# Miura transform for W-algebras of exceptional type

Miura transform for W-algebras of classical types can be found in e.g. Sec. 6.3.3 of Bouwknegt-Schoutens. Is there a similar explicit Miura transform for W-algebras of exceptional types, say, E6? It's 20 years since the review by B-S, so I'd hope somebody worked this out ...

-

## 1 Answer

Yes, for the "quasi-classical" case(i.e. for the case when the $W$-algebra is commutative, which occurs when the level is either infinite or critical) it was defined by Drinfeld and Sokolov long time ago; you can look at Section 4 of http://arxiv.org/PS_cache/math/pdf/0305/0305216v1.pdf for a good review.

For the "quantum" case (i.e. for arbitrary level) it was studied by Feigin and Frenkel, but I am not sure what the right reference is; you can look for example at Section 4 of http://arxiv.org/PS_cache/hep-th/pdf/9408/9408109v1.pdf, but there should be more modern references. In fact, the main tool in the work of Feigin and Frenkel is the screening operators, which describe the $W$-algebra explicitly as a subalgebra of (the vertex operator algebra associated to) the Heisenberg algebra (where the embedding to the Heisenberg algebra is the Miura transformation).

-
Thank you, but my main problem is to explicitly write down the subalgebra commuting with the screening operators. For A and D, it's done by Fateev-Zamolodchikov and Fateev-Lukyanov. Their forms are quite useful because it can be readily implemented in a computer algebra system. I just want to perform a few stupid calculation inside W-algebra of type E6, but I first need to realize it inside computer. –  Yuji Oct 26 '11 at 4:26
Since I don't believe in explicit formulas, I won't be able to say anything intelligent here:) One remark, though: you can describe the image of the W-algebra without the screening operators. It is just equal to the intersection over all simple roots of things like Virasoro$\otimes$Heisenberg of smaller rank (I hope it is clear what I mean) –  Alexander Braverman Oct 26 '11 at 5:48
Yes you're right. Physicists cover their lack of deep thinking by lots of explicit calculation:p I've been using that approach to find generators of W(E6), but that's still quite messy. That's why I asked the question here. –  Yuji Oct 26 '11 at 6:01
Do you want just generators, or generators and relations? –  Alexander Braverman Oct 26 '11 at 6:07
I think he wants the fields for each exponent of $E6$ together with their OPE. I don't think you'll find those Yuji, at least at the principal nilpotent. In the case of the minimal nilpotent, Kac and Wakimoto have explicit formulas in this paper –  Reimundo Heluani Oct 26 '11 at 9:43