1991 Mathematics Subject Classification. Primary 17B69.The author was supported in part by an NSA grant.
<ph f="cmbx">Pseudoderivations, pseudoautomorphisms and simple current modules for vertex algebras</ph>

Haisheng Li

Department of Mathematics, Harbin Normal University, Harbin, China Current address : Department of Mathematical Sciences, Rutgers University, Camden, NJ 08102 E-mail address : hli@camden.rutgers.edu

1 Introduction

Vertex (operator) algebras are analogous to classical Lie algebras and associative algebras in many aspects, though there are also significant differences in many aspects. For a vertex (operator) algebra V   , one has the basic notions of derivation, endomorphism and automorphism. A derivation of a vertex algebra V   is a linear endomorphism d   of V   such that d Y ( u , x ) v = Y ( u , x ) d v + Y ( d u , x ) v for u , v V .   Just as with Lie algebras and associative algebras, the exponential e d   of a locally finite derivation d   of V   is an automorphism of V   .
In a study on formal deformation of vertex operator algebras, Etingof and Kazhdan in [EKintroduced and studied a notion of pseudoderivation for a vertex algebra. By definition, a pseudoderivation of a vertex algebra V   is a linear map a ( z )   from V   to C ( ( z ) ) V   , where z   is a formal variable, such that
[ L ( 1 ) , a ( z ) ] = d d z a ( z ) , (1.1)
[ a ( z ) , Y ( v , x ) ] = Y ( a ( z + x ) v , x ) for v V . (1.2)
It was proved therein that a ( z ) 1 = 0   . Just as derivations are important in the study of formal deformations of classical Lie algebras, pseudoderivations are important in the study of formal deformations of vertex operator algebras.
In another (seemingly unrelated) study [Li2, motivated by the physics superselection principle in (algebraic) conformal field theory, we studied how to deform the adjoint module for a vertex operator algebra V   to obtain new modules. The formulation of the physics superselection principle is based on the following simple fact: For an algebra of many types (Lie algebras, associative algebras and vertex algebras), if σ   is an endomorphism of the algebra, then for any module U   , we have a “new” module structure on U   defined by a * u = σ ( a ) u   for a R , u U   . Having noticed that the σ   -twisting of the adjoint representation by an automorphism σ   is always isomorphic to the adjoint representation we considered twisting by deformed endomorphisms of a certain type. It was proved in [Li2that if Δ ( z ) H o m ( V , C ( ( z ) ) V )   satisfies the conditions
Δ ( z ) 1 = 1 ( = 1 1 ) , (1.3)
[ L ( 1 ) , Δ ( z ) ] = d d z Δ ( z ) , (1.4)
Δ ( z ) Y ( v , x ) = Y ( Δ ( z + x ) v , x ) Δ ( z ) for v V , (1.5)
then for any V   -module ( W , Y W )   , ( W , Y W ( Δ ( x ) , x ) )   carries the structure of a V   -module. (Notice that if Δ ( z ) = σ   is independent of z   , then the above three conditions exactly amount to that σ   is an endomorphism of vertex algebra V   .) Furthermore, we gave an explicit construction of such invertible Δ ( z )   and we applied the results to the vertex operator algebras associated with affine Lie algebras, with Heisenberg Lie algebras and with nondegenerate even lattices. Indeed, the twisting of the adjoint module gives rise to new modules, which were proved to be “simple current modules” in a certain sense. In this way we obtained a construction of simple current modules.
Now, it is manifest that Etingof-Kazhdan's notion of pseudoderivation and our construction of simple current modules are much more than simply related. The main purpose of this short note is to exhibit this connection explicitly and conceptually.
In Section 2, we formulate notions of pseudoendomorphism and pseudoautomorphism, using ( 1.3 )-( 1.5 ). It is proved that that if a ( z )   is a locally nilpotent pseudoderivation of V   such that [ a ( z 1 ) , a ( z 2 ) ] = 0   , e a ( z )   is a pseudoautomorphism of V   . We show that the Lie algebra P D e r ( V )   is canonically isomorphic to the Lie algebra of all C ( ( z ) )   -linear derivations of the tensor product vertex algebra C ( ( z ) ) V   (see [FLM), where C ( ( z ) )   is the Borcherds' vertex algebra associated with the unital commutative associative algebra C ( ( z ) )   together with the derivation D = d / d z   .
