Vertex Algebroids over Veronese Rings
We find a canonical quantization of Courant algebroids over Veronese rings. Part of our approach allows a semi-infinite cohomology interpretation, and the latter can be used to define sheaves of chiral differential operators on some homogeneous spaces including the space of pure spinors punctured at...
Збережено в:
| Дата: | 2008 |
|---|---|
| Автор: | |
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Інститут математики НАН України
2008
|
| Назва видання: | Symmetry, Integrability and Geometry: Methods and Applications |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/148079 |
| 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: | Vertex Algebroids over Veronese Rings / F. Malikov // Symmetry, Integrability and Geometry: Methods and Applications. — 2008. — Т. 4. — Бібліогр.: 25 назв. — англ. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-148079 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-1480792019-02-17T01:25:43Z Vertex Algebroids over Veronese Rings Malikov, F. We find a canonical quantization of Courant algebroids over Veronese rings. Part of our approach allows a semi-infinite cohomology interpretation, and the latter can be used to define sheaves of chiral differential operators on some homogeneous spaces including the space of pure spinors punctured at a point. 2008 Article Vertex Algebroids over Veronese Rings / F. Malikov // Symmetry, Integrability and Geometry: Methods and Applications. — 2008. — Т. 4. — Бібліогр.: 25 назв. — англ. 1815-0659 2000 Mathematics Subject Classification: 14Fxx, 81R10; 17B69 http://dspace.nbuv.gov.ua/handle/123456789/148079 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 |
We find a canonical quantization of Courant algebroids over Veronese rings. Part of our approach allows a semi-infinite cohomology interpretation, and the latter can be used to define sheaves of chiral differential operators on some homogeneous spaces including the space of pure spinors punctured at a point. |
| format |
Article |
| author |
Malikov, F. |
| spellingShingle |
Malikov, F. Vertex Algebroids over Veronese Rings Symmetry, Integrability and Geometry: Methods and Applications |
| author_facet |
Malikov, F. |
| author_sort |
Malikov, F. |
| title |
Vertex Algebroids over Veronese Rings |
| title_short |
Vertex Algebroids over Veronese Rings |
| title_full |
Vertex Algebroids over Veronese Rings |
| title_fullStr |
Vertex Algebroids over Veronese Rings |
| title_full_unstemmed |
Vertex Algebroids over Veronese Rings |
| title_sort |
vertex algebroids over veronese rings |
| publisher |
Інститут математики НАН України |
| publishDate |
2008 |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/148079 |
| citation_txt |
Vertex Algebroids over Veronese Rings / F. Malikov // Symmetry, Integrability and Geometry: Methods and Applications. — 2008. — Т. 4. — Бібліогр.: 25 назв. — англ. |
| series |
Symmetry, Integrability and Geometry: Methods and Applications |
| work_keys_str_mv |
AT malikovf vertexalgebroidsoververoneserings |
| first_indexed |
2025-07-12T18:10:42Z |
| last_indexed |
2025-07-12T18:10:42Z |
| _version_ |
1837465705644032000 |
| fulltext |
Symmetry, Integrability and Geometry: Methods and Applications SIGMA 4 (2008), 086, 28 pages
Vertex Algebroids over Veronese Rings?
Fyodor MALIKOV
Department of Mathematics, University of Southern California, Los Angeles, CA 90089, USA
E-mail: fmalikov@usc.edu
Received July 28, 2008, in final form December 07, 2008; Published online December 13, 2008
Original article is available at http://www.emis.de/journals/SIGMA/2008/086/
Abstract. We find a canonical quantization of Courant algebroids over Veronese rings.
Part of our approach allows a semi-infinite cohomology interpretation, and the latter can be
used to define sheaves of chiral differential operators on some homogeneous spaces including
the space of pure spinors punctured at a point.
Key words: differential graded algebra; vertex algebra; algebroid
2000 Mathematics Subject Classification: 14Fxx, 81R10; 17B69
1 Introduction
Attached to a commutative associative algebra A are the Lie algebra of its derivations, Der(A),
and the module of Kähler differentials, Ω(A). The identities that are satisfied by the clas-
sic differential geometry operations, such as the Lie bracket, the Lie derivative, the de Rham
differential, etc., can be summarized by saying that the A-module Der(A)⊕Ω(A) is a Courant al-
gebroid, [8, 22]. For reasons that will become apparent later, we will use the notation Vpoiss(A) =
Vpoiss(A)0 ⊕ Vpoiss(A)1, where Vpoiss(A)0 = A and Vpoiss(A)1 = Der(A)⊕ Ω(A).
This example can be enriched in two different ways. First, it can be quantized. Attached
to A in [15] is the notion of a vertex algebroid, V(A). This notion is a result of axiomatizing the
structure that is induced on conformal weight 0 and 1 components of a graded vertex algebra.
One has V(A) = A⊕ V(A)1 for some V(A)1, which fits in the exact sequence
0→ Ω(A)→ V(A)1 → Der(A)→ 0.
Therefore, V(A)1 is filtered and the corresponding graded object is GrV(A)1 = Vpoiss(A)1. This
strongly resembles the Poincaré–Birkhoff–Witt filtration, and it is indeed true that the notion
of a Courant algebroid is a quasiclassical limit of that of a vertex algebroid; this important
observation is due to Bressler [7], but the fact that the Courant bracket belongs in the infinite
dimensional world had been discovered by Dorfman much earlier, [9]. The relation of this notion
to various string theory models has been elucidated in [23].
Unlike its quasiclassical counterpart, a vertex algebroid may not exist, and if it exists, it
may not be unique. If Spec(A) is smooth, then [15] the obstruction to existence is the class
ch2(Spec(A)), and if ch2(Spec(A)) = 0, then the isomorphism classes are parameterized by the
hypercohomology group H1(Spec(A),Ω2
A
dDR→ Ω3,cl
A ). But what if Spec(A) is not smooth?
This question suggests the second way to enrich, which is to note that Der(A) and Ω(A) tell
the whole story only if Spec(A) is smooth; if not, then the higher algebras of derivations and
modules of 1-forms must arise. This was made precise by Hinich, [18], who defined Der(A)• as
a kind of a derived functor of the functor Der by applying the latter to a polynomial differential
?This paper is a contribution to the Special Issue on Kac–Moody Algebras and Applications. The full collection
is available at http://www.emis.de/journals/SIGMA/Kac-Moody algebras.html
mailto:fmalikov@usc.edu
http://www.emis.de/journals/SIGMA/2008/086/
http://www.emis.de/journals/SIGMA/Kac-Moody_algebras.html
2 F. Malikov
graded algebra resolution R → A. This gives rise to a graded Courant algebroid functor A 7→
Vpoiss(A)•.
The problem of combining the two, that is to say, finding a quantization, V(A)•, of Vpoiss(A)•
appears to be intellectually attractive and important for applications. If A is a complete intersec-
tion, then the resolving algebra R can be chosen to be a super-polynomial ring on finitely many
generators, the corresponding resolution R → A being none other than the standard Koszul
complex, and the quantization, a differential graded vertex algebroid V(R), is immediate; this
observation has been used in a number of physics and mathematics papers. If, however, A is
not a complete intersection, then any resolving algebra R is infinitely generated in which case
defining a vertex algebroid V(R) becomes problematic because of various divergencies. A re-
gularization procedure for some of these divergencies was suggested in [4] and elaborated on
in [17].
Here is what we do in the present paper. Let VN be the (N + 1)-dimensional irreducible sl2-
module, ON ⊂ P(VN ) the highest weight vector orbit, and AN the corresponding homogeneous
coordinate ring. All of this is a representation theorist’s way of saying that P1 ∼−→ ON ⊂ PN is
a Veronese curve, and AN is a Veronese ring.
AN is a quadratic algebra, in fact it is Koszul [3, 19, 6], but it is not a complete intersection.
The main result of the paper, Theorem 5.1.1, asserts that Vpoiss(AN )• admits a unique quan-
tization. It is no surprise that this quantization, V(AN ), contains a vertex algebroid, V(sl2)k,
attached to ŝl2 with some central charge k. What is more important is that the vertex algebroid
attached to ĝl2 enters the fray. The latter, V(gl2)k1k2 , depends in general on two central charges,
k1, k2, and we find that the quantization conditions imply, first, that k1 +k2 = −2 and, second,
that k1 = −N − 2.
Theorem 5.1.1 and its proof appear in Section 5, and it is for the sake of this section that the
paper was written. Section 4 is to a large extent an exposition of Hinich’s result (see also [2])
with some extensions (Sections 4.3, 4.4, 4.5) that are needed in Section 5. Sections 2 and 3 are
an attempt, perhaps futile, to make the paper self-contained – except Sections 3.7.3, 3.7.4, where
sheaves of vertex algebroids over C2\0 are classified. The classification obtained is instrumental
in proving Theorem 5.1.1; in particular, the vertex algebroid V(gl2)k1,k2 with the compatibility
condition k1 + k2 = −2 makes appearance in Section 3.7.4.
An obvious generalization of AN is provided by the homogeneous coordinate ring of a higher
dimensional Veronese embedding P(Cn)→ P(SN (Cn)). We show (Theorem 5.3.1) that if n > 2
and N > 1, then no quantization exists.
Much of the above carries over to an arbitrary simple g, where AN is replaced with the
homogeneous coordinate ring of the highest weigh vector orbit in the projectivization of a simple
g-module. For example, C2 \0 becomes the Bernstein–Gelfand–Gelfand base affine space, G/N .
Constructed in [16] is the 1-parameter family of sheaves of vertex algebroids H∞/2(Ln,VG,k)
over G/N , where VG,k, k ∈ C, is a family of vertex algebroids over G, [1, 10, 14, 16]. There
is little doubt that the family H∞/2(Ln,VG,k), k ∈ C, is universal in that it classifies vertex
algebroids over G/N equipped with V(g)k-structure. This is a higher rank analogue of the
classification obtained in Section 3.7.3 and alluded to above. Note that just as G/N is a G×T -
space, the maximal torus acting on the right, so there is a diagram of embeddings
V(g)k1 ↪→ H∞/2(Ln,VG,k)←↩ V(t)k2 with k1 + k2 = −ȟ .
Therefore, V(g)k⊕V(t)−k−ȟ is a higher rank analogue of V(gl2)k,−k−2; here g, n and t are the Lie
algebras of the Lie groups G, N , and T resp. We elaborate on these remarks in Section 6, where
we use the technique of semi-infinite cohomology to compute CDO-s on some homogeneous
spaces including the spaces of pure spinors punctured at a point. In the latter case, this gives
an approach alternative to that of the original result by Nekrasov [24].
Vertex Algebroids over Veronese Rings 3
Some aspects of the sl2-case, however, are not that easy to generalize. As they say, we hope
to return to this subject in a separate paper.
We would like to conclude by saying that a major source of inspiration was provided to us
by the work of Berkovits and Nekrasov [5, 4], where similar problems are analyzed in the case
of the spinor representation of the spinor group.
