L-R-smash biproducts, double biproducts and a braided category of Yetter-Drinfeld-Long bimodules
Florin Panaite, Freddy Van Oystaeyen

TL;DR
This paper introduces the concept of L-R-smash biproducts, generalizing Radford biproducts, and constructs a prebraided monoidal category of Yetter-Drinfeld-Long bimodules with applications in bialgebra theory.
Contribution
It defines L-R-smash biproducts, establishes conditions for their structure, and constructs a new prebraided monoidal category of bimodules with bialgebra properties.
Findings
L-R-smash biproducts generalize Radford biproducts.
Conditions for D to form a bialgebra in the category are identified.
A new prebraided monoidal category of bimodules is constructed.
Abstract
Let H be a bialgebra and D an H-bimodule algebra H-bicomodule coalgebra. We find sufficient conditions on D for the L-R-smash product algebra and coalgebra structures on D\otimes H to form a bialgebra (in this case we say that (H, D) is an L-R-admissible pair), called L-R-smash biproduct. The Radford biproduct is a particular case, and so is, up to isomorphism, a double biproduct with trivial pairing. We construct a prebraided monoidal category {\cal LR}(H), whose objects are H-bimodules H-bicomodules M endowed with left-left and right-right Yetter-Drinfeld module as well as left-right and right-left Long module structures over H, with the property that, if (H, D) is an L-R-admissible pair, then D is a bialgebra in {\cal LR}(H).
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAlgebraic structures and combinatorial models · Advanced Topics in Algebra · Homotopy and Cohomology in Algebraic Topology
L-R-smash biproducts, double biproducts and a braided category of
Yetter-Drinfeld-Long bimodules
Florin Panaite
Institute of Mathematics of the Romanian Academy
PO-Box 1-764, RO-014700 Bucharest, Romania
e-mail: [email protected] Research carried out while the first author was visiting the University of Antwerp, supported by a postdoctoral fellowship offered by FWO (Flemish Scientific Research Foundation). This author was also partially supported by the programme CEEX of the Romanian Ministry of Education and Research, contract nr. 2-CEx06-11-20/2006.
Freddy Van Oystaeyen
Department of Mathematics and Computer Science
University of Antwerp, Middelheimlaan 1
B-2020 Antwerp, Belgium
e-mail: [email protected]
Abstract
Let be a bialgebra and an -bimodule algebra and -bicomodule coalgebra. We find sufficient conditions on for the L-R-smash product algebra and coalgebra structures on to form a bialgebra (in this case we say that is an L-R-admissible pair), called L-R-smash biproduct. The Radford biproduct is a particular case, and so is, up to isomorphism, a double biproduct with trivial pairing. We construct a prebraided monoidal category , whose objects are -bimodules -bicomodules endowed with left-left and right-right Yetter-Drinfeld module as well as left-right and right-left Long module structures over , with the property that, if is an L-R-admissible pair, then is a bialgebra in .
Introduction
The L-R-smash product over a cocommutative Hopf algebra was introduced and studied in a series of papers [1], [2], [3], [4], with motivation and examples coming from the theory of deformation quantization. This construction was generalized in [13] to the case of arbitrary bialgebras (even quasi-bialgebras), as follows: if is a bialgebra and is an -bimodule algebra, the L-R-smash product D\mbox{;\natural;}H is an associative algebra structure defined on by the multiplication rule
[TABLE]
It was proved in [13] that, if is moreover a Hopf algebra with bijective antipode, then D\mbox{;\natural;}H is isomorphic to a diagonal crossed product as in [5], [7]; this result was used in [12] to give a very easy proof of the fact that two bialgebroids introduced independently in [6] and [8] are actually isomorphic.
The dual construction of the L-R-smash product was introduced also in [13] under the name L-R-smash coproduct; this is a coassociative coalgebra D\mbox{;\natural;}H, where is an -bicomodule coalgebra. A natural problem, not treated in [13], is to see under what conditions, for a given -bimodule algebra -bicomodule coalgebra , the L-R-smash product and coproduct structures on form a bialgebra. It seems to be difficult to obtain (nicely-looking) necessary and sufficient conditions on for this to happen. The aim of the present paper is to present a list of sufficient conditions, looking resonably nice and being general enough to cover some existing constructions from the literature.
