x2 + 1
3 , x3 + 4
3 , x̄3 − 2
3 , x̄2 − 2
3 , . . .
)
if (F4) and z4 = 1
3 ,(
. . . , x2 + 2
3 , x3 + 2
3 , x̄3 − 4
3 , x̄2 − 1
3 , . . .
)
if (F4) and z4 = 2
3 ,
(. . . , x2 + 1, . . . , x̄3 − 2, . . .) if (F4) and z4 6= 1
3 ,
2
3 ,
(x1 + 1, . . . , x3 − 1, x̄3 − 1, . . .) if (F5),
(. . . , x̄1 − 1) if (F6),
e1b =
(. . . , x̄2 + 1, x̄1 − 1) if x̄2 − x̄3 ≥ (x2 − x3)+,
(. . . , x3 + 1, x̄3 − 1, . . .) if x̄2 − x̄3 < 0 ≤ x3 − x2,
(x1 + 1, x2 − 1, . . .) if (x̄2 − x̄3)+ < x2 − x3,
f1b =
(x1 − 1, x2 + 1, . . .) if (x̄2 − x̄3)+ ≤ x2 − x3,
(. . . , x3 − 1, x̄3 + 1, . . .) if x̄2 − x̄3 ≤ 0 < x3 − x2,
(. . . , x̄2 − 1, x̄1 + 1) if x̄2 − x̄3 > (x2 − x3)+,
e2b =
{(
. . . , x̄3 + 2
3 , x̄2 − 1
3 , . . .
)
if x̄3 ≥ x3,(
. . . , x2 + 1
3 , x3 − 2
3 , . . .
)
if x̄3 < x3,
f2b =
{(
. . . , x2 − 1
3 , x3 + 2
3 , . . .
)
if x̄3 ≤ x3,(
. . . , x̄3 − 2
3 , x̄2 + 1
3 , . . .
)
if x̄3 > x3.
We now set (m0,m1,m2) = (3, 3, 1).
Proposition 1.
(1) For any b ∈ Vl we have eib, fib ∈ Vl t {0} for i = 0, 1, 2.
(2) The equalities ei = êmi
i and fi = f̂mi
i hold on Vl for i = 0, 1, 2.
Proof. (1) can be checked easily.
For (2) we only treat fi. To prove the i = 0 case consider the following table
(F1) (F2) (F3) (F4) (F5) (F6)
z1 −1/3 −1/3 0 0 −1/3 −1/3
z2 0 −1/3 −1/3 2/3 1/3 0
z3 0 1/3 2/3 −1/3 −1/3 0
z4 0 0 −1/3 −1/3 0 0
6 K.C. Misra, M. Mohamad and M. Okado
This table signifies the difference (zj for f̂0b)− (zj for b) when b belongs to the case (Fi). Note
that the left hand sides of the inequalities of each (Fi) (2.3) always decrease by 1/3. Since
z1, z2, z3 ∈ Z, z4 ∈ Z/3 for b ∈ Vl, we see that if b belongs to (Fi), f̂0b and f̂2
0 b also belong to (Fi)
except two cases: (a) b ∈ (F4) and z4 = 1/3, and (b) b ∈ (F4) and z4 = 2/3. If (a) occurs, we
have f̂0b, f̂
2
0 b ∈ (F3). Hence, we obtain f0 = f̂3
0 in this case. If (b) occurs, we have f̂0b ∈ (F4),
f̂2
0 b ∈ (F3). Therefore, we obtain f0 = f̂3
0 in this case as well.
In the i = 1 case, if b belongs to one of the 3 cases, f̂1b and f̂2
1 b also belong to the same case.
Hence, we obtain f1 = f̂3
1 . For i = 2 there is nothing to do. �
Proposition 1, together with Theorem 1, shows that Vl can be regarded as a U ′
q
(
G
(1)
2
)
-crystal
with operators ei, fi (i = 0, 1, 2).
Proposition 2. As a Uq
(
G
(1)
2
)
{1,2}-crystal
Vl '
l⊕
k=0
BG2(kΛ1),
where BG2(λ) is the highest weight Uq(G2)-crystal of highest weight λ.
Proof. For a subset J of {0, 1, 2} we say b is J-highest if ejb = 0 for any j ∈ J . Note from (2.5)
that bk = (k, 0, 0, 0, 0, 0) (0 ≤ k ≤ l) is {1, 2}-highest of weight 3kΛ1 in B̂3l. By setting g = G†
2
(= G2 with the reverse labeling of indices), (m1,m2) = (3, 1), g̃ = G2 in Theorem 1, we know
that the connected component generated from bk by f1 = f̂3
1 and f2 = f̂2 is isomorphic to
BG2(kΛ1). Hence by Proposition 1 (1) we have
l⊕
k=0
BG2(kΛ1) ⊂ Vl. (3.2)
Now recall Weyl’s formula to calculate the dimension of the highest weight representation. In
our case we obtain
]BG2(kΛ1) =
1
120
(k + 1)(k + 2)(2k + 3)(3k + 4)(3k + 5).
However, this is equal to ]{b ∈ Vl | s(b) = k} by Lemma 1. Therefore, ⊂ in (3.2) should be =,
and the proof is completed. �
Proposition 3. As a Uq
(
G
(1)
2
)
{0,1}-crystal
Vl '
bl/2c⊕
i=0
⊕
i≤j0,j1≤l−i
BA2(j0Λ0 + j1Λ1),
where BA2(λ) is the highest weight Uq(A2)-crystal (with indices {0, 1}) of highest weight λ.
Proof. For integers i, j0, j1 such that 0 ≤ i ≤ l/2, i ≤ j0, j1 ≤ l − i, define an element bi,j0,j1
of Vl by
bi,j0,j1 =
{
(0, y1, 3y0 − 2y1 + i, y0 + i, y0 + j0, 0) if j0 ≤ j1,
(0, y0, y0 + i, 2y1 − y0 + i, 2y0 − y1 + j0, 0) if j0 > j1.
Here we have set ya = (l − i − ja)/3 for a = 0, 1. From (2.5) one notices that bi,j0,j1 is {0, 1}-
highest of weight 3j0Λ0 + 3j1Λ1 in B̂3l. For instance, ε̂0(bi,j0,j1) = 0 and ϕ̂0(bi,j0,j1) = 3j0 since
Zero Action on Perfect Crystals for Uq
(
G
(1)
2
)
7
s(bi,j0,j1) = l and maxA = 2z1 + z2 + z3 + 3z4 = j0. By setting g = g̃ = A2, (m0,m1) = (3, 3) in
Theorem 1, the connected component generated from bi,j0,j1 by fi = f̂3
i (i = 0, 1) is isomorphic
to BA2(j0Λ0 + j1Λ1). Hence, by Proposition 1 (1) we have
bl/2c⊕
i=0
⊕
i≤j0,j1≤l−i
BA2(j0Λ0 + j1Λ1) ⊂ Vl.
However, from Proposition 2 one knows that
]Vl =
l∑
k=0
]BG2(kΛ1).
Moreover, it is already established in [7] that
l∑
k=0
]BG2(kΛ1) =
bl/2c∑
i=0
∑
i≤j0,j1≤l−i
]BA2(j0Λ0 + j1Λ1).
Therefore, the proof is completed. �
Theorem 6.1 in [6] shows that if two U ′
q
(
G
(1)
2
)
-crystals decompose into
⊕
0≤k≤lB
G2(kΛ1) as
Uq(G2)-crystals, then they are isomorphic to each other. Therefore, we now have
Theorem 2. Vl agrees with the U ′
q
(
G
(1)
2
)
-crystal Bl constructed in [7]. The values of εi, ϕi with
our representation are given by
ε0(b) = l − s(b) + maxA− (2z1 + z2 + z3 + 3z4), ϕ0(b) = l − s(b) + maxA,
ε1(b) = x̄1 + (x̄3 − x̄2 + (x2 − x3)+)+, ϕ1(b) = x1 + (x3 − x2 + (x̄2 − x̄3)+)+,
ε2(b) = 3x̄2 + 3
2(x3 − x̄3)+, ϕ2(b) = 3x2 + 3
2(x̄3 − x3)+.
(3.3)
4 Minimal elements and a coherent family
The notion of perfect crystals was introduced in [3] to construct the path realization of a highest
weight crystal of a quantum affine algebra. The crystal Bl was shown to be perfect of level l
in [7]. In this section we obtain all the minimal elements of Bl in the coordinate representation
and also show {Bl}l≥1 forms a coherent family of perfect crystals. For the notations such as Pcl,
(P+
cl )l, see [3].
4.1 Minimal elements
From (3.3) we have
〈c, ϕ(b)〉 = ϕ0(b) + 2ϕ1(b) + ϕ2(b)
= l + maxA+ 2(z3 + (z2)+)+ + (3z4)+ − (z1 + z2 + 2z3 + 3z4),
where zj (1 ≤ j ≤ 4) are given in (2.2) and A is given in (2.4). The following lemma was proven
in [6], although Z is replaced with Z/3 here.
