Washing
Why bother?
Which way is best?
Mist-washing
Bubble washing
Oxidation and polymerisation
Stir washing
Washing tanks
Settling and separation
The washing process
Recycling the wash-water
Washing temperature
Emulsions
Using acid
Wash-water disposal
Drying the fuel
Why
bother?
Frequently
Asked Question:
"Do you really need to 'wash' biodiesel? I've been running a 50/50
biodiesel/petro diesel mix for a few months so far and my Dodge Cummins
runs great on it."
Answer:
"Yes, we do. Detergents in gasoline or petroleum diesel? Maybe.
Sodium or potassium soaps in biodiesel? Not. A can of HEET (methanol) in
the tank every once in awhile to resolve a water issue? Maybe. A dose of
methanol to hoses and seals with every inch on the odometer? Not. And
certainly there isn't any benefit to leaving caramelizing glycerin
floating freely in a fuel and headed for an injector. I think it best be
left to the petro-chemical distilleries to leave varnishes, paraffins
and the like in their fuel and serve it up as champagne to the
unsuspecting. As for 'no problems 'til now?' You might take note that
nothing is ever broken until it breaks." -- Todd
Swearingen, Appal Energy,
at the Biofuel mailing list.
Read what the Fuel
Injection Equipment (FIE) Manufacturers
(Delphi, Stanadyne, Denso, Bosch) have to say about the contaminants
left in unwashed biodiesel:
Summary
-- html
Full
document -- Acrobat file, 104kb
See: Determining
the Influence of Contaminants on Biodiesel Properties,
Jon H. Van Gerpen et al., Iowa State University, July 31, 1996 --
12,000-word report on contaminants and their effects. Acrobat file,
2.1Mb
Quality biodiesel must be washed.
Frequently
Asked Question:
"I don't wash my biodiesel, but I filter it through an automotive
oil filter, then through a 3 micron oil filter, this is after allowing
the biodiesel to settle in a 55 gal barrel for a couple days. Is this
adequate?"
Answer:
No. Settling and filtering will not remove the soap, catalyst, glycerin
or excess methanol contaminants.
Frequently
Asked Question:
"Methanol is good fuel, why wash out the excess methanol, what a
waste!"
Answer:
Methanol is good fuel in specially prepared race car engines, which are
not diesels. Excess methanol in unwashed biodiesel is not good fuel and
does harm, corroding the fuel injection system. Recover the methanol for
re-use or wash it out of the fuel.
Which
way is best?
Frequently
Asked Question:
"I was wondering if anyone could tell me which way might be best
for washing large batches of biodiesel? I was wondering if bubble
washing or top-mist washing would be better." -- Biofuel mailing
list, July 2004
Answer:
Perhaps neither.
First you have to know what an emulsion
is and why it matters. Briefly (more below), oil
and water don't mix, and well-made biodiesel should separate quickly and
cleanly from the wash water when it settles. But if the biodiesel isn't
made properly there'll be more soap in it than there should be, along
with half-processed oil molecules, monoglycerides and diglycerides (MGs
and DGs), which are emulsifiers. Emulsifiers are used to make stable
mixtures of oil and water, such as mayonnaise. Soap also does that. If
there's too much of this stuff in your biodiesel, some of the biodiesel
or all of it won't separate from the wash water, or at least not
quickly, and in some cases maybe not for weeks. You have an emulsion
problem.
In fact it's a quality
test -- if it doesn't separate, you need to work on improving your
processing. Properly made biodiesel with good completion of the reaction
will not emulsify.
You can "fix" an emulsion, or prevent
it forming (see below), and emerge
successfully... but with poor-grade fuel. The emulsifiers are fuel
contaminants, they can't be washed out, and the upper limits for them in
the national standards specifications are low. Above those limits
they'll cause injector coking and damage your engine, especially with
prolonged use.
Gentle and super-gentle washing techniques like bubble-washing and
top-mist washing simply mask the problem that the fuel may not have been
made properly in the first place.
Todd
Swearingen of
Appal
Energy
replied:
"Mist- and
bubble-washing have become 'institutionalized' amongst home-brewers
primarily due to the frequency of incomplete reactions that lead to
washing problems. They are both methods of treating incomplete
reactions with little kid's gloves in the hopes that any resulting
emulsion won't be 'too bad' or 'too noticeable.'
"No wash method should be used on any batch unless it is known
for certain that the reaction has completed. (See Quality
testing.) Unfortunately, many people fail to guarantee this for
themselves, opting to just 'go ahead and wash it gently.'"
Quite so. See below, Stir
washing.
There is a lot of confusion about this. Here's a
different reply to the same message, from a champion of super-gentle
washing methods:
"Either mistwash or
bubblewash works fine for big batches, but the same rules apply as for
small batches -- more water is needed for misting and less
emulsification happens, much less water is needed for bubblewashing
but more chance that emulsification can take place.
"I actually use both now -- mist for a few gallons and then I
bubblewash with good use of water recycling during bubblewashing. In
my 350 gallon wash tank, my small aquarium air pump gives fine
results, but not all of the air pumps might be able to handle such
large batches. It looks like a tiny amount of bubbles are rising but
it works really well -- less emulsification than the same air pump
agitating up a smaller batch."