Similarly, we show that any endomorphism of C ( ( z ) ) V   , restricted to V   , is a pseudoendomorphism of V   , and that the C ( ( z ) )   -linear extension of every pseudoendomorphism of V   is an endomorphism of vertex algebra C ( ( z ) ) V   . In Section 3, we reinterpret the construction of simple current modules in terms of pseudoautomorphisms.

2 Pseudoderivations and pseudoendomorphisms

In this section we review Etingof-Kazhdan's notion of pseudoderivation and define a notion of pseudoendomorphism for a vertex algebra V   and we then relate pseudoderivations and pseudoendomorphism of V   with derivations and endomorphisms of the tensor product vertex algebra C ( ( z ) ) V   .
We shall not recall the full definition of a vertex algebra, but for convenience we just mention the main ingredients. For a vertex algebra V   , we have the vacuum vector 1   and the vertex operator map Y : V H o m ( V , V ( ( x ) ) ) ( E n d V ) [ [ x , x 1 ] ]   , where Y ( v , x ) = n Z v n x n 1 ( with v n E n d V ) .   There is a canonical linear operator D   on V   defined by
D ( v ) = v 2 1 for v V . (2.1)
The basic properties about D   are
Y ( u , x ) 1 = e x D u , (2.2)
Y ( u , x ) v = e x D Y ( v , x ) u , (2.3)
[ D , Y ( v , x ) ] = Y ( D v , x ) = d d x Y ( v , x ) for u , v V . (2.4)
An endomorphism of a vertex algebra V   is a linear endomorphism f   of V   such that f ( 1 ) = 1   and
f Y ( u , x ) v = Y ( f ( u ) , x ) f ( v ) for u , v V . (2.5)
An automorphism of a vertex algebra V   is a bijective endomorphism.
A derivation of a vertex algebra V   (cf. [B], [Lia]) is a linear endomorphism d   such that
d Y ( u , x ) v = Y ( u , x ) d v + Y ( d u , x ) v for u , v V . (2.6)
The operator D   is a derivation of V   and it follows from Borcherds' commutator formula that for any v V   , v 0   is a derivation of V   .
If d   is a derivation of V   , one can show that d 1 = 0   and [ D , d ] = 0   . It is straightforward to see that all the derivations of V   form a Lie subalgebra D e r V   of g l ( V )   (the general Lie algebra), with an ideal consisting of v 0   for v V   .
Remark 2.1. If d   is a derivation of V   , which is locally finite in the sense that for any v V   , the vectors d n w   for n 0   span a finite-dimensional subspace of V   , then e d   is an automorphism of the vertex algebra V   .
The following notion is due to Etingof and Kazhdan [EK:
Definition 2.2. Let V   be a vertex algebra. A pseudoderivation of a vertex algebra V   is a linear homomorphism a ( z )   from V   to C ( ( z ) ) V   such that
[ D , a ( z ) ] = d d z a ( z ) , (2.7)
[ a ( z ) , Y ( u , x ) ] = Y ( a ( z + x ) u , x ) for u V , (2.8)
where D   acts on C ( ( z ) ) V   as 1 D   and the map Y   is C ( ( z ) ) [ [ x ] ]   -linearly extended to C ( ( z ) ) [ [ x ] ] V   .
All the pseudoderivations of vertex algebra V   clearly form a C   -subspace of H o m ( V , C ( ( z ) ) V )   , which is denoted by P D e r ( V )   , as in [EK. Note that we have
H o m C ( V , C ( ( z ) ) V ) E n d C ( ( z ) ) ( C ( ( z ) ) V ) (2.9)
as C   -vector spaces. As it was pointed out in [EK, P D e r ( V )   is naturally a Lie subalgebra of g l ( C ( ( z ) ) V )   .
We have the following results, among which the second assertion was proved in [EK:
Proposition 2.3. In Definition  2.2 , ( 2.7 ) follows from ( 2.8 ). Furthermore, for any pseudoderivation a ( z )   we have a ( z ) 1 = 0   .