2 Vertex algebras
2.1. Conventions. Underlying all the constructions in this paper will be the category of Z-
graded vector superspaces and grading preserving linear maps over C. This grading will be called
(and should be thought of as) the homological degree grading. More often, though, the attribute
‘graded’ will be skipped. Thus the phrase ‘let V be a vector space’ will mean that V = ⊕n∈ZV n,
V even = ⊕n∈ZV 2n, V odd = ⊕n∈ZV 2n+1. Likewise, the prefix ‘super-’ will be usually omitted so
that commutative will mean super-commutative, algebra super-algebra, bracket super-bracket:
[a, b] = ab− (−1)abba.
If V and W are vector spaces, then V ⊗W is also a vector space with homological degree
grading defined in the standard way so that (V ⊗ W )n = ⊕i∈ZV i ⊗ Wn−i. Various bilinear
operations (‘multiplications’) to be used below will be morphisms of graded vector spaces V ⊗
W → U .
Along with the homological degree grading, the grading by conformal weight will play a promi-
nent role. The latter will be indicated by a subindex; thus, for example, the phrase ‘a graded
(by conformal weight) vertex algebra’ will mean, in particular, a vector space V with a direct
sum decomposition V = ⊕n≥0Vn valid in the category of graded vector spaces.
Most of the definitions and constructions in this and the following section are well known, and
their graded versions are always straightforward. We recommend [20] and [12] as an excellent
introduction to vertex things and a guide to further reading.
Definition 2.2. A vertex algebra is a collection (V,1, T,(n) , n ∈ Z), where V is a vector space,
1 ∈ V is a distinguished element known as the vacuum vector, T : V → V is a linear operator
known as the translation operator, each (n) is a multiplication
(n) : V ⊗ V → V, a⊗ b 7→ a(n)b s.t. a(n)b = 0 if n� 0,
that is subject to the following axioms:
(1) (vacuum)
1(n) =
{
IdV if n = −1,
0 otherwise,
a(−1)1 = a, ∀ a ∈ V ; (2.1)
(2) (translation invariance)
[T, a(n)]b = (Ta)(n)b = −na(n−1)b, ∀ a, b ∈ V, n ∈ Z; (2.2)
(3) (skew-symmetry)
a(n)b = (−1)ab
∑
j≥0
(−1)n+1+j 1
j!
T j(b(n+j)a), ∀ a, b ∈ V, n ∈ Z; (2.3)
(4) (Jacobi identity)
[a(m), b(n)]c =
∑
j≥0
(
m
j
)
(a(j)b)(m+n−j)c, ∀ a, b, c ∈ V, m, n ∈ Z; (2.4)
4 F. Malikov
(5) (quasi-associativity or normal ordering)
(a(−1)b)(n)c =
∑
j≥0
a(−1−j)b(n+j)c + (−1)ab
∑
j>0
b(n−j)a(−1+j)c, ∀ a, b, c ∈ V, n ∈ Z. (2.5)
The collection of axioms we used in Definition 2.2 is a little redundant but makes the ex-
position a little more transparent. It emphasizes the fact that the notion of a vertex algebra
is a mixture of (appropriate analogues of) that of an associative algebra and a Lie algebra.
Extracting the Lie part of the definition one arrives at the notion of a vertex Lie algebra.
Definition 2.3. A vertex Lie algebra is a collection (V, T, (n), n ∈ Z+), where V is a vector
space, T : V → V is a linear operator, each (n) is a multiplication
(n) : V ⊗ V → V, a⊗ b 7→ a(n)b s.t. a(n)b = 0 if n� 0 (2.6)
that is subject to the following axioms:
(1) translation invariance, that is, (2.2) for n ≥ 0;
(2) skew-symmetry, that is, (2.3) for n ≥ 0;
(3) Jacobi identity, that is, (2.4) for n ≥ 0.
There is an obvious forgetful functor
Φ : {Vertex algebras} → {Vertex Lie algebras}. (2.7)
Its left adjoint functor (the vertex enveloping algebra functor)
U : {Vertex Lie algebras} → {Vertex algebras} (2.8)
is well known to exist, see [25]; it also appears in [20] as ‘the vertex algebra attached to a formal
distribution Lie superalgebra’.
Note a canonical map
ι : L → Φ(UL) (2.9)
that is the image of Id ∈ Hom(UL,UL) under the identification Hom(UL,UL) ∼−→Hom(L,ΦUL).
Example 2.4. Let L′ be a free C[T ]-module on one generator L and let L(Vir)c = L′ ⊕ C,
where C is considered as a trivial C[T ]-module. L(Vir)c carries a unique vertex Lie algebra
structure such that
L(0)L = T (L), L(1)L = 2L, L(2)L = 0, L(3)L = 1
2c, L(n)L = 0 if n > 3.
Upon quotienting out by the relation 1 = 1, the vertex enveloping algebra UL(Vir)c becomes
the vacuum representation of the Virasoro algebra of central charge c.
Example 2.5. Let g be a Lie algebra with an invariant bilinear form (·, ·). Let
L(g)k = C[T ]⊗ g⊕ C.
This space carries an obvious action of T , where again we consider C as a trivial C[T ]-module,
and a unique vertex Lie algebra structure such that
(1⊗ a)(0)(1⊗ b) = 1⊗ [a, b], (1⊗ a)(1)(1⊗ b) = k(a, b). (2.10)
Vertex Algebroids over Veronese Rings 5
Upon quotienting out by the relation 1 = 1, the vertex enveloping algebra UL(g)k becomes the
vacuum representation of the corresponding affine Lie algebra of central charge k.
If g is chosen to be glN = slN ⊕C · I, then this construction has the following version: we let
(a, b) = tr(a · b), L(glN )k1,k2 = L(slN )k1 ⊕ C[t]⊗ C · I and extend (2.10) by
(1⊗ I)(1)(1⊗ I) = k2N, (1⊗ I)(0)(1⊗ I) = (1⊗ I)(n)(1⊗ slN ) = 0 ∀ n. (2.11)
In order to handle the case of the trivial bilinear form (·, ·), or more generally the case where (·, ·)
is not unique even up to proportionality, we will change the notation and denote by L(g)(·,·) the
vertex Lie algebra which is precisely L(g)k except that the last of conditions (2.10) is replaced
with
(1⊗ a)(1)(1⊗ b) = (a, b) (2.12)
for some (·, ·).
A passage to the quasiclassical limit is a gentler way to blend the Lie and commutative/as-
sociative algebra parts of the structure.
Definition 2.6. A vertex Poisson algebra is a collection (V,1 ∈ V, T, (n), n ≥ −1), where V is
a vector space, T : V → V is a linear operator, each (n) is a multiplication
(n) : V ⊗ V → V, a⊗ b 7→ a(n)b s.t. a(n)b = 0 if n� 0
that is subject to the following axioms:
(1) the triple (V,1, T, (−1)) is a unital commutative associative algebra with derivation;
(2) the collection (V,1 ∈ V, T, (n), n ≥ 0) is a vertex Lie algebra;
(3) each multiplication (n), n ≥ 0, is a derivation of (−1).
Vertex Poisson algebras are to vertex algebras what Poisson algebras are to noncommutative
algebras. The following construction (cf. [21]) is meant to illustrate this point.
2.7. Vertex algebras with filtration. Suppose a vertex algebra V carries an exhaustive
increasing filtration by vector spaces
C1 ∈ F 0V ⊂ F 1V ⊂ · · · ⊂ F pV ⊂ · · · , ∪p≥0F
nV = V (2.13)
which satisfies (F pV )(n)F
qV ⊂ F p+qV and T (F pV ) = F pV for all p, q ∈ Z so that (F pV )(n)F
qV
⊂ F p+q−1V if n ≥ 0. By focusing on symbols one discovers that the vertex algebra structure
on V defines the following on the corresponding graded object Gr V = ⊕pF
pV/F p−1V :
1gr = 1 ∈ F 0V ;
T gr : F pV/F p−1V → F pV/F p−1V, T (ā) = T (a) mod F p−1V ;
(−1)gr : (F pV/F p−1V )⊗ (F qV/F q−1V )→ F p+qV/F p+q−1V,
ā⊗ b̄ 7→ ā(−1)gr b̄ = a(−1)b mod F p+q−1V ;
(n)gr : (F pV/F p−1V )⊗ (F qV/F q−1V )→ F p+q−1V/F p+q−2V,
ā⊗ b̄ 7→ ā(n)gr b̄ = a(n)b mod F p+q−2V if n ≥ 0.
It is then immediate to check that (GrV,1gr, T gr, (n)gr , n ≥ −1) is a vertex Poisson algebra.
For example, commutativity of the product (−1) follows from the n = −1 case of (2.3) and
associativity from the n = −1 case of (2.5).
The vertex algebras reviewed in Examples 2.4, 2.5 possess a filtration with the indicated
properties – as does any vertex enveloping algebra:
6 F. Malikov
• in the case of UL(Vir)c the filtration is determined by assigning degree one to ι(L(Vir)c),
see (2.9);
• in the case of UL(g)k the filtration is determined by assigning degree one to ι(L(g)k).
Denote thus defined vertex Poisson algebras as follows:
UpoissL(Vir) = GrUL(Vir)c, UpoissL(g) = GrUL(g)k,
UpoissL(glN ) = GrUL(glN )k,k2 . (2.14)
3 Courant and vertex algebroids
Definition 3.1. A vertex (vertex Poisson) algebra V is called graded if
V =
+∞⊕
n=−∞
Vn so that
Vn = 0, if n < 0, 1 ∈ V0, T (Vm) ⊂ Vm+1, and (Vm)(j)Vn ⊂ Vm+n−j−1. (3.1)
Such grading is usually referred to as conformal, Vn is called the conformal weight n compo-
nent, and v ∈ Vn is said to have conformal weight n; the conformal weight of v ∈ Vn is usually
denoted by ∆(v).
All the vertex algebras we have seen are graded: UL(Vir)c is graded by letting V0 = C,
V1 = {0}, V2 = ι(L), UL(g)k by letting V0 = C, V1 = ι(g), UL(glN )k1,k2 by letting V1 = ι(glN )
A graded vertex (vertex Poisson) algebra structure on V induces the following structure on
the subspace V0 ⊕ V1:
1 ∈ V0, (3.2)
T : V0 → V1, (3.3)
(n) : Vi ⊗ Vj → V(i+j−n−1), n ≥ 0, i, j = 0, 1, (3.4)
(−1) : (V0 ⊗ Vi)⊕ (Vi ⊗ V0)→ Vi, i = 0, 1. (3.5)
These data satisfy a list of identities obtained by inspecting those listed in Definitions 2.2
and 2.6 and choosing the ones that make sense. The meaning of ‘make sense’ is clear in the
case of identities, such as the Jacobi, involving only operations (n) with n ≥ 0, because V0 ⊕ V1
is closed under these operations. Expressions T j(a(n)b), a(i)b(j)c, (a(i)b)(j)c, a, b, c ∈ V0, V1,
are said to make sense if either they are compositions of operations (3.2)–(3.5) or because
∆(a) + ∆(b)−n− 1 < 0, ∆(b) + ∆(c)− j − 1 < 0, ∆(a) + ∆(b)− i− 1 < 0 (resp.) in which case
the expressions are defined to be 0, cf. condition (3.1). Finally, we shall say that an identity
makes sense if all the expressions that it involves make sense.
