[mesa-users] Fwd: 35Cl and K39

[Roni Waldman]

 Hi Dave,

I’ve also experimented in MESA with a network devised with the help of Jim Truran for SN explosive burning (attached).

 

Thanks for sending it, and see below. 

 

In this case I got stuck with an excess of Si30, S34 and so on. Do you have an idea why?

 

Probably real? You have enough nuclei to get that right.  K40, K41 might help, and I would

wonder about the bottleneck to the Fe peak (is it by n-capture or do you have high enough

T for charged-particle links to be open?). Otherwise it seems plausible.

Do you have any specific hints as to what I should check?

 

Why exactly is mixing a problem, wouldn’t mesa automatically do the mixing for all species in the set? Or is it just a matter of time consumption?

 

The problem is  physics. Mixing length theory is local and steady state, stellar convection is

intermittent, turbulent, and nonlocal (Newton's laws of motion). There should be a turbulent

boundary layer, which MLT never can give. Look at ADS for papers by Casey Meakin and me :-)

 

Of course. I am familiar with the excellent work of you and Casey (also met Casey several times and talked with him about this), and I would be delighted to see your results implemented in mesa. 

I thought you meant there is a specific problem with mixing combined with large networks.

For my interest, do you have a list of species and reactions for the 81 and 177 networks that you can send me?

 

Attached. My code automagically adds all the reaction links from the JINA database for

the nuclei requested.

 

OK, thanks.

 

Roni

From: David Arnett [mailto:wdarnett at gmail.com] 
Sent: יום ב 04 יוני 2012 22:02
To: Bill Paxton; mesa-users at lists.sourceforge.net
Subject: [mesa-users] Fwd: 35Cl and K39

 

 

---------- Forwarded message ----------
From: David Arnett <wdarnett at gmail.com>
Date: Mon, Jun 4, 2012 at 12:00 PM
Subject: Re: 35Cl and K39
To: Roni Waldman <roni181066 at gmail.com>


Hi Roni,

On Mon, Jun 4, 2012 at 11:38 AM, Roni Waldman <roni181066 at gmail.com> wrote:

Hi Dave,

I used the ns_he.net which was provided with MESA.

Thanks. It sounded familiar. 

I just naively supposed it has been tested by someone who devised it.

It certainly has been tested (Bill Paxton is good and careful), but maybe not for your problem. 

I attach a file containing the list of nuclei and reactions in the net.

 

Let me clarify, if you are saying that for Si burning you need ~60 isotopes, then do you mean that the approx21 network which is standard in MESA (see attached file) is also wrong?

Yes, in general. No, if you are doing the problem it was devised for.
 

It was my understanding that it is possible to devise small effective networks which give the energy production correctly, without getting stuck into an unrealistic composition.

This is an art rather than a science. The network you sent, ns_he.txt,
will have trouble if the neutron excess deviates much from .002 because you have no Si30 or S34, which sop up extra neutrons. If they are missing, these neutrons glom onto CL35 and K39, which have an extra neutron (Z=17,N=18, e.g.). Because odd Z elements are rare, this gives a big effect in the ELEMENTAL abundance. 
However, the energy generation is probably well represented because the binding energy of  these nuclei is about the same as those omitted :-)
In TYCHO, my default was 81 nuclei, a compromise which got the elemental abundances about right. I am using 177 for core collapse.
I am working with Bill and the MESA team to implement a large network 
option in MESA. The problem is getting the mixing right, which is not now done by anyone. Bill has a version that puts the network inside the Henyey sweep, but it becomes slow for reasonable networks. I have a big network, but my coupling of "chemistry", that is, burning and evolution is operator split. 
I hope this helps.
Dave

Am I wrong?

 

Thanks for your help.

Roni

 

From: David Arnett [mailto:wdarnett at gmail.com] 
Sent: יום ב 04 יוני 2012 20:21
To: roni181066 at gmail.com
Subject: 35Cl and K39

 

Hi Roni,

It depends on your network (which was removed from the email post, so I can't see it).  How may nuclei do you have? For Si burning you need ~60

as a rule, to get the quasi-equilibrium right. With fewer, mass goes to strange places because of constraints on charge and nucleon number.

Maybe you should send more info offline.

Dave


 

-- 
David Arnett
Regents Professor
Steward Observatory
University of Arizona




-- 
David Arnett
Regents Professor
Steward Observatory
University of Arizona




-- 
David Arnett
Regents Professor
Steward Observatory
University of Arizona





 

-- 
David Arnett
Regents Professor
Steward Observatory
University of Arizona

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mesastar.org/pipermail/mesa-users/attachments/20120604/5ef1249f/attachment.html>