[Stoves] Inverse Diffusion Flames in TLUDs

Tue Mar 8 03:49:29 CST 2016

Right.  What happens is that the soot you detect just above the flame is exposed to temps of the order of 600 deg C and it pyrolyses down to almost pure carbon.  The soot is a polyaromatic hydrocarbon in any event, with lots of linked rings.  As a little bit of residual hydrogen is driven off, the carbon becomes more and more refractory and becomes very small particles indeed – like buckyballs (buckminsterfullerene) which are C60 and ~1.1nm in diameter.  At that scale they are too small to deposit in the lungs – respirable fractions start at around 2.5μm and end at about 0.25μm(250nm). The pyrolysing soot doesn’t burn because the carbon is quite refractory – think of trying to ignite graphite (incidentally, diamond burns well at about 700 deg C).

The blue comes from emission at flame temperature – in a really hot flame (e.g. well-mixed oxyacetylene) the emission extends well into the ultraviolet, leading to “welder’s eyes” and the need for dark goggles (at minimum) and full face protection (for big jobs). See Wikipedia

https://upload.wikimedia.org/wikipedia/en/thumb/2/26/Spectrum_of_blue_flame_-_intensity_corrected.png/800px-Spectrum_of_blue_flame_-_intensity_corrected.png

Spectrum of the blue (premixed, i.e., complete combustion) flame from a butane <https://en.wikipedia.org/wiki/Butane>  torch showing molecular radical <https://en.wikipedia.org/wiki/Radical_%28chemistry%29>  band emission and Swan bands <https://en.wikipedia.org/wiki/Swan_bands> . Note that virtually all the light produced is in the blue to green region of the spectrum below about 565 nanometers, accounting for the bluish color of sootless hydrocarbon flames.

Note the significant flux below 350nm i.e uv with the really huge peak at ~305nm, solidly in the dangerous uv region. 

Hope that clarifies things

Philip 

From: Stoves [mailto:stoves-bounces at lists.bioenergylists.org] On Behalf Of Crispin Pemberton-Pigott
Sent: Tuesday, March 8, 2016 9:34 AM
To: Philip Lloyd
Subject: Re: [Stoves] Inverse Diffusion Flames in TLUDs

Philip, in short they become smaller and less sticky, correct?

My mental picture of a paraffin (wax in this case) is that of a long chain being chopped into smaller and smaller pieces so that, given enough time and heat input, it be reduced to gases that are easily burned. 

I was trying to contrast the action above a candle with the model of a particle being processed by a flame as described by Julien which is also a reasonable image. As the colour, or should I say the red and orange colours, of flames results from glowing particles‎. Then there is a sort of conflict between the idea of non-glowing particles being burned away amid glowing particle‎s if the visible flames are themselves a bunch of particles. 

Is it true that the blue portion of the flames is caused by burning CO? And that the cooler colours are from glowing particles? It seems the particles are reduced under a variety of conditions. How do we generalise, or maybe we can't. 

I was under the impression that Buckyballs were constructed from free carbon, not that they were reduced from something. I don't think there are any Buckyballs in paraffin wax. If correct, it means the 'normal particles' would have to be disassembled and the carbon nanostructures assembled afterwards in an environment hot enough to pull apart the bonds but not hot enough to burn. Alternatively, in an oxygen-poor environment like the central region of the flame tip. 

‎How big are Buckyballs?

Thanks

Crispin 

“The humble candle is a good example of PM formation. Particles exist immediately above the flame, in the clear space, but do not exist 60mm higher up, with no flame in between, meaning it is not necessary to have one to have PM burnout. This can be demonstrated using a saucer and passing it at different heights through the gas stream.” 

I think this is not correct. The candle soot changes its character above the flame, the particles do not disappear. Pyrolysis continues, because temperatures are still high at that point.  The products of the pyrolysis become smaller and smaller carbon particles which don’t condense readily – but they are there.  That is why “buckyballs”, widely prevalent above candle flames, were missed for so long, and why the Sistine Chapel has needed drastic cleaning every century or so (the arrival of electric lighting should spare the paintings in future). 