Lemma 2. For (z1, z2, z3, z4) ∈ (Z/3)4 set
ψ(z1, z2, z3, z4) = maxA+ 2(z3 + (z2)+)+ + (3z4)+ − (z1 + z2 + 2z3 + 3z4).
Then we have ψ(z1, z2, z3, z4) ≥ 0 and ψ(z1, z2, z3, z4) = 0 if and only if (z1, z2, z3, z4) =
(0, 0, 0, 0).
8 K.C. Misra, M. Mohamad and M. Okado
From this lemma, we have 〈c, ϕ(b)〉 − l = ψ(z1, z2, z3, z4) ≥ 0. Since 〈c, ϕ(b) − ε(b)〉 = 0, we
also have 〈c, ε(b)〉 ≥ l.
Suppose 〈c, ε(b)〉 = l. It implies ψ = 0. Hence from the lemma one can conclude that such
element b = (x1, x2, x3, x̄3, x̄2, x̄1) should satisfy x1 = x̄1, x2 = x3 = x̄3 = x̄2. Therefore the set
of minimal elements (Bl)min in Bl is given by
(Bl)min = {(α, β, β, β, β, α)| α ∈ Z≥0, β ∈ (Z≥0)/3, 2α+ 3β ≤ l}.
For b = (α, β, β, β, β, α) ∈ (Bl)min one calculates
ε(b) = ϕ(b) = (l − 2α− 3β)Λ0 + αΛ1 + 3βΛ2.
4.2 Coherent family of perfect crystals
The notion of a coherent family of perfect crystals was introduced in [1]. Let {Bl}l≥1 be a family
of perfect crystals Bl of level l and (Bl)min be the subset of minimal elements of Bl. Set
J = {(l, b) | l ∈ Z>0, b ∈ (Bl)min}. Let σ denote the isomorphism of (P+
cl )l defined by σ = ε◦ϕ−1.
For λ ∈ Pcl, Tλ denotes a crystal with a unique element tλ defined in [4]. For our purpose the
following facts are sufficient. For any Pcl-weighted crystal B and λ, µ ∈ Pcl consider the crystal
Tλ ⊗B ⊗ Tµ = {tλ ⊗ b⊗ tµ | b ∈ B}.
The definition of Tλ and the tensor product rule of crystals imply
ẽi(tλ ⊗ b⊗ tµ) = tλ ⊗ ẽib⊗ tµ, f̃i(tλ ⊗ b⊗ tµ) = tλ ⊗ f̃ib⊗ tµ,
εi(tλ ⊗ b⊗ tµ) = εi(b)− 〈hi, λ〉, ϕi(tλ ⊗ b⊗ tµ) = ϕi(b) + 〈hi, µ〉,
wt (tλ ⊗ b⊗ tµ) = λ+ µ+ wt b.
Definition 1. A crystal B∞ with an element b∞ is called a limit of {Bl}l≥1 if it satisfies the
following conditions:
• wt b∞ = 0, ε(b∞) = ϕ(b∞) = 0,
• for any (l, b) ∈ J , there exists an embedding of crystals
f(l,b) : Tε(b) ⊗Bl ⊗ T−ϕ(b) −→ B∞
sending tε(b) ⊗ b⊗ t−ϕ(b) to b∞,
• B∞ =
⋃
(l,b)∈J Im f(l,b).
If a limit exists for the family {Bl}, we say that {Bl} is a coherent family of perfect crystals.
Let us now consider the following set
B∞ =
{
b = (ν1, ν2, ν3, ν̄3, ν̄2, ν̄1) ∈ (Z/3)6
∣∣∣∣ ν1, ν̄1, ν2 − ν3, ν̄3 − ν̄2 ∈ Z,
3ν3 ≡ 3ν̄3 (mod 2)
}
,
and set b∞ = (0, 0, 0, 0, 0, 0). We introduce the crystal structure on B∞ as follows. The actions
of ei, fi (i = 0, 1, 2) are defined by the same rule as in Section 3 with xi and x̄i replaced with νi
and ν̄i. The only difference lies in the fact that eib or fib never becomes 0, since we allow
a coordinate to be negative and there is no restriction for the sum s(b) =
2∑
i=1
(νi+ν̄i)+(ν3+ν̄3)/2.