-- Maria "Girl Mark" Alovert, July 2004
But what sort of fuel will
emulsify at all in a 350-gallon tank with only a small aquarium air pump
to "agitate" it? Well-made fuel can take rough washing with a
paint-stirrer without emulsifying, you can't emulsify it with an
aquarium air pump.
Strangely, the same person
had previously said this about mist-washing:
"The bad news is
that poor quality biodiesel can emulsify just from the agitation of
bubblewashing. The good news is that it's easy to make good
biodiesel... Some people use a super-gentle 'mist washing' method to
take vigorous agitation out of the picture. To me it seems that this
masks the real problem -- which isn't agitation, it's poor fuel
quality. I'd rather make sure I've produced fuel that contains less
soap and less emulsifying monoglycerides and diglycerides, instead of
decreasing the agitation to 'ease washing'."
-- "Bubblewashing 101", Maria Alovert, 12 May 2003
But Ms Alovert was now
using a sealed 200-litre hot water tank "Appleseed" reactor,
based on Dale Scroggins' excellent Touchless
Processor design, and the 1" Clear Water pump she was using to
agitate it is too
small for more than 100 litres -- it won't mix a 200-litre tank
batch thoroughly in the usual time allowed. There were also problems
with the plumbing of the Appleseed reactor, making clear separation of
the biodiesel from the soapy glycerine by-product layer difficult.
Poorly processed and possibly contaminated biodiesel needs super-gentle
washing methods.
But as Ms Alovert said earlier, it's easy to make good biodiesel. These
problems are easily solved.
"I went from 1
& 2 liter test batches to a small 15 liter processor and all went
fine. When I bumped up to a 30 gal batch I got incomplete reactions
and the emulsions during wash that can go with it. The washed and
dried biodiesel looked great, but produced more glycerine when a
sample was reprocessed.
Following advice from Keith Addison at Journey to Forever I scaled
down the volume of the batches, increased the temperature a few
degrees, and increased processing time. It is often necessary to tweak
the process as you scale up. It is much easier to wash good quality
biodiesel. It takes less time and it's less expensive to process it
right the first time than to have to reprocess a batch." -- Tom
Kelly,
Biofuel mailing list, 5 October 2005
Tom left emulsions behind
and now uses stir-washing (see below). We also use
stir-washing, so do many others. We make biodiesel all the time, but
it's years since we had an emulsion problem at Journey to Forever.
Mist-washing
Mist-washing uses a
superfine spray above the wash-tank to send a mist of water droplets
down onto the surface, creating zero agitation. It works, or anyway it
does wash the biodiesel, though it's slow and it uses a lot of water,
and usually the water is not re-used.
Why do it? After all, the basic quality test for biodiesel is to shake
it up violently with some water in a jar and see how long it takes to
separate. (See Quality
testing.) If you can't agitate it during the wash without it foaming
up or forming an emulsion then go back to the beginning and learn how to
make it properly. Then you can use more effective washing methods that
won't just mask poor processing.
Milky
water after the first wash...
2nd
wash -- murky water...
3rd
wash -- clear water at pH7, clear biodiesel on top
|
Bubble
washing
Bubble-washing was
developed at the University of Idaho and is popular among home-brewers.
It uses a small air-pump, usually an aquarium aerator pump with a
bubble-stone. Water is added to the biodiesel in the wash tank (usually
a quarter to a half as much water as biodiesel); the water sinks to the
bottom; throw in the bubble-stone, which also sinks to the bottom, and
switch on the pump. Air-bubbles (lots of little bubbles is best) rise
through the water and into the biodiesel, carrying a film of water
around them, which washes the biodiesel around the bubble. When it
reaches the surface the bubble bursts, leaving the water to sink back
down again, washing the fuel again.
Usually three or four washes are used, each of six to eight hours, often
less for the first wash, with a settling period of at least 1 hour
between washes (some people settle it for much longer). After it's
settled the water is removed via a bottom-drain and replaced with fresh
water.
Washing is completed when the water is clear after settling, with a pH
of 7 (or the same as your tap-water).
Some of the cheap air-bubble stones aren't biodiesel proof and can
crumble away any time from immediately to soon, especially the blue
ones, for some reason.
Get ceramic stones, they'll last indefinitely. A rough-grade carborundum
stone will also work, or a pumice stone, though you'll have to weight
it. Or try experimenting with wood, and weight that too.
Here's Aleks Kac's method: Washing.
Mike Pelly's method: Bubble
washing.
Advantages
of bubble-washing:
It's easy, it works, it doesn't take much effort -- add the water,
switch on the aerator, come back later, do it again. For added
convenience add a timer to switch off the air-pump after eight hours so
you don't have to come back until a couple hours later (or a week later
if you like) when it's settled and you can just change the water.
Disadvantages
of bubble-washing:
It might not take much effort but it takes a lot of time -- there are
quicker (and better) ways (see below).
