/-
Copyright 2026 The Formal Conjectures Authors.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
https://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-/
module
public import FormalConjecturesUtilIs a finite locally free group scheme killed by its order?
A finite locally free group scheme of rank n is the group scheme analogue of a finite group of
order n. Lagrange's theorem gives g ^ n = 1 in a finite group of order n, and Grothendieck
asked whether the n-th power map of such a group scheme is trivial. The answer is affirmative
over a reduced base and for commutative group schemes, and negative in general. Which exponents
kill every group scheme of a given rank is open.
A group scheme is represented here by its coordinate Hopf algebra A over the base ring R.
A is not assumed to be cocommutative, so the group scheme is not assumed to be commutative.
The base is a ring rather than a scheme: a finite locally free group scheme is affine over its
base, and being killed by m is local on the base, so nothing is lost.
References:
M. Demazure and A. Grothendieck, Schémas en groupes (SGA 3), re-edition by P. Gille and P. Polo. Exp. VIII, Rem. 7.3.1 poses the question; an editors' note records that the re-edition drops the commutativity assumption there. Exp. VIIA, Prop. 8.5 settles the case of a field and Cor. 8.5.2 the case of a reduced base.
R. Schoof, Is a finite locally free group scheme killed by its order?, a survey of the question and of the cases that are known, written before the counterexample
J. Tate and F. Oort, Group schemes of prime order, p. 4 for Deligne's theorem and p. 5, which records the general question as open in 1970
Mathlib, A finite free group scheme of rank four that is not killed by four, the counterexample
@[expose] public sectionnamespace FiniteLocallyFreeGroupSchemeExponentopen MonoidAlgebra WithConvuniverse u v
The group scheme Spec A of a commutative Hopf algebra A over R is killed by m when
the m-th convolution power of AlgHom.id R A is the convolution unit.
The convolution power of AlgHom.id R A is the coordinate map of the pointwise power map
x ↦ x ^ m, and 1 is the constant map at the identity section.
def IsKilledBy (R : Type u) [CommRing R] (A : Type v) [CommRing A] [HopfAlgebra R A]
(m : ℕ) : Prop :=
toConv (AlgHom.id R A) ^ m = 1
A positive integer m is a universal killing exponent in rank n if every finite locally
free group scheme of rank n, over every commutative ring, is killed by m.
The hypotheses that A is finite and projective over R, with constant rankAtStalk equal to
n, express that the group scheme is finite locally free of rank n.
def IsUniversalKillingExponent (n m : ℕ+) : Prop :=
∀ (R : Type u) [CommRing R] (A : Type v) [CommRing A] [HopfAlgebra R A]
[Module.Finite R A] [Module.Projective R A],
Module.rankAtStalk (R := R) A = (fun _ ↦ (n : ℕ)) → IsKilledBy R A m
Unfolding IsKilledBy at m = 1: the group scheme is trivial exactly when the identity of
A is the composite of the counit with the unit.
R:Type uinst✝²:CommRing RA:Type vinst✝¹:CommRing Ainst✝:HopfAlgebra R A⊢ toConv (AlgHom.id R A) = toConv ((Algebra.ofId R A).comp (Bialgebra.counitAlgHom R A)) ↔
AlgHom.id R A = (Algebra.ofId R A).comp (Bialgebra.counitAlgHom R A)
exact ⟨fun h ↦ WithConv.toConv_injective h, fun h ↦ congrArg _ h⟩ All goals completed! 🐙
On a group algebra the m-th convolution power of the identity is the pointwise m-th power:
it sends the group-like element single g 1 to single (g ^ m) 1.
