ScalingStacks

6.2. Nil Hecke category

6.2.1. Definition

We now define a groupoid of nn-periodic bijections.

Given II a subset of 𝐙/n{\mathbf{Z}}/n we denote by I~\tilde{I} its inverse image in 𝐙{\mathbf{Z}}.

Let 𝒮n{\mathcal{S}}_{n} be the category with objects the subsets of 𝐙/n{\mathbf{Z}}/n and where Hom𝒮n⁡(I,J)\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J) is the set of nn-periodic bijections σ:I~→∼J~\sigma:\tilde{I}\mathrel{\mathop{\kern 0.0pt\to}\limits^{\sim}}\tilde{J}. The group n​𝐙n{\mathbf{Z}} acts by translation on Hom\operatorname{Hom}\nolimits-sets. Note that 𝔖^n=End𝒮n⁡(𝐙/n)\hat{{\mathfrak{S}}}_{n}=\operatorname{End}\nolimits_{{\mathcal{S}}_{n}}({\mathbf{Z}}/n).

Given i,j∈I~i,j\in\tilde{I} with i−j∉n​𝐙i-j{\not\in}n{\mathbf{Z}}, the element si​j∈𝔖^ns_{ij}\in\hat{{\mathfrak{S}}}_{n} restricts to an nn-periodic bijection I~→∼I~\tilde{I}\mathrel{\mathop{\kern 0.0pt\to}\limits^{\sim}}\tilde{I}, which we also denote by si​js_{ij}.

Let II be a subset of 𝐙/n{\mathbf{Z}}/n. There is a unique increasing bijection βI:{1,…,|I|}→∼I~∩{1,…,n}\beta_{I}:\{1,\ldots,|I|\}\mathrel{\mathop{\kern 0.0pt\to}\limits^{\sim}}\tilde{I}\cap\{1,\ldots,n\}. We extend it to an increasing bijection 𝐙→∼I~{\mathbf{Z}}\mathrel{\mathop{\kern 0.0pt\to}\limits^{\sim}}\tilde{I} by βI​(r+d​|I|)=βI​(r)+d​n\beta_{I}(r+d|I|)=\beta_{I}(r)+dn for r∈{1,…,|I|}r\in\{1,\ldots,|I|\} and d∈𝐙d\in{\mathbf{Z}}. There is an isomorphism of groups

FI:𝔖^|I|→∼End𝒮n⁡(I),σ↦βI∘σ∘βI−1.F_{I}:\hat{{\mathfrak{S}}}_{|I|}\mathrel{\mathop{\kern 0.0pt\to}\limits^{\sim}}\operatorname{End}\nolimits_{{\mathcal{S}}_{n}}(I),\ \sigma\mapsto\beta_{I}\circ\sigma\circ\beta_{I}^{-1}.

6.2.2. Length

Consider σ∈Hom𝒮n⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J). We define

L⁡(σ)={(i,i′)∈I~2|i⁡<i′,σ⁡(i)>​σ​(i′)}L(\sigma)=\{(i,i^{\prime})\in\tilde{I}^{2}\ |\ i<i^{\prime},\ \sigma(i)>\sigma(i^{\prime})\}

and L~​(σ)={(i,i′)∈L⁡(σ)| 1≤i≤n}\tilde{L}(\sigma)=\{(i,i^{\prime})\in L(\sigma)\ |\ 1\leq i\leq n\}. The canonical map L~​(σ)→L​(σ)/n​𝐙\tilde{L}(\sigma)\to L(\sigma)/n{\mathbf{Z}} is bijective. We define ℓ​(σ)=|L~​(σ)|\ell(\sigma)=|\tilde{L}(\sigma)|.

0P7F

Lemma 6.2.1. We have ℓ⁡(σ′∘σ)≤ℓ⁡(σ′)+ℓ⁡(σ)\ell(\sigma^{\prime}\circ\sigma)\leq\ell(\sigma^{\prime})+\ell(\sigma) for all σ∈Hom𝒮n⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J) and σ′∈Hom𝒮n⁡(J,K)\sigma^{\prime}\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(J,K).

0P7G

Proof. We have

L(σ′∘σ)={(i1,i2)∈I~2|i1<i2,σ(i1)>σ(i2),σ′∘σ(i1)>σ′∘σ(i2)}⊔{(i1,i2)∈I~2|i1<i2,σ(i1)<σ(i2),σ′∘σ(i1)>σ′∘σ(i2)}={(i1,i2)∈L⁡(σ)|σ′∘σ⁡(i1)>σ′∘σ⁡(i2)}⊔(σ−1×σ−1)​({(j1,j2)∈L⁡(σ′)|σ−1​(j1)<σ−1​(j2)}).L(\sigma^{\prime}\circ\sigma)=\{(i_{1},i_{2})\in\tilde{I}^{2}\ |\ i_{1}<i_{2},\ \sigma(i_{1})>\sigma(i_{2}),\ \sigma^{\prime}\circ\sigma(i_{1})>\sigma^{\prime}\circ\sigma(i_{2})\}\sqcup\\ \{(i_{1},i_{2})\in\tilde{I}^{2}\ |\ i_{1}<i_{2},\ \sigma(i_{1})<\sigma(i_{2}),\ \sigma^{\prime}\circ\sigma(i_{1})>\sigma^{\prime}\circ\sigma(i_{2})\}\\ =\{(i_{1},i_{2})\in L(\sigma)\ |\ \sigma^{\prime}\circ\sigma(i_{1})>\sigma^{\prime}\circ\sigma(i_{2})\}\sqcup(\sigma^{-1}\times\sigma^{-1})\bigl(\{(j_{1},j_{2})\in L(\sigma^{\prime})\ |\ \sigma^{-1}(j_{1})<\sigma^{-1}(j_{2})\}\bigr).

It follows that

ℓ(σ′)+ℓ(σ)−ℓ(σ′∘σ)=2|{(i1,i2)∈I~2|i1<i2,σ(i1)>σ(i2),σ′∘σ(i1)<σ′∘σ(i2)}/n𝐙|≥0.\ell(\sigma^{\prime})+\ell(\sigma)-\ell(\sigma^{\prime}\circ\sigma)=2|\{(i_{1},i_{2})\in\tilde{I}^{2}\ |\ i_{1}<i_{2},\ \sigma(i_{1})>\sigma(i_{2}),\ \sigma^{\prime}\circ\sigma(i_{1})<\sigma^{\prime}\circ\sigma(i_{2})\}/n{\mathbf{Z}}|\geq 0.

∎

Let σ∈Hom𝒮n⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J). We have ℓ⁡(σ)=0\ell(\sigma)=0 if and only if σ\sigma is an increasing bijection.