Ian
uses three of these pneumatic exhaust silencers for bubble
washing. They don't rot, they're cheap and they work very well.
See: Ian's
vacuum biodiesel processor
|
Bubble-washing is
gentle and can mask an incomplete reaction, which agitation will reveal
immediately. (It's always wise to do a wash-test
first by shaking up a quarter or half a litre of the fuel with some
water in a sealed jar or a PET bottle.)
Oxidation
and polymerisation:
A more complicated problem with bubble-washing is fuel oxidation.
All oils are not the same. Some are 'drying' oils such as linseed oil,
which is why it's used in paints and varnishes. When it dries the oil
irreversibly polymerises
(polymerizes) into a tough, insoluble plastic-like solid. At the high
temperatures in internal combustion engines the process is accelerated.
Steadily accumulating films of tough, insoluble plastic-like solids are
not what you want in your engine and injector pump.
Polymerisation happens when the double bonds in unsaturated oil
molecules are broken by oxygen from the air or water. The oil oxidises,
forming peroxides (hydroperoxides), and the peroxides polymerise,
bonding with carbon to create a long and stable molecule called a
polymer (plastic). Another effect of oxidation is that the
hydroperoxides attack elastomers, such as rubber seals.
Without oxygen the oil can't polymerise.
Oxidation and polymerisation don't only affect the drying oils, there
are also semi-drying oils, many of which are commonly used to make
biodiesel, including sunflower and soy.
Saturated oils don't polymerise, unsaturated oils do. The level of
unsaturation is called the Iodine
Value (IV) --
the higher the IV the more unsaturated the oil, the faster it will
oxidise and the more it will polymerise. Linseed oil, tung oil and some
fish oils have IVs of between 170 and 185. Coconut oil has an IV of 10
and won't polymerise. More information on IV: High
Iodine Values
Converting unsaturated oils to biodiesel lessens the polymerising effect
but doesn't prevent it.
Mixing biodiesel made from high-IV oils with biodiesel made from
lower-IV oils also lessens the polymerising effect but doesn't prevent
it.
The Euro standard for biodiesel, EN
1421 of 2003, is the de facto world standard, now copied by
Australia, and soon by Japan. It sets both a maximum IV limit and an
oxidation stability limit. The IV limit is 120, which excludes soy oil
and sunflower oil as feedstocks for standard biodiesel in Europe, but
includes rapeseed oil, the main oil crop in Europe.
Soy is the main oil crop in the US, and the US ASTM
D-6751 standard sets no IV or oxidation limits.
In a paper published in 1996, leader of the Iowa State University
biodiesel project Prof. Jon H. Van Gerpen stated: "The
specification for biodiesel should include a limit on the tendency of
the fuel to oxidize and a limit on the maximum degree of oxidation
allowable for use of the fuel in diesel engines." See: Determining
the Influence of Contaminants on Biodiesel Properties,
Jon H. Van Gerpen et al., Iowa State University, July 31, 1996 (Acrobat
file, 2.1Mb)
A November 2005 report by the Southwest Research Institute for the
National Renewable Energy Laboratory of the US Dept. of Energy, An
Evaluation and Comparison of Test Methods to Measure the Oxidation
Stability of Neat Biodiesel,
proposes test methods suitable for the inclusion of oxidation stability
limits in the US ASTM D6751 biodiesel standard specifications. Acrobat
file, 816 kb:
http://www.nrel.gov/vehiclesandfuels/npbf/pdfs/38983.pdf
The issue is still being widely ignored or denied in the US, but the
European results are conclusive -- even when made from rapeseed oil with
its lower IV, well within the EU standard limits, oxidised biodiesel can
polymerise and damage pumps and engines. Anti-oxidant additives for
biodiesel sell well in Europe.
Homebrewers
and oxidation:
Many homebrewers in the US think biodiesel shouldn't be used in the
newer TDI or PDI diesel engines because the high injection pressures
cause polymerisation. High pressures cannot cause polymerisation, only
oxidation can do that.
Others think that if the biodiesel meets the US ASTM standard it will
not polymerise -- but the ASTM standard sets no limits for oxidation or
IV levels.
"I agree with Mike
Briggs [ Biodiesel Group, University of New Hampshire] on
polymerization. If we are talking about ASTM fuel, then it is a
non-issue... Polymerization was pretty much dismissed at at least one
session at the NBB [US National Biodiesel Board] conference in
February of this year, as biodiesel performed far better than
anticipated in the lecturers' tests... Meanwhile, quality control is
not enforced in this country, so what do you expect?"
-- Kumar Plocher, owner, Yokayo Biofuels in California, "Concern
over Bio polymerization", BioDieselNow Forums, 21 October 2004
Any problems there may
have been with TDI/PDI engines using biodiesel in the US could be due to
poor fuel quality, and they also could be due to oxidation and
polymerisation of high-IV soy biodiesel. But the problem is
unacknowledged or denied. (See also The
TDI-SVO controversy.)