For εi, ϕi with i = 1, 2 we adopt the formulas in Section 3. For ε0, ϕ0 we define
ε0(b) = −s(b) + maxA− (2z1 + z2 + z3 + 3z4), ϕ0(b) = −s(b) + maxA,
Zero Action on Perfect Crystals for Uq
(
G
(1)
2
)
9
where A is given in (2.4) and z1, z2, z3, z4 are given in (2.2) with xi, x̄i replaced by νi, ν̄i. Note
that wt b∞ = 0 and εi(b∞) = ϕi(b∞) = 0 for i = 0, 1, 2.
Let b0 = (α, β, β, β, β, α) be an element of (Bl)min. Since ε(b0) = ϕ(b0), one can set σ = id.
Let λ = ε(b0). For b = (x1, x2, x3, x̄3, x̄2, x̄1) ∈ Bl we define a map
f(l,b0) : Tλ ⊗Bl ⊗ T−λ −→ B∞
by
f(l,b0)(tλ ⊗ b⊗ t−λ) = b′ = (ν1, ν2, ν3, ν̄3, ν̄2, ν̄1),
where
ν1 = x1 − α, ν̄1 = x̄1 − α,
νj = xj − β, ν̄j = x̄j − β (j = 2, 3).
Note that s(b′) = s(b)− (2α+ 3β). Then we have
wt (tλ ⊗ b⊗ t−λ) = wt b = wt b′,
ϕ0(tλ ⊗ b⊗ t−λ) = ϕ0(b) + 〈h0,−λ〉
= ϕ0(b′) + (l − s(b)) + s(b′)− (l − 2α− 3β) = ϕ0(b′),
ϕ1(tλ ⊗ b⊗ t−λ) = ϕ1(b) + 〈h1,−λ〉 = ϕ1(b′) + α− α = ϕ1(b′),
ϕ2(tλ ⊗ b⊗ t−λ) = ϕ2(b) + 〈h2,−λ〉 = ϕ2(b′) + 3β − 3β = ϕ2(b′).
εi(tλ ⊗ b⊗ t−λ) = εi(b′) (i = 0, 1, 2) also follows from the above calculations.
From the fact that (zj for b) = (zj for b′) it is straightforward to check that if b, eib ∈ Bl
(resp. b, fib ∈ Bl), then f(l,b0)(ei(tλ ⊗ b ⊗ t−λ)) = eif(l,b0)(tλ ⊗ b ⊗ t−λ) (resp. f(l,b0)(fi(tλ ⊗
b ⊗ t−λ)) = fif(l,b0)(tλ ⊗ b ⊗ t−λ)). Hence f(l,b0) is a crystal embedding. It is easy to see that
f(l,b0)(tλ ⊗ b0 ⊗ t−λ) = b∞. We can also check B∞ =
⋃
(l,b)∈J Im f(l,b). Therefore we have shown
that the family of perfect crystals {Bl}l≥1 forms a coherent family.
5 Crystal graphs of B1 and B2
In this section we present crystal graphs of the U ′
q
(
G
(1)
2
)
-crystals B1 and B2 in Figs. 1 and 2.
In the graphs b i−→ b′ stands for b′ = fib. Minimal elements are marked as ∗. Recall that as
a Uq(G2)-crystal
B1 ' B(0)⊕B(Λ1), B2 ' B(0)⊕B(Λ1)⊕B(2Λ1).
We give the table that relates the numbers in the crystal graphs to our representation of elements
according to which Uq(G2)-components they belong to.
B(0) : φ∗ = (0, 0, 0, 0, 0, 0)
B(Λ1):
1 = (1, 0, 0, 0, 0, 0) 2 = (0, 1, 0, 0, 0, 0) 3 =
(
0, 2
3 ,
2
3 , 0, 0, 0
)
4 =
(
0, 1
3 ,
4
3 , 0, 0, 0
)
5 =
(
0, 1
3 ,
1
3 , 1, 0, 0
)
6∗ =
(
0, 1
3 ,
1
3 ,
1
3 ,
1
3 , 0
)
7 = (0, 0, 1, 1, 0, 0) 8 =
(
0, 0, 1, 1
3 ,
1
3 , 0
)
9 =
(
0, 0, 0, 4
3 ,
1
3 , 0
)
10 =
(
0, 0, 0, 2
3 ,
2
3 , 0
)
11 = (0, 0, 0, 0, 1, 0) 12 = (0, 0, 0, 0, 0, 1)
13 = (0, 0, 2, 0, 0, 0) 14 = (0, 0, 0, 2, 0, 0)
10 K.C. Misra, M. Mohamad and M. Okado
Figure 1. Crystal graph of B1. ↙ is f1 and ↘ is f2.
