Explaining observations of rapidly rotating neutron stars in LMXBs

TL;DR

This paper presents a detailed scenario explaining the stability and spin evolution of rapidly rotating neutron stars in LMXBs, emphasizing the role of superfluid mode interactions and CFS instability, and predicts new observable classes.

Contribution

It introduces a novel resonance-based stability mechanism involving superfluid modes that explains observed neutron star spins and predicts new neutron star classes.

Findings

Neutron stars spend significant time on stability peaks near resonance temperatures.

Spin frequencies are limited by m=3 r-mode instability, not m=2.

Scenario predicts hot, rapidly rotating nonaccreting neutron stars ('HOFNARs').

Abstract

In a previous paper [M. E. Gusakov, A. I. Chugunov, and E. M. Kantor, Phys. Rev. Lett. 112, 151101 (2014)], we introduced a new scenario that explains the existence of rapidly rotating warm neutron stars (NSs) observed in low-mass X-ray binaries (LMXBs). Here it is described in more detail. The scenario takes into account the interaction between superfluid inertial modes and the normal (quadrupole) $m=2$ $r$-mode, which can be driven unstable by Chandrasekhar-Friedman-Schutz (CFS) mechanism. This interaction can only occur at some fixed "resonance" stellar temperatures; it leads to formation of the "stability peaks" which stabilize a star in the vicinity of these temperatures. We demonstrate that a NS in LMXB spends a substantial fraction of time on the stability peak, that is, in the region of stellar temperatures and spin frequencies, that has been previously thought to be CFS unstable with respect to excitation of $r$-modes. We also find that the spin frequencies of NSs are limited by the CFS instability of normal (octupole) $m=3$ $r$-mode rather than by $m=2$ $r$-mode. This result agrees with the predicted value of the cutoff spin frequency $\sim 730$ Hz in the spin distribution of accreting millisecond X-ray pulsars. In addition, we analyze evolution of a NS after the end of the accretion phase and demonstrate that millisecond pulsars can be born in LMXBs within our scenario. Besides millisecond pulsars, our scenario also predicts a new class of LMXB descendants - hot and rapidly rotating nonaccreting NSs ("hot widows"/HOFNARs). Further comparison of the proposed theory with observations of rotating NSs can impose new important constraints on the properties of superdense matter.