The eccentric millisecond pulsar, PSR J0955−6150

Context. PSR J0955−6150 is a member of an enigmatic class of eccentric millisecond pulsar (MSP) and helium white dwarf (He WD) systems (eMSPs), whose binary evolution is poorly understood and believed to be strikingly different to that of traditional MSP+He WD systems in circular orbits. Aims. Measuring the masses of the stars in this system is important for testing the different hypotheses for the formation of eMSPs. Methods. We carried out timing observations of this pulsar with the Parkes radio telescope using the 20 cm multibeam and ultra-wide bandwidth low-frequency (UWL) receivers, and the L -band receiver of the MeerKAT radio telescope. The pulse profiles were flux and polarisation calibrated, and a rotating-vector model (RVM) was fitted to the position angle of the linear polarisation of the combined MeerKAT data. Pulse times of arrival (ToAs) were obtained from these using standard pulsar analysis techniques and analysed using the TEMPO 2 timing software. Results. Our observations reveal a strong frequency evolution of this MSP’s intensity, with a flux density spectral index ( α ) of −3.13(2). The improved sensitivity of MeerKAT resulted in a greater than tenfold improvement in the timing precision obtained compared to our older Parkes observations. This, combined with the eight-year timing baseline, has allowed precise measurements of a very small proper motion and three orbital post-Keplerian parameters, namely the rate of advance of periastron, ω̇ = 0.00152(1) deg yr −1 , and the orthometric Shapiro delay parameters, h 3 = 0.89(7) μs and ς = 0.88(2). Assuming general relativity, we obtain M p = 1.71(2) M ⊙ for the mass of the pulsar and M c = 0.254(2) M ⊙ for the mass of the companion; the orbital inclination is 83.2(4) degrees. Crucially, assuming that the position angle of the linear polarisation follows the RVM, we find that the spin axis has a misalignment relative to the orbital angular momentum of > 4.8deg at 99% confidence level. Conclusions. While the value of M p falls well within the wide range observed in eMSPs, M c is significantly smaller than expected from several formation hypotheses proposed, which are therefore unlikely to be correct and can be ruled out; M c is also significantly different from the expected value for an ideal low mass X-ray binary evolution scenario. If the misalignment between the spin axis of the pulsar and the orbital angular momentum is to be believed, it suggests that the unknown process that created the orbital eccentricity of the binary was also capable of changing its orbital orientation, an important evidence for understanding the origin of eMSPs.