Definition 3.2. A Courant algebroid is a vector space V0 ⊕ V1 carrying the data (3.2)–(3.5)
so that all those axioms of Definition 2.6 that make sense are valid.
Definition 3.3. A vertex algebroid is a vector space V0 ⊕ V1 carrying the data (3.2)–(3.5) so
that all those axioms of Definition 2.2 that make sense are valid.
There are two obvious forgetful functors
Φalg : {Vertex algebras} → {Vertex algebroids}, (3.6)
Φpoiss
alg : {Vertex Poisson algebras} → {Courant algebroids}, (3.7)
and both afford the left adjoints Ualg and Upoiss
alg , which are analogous to (2.8).
Vertex Algebroids over Veronese Rings 7
When applied to the algebras of (2.14) and of Examples 2.4, 2.5, they give us first examples
of Courant
Vpoiss(Vir) = C, Vpoiss(g) = C⊕ g, Vpoiss(glN ) = C⊕ glN (3.8)
and vertex algebroids resp.
V(Vir)c = C, V(g)k = C⊕ g, V(glN )k1,k2 = C⊕ glN . (3.9)
Here is an example of geometric nature.
Example 3.4. Let A be a commutative associative algebra with unit 1, Ω(A) the module
of Kähler differentials, Der(A) the algebra of derivations. Note that the homological degree
grading of A determines that of Ω(A) and Der(A). Set V0 = A, V1 = Der(A) ⊕ Ω(A), and let
T
def= d : V0 = A → ΩA ↪→ V1 be the canonical (de Rham) derivation. Then the space V0 ⊕ V1
carries a unique Courant algebroid structure determined by
a(−1)b = ab, a(−1)(τ + ω) = aτ + aω, τ(0)a = τ(a),
τ(0)ω = Lieτω, τ(1)ω = ιτω, (3.10)
τ(0)ξ = [τ, ξ], τ(1)ξ = 0,
for all a, b ∈ A, ω ∈ Ω(A), τ, ξ ∈ Der(A). Denote thus defined Courant algebroid by Vpoiss(A).
Since all the operations recorded in (3.10) are of geometric nature, there arises, for each
scheme X, a sheaf of vertex algebroids
Vpoiss
X
def= TX ⊕ ΩX . (3.11)
Note that in keeping with our convention we assume that X is graded, that is to say, OX is
a sheaf of graded algebras; consequently, TX and ΩX are sheaves of graded OX -modules.
If V = V0 ⊕ V1 is a vertex algebroid, then part of its structure coincides with that of
Vpoiss(A). For example, the triple (V0,(−1) ,1) is a commutative associative algebra with unit,
(V0)(−1)(T (V0)) is a V0-module (although V1 is not), the map T : V0 → (V0)(−1)(T (V0)) is
a derivation. There arises a filtration
V0 ⊕ (V0)(−1)(T (V0)) ⊂ V, (3.12)
and a moment’s thought will show that, absolutely analogously to Section 2.7, the corresponding
GrV carries a canonical Courant algebroid structure.
Definition 3.5.
(a) Given a Courant algebroid Vpoiss, call a vertex algebroid V a quantization of Vpoiss if there
is an isomorphism of Courant algebroids
GrV ∼−→ Vpoiss.
(b) If V is a quantization of Vpoiss(A), see Example 3.4, then we shall denote V by V(A).
Note that in the case of Vpoiss(A), filtration (3.12) implies the following exact sequence of
vector spaces
0→ Ω(A)→ V1
π→ Der(A)→ 0. (3.13)
8 F. Malikov
The problem of quantizing Vpoiss(A) is not trivial and was studied in [15]. Call A suitable
for chiralization if Der(A) is a free A-module on generators τ1, . . . , τN s.t. [τi, τj ] = 0. If A
is suitable for chiralization and a basis τ1, . . . , τN is fixed, then Vpoiss(A) can be quantized by
letting V(A) = Vpoiss(A) as a vector space and requiring all of the relations (3.10) except the
last two; the latter are to hold only for the basis vector fields:
(τi)(0)(τj) = 0, (τi)(1)(τj) = 0. (3.14)
It is easy to show, using axioms (2.1)–(2.5), at least that these choices determine a vertex
algebroid structure.
Let us point out the differences between the Courant and vertex algebroid structures thus
obtained: in the Courant case operation (1) is an A-bilinear pairing and
(f(−1)τi)(1)(g(−1)τj) = 0,
while in the vertex case
(f(−1)τi)(1)(g(−1)τj) = −f(−1)(τj(τi(g))− g(−1)(τi(τj(f))− τi(g)(−1)τj(f); (3.15)
in the Courant case multiplication (−1) is associative, e.g.,
(fg)(−1)τi = f(−1)(g(−1)τi),
in the vertex case it is not as
(fg)(−1)τi = f(−1)(g(−1)τi)− τi(g)df − τi(f)dg; (3.16)
here f, g ∈ A and axioms (2.3)–(2.5) along with simplifications due to grading have been used.
Furthermore, if A is suitable for chiralization, then
{isomorphism classes of quantizations of Vpoiss(A)}
is an
(
Ω3,cl(A)/dΩ2(A)
)0 − torsor, (3.17)
and
Aut(V(A)) ∼−→
(
Ω2,cl(A)
)0
, (3.18)
where the automorphism of V(A) attached to ω ∈ Ω2,cl(A)0 is defined by the assignment
V(A)1 → V(A)1, τ 7→ τ + ω(π(τ)), (3.19)
π being defined in (3.13). Note that we had to pay the price for relentlessly working in the
graded setting by extracting the homological degree 0 subspace in (3.17), (3.18).
Since any smooth algebraic variety X can be covered by the spectra of rings suitable for
chiralization, (3.17), (3.18) create an avenue to define sheaves of vertex algebroids over X, to
be denoted VX or VA if X = Spec(A).
Definition 3.6. Let X be a graded scheme.
(a) Call a sheaf of vertex algebroids over X a quantization of Vpoiss
X , see (3.11), if for each
affine U ⊂ X its space space of sections over U is a quantization of Γ(U,Vpoiss
X ).
(b) Denote by VertX (or VertA if X = Spec(A)) the category of quantizations of the sheaf
Vpoiss
X .
Vertex Algebroids over Veronese Rings 9
The characteristic property of VX ∈ VertX is the existence of the sequence of sheaves
0→ ΩX → VX
π→ TX → 0; (3.20)
this is a sheaf analogue of (3.13).
Here are the (obvious graded versions of the) main results of [15]:
• there is a gerbe of vertex algebroids over a manifold X such that the space of sections over
each U = Spec(A) ⊂ X, A being suitable for chiralization, is the category of quantizations
of Vpoiss(A);
• in the case where OX = O0
X this gerbe possesses a global section, i.e., a sheaf VX ∈ VertX ,
if and only if the 2nd component of the Chern character
ch2(TX) ∈ H2
(
X, Ω2
X → Ω3,cl
X
)
(3.21)
vanishes; if this class vanishes, then
{Isomorphism classes of sheaves VX} is an H1
(
X, Ω2
X → Ω3,cl
X
)
− torsor; (3.22)
• the forgetful functor (3.6) has a left adjoint functor
Ualg : {Vertex algebroids} → {Vertex algebras}; (3.23)
if the obstruction (3.21) vanishes, we call
Dch
X
def= UalgVX (3.24)
a sheaf of chiral differential operators, CDO for short.
Note that proving (3.22) amounts to covering X by open sets that are suitable for chiralization
and re-gluing a given sheaf by composing the old gluing functions with automorphisms (3.18),
(3.19).
3.7. Further examples and constructions.
3.7.1. Localization. For any quantization V(A) and an ideal a ⊂ A a natural quantiza-
tion V(Aa) is defined [15], the reason being that all the (n)-products on V(A) are in fact cer-
tain differential operators. For example, at the quasiclassical level, all the operations recorded
in (3.10) are differential operators of order ≤ 1. Furthermore, (3.15), (3.16) provide examples
of genuine quantum operations being order ≤ 2 differential operators. Therefore, given an A
and V(A), there arises a sheaf of vertex algebroids
VA ∈ VertA s.t. Γ(Spec(A),VA) = V(A). (3.25)
This construction underlies the above discussion of gerbes of vertex algebroids.
A little more generally, if X and Y are manifolds and p : X → Y is a covering, then there is
a functor
p∗ : VertY → VertX .
The reason for this to be true is that the story of quantizing Vpoiss(A), Spec(A) ⊂ Y , starts
with a choice of an Abelian basis {τi} ⊂ Der(A), and any such choice is canonically lifted to an
Abelian basis of Γ(V, TX) for any affine V ⊂ p−1(Spec(A)).
10 F. Malikov
In particular, if X carries a free action of a finite group G, then there arises an equivalence
of categories
VertX/G → VertG
X , (3.26)
where VertG
X is a full subcategory of G-equivariant vertex algebroids. The inverse functor is,
essentially, that of G-invariants
VertG
X → VertX/G, V 7→ p∗(VG), (3.27)
where p∗ is the push-forward in the category of sheaves of vector spaces.
Here is a version of localization called a push-out in [15, 16]. If a Lie group acts on A by
derivations, then
A⊗ V(g)k
def= A⊕A⊗ g⊕ Ω(A) (3.28)
is a vertex algebroid.
3.7.2. Affine space. According to (3.21), (3.22), there is a unique sheaf of vertex algebroids
over CN , VCN . Its space of global sections is V(C[x1, . . . , xN ]), where we take {∂j = ∂/∂xj} for
an Abelian basis of Der(C[x1, . . . , xN ]), and if U = {f 6= 0}, then Γ(U,VCN
) is defined via the
localization of Section 3.7.1.
The corresponding CDO, see (3.24), is Dch
CN = UalgVCN . Its space of global sections, de-
noted by either Dch(CN ) or Dch(C[x1, . . . , xN ]), is likewise obtained via Ualg: Dch(CN ) =
UalgV(C[x1, . . . , xN ]). More explicitly, Dch(CN ) can be defined to be the vertex algebra genera-
ted freely by the vector space CN ⊕ (CN )∗ and relations
(∂i)(0)xj = −(xj)(0)∂i = δij1, a(n)b = 0 for all a, b ∈ CN ⊕ (CN )∗, n > 0,
where {∂i}, {xj} are dual bases of CN and (CN )∗ resp.
3.7.3. Punctured plane. Let X = C2\(0, 0) and choose the trivial grading where OX = O0
X .
The obstruction (3.21) vanishes, because one sheaf, say, the restriction of VC2 to C2\(0, 0), exists.