@[category API, AMS 14 16]
theorem pow_apply_single (R : Type u) [CommRing R] (G : Type v) [CommGroup G] (m : ℕ) (g : G) :
(toConv (AlgHom.id R (MonoidAlgebra R G)) ^ m).ofConv (single g 1) = single (g ^ m) 1 := by R:Type uinst✝¹:CommRing RG:Type vinst✝:CommGroup Gm:ℕg:G⊢ (toConv (AlgHom.id R R[G]) ^ m).ofConv (single g 1) = single (g ^ m) 1
induction m with
| zero => zero R:Type uinst✝¹:CommRing RG:Type vinst✝:CommGroup Gg:G⊢ (toConv (AlgHom.id R R[G]) ^ 0).ofConv (single g 1) = single (g ^ 0) 1
simp [AlgHom.convOne_def, Algebra.ofId_apply, Algebra.algebraMap_eq_smul_one,
MonoidAlgebra.one_def] All goals completed! 🐙
| succ k ih => succ R:Type uinst✝¹:CommRing RG:Type vinst✝:CommGroup Gg:Gk:ℕih:(toConv (AlgHom.id R R[G]) ^ k).ofConv (single g 1) = single (g ^ k) 1⊢ (toConv (AlgHom.id R R[G]) ^ (k + 1)).ofConv (single g 1) = single (g ^ (k + 1)) 1
rw [pow_succ, succ R:Type uinst✝¹:CommRing RG:Type vinst✝:CommGroup Gg:Gk:ℕih:(toConv (AlgHom.id R R[G]) ^ k).ofConv (single g 1) = single (g ^ k) 1⊢ (toConv (AlgHom.id R R[G]) ^ k * toConv (AlgHom.id R R[G])).ofConv (single g 1) = single (g ^ (k + 1)) 1 succ R:Type uinst✝¹:CommRing RG:Type vinst✝:CommGroup Gg:Gk:ℕih:(toConv (AlgHom.id R R[G]) ^ k).ofConv (single g 1) = single (g ^ k) 1⊢ (Algebra.TensorProduct.lift (toConv (AlgHom.id R R[G]) ^ k).ofConv (toConv (AlgHom.id R R[G])).ofConv ⋯)
((TensorProduct.map (lsingle g) (lsingle g)) (CoalgebraStruct.comul 1)) =
single (g ^ (k + 1)) 1 AlgHom.convMul_apply, succ R:Type uinst✝¹:CommRing RG:Type vinst✝:CommGroup Gg:Gk:ℕih:(toConv (AlgHom.id R R[G]) ^ k).ofConv (single g 1) = single (g ^ k) 1⊢ (Algebra.TensorProduct.lift (toConv (AlgHom.id R R[G]) ^ k).ofConv (toConv (AlgHom.id R R[G])).ofConv ⋯)
(CoalgebraStruct.comul (single g 1)) =
single (g ^ (k + 1)) 1 succ R:Type uinst✝¹:CommRing RG:Type vinst✝:CommGroup Gg:Gk:ℕih:(toConv (AlgHom.id R R[G]) ^ k).ofConv (single g 1) = single (g ^ k) 1⊢ (Algebra.TensorProduct.lift (toConv (AlgHom.id R R[G]) ^ k).ofConv (toConv (AlgHom.id R R[G])).ofConv ⋯)
((TensorProduct.map (lsingle g) (lsingle g)) (CoalgebraStruct.comul 1)) =
single (g ^ (k + 1)) 1 MonoidAlgebra.comul_single succ R:Type uinst✝¹:CommRing RG:Type vinst✝:CommGroup Gg:Gk:ℕih:(toConv (AlgHom.id R R[G]) ^ k).ofConv (single g 1) = single (g ^ k) 1⊢ (Algebra.TensorProduct.lift (toConv (AlgHom.id R R[G]) ^ k).ofConv (toConv (AlgHom.id R R[G])).ofConv ⋯)
((TensorProduct.map (lsingle g) (lsingle g)) (CoalgebraStruct.comul 1)) =