Given τ∈Hom𝒮n⁡(J,I)\tau\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(J,I) with ℓ⁡(τ)=0\ell(\tau)=0, we have L⁡(τ∘σ)=L⁡(σ)=(τ×τ)​(L⁡(σ∘τ))L(\tau\circ\sigma)=L(\sigma)=(\tau\times\tau)\bigl(L(\sigma\circ\tau)\bigr), hence ℓ⁡(τ∘σ)=ℓ⁡(σ∘τ)=ℓ⁡(σ)\ell(\tau\circ\sigma)=\ell(\sigma\circ\tau)=\ell(\sigma).

Since L⁡(τ∘σ)=(βI×βI)​(L⁡(FI−1​(τ∘σ)))L(\tau\circ\sigma)=(\beta_{I}\times\beta_{I})(L(F_{I}^{-1}(\tau\circ\sigma))), we have ℓ⁡(σ)=ℓ⁡(FI−1​(τ∘σ))\ell(\sigma)=\ell(F_{I}^{-1}(\tau\circ\sigma)). As a consequence, we deduce the following result from Lemma 3.2.3.

0P7H

Lemma 6.2.2. Let σ∈Hom𝒮n⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J). We have

ℓ⁡(σ)=∑0≤i1<i2<ni1,i2∈I~|⌊σ⁡(i2)−σ⁡(i1)n⌋|.\ell(\sigma)=\sum_{\begin{subarray}{c}0\leq i_{1}<i_{2}<n\\ i_{1},i_{2}\in\tilde{I}\end{subarray}}\bigl|{\lfloor\frac{\sigma(i_{2})-\sigma(i_{1})}{n}\rfloor}\bigr|.

The next lemma relates length and number of intersections of paths on a cylinder.

0P7I

Lemma 6.2.3. Let σ∈Hom𝒮n⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J) where I={i1+n​𝐙,i2+n​𝐙}I=\{i_{1}+n{\mathbf{Z}},i_{2}+n{\mathbf{Z}}\} and J={j1+n​𝐙,j2+n​𝐙}J=\{j_{1}+n{\mathbf{Z}},j_{2}+n{\mathbf{Z}}\} with 1≤i1≠i2≤n1\leq i_{1}\neq i_{2}\leq n, 1≤j1≠j2≤n1\leq j_{1}\neq j_{2}\leq n and σ⁡(ir)=jr(modn)\sigma(i_{r})=j_{r}\pmod{n} for r∈{1,2}r\in\{1,2\}. Fix β:{i1,i2,j1,j2}→𝐑\beta:\{i_{1},i_{2},j_{1},j_{2}\}\to{\mathbf{R}} increasing with |β⁡(u)−β⁡(v)|<1|\beta(u)-\beta(v)|<1 for all u,vu,v.

Consider γr:[0,1]→𝐑\gamma_{r}:[0,1]\to{\mathbf{R}} continuous with γr​(0)=β⁡(ir)\gamma_{r}(0)=\beta(i_{r}) and γr​(1)=β⁡(jr)+σ⁡(ir)−jrn\gamma_{r}(1)=\beta(j_{r})+\frac{\sigma(i_{r})-j_{r}}{n} for r∈{1,2}r\in\{1,2\}. We have

ℓ⁡(σ)≤|{t∈[0,1]|e2​i​π​γ1​(t)=e2​i​π​γ2​(t)}|\ell(\sigma)\leq|\{t\in[0,1]\ |\ e^{2i\pi\gamma_{1}(t)}=e^{2i\pi\gamma_{2}(t)}\}|

with equality if, for all r∈{1,2}r\in\{1,2\}, the map γr\gamma_{r} is affine.

0P7J

Proof. Without loss of generality, we can assume i1<i2i_{1}<i_{2}. The lemma follows by applying the intermediate value theorem to γ2​(t)−γ1​(t)\gamma_{2}(t)-\gamma_{1}(t) and using Lemma 6.2.2, considering four cases according to the signs of j2−j1j_{2}-j_{1} and σ⁡(i2)−σ⁡(i1)\sigma(i_{2})-\sigma(i_{1}). ∎

6.2.3. Filtration

Given I,J⊂𝐙/nI,J\subset{\mathbf{Z}}/n, we define Hom𝒮n≥−r⁡(I,J)={σ∈Hom𝒮n⁡(I,J)|l⁡(σ)≤r}\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}^{\geq-r}}(I,J)=\{\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J)\ |\ l(\sigma)\leq r\} for r∈𝐙≥0r\in{\mathbf{Z}}_{\geq 0}. It follows from Lemma 6.2.1 that this defines a structure of 𝐙≤0{\mathbf{Z}}_{\leq 0}-filtered category on 𝒮n{\mathcal{S}}_{n}. We put ℋn=gr​𝒮n∙{\mathcal{H}}_{n}=\mathrm{gr}{\mathcal{S}}_{n}^{\bullet}, a pointed 𝐙≤0{\mathbf{Z}}_{\leq 0}-graded category.

Note that a map σ\sigma of length 00 is invertible in ℋn{\mathcal{H}}_{n}. Note also that FIF_{I} induces an isomorphism of graded pointed monoids 𝔖^|I|nil→∼Endℋn⁡(I)\hat{{\mathfrak{S}}}_{|I|}^{\operatorname{nil}\nolimits}\mathrel{\mathop{\kern 0.0pt\to}\limits^{\sim}}\operatorname{End}\nolimits_{{\mathcal{H}}_{n}}(I).

6.2.4. Non-commutative degree

Let us consider the free abelian groups Rn=⨁a∈𝐙/n𝐙​αaR_{n}=\bigoplus_{a\in{\mathbf{Z}}/n}{\mathbf{Z}}\alpha_{a} and Ln=⨁a∈𝐙/n𝐙​εaL_{n}=\bigoplus_{a\in{\mathbf{Z}}/n}{\mathbf{Z}}\varepsilon_{a}. We define a linear map ρ:Rn→Ln\rho:R_{n}\to L_{n} by ρ⁡(αa)=εa+1−εa\rho(\alpha_{a})=\varepsilon_{a+1}-\varepsilon_{a} and a representation of the group RnR_{n} on LnL_{n} given by

αa⋅εb=(δa,b+δa+1,b)​εb.\alpha_{a}\cdot\varepsilon_{b}=(\delta_{a,b}+\delta_{a+1,b})\varepsilon_{b}.

Note that δ=∑a∈𝐙/nαa∈ker⁡ρ\delta=\sum_{a\in{\mathbf{Z}}/n}\alpha_{a}\in\ker\rho and δ⋅εb=2​εb\delta\cdot\varepsilon_{b}=2\varepsilon_{b} for all bb.

We define a bilinear map

⟨−,−⟩:Rn×Rn→Ln,⟨α,α′⟩=α⋅ρ⁡(α′).\langle-,-\rangle:R_{n}\times R_{n}\to L_{n},\ \langle\alpha,\alpha^{\prime}\rangle=\alpha\cdot\rho(\alpha^{\prime}).