  • Proof. Since Y ( 1 , x ) = 1   , from ( 2.8 ) we have Y ( a ( z + x ) 1 , x ) = 0   . As the map Y   is injective (see [FHL]), we have a ( z + x ) 1 = 0   . Thus a ( z ) 1 = 0   .
    Now applying both sides of ( 2.8 ) to the vacuum vector 1   and using the fact that a ( z ) 1 = 0   , we get
    a ( z ) Y ( u , x ) 1 = Y ( a ( z + x ) u , x ) 1 .
    Thus
    a ( z ) e x D u = e x D a ( z + x ) u = e x D e x d / d z a ( z ) u .
    That is,
    e x D a ( z ) e x D u = e x d / d z a ( z ) u , (2.10)
    from which we immediately have ( 2.7 ).
The following result, due to [EK, gives a construction of pseudoderivations:
Proposition 2.4. Let V   be a vertex algebra, let v V   and let f ( z ) C ( ( z ) )   . Then
X f , v = R e s x f ( z + x ) Y ( v , x ) = n 0 f ( n ) ( z ) n ! v n (2.11)
is a pseudoderivation of V   .
As we have done so in Definition  2.2 , we extend the C   -linear map Y   from V   to H o m ( V , V ( ( x ) ) )   to a C ( ( z ) )   -linear map from C ( ( z ) ) V   to ( E n d C ( ( z ) ) ( C ( ( z ) ) V ) ) [ [ x , x 1 ] ]   . With C ( ( z ) )   as a field, C ( ( z ) ) V   equipped with the C ( ( z ) )   -linear map Y   is naturally a vertex algebra over C ( ( z ) )   .
On the other hand, C ( ( z ) )   is a unital commutative associative algebra over C   with d / d z   as a derivation. From [B], C ( ( z ) )   is a vertex algebra with 1   as the vacuum vector where
Y ( f ( z ) , x ) g ( z ) = f ( z + x ) g ( z ) = ( e x d / d z f ( z ) ) g ( z ) (2.12)
for f ( z ) , g ( z ) C ( ( z ) )   . Then for any vertex algebra V   , from [FHL], C ( ( z ) ) V   is a vertex algebra where the vertex operator map Y ^   is given by
Y ^ ( f ( z ) u , x ) ( g ( z ) v ) = f ( z + x ) g ( z ) Y ( u , x ) v (2.13)
for f ( z ) , g ( z ) C ( ( z ) ) , u , v V   . Throughout this note, we always refer the vertex algebra C ( ( z C ( ( z ) ) V   to this specific vertex algebra structure.
In the following we shall use both maps Y   and Y ^   on C ( ( z ) ) V   . It is very important for us to distinguish them. For any vector w ( z ) C ( ( z ) ) V   , we have
Y ^ ( w , x ) = Y ( w ( z + x ) , x ) = e x z Y ( w ( z ) , x ) . (2.14)
Denote by D e r C ( ( z ) ) ( C ( ( z ) ) V )   the space of C ( ( z ) )   -linear derivations of vertex algebra C ( ( z ) ) V   .
Proposition 2.5. Let V   be any vertex algebra (over C   ). Every derivation of vertex algebra C ( ( z ) ) V   , restricted to V   , is a pseudoderivation of V   . On the other hand, the C ( ( z ) )   -linear extension of any pseudoderivation of V   is a derivation of C ( ( z ) ) V   . Furthermore, we have
D e r C ( ( z ) ) ( C ( ( z ) ) V ) = P D e r ( V ) . (2.15)
  • Proof. Let σ   be a derivation of the vertex algebra C ( ( z ) ) V   . By restricting (the domain of ) σ   to V   we have a C   -linear map a ( z )   from V   to C ( ( z ) ) V   . Using ( 2.14 ) we get
    a ( z ) Y ( u , x ) v Y ( u , x ) a ( z ) v = a ( z ) Y ^ ( u , x ) v Y ^ ( u , x ) a ( z ) v
    = Y ^ ( a ( z ) u , x ) v
    = Y ( a ( z + x ) u , x ) v .
    In view of Proposition  2.3 , a ( z )   is a pseudoderivation of V   .