single (g ^ (k + 1)) 1succ R:Type uinst✝¹:CommRing RG:Type vinst✝:CommGroup Gg:Gk:ℕih:(toConv (AlgHom.id R R[G]) ^ k).ofConv (single g 1) = single (g ^ k) 1⊢ (Algebra.TensorProduct.lift (toConv (AlgHom.id R R[G]) ^ k).ofConv (toConv (AlgHom.id R R[G])).ofConv ⋯)
((TensorProduct.map (lsingle g) (lsingle g)) (CoalgebraStruct.comul 1)) =
single (g ^ (k + 1)) 1]succ R:Type uinst✝¹:CommRing RG:Type vinst✝:CommGroup Gg:Gk:ℕih:(toConv (AlgHom.id R R[G]) ^ k).ofConv (single g 1) = single (g ^ k) 1⊢ (Algebra.TensorProduct.lift (toConv (AlgHom.id R R[G]) ^ k).ofConv (toConv (AlgHom.id R R[G])).ofConv ⋯)
((TensorProduct.map (lsingle g) (lsingle g)) (CoalgebraStruct.comul 1)) =
single (g ^ (k + 1)) 1
simp [Coalgebra.comul, ih, single_mul_single, pow_succ] All goals completed! 🐙
Every group scheme is killed by the exponent 0, which is why the exponents in
IsUniversalKillingExponent are positive.
@[category test, AMS 14 16]
theorem isKilledBy_zero (R : Type u) [CommRing R] (A : Type v) [CommRing A] [HopfAlgebra R A] :
IsKilledBy R A 0 :=
pow_zero _
IsKilledBy is not vacuous: the group scheme μ₂ = Spec ℚ[ℤ/2] is not killed by 1.
@[category test, AMS 14 16]
theorem not_isKilledBy_one :
¬ IsKilledBy ℚ (MonoidAlgebra ℚ (Multiplicative (ZMod 2))) 1 := by ⊢ ¬IsKilledBy ℚ ℚ[Multiplicative (ZMod 2)] 1
rw [isKilledBy_one_iff ⊢ ¬AlgHom.id ℚ ℚ[Multiplicative (ZMod 2)] =
(Algebra.ofId ℚ ℚ[Multiplicative (ZMod 2)]).comp (Bialgebra.counitAlgHom ℚ ℚ[Multiplicative (ZMod 2)]) ⊢ ¬AlgHom.id ℚ ℚ[Multiplicative (ZMod 2)] =
(Algebra.ofId ℚ ℚ[Multiplicative (ZMod 2)]).comp (Bialgebra.counitAlgHom ℚ ℚ[Multiplicative (ZMod 2)])] ⊢ ¬AlgHom.id ℚ ℚ[Multiplicative (ZMod 2)] =
(Algebra.ofId ℚ ℚ[Multiplicative (ZMod 2)]).comp (Bialgebra.counitAlgHom ℚ ℚ[Multiplicative (ZMod 2)])
intro h h:AlgHom.id ℚ ℚ[Multiplicative (ZMod 2)] =
(Algebra.ofId ℚ ℚ[Multiplicative (ZMod 2)]).comp (Bialgebra.counitAlgHom ℚ ℚ[Multiplicative (ZMod 2)])⊢ False
have h2 := DFunLike.congr_fun h (single (Multiplicative.ofAdd 1) (1 : ℚ)) h:AlgHom.id ℚ ℚ[Multiplicative (ZMod 2)] =
(Algebra.ofId ℚ ℚ[Multiplicative (ZMod 2)]).comp (Bialgebra.counitAlgHom ℚ ℚ[Multiplicative (ZMod 2)])h2:(AlgHom.id ℚ ℚ[Multiplicative (ZMod 2)]) (single (Multiplicative.ofAdd 1) 1) =
((Algebra.ofId ℚ ℚ[Multiplicative (ZMod 2)]).comp (Bialgebra.counitAlgHom ℚ ℚ[Multiplicative (ZMod 2)]))
(single (Multiplicative.ofAdd 1) 1)⊢ False
simp [MonoidAlgebra.single_left_inj (one_ne_zero (α := ℚ))] at h2 All goals completed! 🐙
IsKilledBy is not always false either: the diagonalisable group scheme μₙ = Spec R[ℤ/n]
is killed by n, in accordance with Deligne's theorem.