B(2Λ1):
15 = (2, 0, 0, 0, 0, 0) 16 = (1, 1, 0, 0, 0, 0) 17 =
(
1, 2
3 ,
2
3 , 0, 0, 0
)
18 =
(
1, 1
3 ,
4
3 , 0, 0, 0
)
19 =
(
1, 1
3 ,
1
3 , 1, 0, 0
)
20 =
(
1, 1
3 ,
1
3 ,
1
3 ,
1
3 , 0
)
21 = (1, 0, 1, 1, 0, 0) 22 =
(
1, 0, 1, 1
3 ,
1
3 , 0
)
23 =
(
1, 0, 0, 4
3 ,
1
3 , 0
)
24 =
(
1, 0, 0, 2
3 ,
2
3 , 0
)
25 = (1, 0, 0, 0, 1, 0) 26∗ = (1, 0, 0, 0, 0, 1)
27 = (1, 0, 2, 0, 0, 0) 28 = (1, 0, 0, 2, 0, 0) 29 = (0, 2, 0, 0, 0, 0) 30 =
(
0, 5
3 ,
2
3 , 0, 0, 0
)
31 =
(
0, 4
3 ,
4
3 , 0, 0, 0
)
32 =
(
0, 4
3 ,
1
3 , 1, 0, 0
)
33 =
(
0, 4
3 ,
1
3 ,
1
3 ,
1
3 , 0
)
34 = (0, 1, 1, 1, 0, 0)
35 =
(
0, 1, 1, 1
3 ,
1
3 , 0
)
36 =
(
0, 1, 0, 4
3 ,
1
3 , 0
)
37 =
(
0, 1, 0, 2
3 ,
2
3 , 0
)
38 = (0, 1, 0, 0, 1, 0)
39 = (0, 1, 0, 0, 0, 1) 40 = (0, 1, 2, 0, 0, 0) 41 = (0, 1, 0, 2, 0, 0) 42 =
(
0, 2
3 ,
2
3 , 0, 1, 0
)
43 =
(
0, 1
3 ,
4
3 , 0, 1, 0
)
44 =
(
0, 1
3 ,
1
3 , 1, 1, 0
)
45 =
(
0, 1
3 ,
1
3 ,
1
3 ,
4
3 , 0
)
46 = (0, 0, 1, 1, 1, 0)
47 =
(
0, 0, 1, 1
3 ,
4
3 , 0
)
48 =
(
0, 0, 0, 4
3 ,
4
3 , 0
)
49 =
(
0, 0, 0, 2
3 ,
5
3 , 0
)
50 = (0, 0, 0, 0, 2, 0)
51 = (0, 0, 0, 0, 1, 1) 52 = (0, 0, 2, 0, 1, 0) 53 = (0, 0, 0, 2, 1, 0) 54 =
(
0, 2
3 ,
2
3 , 0, 0, 1
)
55 =
(
0, 1
3 ,
4
3 , 0, 0, 1
)
56 =
(
0, 1
3 ,
1
3 , 1, 0, 1
)
57 =
(
0, 1
3 ,
1
3 ,
1
3 ,
1
3 , 1
)
58 = (0, 0, 1, 1, 0, 1)
59 =
(
0, 0, 1, 1
3 ,
1
3 , 1
)
60 =
(
0, 0, 0, 4
3 ,
1
3 , 1
)
61 =
(
0, 0, 0, 2
3 ,
2
3 , 1
)
62 = (0, 0, 0, 0, 0, 2)
63 = (0, 0, 2, 0, 0, 1) 64 = (0, 0, 0, 2, 0, 1) 65 =
(
0, 2
3 ,
8
3 , 0, 0, 0
)
66 =
(
0, 1
3 ,
10
3 , 0, 0, 0
)
67 =
(
0, 1
3 ,
7
3 , 1, 0, 0
)
68 =
(
0, 1
3 ,
7
3 ,
1
3 ,
1
3 , 0
)
69 = (0, 0, 3, 1, 0, 0) 70 =
(
0, 0, 3, 1
3 ,
1
3 , 0
)
71 =
(
0, 0, 2, 4
3 ,
1
3 , 0
)
72 =
(
0, 0, 2, 2
3 ,
2
3 , 0
)
73 = (0, 0, 4, 0, 0, 0) 74 = (0, 0, 2, 2, 0, 0)
75 =
(
0, 2
3 ,
2
3 , 2, 0, 0
)
76 =
(
0, 1
3 ,
4
3 , 2, 0, 0
)
77 =
(
0, 1
3 ,
1
3 , 3, 0, 0
)
78 =
(
0, 1
3 ,
1
3 ,
7
3 ,
1
3 , 0
)
79 = (0, 0, 1, 3, 0, 0) 80 =
(
0, 0, 1, 7
3 ,
1
3 , 0
)
81 =
(
0, 0, 0, 10
3 ,
1
3 , 0
)
82 =
(
0, 0, 0, 8
3 ,
2
3 , 0
)
83 = (0, 0, 0, 4, 0, 0) 84 =
(
0, 2
3 ,
5
3 , 1, 0, 0
)
85 =
(
0, 1
3 ,
4
3 ,
4
3 ,
1
3 , 0
)
86 =
(
0, 0, 1, 5
3 ,
2
3 , 0
)
87 =
(
0, 2
3 ,
5
3 ,
1
3 ,
1
3 , 0
)
88 =
(
0, 2
3 ,
2
3 ,
4
3 ,
1
3 , 0
)
89 =
(
0, 1
3 ,
1
3 ,
5
3 ,
2
3 , 0
)
90∗ =
(
0, 2
3 ,
2
3 ,
2
3 ,
2
3 , 0
)
91 =
(
0, 1
3 ,
4
3 ,
2
3 ,
