[mesa-users] Approx21

[Ian Foley]

Hi Bill and Rob,

Thanks heaps for your insightful answers. Really helped in many ways.

Now appreciate the goal of approx21 to get energy right and set final Ye. I
now understand the focus on the alpha chain reactions and the use of
compound rates.

I successfully tried the mesa_80.net, but found that it does not burn the
fe, so you end up with near 100% fe at the center. I added f20 and cr 51 56
with success in creating fe infall termination. However, I did not look at
getting the energy right or the final Ye, so I'm not sure if my mesa_87.net
needs further refinement. The simulation with this net took more than 10
times longer than with approx21 at 15MSun.

Overall, my goal has been to teach myself about stellar evolution by
working through all star sizes from pre_ms to end point. But I have been
stuck in the 9 to 13 MSun range with z=0.02. Many of these end up as WD
which is OK. But reading the literature there should be a path in some
cases where core collapse leads to electron capture supernova, but I have
not been able to achieve this yet. It seems one component of achieving this
is electron capture reactions on Mg24, Na23 and f20 which was why I am
interested in the details of the composition of the various nets. approx21
will not achieve this.

I have tried the two sagb nets and they work OK for the 9-11 MSun range,
but all my simulations ended up with cores less than 1.37 MSun, so I think
I will need to look at binary stars and mass accretion to get to core
collapse.

However, in the mass range 12-13 MSun, there was no adequate net in mesa
because O-Ne core ignites and wants to go to si28 and the sagb nets do not
go far enough. Only from 14MSun and larger, does fe infall termination
occur because smaller stars do not get hot enough to ignite si28. The
approx21 net is not adequate because it does not have the required electron
capture reactions to assist in creating the core collapse. So I have
created a net be adding onto sagb up to ca40 and it seems to work OK, but I
do not have the knowledge to test it thoroughly.

So I think you guys are right about there being a sterile desert of nets
between 30 and 150 iso's. The two sagb nets partially fill the gap, but
they are inadequate for 12-13 MSuns stars with z=0.02. The mesa_49.net
might do; I need to try it, but wanted to cut out the less important
reactions if possible. Not sure about the gap for other z's. And it would
be nice to have a net which will work up to fe core collapse without the
computational weight of mesa_80.net and above; something like the sagb nets
with all the key reactions specified.

Thanks heaps for your answers,
Kind regards
Ian





On 17 October 2016 at 15:59, Robert Farmer <rjfarmer at asu.edu> wrote:

> Hi
>
> >In simulations up to Fe core collapse, the primary central isotope which
> remains at the end is cr56 which in my reading conflicts with research.
>
> The cr56 in approx nets is not "real" cr56, it is there to set the final
> ye value. If you look in the net folder there are many approx21_* nets,
> each ones picks a different set of final isotopes (but all use the same
> fudge factored reaction rate) so you can set your desired final ye.
>

>. There are some reactions which contain isotopes such as Fe53 and Fe55
> which are not in the isotope list. This will not work for data files
>
> If you can assume you species is in equilibrium (for instance fe54 <> fe55
> <> fe56) then with a bit of maths you can remove the explicit dependence on
> the amount of fe55 by creating a compound rate based on each of the
> components in the equilibrium chain. If you are making your own net then
> you'll have to add them in yourself, the approx nets have many hard coded
> assumptions.
>
> net/rpivate/net_approx21_procs.inc approx21_special_reactions subroutine
> for all the details
>
> >3. The isotope Na23 is missing. In intermediate stars, research shows the
> main ingredients of the degenerate cores are o16, Ne20, Na23 and Mg24.
>
> The approx nets are only meant to track the energy generation, by
> following alpha chain reactions. Whether the composition is correct is
> irrelevant(-ish) we only want the energy "right", either you need to make
> your own net with isotopes you think are important (or are needed as
> intermediate steps) or post process the model to add in missing reactions
> afterwards (if you can assume they make negligible contributions to the
> energy budget)
>
> I find approx21_cr60_plus_co56.net
> <https://sourceforge.net/p/mesa/code/HEAD/tree/trunk/data/net_data/nets/approx21_cr60_plus_co56.net>
> gets to CC well (though choice of inlist settings is crucial for getting
> there efficiently, especially during si burning). For other nets i'd look
> at the mesa_* nets with at least 80 isotopes to start to get things going
> the "right" way composition wise.
>
> Rob
>
> On Sun, Oct 16, 2016 at 9:23 PM, Ian Foley <ifoley2008 at gmail.com> wrote:
>
>> Hi,
>>
>> Love to get some more insight into this net. It's no longer produced by a
>> data file, but by code. It enables a simple net with broad brush treatment
>> allowing simple simulations up to core collapse.
>>
>> But
>> 1. In simulations up to Fe core collapse, the primary central isotope
>> which remains at the end is cr56 which in my reading conflicts with
>> research.
>>
>> 2. There are some reactions which contain isotopes such as Fe53 and Fe55
>> which are not in the isotope list. This will not work for data files
>>
>> 3. The isotope Na23 is missing. In intermediate stars, research shows the
>> main ingredients of the degenerate cores are o16, Ne20, Na23 and Mg24.
>>
>> So I've been looking for ways to create a simple net which does not have
>> a big performance hit, but gives results closer to research.
>>
>> I also wanted to find a net which can simulate up to core collapse
>> leading to electron capture supernovas. Are there any in mesa which can do
>> this?
>>
>> Kind regards
>> Ian
>>
>> ------------------------------------------------------------
>> ------------------
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
>> _______________________________________________
>> mesa-users mailing list
>> mesa-users at lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/mesa-users
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20161017/8e26a62f/attachment.html>