With braiding
With duals for objects
category with duals (list of them)
dualizable object (what they have)
ribbon category, a.k.a. tortile category
With duals for morphisms
With traces
Closed structure
Special sorts of products
Semisimplicity
Morphisms
Internal monoids
Examples
Theorems
In higher category theory
Definitions
Transfors between 2-categories
Morphisms in 2-categories
Structures in 2-categories
Limits in 2-categories
Structures on 2-categories
The notion of the center of a monoidal category or the Drinfeld center is a categorification of the notion of the center of a monoid(associative algebra, group, etc.) from monoids to monoidal categories.
Where the center of a monoid is just a sub-monoid with the property that it commutes with everything else, under categorification this becomes a structure, since we have to specify how the objects in the Drinfeld center commute (braid) with everything else.
We first give the general-abstract definition
of Drinfeld centers. Then we spell out what this means in components in
For a monoidal category, write for its delooping, the pointed 2-category with a single object such that .
The Drinfeld center of is the monoidal category of endo-pseudonatural transformations of the identity-2-functor on :
Unwinding the definitions, we find that an object of , , has for components pseudonaturality squares
for each . As shown, these consist of a choice of an object together with a natural isomorphism
in .
The transfor-property of says that
And so forth. Writing this out in terms of yields the following component characterization of Drinfeld centers, def. .
Let be a monoidal category. Its Drinfeld center is (cf. EGNO15 ยง7.13) the braided monoidal category whose
objects are pairs of an object and a natural isomorphism (the half-braiding)
such that for all we have
morphisms are given by
the tensor product is given by
the braiding is given (cf. EGNO15 (8.15)) by:
The Drinfeld center is naturally a braided monoidal category.
If is a fusion category over an algebraically closed field of characteristic zero, then the Drinfeld center is also naturally a fusion category.
(Etingof, Nikshych & Ostrik 2005, Thm. 2.15, review in Davydov, Mueger, Nikshych & Ostrik 2003, Sec. 2.3, see also Drinfeld, Gelaki, Nikshych & Ostrik 2010, Cor. 3.9, Mueger 2003, EGNO 2015, Thm. 9.3.2).
Under Tannaka duality, forming the Drinfeld center of a category of modules of some Hopf algebra corresponds to forming the category of modules over the corresponding Drinfeld double algebra (cf. EGNO15, ยง7.14).
Tannaka duality for categories of modules over monoids/associative algebras
| monoid/associative algebra | category of modules |
|---|---|
| -algebra | -2-module |
| sesquialgebra | 2-ring = monoidal presentable category with colimit-preserving tensor product |
| bialgebra | strict 2-ring: monoidal category with fiber functor |
| Hopf algebra | rigid monoidal category with fiber functor |
| hopfish algebra (correct version) | rigid monoidal category (without fiber functor) |
| weak Hopf algebra | fusion category with generalized fiber functor |
| quasitriangular bialgebra | braided monoidal category with fiber functor |
| triangular bialgebra | symmetric monoidal category with fiber functor |
| quasitriangular Hopf algebra (quantum group) | rigid braided monoidal category with fiber functor |
| triangular Hopf algebra | rigid symmetric monoidal category with fiber functor |
| supercommutative Hopf algebra (supergroup) | rigid symmetric monoidal category with fiber functor and Schur smallness |
| form Drinfeld double | form Drinfeld center |
| trialgebra | Hopf monoidal category |
2-Tannaka duality for module categories over monoidal categories
| monoidal category | 2-category of module categories |
|---|---|
| -2-algebra | -3-module |
| Hopf monoidal category | monoidal 2-category (with some duality and strictness structure) |
3-Tannaka duality for module 2-categories over monoidal 2-categories
| monoidal 2-category | 3-category of module 2-categories |
|---|---|
| -3-algebra | -4-module |
The archetypical example of Drinfeld centers is that of the of the category of finite-dimensional -graded vector spaces for a group .
This is also known as the representation category of the Drinfeld double of .
We spell out (following SS26 ยงA.3) the characterization of the simple objects (Prop. , which is classical, cf. EGNO 2015 Example 8.5.4. in) and also the fusion coefficients (Prop. , which, by other means, were given only recently in Li 2026).
We start by recalling definitions and establishing notation:
By we denote the monoidal category of finite-dimensional complex vector spaces with the usual tensor product of vector spaces. For a group, the monoidal category of -graded vector spaces has:
as objects vector spaces equipped with -indexed direct sum decomposition
as morphisms linear maps that respect the -grading,
as monoidal structure the ordinary tensor product but equipped with the convolution grading:
We will be referring to , and its incarnation (6) in the Drinfeld center, as the fusion product, to be distinguished from the degreewise tensor product (cf. also Rem. below).
Equivalently, with
denoting the Hopf algebra of functions (with product the pointwise product of functions and with coproduct induced by precomposition of functions with the group multiplication), we have that is equivalent (as a monoidal category) to the category of modules of :
(The pointwise product in induces the -grading and the coproduct induces the convolution in (1).)
For example, with an ungraded vector space and a group element, we write
for the -graded vector space which is concentrated on in degree .