Prof Philip Lloyd

Energy Institute, CPUT

SARETEC, Sachs Circle

Bellville

Tel 021 959 4323

Cell 083 441 5247

PA Nadia 021 959 4330

From: Stoves [mailto:stoves-bounces at lists.bioenergylists.org] On Behalf Of Crispin Pemberton-Pigott
Sent: Tuesday, March 8, 2016 4:26 AM
To: Julien Winter
Subject: Re: [Stoves] Inverse Diffusion Flames in TLUDs

Dear Julien

This is a most valuable analysis. At the end I was inspired to add something that could be appended. 

The draft of the system can be driven by a chimney. For example the Mongolian TLUD's which are among the cleanest stoves anywhere have to be tested with some length of chimney. The manufacturers make no statements, at least no useful ones, about the chimney height. The height is forgotten by most, save the testers. 

The mixing and the flame length and the flame sheet and the length of flamelets are all affected by the draft. China tests all stoves with a four metre tall chimney under the national protocol. The average installed chimney height is five metres, or six, it has been hard to say with clarity. Current testing of 16 stoves for Hebei Province at CAU (BST Lab) is using a six metre chimney because that is what was observed in Hebei. 

Whatever result is obtained at a given level of draft, it will ‎change if the draft is changed, especially if the pressure changes by 50%.  In Mongolia the tests are done with a 2.8m chimney because the typical application is a yurt. 

The advantage of a fan is that the pressure can be controlled at will, but the added complexity is a problem for many. 

The humble candle is a good example of PM formation. Particles exist immediately above the flame, in the clear space, but do not exist 60mm higher up, with no flame in between, meaning it is not necessary to have one to have PM burnout. This can be demonstrated using a saucer and passing it at different heights through the gas stream. 

Placing a glass tube above the flame that pulls it into the bottom hole can not only pretty much eliminate the PM, it can also nearly eliminate the visible flame. There are YouTube videos of this phenomenon. The additional draft changes almost everything so it forms part of the discussion. Changing the TLUD power level changes the draft which changes the power level which changes the draft...

Regards 

Crispin 

Hi all;

Here is an interesting article about inverse diffusion flames IDF similar to what we see in many TLUDs:

Blevins, L. G., Yang, N. Y., Mulholland, G. W., Davis, R. W., & Steel, E. B. 2002. Early soot from inverse diffusion flames. Prepr. Pap.-Amer. Chem. Soc., DiV. Fuel Chem, 47(2), 740-741. 

https://web.anl.gov/PCS/acsfuel/preprint%20archive/Files/47_2_Boston_10-02_0236.pdf

I have attached a photo of standard diffusion flame and an IDF of ethylene flames near their sooting limits. (Kumfer, BM; Skeen, SA; Axelbaum, RL.  2008.  Soot inception limits in laminar diffusion flames with application to oxy-fuel combustion.  Combustion and Flame 154; 546-556)
Inline image 1

In her article, Linda Blevins argues that soot forms on the outside edge of an IDF, then follows a trajectory away from the flame, and escapes oxidation.  In a standard diffusion flame, soot tends to be oxidized in the tip of the flame.

One common type of syngas burner for TLUDs has secondary air entering the gas burner through small holes.  The flamelets arrising from these holes start out as a type of IDF.  Syngas rises from below, so the underside of the flamelets should be more fuel rich than the top side (which may be lean).  Therefore, soot production in a TLUD burner should be greatest on the underside of the flamelets.  If the flamlets coaless to form a sheet of flame across the width of the gas burner, then (most?) soot particles will have to pass through this flame and oxidized to varying degrees.

However, if we turn down the gasification rate in the TLUD, so there is not a continuous sheet of flame across the width of the burner, is there a change in the nature of soot particles in the exhause gas from the stove?  Will there be a higher proportion of polycyclic aromatic hydrocarbons vs black carbon from a turned down TLUD?  Of course, when turned down the mass of soot emitted may not be very high.  All the same, this is an example of why we should characterize emissions from stoves over a wide range of power levels when assessing health risks.

Cheers,

Julien. 

-- 

Julien Winter
Cobourg, ON, CANADA

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20160308/4b760181/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.png
Type: image/png
Size: 36768 bytes
Desc: not available
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20160308/4b760181/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image005.png
Type: image/png
Size: 101404 bytes
Desc: not available
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20160308/4b760181/attachment-0001.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image009.png
Type: image/png
Size: 25044 bytes
Desc: not available
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20160308/4b760181/attachment-0002.png>