Let Γn′=Ln×Rn\Gamma_{n}^{\prime}=L_{n}\times R_{n}. We define a group structure on Γn′\Gamma_{n}^{\prime} by

(l,α)⋅(l′,α′)=(l+l′+⟨α,α′⟩,α+α′).(l,\alpha)\cdot(l^{\prime},\alpha^{\prime})=(l+l^{\prime}+\langle\alpha,\alpha^{\prime}\rangle,\alpha+\alpha^{\prime}).

Given I⊂𝐙/nI\subset{\mathbf{Z}}/n, we put εI=∑a∈Iεa∈Ln\varepsilon_{I}=\sum_{a\in I}\varepsilon_{a}\in L_{n}. Given i,j∈𝐙i,j\in{\mathbf{Z}}, we put

αi,j=∑i≤r<jαr+n​𝐙−∑j≤r<iαr+n​𝐙.\alpha_{i,j}=\sum_{i\leq r<j}\alpha_{r+n{\mathbf{Z}}}-\sum_{j\leq r<i}\alpha_{r+n{\mathbf{Z}}}.

Note that αi,i+1=αi+n​𝐙\alpha_{i,i+1}=\alpha_{i+n{\mathbf{Z}}}, αi+n,j+n=αi,j\alpha_{i+n,j+n}=\alpha_{i,j} and αi,j+αj,k=αi,k\alpha_{i,j}+\alpha_{j,k}=\alpha_{i,k} for all i,j,k∈𝐙i,j,k\in{\mathbf{Z}}. Note also that δ=αi,i+n\delta=\alpha_{i,i+n} for all i∈𝐙i\in{\mathbf{Z}}. Note finally that ρ⁡(αi,j)=εj+n​𝐙−εi+n​𝐙\rho(\alpha_{i,j})=\varepsilon_{j+n{\mathbf{Z}}}-\varepsilon_{i+n{\mathbf{Z}}}.

Consider σ∈Hom𝒮n⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J). We put

⟦σ⟧=∑i∈I~∩[1,n]αi,σ⁡(i)∈Rn.\llbracket\sigma\rrbracket=\sum_{i\in\tilde{I}\cap[1,n]}\alpha_{i,\sigma(i)}\in R_{n}.

Note that ρ⁡(⟦σ⟧)=εJ−εI\rho(\llbracket\sigma\rrbracket)=\varepsilon_{J}-\varepsilon_{I} and ⟦σ′∘σ⟧=⟦σ′⟧+⟦σ⟧\llbracket\sigma^{\prime}\circ\sigma\rrbracket=\llbracket\sigma^{\prime}\rrbracket+\llbracket\sigma\rrbracket for any σ′∈Hom𝒮n⁡(J,K)\sigma^{\prime}\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(J,K).

We define

m⁡(σ)=⟦σ⟧⋅εI∈Ln​ and ​dm⁡(σ)=(−m⁡(σ),−⟦σ⟧)∈Γn′.m(\sigma)=\llbracket\sigma\rrbracket\cdot\varepsilon_{I}\in L_{n}\text{ and }\operatorname{dm}\nolimits(\sigma)=(-m(\sigma),-\llbracket\sigma\rrbracket)\in\Gamma^{\prime}_{n}.
0P7K

Lemma 6.2.4. Let w∈W|I|w\in W_{|I|}, m∈𝐙m\in{\mathbf{Z}} and let σ=FI​(w​cm)\sigma=F_{I}(wc^{m}) be the element of End𝒮n⁡(I)\operatorname{End}\nolimits_{{\mathcal{S}}_{n}}(I) corresponding to w​cmwc^{m}. We have ℓ⁡(σ)=ℓ⁡(w)\ell(\sigma)=\ell(w), ⟦σ⟧=m⋅δ\llbracket\sigma\rrbracket=m\cdot\delta and m⁡(σ)=2​m​εIm(\sigma)=2m\varepsilon_{I}.

0P7L

Proof. The first statement follows from the fact that FIF_{I} preserves lengths (cf the discussion before Lemma 6.2.2).

Note that ⟦si,j⟧=0\llbracket s_{i,j}\rrbracket=0 for i,j∈I~i,j\in\tilde{I} with i−j∉n​𝐙i-j{\not\in}n{\mathbf{Z}}, while ⟦FI​(c)⟧=δ\llbracket F_{I}(c)\rrbracket=\delta. We deduce that ⟦σ⟧=m⋅δ\llbracket\sigma\rrbracket=m\cdot\delta.

The last statement of the lemma is immediate. ∎

0P7M

Lemma 6.2.5. Consider σ∈Hom𝒮n⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J) and σ′∈Hom𝒮n⁡(J,K)\sigma^{\prime}\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(J,K). We have dm⁡(σ′∘σ)=dm⁡(σ′)⋅dm⁡(σ)\operatorname{dm}\nolimits(\sigma^{\prime}\circ\sigma)=\operatorname{dm}\nolimits(\sigma^{\prime})\cdot\operatorname{dm}\nolimits(\sigma).

0P7N

Proof. We have

m⁡(σ′∘σ)=⟦σ′⟧⋅εI+⟦σ⟧​εI=m⁡(σ′)+m⁡(σ)+⟦σ′⟧⋅(εI−εJ),m(\sigma^{\prime}\circ\sigma)=\llbracket\sigma^{\prime}\rrbracket\cdot\varepsilon_{I}+\llbracket\sigma\rrbracket\varepsilon_{I}=m(\sigma^{\prime})+m(\sigma)+\llbracket\sigma^{\prime}\rrbracket\cdot(\varepsilon_{I}-\varepsilon_{J}),

hence

m⁡(σ′)+m⁡(σ)−m⁡(σ′∘σ)=⟦σ′⟧⋅ρ⁡(⟦σ⟧).m(\sigma^{\prime})+m(\sigma)-m(\sigma^{\prime}\circ\sigma)=\llbracket\sigma^{\prime}\rrbracket\cdot\rho(\llbracket\sigma\rrbracket).

The lemma follows. ∎

We put Γn=12​𝐙×Γn′\Gamma_{n}=\frac{1}{2}{\mathbf{Z}}\times\Gamma^{\prime}_{n}. We endow Γn\Gamma_{n} with a structure of 𝐙{\mathbf{Z}}-monoid by using the canonical embedding 𝐙↪12​𝐙↪Γn{\mathbf{Z}}\hookrightarrow\frac{1}{2}{\mathbf{Z}}\hookrightarrow\Gamma_{n}.

Given σ∈Hom𝒮n⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J), we put deg⁡(σ)=(−ℓ⁡(σ),dm⁡(σ))∈Γn\deg(\sigma)=(-\ell(\sigma),\operatorname{dm}\nolimits(\sigma))\in\Gamma_{n}.

