[Stoves] Effect of mixing sugar and clay?

[Crispin Pemberton-Pigott]

Dear Marc

 

Does anyone know why folks recommend mixing sugar/molasses with clay when
making rocket stoves?



There are two parts to the answer. The first is that any sugary solution
will make the formed product stick together (before firing) and it is
important with ceramics to prevent micro-cracks during drying because later
that will be the place stresses accumulate and where the big cracks start.
When 'folk art ceramics' are being made, they are usually so bad that
anything you can do to improve the product will be acceptable. If adding
sugar will keep the thing together long enough to get fired it is going to
help.

 

When you can choose what to put into the mix, you would add some bentonite
for the same reason, but it would not get removed during the firing process.
Bentonite turns liquid when mechanically disturbed and 'freezes' when that
motion stops so it is good for unfired ('green') strength.

 

The second part is that during firing above 600 C (which most ceramics are)
all the carbonaceous are removed (mostly by CO moving back and forth
rapidly) and there is no carbon in the final product. That means there is
literally nothing from the sugar (sucrose) in the final product so it has to
have something to do with the pre-fired strength and drying process.


* Jon and Flip mention it in the build instructions of a rocket stove on
their new site <http://www.rechoroket.com/Home.html> . Some friends of mine
just tried it in Cameroon, and it seems to work. They didn't have a supply
of sugar, though. So they soaked + squeezed some bananas instead.

Bananas, quite by coincidence, have a lot of potassium in them as well as a
high sugar content. Potassium, lithium and several other things act as a
flux when heated in the presence of clay minerals, reducing the melting
temperature so if the product was being under-fired before, it appears to
have been improved by the sugar, but it was the flux. You could also try
finely chopped fresh grass which has two fluxes in it.

Firing clay is not a single event. There are minerals that melt, one by one,
as the temperature rises. At some point the blend of melted, sticky and
unmelted ingredients has the characteristic you want. In a domestic stove,
it is usually formability prior to firing, green strength and strong
resistance to thermal shock. In some cases, increased resistance to heat
conductance R is possible while retaining thermal shock resistance and
strength and the ability to form, dry and fire the part. In nearly all
cases, simple mixes of ordinary clay and sand are poor on all counts. Adding
sugar can make a mix stronger and slow the drying before it is fired. A lot
of microscopic damage is usually caused during drying of artisanal clay
products but they are not usually put to the fire afterwards, so to speak,
so no one notices. 

Resistance to thermal shock dominates the demands of clay stove parts.
Differential thermal expansion of the surface causes the clay parts to
disintegrate even if never touched by fuel. Clay mixes that are
traditionally made with trial-by-fire experimentation handed down for
generations have two characteristics (that I had heard about) which make
this possible.

The first is they have a low free silica content after firing, though it is
quite by accident and the artisans have no idea what they have got in there.
This causes the material to have a nearly linear expansion plot when fired,
dramatically (not 'significantly') reducing the silica phase change
(expansion from A to B phase) problem at 573 degrees. The addition of fluxes
can contribute to this suppression, but by a different mechanism. It is
possible to get very low deviations from a straight line expansion through
573° C without the product exhibiting a low Δ Length result. Technically it
is not a low thermal expansion material but it appears to work like one in a
limited temperature range. High free silica materials have a pronounced
physical change in size at 573° - I can send you graphs.

Baldosa tiles are an example of this curious characteristic of low free
silica which was discovered by Bruce Berger in a sample that actually
contained more than 50% silica when analysed by XRD and XRF. The tile was
from Guatemala and offered by Dean Still at ETHOS one year. It is an
artisanal tile that works in stoves even though it has a relatively high
thermal expansion. It does not work well because it is 'porous' or
'insulative' or 'traditional' or magical. It has a suppressed silica phase
transition line. It cannot survive repeated high temperature changes for
long but it is a lot better than pot-clay.

The winner is ceramic with a low thermal expansion. Stoves are much more
likely to last longer if they do not crack in the first place. In order to
get heat away from the surface of the combustion chamber and spread as
evenly as possible through the material, it should have the twin features of
high thermal conduction (which spread heat) and low thermal expansion (which
reduces heat stress). You will notice that this heat conduction feature is
directly contrary to the advice given by most stove construction advisors.
That is why most homemade stoves ceramics fall apart as soon as they are
heated a few times. It is hard to make industrial ceramics by accident and
even if you do, can you repeat it?

Adding lithium (often Petalite), particularly in the form of lithium
carbonate, will greatly reduce the thermal expansion of ceramics if they are
'set up' in advance by tuning the mixture to receive it. Correct analysis of
clay minerals can reasonably predict where the points on a phase transition
chart have a workable mix. Working with Bruce for 3 years on this problem
(2005-8) determined that having a low free silica and high strength was
probably more important than ultra-low thermal expansion, though both
approaches work. Flexible high temperature ceramics are another but unlikely
possibility. They are expensive and finicky.

Obviously one way to overcome fracturing is to greatly increase the strength
of the material. Adding sodium silicate to the clay in liquid form is a
proven method. There are some Rocket stoves made in Uganda with the Ugandan
flag colours painted around the outer metal shell. You may have seen them.
The ceramic is very dense and hard and that hardness is from the
introduction of sodium silicate and not adding any porosity. Porosity
aggravates the cracking problem so the next time someone tells you to make a
porous mix that simultaneously has a high thermal expansion and low
strength, stop listening, keep moving. Industrial refractory materials
feature a) low thermal expansion, b) mediocre ability to withstand large
sudden temperature changes like those experienced by the walls of a stove,
c) high working temperature (not usually needed in a stove) and
cost-is-no-object. A domestic stove is one of the most demanding
applications of ceramics. Cement kiln liners have a combination of high
mechanical strength, low thermal expansion and high density, and high cost.
They feel like stones. They make great kilns but you can't cut them and are
rare unless you live near a cement factory. In Dakar you can get them.
"Refractory" means it can take high temperature, but it does not mean it can
take high thermal shock. You can ruin a kiln just by opening the door a few
times when it is 1000 degrees inside. Things fall apart. Fired materials
have a cooling rate that should not be exceeded. One of the first questions
a ceramicist will ask is whether or not the stove part will have water
spilled on it occasionally. Rapid cooling is death to ceramics. Cold air
admitted to the hot combustion chamber is similar death.