    On the other hand, let a ( z )   be a pseudoderivation of V   . Consider a ( z )   as a C ( ( z ) )   -linear endomorphism of C ( ( z ) ) V   . For f ( z ) , g ( z ) C ( ( z ) ) , u , v V   , using ( 2.14 ) we have
    a ( z ) Y ^ ( f ( z ) u , x ) ( g ( z ) v )
    = a ( z ) ( f ( z + x ) g ( z ) Y ( u , x ) v )
    = f ( z + x ) g ( z ) a ( z ) Y ( u , x ) v
    = f ( z + x ) g ( z ) Y ( u , x ) a ( z ) v + f ( z + x ) g ( z ) Y ( a ( z + x ) u , x ) v
    = Y ( f ( z + x ) u , x ) g ( z ) a ( z ) v + Y ( f ( z + x ) a ( z + x ) u , x ) g ( z ) v
    = Y ^ ( f ( z ) u , x ) a ( z ) ( g ( z ) v ) + Y ^ ( a ( z ) f ( z ) u , x ) g ( z ) v .
    This proves that a ( z )   is a derivation of C ( ( z ) ) V   . The identification of the two spaces in ( 2.15 ) is clear.
Remark 2.6. For any f ( z ) C ( ( z ) ) , v V   , ( f ( z ) v ) 0   is a derivation of vertex algebra C ( ( z ) ) V   . We have
( f ( z ) v ) 0 = R e s x Y ^ ( f ( z ) v , x ) = R e s x f ( z + x ) Y ( v , x ) = X f , v . (2.16)
In view of Proposition  2.5 , X f , v   is a pseudoderivation of V   . This gives a (slightly) different proof of Proposition  2.4 .
Define a C   -linear map
X : C ( ( z ) ) V P D e r ( V ) , ( f , v ) X f , v . (2.17)
Denote by D ^   the D   -operator of C ( ( z ) ) V   . We have
D ^ = d / d z 1 + 1 D . (2.18)
Since ( D ^ ( f v ) ) 0 = 0   on C ( ( z ) ) V   for any f C ( ( z ) ) , v V   , we have D ^ ( C ( ( z ) ) V ) ker X   . Then we have a linear map, also denoted by X   , from ( C ( ( z ) ) V ) / D ^ ( C ( ( z ) ) V )   to P D e r ( V )   . Recall from [Bthat for any vertex algebra U   , U / D U   is a Lie algebra with
[ a + D U , b + D U ] = a 0 b + D U for a , b U . (2.19)
Now, taking U = C ( ( z ) ) V   , we have a Lie algebra ( C ( ( z ) ) V ) / D ^ ( C ( ( z ) ) V )   with
[ f ( z ) u , g ( z ) v ] = ( f ( z ) u ) 0 ( g ( z ) v ) = n 0 1 n ! f ( n ) ( z ) g ( z ) u n v (2.20)
for f ( z ) , g ( z ) C ( ( z ) ) , u , v V   . On the other hand, we have
[ X f , u , X g , v ] = m , n 0 1 m ! 1 n ! f ( m ) g ( n ) [ u m , v n ]
= m , n 0 i 0 ( m i ) 1 m ! 1 n ! f ( m ) g ( n ) ( u i v ) m + n i
= m , n 0 i 0 ( m + i i ) 1 ( m + i ) ! 1 n ! f ( m + i ) g ( n ) ( u i v ) m + n
= k 0 n = 0 k i 0 ( k n + i i ) 1 ( k n + i ) ! 1 n ! f ( k n + i ) g ( n ) ( u i v ) k
= k 0 i 0 1 k ! 1 i ! ( f ( i ) g ) ( k ) ( u i v ) k
= X [ f u , g v ] .
Thus X   is a Lie algebra homomorphism.
To summarize we have:
Proposition 2.7. For any vertex algebra V   , the linear map
X : ( C ( ( z ) ) V ) / D ^ ( C ( ( z ) ) V ) P D e r ( V ) ,
( f ( z ) , v ) X f , v (2.21)
is a Lie algebra homomorphism.
Next, we study notions of pseudoendomorphism and pseudoautomorphism.