Isomorphism classes of sheaves of vertex algebroids over X are easy to classify. Indeed, consider
the affine covering
X = U1 ∪ U2, Uj = {(y1, y2) s.t. yj 6= 0}
and the Cech 1-cocycle
ωab : U1 ∩ U2 7→
dy1 ∧ dy2
ya
1yb
2
, a, b ≥ 1.
It is known (and easy to check) that
H1
(
X, Ω2
X → Ω3,cl
X
)
= ⊕a,bCωab.
Hence the isomorphism classes of sheaves of vertex algebroids over X are in 1-1 correspondence
with linear combinations of ωab. Here is an explicit construction of the sheaf attached to kωab:
let VX be the restriction of the standard VC2 to X, VUj its pull-back to Uj , j = 1, 2; now glue VU1
and VU2 over the intersection U1 ∩ U2, cf. (3.18), (3.19), as follows:
∂1 → ∂1 +
kT (y2)
ya
1yb
2
, ∂2 → ∂2 −
kT (y1)
ya
1yb
2
. (3.29)
Vertex Algebroids over Veronese Rings 11
It is immediate to generalize this to the case of an arbitrary ω in the linear span of {ωab}. Denote
the sheaf sheaf thus defined by VX,ω.
This simple example will be essential for our purposes.
3.7.4. Symmetries. Given a Lie algebra morphism
ρ : g→ Der(A),
the composition
g→ Der(A)
(id,0)
↪→ Vpoiss(A) = Der(A)⊕ Ω(A)
defines a Courant algebroid morphism
ρpoiss : Vpoiss(g)→ Vpoiss(A).
Furthermore, any Poisson algebroid morphism ρpoiss, upon quotienting out by Ω(A), defines
a Lie algebra morphism ρ.
A quantization of a Courant algebroid morphism ρpoiss : Vpoiss(g) → Γ(X,Vpoiss
X ) is defined
to be a vertex algebroid morphism ρ̂ : V(g)k → V(A) such that the diagram
0 // Ω(A) // V(A) // Der(A) // 0
V(g)k
ρ //
ρ̂
OO
Der(A) // 0
(3.30)
commutes. Here the arrow V(g)k
ρ→ Der(A) means the composition V(g)k = C⊕ g
0⊕ρ→ Der(A),
see (3.9).
Similarly, if g operates on X, that is, there is a Lie algebra morphism
ρ : g→ Γ(X, TX)
then a quantization of the corresponding Courant algebroid morphism ρpoiss : Vpoiss(g) →
Γ(X,Vpoiss
X ) is defined to be a vertex algebroid morphism ρ̂ : V(g)k → Γ(X,VX) such that
the diagram
0 // Γ(X, ΩX) // Γ(X,VX) // Γ(X, TX) // 0
V(g)k
ρ //
ρ̂
OO
Γ(X, TX) // 0
(3.31)
commutes.
To see an example of importance for what follows, let us consider the tautological action
of gl2 on C2. If we let X = C2 \ (0, 0), then there arises
ρpoiss : Vpoiss(gl2)→ Γ
(
C2 \ (0, 0),Vpoiss
C2\(0,0)
)
, (3.32)
and we ask if this map can be quantized to a map V(gl2)k1,k2 → Γ(C2 \ (0, 0),VC2\(0,0),ω), where
the vertex algebroid VC2\(0,0),ω was defined in Section 3.7.3.
12 F. Malikov
Lemma 3.7.5. Quantization of (3.32),
ρ̂ : V(gl2)k1,k2 → Γ
(
C2 \ (0, 0),VC2\(0,0),ω
)
exists if and only if ω = kdy1 ∧ dy2/y1y2, k1 = −k − 1, k2 = k − 1 for some k ∈ C.
Proof. We shall use the notation of Section 3.7.3. In terms of the coordinates y1, y2 the
morphism ρ is this
ρ(Eij) 7→ yi∂j . (3.33)
If we consider yi∂j as an element of Γ(U1,VX,ω), then over U2 it becomes, according to (3.29),
(yi∂j ± kT (yj±1)yi)/ya
1yb
2 and hence may develop a pole. To compensate for it, we can choose
a different lift of Eij to Γ(U1,VX,ω) by replacing yi∂j with yi∂j + α, where α ∈ Γ(U1,ΩX).
Over U2 this element becomes (yi∂j ± T (yj±1)yi)/ya
1yb
2 + α. Since α may have at most a pole
along {y1 = 0}, for this element to extend to a section over U2, one of the following two things
must happen: either T (yj±1)yi/ya
1yb
2 has no pole along {y1 = 0}, in which case no α is needed,
or T (yj±1)yi/ya
1yb
2 has no pole along {y2 = 0}, in which case a desired α can be found. For
a favorable event to occur for i = 1 and i = 2, both a and b must be at most 1. But by
definition, see Section 3.7.3, a and b are at least 1; therefore a linear map gl2 → Γ(X,VX,ω) may
exist only if ω = kdy1 ∧ dy2/y1y2. On the other hand, if ω = kdy1 ∧ dy2/y1y2, then the map ρ̂
defined so that
ρ̂(E12) = y1∂y2 , (3.34)
ρ̂(E21) = y2∂y1 −
ky′2
y1
, (3.35)
ρ̂(E11) = y1∂y1 , (3.36)
ρ̂(E22) = y2∂y2 +
ky′1
y1
. (3.37)
delivers the desired vertex algebroid morphism
ρ̂ : V(gl2)−k−1,k−1 → Γ
(
C2 \ (0, 0),VC2\(0,0),ω
)
, ω = kdy1 ∧ dy2/y1y2. (3.38)
It is easy to see the vertex algebroid morphism condition determines the map uniquely. �
Note that the top row of (3.31), unlike that of (3.30), does not have to be exact in general.
In the case at hand, however, it is precisely when ω = kω11:
Corollary 3.7.6. The sequence
0→ Γ
(
C2 \ (0, 0),ΩC2\(0,0)
)
→ Γ
(
C2 \ (0, 0),VC2\(0,0),ω
)
1
→ Γ
(
C2 \ (0, 0), TC2\(0,0)
)
→ 0
is exact if and only if ω = kω11 for some k ∈ C.
Proof. Notice that TC2\(0,0) is generated by ρ(gl2) over functions. The “if” part is then seen
to be an immediate consequence of Lemma 3.7.5. The “only if” part was actually proved at the
beginning of the proof of the lemma cited. �
Vertex Algebroids over Veronese Rings 13
3.7.7. Conformal structure. If x1, . . . , xN are coordinates on CN , ∂j = ∂/∂j , then there is
a vertex (Poisson) algebra morphism
UpoissL(Vir)→ Γ
(
CN ,Dpoiss
CN
)
, UL(Vir)2N → Γ
(
CN ,DCN
)
,
L 7→
N∑
j=1
T (xj)(−1)∂j , (3.39)
where the latter is a quantization of the former. A little more generally, if A is an algebra suitable
for chiralization with τ1, . . . , τN an Abelian basis of Der(A), then one can find a coordinate
system, i.e., {x1, . . . , xN} ⊂ A s.t. τi(xj) = δij , and thus obtain
UpoissL(Vir)→ UpoissVpoiss(A), UL(Vir)2N → UV(A), L 7→
N∑
j=1
T (xj)(−1)τj .
In this case, N is the Krull dimension of A.
Another example is provided by the twisted sheaves VC2\0,ω of Section 3.7.3. Somewhat unex-
pectedly, the same definition (3.39), which in the present case becomes L 7→ T (y1)∂1 + T (y2)∂2,
applied locally over both both charts U1 and U2 survives the twisted gluing transformation (3.29)
for any ω and defines a global morphism
UL(Vir)2 → Γ
(
C2 \ 0,VC2\0,ω
)
. (3.40)
4 A graded Courant algebroid
attached to a commutative associative algebra
4.1. Modules of differentials. Even though the assumption that all the vector spaces in
question are Z-graded has been kept since the very beginning of Section 2, it has been barely
used. From now on it will be essential and referred to explicitly as in the following definition.
Definition 4.1.1. A differential graded algebra (DGA) R is a pair (R#, D), where R# =
⊕∞n=0R
n is a graded supercommutative associative algebra with Reven = ⊕∞n=0V
2n, Rodd =
⊕∞n=0R
2n+1, and D is a square 0 degree (−1) (hence odd) derivation. Call a DGA (R#, D)
quasi-free if there is a graded vector superspace V = ⊕∞n=0V
n with V even = ⊕∞n=0V
2n, V odd =
⊕∞n=0V
2n+1 such that R# is the symmetric algebra S•V .
If R is a DGA, then H•
D(R) def= Ker(D)/Im(D) is a a graded supercommutative associative
algebra.
For any commutative associative algebra A there is a quasi-free DGA R and a quasi-isomor-
phism
R→ A, (4.1)
that is to say, a DGA morphism (A being placed in homological degree 0 and equipped with
a zero differential) that delivers a graded algebra isomorphism H•
D(R) ∼−→ A.
If A is finitely generated, then a DGA resolution R can be chosen so that each V j from
Definition 4.1.1 is finite dimensional. These two finiteness assumptions will be made throughout.
A DGA resolution of A is not unique, but for any two such resolutions
R1 → A← R2
14 F. Malikov
there is a homotopy equivalence [2]
f : R1 → R2. (4.2)
If R is a quasi-free DGA, denote by Ω(R) the module of Kähler differentials of R. It is cano-
nically a differential-graded (DG) free R-module with derivation d : R→ Ω(R) and differential
LieD, which we choose to denote by D, too. The correspondence R 7→ Ω(R) is functorial in that
naturally associated to an algebra morphism f : R1 → R2 there is a map of DG R1-modules:
Ω(f) : Ω(R1)→ Ω(R2). (4.3)
Furthermore, we have
Ω(R) =
+∞⊕
n=0
Ω(R)n, d : Rn → Ω(R)n, D : Ω(R)n → Ω(R)n−1, [d,D] = 0. (4.4)
It follows that the homology HD(Ω(R)) is naturally a graded H•
D(R)-module.
For any 2 quasi-free DGA resolutions R1 → A ← R2, we can find a homotopy equivalence
f : R1 → R2, see (4.2), hence a quasi-isomorphism
Ω(f) : Ω(R1)→ Ω(R2) (4.5)
and an isomorphism
H(Ω(f)) : H•
D1
(Ω(R1))→ H•
D2
(Ω(R2)). (4.6)
Definition 4.1.2.
Ω(A)• = H•
D(Ω(R)), (4.7)
where R is a quasi-free DGA resolution of A.
The assignment A 7→ Ω(A)• defines a functor from the category of algebras to the category
of graded vector spaces.
Note that
Ω(A)• =
+∞⊕
n=0
Ω(A)n, (4.8)
is a graded A-module, and Ω(A)0 is the module of Kähler differentials of A, Ω(A).
