Kummer's Criterion and Regular Primes in Lean

7.6. Bernoulli non-divisibility and the plus index🔗

The previous sections combine into the theorem used by the final proof.

Theorem7.6.1
Statement uses 3
Statement dependency previews
Preview
Theorem 7.4.1
Loading preview
Statement dependency preview content is loaded from the Blueprint HTML cache.
used by 1L∃∀N

Bernoulli non-divisibility implies index non-divisibility. Assume p is odd and \forall j,\quad 1\le j,\ 2j\le p-3 \Longrightarrow p\nmid (B_{2j})_{\mathrm{num}}. Then p\nmid [E^+:C^+_{\mathrm{sq}}].

Theorem 7.5.1 Theorem 7.5.2 Theorem 7.4.1

Lean code for Theorem7.6.11 theorem
  • theoremdefined in BernoulliRegular/CyclotomicUnits/UnitsReflection.lean
    complete
    theorem BernoulliRegular.not_dvd_cyclotomicUnitIndex_of_bernoulli_nonzero
      {p : } [Fact (Nat.Prime p)] {K : Type} [Field K] [NumberField K]
      [IsCyclotomicExtension {p}  K] [NumberField.IsCMField K]
      (hp_odd : p  2)
      (hB :
         (j : ), 1  j  2 * j  p - 3  ¬p  (bernoulli (2 * j)).num) :
      ¬p  (BernoulliRegular.CPlus ).index
    theorem BernoulliRegular.not_dvd_cyclotomicUnitIndex_of_bernoulli_nonzero
      {p : } [Fact (Nat.Prime p)] {K : Type}
      [Field K] [NumberField K]
      [IsCyclotomicExtension {p}  K]
      [NumberField.IsCMField K]
      (hp_odd : p  2)
      (hB :
         (j : ),
          1  j 
            2 * j  p - 3 
              ¬p  (bernoulli (2 * j)).num) :
      ¬p  (BernoulliRegular.CPlus ).index
    CU-16: Bernoulli numerator nonvanishing in the classical range implies
    that `p` does not divide the real cyclotomic-unit index. 
Proof for Theorem 7.6.1
uses 0

If p=3, then the generator range is empty. The real subfield of \mathbb{Q}(\zeta_3) is \mathbb{Q}, so its unit group is generated by -1. Hence C^+_{\mathrm{sq}}=E^+ and the index is 1, which is not divisible by p. Now assume p\ge5. The Bernoulli non-divisibility hypothesis makes the Kummer logarithm determinant nonzero by Theorem 7.5.1. The determinant non-vanishing theorem Theorem 7.5.2 gives p-saturation of C^+_{\mathrm{sq}} in E^+. Finally Theorem 7.4.1 converts this saturation into p\nmid [E^+:C^+_{\mathrm{sq}}].