@[category test, AMS 14 16]
theorem isKilledBy_monoidAlgebra (R : Type u) [CommRing R] (n : ℕ) [NeZero n] :
IsKilledBy R (MonoidAlgebra R (Multiplicative (ZMod n))) n := by R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero n⊢ IsKilledBy R R[Multiplicative (ZMod n)] n
refine WithConv.ext (MonoidAlgebra.algHom_ext (fun g ↦ ?_) (by R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero n⊢ (toConv (AlgHom.id R R[Multiplicative (ZMod n)]) ^ n).ofConv.comp singleOneAlgHom = (ofConv 1).comp singleOneAlgHom ext All goals completed! 🐙))
have hg : g ^ n = 1 := by R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero n⊢ IsKilledBy R R[Multiplicative (ZMod n)] n R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero ng:Multiplicative (ZMod n)hg:g ^ n = 1⊢ (toConv (AlgHom.id R R[Multiplicative (ZMod n)]) ^ n).ofConv (single g 1) = (ofConv 1) (single g 1)
have := pow_card_eq_one (G := Multiplicative (ZMod n)) (x := g) R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero ng:Multiplicative (ZMod n)this:g ^ Fintype.card (Multiplicative (ZMod n)) = 1⊢ g ^ n = 1 R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero ng:Multiplicative (ZMod n)hg:g ^ n = 1⊢ (toConv (AlgHom.id R R[Multiplicative (ZMod n)]) ^ n).ofConv (single g 1) = (ofConv 1) (single g 1)
simpa [ZMod.card] using this R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero ng:Multiplicative (ZMod n)hg:g ^ n = 1⊢ (toConv (AlgHom.id R R[Multiplicative (ZMod n)]) ^ n).ofConv (single g 1) = (ofConv 1) (single g 1) R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero ng:Multiplicative (ZMod n)hg:g ^ n = 1⊢ (toConv (AlgHom.id R R[Multiplicative (ZMod n)]) ^ n).ofConv (single g 1) = (ofConv 1) (single g 1)
show (toConv (AlgHom.id R (MonoidAlgebra R (Multiplicative (ZMod n)))) ^ n).ofConv
(single g 1) = _ R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero ng:Multiplicative (ZMod n)hg:g ^ n = 1⊢ (toConv (AlgHom.id R R[Multiplicative (ZMod n)]) ^ n).ofConv (single g 1) = (ofConv 1) (single g 1)
rw [pow_apply_single, R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero ng:Multiplicative (ZMod n)hg:g ^ n = 1⊢ single (g ^ n) 1 = (ofConv 1) (single g 1) R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero ng:Multiplicative (ZMod n)hg:g ^ n = 1⊢ single 1 1 = (ofConv 1) (single g 1) hg R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero ng:Multiplicative (ZMod n)hg:g ^ n = 1⊢ single 1 1 = (ofConv 1) (single g 1) R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero ng:Multiplicative (ZMod n)hg:g ^ n = 1⊢ single 1 1 = (ofConv 1) (single g 1)] R:Type uinst✝¹:CommRing Rn:ℕinst✝:NeZero ng:Multiplicative (ZMod n)hg:g ^ n = 1⊢ single 1 1 = (ofConv 1) (single g 1)
simp [AlgHom.convOne_def, Algebra.ofId_apply, Algebra.algebraMap_eq_smul_one,
MonoidAlgebra.one_def] All goals completed! 🐙
The group scheme μₙ = Spec R[ℤ/n] has rank n.