4.2. Modules of derivations. If R is a quasi-free DGA, we denote by Der(R) the Lie
algebra of derivations of R. Like Ω(R), it is a DG R-module, but unlike Ω(R) it is graded in
both directions:
Der(R) =
⊕
n∈Z
Der(R)n (4.9)
and, which is more serious, not free; in fact, each component Der(R)n is a direct product
Der(R)n =
+∞∏
j=0
(V j)∗ ⊗Rn+j , (4.10)
where V j is one of the ingredients of Definition 4.1.1 assumed to be finite dimensional.
Vertex Algebroids over Veronese Rings 15
The derivation [D, ·] : Der(R) → Der(R) is a differential because D ∈ Der(R)−1 is odd.
Hence a Lie algebra H•
[D,·](Der(R)) arises.
The assignment R 7→ Der(R) is not quite functorial, because even if f : R1 → R2 is a quasi-
isomorphism, a Lie algebra morphism Der(f) : Der(R1) → Der(R2) does not quite exist. It
does exist though at the level of the corresponding homotopy categories. This remark and what
follows belongs to Hinich [18, Section 8].
Decompose f : R1 → R2 as follows
f : R1
i
↪→ S
p→ R2, (4.11)
where i is a standard acyclic cofibration, and p is an acyclic fibration. (Recall that i being
a standard cofibration means S being obtained by adjoining variables to R1, and being a fibration
means being an epimorphism.)
In the case of i, there arises a diagram of quasiisomorphisms
Der(R1)
πi← Der(i) ini→ Der(S), (4.12)
where Der(i) = {τ ∈ Der(S) s.t. τ(R1) ⊂ R1}, πi is the obvious projection, and ini is the obvious
embedding.
Analogously, in the case of p, there is a diagram of quasiisomorphisms
Der(S)
inp← Der(p)
πp→ Der(S), (4.13)
where Der(p) = {τ ∈ Der(R1) s.t. τ(Ker(p)) ⊂ Ker(p)}, inp is the obvious embedding, and πp is
the obvious projection.
Hinich defines
Der(i) = ini ◦ π−1
i , Der(p) = πp ◦ in−1
p , Der(f) = Der(p) ◦Der(i). (4.14)
This map makes sense in the homotopy category and delivers a homotopy category quasi-
isomorphism
Der(f) : Der(R1)→ Der(R2). (4.15)
Hence an isomorphism
H(Der(f)) : H•
[D1,·](R1)→ H•
[D2,·](R2). (4.16)
Theorem 4.2.1 ([18]). If fj : R1 → R2, j = 1, 2, are homotopic to each other, then Der(fj),
j = 1, 2, are also.
Corollary 4.2.2 ([18]).
(i) The assignment R 7→ Der(R) defines a functor from the homotopy category of DG commu-
tative associative algebras with quasi-isomorphisms to the homotopy category of DG Lie
algebras.
(ii) The assignment R 7→ H•
[D,·](Der(R)) defines a functor from the homotopy category of DG
commutative associative algebras with quasi-isomorphisms to the category of graded Lie
algebras.
Definition 4.2.3.
Der(A)• = H•
[D,·](Der(R)),
where R is a quasi-free DGA resolution of A.
16 F. Malikov
4.3. Synthesis: Courant algebroids. The notion of a Courant algebroid allows us to bring
Sections 4.1 and 4.2 under the same roof. Let Vpoiss be a Courant algebroid. It follows from
the Jacobi identity (2.4) that, for any ξ ∈ Vpoiss(R), ξ(0) ∈ End(Vpoiss(R)) is a derivation of
all products. Identity (2.2) implies that ξ(0) commutes with T . If, in addition, ξ is odd and
ξ(0)ξ = 0, then (ξ(0))2 = 0 as another application of (2.4) shows. Therefore, a pair (Vpoiss, ξ) is
a differential Courant algebroid, and the homology Courant algebroid, H•
ξ(0)
(Vpoiss), arises.
Let us now specialize this well-known construction to the Courant algebroid Vpoiss(R) =
Der(R) ⊕ Ω(R), see Example 3.4, in the case of a quasi-free DGA R = (R#, D). By definition
D is odd and, according to (3.10), D(0)D = [D,D] = 0. Hence the pair (Vpoiss(R), D(0)) is
a DG Courant algebroid and the graded Courant algebroid H•
D(0)
(Vpoiss(R)) arises. Again by
virtue of (3.10), the differential D(0) preserves Der(R) ⊂ Vpoiss(R), where it coincides with [D, ·],
and Ω(R), where it coincides with the standard action of D by the Lie derivative, see also (4.4).
We obtain a canonical vector space isomorphism
H•
D(0)
(Vpoiss(R)) = H•
[D,·](Der(R))⊕H•
D(Ω(R)). (4.17)
If f : R1 → R2 is a homotopy equivalence, then
H(Der(f))⊕H(Ω(f)) : H•
D(0)
(Vpoiss(R1))→ H•
D(0)
(Vpoiss(R2)), (4.18)
is a vector space isomorphism by virtue of (4.6) and (4.16). In fact, (4.18) is a graded Courant
algebroid isomorphism. This follows from the fact that the Courant algebroid structure on
Vpoiss(R) consists of classical differential geometry operations, such as the tautological action
of Der(R) on R and the action of Der(R) on Ω(R) by means of the Lie derivative. An inspection
of maps (4.11)–(4.16) shows that Hinich’s construction respects all these operations.
Corollary 4.3.1. The assignment R 7→ H•
D(0)
(Vpoiss(R)) defines a functor from the homotopy
category of DG commutative associative algebras with quasi-isomorphisms to the category of
graded Courant algebroids.
Definition 4.3.2.
Vpoiss(A)• = H•
D(0)
(Vpoiss(R)),
where R is a quasi-free DGA resolution of A.
4.4. Conformal structure. The construction of Section 3.7.7 in the present setting means
the following. If R# = S•V , pick a homogeneous basis {xi} ⊂ V and a dual ‘basis’ {∂i} ⊂ V ∗,
where ∂i(xj) = δij . As in Section 3.7.7, we obtain a morphism
UpoissL(Vir)→ UVpoiss(R), L 7→
∑
j
(T (xj))(−1)∂j . (4.19)
Lemma 4.4.1.
ξ(0)
∑
j
T (xj)(−1)∂j = 0 for any ξ ∈ Der(R). (4.20)
Vertex Algebroids over Veronese Rings 17
Corollary 4.4.2. Assignment (4.19) determines a Courant algebroid morphism
UpoissL(Vir)→ UVpoiss(A)0. (4.21)
Proof of Lemma. Let ξ =
∑
i(fi)(−1)∂i, fi ∈ R#. We have
ξ(0)
∑
j
T (xj)(−1)∂j =
∑
j
(ξ(0)T (xj))(−1)∂j +
∑
j
(−1)ξ·xjT (xj)(−1)(ξ(0)∂j)
=
∑
j
T (ξ(0)xj)(−1)∂j −
∑
j
(−1)ξ·xj+ξ·xjT (xj)(−1)
(
(∂j)(0)
∑
i
(fi)(−1)∂i
)
=
∑
j
T (fj)(−1)∂j −
∑
i
∑
j
T (xj)(−1)
(
∂fi
∂xj
)
(−1)
∂i
=
∑
j
T (fj)(−1)∂j −
∑
i
T (fi)(−1)∂i = 0. �
4.5. Grading of Der(A)• and identif ication of Der(A)0. Recall that R, hence Ω(R)
and Ω(A)•, are all graded by Z+. Contrary to this, although the Lie algebra Der(R) is graded
in both directions, Der(A) is Z−-graded.
Lemma 4.5.1.
(a) Der(A)n = 0 if n ≥ 0;
(b) Der(A)0 is the Lie algebra of derivations of A.
Proof. Consider a quasi-free DGA resolution R → A. The complex (Der(R), [D, ·]) is filtered
as follows, cf. (4.10),
F p Der(R)n =
+∞∏
j=p
(V j)∗ ⊗Rn+j . (4.22)
A spectral sequence (Er
pq, dr)⇒ Der(A)p+q arises so that
(E0
pq, d0) =
(
(V −p)∗ ⊗Rq, 1⊗D
)
.
Since R = (R#, D) is quasi-isomorphic to A placed in degree 0, we have
E1
pq =
{
(V −p)∗ ⊗A if q = 0,
0 otherwise.
It follows at once that the spectral sequence collapses and Der(A)−n is the n-th cohomology of
the complex
0→ (V 0)∗ ⊗A→ (V 1)∗ ⊗A→ · · · → (V n)∗ ⊗A→ · · · . (4.23)
Item (a) of the lemma is thus proven.
In order to prove item (b), we have to write down a formula for the differential of com-
plex (4.23). The resolution R→ A gives an exact sequence
0→ J ↪→ R0 π→ A→ 0.
18 F. Malikov
We shall regard an element τ ∈ (V j)∗ as a derivation of R = S•V . The differential, D, of R can
be written thus: D =
∑
fj∂j + ξ, where ∂j ∈ (V 1)∗, {fj} generate J , and ξ ∈ F 2Der(R)−1. It
easily follows from the construction of the spectral sequence that if τ ∈ (V 0)∗, then
d1(τ ⊗ a) = −aπ(τ(fj))∂j .
It follows at once that Ker{d1 : (V 0)∗ ⊗ A→ (V 1)∗ ⊗ A} is precisely the algebra of derivations
of R0 that preserve the ideal J modulo those derivations whose image is J , and this is Der(A)0
by definition. �
5 Quantization in the case of a Veronese ring
5.1. Set-up. Consider the Veronese ring
AN = C[x0, . . . , xN ]/Q, Q = (xixj − xi+1xj−1). (5.1)
It is known that Spec(AN ) is the cone over the highest weight vector orbit in the projectivization
of the (N + 1)-dimensional representation of sl2. Hence the canonical Lie algebra morphism
sl2 → Der(AN ). (5.2)
An explicit formula for this morphism will appear in Section 5.2.3 below.
Being a cone, Spec(AN ) carries the Euler vector field
∑
j xj∂j . This allows us to extend (5.2)
to an action of gl2:
gl2 → Der(A), where E11 + E22 7→ N
∑
j
xj∂j . (5.3)
Remark. The normalizing factor of N is not particularly important but can be justified by the
geometry of the base affine space SL2/N .
As in Section 3.7.4, this gives a Courant algebroid morphism
Vpoiss(gl2)→ Vpoiss(AN )0 ⊂ Vpoiss(AN )•. (5.4)
The following theorem, the main result of this paper, uses the concept of quantization of
a Courant algebroid, see Definition 3.5, and the notion of quantization of a Courant algebroid
map, see Section 3.7.4, (3.30), (3.31).
Theorem 5.1.1.
(a) The Courant algebroid Vpoiss(AN )• admits a unique quantization V(AN )•.
(b) Maps (5.3) and (4.21) can be quantized to the maps
V(gl2)−N−2,N → V(AN )0, (5.5)
UL(Vir)2 → UalgV(AN )0, (5.6)
where the functors U and Ualg are those defined in (2.8) and (3.23) resp.
The proof of the theorem is constructive, and an explicit construction of V(AN )• will appear
in Section 5.2.4 below.
