Fireballs
Vital Links
Agglomerations of Biomass
In the early 80's I converted a truck to run on woodgas. My main problem was the fuel used hence my motivation for researching other forms of gasifier fuels. The goal is to find a high quality renewable biomass fuel that is suitable for a gasifier that will power internal combustion engines and heat devices. This method should unify many odd fuels such as switchgrass, algae, sawdust, waste paper, crop waste, etc.
I learned of charcoal dust agglomerations, affectionately referred to as fireballs, from the REPP list called stoves [1]. According to Tom Reed, John Tatom (and S.C. Bhattacharya) pioneered the fireball concept. Furthermore my understanding is that the coal fireball was produced before this. For example Helifuel (McDowell & Wellman) converted high sulfur coal into clean fuel. It was a mixture of ground coal and limestone, pelletized in a disc agglomerator, followed by carbonizing in a grate type sinter machine to produce hard, smokeless agglomerations of low ash and sulfur content for combustion or gasification [2].
This has been a stimulating learning experience that I hope to share with you but please understand that this is just the beginning of my personal research. The only information that I had to start with was that John Tatom used a rotating drum with charcoal dust placed inside the drum with a water/starch binder.
People tend to interchange words that describe the art and science of consolidating particulate solids so I will start by defining a few terms that will be used in this article.
Agglomeration: Collected into a mass by means of agitation and or collision alone, thus without mechanical pressure in any kind of mold.
Balling: Producing rounded agglomerations
Binder: In order for agglomerations to stick together they need a binder. There are many binders for example starch, resins, tar, sewage mud [3], fish waste [3] algae [3], clay, cement, waste liquors [4], etc. A binder has been defined as something (such as tar or cement) that produces or promotes cohesion in loosely assembled substance [2]. There is some agreement that four binder classifications exist, 1 matrix type such as asphalt, 2 film type such as starch, chemical type such as lime plus molasses and 4 lubricant type such as oil and water [4]. I have my doubts that we would have a use for number 4.
Fireball: A charcoal and or biomass fuel that has been formed in a balling agglomerator.
Pellet: An agglomeration made in a balling agglomerator.
The danger in using charcoal as a fuel is the possibility of wasted wood. On an average one is fortunate if 30% of the heat value can be carried over to the charcoal , that was in the feedstock wood. This can contribute to deforestation. Ideally when producing charcoal one should find a use for the waste heat and or off gases. For example a dual mode gas producer was mainly used to produce charcoal and the producer gas was a byproduct that was used to fuel a 1000 kW dual fuel-diesel engine for thirty years. (Delacott System) [5].
One of the features of the charcoal fireball is the need for dust or fines as the feedstock. It is difficult to grind wood charcoal into dust but other biomass sources of charcoal or char are easier to reduce into fines. For example grass, leaves, sawdust etc. I have used switchgrass (Panicum virgatum) as a source for char. If wood charcoal is being produced the left over fines could be used for the production of fireballs.
Three balling agglomerators are described below:
Drum Agglomerator: This looks much like a rotating drum whose axis is inclined at a small angle to the horizontal. The entire feedstock bed moves forward in a plug flow fashion. Like the pan agglomerator the drum has many variations from axis tilt toward the inlet, tilt toward outlet, level axis, paddles inside the drum, etc. One of the most interesting variation on the drum is the falling-curtain agglomerator (O'Brien). Inside the drum is a rod-cage that resembles a squirrel cage blower. This rod-cage produces a curtain of dry fines that remain at a constant thickness thus preventing the sprayed liquid/binder coming in contact with the drum wall, as with all agglomerators it is best not to allow the surface to become too wet. Furthermore this curtain which preferentially agglomerates only the small particals lifted with the rod-cage. The falling-curtain also employees a ribbon or auger like device, fastened to the inside diameter, that returns the smaller particals to the inlet end.
Pan Agglomerator: Looks much like a shallow disc set on an angle between 40 and 60 degrees. The largest particles move on the top of the bed and discharge over the rim as more feedstock is added. The growth of the agglomerations can be controlled by the relative position of binder/liquid and feedstock. For example if the binder/liquid is introduced at 3 o'clock and dry feedstock is supplied at the 5 o'clock position the powder feedstock will stick to the wetted agglomerations and cause them to enlarge. Other positions will have different affects. Other factors are angle tilt, rim height, speed of rotation and amount of binder/liquid plus feedstock. Other forms of the pan agglomerator will have stepped pans that allow a separate re-roll area. This will aid in producing a more rounded, smooth and densified balled agglomeration. There is also the deep pan agglomerator.
