Simple output of half-reactions and their rate data?

[Karen Pease]

Extreme newbie question here: can Cantera be used to output the half reactions along with their Arrhenius coefficients (A/pre-exponential factor,B/temperature exponent, and Ea/activation energy), ideally (but not a requirement) in chemkin format?  E.g. something like:

ELEMENTS
 H   O    C   N   AR
END
SPECIE
C7H16   O2              N2              CO      H2O
O               CO2     OH              H       H2
HO2             H2O2    N               NO      C
END
REACTIONS
 C7H16 + 11O2            => 7CO2 + 8H2O        1.00E+8  0.0   15780.0! 1
        FORD    / C7H16 0.25 /
        FORD    / O2 1.5 /
 CO    + O     + M     = CO2   + M            6.170E+14  0.00   3000. ! 96
 CO    + OH            = CO2   + H            3.510E+07  1.30   -758. ! 97
 CO    + O2            = CO2   + O            1.600E+13  0.00  41000. ! 98
 HO2   + CO            = CO2   + OH           5.800E+13  0.00  22930. ! 99
!
 H2    + O2            = OH    + OH           1.700E+13  0.00  47780. !100
 H2    + OH            = H2O   + H            1.170E+09  1.30   3626. !101
 O     + OH            = O2    + H            4.000E+14 -0.50      0. !102
 O     + H2            = OH    + H            5.060E+04  2.67   6290. !103
 H     + HO2           = O     + H2O          3.100E+10  0.00   3590. !104
 O     + OH    + M     = HO2   + M            1.000E+16  0.00      0. !105
       H2O/6.0/  CO2/5.0/  H2/3.3/  CO/2.0/  N2/0.70/
 H     + O2    + M     = HO2   + M            2.800E+18  -.86      0.0!106
 O2/0.00/ H2O/ .00/ CO/0.75/ CO2/1.50/ N2/0.0/
 H     + O2    + O2    = HO2   + O2           2.080E+19 -1.24      0.0!107
 H     + O2    + H2O   = HO2   + H2O          11.26E+18  -.76      0.0!108
 H     + O2    + N2    = HO2   + N2           2.600E+19 -1.24      0.0!109
 OH    + HO2           = H2O   + O2           7.500E+12  0.00      0. !110
 H     + HO2           = OH    + OH           1.700E+14  0.0     875. !111
 O     + HO2           = O2    + OH           1.400E+13  0.00   1073. !112
 OH    + OH            = O     + H2O          6.000E+08  1.30      0. !113
 H     + H     + M     = H2    + M            1.000E+18 -1.00      0. !114
       H2/0./ H2O/0./ CO2/0./
 H     + H     + H2    = H2    + H2           9.200E+16 -0.60      0. !115
 H     + H     + H2O   = H2    + H2O          6.000E+19 -1.25      0. !116
 H     + H     + CO2   = H2    + CO2          5.490E+20 -2.00      0. !117
 H     + OH    + M     = H2O   + M            1.600E+22 -2.00      0. !118
 H     + O     + M     = OH    + M            6.200E+16 -0.60      0. !119
 O     + O     + M     = O2    + M            1.890E+13  0.00  -1788. !120
 H     + HO2           = H2    + O2           1.250E+13  0.00      0. !121
 HO2   + HO2           = H2O2  + O2           2.000E+12  0.00      0. !122
 OH    + OH   (+M)     = H2O2 (+M)            7.600E+13  -.37      0. !123
     LOW  /  4.300E+18    -.900  -1700.00/
     TROE/   .7346   94.00  1756.00  5182.00 /
       H2/2.00/ H2O/6.00/ CO/1.50/ CO2/2.00/ N2/0.70/
 H2O2  + H             = HO2   + H2           1.600E+12  0.00   3800. !124
 H2O2  + OH            = H2O   + HO2          1.000E+13  0.00   1800. !125
 H2O2  + H             = H2O   + OH           1.000E+13  0.00   3590. !126
 H2O2  + O             = H2O   + O2           8.400E+11  0.00   4260. !127
 H2O2  + O             = OH    + HO2          2.000E+13  0.00   5900. !128
 H2    + HO2           = H2O   + OH           6.500E+11  0.00  18800. !129
!
 CO2   + N             = NO    + CO           1.900E+11  0.00   3400. !130
 N     + NO            = N2    + O            3.270E+12  0.30      0. !136
 N     + O2            = NO    + O            6.400E+09  1.00   6280. !137
 N     + OH            = NO    + H            7.333E+13  0.00   1120. !138
!
END


I need to be able to generate such files to use with OpenFOAM. While I've found a great amount of such data for C, H, O and N, I also need such data for Al and preferably also Li.  Really, I can take the data in any format and change it to the above format if need -- I just need to get it somehow.

If Cantera can't do this, do you know what can?

Also: minor bug in your docs:

http://www.cantera.org/docs/sphinx/html/cti/example-combustion.html

That page is blank.

 - kv, Karen

[Bryan W. Weber]

Hi Karen,

Welcome to Cantera! Can you be a bit more specific about what you need? Where are you getting the reactions from, and which half should be output (that is, how to pick which reactions to output)? As far as I know, Cantera does not have any built-in aluminum or lithium chemistry, so you'll have to find the reaction parameters from somewhere else.

