Thermodynamic limit in number theory: Riemann-Beurling gases (Q1330320): Difference between revisions

As mentioned by the author, it is generally recognized nowadays that the Riemann zeta function is quite naturally the canonical partition function of a free bosonic gas (the localization of the pole at \(s=1\) being associated with the Hagedorn temperature). In the paper the grand canonical version of a well-known statistical model, the Riemann gas -- which consists of a collection of bosonic oscillators with energies the logarithms of the prime numbers -- is studied. The introduction of the chemical potential \(\mu\) amounts to multiplying each prime number by \(e^{-\mu}\) and the author chooses to call the resulting gas a Beurling one, because these gases are defined by an appropriate choice of generalized primes, when considered as forming canonical ensembles. For these gases, generalized Hagedorn singularities in the temperature are found. The usual approximation is used in the paper, that in the high- energy region the spectrum can be treated as continuous, what allows the author to study the level density in that region. This idea is applied specifically to the Beurling gases by the author. The case of the logarithmic gas is studied in the paper in more detail. It corresponds to the choice of all the integers strictly larger than one as Beurling primes. An explicit formula is obtained in the paper for the corresponding grand-canonical thermodynamic potential \(F-\mu N\) in terms of a hypergeometric function and a check of the approximation on the Hagedorn phenomenon is carried out. Related physical situations include the string theory approximation to QCD [see for instance \textit{E. Elizalde} et al., Phys. Rev. D 48, 1757- 1767 (1993)] and also the quark deconfinement transition. In both cases a much better understanding of the specific nature of the associated Hagedorn transitions is badly needed. An exciting open question is then to perform explicitly the analytic continuation across the Hagedorn singularity and see in detail to what kind of known phase transition (if any) this continuation actually corresponds.