Let DD be a subset of {1,…,n}×{±1}\{1,\ldots,n\}\times\{\pm 1\} that embeds in its projection on {1,…,n}\{1,\ldots,n\}. We denote by ΓD\Gamma_{D} the quotient of Γn\Gamma_{n} by the subgroup generated by (0,εi+n​𝐙)+(12​νi,0)(0,\varepsilon_{i+n{\mathbf{Z}}})+(\frac{1}{2}\nu_{i},0), where (i,νi)∈D(i,\nu_{i})\in D. We identify 12​𝐙\frac{1}{2}{\mathbf{Z}} with the image of 12​𝐙×0\frac{1}{2}{\mathbf{Z}}\times 0 in ΓD\Gamma_{D}. We define a partial order on ΓD\Gamma_{D} by h≥gh\geq g if h​g−1hg^{-1} is in 12​𝐙≥0\frac{1}{2}{\mathbf{Z}}_{\geq 0}. We denote by degD⁡(σ)\deg_{D}(\sigma) the image of deg⁡(σ)\deg(\sigma) in ΓD\Gamma_{D}.

Given EE a subset of {1,…,n}\{1,\ldots,n\}, we put E+={(i,1)|i∈E}E^{+}=\{(i,1)\ |\ i\in E\}.

By Lemmas 6.2.1 and 6.2.5, we obtain a ΓD\Gamma_{D}-filtration on 𝒮n{\mathcal{S}}_{n} by defining

Hom𝒮n≥g⁡(I,J)={σ∈Hom𝒮n⁡(I,J)|degD⁡(σ)≥g}.\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}^{\geq g}}(I,J)=\{\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J)\ |\ \deg_{D}(\sigma)\geq g\}.

It follows from Lemma 6.2.5 that the pointed category ℋn{\mathcal{H}}_{n} is isomorphic to the graded pointed category associated to the ΓD\Gamma_{D}-filtration of 𝒮n{\mathcal{S}}_{n} (after forgetting the ΓD\Gamma_{D}-grading to 𝐙{\mathbf{Z}}).

Note that if D=∅D=\emptyset, then ΓD=Γn\Gamma_{D}=\Gamma_{n}, degD=deg\deg_{D}=\deg and the 𝐙≤0{\mathbf{Z}}_{\leq 0} grading on ℋn{\mathcal{H}}_{n} given by the length can be recovered from the Γn\Gamma_{n}-grading by using the quotient map Γn→Γn/Γn′=12​𝐙\Gamma_{n}\to\Gamma_{n}/\Gamma^{\prime}_{n}=\frac{1}{2}{\mathbf{Z}}.

This quotient map provides a 𝐙{\mathbf{Z}}-grading on the Γn\Gamma_{n}-graded pointed category associated to the Γn\Gamma_{n}-filtration of 𝒮n{\mathcal{S}}_{n}. This 𝐙{\mathbf{Z}}-graded pointed category is isomorphic to ℋn{\mathcal{H}}_{n}.

0P7P

Remark 6.2.6. The bilinear form ⟨⟨−,−⟩⟩:Rn×Rn→12​𝐙\langle\langle-,-\rangle\rangle:R_{n}\times R_{n}\to\frac{1}{2}{\mathbf{Z}} obtained from ⟨−,−⟩\langle-,-\rangle by composing with the morphism Ln→12​𝐙,εi↦−12L_{n}\to\frac{1}{2}{\mathbf{Z}},\ \varepsilon_{i}\mapsto-\frac{1}{2} is given by ⟨⟨αa,αb⟩⟩=12​(δb,a+1−δb+1,a)\langle\langle\alpha_{a},\alpha_{b}\rangle\rangle=\frac{1}{2}(\delta_{b,a+1}-\delta_{b+1,a}). It is antisymmetric.

Assume D=[1,n]+D=[1,n]^{+}. Composing with the quotient map Γn↠ΓD\Gamma_{n}\twoheadrightarrow\Gamma_{D}, the embedding 12​𝐙↪Γn,r↦(r,0)\frac{1}{2}{\mathbf{Z}}\hookrightarrow\Gamma_{n},\ r\mapsto(r,0) and the quotient map Γn↠Rn,(r,(l,α))↦α\Gamma_{n}\twoheadrightarrow R_{n},\ (r,(l,\alpha))\mapsto\alpha induce an embedding of 12​𝐙\frac{1}{2}{\mathbf{Z}} as a central subgroup of ΓD\Gamma_{D} with quotient map ΓD↠Rn\Gamma_{D}\twoheadrightarrow R_{n}. So, Γ[1,n]+\Gamma_{[1,n]^{+}} identifies with the set 12​𝐙×Rn\frac{1}{2}{\mathbf{Z}}\times R_{n}, with multiplication given by (r,α)⋅(r′,α′)=(r+r′+⟨⟨α,α′⟩⟩,α+α′)(r,\alpha)\cdot(r^{\prime},\alpha^{\prime})=(r+r^{\prime}+\langle\langle\alpha,\alpha^{\prime}\rangle\rangle,\alpha+\alpha^{\prime}). When n≥3n\geq 3, the group Γ[1,n]+\Gamma_{[1,n]^{+}} has a presentation with generators z=(12,0)z=(\frac{1}{2},0), ga=(0,αa)g_{a}=(0,\alpha_{a}), a∈𝐙/na\in{\mathbf{Z}}/n and relations

z​ga=ga​z,ga​gb​ga−1​gb−1={z if ​b=a+1z−1 if ​b=a−11 otherwise.zg_{a}=g_{a}z,\ g_{a}g_{b}g_{a}^{-1}g_{b}^{-1}=\begin{cases}z&\text{ if }b=a+1\\ z^{-1}&\text{ if }b=a-1\\ 1&\text{ otherwise.}\end{cases}

We define a morphism of groups

ϵ:Γ[1,n]+→𝐙/2,z↦1,ga↦1.\epsilon:\Gamma_{[1,n]^{+}}\to{\mathbf{Z}}/2,\ z\mapsto 1,\ g_{a}\mapsto 1.
0P7Q

Lemma 6.2.7. Given (r,∑ava​αa)∈Γ[1,n]+(r,\sum_{a}v_{a}\alpha_{a})\in\Gamma_{[1,n]^{+}}, we have ϵ⁡(r,∑ava​αa)=2​r+12​|{a∈𝐙/n|va+va+1​ odd}|\epsilon(r,\sum_{a}v_{a}\alpha_{a})=2r+\frac{1}{2}\bigl|\{a\in{\mathbf{Z}}/n\ |\ v_{a}+v_{a+1}\text{ odd}\}\bigr|.

Given σ∈Hom𝒮n⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J), we have ϵ⁡(deg[1,n]+⁡(σ))=0\epsilon(\deg_{[1,n]^{+}}(\sigma))=0.

0P7R

Proof. Denote by ϵ~\tilde{\epsilon} the map defined by the right hand side of the equality of the lemma.

