Coxeter graphs for super weyl groups of exceptional classical Lie superalgebras
Yuhui shen and Zhiyang Tan · 2026
Read Coxeter graphs for super weyl groups of exceptional classical Lie superalgebras by Yuhui shen and Zhiyang Tan (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-28312.txt.
Introduction
Shen and Tan work out the Coxeter graphs for the super Weyl groups of three exceptional Lie superalgebras. Super Weyl groups shape how these algebras' representations behave. Earlier work had proposed graphs for the D(2,1,alpha), F(4), and G(3) cases but left them unchecked. The authors list all fundamental systems for these algebras and derive the graphs from them. Their results confirm the earlier proposal in full.
Catalog record
| CREATOR | Yuhui shen and Zhiyang Tan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51131 |
| LICENSE | http://creativecommons.org/publicdomain/zero/1.0/ |
| RIGHTS | CC0 |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.28312
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Coxeter graphs for super weyl groups of exceptional classical Lie superalgebras Yuhui shen and Zhiyang Tan Abstract . Super Weyl group plays an important role in the study of representations of basic classical Lie superalgebras. The Coxeter graphs for super Weyl groups of basis classical Lie superalgebras have been given in [ 1 ] , where the authors also made a proposal on the Coxeter graphs for the super Weyl groups of exceptional classical Lie superalgebras D ( 2 , 1 , α ) D(2,1,\alpha) , F ( 4 ) F(4) and G ( 3 ) G(3) . In this paper, we present all fundamental systems of the exceptional Lie superalgebras and get the Coxeter graphs for the corresponding super Weyl groups, verifying a proposal in [ 1 ] . 1. Introduction It is well-known that all Cartan subalgebras of a semisimple Lie algebra over the complex number field are conjugate and all Borel subalgebras are also conjugate. The Weyl group plays an important role in the representation theory of complex semisimple Lie algebras. However, it is not the case for Lie superalgebras. The Weyl groups for basical classical Lie superalgebras only reflect partial information in representation theory of those Lie superalgebras. In [ 9 ] , the authors introduced the super Weyl group \widehat{\mathscr{C}} for a basical Lie superalgebra, which is the transformation group of the set of all Borel subalgebras containing a given Cartan subalgebra, generated by ordinary reflections and odd reflections. The introduction of super Weyl groups integrates ordinary reflections and odd reflections in a body. With aid of super Weyl groups, the authors in [ 9 ] and [ 8 ] successfully obtained the results on Jantzen filtrations and Jantzen sum formulas in modular representations of basic classical Lie algebras and general linear supergroups, respectively. This demonstrates that super Weyl groups deserve further study. Generally, a super Weyl group must be a quotient of the corresponding Coxeter group. In [ 1 ] , the authors initiate to study the Coxeter systems for super Weyl groups, by drawing the Coxeter graphs for basic classical Lie superalgebras of type A , B , C and D . However, they didn’t finish the arguments for the exceptional cases. In this paper, we continue to study the Coxeter system for super Weyl groups of exceptional cases. We also follow the notaions in [ 1 ] and denote super