@[category test, AMS 14 16]
theorem rankAtStalk_monoidAlgebra (R : Type u) [CommRing R] [Nontrivial R] (n : ℕ) [NeZero n] :
Module.rankAtStalk (R := R) (MonoidAlgebra R (Multiplicative (ZMod n))) = fun _ ↦ n := by R:Type uinst✝²:CommRing Rinst✝¹:Nontrivial Rn:ℕinst✝:NeZero n⊢ Module.rankAtStalk R[Multiplicative (ZMod n)] = fun x ↦ n
ext p R:Type uinst✝²:CommRing Rinst✝¹:Nontrivial Rn:ℕinst✝:NeZero np:PrimeSpectrum R⊢ Module.rankAtStalk R[Multiplicative (ZMod n)] p = n
rw [Module.rankAtStalk_eq_finrank_of_free, R:Type uinst✝²:CommRing Rinst✝¹:Nontrivial Rn:ℕinst✝:NeZero np:PrimeSpectrum R⊢ ↑(Module.finrank R R[Multiplicative (ZMod n)]) p = n R:Type uinst✝²:CommRing Rinst✝¹:Nontrivial Rn:ℕinst✝:NeZero np:PrimeSpectrum R⊢ ↑(Fintype.card (Multiplicative (ZMod n))) p = n
Module.finrank_eq_card_basis (MonoidAlgebra.basis (Multiplicative (ZMod n)) R) R:Type uinst✝²:CommRing Rinst✝¹:Nontrivial Rn:ℕinst✝:NeZero np:PrimeSpectrum R⊢ ↑(Fintype.card (Multiplicative (ZMod n))) p = n R:Type uinst✝²:CommRing Rinst✝¹:Nontrivial Rn:ℕinst✝:NeZero np:PrimeSpectrum R⊢ ↑(Fintype.card (Multiplicative (ZMod n))) p = n] R:Type uinst✝²:CommRing Rinst✝¹:Nontrivial Rn:ℕinst✝:NeZero np:PrimeSpectrum R⊢ ↑(Fintype.card (Multiplicative (ZMod n))) p = n
simp All goals completed! 🐙
The hypotheses of IsUniversalKillingExponent are satisfiable in every rank n: they apply
to μₙ = Spec R[ℤ/n] over any nontrivial ring.
@[category test, AMS 14 16]
theorem isKilledBy_of_isUniversalKillingExponent (n m : ℕ+)
(h : IsUniversalKillingExponent.{u, u} n m) (R : Type u) [CommRing R] [Nontrivial R] :
IsKilledBy R (MonoidAlgebra R (Multiplicative (ZMod (n : ℕ)))) (m : ℕ) :=
haveI : NeZero (n : ℕ) := ⟨n.ne_zero⟩
h R _ (rankAtStalk_monoidAlgebra R n)set_option linter.style.formal_proof_palomar false inGrothendieck's question: is every finite locally free group scheme of rank $n$, over every commutative ring, killed by $n$?
The answer is no. There is a commutative Hopf algebra of rank $4$ over a finite non-reduced ring
whose fourth power map is not trivial; its eighth power map is trivial. The formalised
counterexample lives in Type, so the statement quantifies over rings and algebras in Type.
@[category research solved, question, AMS 14 16, formal_proof using lean4 at
"https://github.com/leanprover-community/mathlib4/blob/0df444a360eaa60ab8c11dca51a86af692955474/Counterexamples/GrothendieckPower.lean#L882-L892"]
theorem killed_by_rank :
¬ ∀ n : ℕ+, IsUniversalKillingExponent.{0, 0} n n := by ⊢ ¬∀ (n : ℕ+), IsUniversalKillingExponent n n
sorry All goals completed! 🐙Over a field the answer is affirmative: a finite group scheme of rank $n$ over a field is killed by $n$. This is [SGA 3], Exp. VIIA, Prop. 8.5.
