Pyrolysis using TGA

256 views
Skip to first unread message

ChrisDagenais

unread,
May 24, 2022, 12:06:50 PM5/24/22
to FDS and Smokeview Discussions
Good morning,
I am trying to model pyrolysis using TGA data. From many of the posts in this forum, I found this very useful thesis (http://vuir.vu.edu.au/31007/) which also has code examples for both the TGA and cone calorimeter.

When using the TGA code and my own TGA data, I can verify my input pyrolysis parameters and results. However when using the code of the cone calorimeter and the Pine data from the thesis example, I don't get any HRR. Ignition does not seem to be happening. I therefore cannot verify my own sets as the exampe does not seem to work.

Has anyone been able to run the cone model and obtain HRR?
Am I missing a parameter in the command lines?

I'm using FDS 6.7.6.

Thanks,
Christian

P.S. For some reason, I can't upload any files in my post...I'll try later to upload the .fds file from the thesis.

dr_jfloyd

unread,
May 24, 2022, 12:21:53 PM5/24/22
to FDS and Smokeview Discussions
I recommend you first follow the process discussed in the FDS User's Guide for doing virtual TGA and CONE tests (11.6 Testing Your Pyrolysis Model).  First do the TGA test and make sure your kinetic properties are OK. FDS has a special mode of operation now for doing TGA testing. Then do the virtual cone test. Only once you are satisfied that your full set of kinetic and bulk properties are reasonable should you move to the next step of doing a simulation with gas phase combustion.

ChrisDagenais

unread,
Jul 12, 2022, 2:53:00 PM7/12/22
to FDS and Smokeview Discussions
I figured it out now, everything works well.
It's actually quite easy!
Thanks!

Christian

Khalid Moinuddin

unread,
Jul 12, 2022, 7:03:11 PM7/12/22
to fds...@googlegroups.com

Hi Christian,

 

We used to model TGA and cone calorimeter as shown in that thesis.

 

We also published some results in a journal paper. See this link:

 

https://www.mdpi.com/2073-4360/12/9/2075

 

This is an open-access journal. You should be able to access it easily.

 

Kind regards

 

Khalid

 

 

 

From: 'ChrisDagenais' via FDS and Smokeview Discussions <fds...@googlegroups.com>
Sent: Wednesday, 13 July 2022 4:53 AM
To: FDS and Smokeview Discussions <fds...@googlegroups.com>
Subject: [fds-smv] Re: Pyrolysis using TGA

 

EXTERNAL EMAIL: Please be cautious before clicking on any links or downloading attachments.

 

--
You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/134316aa-c76e-4447-96d6-7a265b738c37n%40googlegroups.com.

ChrisDagenais

unread,
Jul 13, 2022, 9:58:56 AM7/13/22
to FDS and Smokeview Discussions
Thank you for the follow-up.
As I mentioned, I did managed to have both TGA and cone validated. The cone was done using a VENT with specific thickness and pyrolysis parameters. Both worked well and quite close to actual test data.
I am now scaling-up and trying to model an actual OBST that is 25 mm thick (5 cells of 5 mm each), in which I am assigning the SURF to the MATL (wood in this case).
I also used the BULK DENSITY and BURN_AWAY but the predictions are now far off. Cells are not even being "burned away"...
I am attaching the code lines for the MATL and a graph showing both methods (1-being using the "real" model and 2-being using a VENT of 25mm thick).

What am I missing?
Thanks.

Test1 - ForForum.fds
HRR_Graph.jpg

Kevin

unread,
Jul 13, 2022, 10:09:53 AM7/13/22
to FDS and Smokeview Discussions
Don't use BURN_AWAY. It was not designed for this kind of application. 

ChrisDagenais

unread,
Jul 13, 2022, 11:06:01 AM7/13/22
to FDS and Smokeview Discussions
Thank you. I will launch the analysis without this feature.
I'll repost the outcomes in a few hours...Stay tuned!

Franco Marra

unread,
Jul 18, 2022, 1:44:09 PM7/18/22
to FDS and Smokeview Discussions
Dear All,

I am facing a very similar problem.

