Phase variation and angular momentum of the Riemann and Dirichlet
ξ · 2025
Read Phase variation and angular momentum of the Riemann and Dirichlet by ξ (2025) 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-2501-10826.txt.
Introduction
Giovanni Lodone borrows a notion from physics, angular momentum, to study two classic functions of number theory. The functions, tied to Riemann and to Dirichlet, sit at the heart of deep open questions such as the Riemann hypothesis. Lodone tracks how the phase of these functions varies and reads that motion through the angular momentum lens. From this view he draws new statements that are equal in force to the famous hypothesis itself. He also stretches some known results from the Riemann case to the wider Dirichlet family.
Catalog record
| CREATOR | ξ |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 19 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-10826
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Phase variation and angular momentum of the Riemann and Dirichlet ξ Abstract \setcaptionmargin 1cm MSC-Class : 11M06, 11M26, 11M99 Keywords : Riemann Hypothesis ; Generalized Riemann Hypothesis 1. Introduction In [ 20 ] and [ 21 ] it is used a function called, angular momentum ( 2.2 ), in order to not reinvent existent named quantities, to attempt an RH proof for two types of Dirichlet function: Riemann \zeta(s) function, and, L(s,\chi_{odd\ primitive}) Dirichlet functions. See ( [ 7 ] , [ 2 ] , and, [ 29 , p. 1] ). In this paper we show that this concept is useful also to give an unitary treatment of some aspects of Dirichlet functions. Besides some known results about Riemann \xi function can be extended easily to primitive Dirichlet function. We apologize for unavoidable overlapping with previous papers. We express the complex argument s , as usual, as: (1.1) s=\frac{1}{2}+\epsilon+it where the real values \epsilon and t are respectively the distance from the critical line ( i.e \Re(s)=1/2 ) and the imaginary part of s , while, 0<\Re(s)<1 is the critical strip. As well known Rieman \zeta(s) function, [ 27 ] (see also [ 3 ] , [ 5 ] , [ 8 , p. 229] ), is defined as the analytic continuation of the function: (1.2) \zeta(s)=\sum_{n=1}^{\infty}\frac{1}{n^{s}}=\prod_{\forall p}\frac{1}{1-\frac{1}{p^{s}}}\quad\mbox{ ($p$ prime)}\quad\quad\Re(s)>1 [RH ( Riemann Hypothesis) ] \equiv [ the analytic continuation of ( 1.2 ) has all non trivial zeros, s_{0} , on critical-line] . Using Dirichlet characters \chi(n) , L functions ( [ 7 ] and [ 2 ] , [ 29 , p. 1] ) are defined as : (1.3) \L(s.\chi)=\sum_{n=1}^{\infty}\ \chi(n)/n^{s}=\prod_{\forall p}\frac{1}{1-\frac{\chi(p)}{p^{s}}}\quad p\quad prime\quad,\quad\quad\quad\Re(s)>1 [GRH (Generalized Riemann Hypothesis)] \equiv [ all the zeros inside critical strip of the ( 1.3 ), or of their analytic continuation, are on critical-line] , i.e \forall s_{0} inside critical strip :\Re(s_{0})=1/2 for ( 1.2 ) and ( 1.3 ). Because of ( 1.5 ) we focus mainly on \chi_{primitive} , defined in [ 2 , p. 168] . Companion functions \xi(s) , ( [ 7 , p. 62] ), and \xi(s,\chi_{primitive}) , ( [ 7 , p. 71] ), are defined : (1.4) \xi(s)=\Gamma\left(\frac{s}{2}+1\right)(s-1)\frac{\zeta(s)}{\pi^{s/2}}\ ;\xi(s,\chi)=\left(\frac{q}{\pi}\right)^{\frac{s+\alpha}{2}}\Gamma\left(\frac{s+\alpha}{2}\right)L(s,\chi) where q=congruences\quad modulus , with \alpha=0 if \chi_{primitive}(-1)=1 , even character, and, \alpha=1 if \chi_{primitive}(-1)=-1 , odd character. Both have the same zeros of the \zeta(s) or L(s,\chi) functions respectively in the critical strip [ 27 ] [ 7 ] [ 8 ] . Following [ 5 , p. 14] , and [ 7 , p. 71] the functional equation for \xi(s) and \xi(s,\chi) can be defined : (1.5) \quad\xi(1-s)=\xi(s)\ \ \ ;\ \ \ \xi(1-s,\bar{\chi})=\frac{i^{\alpha}q^{\frac{1}{2}}}{\tau(\chi)}\xi(s,\chi)\quad with\quad\chi_{primitive}\ \ \quad Where \bar{\chi} is the complex conjugate of \chi and \tau(\chi) is the gauss sum \tau(\chi):=\sum_{m=1}^{k}\chi(m)e^{2\pi im/k} see [ 7 , p. 65] , or [ 2 , p. 165] . For [ 2 , p. 168] , and, [ 7 , p. 66] : |\tau(\chi)|^{2}=q where q is the arithmetic congruence modulus, ( 1.5 ) leads to mirror symmetry around \Re s=1/2 and real axis for \xi(s) : (1.6) \xi\left(\frac{1}{2}+\epsilon+it\right)=\bar{\xi}\left(\frac{1}{2}-\epsilon+it\right) Instead if L(s,\chi) ( 1.3 ) or \xi(s,\chi) have off-critical line zeros they are bound to respect in general only mirror symmetry with respect to critical line, i.e. a zero in (t,\epsilon) means a zero in (t,-\epsilon) . See [ 7 , p. 83] . In [ 7 , p. 80] and in [ 7 , p. 82] similar expressions are given for Hadamard infinite product of ( 1.4 ): (1.7) \xi(s)=e^{g(s)}\prod_{\rho\hskip 8.19447ptpaired}\left(1-\frac{s}{\rho}\right)e^{s/\rho}\quad;\quad\xi(s,\chi)=e^{g(s,\chi)}\prod_{|\Im\rho|<T\rightarrow\infty}\left(1-\frac{s}{\rho}\right)e^{s/\rho} where g(z,\chi)=A(\chi)+B(\chi)z . (1.8) For\quad\xi(s):\ \ B=-\sum_{m=-\infty}^{m=+\infty}\frac{1}{\rho_{m}}\quad;\quad For\quad\xi(s,\chi):\ \ \Re[B(\chi)[=-\sum_{\Im\rho_{m}>-T}^{\Im\rho_{m}<T}\Re\left[\frac{1}{\rho_{m}}\right]\ \ ,\ T\rightarrow\infty Throughout we attribute to \xi(s.\chi) and to \xi\left(\frac{1}{2}+it,\chi\right) and \xi\left(t,\epsilon,\chi\right) the same meaning. As at \xi(t,\chi) and to \xi(t,\epsilon=0,\chi) ,or at \xi(t) and to \xi(t,\epsilon=0) . This article is structured as follows. • In section 2 we define our angular momentum \mathcal{L} connecting it with the phase variation along t . • In section 3 we give some equivalence theorems linking the RH or GRH to the behavior of \mathcal{L} . 2. Angular Momenta 2.1. Phase variations and Angular Momentum of the \xi function Derivative of the phase of \xi(s) for constant \epsilon with respect to t : (2.1) \displaystyle\frac{\partial}{\partial t}\arctan\left(\frac{\Im\xi(s)}{\Re\xi(s)}\right) \displaystyle= \displaystyle\frac{1}{1+\left(\frac{\Im\xi(s)}{\Re[\xi(s)]}\right)^{2}}\frac{\frac{\partial\Im\xi(s)}{\partial t}\Re\xi(s)-\frac{\partial\Re\xi(s)}{\partial t}\Im\xi(s)}{\left(\Re\xi(s)\right)^{2}} \displaystyle= \displaystyle\frac{\frac{\partial\Im\xi(s)}{\partial t}\Re\xi(s)-\frac{\partial\Re\xi(s)}{\partial