Weyl group by \widehat{\mathscr{C}} . For super case, the super Weyl groups are independent of the choices of Borel subalgebras and Weyl groups can be realized as the subgroups of super Weyl groups. The purpose of this paper is to calculate the m_{xy} defined in [ 1 ] , where x is an odd reflection and get the Coxeter graphs for exceptional classical Lie superalgebras D(2,1,\alpha) , F(4) and G(3) . In order to calculate m_{xy} , we need to first describe all fundamental systems and consider the action of the multiplication of odd reflection and ordinary reflection on the fundamental systems. There are too many fundamental systems \widetilde{\Pi}^{{}^{\prime}} for the exceptional classical Lie superalgebras and the computation will be very tedious. However, one can see that if a fundamental system \widetilde{\Pi}^{{}^{\prime}} contains some simple root \alpha transformed into the corresponding isotropic odd root up to a sign under the reflections, then the order on this fundamental system is bigger than 2 . Otherwise, the order will be 2 . Fortunately, we can classify the fundamental systems into some cases to simplify the calculation of m_{xy} . Finally, the Coxeter groups for D(2,1,\alpha) , F(4) and G(3) can be obtained. In this paper, we directly begin with our arguments on Coxeter graphs for super Weyl groups. The reader can refer to [ 2 ] , [ 7 ] for exceptional super Lie superalgebras, to [ 3 ] , [ 2 ] , or [ 7 ] for their root systems, and to [ 4 ] or [ 5 , 6 ] for Coxeter groups. 2. The root system and Coxeter graph for D(2,1,\alpha) 2.1. The root system Recall that the Lie superalgebra \mathfrak{g}=D(2|1;\alpha) is a family of simple contragredient Lie superalgebras, which depends on a parameter \alpha . There are isomorphisms of Lie superalgebras with different parameters D(2|1;\alpha)\cong D(2,1;-1-\alpha^{-1})\cong D(2|1;\alpha^{-1}) when \alpha is nonzero. Let \mathfrak{g}=D(2,1;\alpha) . Recall that \mathfrak{g}=\mathfrak{g}_{\bar{0}}\oplus\mathfrak{g}_{\bar{1}} with \mathfrak{g}_{\bar{0}}\cong\mathfrak{sl}(2)^{\oplus 3} and as a \mathfrak{g}_{\bar{0}} -module, \mathfrak{g}_{\bar{1}}\cong(\mathbb{C}^{2})^{\oplus 2} . For a Cartan subalgebra \mathfrak{h} of \mathfrak{g}_{\bar{0}} , its linear dual \mathfrak{h}^{*} has a basis \{\delta,\varepsilon_{1},\varepsilon_{2}\} with (\delta,\delta)=-(1+\alpha),(\varepsilon_{1},\varepsilon_{1})=1,(\varepsilon_{2},\varepsilon_{2})=\alpha,(\varepsilon_{1},\varepsilon_{2})=0 and (\delta,\varepsilon_{i})=0,i=1,2 (see for example, [ 7 ] or [ 3 ] for the details). Its root system \Phi=\Phi_{\bar{0}}\bigcup\Phi_{\bar{1}} is presented as \{\pm 2\delta,\pm 2\varepsilon_{i}\}\cup\{\pm\delta\pm\varepsilon_{1}\pm\varepsilon_{2}\}. We can take \Pi=\{\delta+\varepsilon_{1}+\varepsilon_{2},-2\varepsilon_{1},-2\varepsilon_{2}\} as the standard fundamental system of D(2,1,\alpha) . Then the corresponding extended standard fundamental system is \widetilde{\Pi}=\{-2\delta\}\cup\{\delta+\varepsilon_{1}+\varepsilon_{2},-2\varepsilon_{1},-2\varepsilon_{2}\}. What’s more, we can classify all fundamental systems \widetilde{\Pi}^{{}^{\prime}} of D(2,1,\alpha) into following 4 cases: \displaystyle\widetilde{\Pi}^{{}^{\prime}}=\{s\delta+t_{1}\varepsilon_{1}+t_{2}\varepsilon_{2},-2t_{1}\varepsilon_{1},-2t_{2}\varepsilon_{2}\} \displaystyle\widetilde{\Pi}^{{}^{\prime}}=\{s\delta+t_{1}\varepsilon_{1}+t_{2}\varepsilon_{2},-s\delta+t_{1}\varepsilon_{1}-t_{2}\varepsilon_{2},-s\delta-t_{1}\varepsilon_{1}+t_{2}\varepsilon_{2}\} \displaystyle\widetilde{\Pi}^{{}^{\prime}}=\{2t_{1}\varepsilon_{1},s\delta-t_{1}\varepsilon_{1}+t_{2}\varepsilon_{2},-2s\delta\} \displaystyle\widetilde{\Pi}^{{}^{\prime}}=\{2t_{2}\varepsilon_{2},-s\delta+t_{1}\varepsilon_{1}-t_{2}\varepsilon_{2},2s\delta\}, where s,t_{1},t_{2}\in\{1,-1\} . 