perso.ens-lyon.fr/laurent.berger/
The perfectoid commutant
of Lubin-Tate power series
Abstract
Let be a Lubin-Tate formal group attached to a finite extension of . By a theorem of Lubin-Sarkis, an invertible characteristic power series that commutes with the elements is itself in . We extend this result to perfectoid power series, by lifting such a power series to characteristic zero and using the theory of locally analytic vectors in certain rings of -adic periods. This allows us to recover the field of norms of the Lubin-Tate extension from its completed perfection.
Key words and phrases:
Lubin-Tate group; field of norms; -adic period; locally analytic vector; -adic dynamical system; perfectoid field1991 Mathematics Subject Classification:
11S; 12J; 13JIntroduction
Let be a finite extension of , with ring of integers and residue field . Let and let be a uniformizer of . Let be the Lubin-Tate formal -module attached to . Let denote the extension of generated by the torsion points of , and let . The Lubin-Tate character gives rise to an isomorphism .
The field of norms ([Win83]) of the extension is a local field of characteristic , endowed with an action of , that can be explicitly described as follows. We choose a coordinate on , so that for each we get a power series . We then have , on which acts via the formula . In -adic Hodge theory, we consider the field , which is the -adic completion of the maximal purely inseparable extension of inside an algebraic closure. The action of extends to the field . If and , we still have . The question that motivated this paper is the following.
Question.
Can we recover from the data of the valued field endowed with the action of ?
If , then is an element of of valuation that satisfies the functional equation for all . Conversely, we prove the following theorem, which answers the question, as it allows us to find a uniformizer of from the data of the valued field endowed with the action of .
Theorem A.
If is such that and for all , then there exists such that .
In particular, for any as in theorem A. The main difficulty in the proof of theorem A is to prove that if is as in the statement of theorem A, then there exists such that . If and , namely in the cyclotomic situation, this follows from the main result of [BR22]. However, a crucial ingredient in that paper does not generalize to . In order to go beyond the cyclotomic case, we instead use a result of Colmez ([Col02]) to lift to an element of a ring (the Witt vectors over the ring of integers of , as well as a completion of , where ), that will satisfy a similar functional equation. In particular, is a locally analytic element of a suitable ring of -adic periods. By previous results of the author ([Ber16]), belongs to for a certain . This allows us to prove that there exists such that . By replacing with for a well chosen , we are led to the study of elements of under composition. We prove that is invertible for composition, and to conclude we use a theorem of Lubin-Sarkis ([LS07]) saying that if an invertible series commutes with a nontorsion element of , then that series is itself in . We finish this paper with an explanation of why the βTate tracesβ on used in [BR22] donβt exist if .
1. Locally analytic vectors
We use the notation that was introduced in the introduction. In order to apply lemma 9.3 of [Col02], we assume that the coordinate on is chosen such that is a monic polynomial of degree (for example, we could ask that ).
Let . Let denote the ring of integers of and let be the -Witt vectors over .
\propname \the\smf@thm.
If is such that for all , then has a lift such that for all .
Proof.
By lemma 9.3 of [Col02], there is a unique lift of such that (in ibid., this element is denoted by ). If , then both and are lifts of that are compatible with Frobenius as above. By unicity, they are equal. β
Let and be the logarithm and exponential series for . Write and for .
\lemmname \the\smf@thm.
We have for all .
Proof.
Fix such that and let . Recall that . If and , then . This implies that . Its composition inverse therefore also belongs to . This implies the claim for . The claim for follows easily. β
We use a number of rings of -adic periods in the Lubin-Tate setting, whose construction and properties were recalled in Β§3 of [Ber16]. Proposition 1 gives us an element (denoted by in ibid.). Let . Given an interval , a valuation on is constructed in ibid., as well as various completions of that ring. We use , the completion of for and . Inside , there is the ring of power series with coefficients in , where converges on a certain annulus depending on .
\lemmname \the\smf@thm.
If , then .
Proof.