Vertex Algebroids over Veronese Rings 19
5.2. Proof.
5.2.1. Beginning of the proof: a reduction to the homological degree 0. Suppose one
quantization, V(AN )•, is given. We have the direct sum decomposition
V(AN )• = ⊕n∈ZV(AN )n,
where V(AN )0 ⊂ V(AN ) is a vertex subalgebroid. Filtration (3.12) in the present situation
becomes
AN ⊕ Ω(AN )0 ⊂ V(AN )•.
Since AN ⊕Ω(AN )0 ⊂ V(AN )0, V(AN )0 is a quantization of Vpoiss(AN )0. This and the fact that
Ω(AN )• and Der(AN )• are graded in opposite directions, cf. (4.8) and Lemma 4.5.1(a), imply
a canonical vector space isomorphism
V(AN )• ∼−→ V(AN )0
⊕
(⊕n<0Der(AN )n)
⊕
(⊕n>0Ω(AN )n). (5.7)
Now suppose that only V(AN )0 is given.
Lemma. If V is a quantization of Vpoiss(AN )0, then there is a unique quantization, V(AN )•, of
Vpoiss(AN )• such that V(AN )0 = V.
Proof. (i) Uniqueness. Pick a splitting (over C) of the exact sequence of graded vector spaces,
cf. (3.13),
0→ Ω(AN )0 → V(AN )01 → Der(AN )0 → 0
so as to identify V(AN )1 with Ω(AN )⊕Der(AN ) and obtain the projection π : V(AN )1 → Ω(AN )
that is compatible with (5.7). It follows from Definition 3.5 that the only multiplications on
V(AN ) that are not immediately determined by those on Vpoiss(AN ) are the following compo-
nents of (0) and (1):
π ◦ ((0)) : Der(AN )• ⊗Der(AN )• → Ω(AN )•, ξ ⊗ τ 7→ π(ξ(0)τ),
(1) : Der(AN )• ⊗Der(AN )• → AN , ξ ⊗ τ 7→ ξ(1)τ.
The homological degree of the l.h.s. of these is non-positive, see (5.7), of the r.h.s. is non-negative;
therefore, the operations may be non-zero only if both ξ, τ ∈ Der(A)0; the uniqueness follows.
(ii) To prove the existence, note that Definition 3.2 differs from Definition 3.3 in the following
two respects only:
• the associativity of (−1) in the former is replaced with quasi-associativity (2.5) in the latter;
• the requirements of the former that multiplications (n), n ≥ 0, be derivations of multipli-
cation (−1) and that multiplication (−1) be commutative are simultaneously replaced with the
Jacobi identity (2.4) with m or n equal to −1.
Upon choosing a splitting as at the beginning of the proof, in each of the cases the identities of
Definition 2.2 exhibit quantum corrections, i.e., the terms that measure the failure of a quantum
object to be a classical one. It is easy to notice, by inspection, that in our situation the quantum
corrections may be non-zero only if all the terms involved belong to V = V(AN )0.
One such example is provided by formula (3.16), where the failure of multiplication (−1) to be
associative is measured by −τ(g)df−τ(f)dg; both the summands vanish unless τ ∈ V, f , g being
in V automatically.
20 F. Malikov
Another example deals with the commutativity of (−1). Let f ∈ AN , τ ∈ Der(AN )•; then (2.4)
reads
[τ(−1), f(−1)] = τ(f)(−2),
which is 0 unless τ ∈ V. We leave it to the untiring reader to check the validity of all the
remaining requirements of Definition 3.3. �
In order to prove Theorem 5.1.1, it remains to quantize Vpoiss(AN )0. We shall do this in
a somewhat roundabout manner.
5.2.2. Localization. To return to the hypothetical vertex algebroid V(AN )•. By virtue of
Section 5.2.1, it is enough to consider V(AN ) def= V(AN )0. Since C = Spec(AN ) is affine, we
can localize V(AN ), see (3.25) in Section 3.7.1, so as to get a sheaf VC ∈ VertC . Let Č be
C \ {0} and VČ the restriction of VC to Č. Apparently, VČ ∈ VertČ and, the manifold Č being
smooth, our strategy will be to use the classification of the objects of VertČ , Section 3.7.3, so
as to identify those vertex algebroids over Č that may have come from C as above.
We begin by realizing Č as a quotient of a manifold w.r.t. a finite group action. Consider
the action
ZN × C2 → C2, m̄(y1, y2) =
(
exp (2π
√
−1m/N)y1, exp (2π
√
−1m/N)y2
)
. (5.8)
The map
AN → C[y1, y2]ZN , xj 7→ yN−j
1 yj
2 (5.9)
is an isomorphism; hence isomorphisms
C2/ZN
∼−→ C, (C2 \ 0)/ZN
∼−→ Č. (5.10)
There arises a projection
p : C2 \ 0→ Č (5.11)
and a faithful functor
p∗ : VertČ → VertC2\0 (5.12)
It is an equivalence of categories
p∗ : VertČ → VertZN
C2\0, (5.13)
where VertZN
C2\0 is the full subcategory of ZN -equivariant vertex algebroids; the inverse functor
is that of ZN -invariants: VC2\0 7→ V
ZN
C2\0; cf. Section 3.7.1, (3.26), (3.27).
The objects of the category VertC2\0 were classified in Section 3.7.3 to the effect that there is
a 1-1 correspondence between isomorphism classes of vertex algebroids and linear combinations
ω =
∑
a,b>0 kabωab, where ωab is the 2-form dy1∧dy2/ya
1yb
2. It follows from the construction that
the vertex algebroid, VC2\0,ω is ZN -equivariant if and only if ω is, hence if and only if
ω =
∑
a,b>0,N divides a+b−2
kabωab. (5.14)
It is from this list that we have to select.
Vertex Algebroids over Veronese Rings 21
5.2.3. Conclusion of the proof. By definition, our hypothetical sheaf VC must fit, for
some ω, in the following commutative diagram:
0 // Γ(C,ΩC) //
� _
��
Γ(C,VC) //
� _
��
Γ(C, TC) //
��
0
0 // Γ(Č, ΩČ) // Γ(Č,VČ,ω) // Γ(Č, TČ) // 0
(5.15)
Note that the vertical arrows are all the restriction (from C to Č) maps. Furthermore, the
rightmost vertical arrow is an equality. To see this, note that Γ(Č, TČ) = Γ(C2 \ 0, TC2\0)ZN .
The latter is generated, over functions, by the tautological action of gl2, see Section 3.7.4, formu-
las (3.32), (3.33). (Indeed, an element of Γ(C2 \0, TC2\0)ZN is a linear combination of f(y1, y2)∂1
and g(y1, y2)∂2, where N divides deg(f) − 1 and deg(g) − 1. This implies that f(y1, y2)∂1 is
proportional to either ρ(E11) or to ρ(E21) and g(y1, y2)∂2 is proportional to either ρ(E12) or to
ρ(E22), see (3.33).) Therefore, so is the former. But this action is precisely the action of gl2 on
C = Spec(AN ) described somewhat implicitly in (5.2), hence the equality Γ(C, TC) = Γ(Č, TČ).
Contrary to this, the leftmost vertical arrow is not an equality; e.g. it is easy to check that
yr
1y
N−r−1
2 dy1 ∈ Γ(Č, ΩČ) but yr
1y
N−r−1
2 dy1 6∈ Γ(C,ΩC) if 0 ≤ r ≤ N − 2. (5.16)
This simple remark is the reason why the quantization of Vpoiss(AN ) is unique.
Now, the upper row of (5.15) is exact by virtue of Definition 3.5. This and the fact that
the rightmost arrow is an equality imply that the lower row must also be exact, at least on the
right. By virtue of Corollary 3.7.6,
ω = kω11. (5.17)
Now our task is to determine k.
Define W to be the vertex subalgebroid of Γ(Č,VČ,kω11
) generated by AN = Γ(Č,OČ) and
ρ̂(V(gl2)−k−1,k−1), where ρ̂ is the one from Lemma 3.7.5.
Since Γ(C, TC) is generated by ρ(gl2) over AN , W = Γ(C,VC). It is clear that W = AN ⊕
Γ(C,ΩC) + AN(−1)ρ̂Vch(gl2)−k−1,k−1 and were the elements ρ(Eij) independent over AN , we
would be done: W would be the sought after quantization for any k. (In fact, were that true, we
could equivalently define V(AN ) to be the push-out AN ⊗V(gl2)−k−1,k−1, see (3.28).) But they
are not, and the problem with this is that an element of AN(−1)ρ̂Vch(gl2)−k−1,k−1 may belong
to Γ(Č, ΩČ) and not to Γ(C,ΩC), cf. (5.16).
In fact, AN is a quadratic algebra, see (5.1), and Γ(C, TC) is a quadratic AN -module generated
by {Eij , 1 ≤ i, j ≤ 2}. The relations, in terms of y1, y2, are
ya
1yb
2ykρ̂(Eij)− ya
1yb
2yiρ̂(Ekj) = 0 for all a + b + 1 = N, (5.18)
as it easily follows from (3.33).
Our task then is to ensure that
(ya
1yb
2yk)(−1)ρ̂(Eij)− (ya
1yb
2yi)(−1)ρ̂(Ekj) ∈ Γ(C,ΩC) for all a + b + 1 = N. (5.19)
Let us consider for the sake of definiteness the case of i = 1, k = j = 2. We have to compute
the following section of Γ(Č,VČ,kω11
(Č)):
(yr
1y
N−r
2 )(−1)ρ̂(E12)− (yr+1
1 yN−r−1
2 )(−1)ρ̂(E22) for all 0 ≤ r < N. (5.20)
Formulas (3.34)–(3.37) allow us to re-write (5.20) as follows
(yr
1y
N−r
2 )(−1)(y1∂2)− (yr+1
1 yN−r−1
2 )(−1)
(
y2∂y2 +
kT (y1)
y1
)
.
22 F. Malikov
A little thought (or formula (3.16)) will show that the first term yields
yr+1
1 yN−r
2 ∂2 − r(N − r)yr
1y
N−r−1
2 T (y1)− (N − r)(N − r − 1)yr+1
1 yN−r−2
2 T (y2),
where it is understood that
yr+1
1 yN−r
2 ∂2
def= (y1(−1)(y1(−1)(· · · (y1(−1)(y2(−1)(· · · (y2(−1)∂2) · · · )))))).
The second one will likewise give
−yr+1
1 yN−r
2 ∂2 + (N − r − 1)(r + 1)yr
1y
N−r−1
2 T (y1)
+ (N − r − 1)(N − r − 2)yr+1
1 yN−r−2
2 T (y2)− kyr
1y
N−r−1
2 T (y1).
Adding one to another makes expression (5.20) into
(N − 1− 2r − k)yr
1y
N−r−1
2 T (y1) + (−2N + 2r + 2)yr+1
1 yN−r−2
2 T (y2).
The latter equals the total derivative
−2T (yr+1
1 yN−r−1
2 ) = −2T (xN−r−1),
and is therefore an element of Γ(C,ΩC), precisely when k = N + 1.