Thanks for the bug report, we'll make sure it gets fixed.

Best regards,

Bryan

[Karen Pease]

Thanks for the quick response, Bryan!

I'm looking to use OpenFOAM to simulate a virtual expansion nozzle for a rocket, using the CEA2 calculations for the conditions at the throat (I'm not looking to model the combustion chamber itself).  The exhaust in the simulation is from the combustion of high molecular weight paraffin wax (using C30H62 as the baseline) and aluminum with liquid oxygen. CEA2 suggests that at the throat (using a bunch of rough assumptions on my part) the aluminum-containing fraction of the exhaust would be about 43% Al2O3, 25% AlOH, 13% Al2O, 11% Al (partially ionized) and the rest miscellaneous (AlO, Al2O2, AlH, Al(OH)2, etc).  Since that's obviously only partially burned, the combustion will continue as it expands (ultimately nearly all ending up condensing out as Al2O3, further imparting the heat of condensation to the exhaust stream).  So I can't just model it as a frozen combustion process as it exits the throat.  OpenFOAM has a number of different combustion solvers but they require a chemkin format file (as presented above) with the arrhenius parameters for all of the half reactions under consideration in order to operate.

It's a shame to hear that you don't have any data for aluminum.  It's very important to high performance solid and hybrid propellants, giving a significant boost to specific impulse and damping out combustion instabilities..  The "even higher performance" alternative to aluminum is lithium, although it's rarely used due to handling challenges.  You wouldn't perchance have any ideas where I could look to find reaction rate data on them, would you?  I found a paper recently that did tests on aluminum powder oxidation in the 425-525° range and came up with arrhenius parameters A=7,41e10 s^-1, B=0, and Ea=222kg/mol... but they then cited values for comparison from other papers which ranged from A=6,31e3 to 1,84e12 and Ea=119-230, so nine orders of magnitude range... ;)  And even if I just presume that it's due to different powder formulations** and that anything in that range could be valid for a properly-produced powder, and if I further assume that those parameters hold at higher temperatures, the bigger problem is that those are the parameters for the reaction 4 Al + 3 O2 <=> 2 Al2O3; that doesn't help me with the half reactions in order to be able to deal with the wide range of aluminum-containing species that exist at the throat.  I tried searching for, for example, for anything on the combustion of AlOH (2+) but drew a blank.

Come to think of it, I have to wonder how CEA2 would be working out reaction rates. I searched through the source tree and didn't see any arrhenius parameters in it.

 - kv, Karen

** Aluminum combustion is kind of tricky as the oxide has a very high melting point and is very strong, so it tends to encapsulate the burning droplets.

[Karen Pease]

Oops, that should read "kJ/mol", not "kg/mol"  :)

 - kv, Karen

[Nick Curtis]

Karen,
A couple of things.  First CEA is an equilibrium solver, i.e. it doesn't require reaction rates but instead solves for chemical equilibrium based on the properties of the species (typically the gibbs free energy parameterization).
Second, if you had a Cantera format mechanism it wouldn't be impossible to output the chemkin format mechanism from the cantera mechanism; that tool does not exist currently, and you would have to do some work to get the proper formatting etc. but there's nothing fundamentally wrong with the idea.
Finally Aluminum combustion mechanisms in literature look to be fairly sparse, but I did (seemingly) find a few.  Furthermore the modeling is going to be complicated in that you have to deal with multiple phases, but here's what I found:

http://enu.kz/repository/2010/AIAA-2010-6677.pdf
http://www.wiley.com/WileyCDA/WileyTitle/productCd-1118127560,subjectCd-ME70.html
https://books.google.com/books?id=OmzSN4GADBUC&pg=PA207&lpg=PA207&dq=%22aluminum%22+chemical+kinetic+mechanisms+for+combustion+applications&source=bl&ots=pRRfbp9DKK&sig=KncfkJOmmc60hlJqg78_GjsMAWY&hl=en&sa=X&ved=0ahUKEwi57rLy5cfJAhWRsoMKHeI6AZ0Q6AEIOjAD#v=onepage&q=%22aluminum%22&f=false
http://www.tfd.chalmers.se/~valeri/Ajax/CERC_2011_newAL_H2O.pdf

Nick

[Ray Speth]

Karen,

I think the key thing to understand is that Cantera, like OpenFOAM and Chemkin, is a tool that takes chemical kinetic mechanisms as input. It includes a few simple mechanisms, e.g. the GRI 3.0 mechanism for methane combustion, but generally the idea is that you need to provide the data about reactions for whatever system you're studying.

Regards,

Ray

[Karen Pease]

Æji... of course (re CEA); I don't know why that didn't occur to me.  :Þ 

Re: the literature on aluminum combustion: indeed, "sparse" was my take as well, and I'd already read most of what was linked there. However, while I'd read about ALICE before (neat concept IMHO), I hadn't read that particular paper, and it apparently has the reaction rate parameters for a number of the common aluminum species - that's an excellent find, thank you!  I should probably be able to piece together a pretty reasonable input file from that.

 - kv, Karen