Chemical reaction delay

MFiX
#1

Dear all,

I’m running a DEM file on version 21.1.4 and I’m facing some issue that I’d like to discuss with you.

I’m setting several reactions, among which, is pyrolysis reaction that converts solid biomass (species defined by me) into carbon + gases + ash (defined by me too). The thing is, the ash species mass fraction appear normally at the beginning of the simulation, but the carbon and probably the gases do not appear unless the particle temperature reaches 665K knowing that the biomass enters at 363K so it takes about 5 sec simulation time to reach the mentioned temperature which is too long to have steady state values.

I wonder if the code has some condition to start this reaction, however I’m not setting any condition in my udf. If so, how can I fasten the reaction or the heat transfer?

I’ve attached both the project file and the result file that shows the temperature and particle mass fractions where x_p(1) refers to char, x_p(2) is ash, x_p(3) is moisture and x_p(4) is wood (biomass).

wang_new_2021-08-10T114526.807724.zip (26.7 MB)
BIOMASS.zip (15.1 KB)

Thanks,
Mariam

#2

You are using an arrhenius reaction rate which has the form:

5.0d6*exp(-14433.0d0/Tp)

Notice how the particle temperature is in the exponential, which gives you a curve like this:

image

So if you want the pyrolysis to be faster, play with the rate expression. Although I suspect this expressions is from an experiment?

1 Like
#3

Thanks a lot, Justin. That explains the issue. Now all I need is to speed up the heat transfer to the biomass particles as they enter at 298K. So, if you could help me find out what parameters that matter the most in transferring heat to them.

And yes, these Arrhenius constants are in accordance with the journal paper: P. N. and T. Karunanithi, “Kinetic Modeling in Biomass Pyrolysis – A Review,” Appl. Sci. Res., vol. 4, no. 12, pp. 1627–1636, 2008.

Thanks again,
Mariam

#4

Are you saying you have experimental data that shows steady state is reached in less than 5 seconds?

#5

Hello Jeff,

Not sure about the experimental steady state, however there is a similar simulation (CFD-DEM on MFIX) that reached final steady state after 5 seconds. The problem is, in my case, when the pyrolysis reaction starts after 5 sec, it takes almost 3 sec more to reach the steady state, that’s 8 sec in total. Both simulations apply the same chemical rates parameters.

Here is a sample of my result, and Wang’s result that I just mentioned above:

1- my result:

Picture3
Picture31653×993 82.3 KB

2- Wang’s result:
image

ref.: S. Wang, K. Luo, and J. Fan, “CFD-DEM coupled with thermochemical sub-models for biomass gasification: Validation and sensitivity analysis,” Chem. Eng. Sci., vol. 217, May 2020, doi: 10.1016/j.ces.2020.115550.

Thanks,
Mariam

#6

If everything’s truly the same, you should get the same result (ICs, BCs, Kinetics, model selection, grid, etc.).

That being said, who knows what Wang actually did to get that result. The only way to truly know is to get and read the source code.

#7

Maybe he didn’t want to show the unsteady process by cutting the results from 0 to 5 on the graph.