Defining autocatalytic reaction in Gpyro

[Takashi Odagiri]

Hello all,

 

I am working on a smoldering problem of a hot polymer material.

 

Typical polymer have stabilizer (anti-oxidant) added and therefore there is some induction period before the polymer start oxidation reaction and heat up.

 

Is there any way to model such induction time by defining a autocatalytic reaction or any other method?

 

Any help or comments are greatly appreciated.

 

Best regards,

Tak Odagiri

[Guillermo Rein]

Hello Tak,

I have not heard of such use in Gpyro. Maybe you could add a very slow reaction (high E) in which a product is a reactant for the subsequent oxidation reaction (like the activation reaction in wood pyrolysis).  There are several papers on Gpyro with wood pyrolysis. 

Also, Dr Han Yuan (cc) did his PhD with me on self-heating with Gpyro and those papers include several low temperature reactions.

Cheers,

G.

Prof. Guillermo Rein

Imperial College London, UK

“Hope for the best and prepare for the worst” British proverb and B. Disraeli, 1833.

On 23 Jan 2025, at 00:58, Takashi Odagiri <takashi...@gmail.com> wrote:



--
You received this message because you are subscribed to the Google Groups "Gpyro Working Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gpyro-working-g...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/gpyro-working-group/d9f95031-afbc-4330-9d78-0c7aa4fc6501n%40googlegroups.com.

[Alexander Castagna]

Hi Tak,

A very interesting problem indeed! I have never dealt with anti-oxidants but I can second Guillermo's suggestion. I would add to Guillermo's proposed solution that you may wish to set the heat of reaction to zero. Gpyro offers the possibility to model a catalytic reaction (kinetic model i_{kinetic model} = 9), in which case you must specify the catalytic species i_cat to correspond to the ISPEC number in your SPROPS sheet and choose a suitable coefficient K_{cat}. I have never used this kinetic model in my work but I hope this suggestion can help you progress.

If you manage to make your case work, would you consider sharing  it with the wider gpyro community so all can profit?

Happy to help out further if needed,

Best,

Alexander

-------------------------------------------------------------------------

Alexander Castagna MEng  ACGI FFSC  AMIMechE MInstP

Research Postgraduate (PhD candidate) in Fire Science

HAZELAB research group

Department of Mechanical Engineering

Room 609 – City and Guilds Building

Imperial College London

[Davood Zeinali]

Hi Tak,

If you couple the Gpyro simulation with FDS, you can also define an igniter, which is basically a volumetric heat source. You can control the igniter via the &MISC Namelist group in the FDS file using parameters TIGNSTART, TIGNSTOP, IGNPOW, and IGNLOC. The igniter has a power output of IGNPOW (units of W/m3), but only between time TIGNSTART and time TIGNSTOP. The location of the igniter is given by the sextuplet IGNLOC. A related sample from the Google repository is attached.

I see that your current email is not part of the Gpyro group. In case you do not have access to the group emails, a response from Alexander is also included below.

Best regards,

Dave

──────────────────────────    ──    ──    ──    ──

Davood Zeinali

Research Scientist

PhD, MSc, BSc in Fire Safety Eng.

RISE Fire Research AS | Subsidiary of RISE Research Institutes of Sweden

M: +47 922 535 15

davood....@risefr.no

risefr.com | ri.se

Visiting adress: Tillerbruvegen 202, 7092 Tiller, Norway

Postal adress: Box 4767 Torgarden, 7465 Trondheim, Norway

Read and subscribe to the Brandposten magazine here

To view this discussion visit https://groups.google.com/d/msgid/gpyro-working-group/a862cc75-2fc3-425b-922a-95fd62036e01n%40googlegroups.com.

[Davood Zeinali]

PS: my assumption is that the volumetric heat source can be defined within the material, but this may not be allowed by FDS/GPyro. Please do share what you find out in any case.

Best regards,

Dave

──────────────────────────    ──    ──    ──    ──

Davood Zeinali

Research Scientist

PhD, MSc, BSc in Fire Safety Eng.

RISE Fire Research AS | Subsidiary of RISE Research Institutes of Sweden

M: +47 922 535 15

davood....@risefr.no

risefr.com | ri.se

Visiting adress: Tillerbruvegen 202, 7092 Tiller, Norway

Postal adress: Box 4767 Torgarden, 7465 Trondheim, Norway

Read and subscribe to the Brandposten magazine here

[Takashi Odagiri]

Dear all,

Thank you all very much for such quick responce and many ideas on possible methods of over comming this problem.

I will look into all the iseas ASAP.

And if it works out, I have no problem sharing the infomation with the groupe.

Best regards,

Tak Odagiri

2025年1月24日金曜日 4:32:15 UTC+9 zeinali.davood:

[Takashi Odagiri]

Dear all,

 

I am now looking into use of kinetic model 9 (Catalytic) to express an induction time in my polymer smoldering problem.

 

However, I find that there may be a mistake in GPYRO documentation.

 

The document states that the model is as follows.

 

Eq1.       (1-α)ⁿ(1+K_catα_icat)

 

On the other hand, other documentations I find in other literature states that the model should be as follows.

 

Eq2.       (1-α)ⁿ(1+K_catα)

 

The difference is that the second conversion term (α) is of the reactant and not of the catalyst specie.  In ordinary catalytic reaction, catalyst is not consumed nor produced in reaction, therefore the amount of catalyst is constant and thereby the conversion of catalyst must be zero (α_icat=0).

 

In GPYRO, if I specify SPECIES_icat to be the reactant material then the model will run just as I expect.

 

Can anyone confirm that Eq2 is the correct form ?

I will attach a document I found (page 9, equation 13)

 

Best regards,

Tak Odagiri

 An introduction to n-th order and autocatalysis reactions

2025年1月24日金曜日 9:53:19 UTC+9 Takashi Odagiri: