[Stoves] Blue flames in char combustion

Fri Sep 6 02:30:55 CDT 2013

Dear Paul and all interested in blue flames and effect of H2O on burning rate of CO. Thanks also for Kobus for interesting aspects concerning blue flames.

Dear Jaakko,

Thank you for your excellent message.   It is also helpful about understanding charcoal-burning stoves, which tend to have a rather shallow thickness of char.

Can you please elaborate more about the
two different stable (quasi)steady state solutions for the temperature and species distributions in the char bed, a high temperature solution and a low temperature solution.

Paul

Paul S. Anderson, PhD  aka "Dr TLUD"

Email:  psanders at ilstu.edu<mailto:psanders at ilstu.edu>   Skype: paultlud  Phone: +1-309-452-7072

Website:  www.drtlud.com<http://www.drtlud.com>

***
I try to explain the term "quasi-steady". Grate combustion is similar to TLUD, since a TLUD is like a small fuel batch moving on a grate. I have made models including both pyrolysis and char combustion stages [1] and also only for char combustion stage in a fireplace [2]. One can calculate profiles inside the char bed and the possibility for two solutions [2]. The model includes conservation equations for energy, mass and species (O2, CO2, CO, H2O). Temperature and concentrations can be calculated as function of time (model [1]) and distance from the air inlet. In addition, equations for local heat and mass transfer (for example O2 from gas to particle surface) between gas and particles inside the bed, kinetic reaction Arrhenius-type equations for the reactions of char with O2, CO2 and H2O and the kinetic reaction rate for burning of CO are required. It is possible to calculate the combustion from start to end with a time dependent model by solving the equations numerically. Nowadays, very sophisticated models have been published on grate combustion.

A time dependent model includes the combustion history of the fuel bed. This "memory of the past" is shown by the present particle and temperature distributions (species distributions, char particle size distribution) as function of distance from the air inlet. A continuous time dependent model is presented by partial differential equations for temperature and species. These equations contain "storage terms", which are proportional to the rate of change. Usually their contribution is small, but however, they contain the knowledge of the "past". Simpler model equations are obtained, if one ignores the "storage terms". By putting zero value, one gets a quasi (or pseudo)-steady model. This model can give good predictions, but because the memory is not included (it is forgotten by neglecting the storage terms), it can give different stable results. So by a quasi-steady model we look "a slice of time" and seek the solution for an assumed problem without knowing how we get there. First we make a guess and find by iterative process (some kind of a trial and error method) the solution agreeing with all the model equations. Reactor calculations are sometimes based on quasi-steady analysis and sometimes different outcomes are found. Maybe the climate scientists are using similar method to examine possible states of the globe.

The difference between a time-dependent model and the one predicted by quasi-steady model can be very small. For example when one calculates the temperature T of a single burning char particle in air, one solves a quasi-steady equation (I do not explain the terms in details here)
G=L
G = heat generated by combustion in the particle is equal to L= heat lost from the particle to ambient air by convection and radiation. This depends on particle (T) and ambient temperatures and T can be solved.
However, the above heat balance equation is not accurate. Time dependent more accurate equation is
G=L+S
S is the storage term, which is proportional to heat flow consumed to heat up the particle (~dT/dt = rate of change of temperature, t is time). S can be neglected because the burning particle is already hot and S<<G in this "slice of time". However, it still has an (very small) effect because the particle temperature does not remain exactly constant. It is also possible to calculate the temperature of the particle as function of time to the end of burning with the simpler quasi-steady model agreeing well with the more exact time-dependent model, because S<<G. However, it is not possible to predict the temperature during the fast and short heating up period with the quasi-steady model.  We might also in some special conditions have two solutions. CO (produced in char combustion) might be burning close to the particle giving heat to keep the particle and its vicinity hot enough so that CO can be burning. In the other case the particle is not hot enough to keep CO burning. The particle might also not be burning at all and this depends on its history (was it heated to be ignited?). So there is still another stable solution for the particle temperature (because someone forgot to ignite the grill).

Effect of H2O on promoting blue flame by burning of CO is interesting. The combustion of CO takes place by the reaction CO+½O2=CO2.  There is no H2O in this simple overall reaction. However, radicals formed from H2O are important in the reaction path affecting the reaction rate. CO hardly burns without any H2O. The combustion rate of CO is proportional to concentrations [O2]^n and [H2O]^k . In some models the exponents are n=1/4 and k=1/2, which shows the great effect of H2O (because k> n !). So in a dry desert CO is not burned so well as in wet climate. However, the fuel itself may contain some hydrogen producing H2O, which promotes burning of CO even if there is not much H2O in the combustion air. This is the reason for including H2O (from air) in the char bed model even char is assumed to consist only of carbon and ash.

[1] NOx formation in grate combustion of wood. Clean Air: International Journal on Energy for a Clean Environment, Vol. 4 (3), 30 p., 2003.
[2] Emission formation during wood log combustion in fireplaces - Part II: Char combustion stage. Progress in Computational Fluid Dynamics 6 (4/5), 209-216, 2006.

Regards

Jaakko

From: Stoves [mailto:stoves-bounces at lists.bioenergylists.org] On Behalf Of Saastamoinen Jaakko
Sent: 3. syyskuuta 2013 16:57
To: Discussion of biomass cooking stoves
Subject: Re: [Stoves] Blue Flame -- Natural Draft -- Rice Husk

Dear all,

I have seen blue flames in my heat storing stove under natural draft using wood logs in char combustion stage, when the logs have been broken into pieces. It is CO burning as has been discussed here by others. Later the blue flame disappears due to lower bed temperature so that CO cannot burn but goes to chimney.

I have also calculated this phenomena. One can see blue flames  (CO is burning) if the thickness of bed of char particles and air rate are suitable.  If the bed is too thick, blue flames are inside the bed, increasing gas temperature and gasification takes place above producing CO from the bed. (This CO could be burned if somehow oxygen could be mixed and temperature of the gas is high enough). If the bed is too thin, excess air (due to lower flow resistance) cools the gas so that CO is not burning or it burning rate is very low.  So it needs suitable bed thickness and air rate which are difficult to maintain with natural draft.

As Richard Stanley has experienced, blue flame is very sensitive to air rate.  I noticed this when calculating the burning with a model. The calculation was based on iteration and on an initial guess. I was astonished that depending on this initial guess I got, not a chaotic solution, but two different stable (quasi)steady state solutions for the temperature and species distributions in the char bed, a high temperature solution and a low temperature solution.  Conclusion is that both solutions could be right (CO is either burning or it is not). It depends on the burning history of the stove which solution is the right one. It the fire is disturbed and cooled down a little, CO does not burn but if it remains hot, CO can burn.

I throw small twigs in the end stage to get some CO escaping to chimney to burn in the volatiles flames, increase the draft and speed up the burning of residual char at the end of the heating.

Jaakko

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20130906/be9e317e/attachment.html>