More precisely, if satisfies those conditions, we say that is an L-R-admissible pair and the bialgebra D\mbox{;\natural;}H is called an L-R-smash biproduct. The Radford biproduct is a particular case, corresponding to the situation when the right action and coaction are trivial. We prove that a double biproduct (as in [10], [15]) with trivial pairing is isomorphic to an L-R-smash biproduct (A\otimes B)\mbox{;\natural;}H. Also, we show that a construction introduced in [16] is a particular case of an L-R-smash biproduct.
It is known that the Radford biproduct has a categorical interpretation (due to Majid): is an admissible pair (as in [14]) if and only if is a bialgebra in the Yetter-Drinfeld category . We give a similar interpretation for L-R-admissible pairs. Namely, we define a prebraided category (which is braided if has a skew antipode) consisting of -bimodules -bicomodules which are left-left and right-right Yetter-Drinfeld modules as well as left-right and right-left Long modules over (this category contains and as braided subcategories). We prove that all except one of the conditions for to be an L-R-admissible pair are equivalent to being a bialgebra in . The extra condition reads
[TABLE]
and unfortunately does not seem to have a categorical interpretation inside .
1 The L-R-smash biproduct
We work over a field . All algebras, linear spaces etc. will be over ; unadorned means . For a bialgebra with comultiplication we denote , for . For terminology concerning bialgebras, Hopf algebras and monoidal categories we refer to [9], [11].
Let be a bialgebra and let be a vector space satisfying the following conditions:
(i) is an -bimodule, with actions and , for and ;
(ii) is an algebra, with unit and multiplication , for ;
(iii) is an -bimodule algebra, that is , , and , for all and ;
(iv) is an -bicomodule, with structures (for all ):
[TABLE]
(v) is a coalgebra, with comultiplication , , and counit ;
(vi) is an -bicomodule coalgebra, that is, for all :
[TABLE]
We denote the vector space by D\mbox{;\natural;}H and elements by d\mbox{;\natural;}h. By [13], D\mbox{;\natural;}H becomes an algebra (called L-R-smash product) with unit 1_{D}\mbox{;\natural;}1_{H} and multiplication
[TABLE]
and a coalgebra (called L-R-smash coproduct) with comultiplication and counit given by
[TABLE]
We consider now the following list of conditions, for and as above, corresponding to elements and :
[TABLE]
If all these conditions hold, for all and , by analogy with [14] we will say that is an L-R-admissible pair.
Theorem 1.1
If is an L-R-admissible pair, then D\mbox{;\natural;}H with structures as above is a bialgebra, called the L-R-smash biproduct of and .
Proof. It is very easy to see that \varepsilon_{D\mbox{;\natural;}H} is an algebra map and \Delta_{D\mbox{;\natural;}H} is unital, so we will only prove that \Delta_{D\mbox{;\natural;}H} is multiplicative. We will prove first two auxiliary relations:
[TABLE]
[TABLE]
for all , ; we compute:
[TABLE]
[TABLE]
Let now and ; we compute:
\Delta((c\mbox{;\natural;}h)(d\mbox{;\natural;}g))
[TABLE]
[TABLE]
and we see that the two terms are equal.
Remark 1.2
Obviously, the Radford biproduct (cf. [14]) is a particular case of the L-R-smash biproduct, corresponding to the case when the right action and coaction are trivial.**
We recall now from [10], [15] the construction of the so-called double biproduct, more precisely a particular case of it (corresponding to a trivial pairing, in the terminology of [10]). Let be a bialgebra, a bialgebra in the Yetter-Drinfeld category and a bialgebra in the Yetter-Drinfeld category , with the following notation for the structure maps: counits and , comultiplications and , and actions and coactions
[TABLE]
for all , , . We denote by the vector space (the element is denoted by ), which becomes an algebra (called two-sided smash product) with unit and multiplication
[TABLE]
and a coalgebra (called two-sided smash coproduct) with counit and comultiplication
[TABLE]
Proposition 1.3
([10], [15]) Assume that moreover the following condition holds:
[TABLE]
Then is a bialgebra, called the double biproduct.
Proposition 1.4
Let be a double biproduct bialgebra. Define , with tensor product algebra and coalgebra structures and with two-sided actions and coactions given by
[TABLE]
Then is an L-R-admissible pair and we have a bialgebra isomorphism
[TABLE]
Proof. The fact that is an L-R-admissible pair follows by direct computation; let us only check (1.14), for and :
[TABLE]
We know from [13], Proposition 2.4, that is an algebra isomorphism, and an easy computation shows that is also a coalgebra map.
