Riemann hypothesis and quantum mechanics

In their 1995 paper, Jean-Beno\^{i}t Bost and Alain Connes (BC) constructed a quantum dynamical system whose partition function is the Riemann zeta function $\zeta(\beta)$, where $\beta$ is an inverse temperature. We formulate Riemann hypothesis (RH) as a property of the low temperature Kubo-Martin-Schwinger (KMS) states of this theory. More precisely, the expectation value of the BC phase operator can be written as $$\phi_{\beta}(q)=N_{q-1}^{\beta-1} \psi_{\beta-1}(N_q), $$ where $N_q=\prod_{k=1}^qp_k$ is the primorial number of order $q$ and $ \psi_b $ a generalized Dedekind $\psi$ function depending on one real parameter $b$ as $$ \psi_b (q)=q \prod_{p \in \mathcal{P,}p \vert q}\frac{1-1/p^b}{1-1/p}.$$ Fix a large inverse temperature $\beta >2.$ The Riemann hypothesis is then shown to be equivalent to the inequality $$ N_q |\phi_\beta (N_q)|\zeta(\beta-1) >e^\gamma \log \log N_q, $$ for $q$ large enough. Under RH, extra formulas for high temperatures KMS states ($1.5< \beta
Comment: version to appear in J. Phys. A: Math. Theor