Take , and let . We have , so that is bounded by on the corresponding annulus. This is true for all , so that . We now have . β
Let be a Banach space with a continuous action of . The notion of locally analytic vector was introduced in [ST03]. Recall (see for instance Β§2 of [Ber16]; the definition given there is easily seen to be equivalent to the following one) that an element is locally -analytic if there exists a sequence of such that , and an integer such that for all such that with , we have . If is a FrΓ©chet representation of , we say that is pro--analytic if its image in is locally -analytic for all .
\propname \the\smf@thm.
If and is such that and for all , then is a pro--analytic element of .
Proof.
We prove that for all , is a locally -analytic vector of . The proposition then follows, since as FrΓ©chet spaces.
Let be the power series that gives the addition in . We have . Take such that . We have , so that . If , then
We have by lemma 1, by hypothesis, and . This implies the claim. β
\propname \the\smf@thm.
If and is a pro--analytic element of , then there exists such that .
2. Composition of power series
Recall that a power series is separable if . If , we say that is invertible if , which is equivalent to being invertible for composition (denoted by ). We say that is nontorsion if for all . If and , let . Note that .
\propname \the\smf@thm.
Let be an invertible nontorsion series, and let be a separable power series. If , then is invertible.
Proof.
This is a slight generalization of lemma 6.2 of [Lub94]. Write
with . Since is nontorsion, we can replace by for and assume that . We have
If , then and we are done, so assume that . We have
This implies that , hence , which is impossible if unless . Hence and is invertible. β
We now prove theorem A. Take such that and for all . By proposition 1, has a lift such that for all . By proposition 1, is a pro--analytic element of . By proposition 1, there exists such that . This implies that . Hence there is an such that belongs to and is separable. Note that . Take such that is nontorsion, and let so that . We have . By proposition 2, is invertible. This implies that , so that and itself is invertible. Since for all , theorem 6 of [LS07] implies that . Hence there exists such that .
3. Tate traces in the Lubin-Tate setting
If and (namely in the cyclotomic situation) the fact that, in the proof of theorem A, there exists such that follows from the main result of [BR22]. We now explain why the methods of ibid donβt extend to the Lubin-Tate case. More precisely, we prove that there is no -equivariant -linear projector if . Choose a coordinate on such that , so that . Let be the invariant derivative on .
\lemmname \the\smf@thm.
We have in for all .
Proof.
Since , we have in . Applying to , we get the claim. β
\lemmname \the\smf@thm.
If is nontorsion, then .
\propname \the\smf@thm.
If , there is no -equivariant map such that for all .
Proof.
Suppose that such a map exists, and take nontorsion and such that . We first show that if is such that , then . Write where , so that and . Applying , we get . Hence so that . Since by lemma 3, this implies , so that .
However, lemma 3 and the fact that imply that for some , so that . Hence even though does not belong to . Therefore, no such map can exist. β
\coroname \the\smf@thm.
If , there is no -equivariant -linear projector . A fortiori, there is no -equivariant -linear projector .
Proof.
Given such a projector , we could define as in prop 3 by . β
Acknowledgements. I thank Juan Esteban RodrΓguez Camargo for asking me the question that motivated both this paper and [BR22].
References
- [Ber16] L. Berger β βMultivariable -modules and locally analytic vectorsβ, Duke Math. J. 165 (2016), no. 18, p. 3567β3595.
- [BR22] L. Berger & S. Rozensztajn β βDecompletion of cyclotomic perfectoid fields in positive characteristicβ, preprint, 2022.
- [Col02] P. Colmez β βEspaces de Banach de dimension finieβ, J. Inst. Math. Jussieu 1 (2002), no. 3, p. 331β439.
- [LS07] J. Lubin & G. Sarkis β βExtrinsic properties of automorphism groups of formal groupsβ, J. Algebra 315 (2007), no. 2, p. 874β884.
- [Lub94] J. Lubin β βNonarchimedean dynamical systemsβ, Compositio Math. 94 (1994), no. 3, p. 321β346.
- [ST03] P. Schneider & J. Teitelbaum β βAlgebras of -adic distributions and admissible representationsβ, Invent. Math. 153 (2003), no. 1, p. 145β196.
- [Win83] J.-P. Wintenberger β βLe corps des normes de certaines extensions infinies de corps locaux; applicationsβ, Ann. Sci. Γcole Norm. Sup. (4) 16 (1983), no. 1, p. 59β89.