Therefore I would ask for the help of the experts to understand the correct definition of the reactions for a material, sucrose, supposed to produce "CARBON MONOXIDE" by pyrolysis, which should then burn in the gas phase. I must be doing something wrong, but I cannot figure it out after so many attempts and careful inspection of all the manuals, so maybe the solution to this problem could be of help also for others.
Shortly saying I do not get any heat production (apparently both in the solid and gas phase), despite the material I am defining getting consumed by the reaction and having specified a "HEAT OF COMBUSTION".
Among the several trials I have made, I have also made a simulation with extinction switched off, because I have read in several other posts on this mailing list that this could be a common cause for the heat not being produced (however in this case the material is consumed and disappear from the domain with "BURN_AWAY=.TRUE.") but the result does not change.

I am attaching the fds input file (TestBurnMaterial.fds) of my simple test case. I am using the 6.7.9 version of FDS. It is a very simple test, a box of the defined material is heated by a side object at a very high temperature.

As for comparison, I am attaching a second case (TestBurnSugarOK.fds) in which the same material burns correctly, but in this case, I have followed a different approach: all the heat is released by a gas species whose composition corresponds to the composition of the solid material and the "HEAT OF COMBUSTION" is the one expected to be fully released by the burning of the solid material.

I am now confused about the meaning of the "HEAT OF COMBUSTION" for the solid phase and why the first approach does not work.

Any hint on what is wrong and why I cannot adopt the first approach will be very much appreciated.

Thank you in advance to anyone who will have a look at this post.

Best regards,

Franco
TestBurnSugarOK.fds
TestBurnMaterial.fds

Kevin McGrattan

unread,
Jul 18, 2022, 2:02:21 PM7/18/22
to fds...@googlegroups.com
&REAC FUEL='SugarFuel'
      C = 12
      H = 22
      O = 11
      IDEAL=.FALSE.
      HEAT_OF_COMBUSTION=16459
      SOOT_YIELD = 0.0001 /

You have not told FDS how much CO to generate per unit mass of "SugarFuel" consumed. 


dr_jfloyd

unread,
Jul 18, 2022, 3:52:57 PM7/18/22
to FDS and Smokeview Discussions
What is the sugar decomposition reaction you are trying to achieve? Right now you have sugar decomposing into water vapor plus the original sugar. Are you actually trying to have the sugar decompose into water vapor, CO, and then whatever stuff is left?  If this is the case you would need two REAC lines. One REAC line for CO, and a second REAC line for whatever is left in the sugar after making the water vapor and CO. Then you would make three gas products on the MATL input. With two reactions you will no longer be able to use simple chemistry on REAC (C,H,O,N) and will need to follow the User's Guide example for multiple fuels.

As an example if you said 30 % of the sugar mass went to water vapor and 30 % to CO, then 40 % of the mass would go to whatever is left. If you subtract out the H,C, and O associated with the water vapor and CO the "fuel" that will be left would be ~C8.33H10.6O1.64 which is a high molecular weight so would be better off expressed as C1H1.27O0.20.

Franco Marra

unread,
Jul 20, 2022, 12:55:21 PM7/20/22
to FDS and Smokeview Discussions
Dear Kevin and JFloyd,

first, thank you so much for your kind answers.

Kevin, I do not understand why I should add the produced CO amount in the lines of file  TestBurnSugarOK.fds  you pointed out when adopting the simple chemistry approach in the REAC line. If I assume the complete conversion of sucrose, products are only H2O and CO2, following the reaction:
C12H22O11 + 12 O2 -> 12 CO2 + 11 H2O .
Probably it is not fully true that the only products are CO2 and H2O, but for the moment this is my assumption. So the schematization of the process is that solid sugar transforms (pyrolysis) into "gas sugar" (SugarFUEL in my example file TestBurnSugarOK.fds) at a rate defined by the properties of the two materials Sugar1 and Sugar2 in MATL lines (whose parameters were valued from a TGA analysis found in the literature and following the procedure explained in the FDS User Guide). Sorry for not having well explained my assumptions in my original post.