Let NN (resp. N′N^{\prime}) be the cardinality of the set of a∈𝐙/na\in{\mathbf{Z}}/n such that va+va+1v_{a}+v_{a+1} (resp. va′+va+1′v^{\prime}_{a}+v^{\prime}_{a+1}) is odd, where va′=va+δa​bv^{\prime}_{a}=v_{a}+\delta_{ab}. The integers NN and N′N^{\prime} are even. We have

ϵ~​(r,∑ava​αa)+ϵ~​((r,∑ava​αa)​(s,αb))\displaystyle\tilde{\epsilon}(r,\sum_{a}v_{a}\alpha_{a})+\tilde{\epsilon}((r,\sum_{a}v_{a}\alpha_{a})(s,\alpha_{b})) =ϵ~​(r,∑ava​αa)+ϵ~​(r+s+12​(vb−1−vb+1),αb+∑ava​αa)\displaystyle=\tilde{\epsilon}(r,\sum_{a}v_{a}\alpha_{a})+\tilde{\epsilon}(r+s+\frac{1}{2}(v_{b-1}-v_{b+1}),\alpha_{b}+\sum_{a}v_{a}\alpha_{a})
=2​s+vb+1+vb−1+12​(N+N′).\displaystyle=2s+v_{b+1}+v_{b-1}+\frac{1}{2}(N+N^{\prime}).

We have

N′={N+2 if ​vb−1,vb​ and ​vb+1​ have the same parityN−2 if ​vb−1,vb+1​ and ​vb+1​ have the same parityNotherwise.N^{\prime}=\begin{cases}N+2&\text{ if }v_{b-1},\ v_{b}\text{ and }v_{b+1}\text{ have the same parity}\\ N-2&\text{ if }v_{b-1},\ v_{b}+1\text{ and }v_{b+1}\text{ have the same parity}\\ N&\text{otherwise}.\end{cases}

It follows that

ϵ~​(r,∑ava​αa)+ϵ~​((r,∑ava​αa)​(s,αb))=2​s+1.\tilde{\epsilon}(r,\sum_{a}v_{a}\alpha_{a})+\tilde{\epsilon}((r,\sum_{a}v_{a}\alpha_{a})(s,\alpha_{b}))=2s+1.

We deduce by induction on ∑a∈𝐙/n|va|\sum_{a\in{\mathbf{Z}}/n}|v_{a}| that ϵ~​(r,∑ava​αa)=ϵ⁡(r,∑ava​αa)\tilde{\epsilon}(r,\sum_{a}v_{a}\alpha_{a})=\epsilon(r,\sum_{a}v_{a}\alpha_{a}).

Given a,b∈𝐙/na,b\in{\mathbf{Z}}/n and i∈Ii\in I, we have

αa,b⋅εi=δi∈{a,b}a≠b​εimod2​Ln.\alpha_{a,b}\cdot\varepsilon_{i}=\delta_{\begin{subarray}{c}i\in\{a,b\}\\ a\neq b\end{subarray}}\ \varepsilon_{i}\mod 2L_{n}.

It follows that m⁡(σ)≡∑i∈I∖(I∩J)εimod2​Lnm(\sigma)\equiv\sum_{i\in I\setminus(I\cap J)}\varepsilon_{i}\mod{2L_{n}}. Write ⟦σ⟧=∑ava​αa\llbracket\sigma\rrbracket=\sum_{a}v_{a}\alpha_{a}. Given a∈𝐙/na\in{\mathbf{Z}}/n, the integer va+va+1v_{a}+v_{a+1} is odd if and only if a∈I​Δ​Ja\in I\Delta J, hence ϵ⁡(0,⟦σ⟧)=12​|I​Δ​J|=|I∖(I∩J)|\epsilon(0,\llbracket\sigma\rrbracket)=\frac{1}{2}|I\Delta J|=|I\setminus(I\cap J)|. It follows that ϵ⁡(deg[1,n]+⁡(σ))=0\epsilon(\deg_{[1,n]^{+}}(\sigma))=0. ∎

It follows from Lemma 6.2.7 that the Γn\Gamma_{n}-grading on ℋn{\mathcal{H}}_{n} comes from a grading by the kernel of the composition Γn→canΓ[1,n]+→ϵ𝐙/2\Gamma_{n}\xrightarrow{{\mathrm{can}}}\Gamma_{[1,n]^{+}}\xrightarrow{\epsilon}{\mathbf{Z}}/2.

6.2.5. Differential

Given σ∈Hom𝒮n⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J), let D⁡(σ)D(\sigma) be the set of pairs (i1,i2)∈L⁡(σ)(i_{1},i_{2})\in L(\sigma) such that

  • •

    i2−i1<ni_{2}-i_{1}<n or σ⁡(i1)−σ⁡(i2)<n\sigma(i_{1})-\sigma(i_{2})<n and

  • •

    given i∈I~i\in\tilde{I} with i1<i<i2i_{1}<i<i_{2}, we have σ⁡(i1)<σ⁡(i)\sigma(i_{1})<\sigma(i) or σ⁡(i)<σ⁡(i2)\sigma(i)<\sigma(i_{2}).

We put D~​(σ)=D⁡(σ)∩L~​(σ)\tilde{D}(\sigma)=D(\sigma)\cap\tilde{L}(\sigma). The diagonal action of n​𝐙n{\mathbf{Z}} on L⁡(σ)L(\sigma) preserves D⁡(σ)D(\sigma) and we have a canonical bijection D~​(σ)→∼D​(σ)/n​𝐙\tilde{D}(\sigma)\mathrel{\mathop{\kern 0.0pt\to}\limits^{\sim}}D(\sigma)/n{\mathbf{Z}}.

Given (i1,i2)∈L⁡(σ)(i_{1},i_{2})\in L(\sigma), we put σi1,i2:=σ∘si1,i2\sigma^{i_{1},i_{2}}:=\sigma\circ s_{i_{1},i_{2}}.

We define a partial order on Hom𝒮n⁡(I,J)\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J) as the transitive closure of σ′<σ\sigma^{\prime}<\sigma if σ′=σi1,i2\sigma^{\prime}=\sigma^{i_{1},i_{2}} for some (i1,i2)∈D⁡(σ)(i_{1},i_{2})\in D(\sigma).

When I=J=𝐙/nI=J={\mathbf{Z}}/n, this coincides with the extended Chevalley-Bruhat order on 𝔖^n\hat{{\mathfrak{S}}}_{n} by Lemma 3.2.4 and given (i1,i2)∈L⁡(σ)(i_{1},i_{2})\in L(\sigma), we have σi1,i2<σ\sigma^{i_{1},i_{2}}<\sigma (Lemma 3.2.3). The next lemma shows that this holds for general maps in 𝒮n{\mathcal{S}}_{n}.

0P7S

Lemma 6.2.8. Let σ,σ′∈Hom𝒮n⁡(I,J)\sigma,\sigma^{\prime}\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J). Given τ∈Hom𝒮n⁡(J,I)\tau\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(J,I) with ℓ⁡(τ)=0\ell(\tau)=0, we have σ′<σ\sigma^{\prime}<\sigma if and only if τ∘σ′<τ∘σ\tau\circ\sigma^{\prime}<\tau\circ\sigma if and only if σ′∘τ<σ∘τ\sigma^{\prime}\circ\tau<\sigma\circ\tau.

