2025ApJ...987..171X

We study the secular periodic evolution of quasiperiodic eruptions (QPEs) for GSN069 and eRO-QPE2 assuming that they are driven by star-disk collisions. We set up numerical simulations and compared them with the observed periodic decay of ∼-3160 ± 720 s yr^–1 in GSN069 and ∼-370 ± 40 s yr^–1 in eRO-QPE2. We find the following: (1) Stellar-mass black holes are unlikely to be the orbiters in these two sources, as their periodic decays are on the order of <10 s yr^–1. (2) A naked degenerate core (including a white dwarf) is unlikely to be the orbiter in GSN069, as the decay is on the order of <200 s yr^–1. However, it is possible in eRO-QPE2, although the required surface density of the accretion disk is relatively high (e.g., Σ ≳ 10⁷-10⁸ g cm^–2). (3) Both the orbiters in GSN069 and eRO-QPE2 can be solar-like main-sequence (MS) stars. However, each collision can lead to gradual ablation of the stellar envelope on the order of 10^–5 to 10^–3M_⊙. To reproduce the observed decay while surviving for ≳3 yr, the surface density of the disk needs to be within a certain range. For example, given a 1M_⊙ MS orbiter, the surface density of the disk gas should be in the range of 3 × 10⁵ g cm^–2 to 2 × 10⁶g cm^–2 for GSN069 or 5 × 10⁴ g cm^–2 to 10⁶ g cm^–2 for eRO-QPE2. In both of these sources, the MS star cannot survive for more than ∼12 yr. We expect that future observations of these two sources can help to distinguish whether the orbiters are degenerated compact objects or gaseous stars.