The case where i = j = 1, k = 2 works out similarly and gives the same answer k = N + 1.
This concludes the proof of item (a) of Theorem 5.1.1.
As to item (b), (5.5) follows from the k = N + 1 case of Lemma 3.7.5, the assertion that has
been instrumental for the proof anyway, and (5.6) follows from (3.40). �
Corollary 5.2.4. The unique quantization V(AN ) = V(AN )0 ⊕ Der(AN )• ⊕ Ω(AN )•, where
V(AN )0 is the vertex subalgebroid of Γ(Č,VČ,(N+1)ω11
) generated by AN = Γ(Č,OČ) and
ρ̂(V(gl2)−N−2,N ), ρ̂ being the one from Lemma 3.7.5.
5.3. Higher dimensional Veronese embeddings. Regard Cn as the tautological represen-
tation of gln and let V = (Cn)∗. Let
ιN : P(V )→ P(SNV ), l 7→ l⊗N (5.21)
be the classical Veronese embedding. By AnN let us denote the homogeneous coordinate ring of
ιN (P(V )). It is clear that if n = 2, then AnN is the algebra AN we dealt with above. It is now
natural to ask if Vpoiss(AnN ) affords a quantization. The result is a bit disheartening.
Theorem 5.3.1. The vertex Poisson algebroid Vpoiss(AnN ) cannot be quantized if N > 1 and
n > 2.
Proof. Consider the action
ZN × V → V, (m̄, v) 7→ exp (2π
√
−1m/N)v. (5.22)
Analogously to (5.9), (5.10), we obtain isomorphisms
V/ZN
∼−→ Spec(AnN ), (V \ 0)/ZN
∼−→ Spec(AnN ) \ 0. (5.23)
Thus we are led to the question, “How many vertex algebroids are there on (V \ 0)/ZN?” That
such vertex algebroids exist is obvious because VZN
V \0 is one; here VV \0 is the pull-back of the
standard VV , cf. Section 3.7.2, on V \0. Note that if x1, . . . , xn is a basis of Cn – remember that
we think of Cn as the space of linear functions on V – then the assignment
ρ̂ : Eij 7→ xi(−1)∂j (5.24)
defines a vertex algebroid morphism
ρ̂ : V(gln)−1,−1 → Γ
(
(V \ 0)/ZN ,VZN
V \0
)
. (5.25)
Vertex Algebroids over Veronese Rings 23
Lemma 5.3.2. The manifolds V \ 0, (V \ 0)/ZN carry a unique up to isomorphism sheaf of
vertex algebroids. It is isomorphic to VV \0 in the former case and to VZN
V \0 in the latter case.
Proof of Lemma 5.3.2. By virtue of (3.22) and (3.26), it suffices to show that
H1
(
V \ 0,Ω2
V \0 → Ω3,cl
V \0
)
= 0. (5.26)
Converging to the hypercohomology H∗(V \ 0,Ω2
V \0 → Ω3,cl
V \0) is a standard spectral sequence
with
E1
00 = H0
(
V \ 0,Ω2
V \0
)
,
E1
01 ⊕ E1
10 = H1
(
V \ 0,Ω2
V \0
)
⊕H0
(
V \ 0,Ω3,cl
V \0
)
.
The next differential is
d2 = dDR : H0
(
V \ 0,Ω2
V \0
)
→ H0
(
V \ 0,Ω3,cl
V \0
)
,
and it is clear that it is surjective. Finally, if n > 2, then H1(V \ 0,Ω2
V \0) = 0; this concludes
the proof of Lemma 5.3.2. �
It is clear now why Vpoiss(AnN ) cannot be quantized if n > 2: it is because Lemma 5.3.2 has
left us no room for manoeuvre that was helpfully provided by the analysis of Section 3.7.3 in
the n = 2 case. Indeed, one can now repeat the entire argument of Sections 5.2.1–5.2.3 only to
find out that an obvious analogue of (5.19) is false. Here are some details:
According to Lemma 5.3.2, a quantization of Vpoiss(AnN ), if existed, upon localizing to
(V \ 0)/ZN would give VZN
V \0. Hence any such quantization can be equal only to the vertex
subalgebroid of Γ(V \ 0,VV \0)ZN generated by AnN and ρ̂(V(gln)−1,−1, see (5.24-5.25). But this
subalgebroid necessarily contains elements from Γ((V \ 0)/ZN ,Ω(V \0)/ZN
) \ Ω(AnN ). Indeed, a
computation analogous to the one performed at the end of Section 5.2.3 shows that if n ≥ 3,
then (
x3x
N−1
2
)
(−1)
(x1(−1)∂1)−
(
x3x
N−2
2 x1
)
(−1)
(x2(−1)∂1)
=
(
T
(
x3x
N−2
2
))
(−1)
x2 ∈ Γ((V \ 0)/ZN ,Ω(V \0)/ZN
)) \ Ω(AnN ).
This concludes the proof of Theorem 5.3.1. �
6 Chiral Hamiltonian reduction interpretation
We will now interpret some of the constructions above in the language of semi-infinite coho-
mology. Our exposition will be brief and almost no proofs will be given. In some respects, the
material of this section is but an afterword to [16].
Since we will be mostly concerned with smooth varieties, we will find it convenient to work
not with vertex algebroids, such as V(g)k, VX , but with the corresponding vertex algebras or
CDO-s, such as UalgV(g)k = UL(g)k, Dch
X = UalgVX .
6.1. Semi-infinite cohomology. Let V be a vertex algebra, g a finite dimensional Lie algebra,
(·, ·) an invariant bilinear form on g, and ρ a vertex algebra morphism
ρ : UalgV(g)(·,·) → V, (6.1)
see Example 2.5, especially (2.12).
24 F. Malikov
Introduce the Clifford vertex algebra built on Π(g⊕ g∗), Π being the parity change functor.
This vertex algebra is nothing but the space of global sections of the standard CDO on superspace
Π(g⊕g∗), see Section 3.7.2 for a discussion of a purely even analogue. Denote this vertex algebra
by Dch(Π(g⊕ g∗)).
By definition, if we let {xi} be a basis of g, {φi} the corresponding basis of Π(g), {φ∗i } the
dual basis of Π(g∗), then Dch(Π(g ⊕ g∗)) is the vertex algebra generated by the vector space
Π(g⊕ g∗) and relations
(φ∗i )(0)φj = (φi)(0)φ
∗
j = δij1,
(φ∗i )(n+1)φj = (φi)(n+1)φ
∗
j = (φ∗i )(n)φ
∗
j = (φi)(n)φ
∗
j = 0 if n ≥ 0.
There arises the vertex algebra V⊗Dch(Π(g⊕ g∗)).
If {ck
ij} are the structure constants of g relative to {xi}, that is to say, if [xi, xj ] =
∑
k ck
ijxk,
then following [13] one considers the element
d∞/2 =
∑
k
ρ(xk)(−1)φ
∗
k −
1
2
∑
i,j,k
ck
ijφk(−1)(φ
∗
i(−1)φ
∗
j ) ∈ V⊗Dch(Π(g⊕ g∗)).
A direct computation shows that
(d∞/2)(0)d
∞/2 = 0 if (·, ·) = −K(·, ·), (6.2)
where K(·, ·) is the Killing form on g: K(a, b) = tr(ada · adb).
If condition (6.2) is satisfied, then we obtain a DGVA C∞/2(Lg; V) def= (V⊗Dch(Π(g⊕ g∗)),
d
∞/2
(0) ). The cohomology vertex algebra H∞/2(Lg; V) is due to Feigin [13] and well known as
either semi-infinite or BRST cohomology of the loop algebra Lg with coefficients in V. If one
chooses to think of V as an algebra of (‘chiral’) functions on a symplectic manifold with g-
structure, then H∞/2(Lg, V) is to be thought of as an algebra of functions on the symplectic
quotient M//g, hence the title of this section.
One similarly defines the relative version H∞/2(Lg, g; V), see [16] for some details; the con-
dition (6.2) remains the same in this case.
6.2. The sl2 case. Let us return to the set-up of Section 5.2.2, where we had the Veronese
cone C = Spec(AN ), Č = C \ 0, and consider LN , the degree N line bundle over P1, and
ĽN = LN \ {the zero section}. We obtain the commutative square
LN
// C
ĽN
∼ //
?�
OO
Č
?�
OO (6.3)
where the upper horizontal map is a surjective birational isomorphism, a blow-up of the vertex
of the cone. We have seen that Č carries a family of CDO-s, Dch
Č,ω
, ω ∈ H1(Č, Ω2
Č
). Denote the
coorresponding family of CDO-s on ĽN by Dch
ĽN ,ω
. Theorem 5.1.1 says that Dch
Č,ω
is a pull-back
of a CDO on C iff ω = (N − 1)ω11, in which case it contains V(sl2)−N−2.
Now a question arises, “For what, if any, ω is Dch
ĽN ,ω
a pull-back of a CDO from LN?” The
existence of such ω depends on the vanishing of the characteristic class ch2(LN ), (3.21). A simple
way to prove the vanishing result, and to compute a possible ω, is provided by the semi-infinite
cohomology.
Vertex Algebroids over Veronese Rings 25
One has the base affine space C2 \ 0, the principal N -bundle p : SL2 → C2 \ 0, where N is
the subgroup of upper-triangular matrices and a family of CDO-s, DSL2,(·,·) on SL2, over SL2.
This family enjoys [1, 16] the 2 vertex algebra embeddings
UalgV(g)(·,·)
ρl→ Γ(SL2, D
ch
SL2,(·,·))
ρr← UalgV(g)−(·,·)−K(·,·),
s.t. ρl(a)(n)ρr(b) = 0 if n ≥ 0. (6.4)
In this case condition (6.2) is satisfied for all forms (·, ·). Therefore, for any U ⊂ SL2/N ,
there arises a vertex algebra H∞(Ln,Γ(p−1(U), Dch
SL2,k)), where we use ρr in place of ρ, see (6.1).
Denote by H∞/2(Ln, Dch
SL2,(·,·)) the sheaf associated with the presheaf U 7→ H∞(Ln,Γ(p−1(U),
DSL2,(·,·))). It was noted in [16] that this sheaf is a CDO, which re-proves the obvious fact that
ch2(C2 \ 0) = 0.
Note that the left one of the embeddings (6.4) survives the passage to the cohomology. The
right one does not, not entirely at least, but the embedding of the torus part does, albeit with
a shifted central charge. We obtain
UalgV(g)(·,·)
ρl→ Γ
(
C2 \ 0,H∞/2(Ln, Dch
SL2,(·,·))
) ρr← UalgV(t)−(·,·)|t−1/2K(·,·)|t (6.5)
so that ρl(a)(n)ρl(b) = 0 if n ≥ 0; here (·, ·)|t and K(·, ·)|t stand for the restrictions of the
corresponding forms to t.