From a more recent discussion, at the "wastewatts" group at
Yahoo, "Biodiesel, comparison of oils", 2 May 2006:
"Is there any one
out there that has experience with using different oils and comparison
of results when making biodiesel? I have been using Canola oil [IV 98]
for the last 3 years, but now need to find another oil supply in
addition to the canola I am getting. Have found Soya [IV 130] and
Cotton seed oil [IV 105] users. So would like to know what to expect
if I was to start processing these oils?"
-- Trevor, "wastewatts" group at yahoo, May 01, 2006
"There is no difference between soy, canola, cotton or the myriad
of other oils."
-- Steve Spence, Dir., Green Trust Sustainability & Renewable
Energy
"There's a comparison of various vegetable oils on the
journeytoforever.org site [linked] that details the cetane ratings,
melting points, Iodine Value, and oil yields from seed and nut crops.
High Iodine Values are reportedly associated with polymerisation and
subsequent possible engine damage (not just when using SVO but
biodiesel made from these oils as well). A link within this page
references the Euro standard for biodiesel, EN 1421, which sets the
maximum Iodine Value (IV) at 120. The IV level of 120 reportedly
includes rapeseed oil but excludes soybean and sunflower oil as
sources of standard biodiesel in Europe. ASTM standards in the United
States don't set Iodine Value limits."
-- Rosannah
"Very thorough article. Nice find!"
-- tinkerbill (william)
"The key word being "reportedly". In reality, when
picking up used fryer oil from a variety of sources, it tends to
average values out and there are no real world issues."
-- Steve Spence, Dir., Green Trust Sustainability & Renewable
Energy
In reality, mixing oils
with different IVs or mixing the biodiesel made from them has no effect
on the IV or the oxidation stability of a higher IV oil: if it is
oxidised it will polymerise no matter what other biodiesel you mix it
with. Different IVs are not "averaged out", and many biodiesel
homebrewers only have one source of used oil anyway.
Iodine Values are not "reportedly" associated with
polymerisation, they ARE associated with polymerisation. And there have
indeed been "real world issues" in the shape of damaged
engines.
All oils are not the same.
Another issue remains ignored -- the hydroperoxides formed in high-IV
biodiesel when it oxidises are more likely to rot elastomer injector
seals than biodiesel is.
Oil paint is based on high-IV vegetable oils. Paint doesn't dry when
it's in the paint tin with the lid sealed. It does dry if you paint a
thin film of it on a surface exposed to the air.
Biodiesel tends to cover everything it touches with a thin film. When
motors stand idle overnight or longer, thin films of biodiesel cover
many engine parts, including elastomer seals.
This isn't exactly scientific, but you could try painting a thin film of
your biodiesel on a piece of glass, or any clean, smooth surface, along
with a thin film of the oil you made it from, see if it starts getting
sticky to the touch and dries. Pour about 1/8" (2-3mm) of your
biodiesel into an uncovered shallow dish and keep an eye on it, watch
for it turning a reddish colour, darker and more viscous. If it happens,
check how long it took. You might want to consider finding a source of
better oil.
If you're using a higher-IV oil for biodiesel, make it well, store it
well, and use it fast.
See also Storing
biodiesel
Bubble-washing
and oxidation:
Without oxygen the oil can't oxidise and polymerise. Everyone knows what
a fish-tank aerator does: it pumps the water full of oxygen so the fish
can breathe. Bubble-washing does the same thing to biodiesel, pumping it
full of oxygen.
Bubble-washing certainly increases oxidation. We've been sent some lab
test results from Europe, or rather a precis of them. The home-brewed
WVO biodiesel tested ("Foolproof"
method) passed all the standards requirements except the Oxidation
stability level. Those home-brewers no longer use bubble-washing for
their fuel. They use stirrers.
If you do use bubble-washing, use the fuel quickly, especially if your
feedstock was soy or sunflower oil or another semi-drying oil. Storing
it for a long time might not be a good idea. There are a lot of
variables: the kind of oil you used, how you store it, the weather...
Keep it airtight in full containers in a cool, dark place. There are
also anti-oxidant additives which you could try. We have no experience
of them. With semi-drying or higher IV oils, try alternating or mixing
with batches of biodiesel made from lower IV oils in your tank. It won't
solve the problem but it will help.
We seldom use bubble-washing, we use a stirrer (see next).
Stir-washing,
the first wash -- homogenised biodiesel and water immediately
after a thorough mix with a paint stirrer.
3rd
wash -- clear water at pH7, clear biodiesel on top, no loss of
yield.
|
Stir
washing
Frequently
Asked Question:
"Is there a way to speed-up the water-biodiesel wash? I've read the
bubble washing techniques and understand it, but for the best quality of
biodiesel, it can take up to a week, with several washes."
Answer:
"You can speed up the washing process considerably. It involves the
following:
1) Throw out your mist washer.
2) Box up your bubble washer.
3) Make absolutely sure that you never try to wash an incomplete
reaction by test-washing a 1-ounce sample in a sealed jar.
4) Use a motor-driven impeller to mix the water/fuel mixture to the
point of appearing homogenous for about 5 minutes.
5) Let settle 1 hour.
6) Syphon off the top layer of fuel and repeat steps 4, 5 and 6 two more
cycles.