@[category research solved, AMS 14 16]
theorem killed_by_rank.variants.field (n : ℕ+) (k : Type*) [Field k] (A : Type*) [CommRing A]
[HopfAlgebra k A] [Module.Finite k A] (hn : Module.finrank k A = (n : ℕ)) :
IsKilledBy k A n := by n:ℕ+k:Type u_1inst✝³:Field kA:Type u_2inst✝²:CommRing Ainst✝¹:HopfAlgebra k Ainst✝:Module.Finite k Ahn:Module.finrank k A = ↑n⊢ IsKilledBy k A ↑n
sorry All goals completed! 🐙Over a reduced base ring the answer is affirmative: every finite locally free group scheme of rank $n$ over a reduced commutative ring is killed by $n$. This is [SGA 3], Exp. VIIA, Cor. 8.5.2, deduced there from the case of a field.
@[category research solved, AMS 14 16]
theorem killed_by_rank.variants.reduced_base (n : ℕ+) (R : Type*) [CommRing R] [IsReduced R]
(A : Type*) [CommRing A] [HopfAlgebra R A] [Module.Finite R A] [Module.Projective R A]
(hn : Module.rankAtStalk (R := R) A = fun _ ↦ (n : ℕ)) :
IsKilledBy R A n := by n:ℕ+R:Type u_1inst✝⁵:CommRing Rinst✝⁴:IsReduced RA:Type u_2inst✝³:CommRing Ainst✝²:HopfAlgebra R Ainst✝¹:Module.Finite R Ainst✝:Module.Projective R Ahn:Module.rankAtStalk A = fun x ↦ ↑n⊢ IsKilledBy R A ↑n
sorry All goals completed! 🐙Deligne's theorem: a finite locally free commutative group scheme of rank $n$ over any commutative ring is killed by $n$. Commutativity of the group scheme is cocommutativity of its coordinate Hopf algebra. See [SGA 3], Exp. VIIA, Rem. 8.5.3 and [TO70], p. 4.
@[category research solved, AMS 14 16]
theorem killed_by_rank.variants.commutative (n : ℕ+) (R : Type*) [CommRing R] (A : Type*)
[CommRing A] [HopfAlgebra R A] [Coalgebra.IsCocomm R A] [Module.Finite R A]
[Module.Projective R A]
(hn : Module.rankAtStalk (R := R) A = fun _ ↦ (n : ℕ)) :
IsKilledBy R A n := by n:ℕ+R:Type u_1inst✝⁵:CommRing RA:Type u_2inst✝⁴:CommRing Ainst✝³:HopfAlgebra R Ainst✝²:Coalgebra.IsCocomm R Ainst✝¹:Module.Finite R Ainst✝:Module.Projective R Ahn:Module.rankAtStalk A = fun x ↦ ↑n⊢ IsKilledBy R A ↑n
sorry All goals completed! 🐙def optimal_exponent.answer : ℕ+ → ℕ := sorry
Determine the function assigning to each positive integer $n$ the least positive integer $m$
such that every finite locally free group scheme of rank $n$, over every commutative ring, is
killed by $m$. The value is $0$ if there is no such $m$, since sInf ∅ = 0 in ℕ. As in
killed_by_rank, the rings and algebras are in Type.
Grothendieck asked whether this least exponent is $n$ itself. It is not, already for $n = 4$. If some positive exponent kills every group scheme of rank $n$, then these exponents are the multiples of the least one, so minimality for $\le$ agrees with minimality for divisibility.
@[category research open, question value, AMS 14 16]
theorem optimal_exponent :
∀ n, sInf ((↑) '' {m | IsUniversalKillingExponent.{0, 0} n m} : Set ℕ) = optimal_exponent.answer
n := by ⊢ ∀ (n : ℕ+), sInf (PNat.val '' {m | IsUniversalKillingExponent n m}) = optimal_exponent.answer n
sorry All goals completed! 🐙end FiniteLocallyFreeGroupSchemeExponent