Definition 2.8. A pseudoendomorphism of a vertex algebra V   is a C   -linear map A ( z )   from V   to C ( ( z ) ) V   such that
A ( z ) 1 = 1 ( = 1 1 ) , (2.22)
[ D , A ( z ) ] = d d z A ( z ) , (2.23)
A ( z ) Y ( v , x ) = Y ( A ( z + x ) v ) , x ) A ( z ) for v V . (2.24)
A pseudoautomorphism of a vertex algebra V   is an invertible pseudoendomorphism of V   . Clearly, all the pseudoautomorphisms of V   form a group, which we denote by P A u t ( V )   .
Remark 2.9. It is clear that a constant pseudoendomorphism of V   exactly amounts to an endomorphism of vertex algebra V   .
The following result, analogous to Proposition  2.3 , shows that condition ( 2.23 ) in Definition  2.8 is redundant:
Lemma 2.10. Let V   be a vertex algebra and let A ( z ) H o m ( V , C ( ( z ) ) V )   be such that ( 2.22 ) and ( 2.24 ) hold. Then A ( z )   is a pseudoendomorphism of V   .
  • Proof. Applying ( 2.24 ) to 1   and using the assumption that A ( z ) 1 = 1   we get
    A ( z ) Y ( v , x ) 1 = Y ( A ( z + x ) v , x ) 1 = e x z Y ( A ( z ) v , x ) 1 . (2.25)
    Just as in the second paragraph of the proof of Proposition  2.3 , we get [ A ( z ) , D ] v = d d z A ( z ) v ,   proving ( 3.2 ). Thus A ( z )   is a pseudoendomorphism of V   .
We also have the following analogue of Proposition  2.5 :
Proposition 2.11. Let V   be a vertex algebra. Every endomorphism of vertex algebra C ( ( z ) ) V   , restricted to V   , is a pseudoendomorphism of vertex algebra V   . On the other hand, the C ( ( z ) )   -linear extension of any pseudoendomorphism of vertex algebra V   is an endomorphism of vertex algebra C ( ( z ) ) V   . Furthermore, we have
P E n d ( V ) = E n d C ( ( z ) ) ( C ( ( z ) ) V ) , (2.26)
where E n d C ( ( z ) ) ( C ( ( z ) ) V )   stands for the space of all the C ( ( z ) )   -linear endomorphisms of vertex algebra C ( ( z ) ) V   .
  • Proof. Let σ   be an endomorphism of vertex algebra C ( ( z ) ) V   . Denote by A σ ( z )   the restriction of σ   as a C   -linear map from V   to C ( ( z ) ) V   . Then
    A σ ( z ) 1 = σ ( 1 1 ) = 1 1 = 1
    and for u , v V   , using ( 2.14 ) we get
    A σ ( z ) Y ( u , x ) v = σ Y ^ ( u , x ) v = Y ^ ( σ ( u ) , x ) σ ( v )
    = Y ^ ( A σ ( z ) u , x ) A σ ( z ) v
    = Y ( A σ ( z + x ) u , x ) A σ ( z ) v .
    In view of Lemma  2.10 , A σ ( z )   is a pseudoendomorphism of vertex algebra V   .
    On the other hand, let A ( z )   be a pseudoendomorphism of vertex algebra V   .
    Denote by A ¯ ( z )   the corresponding C ( ( z ) )   -linear endomorphism of C ( ( z ) ) V   . Then for f ( z ) C ( ( z ) ) , v V   , we have
    A ¯ ( z ) Y ^ ( f ( z ) u , x ) ( g ( z ) v )
    = A ¯ ( z ) ( f ( z + x ) g ( z ) Y ( u , x ) v )
    = f ( z + x ) g ( z ) A ( z ) Y ( u , x ) v
    = f ( z + x ) g ( z ) Y ( A ( z + x ) u , x ) A ( z ) v
    = Y ( f ( z + x ) A ( z + x ) u , x ) g ( z ) A ( z ) v
    = Y ( A ¯ ( z + x ) f ( z + x ) u , x ) g ( z ) A ( z ) v
    = Y ^ ( A ¯ ( z ) ( f ( z ) u ) , x ) A ¯ ( z ) ( g ( z ) v ) . (2.27)
    That is, A ¯ ( z )   is an endomorphism of vertex algebra C ( ( z ) ) V   .