7) Let the fuel air-dry or heat to 120 deg F (48 deg C) to dry.
Depending upon the volume of oil in your batch, the hp/size of your
reactor tank and wash tank motors/impellers, you could get your entire
batch process time down to 24 hours.
Most people avoid mechanical mixing of the fuel at the wash stages,
thinking that it will create emulsion problems. And they're right if
they try to wash 'fuel' from incomplete reactions." -- Todd
Swearingen, Appal Energy,
at the Biofuel mailing list.
Stir-washing
a 60-litre batch -- in goes the first wash water, 20 litres of
it.
A
paint-stirrer on a drill agitates the whole tank (avoid making a
vortex, but it helps for photographs!).
The
double paint-stirrer. Three washes in a day leaves the biodiesel
clean and clear, with no emulsion formed and no loss of yield.
|
Comments:
- "Todd, I had a
chance to try your 'fast' wash method today (35 gal white poly drum)
with a simple 55 gal drum mixer and wanted to thank you for the time
you just gave back to me! Success! At least for me, this procedure
is much better and faster to wash biodiesel this way, compared to
the bubble-wash method. Just five minutes of 'appearing homogenous'
and allowing the 1 hour settling, drained-off and replaced the water
and repeated for a total of three wash cycles. The water and
biodiesel layers are currently clear and it's settling for 24 hours.
Excellent separation, no emulsion!" -- Kevin
Shea at
the Biofuel mailing list.
- Thank you Keith and
Todd -- I followed your advice and stir-washed the 15L batch instead
of pump-washing it. The first wash water came out a bit cloudy. The
second was crystal clear. Thanks again. -- Tom
Kelly,
Biofuel mailing list, 14 Aug 2005.
- It is my belief (may be
wrong) that the higher the quality of biodiesel the more rigorous
the wash agitation can be. The batches that I have made that cannot
take stir washing (emulsions occurred) have invariably been the
result of incomplete reactions. As the biodiesel I make has
increased in quality, it has stir washed very easily. -- Tom
Kelly,
Biofuel mailing list, 5 Nov 2005.
Advantages
of stir-washing:
Quick and effective, no masking of a poor reaction, no oxidation.
Disadvantages
of stir-washing:
There aren't any. BUT you have to process the fuel properly in the first
place. Some beginners have difficulty with this, all seems well until
they start the wash, and hit problems -- it foams, or it turns into
chicken soup and won't separate. And then they get discouraged. Don't be
discouraged, keep
trying! It's just a matter of practice and getting a good feel for
the biodiesel process. Make sure your titration and measurements are
accurate, be as meticulous as you can, make sure your chemicals are
fresh and high-quality, that you're using enough methanol, that you're
processing the fuel at the right temperature and agitating it for long
enough. Make small test batches for comparison -- see Poor
man's titration.
More advice from Todd Swearingen:
"You can use a prop
agitator for not much more than 5 minutes, enough to homogenize the
fuel and water, and 99% of the water will settle out of a 500 gallon
batch in an hour. You could probably conduct one wash every three
hours. Leave the last wash to settle 12-24 hours before drying the
fuel.
"Some people express concern with pump- or prop-washed fuel not
clearing as quickly as mist- or bubble-washed. There's sound reason
for this. Pumps and propellers have the ability of better mixing the
fuel and water ('atomizing' it), bringing both in more frequent
contact with each other. This means greater surface to surface contact
between water molecules and all suspended/dissolved impurities.
Fifteen minutes with a 1/2 hp motor and 4"-6" prop in a 200
gallon wash tank or bigger will achieve the same thing or more as an
all-day-affair with a mist- or bubble-washer. This allows for hours of
washing time to be converted to settling time, in turn hastening the
entire wash process.
"Pump washing has the potential to more quickly and thoroughly
'atomize' the two fluids and generate an emulsion in washes of
biodiesel that are borderline to or perhaps far less than a reaction
completion than does mechanical agitation."
Using the same pump you
use for the biodiesel reaction to wash the fuel might save some money
but before you process your next batch you'll have to flush it out
thoroughly with biodiesel. The same goes for using the reactor vessel as
a wash-tank. It is an option if you're short of space, but having a
separate wash-tank with its own stirrer is a much better option.
If you use a sealed water heater tank as a reactor, there'll probably be
residues of soapy glycerine by-product that didn't drain left in it to
foul up the washing process, and the wash process will leave traces of
water to foul up the next biodiesel batch. Get a separate wash-tank.
(If you do manage to get some by-product in the fuel when you're
transferring it to the wash-tank, let it all settle again overnight in
the wash tank, then add up to a quarter as much hot water, pour it over
the surface of the fuel. Let it settle for a few hours and separate it.
Do a wash-test -- see Quality
testing. If the fuel separates quickly from the water, continue with
washing the batch as usual. If not, treat again with hot water, or use a
gentle washing method.)
A
revolving garden sprinkler does a good job of mixing fuel and
water. Water is drawn from the bottom of the tank and sent to
the sprinkler via a small pump.
|
An alternative to stir washing is to use a small pump and a garden
sprinkler.
