p-adic Lie group and the p-adic exponential
Anchor (Master): Lazard 1965 Publ. Math. IHES; Bourbaki 1972; Serre 1965; Dixon-du Sautoy-Mann-Segal Analytic Pro-p Groups
Intuition [Beginner]
The real exponential map sends addition to multiplication. In the world of -adic numbers (where distances are measured by divisibility by a prime ), there is an analogous exponential: a power series that sends -adic addition to -adic multiplication. This -adic exponential is the foundation of -adic Lie group theory.
A -adic Lie group is a group that is also a -adic manifold, just as an ordinary Lie group is a group that is also a smooth manifold. The -adic exponential map provides coordinates near the identity, turning the group into a -adic version of a vector space (the Lie algebra).
The key difference from the real case is convergence. The -adic exponential only converges when , because the denominators have large -adic valuation. This restricted domain shapes the entire theory: the exponential is only locally defined, and the group structure it captures is inherently -adic analytic.
Visual [Beginner]
A number line labelled with -adic distances (small near 0, growing by powers of ). A small disc around the origin is highlighted as the "convergence disc" of the -adic exponential. Arrows show the exponential mapping the additive disc to a multiplicative disc inside the -adic group, with the logarithm as the reverse arrow.
The -adic exponential maps a small additive disc to a multiplicative disc inside the -adic group.
Worked example [Beginner]
The -adic exponential for . Consider the 2-adic numbers . The exponential series is .
Step 1. For , the convergence condition requires . So must be divisible by 2.
Step 2. Take . Then , which is on the boundary. The series converges because the -adic valuation of grows as increases.
Step 3. Compute . In 2-adic arithmetic, the partial sums converge to a well-defined 2-adic number. The result is a 2-adic unit (a number with ).
What this tells us: the -adic exponential works, but only inside a disc of controlled -adic radius. The group law it produces is the formal group law of the Lie group, evaluated in the convergent region.
Check your understanding [Beginner]
Formal definition [Intermediate+]
Let be a prime. A -adic analytic manifold of dimension is a topological space locally modelled on with transition functions given by convergent -adic power series.
Definition (-adic Lie group). A -adic Lie group is a group equipped with the structure of a -adic analytic manifold such that the multiplication map and the inversion map are -adic analytic.
Definition (-adic exponential). The -adic exponential is the power series
It converges in for . The -adic logarithm is
converging for . On their common domain, and are inverse functions.
Definition (-adic Lie algebra). The Lie algebra of a -adic Lie group is the tangent space at the identity , equipped with the Lie bracket defined by the -adic Baker-Campbell-Hausdorff formula: .
Counterexamples to common slips
- The -adic exponential converges everywhere. False. The -adic valuation (where is the digit sum of in base ) grows linearly, so the convergence radius is , strictly less than 1.
- Every -adic Lie group is a matrix group. False in general. Over , the Ado-Iwasawa theorem gives a local embedding into , but the global structure can be more complicated.
- The -adic BCH series is the same as the real BCH series. True as a formal power series. The difference is in the convergence domain: the -adic BCH converges on a smaller disc than the real one.
Key theorem with proof [Intermediate+]
Theorem (-adic Lie group-Lie algebra correspondence). Let be a -adic Lie group with Lie algebra . The -adic exponential map (defined in coordinates via the power series) is a local -adic analytic isomorphism from a neighbourhood of in to a neighbourhood of in . The group law in exponential coordinates is given by the -adic BCH formula: .
Proof sketch. The -adic exponential is defined by the same power series as in the real case. The derivative at zero is the identity map, so the -adic inverse function theorem guarantees a local analytic inverse (the -adic logarithm). The group multiplication, expressed in exponential coordinates, is analytic and agrees with the BCH series to all orders because both are determined by the same formal identities.
The key technical step is the convergence of BCH in the -adic setting. Since BCH involves denominators , the -adic valuation of the coefficients must be tracked. The -adic BCH converges on the domain because the -adic valuation of the denominators grows fast enough.
Bridge. This theorem builds toward Lazard's characterisation of -adic analytic groups via uniform pro- groups in the Master section, and the foundational reason the correspondence works is that the formal group law 03.03.04 of the Lie group, evaluated in the convergent region of the -adic exponential, produces the -adic BCH formula. This is exactly the mechanism by which the algebraic formal group law theory connects to the analytic -adic group theory, and the bridge is that the -adic exponential converts the formal group law into a convergent analytic group law. The result pairs with the real Lie group-Lie algebra correspondence where the same BCH formula appears, but the -adic convergence analysis is different due to the ultrametric inequality, and this leads to the Lazard theory of -adic analytic groups 03.03.06.
Exercises [Intermediate+]
Advanced results [Master]
Theorem 1 (Lazard's theorem on -adic analytic groups). A topological group is -adic analytic if and only if it contains an open subgroup which is a uniform pro- group of finite rank. The rank equals the dimension of the -adic Lie group.
Theorem 2 (Powerful -groups and analyticity). A finitely generated pro- group is -adic analytic if and only if is powerful: (for odd ) or (for ). This algebraic criterion detects -adic analyticity without reference to manifolds.