0P7T

Proof. Note that τ\tau is an increasing bijection since ℓ⁡(τ)=0\ell(\tau)=0. We have D⁡(τ∘σ)=D⁡(σ)D(\tau\circ\sigma)=D(\sigma) and given (i1,i2)∈D⁡(σ)(i_{1},i_{2})\in D(\sigma), we have (τ∘σ)i1,i2=τ∘σi1,i2(\tau\circ\sigma)^{i_{1},i_{2}}=\tau\circ\sigma^{i_{1},i_{2}}. This shows the first equivalence. The second equivalence follows from the fact that D⁡(σ∘τ)=(τ−1×τ−1)​(D⁡(σ))D(\sigma\circ\tau)=(\tau^{-1}\times\tau^{-1})(D(\sigma)) and given (i1,i2)∈D⁡(σ)(i_{1},i_{2})\in D(\sigma), we have (σ∘τ)τ−1​(i1),τ−1​(i2)=σi1,i2∘τ(\sigma\circ\tau)^{\tau^{-1}(i_{1}),\tau^{-1}(i_{2})}=\sigma^{i_{1},i_{2}}\circ\tau. ∎

0P7U

Lemma 6.2.9. Given σ∈Hom𝒮n⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J), there is a bijection

D~(σ)→∼{σ′∈Hom𝒮n(I,J)|σ′<σ,ℓ(σ′)=ℓ(σ)−1},(i1,i2)↦σi1,i2.\tilde{D}(\sigma)\mathrel{\mathop{\kern 0.0pt\to}\limits^{\sim}}\{\sigma^{\prime}\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J)\ |\ \sigma^{\prime}<\sigma,\ \ell(\sigma^{\prime})=\ell(\sigma)-1\},\ (i_{1},i_{2})\mapsto\sigma^{i_{1},i_{2}}.

Note that

{σ′∈Hom𝒮n(I,J)|σ′<σ,ℓ(σ′)=ℓ(σ)−1}={σ′∈Hom𝒮n(I,J)|σ′<σ,deg(σ′)=deg(σ)+1}.\{\sigma^{\prime}\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J)\ |\ \sigma^{\prime}<\sigma,\ \ell(\sigma^{\prime})=\ell(\sigma)-1\}=\{\sigma^{\prime}\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J)\ |\ \sigma^{\prime}<\sigma,\ \deg(\sigma^{\prime})=\deg(\sigma)+1\}.

Given (i1,i2)∈L⁡(σ)(i_{1},i_{2})\in L(\sigma), we have (i1,i2)∈D⁡(σ)(i_{1},i_{2})\in D(\sigma) if and only if degD⁡(σ)=degD⁡(σi1,i2)−1\deg_{D}(\sigma)=\deg_{D}(\sigma^{i_{1},i_{2}})-1 for some subset (equivalently, for any subset) DD of {1,…,n}×{±1}\{1,\ldots,n\}\times\{\pm 1\} that embeds in its projection on {1,…,n}\{1,\ldots,n\}.

0P7V

Proof. Let τ∈Hom𝒮n⁡(J,I)\tau\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(J,I) be an increasing bijection. We have D⁡(τ∘σ)=D⁡(σ)D(\tau\circ\sigma)=D(\sigma) and

{σ′′∈End𝒮n(I)|σ′′<τ∘σ,ℓ(σ′′)=ℓ(τ∘σ)−1}={τ∘σ′|σ′∈Hom𝒮n(I,J),σ′<σ,ℓ(σ′)=ℓ(σ)−1}\{\sigma^{\prime\prime}\in\operatorname{End}\nolimits_{{\mathcal{S}}_{n}}(I)\ |\ \sigma^{\prime\prime}<\tau\circ\sigma,\ \ell(\sigma^{\prime\prime})=\ell(\tau\circ\sigma)-1\}=\{\tau\circ\sigma^{\prime}\ |\ \sigma^{\prime}\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J),\ \sigma^{\prime}<\sigma,\ \ell(\sigma^{\prime})=\ell(\sigma)-1\}

by Lemma 6.2.8. Since the first statement of the lemma holds for τ∘σ\tau\circ\sigma by Lemma 3.2.4, it holds for σ\sigma.

The other statements follow from Lemmas 6.2.4 and 6.2.5. ∎

0P7W

Lemma 6.2.10. Consider σ′′∈Hom𝒮n⁡(I,J)\sigma^{\prime\prime}\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J) and σ′∈Hom𝒮n⁡(J,K)\sigma^{\prime}\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(J,K) and let σ=σ′​σ′′\sigma=\sigma^{\prime}\sigma^{\prime\prime}. Assume ℓ⁡(σ)=ℓ⁡(σ′)+ℓ⁡(σ′′)\ell(\sigma)=\ell(\sigma^{\prime})+\ell(\sigma^{\prime\prime}).

Let (i1,i2)∈D⁡(σ)∖(D⁡(σ)∩D⁡(σ′′))(i_{1},i_{2})\in D(\sigma)\setminus(D(\sigma)\cap D(\sigma^{\prime\prime})). Let α′′=σ′′​si1,i2\alpha^{\prime\prime}=\sigma^{\prime\prime}s_{i_{1},i_{2}} and α′′=(σ′)σ′′​(i1),σ′′​(i2)\alpha^{\prime\prime}=(\sigma^{\prime})^{\sigma^{\prime\prime}(i_{1}),\sigma^{\prime\prime}(i_{2})}. We have σ=α′​α′′\sigma=\alpha^{\prime}\alpha^{\prime\prime} and ℓ⁡(σ)=ℓ⁡(α′)+ℓ⁡(α′′)\ell(\sigma)=\ell(\alpha^{\prime})+\ell(\alpha^{\prime\prime}).

0P7X

Proof. Assume first I=J=KI=J=K. The lemma follows in that case from Lemmas 3.2.4 and 3.2.2.

Consider now the general case. There are increasing bijections τ:J→I\tau:J\to I and τ′:K→J\tau^{\prime}:K\to J. We have D⁡(σ)=τ−1​(D⁡(τ′​σ​τ))D(\sigma)=\tau^{-1}(D(\tau^{\prime}\sigma\tau)) and D⁡(σ′′)=τ−1​(D⁡(σ′′​τ))D(\sigma^{\prime\prime})=\tau^{-1}(D(\sigma^{\prime\prime}\tau)) (proof of Lemma 5.4.7). The lemma follows now from the previous case applied to the decomposition τ′​σ​τ=(τ′​σ′)​(σ′′​τ)\tau^{\prime}\sigma\tau=(\tau^{\prime}\sigma^{\prime})(\sigma^{\prime\prime}\tau). ∎

Consider σ∈Homℋn⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{H}}_{n}}(I,J) non-zero. We put

d⁡(σ)=∑(i1,i2)∈D~​(σ)σi1,i2∈Hom𝐅2​[ℋn]⁡(I,J).d(\sigma)=\sum_{(i_{1},i_{2})\in\tilde{D}(\sigma)}\sigma^{i_{1},i_{2}}\in\operatorname{Hom}\nolimits_{{\mathbf{F}}_{2}[{\mathcal{H}}_{n}]}(I,J).
0P7Y

Proposition 6.2.11. The maps dd equip the 𝐅2{\mathbf{F}}_{2}-linear Γn\Gamma_{n}-graded category 𝐅2​[ℋn]{\mathbf{F}}_{2}[{\mathcal{H}}_{n}] with a differential Γn\Gamma_{n}-graded structure, hence equip ℋn{\mathcal{H}}_{n} with a differential Γn\Gamma_{n}-graded pointed structure.