This has nothing in common with top-mist washing, there's nothing very
gentle about it -- it homogenises biodiesel and water the same as
mechanical stirring does though it takes a bit longer, about 15 or 20
minutes. It splashes a bit, but the danger of oxidation is very much
lower than with bubble-washing's millions of little air-bubbles.
Washing
tanks
You need a separate
washing tank. If you have the space have two wash tanks, it really
increases your production options. If you're short of space you can
compromise by forcing the reactor vessel to serve as a washing tank as
well (see above), but don't expect to like it.
The wash tank should be the same size as the processor tank or bigger.
We wash 60-litre batches in 90-litre tanks, with 1/3 water v/v oil for
each wash, ie 60 litres of biodiesel and 20 litres of water. That leaves
a few inches of headroom, enough for stirring or sprinkling or bubbling.
(See photographs in Stir washing section, above.)
 The
tank should have a bottom drain fitted with a valve. Cone-shaped
tank bottoms help, but it also helps if it's convex, which you can
achieve by standing a flat-bottomed tank on a few bricks under the
bottom edge (which is usually a lip) and bashing the bottom down a few
millimetres with a heavy bit of lumber wood. Often the drain fitting has
a raised edge a millimetre or so above the floor of the tank, which is
annoying but you can live with it.
If there's no bottom drain and it's a closed drum there should be filler
caps in the lid and one of them might take the standard plumbing fitting
thread on a 3/4-inch ball valve or gate valve. Closed 200-litre 55-gal
drums have two caps, one with a standard 3/4" thread. Cut the
bottom of the tank out instead of the top and turn the tank upside down,
fit a drain valve to the cap.
Otherwise cut a hole for the drain and weld a female-threaded valve
fitting into place. Use the fitting to mark out the size of the hole and
carefully drill a circle of small holes, maybe 2mm, next to each other
just inside the circle. Cut through the metal between the holes, remove
the inner bit of metal, and use a half-round file to file the edges of
the hole smooth until it fits the valve fitting. If you do it carefully
you should be able to screw the valve fitting into the hole. Then weld
or braze the fitting in place.
Some people use a standpipe in the washing tank, with two drain outlets.
One has a length of pipe inserted into the outlet inside the tank,
calculated so that the open end will be higher than the depth of the
wash water in the bottom of the tank. You drain the wash water from the
ordinary outlet for the first washes, but after the final wash is
finished use the standpipe outlet to drain the biodiesel before draining
the wash water, which makes certain you don't get any of the water along
with the biodiesel.
We use standpipes in the oil pre-heating tank, the settling tank and the
fuel and oil storage tanks, but not in the washing tank. A standpipe
only helps with standard-sized batches and the standard amount of
wash-water, and we don't find any difficulty getting clean separation
after the final wash anyway. See the standpipe
set-up in the settling tank used with the Journey to Forever
90-litre processor, which works the same way. Glycerine moves like
treacle and sticks to the inside walls; bottom-drain separation isn't
very efficient. The standpipe keeps the biodiesel above the problem. The
same problem doesn't exist with water removal. Nonetheless, many people
find standpipe wash tanks a useful option.
Wash
tank lid, on the right the hose-barb connection for the transfer
pipe from the settling tank. The other fitting is a water inlet
but we never use it.
|
You can make a lid for the tank from a piece of compound board cut to
fit with sheet plastic glued on the underneath surface and four toggle
latches. Give it a silicon seal too if you like -- apply a generous
helping of silicon round the underside edge, allow to set for 6-8 hours
(depending on the weather) until it isn't sticky to the touch anymore
but is still soft. Then place the lid carefully in position on top of
the processor and weigh it down with something heavy, making sure the
weight is spread evenly around the edge. The silicon takes the shape of
the rim but doesn't stick -- no need for a release agent.
Rig a stand for the wash tank that's high enough for a bucket to fit
easily under the outlet valve. We use steel stands cut and welded from
angle iron. Consider that a 200-litre 55-gal steel drum with a batch of
biodiesel and wash-water weighs about 350 lbs, 160 kg, don't use a
flimsy or wobbly stand for it.
Settling
and separation
After processing, settle
for at least 12 hours, preferably 24 hours, before separating the
glycerine by-product and starting with the wash.
Be careful separating the by-product layer from the biodiesel, it's very
soapy stuff -- much rather have some biodiesel in the by-product than
by-product in the biodiesel when you want to wash it.
If you do manage to get some by-product in the fuel when you're
transferring it to the wash-tank, let it all settle again overnight in
the wash tank, then add up to a quarter as much hot water, pour it over
the surface of the fuel. Do a wash-test -- see Quality
testing. If the fuel separates quickly from the water, continue with
washing the batch as usual. If not, treat again with hot water, or use a
gentle washing method.
This is how we separate
the glycerine by-product from the biodiesel, via a standpipe in the
settling tank (see caption).
The
washing process
Transfer the biodiesel
from the settling tank/processor, leaving ALL the glycerine by-product
behind. Add water, wash.
Most people use 25% to 50% as much water as biodiesel or more for each
wash. We use one-third as much, 20 litres of water and 60 litres of
biodiesel.