2.2. The Weyl group and Coxeter graph Recall that \mathfrak{g}_{\bar{0}} admits the Weyl group \langle\widehat{r}_{2\delta}\rangle\times\langle\widehat{r}_{2\varepsilon_{1}}\rangle\times\langle\widehat{r}_{2\varepsilon_{2}}\rangle which is isomorphic to \mathbb{Z}_{2}\times\mathbb{Z}_{2}\times\mathbb{Z}_{2} . Now we can calculate the Coxeter graph for super Weyl group of D(2,1,\alpha) . Theorem 2.1 . The Coxeter graph of \widehat{\mathscr{C}} for D(2,1,\alpha) is 12 12 12 \widehat{r}_{2\varepsilon_{2}} \widehat{r}_{2\delta~~~~} \widehat{r}_{\delta+\varepsilon_{1}+\varepsilon_{2}} \widehat{r}_{2\varepsilon_{1}} Note that the Coxeter graph of the subsystem (\mathscr{W},\textsf{S}_{\Pi_{0}}) is already known, which is just the one of three vertices mutually isolated. We only need to calculate the order of \widehat{r}_{\delta+\varepsilon_{1}+\varepsilon_{2}}\cdot\widehat{r}_{2\delta} , \widehat{r}_{\delta+\varepsilon_{1}+\varepsilon_{2}}\cdot\widehat{r}_{2\varepsilon_{1}} and \widehat{r}_{\delta+\varepsilon_{1}+\varepsilon_{2}}\cdot\widehat{r}_{2\varepsilon_{2}} . First, we need to calculate the actions of \widehat{r}_{\delta+\varepsilon_{1}+\varepsilon_{2}}\cdot\widehat{r}_{2\delta} on all the fundamental systems, and get the order of \widehat{r}_{\delta+\varepsilon_{1}+\varepsilon_{2}}\cdot\widehat{r}_{2\delta} on each fundamental system. The least common multiple of these orders is 12 . So the order of \widehat{r}_{\delta+\varepsilon_{1}+\varepsilon_{2}}\cdot\widehat{r}_{2\delta} is 12 . Denote \widehat{r}_{\delta 12} by \widehat{r}_{\delta+\varepsilon_{1}+\varepsilon_{2}} . We can discuss the order of \widehat{r}_{\delta 12}\cdot\widehat{r}_{2\delta} on different fundamental systems in following cases: (i) \widetilde{\Pi}^{{}^{\prime}} contains elements \pm(-\delta+\varepsilon_{1}+\varepsilon_{2}) and \pm(\delta+\varepsilon_{1}+\varepsilon_{2}) . By observing the fundamental systems, we can see that \widetilde{\Pi}^{{}^{\prime}}=\{s(\delta+\varepsilon_{1}-\varepsilon_{2}),s(-\delta+\varepsilon_{1}+\varepsilon_{2}),s(-\delta-\varepsilon_{1}-\varepsilon_{2})\} or \{s(\delta-\varepsilon_{1}+\varepsilon_{2}),s(-\delta-\varepsilon_{1}-\varepsilon_{2}),s(-\delta+\varepsilon_{1}+\varepsilon_{2})\} , where s\in\{\pm 1\} . By calculation, we have \displaystyle\{s(\delta+\varepsilon_{1}-\varepsilon_{2}),s(-\delta+\varepsilon_{1}+\varepsilon_{2}),s(-\delta-\varepsilon_{1}-\varepsilon_{2})\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(-\delta+\varepsilon_{1}-\varepsilon_{2}),s(\delta+\varepsilon_{1}+\varepsilon_{2}),s(\delta-\varepsilon_{1}-\varepsilon_{2})\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(2\varepsilon_{1}),s(-\delta-\varepsilon_{1}-\varepsilon_{2}),s(2\delta)\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(2\varepsilon_{1}),s(\delta-\varepsilon_{1}-\varepsilon_{2}),s(-2\delta)\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(2\varepsilon_{1}),s(\delta-\varepsilon_{1}-\varepsilon_{2}),s(-2\delta)\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(2\varepsilon_{1}),s(-\delta-\varepsilon_{1}-\varepsilon_{2}),s(2\delta)\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(-\delta+\varepsilon_{1}-\varepsilon_{2}),s(\delta+\varepsilon_{1}+\varepsilon_{2}),s(\delta-\varepsilon_{1}-\varepsilon_{2})\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(\delta+\varepsilon_{1}-\varepsilon_{2}),s(-\delta+\varepsilon_{1}+\varepsilon_{2}),s(-\delta-\varepsilon_{1}-\varepsilon_{2})\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(-2\varepsilon_{2}),s(-2\delta),s(\delta+\varepsilon_{1}+\varepsilon_{2})\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(-2\varepsilon_{2}),s(2\delta),s(-\delta+\varepsilon_{1}+\varepsilon_{2})\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(-2\varepsilon_{2}),s(2\delta),s(-\delta+\varepsilon_{1}+\varepsilon_{2})\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(-2\varepsilon_{2}),s(-2\delta),s(\delta+\varepsilon_{1}+\varepsilon_{2})\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(\delta+\varepsilon_{1}-\varepsilon_{2}),s(-\delta+\varepsilon_{1}+\varepsilon_{2}),s(-\delta-\varepsilon_{1}-\varepsilon_{2})\}. So the order of the action \widehat{r}_{\delta 12}\cdot\widehat{r}_{2\delta} on \widetilde{\Pi}^{{}^{\prime}}=\{s(\delta+\varepsilon_{1}-\varepsilon_{2}),s(-\delta+\varepsilon_{1}+\varepsilon_{2}),s(-\delta-\varepsilon_{1}-\varepsilon_{2})\} is 6 . Similarly, the order of the action \widehat{r}_{\delta 12}\cdot\widehat{r}_{2\delta} on \widetilde{\Pi}^{{}^{\prime}}=\{s(\delta-\varepsilon_{1}+\varepsilon_{2}),s(-\delta-\varepsilon_{1}-\varepsilon_{2}),s(-\delta+\varepsilon_{1}+\varepsilon_{2})\} is also 6 , which is left to readers. (ii) \widetilde{\Pi}^{{}^{\prime}} contains elements \pm(\delta+\varepsilon_{1}+\varepsilon_{2}) and \pm 2\delta , then \widetilde{\Pi}^{{}^{\prime}}=\{s(\delta+\varepsilon_{1}+\varepsilon_{2}),s(-2\delta),s(-2\varepsilon_{i})\} (i=1,2) . \displaystyle\{s(\delta+\varepsilon_{1}+\varepsilon_{2}),s(-2\delta),s(-2\varepsilon_{i})\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(-\delta+\varepsilon_{1}+\varepsilon_{2}),s(2\delta),s(-2\varepsilon_{i})\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(-\delta+\varepsilon_{1}+\varepsilon_{2}),s(2\delta),s(-2\varepsilon_{i})\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(\delta+\varepsilon_{1}+\varepsilon_{2}),s(-2\delta),s(-2\varepsilon_{i})\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(-\delta-\varepsilon_{1}-\varepsilon_{2}),s(-\delta+\varepsilon_{1}+\varepsilon_{2}),s(\delta+\varepsilon_{1}+\varepsilon_{2}-2\varepsilon_{i})\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(\delta-\varepsilon_{1}-\varepsilon_{2}),s(\delta+\varepsilon_{1}+\varepsilon_{2}),s(-\delta+\varepsilon_{1}+\varepsilon_{2}-2\varepsilon_{i})\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(2\delta),s(-\delta-\varepsilon_{1}-\varepsilon_{2}),s(2\varepsilon_{1}+2\varepsilon_{2}-2\varepsilon_{i})\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(-2\delta),s(\delta-\varepsilon_{1}-\varepsilon_{2}),s(2\varepsilon_{1}+2\varepsilon_{2}-2\varepsilon_{i})\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(-2\delta),s(\delta-\varepsilon_{1}-\varepsilon_{2}),s(2\varepsilon_{1}+2\varepsilon_{2}-2\varepsilon_{i})\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(2\delta),s(-\delta-\varepsilon_{1}-\varepsilon_{2}),s(2\varepsilon_{1}+2\