Given I⊂𝐙/nI\subset{\mathbf{Z}}/n, the morphism FIF_{I} induces an isomorphism of differential 𝐙{\mathbf{Z}}-graded pointed monoids

𝔖^|I|nil→∼Endℋn⁡(I).\hat{{\mathfrak{S}}}_{|I|}^{\operatorname{nil}\nolimits}\mathrel{\mathop{\kern 0.0pt\to}\limits^{\sim}}\operatorname{End}\nolimits_{{\mathcal{H}}_{n}}(I).
0P7Z

Proof. Note that Lemma 6.2.9 shows that dd is homogeneous of degree 11. The compatibility of dd with FIF_{I} follows from Lemma 3.2.4.

Consider now σ∈Homℋn⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{H}}_{n}}(I,J) non-zero. There exists τ∈Homℋn⁡(J,I)\tau\in\operatorname{Hom}\nolimits_{{\mathcal{H}}_{n}}(J,I) with ℓ⁡(τ)=0\ell(\tau)=0. We have d⁡(τ∘σ)=τ∘d⁡(σ)d(\tau\circ\sigma)=\tau\circ d(\sigma), hence d2​(τ∘σ)=τ∘d2​(σ)d^{2}(\tau\circ\sigma)=\tau\circ d^{2}(\sigma). The compatibility of FIF_{I} with dd shows that d2​(τ∘σ)=0d^{2}(\tau\circ\sigma)=0. Since τ\tau is invertible, we deduce that d2​(σ)=0d^{2}(\sigma)=0.

Consider finally σ′∈Homℋn⁡(J,K)\sigma^{\prime}\in\operatorname{Hom}\nolimits_{{\mathcal{H}}_{n}}(J,K) and fix τ′∈Homℋn⁡(K,J)\tau^{\prime}\in\operatorname{Hom}\nolimits_{{\mathcal{H}}_{n}}(K,J) with ℓ⁡(τ′)=0\ell(\tau^{\prime})=0. We have d⁡(τ′∘σ′∘σ∘τ)=τ′∘d⁡(σ′∘σ)∘τd(\tau^{\prime}\circ\sigma^{\prime}\circ\sigma\circ\tau)=\tau^{\prime}\circ d(\sigma^{\prime}\circ\sigma)\circ\tau and it follows from the compatibility of FJF_{J} with dd that

d⁡(τ′∘σ′∘σ∘τ)\displaystyle d(\tau^{\prime}\circ\sigma^{\prime}\circ\sigma\circ\tau) =FJ​(d⁡(FJ−1​(τ′∘σ′∘σ∘τ)))=FJ​(d⁡(FJ−1​(τ′∘σ′)∘FJ−1​(σ∘τ)))\displaystyle=F_{J}\bigl(d(F_{J}^{-1}(\tau^{\prime}\circ\sigma^{\prime}\circ\sigma\circ\tau))\bigr)=F_{J}\bigl(d(F_{J}^{-1}(\tau^{\prime}\circ\sigma^{\prime})\circ F_{J}^{-1}(\sigma\circ\tau))\bigr)
=FJ​(d⁡(FJ−1​(τ′∘σ′))∘FJ−1​(σ∘τ))+FJ​(FJ−1​(τ′∘σ′))∘d⁡(FJ−1​(σ∘τ))\displaystyle=F_{J}\bigl(d(F_{J}^{-1}(\tau^{\prime}\circ\sigma^{\prime}))\circ F_{J}^{-1}(\sigma\circ\tau)\bigr)+F_{J}\bigl(F_{J}^{-1}(\tau^{\prime}\circ\sigma^{\prime}))\circ d(F_{J}^{-1}(\sigma\circ\tau)\bigr)
=d⁡(τ′∘σ′)∘σ∘τ+τ′∘σ′∘d⁡(σ∘τ).\displaystyle=d(\tau^{\prime}\circ\sigma^{\prime})\circ\sigma\circ\tau+\tau^{\prime}\circ\sigma^{\prime}\circ d(\sigma\circ\tau).

∎

0P80

Example 6.2.12. Elements of L~​(σ)\tilde{L}(\sigma) correspond to intersections in a representing diagram. Given (i1,i2)∈L~​(σ)(i_{1},i_{2})\in\tilde{L}(\sigma), the element σi1,i2\sigma^{i_{1},i_{2}} correspond to the diagram obtained by smoothing the intersection point corresponding to (i1,i2)(i_{1},i_{2}). If (i1,i2)∉D~​(σ)(i_{1},i_{2}){\not\in}\tilde{D}(\sigma), the element associated to the diagram will vanish in ℋn{\mathcal{H}}_{n}.

[Uncaptioned image]

6.2.6. Change of nn

Fix a positive integer n′≤nn^{\prime}\leq n and an increasing injection α:{1,…,n′}↪{1,…,n}\alpha:\{1,\ldots,n^{\prime}\}\hookrightarrow\{1,\ldots,n\}. We extend α\alpha to an increasing injection 𝐙→𝐙{\mathbf{Z}}\to{\mathbf{Z}} by α⁡(r+d​n′)=α⁡(r)+d​n\alpha(r+dn^{\prime})=\alpha(r)+dn for r∈{1,…,n′}r\in\{1,\ldots,n^{\prime}\} and d∈𝐙d\in{\mathbf{Z}}.