Take no notice of advice that you have to pour the water in carefully
and gently, that you mustn't pour it straight into the biodiesel, pour
it against the side of the tank instead like you fill a beer mug or you
might emulsify it. If you've used wash-tests to tune your processing and
you test each batch before washing, just tip the water into the tank, it
won't froth over like beer. Or pump it in through the lid.
The water sinks to the bottom. Proceed with the wash, whether stir,
sprinkle or bubble.
Settle the fuel between washes for as long as you practically can before
replacing the water and starting the next wash. Generally, the longer it
settles, the more effective the next wash will be.
The first wash water turns white, like milk. Washing is completed when
the water is clear after settling, with a pH of 7 (or the same as your
tap-water), usually after the third or fourth wash.
Freshly washed biodiesel is often a bit cloudy. It should clear in
anything from a day to a week. Leave it in a container out in the sun,
or heat it to 45-50 deg C (113-122 deg F) and let it cool in a vented
container. If it goes cloudy again on cooling you probably haven't
washed it thoroughly enough. Give it another wash, or if you're in a
hurry, heating it to 45-50 deg C a second time might clear it. When it's
clear go ahead and use it. See Drying the
fuel, below.
If it doesn't clear, it probably needs reprocessing. Try the Reprocessing
test and the Methanol
test.
Recycling
the wash-water
The counter-current method
of re-using the washwater reduces water use for three washes by
two-thirds.
The water doesn't become saturated with the contaminants it removes from
the biodiesel. Instead a sort of equilibrium is reached between the
contaminants in the water and the contaminants still left in the
biodiesel. Water from the first wash contains the most, with less in the
second wash water and less again in the third. Water from the second and
third washes can be re-used in washing your next batch of biodiesel. The
second wash water is used for the first wash of the next batch, the
third wash water for the second wash, with fresh water for the third
wash. Only the water from the first wash is disposed of, and only when
it has already been used for three washes in three successive batches of
biodiesel.
See Todd
Swearingen's
explanation of why it's a good idea to recycle the water when washing
biodiesel and how it works.
The
Economy of Wash Water Recycling
Washing
temperature
It works better in summer
than in winter, and the more cold it gets the less effective the washing
-- it takes longer and uses more water.
If your processor and wash-tank are in a heated room you shouldn't have
any trouble. Otherwise, either heat the biodiesel or, easier, use hot
water. Try to get the overall temperature to at least 30 deg C (86 deg
F) or more. We use hot water in the winter. It's cheap for us, but it
still uses energy, and energy isn't cheap no matter how little you pay
for it. We have a Kyocera solar water heater, we just have to plumb it
in, and then we'll use that.
Emulsions
... shouldn't happen if
you processed the fuel properly (see above). When they do happen, the
first thing to think about is why it happened so you can prevent it
happening next time -- make good fuel!
The second thing to think about is how to save the batch. One batch of
sub-standard biodiesel with too much glycerine in it won't hurt your
motor (though constant use of it will), so rescue your bad batch and use
it. Best to mix the bad batch in with better batches in your fuel tank.
A mild emulsion makes a third layer between the biodiesel and the
wash-water after your first wash has settled. A paper-thin layer is
fairly normal, just go on with the second wash. Thicker layers are not
normal. Drain the layer off, with a bit of water and biodiesel on either
side, and set it aside to settle, it will eventually separate. Heating
it will usually separate it. Leave it to dry, then add the top layer to
your next batch of oil for reprocessing. Meanwhile continue with the
second wash, which shouldn't produce an emulsion, or much less anyway.
A severe emulsion is when the entire tankful turns into a third-layer --
it looks like you have a tankful of chicken-soup or dirty-looking
mayonnaise.
If you leave it overnight or for 24 hours it will usually separate, and
again you'll be left with a third layer in the middle to deal with, as
above. It might take longer to separate, several days or a week,
especially in cold weather. Meanwhile your wash-tank is full of
mayonnaise and you can't wash the next batch.
There are several ways of breaking the emulsion. Heating it will usually
work -- 50-60 deg C (122-140 deg F) is usually enough. As with drying
the oil before processing, if you use higher temperatures beware of
causing steam explosions. Another way is to add hot water to the mix, if
there's enough headroom in your wash-tank. The hotter the better. That
should at least start the separation process: allow it to settle, drain
off any water that's separated, add more hot water, and so on.
Another way is to add ordinary table-salt, sodium chloride. Dilute the
salt in water first, add while stirring gently, keep adding until it
starts to separate.
This works well to break the emulsion, the trouble is that salt is an
impurity that you should be washing out instead of washing in. It will
take more washing than usual to get all the salt out and end up with
clear, neutral water. And salty wash-water is not as innocuous as
ordinary wash-water, so take care in disposing of it.
A third way is to use some acid to "quench" the wash, which
will also break an emulsion. See next section.
Whichever method you used, once the emulsion is broken, continue with
the rest of the washes until clear and dry as usual. Now you can use
your rescued batch while pondering how
to make better fuel next time.
