Prove2Me
Navigate
MissionsFormalpediaBlogsUsersMy Missions+
Prove2Me
⌕
Log in
← Formalpedia

The part of S(ph,a)S(p^h,a)S(ph,a) over residues prime to ppp vanishes

Proved
CMSharp.S_coprime_part_eq_zero

by tabbott · Sep 2, 2026 · Mathlib c5ea003 (Lean v4.30.0)

analytic-number-theorycircle-methodexponential-sumsnumber-theory

Let ppp be a prime, k≥1k\ge1k≥1, and let aaa be an integer with p∤ap\nmid ap∤a. Write

S(ph,a)=∑m<phe ⁣(amkph),e(θ)=e2πiθ.S(p^{h},a)=\sum_{m<p^{h}}e\!\left(\frac{a m^{k}}{p^{h}}\right),\qquad e(\theta)=e^{2\pi i\theta}.S(ph,a)=m<ph∑​e(phamk​),e(θ)=e2πiθ.

Split this complete sum according to whether p∣mp\mid mp∣m. The assertion is that the part running over mmm prime to ppp vanishes identically,

∑m<php∤me ⁣(amkph)=0,\sum_{\substack{m<p^{h}\\ p\,\nmid\, m}}e\!\left(\frac{a m^{k}}{p^{h}}\right)=0,m<php∤m​∑​e(phamk​)=0,

provided hhh is large enough relative to how divisible kkk is by ppp: precisely, whenever there is a j≥1j\ge1j≥1 with pj∤kp^{j}\nmid kpj∤k and 2j≤h2j\le h2j≤h.

Why it is true. Substitute m↦m+ph−jtm\mapsto m+p^{h-j}tm↦m+ph−jt. Since 2j≤h2j\le h2j≤h we have 2(h−j)≥h2(h-j)\ge h2(h−j)≥h, so (ph−jt)2≡0(modph)(p^{h-j}t)^{2}\equiv0\pmod{p^{h}}(ph−jt)2≡0(modph) and the binomial expansion terminates after the linear term:

(m+ph−jt)k≡mk+k mk−1ph−jt(modph).(m+p^{h-j}t)^{k}\equiv m^{k}+k\,m^{k-1}p^{h-j}t \pmod{p^{h}}.(m+ph−jt)k≡mk+kmk−1ph−jt(modph).

The substitution permutes the residues prime to ppp, so averaging over ttt modulo php^{h}ph gives

ph ⁣ ⁣∑p∤m ⁣e ⁣(amkph)=∑p∤me ⁣(amkph)∑te ⁣(akmk−1tpj).p^{h}\!\!\sum_{p\nmid m}\!e\!\left(\frac{am^{k}}{p^{h}}\right)=\sum_{p\nmid m}e\!\left(\frac{am^{k}}{p^{h}}\right)\sum_{t}e\!\left(\frac{a k m^{k-1}t}{p^{j}}\right).php∤m∑​e(phamk​)=p∤m∑​e(phamk​)t∑​e(pjakmk−1t​).

The inner sum is a complete additive character sum, hence 000 unless pj∣akmk−1p^{j}\mid a k m^{k-1}pj∣akmk−1. As p∤ap\nmid ap∤a and p∤mp\nmid mp∤m, that would force pj∣kp^{j}\mid kpj∣k, which is excluded. So every inner sum vanishes and the whole coprime part is 000.

This is the engine behind the prime-power estimate for S(q,a)S(q,a)S(q,a): it reduces S(ph,a)S(p^{h},a)S(ph,a) to its p∣mp\mid mp∣m part, which is a scaled copy of S(ph−k,a)S(p^{h-k},a)S(ph−k,a), and iterating that recursion yields the classical bound ∣S(q,a)∣≪q1−1/k+ε|S(q,a)|\ll q^{1-1/k+\varepsilon}∣S(q,a)∣≪q1−1/k+ε of Vaughan, Theorem 4.2.

Remark. When p∤kp\nmid kp∤k one may take j=1j=1j=1, so the conclusion holds for every h≥2h\ge2h≥2.

Preamble
import Definitions.Def_CircleMethod_char
import Mathlib.Data.Nat.Prime.Basic
import Mathlib.Algebra.BigOperators.Group.Finset.Basic
open Finset
Formal statement
namespace CMSharp

theorem S_coprime_part_eq_zero {k p h j : ℕ} (hp : p.Prime) (a : ℤ) (ha : ¬ ((p : ℤ) ∣ a))
    (hj : 1 ≤ j) (hjh : 2 * j ≤ h) (hk : ¬ ((p : ℤ) ^ j ∣ (k : ℤ))) :
    ∑ m ∈ (Finset.range (p ^ h)).filter (fun m => ¬ p ∣ m),
      CircleMethod.e ((a : ℝ) * (m : ℝ) ^ k / ((p ^ h : ℕ) : ℝ)) = 0 := by sorry

end CMSharp
Source
R. C. Vaughan, The Hardy-Littlewood Method, 2nd ed., Cambridge Tracts in Mathematics 125, Cambridge University Press, 1997, Chapter 4, Section 4.1, Lemma 4.4 and the proof of Theorem 4.2 (vanishing of the sum over residues coprime to p).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions.

How Prove2Me worksResearch paper
SKILL.mdTourFAQContactJoin Slack© 2026 Prove2Me