Powers of matrices with all principal minors equal to 1

Abstract: Consider a square matrix $A$ whose all principal minors are equal to $1$. Over a field, this property is inherited by any power of $A$, but this is not the case over an arbitrary commutative ring. We show that it is the case over any regular ring, and also over the ring $\mathbb{Z} / d$ for any integer $d$, and in some other settings (quotients of Prüfer domains and principal quotients of normal domains). This generalizes Problem B5 of the 2021 Putnam contest. Over arbitrary commutative rings, we identify a stronger property that is always inherited by powers: We say that a matrix $A = \left(a_{i,j}\right)<em>{i,j\in\left[n\right]}$ is strongly $1$-principled if all its diagonal entries are $1$ and if all the cyclic products $a</em>{i_1, i_2} a_{i_2, i_3} \cdots a_{i_k, i_1}$ with $k&gt;1$ vanish. We show that the latter products are always integral over the ideal generated by the principal minors of $A$ minus $1$.

• The paper establishes that if a 1-principled matrix is defined over a reduced or integrally controlled ring, then all its powers remain 1-principled.
• It distinguishes between 1-principled and strongly 1-principled matrices, showing that enforcing diagonal unity and controlling cycle weights guarantees closure under powering.
• The study leverages combinatorial graph theory and integrality arguments to link cycle weights with principal minor defects, providing insights for both theoretical and computational applications.

Matrix Powers with All Principal Minors Equal to 1: Structural and Ring-Theoretic Insights

Introduction

The paper "Powers of matrices with all principal minors equal to 1" (2606.28976) investigates the algebraic and combinatorial structure of $n \times n$ matrices $A$ over commutative rings with the striking property that all principal minors of $A$ equal $1$. The principal focus is on the behavior of these matrices under matrix powering: when does $A^m$ retain the same property for all $m \geq 1$? The analysis generalizes a result stemming from the 2021 Putnam contest and explores a range of commutative rings, highlighting subtle failures and successes of the inheritance of the principal minor property. The work provides algebraic classifications and explicit criteria in terms of ideal-theoretic and integrality conditions.

Definitions and Key Properties

The central definitions are as follows:

• A matrix $A$ is $1$-principled if every principal minor $\det(A_S)$, for any index set $S \subseteq [n]$, equals $A$0.
• A matrix is strongly $A$1-principled if all diagonal entries are $A$2 and every nontrivial cycle $A$3 (in the underlying directed graph of $A$4) has weight $A$5.
• The authors distinguish between $A$6-principled and strongly $A$7-principled matrices, notably establishing that the latter property is strictly stronger and more tractable for inductive and closure operations.

Main Theorems and Algebraic Results

The paper establishes several significant algebraic theorems:

1. Power Inheritance over Fields and Reduced Rings: If $A$8 is $A$9-principled over a reduced (i.e., non-nilpotent) commutative ring, then $A$0 is $A$1-principled for all $A$2. This result also holds trivially over fields and, by extension, over finite quotient rings of the form $A$3.
2. Failure over Non-reduced Rings: There exist explicit counterexamples showing that the property does not generally persist for arbitrary commutative rings with nilpotent elements [(Grinberg, 2022), §6]. Thus, the context of the ambient ring is crucial for the stability property.
3. Main Generalization: For $A$4 over a quotient $A$5 where $A$6 is integrally closed in $A$7, $A$8 is guaranteed to be $A$9-principled for all $1$0. This encompasses the substantial classes:
   • Quotients of Pr\"ufer domains
   • Quotients of normal domains by principal ideals
   • $1$1 for any $1$2

This is formally stated:

If $1$3 is a $1$4-principled $1$5 matrix over $1$6 with $1$7 integrally closed in $1$8, then $1$9 is again $A^m$0-principled for all $A^m$1.

Corollaries provide the analogous results for the special cases above.

1. Structural Results on Cycle Weights: The $A^m$2-weights of nontrivial cycles are shown to be integral over the ideal generated by the defects of the principal minors ($A^m$3), fundamentally tying matrix structure to the algebraic notion of integral closure. In $A^m$4-principled matrices, all nontrivial cycle weights are nilpotent. Over reduced rings, these weights are forced to vanish, providing the equivalence of being $A^m$5-principled and strongly $A^m$6-principled.

Technical Innovations and Combinatorics

Key proofs utilize combinatorial graph theory and algebraic techniques:

• The concept of walks and cycles in the complete digraph $A^m$7 is leveraged to analyze matrix powers and the propagation of the $A^m$8-principled property.
• The authors show that strongly $A^m$9-principled matrices are closed under powering, whereas arbitrary products (even if matrices commute) need not inherit the property, as illustrated by explicit counterexamples over small finite fields.
• The analysis of cycle weights uses elementary symmetric functions and integrality arguments, with induction on the cycle length and essential use of Viète's formula, to embed the problem into the study of integral closures of ideals.

Implications and Applications

Theoretical Significance

This work systematically extends a classical problem motivated by the Putnam competition to a general algebraic setting, uncovering deep connections between combinatorial matrix properties, ring-theoretic invariants, and ideal theory. The characterization of rings (and, ultimately, matrix classes) that inherit the $m \geq 1$0-principled property through powers has implications for understanding invariant subspaces, the closure of matrix properties under semigroup operations, and related questions in algebraic combinatorics.

The study shows that the critical dividing line is not simply reduction modulo ideals, but rather the integrality properties of those ideals and the presence (or lack thereof) of nilpotents. Notably, the analysis gives a clear mechanism: nontrivial cycle weights are "controlled" by the principal minor defects.

Practical and Computational Implications

Although the work is theoretical, insights into the preservation of structural matrix properties under powering are highly relevant to computer algebra systems, matrix group classifications, and computational invariant theory. The explicit identification of permissible ambient rings—such as all $m \geq 1$1 or quotients of valuation domains—provides algorithmically checkable criteria for applications.

Future Developments

Future research directions could include:

• Classification of all (finite) commutative rings for which the principal minor property is power-invariant.
• Extension to noncommutative settings, where principal minors and their generalizations play roles in representation theory and noncommutative invariant theory.
• Investigation of analogous phenomena for matrices with prescribed structure on other minors (not just principal) or for block matrices.
• Algorithmic applications: efficient recognition of strongly $m \geq 1$2-principled matrices in symbolic computation.

Conclusion

This paper delivers a rigorous and comprehensive analysis of the structural stability of matrices with all principal minors equal to $m \geq 1$3 under powers, across a broad landscape of commutative rings. By identifying integrally closed ideals as the essential algebraic criterion for invariance, the authors unify and generalize several prior results and provide a robust combinatorial and algebraic framework. The theoretical implications extend into ideal theory and combinatorial matrix theory, with potential computational applications in algebraic software and further research in generalizing matrix minor constraints.