Using
acid
Some people say it's a
mistake to "quench" the wash with acid. They say that while
the acid will neutralize the alkali catalyst, it will also convert some
of the soap to Free Fatty Acids (FFAs), which dissolve back into the
biodiesel, and which you don't want in your fuel.
We seldom use acid in the wash, but many people get good results using
an acid quench. Some of them have had their biodiesel analysed, again
with good results -- acid-quenched biodiesel can and does meet and
exceed standard specifications such as EN
14214 and ASTM D-6751. It's a normal procedure in commercial
production.
Does it in fact convert soap to FFAs? Typical quantities of acid used
are: 8 milliliters of 5% vinegar per liter of wash water used (usually a
quarter to half as much water as the amount of oil used); 2 milliliters
of 10% phosphoric acid per liter of washing water; enough ("not
much") 5% phosphoric acid to bring the pH down to 7 (neutral).
(Other acids used are sulphuric, citric or acetic, all well-diluted.)
This is how to convert soap to FFAs: Separating
glycerine/FFAs. What it says is that for
"average" WVO, titrating at 3 ml, you'd need 9.75 ml of 85%
phosphoric acid per liter of oil used to convert the soap to FFAs, thus
separating it from the glycerine. That's about 50 times as much acid as
you'd use to quench the wash, and in the wash it's buffered by all that
water too. FFAs? Naah.
But be sparing with the acid anyway.
An acid-quench will also break an emulsion or prevent it forming, which
might disguise a processing problem that the emulsion would reveal.
Always do a wash-test
first! You can "break" an emulsion by slowly adding dilute
acid (phosphoric, acetic, citric, sulphuric, 10% concentration) until
you see it start to separate (see "Emulsions", above).
Wash-water
disposal
Used wash-water is about
as innocuous as the average family's laundry water, or probably more
innocuous if they have infants in diapers. It's at its most innocuous if
you recover
the excess methanol for reuse before washing. Most of the excess
methanol and lye are in the glycerine by-product layer, not the
wash-water. Lye is used as drain-cleaner anyway, and any waste-water
management system handles soaps all the time. Even the methanol might be
an advantage (some of the waste-water digester bacteria like it) --
check with your local authority. See More
about methanol. The rest of it is probably not much
different to washing up after dinner. So you won't be causing any
problems if you simply flush it down the drain. Or you can treat used
wash-water in a simple greywater system using biological filtration with
water plants such as water-hyacinth and duckweed, and then reuse it.
There are many resources on the web to teach you how to do this. You can
also dilute it and spray it over meadows, fields, lawns, do some
small-scale tests first to get the dilution right. See Methanol
as a plant nutrient.
Drying
the fuel
When the fuel is clear --
not colourless but translucent, you can see through it, there's no haze
or cloudiness -- then it's dry. Actually it's never dry: despite what
the standards say, it always absorbs some water from the atmosphere,
between 1,200 and 1,500 ppm, but this is dissolved water, which is
harmless, unlike suspended water.
It should clear by itself in anything from a day or so to a week after
you separate the final batch of wash-water. Leaving it in the sun will
help, or heat it to 45-50 deg C (113-122 deg F) and let it cool in a
vented container. If it goes cloudy again on cooling you probably
haven't washed it thoroughly enough. Give it another wash. If you're in
a hurry heating it to 45-50 deg C a second time might clear it.
If it doesn't clear, it probably needs reprocessing. Try the Reprocessing
test and the Methanol
test.
Another method is bubble-drying -- the same as bubble-washing but
without the water. Drain the final batch of wash-water and use a
bubble-stone and an aquarium air-pump to bubble air through the washed
fuel. This will usually clear the fuel overnight. Sometimes it doesn't
stay clear, in which case it might need another wash.
The disadvantage of bubble-drying is the same as with bubble-washing --
it oxidises the fuel, and you could
have problems with the fuel polymerising unless you use it quickly. It
won't pass the European EN 1421 standards requirements. We don't
recommend bubble-drying for this reason. Our fuel is usually clear in a
day or two anyway.
Biofuels at
Journey to Forever
Biofuel
En español
-- Biocombustibles, biodiesel
Biofuels
Library
Biofuels
supplies and suppliers
Biodiesel
Make your own
biodiesel
Mike Pelly's
recipe
Two-stage
biodiesel process
FOOLPROOF
biodiesel process
Biodiesel
processors
Biodiesel in
Hong Kong
Nitrogen Oxide
emissions
Glycerine
Biodiesel
resources on the Web
Do diesels
have a future?
Vegetable oil
yields and characteristics
Washing
Biodiesel
and your vehicle
Food or fuel?
Straight
vegetable oil as diesel fuel
Ethanol
Ethanol
resources on the Web
Is ethanol
energy-efficient?
© Copyright of all original material on
this website is the property of Keith Addison, unless otherwise stated,
and may be copied and distributed for non-commercial education purposes
only as long as the source of the material is stated and a reference to
the Journey to Forever website URL is included (http://journeytoforever.org/).
All material is provided “as is” without guarantees or warranty of
any kind, either expressed or implied.
|