Now, the Drinfeld center is, by its general definition (Def. , but we now change the conventuion for the handedness of the half-braiding in order to end up with left group actions), the monoidal category which has:
as objects pairs, , consisting of a and a natural isomorphism (the \emph{half-braiding})
satisfying for all the relation
as morphisms linear maps such that for all we have
as tensor product the operation which on the graded vector spaces is the convolution product from (1), extended to the half-braiding as follows:
The above general definition of Drinfeld center applied to the case of may be further simplified:
Since every vector space is isomorphic to a direct sum of copies of , the linear maps (3) are determined already by their values on (2); and if we understand that , then, by degree reasons, their -components must be linear isomorphisms of this form:
For these, the coherence condition (4) reduces to the action property
This way, the objects of are bijectively identified with (complex, finite-dimensional) groupoid representations of the action groupoid
of the adjoint action of (its inertia groupoid):
In this representation-theoretic description, the morphisms (5) of are equivalently the homomorphisms of groupoid representations (the intertwiners). This shows that:
For a group, the Drinfeld center of is equivalently the representation category of the inertia groupoid of :
The equivalence (7) implies at once that the simple objects in (those admitting no nontrivial direct sum decomposition) are supported on conjugacy classes of group elements
where they form an (irreducible) groupoid representation of the connected component . Since these connected groupoids are equivalent to the delooping of their isotropy group, (cf. this prop.), these groupoid representations are equivalently irreducible linear representations of the centralizer group :
The simple objects of are pairs consisting of a conjugacy class and an irrep of :
Beware that the above equivalence (7) is not a monoidal equivalence, as the standard tensor product on groupoid representations is degree-wise, not convolutive as in (1).
Instead, the fusion product in the Drinfeld center is category-theoretically given as follows:
For a pair of objects (?), their fusion tensor product (?) is equivalently given by the following pull-tensor-push operation:
in that there is a natural isomorphism
Since the functor is a Kan fibration, the left base change is given by forming the direct sum over fibers. (It is immediate to check the universal property of the left adjoint explicitly.) This reproduces the definition (1). The induced -action on these direct sums is the diagonal one, which under the equivalence (7) reproduces the definition (6).
To determine concretely the convolution tensor product of a pair of such simple objects, we first need:
For , with denoting the Cartesian product of their conjugacy classes (?), the action groupoid of the diagonal adjoint action is equivalent to the following disjoint union of delooping groupoids:
where denotes the following double coset:
Consider the group multiplication map . Since this is -equivariant for the (diagonal) adjoint action, the set of -orbits in is partitioned into subsets labeled by conjugacy classes . An orbit in is in the subset labeled by iff it contains an element such that , for some . With the representative of held fixed, the latter pair has a unique class , and hence the set of orbits labeled by is in bijection to (9).
This shows that the disjoint union on the right of (8) is indexed by the set of orbits of the groupoid on the left. Since the disjoint union is over the delooping groupoids of the isotropy groups of these orbits, the claim follows.
It follows that:
The fusion product of a pair of simple objects contains any simple object with the following multiplicity:
By Lem. we have that the fusion product is given by pull-push through
By Lemma , this yields on the orbit the isotropy representation
where we are using that on group representations the left base change is given by forming induced representations.
This yields the claim: By Schur's lemma, the multiplicity of the irrep in this expression is the dimension of the hom space from the latter to the former. Finally, by Frobenius reciprocity, , the induction is left adjoint to restriction, whence we get (10) from the hom-isomorphism
Up to immediate translation, Prop. coincides with Li 2026 Thm. 3.2, proven there by different methods (Mackey theory applied to the Drinfeld double of ). As remarked there, this is in turn the untwisted specialization of a formula given by Goff 2012 Thm. 4.5.
Original articles:
Michael Mueger, From Subfactors to Categories and Topology II. The quantum double of tensor categories and subfactors, J. Pure Appl. Alg. 180, 159-219 (2003) (arXiv:math/0111205)
Pavel Etingof, Dmitri Nikshych, Victor Ostrik, On fusion categories, Annals of Mathematics Second Series, Vol. 162, No. 2 (Sep., 2005), pp. 581-642 (arXiv:math/0203060, jstor:20159926)
Vladimir Drinfeld, Shlomo Gelaki, Dmitri Nikshych, Victor Ostrik, On braided fusion categories I, Selecta Mathematica. New Series 16 (2010), no. 1, 1โ119 (doi:10.1007/s00029-010-0017-z)
Review:
Textbook accounts:
Shahn Majid: Foundations of quantum group theory, Cambridge University Press (1995, 2000) [doi:10.1017/CBO9780511613104]
Christian Kassel: Quantum groups, Graduate Texts in Mathematics 155, Springer (1995) [doi:10.1007/978-1-4612-0783-2, webpage, errata pdf]
Bojko Bakalov, Alexandre Kirillov; section 3.2 in: Lectures on tensor categories and modular functors, University Lecture Series 21, Amer. Math. Soc. (2001) [webpage, ams:ulect/21, pdf]
Pavel Etingof, Shlomo Gelaki, Dmitri Nikshych, Victor Ostrik: Tensor Categories, Mathematical Surveys and Monographs 205, AMS (2015) [ISBN:978-1-4704-3441-0, pdf]
A general discussion of centers of monoid objects in braided monoidal 2-categories (which reduces to the above for the 2-category Cat with its cartesian product) is in
An application to character sheaves is in
In relation to spectra of tensor triangulated categories:
Relation to Frobenius monoidal functors:
Explicit derivation of the fusion rules in the Drinfeld center of :
Last revised on March 25, 2026 at 09:11:22. See the history of this page for a list of all contributions to it.