Cone Agglomerator: This it a compromise between the pan and drum balling agglomerator. The cone angle and rotation speed control the pellet size. In order to provide a good surface for balling this agglomerator has a stationary and moving cutter bars. The axis angle can also be adjusted.
The list of balling agglomerators goes on and on. It seems to be more of an art than science, although the science has been greatly improved, hence one can dream up just about any form of balling agglomerator and make improvement until it works sufficiently for your needs. Or at least that is my impression of this method of pelleting.
I choose to employee a cement mixer for my experimental balling agglomerator. First the paddles need to be removed. Next add the char/charcoal fines, see picture 1. Now comes the difficult part of adding the liquid/binder. If the cement mixer is turning before the addition of liquid/binder you will have a sever dust problem. Ideally it would be beneficial to have a device that would spray the liquid/binder, I did not. So just before the cement mixer starts to rotate the spray would begin hence a form of dust control. In an experimental fashion, I have found that I could wet the dust lightly then start the cement mixer and continue to add the liquid/binder. I have made many mistakes and still have been able to produce some kind of balled agglomerations. Picture 2 shows the cement mixer in kind of a drum agglomerator mode and picture 3 shows it in more of a pan mode.
 Picture #1 |
 Picture #2 |
 Picture #3 |
My most embarrassing mistake was not knowing that the starch water solution (liquid/binder) needs to be boiled first! I used about 6% starch to water. I have also added too much liquid/binder. This will create a condition that causes the bed to lose traction with the drum and stop circulating. The material will also stop agglomerating and develop into one large clump. If you end up with a large clump, simply break it up with a stick and start over. My solution was to add more dry fines to the mix thus allowing formation of nuclei and next growth. The amount of moisture seems to be paramount.
Other fines can be blended with the charcoal fines such as sawdust.
I did measure the density of one batch of charcoal/sawdust fireballs. It was about 0.500 g/cm^3. This was not a 100% charcoal feedstock but had about 20% sawdust. How fine the feedstock is reduced and the amount of dwell time in the agglomerator has an influence on the density.
PICTURE #4
At this point I am concentrating my efforts in the direction of the inverted down-draft gasifier. This gasifier produces gas from the volatiles in biomass and as a by-product charcoal or char is made. This is the reason for favoring the fibrillated bio-fiber fireball over the charcoal fireball. Examples of this species is the paper and switchgrass fireball. Picture 4 shows a paper/switchgrass (chopped), paper, and charcoal fireball from left to right respectively. Picture 5 shows three 100% switchgrass fireballs. They where made by hand, simulating the action of an agglomerator. This classification of fireball is binderless hence hydrogen bonding. Producing them in an agglomerator is totally a different procedure than making fireballs from dust feedstock. With the fiber fireball, pulp is made out of the biomass first.
 Picture #5 |
 Picture #6 |
My best laboratory investment, to date, is the Woodgas Stove [6], a Tom Reed design. Picture 6 shows the stove loaded with paper fireballs. Picture 7 shows the stove gasifing/burning paper fireballs and picture 8 shows the charcoal produced. Simply put, what better way to test your fuel! The plan is to oven dry the fireballs, weigh the sample, then load the Woodgas Stove with them. Ignite the stove and place a measured quantity of water, in a pan, on top. Time the burn and measure how much of the water has been boiled away and measure how much charcoal was produced in the stove.
 Picture #7 |
 Picture #8 |
The charcoal, from the inverted down draft gasifier, will be used as a soil enrichment hence Terra Preta [7], this stands out among the opportunities for sustainable soil management. Basically it is a black soil found in parts of the world that had carbon added. This is a very fertile environment for plants to flourish in.
INFORMATION:
1 www.repp.org
2 “Elements II Briquetting and Agglomeration” (1983)
3 VITA, “Technical Paper #31, Understanding Briquetting” (1985)
4 “Elements of Briquetting and Agglomeration” (1977)
ISBN 0-920292-00-3
5 “Small Scale Gas Producer-Engine Systems; State of the Art”
Biomass Energy Foundation Press (www.woodgas.com)
7 www.css.cornell.edu/faculty/lehmann/index.htm