Q Q -difference analogue of the Stothers-Mason theorem
Jian-Tang Lu, Xing-Xing Lu, Zhi-Tao Wen · 2026
Read Q Q -difference analogue of the Stothers-Mason theorem by Jian-Tang Lu, Xing-Xing Lu, Zhi-Tao Wen (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-2605-27876.txt.
Introduction
Jian-Tang Lu, Xing-Xing Lu, and Zhi-Tao Wen extend a classic theorem about polynomial zeros to the world of q-differences, where steps multiply rather than add. They define a new q-weight for zeros and a matching q-radical, recovering the classical Stothers-Mason result as q approaches one. The machinery then handles polynomial solutions of q-difference Fermat-type equations.
Catalog record
| CREATOR | Jian-Tang Lu, Xing-Xing Lu, Zhi-Tao Wen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 29365 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.27876
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Q Q -difference analogue of the Stothers-Mason theorem Jian-Tang Lu, Xing-Xing Lu, Zhi-Tao Wen Abstract In this paper, we give a new definition of the q q -weight of zeros, which reduces to the multiplicity of zeros as q → 1 q\to 1 . Furthermore, we obtain a q q -difference version of the Stothers-Mason theorem by means of the new definition of the q q -difference radical, which covers the classical Stothers-Mason theorem as q → 1 q\to 1 . As applications, we study the polynomial solutions of q q -difference Fermat type functional equations. Keyword : q q -weight of zeros, q q -difference radical, Stothers–Mason theorem, Fermat type functional equation 2020MSC : 39B32, 30D35 1 Introduction Let P be a polynomial. The radical {\rm rad}(P) is the product of distinct linear factors of P . Let a , b and c be relatively prime polynomials such that not all of them are identically zero. The Stothers–Mason theorem [ 12 ] , [ 14 ] , see also [ 13 ] , states that if they satisfy a+b=c , then \max\{\deg(a),\deg(b),\deg(c)\}\leq\deg({\rm rad}(abc))-1. A difference analogue of the Stothers-Mason theorem or difference abc theorem for polynomials was given by Ishizaki et al. in [ 7 , Theorem 3.1] . Let a,b and c be relatively prime polynomials in \mathbb{C}[z] , which are not all constants, satisfying a+b=c . Then, \max\{\deg(a),\deg(b),\deg(c)\}\leq\tilde{n}_{\kappa}(a)+\tilde{n}_{\kappa}(b)+\tilde{n}_{\kappa}(c)-1, where \kappa\in\mathbb{C}\setminus\{0\} , and \tilde{n}_{\kappa}(p)=\sum_{w\in\mathbb{C}}(\text{ord}_{w}(p)-\min\{\text{ord}_{w}(p),\text{ord}_{w+\kappa}(p)\}) for a polynomial p by \text{ord}_{w}(p) denoting the order of zeros of p at z=w . Another difference analogue of the Stothers-Mason theorem was given by Ishizaki and Wen in [ 8 , Theorem 3.5] . Let a,b and c be relatively shifting prime polynomials, see [ 8 , pages 736-737] , such that a,b and c are not all constants satisfying a+b=c . Then \max\{\deg(a),\deg(b),\deg(c)\}\leq\deg(\text{rad}_{\Delta}(abc))-1, where \text{rad}_{\Delta}(abc) is difference radical of abc defined in [ 8 ] . One of the purposes of this paper is to introduce a q -difference analogue of the radical of polynomials and a q -difference analogue of the Stothers-Mason theorem, which are different from the results in [ 3 ] . These q -difference counterparts reduce to the classical radical of polynomials and Stothers-Mason theorem, respectively, as q\to 1 . As applications, we study polynomial solutions of q -difference Fermat type functional equations. In Section 2, we define the q -weight of a zero of an analytic function in a domain G , and investigate its properties via q -difference calculus. We then introduce the q -difference radical by means of the q -weight of zeros and focus on the q -difference analogue of the Stothers-Mason theorem by using the q -difference radical in Section 3. Section 4 is concerned with an extension of the aforementioned q -difference Stothers-Mason theorem. Finally, in Section 5 we study the polynomial solutions of q -difference version of the Fermat type functional equations. 2 Q -weight of zeros Following [ 11 ] , we denote q -number which is the q -analogue of the natural number n by [n]_{q}=\frac{q^{n}-1}{q-1}=q^{n-1}+\cdots+q+1. We set the q -analogue of factorial n! by [n]_{q}!=\begin{cases}1\quad&n=0;\\ [n]_{q}\times[n-1]_{q}\times\cdots\times[1]_{q}\quad&n\in\mathbb{N}^{+}.\end{cases} Let us set the q -analogue of power function (z-a)^{n} by [z-a]^{n}_{q}=\begin{cases}1\quad&n=0;\\ (z-a)(z-aq)\cdots(z-aq^{n-1})\quad&n\in\mathbb{N}^{+}.