Porous ceramics: just a little more on this topic. One can imagine a
flexible ceramic sponge that will survive repeated heating and cooling
cycles through the silica transition temperature (573). OK, nice to have if
you can find it. Hugh Allen, who made the Improved Kenyan Jiko Stove (IKJ as
it is properly known) into a commercially viable product, says it takes a
year of trial and error research to develop a rural, artisanal mix that will
survive for a year in use as a charcoal stove liner. Note that is not a
structural part, it is a part held together by steel sheeting. He wrote an
excellent book on the subject in about 1991 outlining the artisanal
approach.

The POCA stove (Maputo Ceramic Stove, MCS) uses the approach of limiting the
thermal expansion (to prevent cracks starting) and high strength so as to
provide structural components. It has no metal supports so is potentially a
much cheaper product than a metal stove, with many added material
advantages. The stove is produced by Peter Coughlin who subscribes to this
list and he is probably getting nervous as I raise his product's design
features. He is bringing out a new product in a few weeks which features a
blend that is a shift in approach from one of those above to another. This
will bring down his cost and improve the product because he is taking care
of multiple issues simultaneously. It is a technically advanced product
though looks quite simple. The intelligence is applied early on resulting in
a cheap, strong, durable stove body.

* In rural Vietnam, a brick kiln engineer explained to me that they lined
the inside of kilns with a traditional refractory material consisting of
clay + rice hull ash + molasses.

That is for the structural support prior to firing and limiting the
shrinkage cracks. It has to (obviously) be fired very slowly in order to get
the heat conducted evenly through the walls as they dry and de-carbonise.

* While talking to some old colleagues, I heard that old moonshine stills in
Georgia used to have a clay + sugar mix to seal pipe connections (very much
word of mouth, could have been inaccurate)

Ditto. When fired, there is nothing left from the sugar which is organic.
The mass lost is accounted for in the clay analysis report under 'loss on
ignition' or LOI. Oil, decomposed wood etc is in the same category: organic
material. It can increase the porosity and weaken the material if there is
enough of it.

* Apparently, in construction, you can retard the setting of concrete by
pouring Coca Cola onto it while it's wet. Useful if a truck gets stuck in
traffic when you're halfway through a pour.

That is because the sugar (not the Coke per se) destroys the bond between
the aggregate and the hydrating lime. Concrete trucks often carry a 50 pound
bag of sugar in the cab for just the event you foresaw. There was a company
in Canada that used to remove set concrete from large mixers if something
broke down. They drilled thousands of tiny holes all over the rotating drum
and used little explosive charges to break up the giant lump inside. My
friend David Hadden asked the guy how often his services were needed and he
said, "about once a year." I guess that was enough.  If a cement truck
breaks down at the roadside, the driver is told to put the sugar into the
mix. It prevents setting. If one put in coke, it would not delay setting, it
would ruin the mix, guaranteed.

Does anyone know what effect the sugar actually has? Does it just retard the
setting of the clay? 

It probably retards drying.  Clay does not set. Elements within it melt and
stick together (vitrification).

Or does it help improve the material properties? (or does retarding its
setting improve material properties when it's eventually fired?!?)

It retards drying in an environment where drying is poorly controlled.
Because this sort of forces 'proper drying' as would be done by a good
ceramics shop, it appears to be 'doing something'. It is compensating for a
lack of skill or facilities. At least partially. Proper milling/mixing
first, Bentonite and controlled drying would do a better job.

There is another approach that will also help which is to reduce the amount
of water used in the mix. If it is not in the clay, you don't have to remove
it.  Clay is hydrophilic so it usually expands when wetted.  This approach
was used by ProBEC in Malawi making very low cost ceramic top plates for
single pot stoves and was discussed at the time at length on this list so
the messages, drawings and photos are on file. The clay was dampened (if at
all) with a little water then it was pressed in a manual device which formed
the part with a very low water content. Wall tiles in bathrooms are made
this way, with moisture as low as 4%, under very high pressure. This solves
all sorts of problems. There is nearly no water in it to be removed so
shrinkage cracks are zero. Storing green parts outside for a while is not
likely to reduce the moisture content so if done properly, even at the
artisanal level, they can by put directly from the forming process into the
kiln and slowly cooked up to 120, held for a couple of hours, then heated to
the final temp, ramping at perhaps 135 deg per hour to 600 and then 150 or
more per hour to ...900, 1100 or whatever. If at any time the heating rate
exceeds these values, it is likely spalling, fractures and cracks will
appear on what could have been perfect products. The organics and water and
crystal water have to be removed carefully.

Incidentally regarding phosphates in the clay: phosphates act as a
surfactant and aid densification with little pressure. You can turn a
dry-ish mix liquid by adding phosphates. It is then possible to mix together
large and small amounts of additives evenly and quickly for a homogeneous
(accurate) mix. If someone put in phosphates accidentally they would also
see product improvement without perhaps realising that they are compensating
for poor preparation. You can re-stiffen the mix by adding a little HCl.
Neat, huh?

Regards

Crispin

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20120118/98668984/attachment.html>