JFloyd, you are absolutely right, the process of sugar burning is much more complex than the one I am trying to implement: I neglected the release of water vapour before the flame ignition. The problem is that I do not know, and I did not find, enough data about the amount of water that would be released at a given heating rate nor about the energy required (I am continuing to search the literature). The steps of sucrose transformation before burning lead to the caramelization with the release of H2O as you say. Unfortunately, the products left in the caramel are at least three as explained at this link:
https://www.scienceofcooking.com/caramelization.htm.
Which composition exactly I would obtain before the flame, as a function of the heating rate, is not clear to me at the moment.
Nevertheless, I am interested in a fire scenario, so in cases where the heating rate is high enough to assume that even caramel, whatever its composition, will immediately transform into the final products, CO2 and H2O, that is what I am assuming in the file TestBurnSugarOK.fds. I thank you so much for suggesting further investigation of sucrose's physical and chemical reactions. I will continue to investigate this complex process.


However, let me go back to my original questions. Whatever intermediate products would be created by the heating and pyrolysis of sucrose, why, if I assume that CO is the gaseous fuel produced in the pyrolysis steps (described in the MATL lines of file TestBurnMaterial.fds), this setup does not ignite this gaseous fuel? The question is relevant to me because even if I change the composition of the gaseous fuel, I expect that, being my setup wrong in some way, I would obtain the same behaviour. What is it the relation between the HEAT_OF_COMBUSTION specified in the MATL lines and the HEAT_OF_COMBUSTION of the gaseous phase? Should not they be independent of each other?

I am sorry for the numerous questions I am contemporary asking you!

Warmest regards,

Franco

Kevin

unread,
Jul 20, 2022, 1:18:22 PM7/20/22
to FDS and Smokeview Discussions
You have specified only one gas phase reaction for SugarFuel. CO is not involved in this reaction, either as a fuel or product. If you assume that CO is produced by the pyrolysis, you can write a REAC line for CO instead of SugarFuel. If you want to react in the gas phase both SugarFuel and CO, you cannot use the "simple chemistry" approach and you must fully specify all reactants and products of the two reactions.

Franco Marra

unread,
Jul 22, 2022, 11:12:15 AM7/22/22
to FDS and Smokeview Discussions
Dear Kevin,

thank you for your attention to this post. I really appreciate your effort to help me, but, I am really sorry, I am not understanding your hints.

I think that your suggestion is exactly what I tried in the second file I attached to my original post, already trying to follow the hint proposed by JFloyd. I am now attaching a slightly modified version of that file where I just renamed the materials to make it clearer that in this second case, I was trying to implement complex chemistry for sugar. In this setup, CO is the only gaseous product of both reactions of sugar pyrolysis (Sugar1 e Sugar2 to reproduce the TGA results). This of course could be not a correct representation of this fire, but before trying to make the reaction even more complex with multiple REAC lines, I have to understand what I am doing wrong. Indeed, the problem is that this setup does not work in several ways:
- no heat is produced by the solid and gas phase reactions;
- I had to define a background Air species (line 22) to get Oxygen and Nitrogen entering the domain from the open vent boundaries. Otherwise, the initial air is consumed and not replenished from the boundaries.

I hope I am not abusing your patience.

Thank you again and best regards,

Franco


TestBurnSugarCChem.fds

Kevin McGrattan

unread,
Jul 22, 2022, 11:47:42 AM7/22/22
to fds...@googlegroups.com
I removed the AIR and just left these lines

&SPEC ID='NITROGEN', BACKGROUND=T /
&SPEC ID='WATER VAPOR' /
&SPEC ID='CARBON MONOXIDE' /
&SPEC ID='CARBON DIOXIDE' /
&SPEC ID='OXYGEN', MASS_FRACTION_0=0.23 /


Your REACtion involves OXYGEN, not AIR. 


Reply all
Reply to author
Forward
0 new messages