varepsilon_{2}-2\varepsilon_{i})\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(\delta-\varepsilon_{1}-\varepsilon_{2}),s(\delta+\varepsilon_{1}+\varepsilon_{2}),s(-\delta+\varepsilon_{1}+\varepsilon_{2}-2\varepsilon_{i})\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(-\delta-\varepsilon_{1}-\varepsilon_{2}),s(-\delta+\varepsilon_{1}+\varepsilon_{2}),s(\delta+\varepsilon_{1}+\varepsilon_{2}-2\varepsilon_{i})\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(\delta+\varepsilon_{1}+\varepsilon_{2}),s(-2\delta),s(-2\varepsilon_{i})\}. So the order of the action of \widehat{r}_{\delta+\varepsilon_{1}+\varepsilon_{2}}\cdot\widehat{r}_{2\delta} on this fundamental system is 6 . (iii) \widetilde{\Pi}^{{}^{\prime}} contains elements \pm(-\delta+\varepsilon_{1}+\varepsilon_{2}) and \pm(2\delta) , then \widetilde{\Pi}^{{}^{\prime}}=\{s(-\delta+\varepsilon_{1}+\varepsilon_{2}),s(2\delta),s(-2\varepsilon_{i})\} (i=1,2) . Similar to the case \widetilde{\Pi}^{{}^{\prime}}=\{s(\delta+\varepsilon_{1}+\varepsilon_{2}),s(-2\delta),s(-2\varepsilon_{i})\} (i=1,2) , one can easily check that the order in this case is also 6 . (iv) \widetilde{\Pi}^{{}^{\prime}}=\{s(\delta+\varepsilon_{1}+\varepsilon_{2}),\alpha,\beta\} , where \alpha,\beta\neq\pm(\delta-\varepsilon_{1}-\varepsilon_{2}),\pm(2\delta) . By observing all fundamental systems, one can easily see that \widehat{r}_{2\delta}(\alpha),\widehat{r}_{2\delta}(\beta),\widehat{r}_{2\delta}\cdot\widehat{r}_{\delta 12}(\alpha),\widehat{r}_{2\delta}\cdot\widehat{r}_{\delta 12}(\beta)\neq\pm(\delta+\varepsilon_{1}+\varepsilon_{2}) . Then it follows that \displaystyle\{s(\delta+\varepsilon_{1}+\varepsilon_{2}),\alpha,\beta\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(-\delta+\varepsilon_{1}+\varepsilon_{2}),\widehat{r}_{2\delta}(\alpha),\widehat{r}_{2\delta}(\beta)\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(-\delta+\varepsilon_{1}+\varepsilon_{2}),\widehat{r}_{2\delta}(\alpha),\widehat{r}_{2\delta}(\beta)\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(\delta+\varepsilon_{1}+\varepsilon_{2}),\alpha,\beta\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(-\delta-\varepsilon_{1}-\varepsilon_{2}),\widehat{r}_{\delta 12}(\alpha),\widehat{r}_{\delta 12}(\beta)\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(\delta-\varepsilon_{1}-\varepsilon_{2}),\widehat{r}_{2\delta}\cdot\widehat{r}_{\delta 12}(\alpha),\widehat{r}_{2\delta}\cdot\widehat{r}_{\delta 12}(\beta)\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(\delta-\varepsilon_{1}-\varepsilon_{2}),\widehat{r}_{2\delta}\cdot\widehat{r}_{\delta 12}(\alpha),\widehat{r}_{2\delta}\cdot\widehat{r}_{\delta 12}(\beta)\}\overset{\widehat{r}_{2\delta}}{\longrightarrow}\{s(-\delta-\varepsilon_{1}-\varepsilon_{2}),\widehat{r}_{\delta 12}(\alpha),\widehat{r}_{\delta 12}(\beta)\}\overset{\widehat{r}_{\delta 12}}{\longrightarrow} \displaystyle\{s(\delta+\varepsilon_{1}+\varepsilon_{2}),\alpha,\beta\}. Hence, the order of \widehat{r}_{\delta 12}\cdot\widehat{r}_{2\delta} in this case is 4 . Similarly, one can check the following case (v) and we leave the proof to readers. (v) \widetilde{\Pi}^{{}^{\prime}}=\{s(-\delta+\varepsilon_{1}+\varepsilon_{2}),\alpha,\beta\} , where \alpha,\beta\neq\pm(\delta+\varepsilon_{1}+\varepsilon_{2}),\pm(2\delta) . The order of \widehat{r}_{\delta 12}\cdot\widehat{r}_{2\delta} on these fundamental