Ge stability criteria accidentally applied to WDs

[reinhold-willcox]

Describe the bug
The Ge stability criteria for MT have been calculated for H and He rich stars, but we currently treat WDs according to the same prescription (inadvertently). In lieu of detailed models for WD stability from the Ge team, we should default back to the Soberman stability criteria for WDs.

Label the issue
urgency_low - This issue is not urgent
severity_minor - This is a minor bug with minimal impact

Versioning (please complete the following information):

• COMPAS v03.19.00

[reinhold-willcox]

This should be a straightforward bug fix for an interested newcomer, but otherwise I can tackle this.

[SimonStevenson]

To be clear, this is for WDs as mass donors? Does this only apply to WD, or also NS and BH?

What is the Soberman+ stability criteria for WDs?

In Remnants.h, we currently have

    double          CalculateCriticalMassRatio(const bool p_AccretorIsDegenerate,
                                               const double p_massTransferEfficiencyBeta)                       { return 0.0; }                                                         // Should never be called...

"Should never be called." is an interesting comment 🤣 Are we saying that we do now care about this value. If so, we should remove this or change this value.

We presumably also need to add CalculateCriticalMassRatioGeEtAl which should call this function. Should was also add CalculateCriticalMassRatioClaeys14 to Remnants.h or WhiteDwarfs.h? @nrsegovia has added CalculateCriticalMassRatioHurleyHjellmingWebbink in #1384

Claeys et al. 2014 (https://ui.adsabs.harvard.edu/abs/2014A%26A...563A..83C/abstract) suggest a qcrit of 1.6 for a WD with a degenerate accretor.

If this gets cleared up soon, we can put a fix into #1384 , otherwise we can make a new PR to address this.

[ilyamandel]

"Should never be called." is an interesting comment 🤣

I think it may have been written when we did not consider mass transfer from WDs (stopped evolution when two remnants formed).

In any case, I am not convinced that it makes sense to talk about "stability" for WD donors. They are effectively very cold objects supported by Fermi degeneracy pressure and their Kelvin-Helmholtz is not relevant, they will always respond to mass loss on a dynamical timescale. We therefore set their thermal timescale to be the dynamical timescale; see Remnants.h:

double CalculateThermalTimescale(const double p_Radius = 1.0) const { return CalculateDynamicalTimescale(); } // Parameter is ignored

Nor does it make sense to talk about a "common envelope" from a white dwarf donor in our usual framework. For now, I do not see a viable alternative to treating all mass transfer from WDs as stable (but likely very non-conservative), governed by angular momentum conservation.

Thus, my proposed fix is to always label MT from WDs as stable.

[ilyamandel]

I think we should make the following change in Remnants.h:

double CalculateCriticalMassRatio(const bool p_AccretorIsDegenerate,
const double p_massTransferEfficiencyBeta)
{ return std::numeric_limits::infinity(); }
// Mass transfer from a compact remnant is always labeled as stable since their thermal timescale is set to be their dynamical timescale (and since it doesn't make much sense to talk about a "common envelope" for such a donor, anyway). Happy to make this change in an upcoming PR unless someone tells me otherwise.

[ilyamandel]

In #1386 , we decided that if a WD donor does enter a CE, we shall always call it a merger. So I suppose "stability" here means "does a WD donor survive mass transfer or inevitably merge". If Ge+ don't have a threshold for this, we could indeed default to something else -- perhaps HURLEY_HJELLMING_WEBBINK?

[ilyamandel]

No response from @reinhold-willcox , so I just made our behaviour uniform across all critical mass ratios in #1397.
All critical mass ratios now return the HURLEY_HJELLMING_WEBBINK_QCRIT_WD for white dwarfs and 0 for other remnant donors (only stable mass transfer -- though we should not be handling mass transfer from neutron stars or black holes :-) ). Happy to revise this if there are better suggestions for what to do (@BalexBalex says that the vast majority of ultra-compact binaries with WD donors do merge), but closing for now.

[reinhold-willcox]

Sorry for the lack of response. Thanks for taking care of this Ilya!