Just as the exponential of a locally nilpotent derivation of a Lie algebra is an automorphism we have:
Proposition 2.12. Let V   be a vertex algebra and let a ( z )   be a pseudoderivation of V   such that [ a ( z ) , a ( x ) ] = 0   and such that for every v V   , there exists a positive integer k   such that a ( z ) k v = 0   . Then exp a ( z )   is a pseudoautomorphism of V   .
  • Proof. It is clear that under the assumptions, exp a ( z )   is a well defined linear map from V   to C ( ( z ) ) V   . As a ( z ) 1 = 0   , we have ( exp a ( z ) ) 1 = 1   . Since [ a ( z ) , Y ( v , x ) ] = Y ( a ( z + x ) v , x )   for v V   , it follows immediately that
    ( exp a ( z ) ) Y ( v , x ) exp ( a ( z ) ) = Y ( ( exp a ( z + x ) ) v , x ) . (2.28)
    By Lemma  2.10 , exp a ( z )   is a pseudoautomorphism of V   .
Example 2.13. Let V = L g ^ ( , 0 )   be the simple vertex operator algebra associated with an affine Kac-Moody Lie algebra g ^   and with a positive integer   . Then L g ^ ( , 0 )   is an integrable g ^   -module (see [K). Let u   be a root vector of g   . We have [ u ( x 1 ) , u ( x 2 ) ] = 0   and for every n Z   , u ( n )   acts locally nilpotently on L g ^ ( , 0 )   . Furthermore, for any w L g ^ ( , 0 )   we have u ( n ) w = 0   for n   sufficiently large. It follows that for any f ( z ) C ( ( z ) )   and for any w L g ^ ( , 0 )   , there exists a nonnegative integer k   such that X f , u k w = 0   . (Recall that X f , u = n 0 f ( n ) ( z ) u ( n )   .) By Proposition  2.12 , we have a pseudoautomorphism exp X f , u   . It will be interesting to study the subgroup of P A u t ( V )   , generated by all such exp X f , u   .

3 Simple current modules and pseudoautomorphisms

In this section we reveal a connection between pseudoautomorphisms and simple current modules for a vertex algebra.
Let us first recall the following results from [Li2:
Proposition 3.1. Let V   be a vertex algebra and let Δ ( z ) H o m ( V , C ( ( z ) ) V )   . If ( V , Y ( Δ ( x ) , x ) )   carries the structure of a V   -module, then
Δ ( z ) 1 = 1 ( 1 1 ) , (3.1)
[ D , Δ ( z ) ] = d d z Δ ( z ) , (3.2)
Δ ( z ) Y ( v , x ) = Y ( Δ ( z + x ) v , x ) Δ ( z ) for v V . (3.3)
On the other hand, if Δ ( z ) H o m ( V , C ( ( z ) ) V )   satisfies the above three conditions, then for any V   -module ( W , Y W )   , ( W , Y W ( Δ ( x ) , x ) )   carries the structure of a V   -module.
The following notion is due to [SY:
Definition 3.2. Let V   be a simple vertex operator algebra. An irreducible (simple) V   -module W   is called a simple current V   -module if the matrix of the left multiplication associated to the equivalence class of W   in the fusion algebra of V   is a permutation.
Proposition 3.3. Let V   be an irreducible vertex algebra and let Δ ( z ) H o m ( V , C ( ( z ) ) V )   such that the three conditions in Proposition  3.1 hold. Suppose that Δ ( z )   is invertible in the sense that there exists A ( z ) H o m ( V , C ( ( z ) ) V )   such that A ( z ) Δ ( z ) = Δ ( z ) A ( z ) = 1   . Then ( V , Y ( Δ ( x ) , x ) )   carries the structure of a simple current V   -module.
Remark 3.4. In [Li2, it was assumed that Δ ( z ) H o m ( V , V [ z , z 1 ] )   , however, all the proofs just work fine if Δ ( z ) H o m ( V , C ( ( z ) ) V )   .
