Definition

Let kk be a . A kk-bialgebra is a unital associative kk- (H,m,u)(H,m,u) and a (H,Δ,ε)(H,\Delta,\varepsilon) on the same module such that Δ:HHkH\Delta:H\to H\otimes_k H and ε:Hk\varepsilon:H\to k are unital algebra homomorphisms. The tensor-product algebra structure used here is

(ab)(cd)=acbd.(a\otimes b)(c\otimes d)=ac\otimes bd.

Equivalently, mm and uu are coalgebra homomorphisms.

Compatibility in formulas

The algebra-homomorphism formulation means

Δ(ab)=Δ(a)Δ(b),Δ(1H)=1H1H,\Delta(ab)=\Delta(a)\Delta(b),\qquad \Delta(1_H)=1_H\otimes1_H,

and

ε(ab)=ε(a)ε(b),ε(1H)=1k.\varepsilon(ab)=\varepsilon(a)\varepsilon(b),\qquad \varepsilon(1_H)=1_k.

Together with coassociativity and the counit identities, these conditions say that HH is simultaneously a monoid and a comonoid, with each structure compatible with the other, in the of kk-modules.

Morphisms

A bialgebra homomorphism is a unital that is also a coalgebra homomorphism. Bialgebras may be commutative as algebras, cocommutative as coalgebras, both, or neither; these are independent conditions.

Examples
  • For a GG, the k[G]k[G] has Δ(g)=gg\Delta(g)=g\otimes g and ε(g)=1\varepsilon(g)=1, extended linearly and multiplicatively. It is cocommutative, but it is commutative exactly when GG is abelian.
  • The polynomial algebra k[x]k[x] becomes a commutative and cocommutative bialgebra with Δ(x)=x1+1x\Delta(x)=x\otimes1+1\otimes x and ε(x)=0\varepsilon(x)=0.
Why an antipode is extra

A bialgebra encodes multiplication and comultiplication but need not have an operation representing inversion. A bialgebra equipped with such a convolution inverse is a .

References
  1. Christian Kassel, Quantum Groups, Graduate Texts in Mathematics 155, Springer, 1995. Publisher record. Relevant: Chapter III, §§1–2.
  2. Susan Montgomery, Hopf Algebras and Their Actions on Rings, CBMS Regional Conference Series 82, American Mathematical Society, 1993. Publisher record. Relevant: Chapter 1.