t}\Im\xi(s)}{\left(\Re\xi(s)\right)^{2}+\left(\Im\xi(s)\right)^{2}}=\frac{\partial\angle[\xi(s)]}{\partial t} Where \angle[\psi] means phase of \psi . The numerator, that determines the sign of ( 2.1 ), can be seen as the angular momentum with respect to the origin of an unitary mass positioned in (\Re\xi(s),\Im\xi(s)) at time t for constant \epsilon . We will name it \mathcal{L}[\xi(s)] , so that: (2.2) \mathcal{L}[\xi(s)]=\det\left(\begin{matrix}\Re\xi(s)&\Im\xi(s)\\ \frac{\partial}{\partial t}\Re\xi(s)&\frac{\partial}{\partial t}\Im\xi(s)\end{matrix}\right) Since \Im\xi=0 for \epsilon=0 , we have \mathcal{L}[\xi(s)]=0 on the critical line. 2.2. First lemma on angular momentum If A(s) is a derivable function and F is a complex constant, then: (2.3) \mathcal{L}[FA(s)]=|F|^{2}\mathcal{L}[A(s)] PROOF: \angle[F\ A]=\angle[F]+\angle[A] so \frac{\partial\angle[F\ A]}{\partial t}=\frac{\partial\angle[A]}{\partial t} .From ( 2.1 ) ( 2.3 ) follows. Let us apply First Lemma to 1.5 ( remember \chi_{primitive} ). For [ 7 , p. 66] |\tau(\chi)|^{2}=q , then \left|\frac{i^{\alpha}q^{\frac{1}{2}}}{\tau(\chi)}\right|=1 . But, at \epsilon=0 , for ( 1.4 ), and ( 1.5 ): \xi(\bar{s},\bar{\chi})e^{-\frac{i}{2}\angle\left[\frac{i^{a}q^{\frac{1}{2}}}{\tau(\chi)}\right]}=\xi(s,\chi)e^{\frac{i}{2}\angle\left[\frac{i^{a}q^{\frac{1}{2}}}{\tau(\chi)}\right]}=\eta(t)\ \ \in\Re So: \mathcal{L}[\xi(t,\epsilon=0,\chi)]=\left|e^{+\frac{i}{2}\angle\left[\frac{i^{a}q^{\frac{1}{2}}}{\tau(\chi)}\right]}\right|^{2}\mathcal{L}[\eta(t)]=0 . 2.3. Second lemma on angular momentum If A(s) and F(s) are derivable functions, then: (2.4) \mathcal{L}[F(s)A(s)]=|F(s)|^{2}\mathcal{L}[A(s)]+|A(s)|^{2}\mathcal{L}[F(s)] From ( 1.7 ): \xi(s)=Z^{T}(z)E^{T}(z)=\prod_{\rho}\left(1-\frac{z}{\rho}\right)E^{T}(z) ,where [ 8 , p. 39,46] , E^{T}(z)=e^{g(z)}e^{\sum_{m=-N}^{N}\frac{z}{z_{m}}}=\xi(0) , so, \mathcal{L}\left[E^{T}(z)\right]=0 . Instead: (2.5) \xi(z,\chi)=Z^{T}(z)\times E^{T}(z,\chi)=\left[\prod_{m=-N^{\prime}}^{N}\left(1-\frac{z}{z_{m}}\right)\right]\times\left[e^{g(z,\chi)}e^{\sum_{m=-N^{\prime}}^{N}\frac{z}{z_{m}}}\right]\quad;T\rightarrow\infty Where N are the zeros within 0<t<T while N^{\prime} are the zeros within 0>t>-T . For Dirichlet primitive characters g(z)=A(\chi_{primitive})+B(\chi_{primitive})z . Where Z^{T}(z)=\prod_{m=-N^{\prime}}^{N}\left(1-\frac{z}{z_{m}}\right) , and E^{T}(z,\chi)=e^{g(z,\chi)}e^{\sum_{m=-N^{\prime}}^{N}\frac{z}{z_{m}}} . But we state also that : \mathcal{L}\left[E^{T}(z,\chi)\right]=\mathcal{L}\left[e^{g(z,\chi)}e^{\sum_{m=-N^{\prime}}^{N}\frac{z}{z_{m}}}\right]=0 In other words E^{T}(z,\chi) factor behaves like a constant in ( 2.4 ) for \xi(s,\chi) as well as E^{T}(z) for \xi(s) . PROOF: from ( 2.4 ), for \mathcal{L}[Z^{T}E^{T}(z,\chi)] we have (dropping : ..[(z,\chi)].. ) (2.6) \mathcal{L}(Z^{T}E^{T})=det\left(\begin{matrix}\Re[Z^{T}]&\Im[Z^{T}]\\ \Re\left[\frac{\partial}{\partial y}Z^{T}\right]&\Im\left[\frac{\partial}{\partial y}Z^{T}\right]\end{matrix}\right)\left\{\Re^{2}[E^{T}]+\Im^{2}[E^{T}]\right\}+.. ..