The Ultra-Radical: Analytic Continuation, Branching, and Stability of the Principal Branch

TL;DR

This work introduces the ultra-radical as the analytic continuation of the multivalued solution to y^a = 1 + a x y^b, providing a deterministic angular (sector) criterion to select the correct conjugate series for each branch beyond the convergence radius. A key finding is that only the principal branch (n = 0) remains continuously connected under smooth variation of parameters, with the principal branch reducing to classical solutions in the limits a → 0 or b → 0, while other branches exhibit oscillatory divergence. The authors develop a general Master-J framework, including generalizations to arbitrary coefficients and multiple terms via a merge operation, and apply it to nonlinear ODEs and nonlinear circuits, where the ultra-radical offers explicit analytic representations and guaranteed branch continuity. They also establish a deep logarithmic counterpart, the ultralogarithm, linking ultra-radicals to a canonical logarithmic theory and enabling robust algebraic and differential manipulations. The work culminates in practical tools for analytic continuation, integration, and applications in nonlinear physics, providing a unified, computationally friendly approach to multi-valued nonlinear equations and their stable branches.

Abstract

We study the ultra-radical $\sqrt[{n;a;b}]{x}$, the multi-valued solution to $y^{a}=1+a x y^{b}$. Inside the convergence radius $|x|<R$, every branch is given by a Master-J power series; for $|x|\ge R$, analytic continuation requires switching to one of two conjugate series. We introduce a deterministic geometric criterion that selects, for each branch index $n$, the correct conjugate series, thereby eliminating heuristic search and guaranteeing branch continuity across $|x|=R$. Key finding: Only the principal branch ($n=0$) remains continuous when the parameters $a$, $b$, and $x$ vary smoothly. This includes the critical limits $a\to0$ (transition to an exponential equation) and $b\to0$ (transition to a binomial root), where the principal branch converges to the corresponding classical solution. In contrast, branches with $n\neq0$ exhibit oscillatory divergence as $a\to0$ and lose their identity in these limits. This structural continuity singles out the principal branch for applications where parameters may vary with the system's state, such as in nonlinear media with field-dependent exponents or adaptive dynamical systems.