Problems related to type-\(A\) and type-\(B\) matrices of chromatic joins (Q1433012)

In this paper results and observations based on work by Rodica Simion concerning the determinants of matrices \(T_{n(q)}\) and generalizations \(M(L',\wedge_L,q)\) needed in the inversion formulas for these matrices is discussed at some length with a listing of problems in this area included. The matrix \(T_n(q)\) is the matrix of chromatic joins with rows and columns indexed by elements \(\alpha\), \(\beta\) of \(\text{NC}_n\), the lattice of non-crossing partitions of \(\{1,2,\dots, n\}\) ordered by refinement, the corresponding element \(q^m\) having \(m\) being the index of the block \(\alpha\vee\beta\) in \(\Pi_n\) , the partition lattice of \(\{1,2,\dots, n\}\) of which \(\text{NC}_n\) is a subposet, which leads to a curious formula, among others viz., \(\text{det\,}T_n(q)= \prod^n_{i=1} F_{i+1}(q)^{\delta(n,i)}\), where \(F_n(q)\) is the unique irreducible (Beraha) factor of the \(n\)th Beraha polynomial which does not divide any of the Beraha polynomials of lower index. With \(\Pi_n\) replaced by a ranked lattice \(L\) and an induced subposet \(L'\), interesting questions about the \(T_n(q)\) analog \(M(L',\wedge_L, q)\), which for \(\Pi^B_n= L\) and \(\text{NC}^B_n= L'\) suggest that all factors of det \(M\) have the form \(F_{3k}(q)\) along with a list of other observations and guesses which leave a definite suspicion that deeper and as yet unelucidated principles may be at work here.