On the rigidity of certain Pham-Brieskorn rings (Q2295390)

Let $k$ be a field of characteristic zero and $B$ a commutative ring containing $k$. $B$ is called \textit{rigid} if the only locally nilpotent derivation $D: B \longrightarrow B$ is the zero derivation. Let $B[X_1, \dots, X_n]$ denote the polynomial ring in $n$ indeterminates $X_1, \dots, X_n$ over $B$. $B$ is said to be \textit{stably rigid} if for any $n \ge 0$ and any locally nilpotent derivation $D: B[X_1, \dots, X_n] \longrightarrow B[X_1, \dots, X_n]$, we have $D(B) = \{0\}$. For $a_1, \dots, a_n \in \mathbb{N}$ and $i = 1, \dots, n$, let $\{ a_1, \dots, \hat{a_i}, \dots, a_n \}$ denote the set $\{ a_1, \dots, a_n \} \backslash \{ a_i \}$. For $a_1, \dots, a_n \in \mathbb{N}$ ($n \ge 3$), set \[B_{a_1, \dots, a_n} := k[X_1, \dots, X_n]/(X_1^{a_1} + \dots + X_n^{a_n})\] and \[T_n := \{ (a_1, \dots, a_n) \in \mathbb{N}^n \mid a_i \ge 2 \text{ for all } i \text{ and } a_i =2 \text{ for at most one } i\}.\] The integral domain $B_{a_1, \dots, a_n}$ is a called Pham-Brieskorn ring. While it is known that if $B_{a_1, \dots, a_n}$ is rigid, then $(a_1, \dots, a_n) \in T_n$; the converse remains as a conjecture. For the convenience of the reader, we define the following. $\mathcal{PB-R}(n)$ := ``If $(a_1, \dots, a_n) \in T_n$, then $B_{a_1, \dots, a_n}$ is rigid.'' $\mathcal{PB-SR}(n)$ := ``If $(a_1, \dots, a_n) \in T_n$, then $B_{a_1, \dots, a_n}$ is stably rigid.'' It has been proven in [\textit{G. Freudenburg} and \textit{L. Moser-Jauslin}, Mich. Math. J. 62, No. 2, 227--258 (2013; Zbl 1321.14046)] that $\mathcal{PB-R}(3)$ holds true. Again, under several special cases, $\mathcal{PB-R}(n)$ for $n \ge 4$ and $\mathcal{PB-SR}(3)$ have been established in [\textit{S. Kaliman} and \textit{M. Zaidenberg}, Ann. Inst. Fourier 50, No. 6, 1649--1669 (2000; Zbl 0971.14044); \textit{D. Finston} and \textit{S. Maubach}, Can. Math. Bull. 53, No. 1, 77--86 (2010; Zbl 1210.14073); \textit{D. Daigle} et al., Adv. Stud. Pure Math. 75, 29--48 (2017; Zbl 1396.14060)] (also see [\textit{J. Seade}, On the topology of isolated singularities in analytic spaces. Basel: Birkhäuser (2006; Zbl 1091.32011)]). In this article, Chitayat-Daigle, shows that $\mathcal{PB-R}(n)$ and $\mathcal{PB-SR}(n)$ holds true under a number of special cases which were not covered earlier, specifically, \begin{itemize} \item [(a)] If $a \ge n \ge 4$, then $B_{a, \dots ,a}$ is rigid where $a, \dots, a$ is a $n$-many $a \in \mathbb{N}$. \item [(b)] If $\sum_{i-1}^{n} \frac{1}{a_i} \le \frac{1}{n-2}$, then $B_{a_1, \dots, a_n}$ is stably rigid. \item [(c)] If $\sum_{i \in I} \frac{1}{a_i} < \frac{1}{n-2}$, then $B_{a_1, \dots, a_n}$ is rigid, where $I = \{i \mid a_i \text{ divides } \text{lcm}(a_1, \dots, \hat{a_i}, \dots, a_n)\}$. \item [(d)] If $a,b,c,d \ge 1$ are such that $a \nmid \text{lcm}(b,c,d)$ and $\frac{1}{b} + \frac{1}{c} + \frac{1}{d} < \frac{1}{2}$, then $B_{a,b,c,d}$ is rigid. \item [(e)] If $(a_1, \dots, a_n) \in T_n$ and $\text{cotype}(a_1, \dots, a_n) \ge n-2$, then $B_{a_1, \dots, a_n}$ is rigid. \item [(f)] If $k_1, k_2, k_3, k_4 \ge 1$ are pairwise relatively prime and $a \ge 3$, then $B_{ak_1, \dots, ak_n}$ is rigid. \item [(g)] If $k_1, k_2, k_3, k_4 \ge 1$ are pairwise relatively prime and $a \ge n \ge 4$, then $B_{ak_1, \dots, ak_n}$ is rigid. \item [(h)] If $a_1, \dots, a_m \ge 1$ ($m \ge 1$) satisfy $a_i \nmid \text{lcm}(3, a_1, \dots, \hat{a_i}, \dots, a_m)$ for all $i = 1, \dots, m$, then $B_{a_1, \dots, a_m, 3,3,3}$ is rigid. \end{itemize}