We recall now the following result from [16]. Let be a bialgebra and an -bimodule bialgebra (i.e. is a bialgebra which is an -bimodule algebra and an -bimodule coalgebra). Consider the L-R-smash product algebra D\mbox{;\natural;}H, together with the tensor product coalgebra structure on it (i.e. \Delta(d\mbox{;\natural;}h)=(d_{1}\mbox{;\natural;}h_{1})\otimes(d_{2}\mbox{;\natural;}h_{2}) and \varepsilon(d\mbox{;\natural;}h)=\varepsilon_{D}(d)\varepsilon_{H}(h)). Then D\mbox{;\natural;}H with these structures is a bialgebra if and only if the following conditions are satisfied, for all , :
[TABLE]
This result is a particular case of Theorem 1.1. Indeed, consider on the left and right trivial -coactions (i.e. and , for ). Then one can easily check that is an L-R-admissible pair ((1.18) and (1.19) are precisely (1.10) and respectively (1.12)) and the L-R-smash coproduct coalgebra structure in this case coincides with the tensor product coalgebra structure.
2 A braided category related to L-R-smash biproducts
Let be a bialgebra. We will introduce a prebraided category associated to , denoted by . The objects of are vector spaces endowed with -bimodule and -bicomodule structures (denoted by , , , , for all , ), such that is a left-left Yetter-Drinfeld module, a left-right Long module, a right-right Yetter-Drinfeld module and a right-left Long module, i.e.
[TABLE]
for all , . The morphisms in are the -bilinear -bicolinear maps.
One can check that becomes a strict monoidal category, with unit endowed with usual -bimodule and -bicomodule structures, and tensor product given as follows: if then with structures (for all , , ):
[TABLE]
Proposition 2.1
The monoidal category is a prebraided category, with braiding defined, for all , , , by
[TABLE]
If has a skew antipode , then is braided, the inverse of being given by
[TABLE]
Proof. We only check that is left -linear, right -colinear and satisfies one of the two hexagonal equations, and leave the rest to the reader. For and , , , , we compute:
[TABLE]
[TABLE]
[TABLE]
Also, the bijectivity of in the presence of a skew antipode follows by a direct computation which is left to the reader.
Remark 2.2
We denote as usual by and the categories of left-left and respectively right-right Yetter-Drinfeld modules over . One can check that, if V\in$$\;{}^{H}_{H}{\cal YD} and , then , with structures as in Proposition 1.4. In particular, for and respectively , we obtain that and are subcategories of , and one can see that they are actually braided subcategories, i.e. the braiding of restricts to the usual braidings of and . **
We can state now the categorical interpretation of L-R-admissible pairs:
Proposition 2.3
Let be a bialgebra and a vector space. Then is an L-R-admissible pair if and only if is a bialgebra in satisfying (1.14).
Proof. A straightforward verification; we only note that (1.9) expresses the fact that the comultiplication of is an algebra map inside the category .
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1[1] P. Bieliavsky, P. Bonneau, Y. Maeda, Universal deformation formulae, symplectic Lie groups and symmetric spaces, Pacific J. Math. 230 (2007), 41–57.
- 2[2] P. Bieliavsky, P. Bonneau, Y. Maeda, Universal deformation formulae for three-dimensional solvable Lie groups, in “Quantum field theory and noncommutative geometry”, 127–141, Lecture Notes in Phys., 662 , Springer, Berlin, 2005.
- 3[3] P. Bonneau, M. Gerstenhaber, A. Giaquinto, D, Sternheimer, Quantum groups and deformation quantization: Explicit approaches and implicit aspects, J. Math. Phys. 45 (2004), 3703–3741.
- 4[4] P. Bonneau, D. Sternheimer, Topological Hopf algebras, quantum groups and deformation quantization, in ”Hopf algebras in noncommutative geometry and physics”, 55–70, Lecture Notes in Pure and Appl. Math. 239 , Marcel Dekker, New York, 2005.
- 5[5] D. Bulacu, F. Panaite, F. Van Oystaeyen, Generalized diagonal crossed products and smash products for quasi-Hopf algebras. Applications, Comm. Math. Phys. 266 (2006), 355–399.
- 6[6] A. Connes, H. Moscovici, Rankin-Cohen brackets and the Hopf algebra of transverse geometry, Mosc. Math. J. 4 (2004), 111–130.
- 7[7] F. Hausser, F. Nill, Diagonal crossed products by duals of quasi-quantum groups, Rev. Math. Phys. 11 (1999), 553–629.
- 8[8] L. Kadison, Pseudo-Galois extensions and Hopf algebroids, ar Xiv:math.QA/0508411.