The following result was obtained in [Li2(see also [Li1):
Proposition 3.5. Let V   be a vertex algebra and let h V   be such that
h 1 h C 1 and h n h = 0 for n 0 , n 1 (3.4)
and such that h 0   acts semisimply on V   with integral eigenvalues and h n   for n 1   act locally nilpotently on V   . Set
Δ ( h , z ) = z h ( 0 ) exp ( n 1 h ( n ) n ( z ) n ) . (3.5)
Then Δ ( h , z )   satisfies the conditions ( 3.1 )-( 3.3 ).
Now, in terms of our notions of pseudoendomorphism and pseudoautomorphism Propositions  3.1 and  3.3 can be reformulated as follow:
Proposition 3.6. Let V   be a vertex algebra and let Δ ( z ) H o m ( V , C ( ( z ) ) V )   . If ( V , Y ( Δ ( x ) , x ) )   carries the structure of a V   -module, then Δ ( z )   is a pseudoendomorphism of V   . On the other hand, if Δ ( z )   is a pseudoendomorphism of V   , then for any V   -module ( W , Y W )   , ( W , Y W ( Δ ( x ) , x ) )   carries the structure of a V   -module. Furthermore, if Δ ( z )   is invertible, ( V , Y ( Δ ( x ) , x ) )   is a simple current V   -module.
Next, we discuss certain connections between V   -modules and C ( ( z ) ) V   -modules.
Since V   is a vertex subalgebra of C ( ( z ) ) V   , any C ( ( z ) ) V   -module is naturally a V   -module. On the other hand, any V   -module can be naturally made a C ( ( z ) ) V   -module by the following result:
Proposition 3.7. Let ( W , Y W )   be a V   -module. Define a C   -linear map Y ^ W : C ( ( z ) ) V ( E n d W ) [ [ x , x 1 ] ]   by
Y ^ W ( f ( z ) v , x ) = f ( x ) Y W ( v , x ) (3.6)
for f ( z ) C ( ( z ) ) , v V   . Then ( W , Y ^ W )   carries the structure of a C ( ( z ) ) V   -module.
  • Proof. It is clear that Y ^ W ( f ( z ) v , x ) H o m ( W , W ( ( x ) ) )   for f ( z ) C ( ( z ) ) , v V   .
    We have Y ^ W ( 1 1 , x ) = Y W ( 1 , x ) = 1 W .   Furthermore, for f ( z ) , g ( z ) C ( ( z ) ) , u , v V   , we have
    Y ^ W ( f ( z ) u , x 1 ) = f ( x 1 ) Y W ( u , x 1 ) ,
    Y ^ W ( g ( z ) v , x 2 ) = g ( x 2 ) Y W ( v , x 2 ) ,
    and
    x 2 1 δ ( x 1 x 0 x 2 ) Y ^ W ( Y ^ ( f ( z ) u , x 0 ) ( g ( z ) v , x 2 )
    = x 2 1 δ ( x 1 x 0 x 2 ) Y ^ W ( f ( z + x 0 ) g ( z ) Y ( u , x 0 ) v , x 2 )
    = x 2 1 δ ( x 1 x 0 x 2 ) f ( x 2 + x 0 ) g ( x 2 ) Y W ( Y ( u , x 0 ) v , x 2 )
    = x 2 1 δ ( x 1 x 0 x 2 ) f ( x 1 ) g ( x 2 ) Y W ( Y ( u , x 0 ) v , x 2 ) .
    It follows immediately that the Jacobi identity holds. Thus ( W , Y ^ W )   carries the structure of a C ( ( z ) ) V   -module.
Remark 3.8. We here give a new proof for the assertion in Proposition  3.6 that if Δ ( z )   is a pseudoendomorphism of V   , for any V   -module ( W , Y W )   , ( W , Y W ( Δ ( x ) , x ) )   is a V   -module. In view of Proposition  2.11 , a pseudoendomorphism Δ ( z )   of V   is naturally an endomorphism of the tensor product vertex algebra C ( ( z ) ) V   . For any V   -module ( W , Y W )   , by Proposition  3.7 , we have a C ( ( z ) ) V   -module ( W , Y ^ W )   . By the simple fact, ( W , Y ^ W Δ ( z ) )   is also a C ( ( z ) ) V   -module. In particular, ( W , Y ^ W Δ ( z ) )   is a V   -module.
