2 replies to this topic

[jimmason]

yesterday Bear found a quote on the energy needed for pyrolysis.  
"The heat of devolatilisation (Hdevol) [pyrolysis?] is generally taken to be equal to -200 kJ/kg at the temperature of devolatilisation."  
- Richard &Thunman, 2002.

remember modelling this is/was the big unknown in the energy balance work of a few months ago.  see here: http://gekgasifier.p...-Energy-Balance.   an accurate answer is important to understand how much process improvement we can realize with rigorous heat recycling architectures.  and relatedly, the relative impact of moisture and pyrolysis loads on the combustion zone in a downdraft gasifier.  the goal, of course, is to get rid of these preparatory loads on combustion that drag down our temps and generally make a typical gasifier so thermally precarious.  that is what i'm trying to do with the "Hot TOTTI" architecture.

so newly inspired, i went looking again at this.  in the process, i found a very interesting series of publications of a 4 year conference cycle called "advances in thermochemical biomass conversion".  the are at least three published tomes of the conference papers.  each volume is about 1500 pages.  they are a treasure of testing and modelling of most of the issues we debate regularly in these parts.  here's an example of one of the volumes.  note the large collection of papers on modelling pyrolysis.  
http://books.google....ndbook"&f=false  unfortuantely, they sell online for between 1 and 2k.  but they are likely findable somewhere for less.  anyone here have a collection?  i'm sure some of you here were at the conferences.  are these still continuing?

in the interim, here is a paper i found by tom reed that answers in reasonable specifics my longstanding question of the energy needed for pyrolysis.  this is formally called the "heat of pyrolysis" or "heat of devolatization".   tom actually fudges it to an engineering number of heat needed from atm conditions through finish of pyrolysis, at both dry and % wet wood.  he calls this the "heat FOR pyrolysis".  he generates his numbers not with modelling, but with a clever test using a drill motor to spin wood chunks over a meeker burner.  this is a method he was telling me about at the biochar conference last week when i was drilling him as to whether densified biomass behaves differently in pyrolysis than natural wood, given the lack of pores.  his answer in short, "go test it jim, here's how".

http://books.google....page&q=&f=false

unfortunately, tom's answer is very different than the answer bear found.  or really, i think the -200kj/kg is just for the chemical transformation in isothermic conditions, and does not include the mass heating before and after the devolatilization.  tom's numbers are more of an insitu "here's the heat needed to make pyrolysis happen from standard atm conditions and heat the various masses to combustion ready temps".  tom prefaces his "answer" by saying there is a giant swing in numbers reported.  he claims this large range relates to the rate at which the pyrolysis proceeds and the difference in char to gas products that come out.  slow pyro that produces lots of char can be net exo once at devolatilization temps.  faster pyro that produces little char is net endothermic.  the specific heat numbers vary with the ratios of output products.  antal has an equation to calc heat of pyro in relation to the ratio of char to tar out.  where we should understand our pyrolysis rate in a downdraft remains unclear.  

the summary:  tom's numbers for the heat needed for pyro are about 2x the ones i had.  my simplification of "let's average the whole problem to the spec heat of wood from 25c to 1000c, and ignore mass and chemical changes", turns out to not be a bad approximation.  problem is i did the math wrong over this approximation (how did none of you catch this?).  that, and the spec heat of wood curves upward with temp quite signiicantly.  there is a nice chart in tom's paper.

in engineering terms, pyrolysis and related solid fuel specific heating needs to combustion temps are about the same range of thermal load as processing fuel moisture to combustion temps.  both are in the 2000-3000kj/kg range with typical wet woods.  with dry wood, the pyrolysis load is actually larger than the drying load.  both are in the 2000-3000kj/kg range with typical wet woods.  this means each is on the order of 15% of the LHV of the wood.  or 30% total, which makes sense given what we know about efficiencies of gasification.  the pyrolysis and fuel heating load is much larger than i thought (assuming these numbers are correct, which is still very much in debate).  

if they are in the realm of correct, this suggests that external heat sources to support pyrolysis in a reactor can have a much larger contribution to the total solution than previously thought.   pyrolysis and drying are on the order of equal thermal problems.  the air preheating is on the order of 1/4 the load of the pyrolysis and drying loads.  it seems the thermal battle is largely won or lost in the drying and pyrolysis stages.

my previous mistakes are here: http://gekgasifier.p...-Energy-Balance  i'll leave it uncorrected for a bit so you can see what i did previously.

thus we will continue with slightly improved understanding, as we continue to peel back the layers of this most interesting problem . . .

jim

[JayMart]

I started reading up on pyrolysis today and found a few interesting tidbits.

First, I am only repeating information presented on the pyrolysis of pure cellulose. Wood is made up of varying amounts of cellulose, hemicellulose, and lignin.

The pyrolysis of cellulose forms the volatile compound levoglucosan. The reaction is strictly exothermic with a high activation energy (238 kJ/mol). Levoglucosan is a water soluble product that can be further pyrolyzed into shorter chains through the removal of H20. An interesting observation of this reaction is if you draw the levoglucosan away from the cellulose, nearly no char will be formed. If you leave the levoglucosan in proximity of the cellulose and/or increase the pressure, significant amounts of char (upto 40%) can be formed and the overall reaction can become slightly exothermic. The presence of water vapor seems to be a catalyst for the creation of char.

All in all there are a lot of complicated reactions going on. It seems that if you drive the reaction hard by pushing a lot of heat into it, an increased rate of pyrolysis is possible but this occurs at the expense of a lower char yield and the consumption of the energy. If the applied heat is reduced or nearly removed, the char yield can be maximized, but the reaction rates decrease and with out the continued addition of some energy will stop, i.e. the exothermic reaction that creates the char is not sufficient to heat the raw material to the temperatures required for the reaction to start.

This summary is derived from "Cellulose Pyrolysis Kinetics: The Current State of Knowledge" by Antal and Varhegyi, 1995, Ind. Eng. Chem. Res. 34, 703-717.

Jay Martin

[solarbobky]

According to WorldCat, UC Davis has Developments in thermochemical biomass conversion and the later Progress in thermochemical biomass conversion in their library.