\end{cases} We note that if a=0 , then [z]^{n}_{q}=z^{n} for all n\in\mathbb{N}^{+}\cup\{0\} . For a given function f , we define q -derivative of f [ 9 ] by D_{q}f(z)=\frac{f(qz)-f(z)}{qz-z} for z\in\mathbb{C}\setminus\{0\} and q\neq 1 . In addition, q -derivative of f at zero is defined by D_{q}f(0)=\lim_{n\to\infty}\frac{f(z^{\prime}q^{n})-f(0)}{z^{\prime}q^{n}} for z^{\prime}\in\mathbb{C}\setminus\{0\} . Note that if f(z) is analytic at z=0 , then D_{q}f(0)=f^{\prime}(0) . Moreover, if f is an analytic function, then \lim_{q\to 1}D_{q}f=f^{\prime} , and D_{q} is known as the Jackson operator , see e.g., [ 10 , Page 41] . For any n\in\mathbb{N}^{+} , set D_{q}^{n}f=D_{q}^{n-1}(D_{q}f) and D_{q}^{0}f=f . Obviously, for all n\in\mathbb{N}^{+} we have D_{q}([z-a]_{q}^{n})=[n]_{q}[z-a]^{n-1}_{q}, which is q -analogue of differential calculus ((z-a)^{n})^{\prime}=n(z-a)^{n-1} . It is a fact that D_{q} reduces the degree of a polynomial by one. Let f be analytic in a domain G\subset\mathbb{C} and let n\in\mathbb{N} and q\in\mathbb{C}\setminus\{0\} . Suppose that z_{0},qz_{0},\ldots,q^{n}z_{0}\in G . The point z_{0} is called a zero of f(z) with q -weight n in G provided f(z_{0}) and all q -derivative D_{q}^{k}f(z_{0}) vanish for all 0\leq k<n , but D_{q}^{n}f(z_{0})\neq 0 . In general, the point z_{0} is called an a -point of f(z) with q -weight n in G provided f(z_{0})-a and all q -derivative D_{q}^{k}f(z_{0}) vanish for all 0\leq k<n , but D_{q}^{n}f(z_{0})\neq 0 . In particular, if z_{0} is an a -point of f with q -weight 1, then we also call z_{0} is an a -point of f with simple q -weight. Obviously, z_{0}=0 being a zero of f with q -weight n is equivalent to z_{0}=0 being a zero of f with multiplicity n . We emphasize here that q -weight of zeros is a generalization of multiplicity of zeros of an analytic function. As q\to 1 , we see that q -weight of zeros reduces to the multiplicity of zeros of an analytic function. Example 2.1 Suppose that n\in\mathbb{N}^{+} and a,q\in\mathbb{C} such that q\neq 0 , and f(z)=[z-a]_{q}^{n} . Then we have D_{q}^{k}f(a)=0 for any k\in\mathbb{N} satisfying 0\leq k<n , but D_{q}^{n}f(a)\neq 0 . Thus, it is obvious that z=a is a zero of f with q -weight n . In addition, we see that z=a becomes a zero of f with multiplicity n as q\to 1 . The following example illustrates that a transcendental function may have a zero with infinite q -weight. Example 2.2 Suppose that 0<|q|<1 , and a function f(z)=\prod_{n=0}^{\infty}(1+q^{n}z), which converges uniformly on any compact subset of \{z\in\mathbb{C}:|z|<R\} . The entire function f(z) is a solution of linear q -difference equations f(qz)-(1+qz)f(z)=0. In addition, f is of logarithmic order \rho_{\log}(f)=2 , see [ 15 , Theorem 1.1] . The logarithmic order is defined by \rho_{\log}(f):=\limsup_{r\to\infty}\frac{\log\log M(r,f)}{\log\log r}. Obviously, the distinct zero sequence of f is z_{n}=-(1/q)^{n} for n=0,1,\ldots . Thus, z_{0}=-1 is a zero of f with infinite q -weight. It is well known that z=z_{0} is a zero of an analytic function f with multiplicity n if and only if f is of the form f(z)=(z-z_{0})^{n}g(z) , where g(z) is analytic at z=z_{0} and g(z_{0})\neq 0 . The q -analogue of this result is given as follows. Theorem 2.1 Let f be analytic in the domain G\subset\mathbb{C} and let n\in\mathbb{N} and q\in\mathbb{C}\setminus\{0\} . Suppose that z_{0},qz_{0},\ldots,q^{n}z_{0}\in G . The point z_{0} is a zero of f(z) with q -weight n in G if and only if there exists an analytic function g(z) in G such that f(z)=[z-z_{0}]_{q}^{n}g(z) (2.1) and g(q^{n}z_{0})\neq 0 . Before the proof of Theorem 2.1 , let us recall a lemma, see e.g., [ 2 , Pages 12-13] , which reveals the relation between f(q^{n}z) and D_{q}^{n}f(z) for n\in\mathbb{N} . Lemma 2.2 For any z\in\mathbb{C} , k\in\mathbb{N} , we have f(q^{k}z)=\sum_{j=0}^{k}q^{j(j-1)/2}(qz-z)^{j}\left[\begin{array}[]{cc}k\\ j\end{array}\right]_{q}D_{q}^{j}f(z), where \left[\begin{array}[]{cc}k\\ j\end{array}\right]_{q}=\frac{[k]_{q}!}{[j]_{q}![k-j]_{q}!} are called q -binomial coefficients. In addition, for z\in\mathbb{C}\setminus\{0\} and k\in\mathbb{N} , we have D_{q}^{k}f(z)=\frac{1}{q^{k(k-1)/2}(qz-z)^{k}}\sum_{j=0}^{k}(-1)^{k-j}q^{(k-j)(k-j-1)/2}\left[\begin{array}[]{cc}k\\ j\end{array}\right]_{q}f(q^{j}z). Proof of Theorem 2.1 . It is easy to get our assertion when q=1 . In the following, we assume that q\neq 1 . Let z_{0}\neq 0 be a zero of f with q