Consider α:{1,…,n′}↪{1,…,n}\alpha:\{1,\ldots,n^{\prime}\}\hookrightarrow\{1,\ldots,n\} an increasing injection as in §6.2.1. We define two injective morphisms of groups

Rα:Rn→Rn′,αi+n′​𝐙↦αα⁡(i),α⁡(i+1)​ and ​Lα:Ln→Ln′,εi+n′​𝐙↦εi+n​𝐙R_{\alpha}:R_{n}\to R_{n^{\prime}},\ \alpha_{i+n^{\prime}{\mathbf{Z}}}\mapsto\alpha_{\alpha(i),\alpha(i+1)}\text{ and }L_{\alpha}:L_{n}\to L_{n^{\prime}},\ \varepsilon_{i+n^{\prime}{\mathbf{Z}}}\mapsto\varepsilon_{i+n{\mathbf{Z}}}

for 1≤i≤n′1\leq i\leq n^{\prime}. We have commutative diagrams

Rn′\textstyle{R_{n^{\prime}}\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces}ρ\scriptstyle{\rho}Rα\scriptstyle{R_{\alpha}}Ln′\textstyle{L_{n^{\prime}}\ignorespaces\ignorespaces\ignorespaces\ignorespaces}Lα\scriptstyle{L_{\alpha}}Rn′×Ln′\textstyle{R_{n^{\prime}}\times L_{n^{\prime}}\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces}⋅\scriptstyle{\cdot}Rα×Lα\scriptstyle{R_{\alpha}\times L_{\alpha}}Ln′\textstyle{L_{n^{\prime}}\ignorespaces\ignorespaces\ignorespaces\ignorespaces}Lα\scriptstyle{L_{\alpha}}Rn′×Rn′\textstyle{R_{n^{\prime}}\times R_{n^{\prime}}\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces\ignorespaces}⟨−,−⟩\scriptstyle{\langle-,-\rangle}Rα×Rα\scriptstyle{R_{\alpha}\times R_{\alpha}}Ln′\textstyle{L_{n^{\prime}}\ignorespaces\ignorespaces\ignorespaces\ignorespaces}Lα\scriptstyle{L_{\alpha}}Rn\textstyle{R_{n}\ignorespaces\ignorespaces\ignorespaces\ignorespaces}ρ\scriptstyle{\rho}Ln\textstyle{L_{n}}Rn×Ln\textstyle{R_{n}\times L_{n}\ignorespaces\ignorespaces\ignorespaces\ignorespaces}⋅\scriptstyle{\cdot}Ln\textstyle{L_{n}}Rn×Rn\textstyle{R_{n}\times R_{n}\ignorespaces\ignorespaces\ignorespaces\ignorespaces}⟨−,−⟩\scriptstyle{\langle-,-\rangle}Ln\textstyle{L_{n}}

As a consequence, we have two injective morphisms of groups

Γα′=Lα×Rα:Γn′′→Γn′​ and ​Γα=id×Lα×Rα:Γn′→Γn,\Gamma^{\prime}_{\alpha}=L_{\alpha}\times R_{\alpha}:\Gamma^{\prime}_{n^{\prime}}\to\Gamma^{\prime}_{n}\text{ and }\Gamma_{\alpha}=\operatorname{id}\nolimits\times L_{\alpha}\times R_{\alpha}:\Gamma_{n^{\prime}}\to\Gamma_{n},

the last of which induces an injective morphism of groups ΓD→Γ(α×id)(D)\Gamma_{D}\to\Gamma_{(\alpha\times\operatorname{id}\nolimits)(D)}, for DD a subset of {1,…,n′}×{±1}\{1,\ldots,n^{\prime}\}\times\{\pm 1\} that embeds in its projection on {1,…,n′}\{1,\ldots,n^{\prime}\}.

We define now a fully faithful functor F=Fα:𝒮n′→𝒮nF=F_{\alpha}:{\mathcal{S}}_{n^{\prime}}\to{\mathcal{S}}_{n}. Given II a subset of 𝐙/n′{\mathbf{Z}}/n^{\prime}, we define F⁡(I)F(I) to be the image of α⁡(I~∩[1,n′])\alpha(\tilde{I}\cap[1,n^{\prime}]) in 𝐙/n{\mathbf{Z}}/n. Given σ∈Hom𝒮n′⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n^{\prime}}}(I,J), we put F⁡(σ)=α∘σ∘α−1F(\sigma)=\alpha\circ\sigma\circ\alpha^{-1}.

Note that the isomorphism of groups 𝔖^n′=End𝒮n′⁡(𝐙/n′)→∼End𝒮n⁡(F⁡(𝐙/n′))\hat{{\mathfrak{S}}}_{n^{\prime}}=\operatorname{End}\nolimits_{{\mathcal{S}}_{n^{\prime}}}({\mathbf{Z}}/n^{\prime})\mathrel{\mathop{\kern 0.0pt\to}\limits^{\sim}}\operatorname{End}\nolimits_{{\mathcal{S}}_{n}}(F({\mathbf{Z}}/n^{\prime})) induced by FF coincides with FF⁡(𝐙/n′)F_{F({\mathbf{Z}}/n^{\prime})} defined in §6.2.1.

As a consequence, FαF_{\alpha} induces a fully faithful graded functor ℋn′→ℋn{\mathcal{H}}_{n^{\prime}}\to{\mathcal{H}}_{n}.

0P81

Lemma 6.2.13. Given n′≤nn^{\prime}\leq n and α:{1,…,n′}↪{1,…,n}\alpha:\{1,\ldots,n^{\prime}\}\hookrightarrow\{1,\ldots,n\} an increasing injection, the functor FαF_{\alpha} induces a differential Γn\Gamma_{n}-graded pointed functor ℋn′→ℋn{\mathcal{H}}_{n^{\prime}}\to{\mathcal{H}}_{n}.

0P82

Proof. Let σ∈Hom𝒮n⁡(I,J)\sigma\in\operatorname{Hom}\nolimits_{{\mathcal{S}}_{n}}(I,J). We have L⁡(Fα​(σ))=(α×α)​(L⁡(σ))L(F_{\alpha}(\sigma))=(\alpha\times\alpha)(L(\sigma)), hence ℓ⁡(Fα​(σ))=ℓ⁡(σ)\ell(F_{\alpha}(\sigma))=\ell(\sigma). We have Rα​(⟦σ⟧)=⟦Fα​(σ)⟧R_{\alpha}(\llbracket\sigma\rrbracket)=\llbracket F_{\alpha}(\sigma)\rrbracket, hence Lα​(m⁡(σ))=m⁡(Fα​(σ))L_{\alpha}(m(\sigma))=m(F_{\alpha}(\sigma)). We deduce that Γσ​(deg⁡(σ))=deg⁡(Fα​(σ))\Gamma_{\sigma}(\deg(\sigma))=\deg(F_{\alpha}(\sigma)).

We have D⁡(Fα​(σ))=(α×α)​(D⁡(σ))D(F_{\alpha}(\sigma))=(\alpha\times\alpha)(D(\sigma)) and Fα​(si1,i2)=sα⁡(i1),α⁡(i2)F_{\alpha}(s_{i_{1},i_{2}})=s_{\alpha(i_{1}),\alpha(i_{2})} for i1,i2∈I~i_{1},i_{2}\in\tilde{I} with i1−i2∉n​𝐙i_{1}-i_{2}{\not\in}n{\mathbf{Z}}, hence FαF_{\alpha} is compatible with dd. ∎

Original mathematics by the credited authors. Source collection and HTML conversion remain in progress.

Andrew Manion, Raphael Rouquier

Original source: arXiv:2009.09627v2