Theorem 3 (-adic Ado theorem). Every finite-dimensional Lie algebra over has a faithful finite-dimensional representation. Consequently, every -adic Lie group locally embeds into for some .
Theorem 4 (Iwasawa decomposition for -adic groups). A semisimple -adic Lie group admits an Iwasawa decomposition where is a maximal compact subgroup, is a split torus, and is a unipotent group. This mirrors the real Iwasawa decomposition but with -adic components.
Theorem 5 (Mahler expansion). Every continuous function has a unique Mahler expansion with in . This is the -adic analogue of Taylor expansion and provides the analytic toolkit for -adic Lie groups.
Synthesis. The -adic exponential is the foundational reason that -adic Lie groups admit a Lie theory parallel to the real case; the central insight is that the formal group law 03.03.04 of the Lie group, when evaluated on the convergent disc of the -adic exponential, produces an analytic group law whose algebraic structure is captured by the Lie bracket. Putting these together with Lazard's theorem, the -adic analytic groups are precisely the topological groups containing a uniform pro- subgroup, which is exactly the bridge between the topological and algebraic viewpoints on -adic group theory. This pattern recurs throughout the subject, appearing in the powerful -group criterion for analyticity, the -adic BCH formula that converts Lie algebra data into group data, and the connection to the formal group law height 03.03.04 which governs the -adic structure through the -series. The result also connects to number theory via the Galois representations attached to -adic Lie groups by -adic Hodge theory, where the Lie algebra of the Galois group controls the arithmetic of the field extension.
Full proof set [Master]
Proposition (Convergence of the -adic exponential). The -adic exponential series converges for .
Proof. By Legendre's formula, . For with :
Since and is bounded, , so .
Proposition (-adic logarithm as inverse). The series and are inverse functions on their common domain of convergence.
Proof. Both series satisfy the formal identity and as identities in . Since both sides converge -adically on the common domain (where and ), the formal identity becomes an analytic identity.
Connections [Master]
Formal group law
03.03.04. The -adic exponential evaluates the formal group law of a Lie group in the convergent -adic regime. The height of the formal group law03.03.04controls the -adic structure of the group through the -series, and the -adic BCH formula is the convergent realisation of the formal group law.Lie groups and Lie algebras
03.03.01. The -adic Lie group-Lie algebra correspondence parallels the real correspondence from03.03.01, but the -adic exponential has a restricted convergence domain. The Lie algebra structure (bracket, BCH formula) is formally identical; only the analytic properties differ.Lie's third theorem
03.03.06. Lie's third theorem03.03.06asserts that every Lie algebra integrates to a Lie group in the simply-connected case. The -adic version uses the -adic exponential and BCH formula to construct the group law on the Lie algebra disc, and the convergence analysis provides the integration in the -adic setting.-extensions and Iwasawa theory
21.07.01. Number-theoretic application unit. The Galois group of the cyclotomic -extension is the simplest non-discrete -adic Lie group: a -dimensional pro- group whose Lie algebra is . The Iwasawa algebra is the completed group algebra of this -dimensional -adic Lie group, providing the algebraic carrier on which Iwasawa-theoretic class-group and Selmer-group invariants live. The non-commutative Iwasawa theory of Coates-Fukaya-Kato-Sujatha-Venjakob 2005 extends this picture to higher-dimensional pro- Lie groups arising from non-abelian -towers, with the -adic Lie group structure of the present unit as the foundational substrate.
Historical & philosophical context [Master]
The theory of -adic Lie groups was created by Michel Lazard in his monumental 1965 paper "Groupes analytiques p-adiques" [Lazard 1965], published in Publications Mathematiques de l'IHES. Lazard established the equivalence between -adic analytic groups, uniform pro- groups, and Lie algebras over , creating a complete parallel to the classical theory of real Lie groups.
Jean-Pierre Serre systematised the theory in his 1965 lectures "Lie Algebras and Lie Groups" [Serre 1965], emphasising the role of the formal group law and the BCH formula as the bridge between the algebraic and analytic viewpoints.
The modern treatment via powerful pro- groups is due to Lubotzky and Mann (1987), and the definitive reference is Dixon-du Sautoy-Mann-Segal "Analytic Pro-p Groups" (1999) [Dixon et al. 1999], which gives a purely group-theoretic characterisation of -adic analytic groups without any reference to manifolds.
Bibliography [Master]
@article{lazard1965,
author = {Lazard, Michel},
title = {Groupes analytiques $p$-adiques},
journal = {Publ. Math. Inst. Hautes \'Etudes Sci.},
volume = {26},
year = {1965}
}
@book{bourbaki1972,
author = {Bourbaki, Nicolas},
title = {Lie Groups and Lie Algebras, Ch. 4},
publisher = {Hermann},
year = {1972}
}
@book{serre1965,
author = {Serre, Jean-Pierre},
title = {Lie Algebras and Lie Groups},
publisher = {Benjamin},
year = {1965}
}
@book{dixon1999,
author = {Dixon, J. D. and du Sautoy, M. P. F. and Mann, A. and Segal, D.},
title = {Analytic Pro-$p$ Groups},
edition = {2},
publisher = {Cambridge Univ. Press},
year = {1999}
}