2
3 , 0
)
Comparing our crystal graphs with those in [7] we found that some 2-arrows are missing in
Fig. 3 of [7].
Zero Action on Perfect Crystals for Uq
(
G
(1)
2
)
11
Figure 2. Crystal graph of B2. ↘ is f0, ↙ is f1 and others are f2.
Acknowledgements
KCM thanks the faculty and staff of Osaka University for their hospitality during his visit in
August, 2009 and acknowledges partial support from NSA grant H98230-08-1-0080. MM would
like to thank Universiti Tun Hussein Onn Malaysia for supporting this study. MO would like
to thank the organizers of the conference “Geometric Aspects of Discrete and Ultra-Discrete
Integrable Systems” held during March 30 – April 3, 2009 at Glasgow for a warm hospitality
and acknowledges partial support from JSPS grant No. 20540016.
12 K.C. Misra, M. Mohamad and M. Okado
References
[1] Kang S.-J., Kashiwara M., Misra K.C., Crystal bases of Verma modules for quantum affine Lie algebras,
Compositio Math. 92 (1994), 299–325.
[2] Kang S.-J., Misra K.C., Crystal bases and tensor product decompositions of Uq(G2)-module, J. Algebra 163
(1994), 675–691.
[3] Kang S.-J., Kashiwara M., Misra K.C., Miwa T., Nakashima T., Nakayashiki A., Affine crystals and vertex
models, in Infinite Analysis, Part A, B (Kyoto, 1991), Adv. Ser. Math. Phys., Vol. 16, World Sci. Publ.,
River Edge, NJ, 1992, 449–484.
[4] Kashiwara M., Crystal bases of modified quantized enveloping algebra, Duke Math. J. 73 (1994), 383–413.
[5] Kashiwara M., Similarity of crystal bases, in Lie Algebras and Their Representations (Seoul, 1995), Contemp.
Math., Vol. 194, Amer. Math. Soc., Providence, RI, 1996, 177–186.
[6] Kashiwara M., Misra K.C., Okado M., Yamada D., Perfect crystals for Uq
(
D
(3)
4
)
, J. Algebra 317 (2007),
392–423, math.QA/0610873.
[7] Yamane S., Perfect crystals of Uq
(
G
(1)
2
)
, J. Algebra 210 (1998), 440–486, q-alg/9712012.
http://dx.doi.org/10.1006/jabr.1994.1037
http://dx.doi.org/10.1215/S0012-7094-94-07317-1
http://dx.doi.org/10.1016/j.jalgebra.2007.02.021
http://arxiv.org/abs/math.QA/0610873
http://dx.doi.org/10.1006/jabr.1998.7597
http://arxiv.org/abs/q-alg/9712012
1 Introduction
2 Review on U'q(D4(3))-crystal l
3 U'q(G2(1))-crystal
4 Minimal elements and a coherent family
4.1 Minimal elements
4.2 Coherent family of perfect crystals
5 Crystal graphs of B1 and B2
References
|