Furthermore, for u V   , we have
Y ^ W ( Δ ( z ) u , x ) = Y W ( Δ ( x ) u , x ) . (3.7)
This shows that ( W , Y W ( Δ ( x ) , x ) )   is a V   -module.
Remark 3.9. Recall from Proposition  3.5 the pseudoautomorphism Δ ( h , z )   . Since [ h ( m ) , h ( n ) ] = 0   for m , n 0   , we can write Δ ( h , z )   as
Δ ( h , z ) = exp ( h ( 0 ) log z + n 1 ( 1 ) n 1 h ( n ) n z n ) . (3.8)
Notice that if we could take f ( z ) = log z   in the formula X f , v = n 0 1 n ! f ( n ) ( z ) v n = R e s x e x ( d / d z ) f ( z ) Y ( v , x ) ,   we would have X log z , v = n 0 1 n ! ( d / d z ) n ( log z ) v n = v 0 log z + n 1 ( 1 ) n 1 1 n v n z n .   Now, X log z , v   is a C   -linear map from V   to C ( ( z ) ) [ log z ] V   . One can easily proves
[ X log z , v , Y ( u , x ) ] = Y ( e x ( d / d z ) X log z , v u , x ) for u V . (3.9)
This suggests us to study “logarithmic pseudoderivations” of V   , which are C   -linear maps a ( z )   from V   to C ( ( z ) ) [ log z ] V   satisfying the condition
[ a ( z ) , Y ( u , x ) ] = Y ( e x ( d / d z ) a ( z ) u , x ) for u V . (3.10)
One can also define a notion of logarithmic pseudoendomorphism of V   as a linear map A ( z )   from V   to C ( ( z ) ) [ log z ] V   such that
A ( z ) Y ( u , x ) v = Y ( e x ( d / d z ) A ( z ) u , x ) A ( z ) v for u , v V . (3.11)
We shall study this in a future publication.
References

  1. R. E. Borcherds, Vertex algebras, Kac-Moody algebras, and the Monster, Proc. Natl. Acad. Sci. USA 83 (1986), 3068-3071.
  2. C. Dong, H.-S. Li and G. Mason, Simple currents and extensions of vertex operator algebras, Commun. Math. Phys. 180 (1996), 671-707.
  3. P. Etingof and D. Kazhdan, Quantization of Lie bialgebras, V, Selecta Math. (N.S.) 6 (2000), 105-130.
  4. J. Fuchs, Simple WZW currents, Commun. Math. Phys. 136 (1991), 345-356.
  5. I. Frenkel, Y.-Z. Huang and J. Lepowsky, On axiomatic approaches to vertex operator algebras and modules, Memoirs Amer. Math. Soc. 104, 1993.
  6. I. Frenkel, J. Lepowsky and A. Meurman, Vertex Operator Algebras and the Monster, Pure and Appl. Math. Vol. 134, Academic Press, Boston, 1988.
  7. I. B. Frenkel and Y. Zhu, Vertex operator algebras associated to representations of affine and Virasoro algebras, Duke Math. J. 66 (1992), 123-168.
  8. V. G. Kac, Infinite-dimensional Lie Algebras, 3rd ed., Cambridge Univ. Press, Cambridge, 1990.
  9. H.-S. Li, Local systems of twisted vertex operators, vertex superalgebras and twisted modules, Contemp. Math. 193 (1995), 203-236.
  10. H.-S. Li, The physics superselection principle in vertex operator algebra theory, J. Alg. 196 (1997), 436-457.
  11. B.-H. Lian, On the classification of simple vertex operator algebras, Commun. Math. Phys. 163 (1994), 307-357.
  12. Schellekens, A. N. and Yankielowick, S., Modular invariants from simple currents. An explicit proof. Phys. Lett. B227 (1989), 387-391.

Department of Mathematics, Harbin Normal University, Harbin, China Current address : Department of Mathematical Sciences, Rutgers University, Camden, NJ 08102 E-mail address : hli@camden.rutgers.edu