+\left\{\Re^{2}[Z^{T}]+\Im^{2}[Z^{T}]\right\}det\left(\begin{matrix}\Re[E^{T}]&\Im[E^{T}]\\ \Re\left[\frac{\partial}{\partial y}E^{T}\right]&\Im\left[\frac{\partial}{\partial y}E^{T}\right]\end{matrix}\right) We want to show that det\left(\begin{matrix}\Re[E^{T}]&\Im[E^{T}]\\ \Re\left[\frac{\partial}{\partial y}E^{T}\right]&\Im\left[\frac{\partial}{\partial y}E^{T}\right]\end{matrix}\right)=0 for \xi(s,\chi) . Applying ( 1.8 ): (2.7) \frac{\partial}{\partial y}E^{T}=\frac{\partial}{\partial y}\left[e^{g(z,\chi)}e^{\sum_{m=-N^{\prime}}^{N}\frac{z}{z_{m}}}\right]=e^{g(z,\chi)+\sum_{m=-N^{\prime}}^{N}\frac{z}{z_{m}}}\left(iB(\chi)+\sum_{m=-N}^{N}\frac{i}{z_{m}}\right)=E^{T}\left(iB(\chi)+\sum_{m=-N^{\prime}}^{N}\frac{i}{z_{m}}\right) =E^{T}\left\{-\Im\left[B(\chi)+\sum_{m=-N^{\prime}}^{N}\frac{1}{z_{m}}\right]+i\times\left(\Re\left[B(\chi)+\sum_{m=-N^{\prime}}^{N}\frac{1}{z_{m}}\right]\right)\right\}= E^{T}\left\{-\Im\left[B(\chi)+\sum_{m=-N^{\prime}}^{N}\frac{1}{z_{m}}+i\times 0\right]\right\}=E^{T}\ C . C\in\Re , so : det\left(\begin{matrix}\Re[E^{T}]&\Im[E^{T}]\\ \Re\left[\frac{\partial}{\partial y}E^{T}\right]&\Im\left[\frac{\partial}{\partial y}E^{T}\right]\end{matrix}\right)=det\left(\begin{matrix}\Re[E^{T}]&\Im[E^{T}]\\ \Re\left[CE^{T}\right]&\Im\left[CE^{T}\right]\end{matrix}\right)=0 . As \Re[C\times E^{T}]=C\Re[E^{T}] , and, \Im[C\times E^{T}]=C\Im[E^{T}] . So, for both \xi(s) and \xi(s,\chi_{primitive}) , only determinant with Z^{T} in ( 2.6 ) is left, as T\rightarrow\infty . 3. Some RH-GRH equivalences The infinite products 1.7 uses the zeros of the functions \xi(s)\ ,\ \xi(s,\chi) and the convergence in the expansion is ensured if the symmetrical factors with respect to the real axis are joined together, that is if the zeros of the form \rho and 1-\rho are paired for \xi(s) , or are at least included symmetrically for \xi(s,\chi) . 3.1. Third lemma on angular momentum If all the zeros of the \xi function are on the critical line then the sign of the angular momentum equals the sign of \epsilon . That is: (3.1) \forall t:\text{sign}(\mathcal{L}[\xi(t,\epsilon)])=\text{sign}(\epsilon)\quad and\ \forall t:\text{sign}(\mathcal{L}[\xi(t,\epsilon,\chi_{primitive)}])=\text{sign}(\epsilon) PROOF: to compute ( 2.4 ) we can write : (3.2) \frac{\partial\xi(s)}{\partial t}=E^{T}(s)\frac{\partial Z^{T}(s)}{\partial t}=\xi(s)\sum_{\rho}\frac{-i}{\rho-\frac{1}{2}-\epsilon-it}=\xi(s)\sum_{\rho}\frac{(t-\Im[\rho)]+i(\frac{1}{2}+\epsilon-\Re[\rho])}{|\rho-s|^{2}} For \xi(s) because E^{T}(s) is constant, but, only in order to compute ( 2.4 ), also for \xi(s,\chi) , even if E^{T}(s,\chi) is not constant (Lemma 2). Taking now the real and imaginary parts to compute the angular momentum: (3.3) \mathcal{L}[\xi(s)]=\det\left(\begin{matrix}\Re[\xi]&\Im[\xi]\\ \sum_{\rho}\frac{\Re[\xi](t-\Im\rho)-\Im[\xi](\frac{1}{2}+\epsilon-\Re\rho)}{|\rho-s|^{2}}&\sum_{\rho}\frac{\Re[\xi](\frac{1}{2}+\epsilon-\Re\rho)+\Im[\xi](t-\Im\rho)}{|\rho-s|^{2}}\end{matrix}\right) so that, after cancellations: (3.4) \mathcal{L}[\xi(s)]=|\xi(s)|^{2}\sum_{\rho}\frac{\frac{1}{2}+\epsilon-\Re[\rho]}{|\rho-s|^{2}} Notice that