Altogether, the 2 embeddings provide a vertex algebra embedding of UalgV(gl2)k1,k2 with
appropriate central charges k1 and k2. In fact, if we let (a, b) = k tr(a·b) as we did in Section 3.7.4,
then we obtain that k1 = k, k2 = −k − 2 and a diagram, cf. Lemma 3.7.5 and (3.38).
UalgV(g)k ↪→ Γ(C2 \ 0,H∞/2(Ln, Dch
SL2,(·,·)))
∼−→ Γ(C2 \ 0, Dch
C2\0,−(k+1)ω11
). (6.6)
Lemma 3.7.5 shows that the chiral hamiltonian reduction technology reproduces precisely those
CDO-s on the punctured plane that carry an affine Lie algebra action.
In order to try and define a CDO on LN , we represent the latter as
LN = (C2 \ 0)×C∗ CN , (6.7)
where CN is the character C∗ 3 z 7→ zN . This suggests to define a CDO on LN as the chiral
hamiltonian reduction of Dch
C2\0,−(k+1)ω11
⊗Dch
C using embedding (6.5) twisted by action (6.7);
in practice that means that if h is the standard generator of t, then one has to replace ρr(h)
in (6.5) with ρr(h) + Ny(−1)∂y, where y is a coordinate on CN . Two things are to be kept in
mind: first, since the topology of C∗ is non-trivial [16], one has to use the relative version of
the semi-infinite cohomology; second, and most important, condition (6.2) is not automatically
satisfied. In fact, (6.2) is equivalent to
(ρr(h) + Ny(−1)∂y)(1)(ρr(h) + Ny(−1)∂y) = 0, (6.8)
which gives
(·, ·) = −
(
N2
8
+
1
2
)
K(·, ·). (6.9)
It follows thatH∞/2(Lt, t;Dch
C2\0,−(k+1)ω11
⊗Dch
C ) is well defined and gives a CDO on LN precisely
if (6.9) holds.
To conclude,
(1) the manifold ĽN carries a 1-parameter family of CDO-s, Dch
C2\0,−(k+1)ω11
, with
UalgV(sl2)(·,·)-structure, see Section 3.7.4;
26 F. Malikov
(2) the condition that a CDO on Č extends to one on C picks a unique (·, ·); the latter depends
on N linearly, see Theorem 5.1.1;
(3) the family Dch
C2\0,−(k+1)ω11
contains at least one representative that extends to a CDO
on LN ; the condition that a CDO on ĽN extends to LN and affords a realization via
the chiral Hamiltonian reduction picks a unique (·, ·); the latter depends on N quadrati-
cally, (6.9).
In fact, there is a third way to fix a (·, ·). This one amounts to carrying a regularization
procedure à la Lambert, used in [4] in a similar but different situation, and gives another
quadratic dependence on N . The importance of this approach is yet to be worked out.
6.3. Higher rank generalization. Let G be a simple complex Lie group, P ⊂ G a parabolic
subgroup, R ⊂ P the unipotent radical of R, M = P/R. Let us make the following assumption
M = M0 ×M1, where M0
∼−→ C∗, M1 is simple. (6.10)
Let Q ⊂ G be the extension of M1 by R. Thus we obtain a C∗-bundle
G/Q→ G/P
and the associated line bundle
LQ → G/P.
We have
ch2(G/Q) = 0. (6.11)
Indeed, there is at least one CDO on G/Q that can be defined via the chiral Hamiltonian reduc-
tion as follows. By analogy with Section 6.2, since (6.4) holds true with SL2 replaced with an
arbitrary simple complex Lie group G, we observe that if q = Lie(Q), then H∞/2(Lq, q;Dch
G,(·,·))
is well defined for precisely one choice of (·, ·). In fact, in this case condition (6.2) amounts
to the requirement that the restriction (·, ·) to M1 be equal to the Killing form on M1, and
there is one and only one way to achieve that by appropriately rescaling (·, ·) – this is where
assumption (6.10) is crucial. It is rather clear that H∞/2(Lq, q;Dch
G,(·,·)) is a CDO on G/Q.
As a consequence, we obtain a vertex algebra morphism
UalgV(g)(·,·) → H∞/2(Lq, q;Dch
G,(·,·)) (6.12)
for a uniquely determined bilinear form (·, ·).
Assertion (6.11) is of course analogous to the fact that ch2(SL2/N) = 0, which was discussed
in Section 6.2. Unlike the sl2-case, however, we have obtained not a family but a single CDO on
G/Q, and this precludes the definition of a CDO on LQ via the chiral Hamiltonian reduction.
It is then natural to expect that ch2(LQ) 6= 0 and that each CDO on G/Q with UalgV(g)(·,·)-
structure is isomorphic to H∞/2(Lq, q;Dch
G,(·,·)).
One example of this analysis is provided by G = SLn with Q chosen to be the subgroup with
1st column equal to (1, 0, 0, . . . , 0). Then G/Q is Cn \ 0, the corresponding CDO has been used
in Section 5.3, and it is easy to check that in this case the embedding (6.12) becomes precisely
(5.24), (5.25).
Another example is provided by the space of pure spinors punctured at a point. It is a ho-
mogeneous space which satisfies assumption (6.10). Therefore, our analysis is an alternative
Vertex Algebroids over Veronese Rings 27
way to prove the vanishing of the 2nd component of the Chern character, originally verified by
Nekrasov [24].
Needless to say, our discussion is very close in spirit to the definition of Wakimoto modules
due to Wakimoto and Feigin–Frenkel, see [11] and references therein. In fact, it is easy to see
that the spaces of sections over ‘the big cell’ of the sheaves constructed in Section 6.2 contain
Wakimoto modules over ŝl2, and those of the present section contain the so-called generalized
Wakimoto modules corresponding to R.
Acknowledgements
The author would like to thank N. Nekrasov and especially V. Hinich for interesting discussions
and for bringing [18] to his attention. The paper was completed at the IHES in Bures-sur-Yvette.
We are grateful to the institute for hospitality and excellent working conditions. This work was
partially supported by an NSF grant. Special thanks go to V. Gorbounov and V. Schechtman.
References
[1] Arkhipov S., Gaitsgory D., Differential operators on the loop group via chiral algebras, Int. Math. Res. Not.
2002 (2002), no. 4, 165–210, math.AG/0009007.
[2] Behrend K., Differential graded schemes I: Perfect resolving algebras, math.AG/0212225.
[3] Backelin J., Fröberg R., Koszul algebras, Veronese subrings and rings with linear resolutions, Rev. Roumaine
Math. Pures Appl. 30 (1985), no. 2, 85–97.
[4] Berkovits N., Nekrasov N., The character of pure spinors, Lett. Math. Phys. 74 (2005), 75–109,
hep-th/0503075.
[5] Berkovits N., Super-Poincaré covariant quantization of the superstring, J. High Energy Phys. 2000 (2000),
no. 4, 18, 17 pages, hep-th/0001035.
[6] Bezrukavnikov R., Koszul property and Frobenius splitting of Schubert varieties, alg-geom/9502021.
[7] Bressler P., The first Pontryagin class, Compos. Math. 143 (2007), 1127–1163, math.AT/0509563.
[8] Courant T.J., Dirac manifolds, Trans. Amer. Math. Soc. 319 (1990), 631–661.
[9] Dorfman I., Dirac structures of integrable evolution equations, Phys. Lett. A 125 (1987), no. 5, 240–246.
[10] Frenkel E., Private communication.
[11] Frenkel E., Wakimoto modules, opers and the center at the critical level, Adv. Math. 195 (2005), 297–404,
math.QA/0210029.
[12] Frenkel E., Ben-Zvi D., Vertex algebras and algebraic curves, 2nd ed., Mathematical Surveys and Mono-
graphs, Vol. 88, American Mathematical Society, Providence, RI, 2004.
[13] Feigin B., Semi-infinite homology of Lie, Kac–Moody and Virasoro algebras, Uspekhi Mat. Nauk 39 (1984),
no. 2, 195–196 (in Russian).
[14] Feigin B., Parkhomenko S., Regular representation of affine Kac–Moody algebras, in Algebraic and geo-
metric methods in mathematical physics (Kaciveli, 1993), Math. Phys. Stud., Vol. 19, Kluwer Acad. Publ.,
Dordrecht, 1996, 415–424, hep-th/9308065.
[15] Gorbounov V., Malikov F., Schechtman V., Gerbes of chiral differential operators. II. Vertex algebroids,
Inv. Math. 155 (2004), 605–680, math.AG/0003170.
[16] Gorbounov V., Malikov F., Schechtman V., On chiral differential operators over homogeneous spaces, Int.
J. Math. Math. Sci. 26 (2001), no. 2, 83–106, math.AG/0008154.
[17] Gorbounov V., Schechtman V., Homological algebra and divergent series, arXiv:0712.3670.
[18] Hinich V., Homological algebra of homotopy algebras, Comm. Algebra 25 (1997), 3291–3323, q-alg/9702015.
[19] Inamdar S.P., Mehta V.B., Frobenius splitting of Schubert varieties and linear syzygies, Amer. J. Math.
116 (1994), 1569–1586.
[20] Kac V., Vertex algebras for beginners, 2nd ed., University Lecture Series, Vol. 10, American Mathematical
Society, Providence, RI, 1998.
http://arxiv.org/abs/math.AG/0009007
http://arxiv.org/abs/math.AG/0212225
http://arxiv.org/abs/hep-th/0503075
http://arxiv.org/abs/hep-th/0001035
http://arxiv.org/abs/alg-geom/9502021
http://arxiv.org/abs/math.AT/0509563
http://arxiv.org/abs/math.QA/0210029
http://arxiv.org/abs/hep-th/9308065
http://arxiv.org/abs/math.AG/0003170
http://arxiv.org/abs/math.AG/0008154
http://arxiv.org/abs/arXiv:0712.3670
http://arxiv.org/abs/q-alg/9702015
28 F. Malikov
[21] Li H., Abelianizing vertex algebras, Comm. Math. Phys. 259 (2005), 391–411, math.QA/0409140.
[22] Liu Z.-J., Weinstein A., Xu P., Manin triples for Lie bialgebroids, J. Differential Geom. 45 (1997), 547–574,
dg-ga/9508013.
[23] Malikov F., Lagrangian approach to sheaves of vertex algebras, Comm. Math. Phys. 278 (2008), 487–548,
math.AG/0604093.
[24] Nekrasov N., Lectures on curved beta-gamma systems, pure spinors, and anomalies, hep-th/0511008.
[25] Primc M., Vertex algebras generated by Lie algebras, J. Pure Appl. Algebra 135 (1999), 253–293,
math.QA/9901095.
http://arxiv.org/abs/math.QA/0409140
http://arxiv.org/abs/dg-ga/9508013
http://arxiv.org/abs/math.AG/0604093
http://arxiv.org/abs/hep-th/0511008
http://arxiv.org/abs/math.QA/9901095
1 Introduction
2 Vertex algebras
3 Courant and vertex algebroids
4 A graded Courant algebroid attached to a commutative associative algebra
5 Quantization in the case of a Veronese ring
6 Chiral Hamiltonian reduction interpretation
References
|