Documentation

TauCeti.Algebra.Lie.Sl2.Associative

sl₂ commutation relations in an associative algebra #

Let A be an associative ring and let H, E, F be elements of A satisfying the sl₂ relations for the ring commutator,

E * F - F * E = H,   H * E - E * H = 2 • E,   H * F - F * H = -(2 • F).

This file computes how E and F commute past powers, and past divided powers, of one another. It uses the general integer-eigenvalue identities from TauCeti.Algebra.Ring.Commutator and TauCeti.RingTheory.DividedPowers.Associative. Over a ℚ-algebra the divided-power forms are

E * F⁽ⁿ⁺¹⁾ = F⁽ⁿ⁺¹⁾ * E + F⁽ⁿ⁾ * (H - n),
F * E⁽ⁿ⁺¹⁾ = E⁽ⁿ⁺¹⁾ * F - E⁽ⁿ⁾ * (H + n),
H * F⁽ⁿ⁾   = F⁽ⁿ⁾ * (H - 2 * n),
H * E⁽ⁿ⁾   = E⁽ⁿ⁾ * (H + 2 * n),

and the point of the normalisation is visible in them: although each divided power is a rational multiple of a power, no denominator survives on the right-hand sides. This is the first characteristic instance of Kostant's normal-ordering formula needed toward proving that the subring generated by divided powers of root vectors and by binomial coefficients of Cartan elements is closed under multiplication.

The relations are stated with the hypotheses each one needs, and not as a bundled IsSl2Triple: the intended consumer is the universal enveloping algebra, where the image of the Cartan element is not known to be nonzero, so the IsSl2Triple.h_ne_zero field is unavailable there. The last section transports the individual bracket hypotheses through the canonical map to the enveloping algebra; a genuine IsSl2Triple supplies them through its bracket fields.

Main results #

Mathlib's IsSl2Triple.HasPrimitiveVectorWith.lie_e_pow_succ_toEnd_f and lie_h_pow_toEnd_f prove the corresponding identities applied to a primitive vector, that is under the extra hypothesis ⁅e, m⁆ = 0, and follow by applying the unconditional operator identities below to that vector. The converse does not follow: identities on one primitive vector do not determine the operators. The operator identities are what closure of an integral form under multiplication requires.

References #

theorem TauCeti.Sl2.h_mul_f_pow {A : Type u_1} [Ring A] {H F : A} (hhf : H * F - F * H = -(2 • F)) (n : ℕ) :
H * F ^ n = F ^ n * (H - 2 * ↑n)

Moving a Cartan element past a power of the lowering element translates it by -2n.

theorem TauCeti.Sl2.h_mul_e_pow {A : Type u_1} [Ring A] {H E : A} (hhe : H * E - E * H = 2 • E) (n : ℕ) :
H * E ^ n = E ^ n * (H + 2 * ↑n)

Moving a Cartan element past a power of the raising element translates it by 2n.

theorem TauCeti.Sl2.e_mul_f_pow_succ {A : Type u_1} [Ring A] {H E F : A} (hef : E * F - F * E = H) (hhf : H * F - F * H = -(2 • F)) (n : ℕ) :
E * F ^ (n + 1) = F ^ (n + 1) * E + (n + 1) • (F ^ n * (H - ↑n))

The raising element commuted past a power of the lowering element: this is the classical ⁅E, Fⁿ⁆ = n Fⁿ⁻¹ (H - (n - 1)), with the index shifted so that no truncated subtraction appears.

theorem TauCeti.Sl2.f_mul_e_pow_succ {A : Type u_1} [Ring A] {H E F : A} (hef : E * F - F * E = H) (hhe : H * E - E * H = 2 • E) (n : ℕ) :
F * E ^ (n + 1) = E ^ (n + 1) * F - (n + 1) • (E ^ n * (H + ↑n))

The lowering element commuted past a power of the raising element: ⁅F, Eⁿ⁺¹⁆ = -(n + 1) Eⁿ (H + n), the mirror of e_mul_f_pow_succ under the Chevalley involution.

theorem TauCeti.Sl2.h_mul_dividedPower_f {A : Type u_1} [Ring A] [Algebra ℚ A] {H F : A} (hhf : H * F - F * H = -(2 • F)) (n : ℕ) :

Moving a Cartan element past a divided power of the lowering element translates it by -2n.

theorem TauCeti.Sl2.h_mul_dividedPower_e {A : Type u_1} [Ring A] [Algebra ℚ A] {H E : A} (hhe : H * E - E * H = 2 • E) (n : ℕ) :

Moving a Cartan element past a divided power of the raising element translates it by 2n.

theorem TauCeti.Sl2.e_mul_dividedPower_succ_f {A : Type u_1} [Ring A] [Algebra ℚ A] {H E F : A} (hef : E * F - F * E = H) (hhf : H * F - F * H = -(2 • F)) (n : ℕ) :

The raising element commuted past a divided power of the lowering element. Every structure constant on the right is 1: this is the first case of Kostant's normal-ordering formula, and the first step toward proving that the relevant ℤ-form is closed under multiplication.

theorem TauCeti.Sl2.f_mul_dividedPower_succ_e {A : Type u_1} [Ring A] [Algebra ℚ A] {H E F : A} (hef : E * F - F * E = H) (hhe : H * E - E * H = 2 • E) (n : ℕ) :

The lowering element commuted past a divided power of the raising element.

theorem TauCeti.Sl2.dividedPower_succ_e_mul_f {A : Type u_1} [Ring A] [Algebra ℚ A] {H E F : A} (hef : E * F - F * E = H) (hhe : H * E - E * H = 2 • E) (n : ℕ) :

The reversed-product form of f_mul_dividedPower_succ_e: E⁽ⁿ⁺¹⁾ F = F E⁽ⁿ⁺¹⁾ + (H - n) E⁽ⁿ⁾. This is the orientation that occurs as the one-lowering-vector case of the full Kostant straightening formula.

theorem TauCeti.Sl2.dividedPower_succ_f_mul_e {A : Type u_1} [Ring A] [Algebra ℚ A] {H E F : A} (hef : E * F - F * E = H) (hhf : H * F - F * H = -(2 • F)) (n : ℕ) :

The reversed-product form of e_mul_dividedPower_succ_f: F⁽ⁿ⁺¹⁾ E = E F⁽ⁿ⁺¹⁾ - (H + n) F⁽ⁿ⁾.

The divided-power relation e * f⁽ⁿ⁺¹⁾ = f⁽ⁿ⁺¹⁾ * e + f⁽ⁿ⁾ * (h - n) in a universal enveloping algebra over ℚ. This is the relation used after specialization to the enveloping algebra of a Chevalley Lie algebra.

The divided-power relation f * e⁽ⁿ⁺¹⁾ = e⁽ⁿ⁺¹⁾ * f - e⁽ⁿ⁾ * (h + n) in a universal enveloping algebra over ℚ.

The reversed-product form of ι_f_mul_dividedPower_succ_ι_e: e⁽ⁿ⁺¹⁾ f = f e⁽ⁿ⁺¹⁾ + (h - n) e⁽ⁿ⁾. This is the orientation that occurs as the one-lowering-vector case of the full Kostant straightening formula.