Once biodiesel has been properly settled and the glycerine drained it must then be washed. In this section we will be dealing with the water wash.One of the primary problems faced in water washing has been creating an emulsion. An emulsion can occur due to two major factors; the first being that the reaction is not complete and residual triglycerides remain allowing the bonding of glycerine molecules to the esters portion.These tend to not be very water soluble and thicken up creating a creamy-like textured emulsion. The other primary factor is allowing the glycerine layer to find it's way into the wash cycle.This glycerine layer is essentially a concentrated soap solution and if allowed into the wash cycle would result in an emulsion that must be broken, when possible. There are methods available to help in breaking an emulsion ranging from adding salt to the wash water by sprinkling it over the biodiesel as it is being washed to introducing raw glycerine by-product into the wash cycle. See Breaking an Emulsion . This sounds counter indicated from what was just said above, but with sufficient raw glycerine added to an already existing emulsion it can be broken this way. A little in mindset like using bubble gum to remove bubble gum from clothing. The easiest to avoid is the glycerine in the wash cycle aspect.

There are a couple of ways to work around this glycerine by-product getting into the wash cycle; one is to have the reactor tilted to the drain side by putting a 1" flat board on the drain side and then raising it up on the back side so that the slope of the draining liquid is accentuated. Should the NaOH-based glycerine by-product harden, as it is prone to do at times, especially when it gets colder and not in a heated space, then only part of it will drain at the opening of the glycerine drain (BP-2 on the main page) so it will require that the mix be reheated using the reactor's heating element, but just enough to liquify everything again, not to the 55C processing temps. Then a circulation of the mix must again take place completely remixing the biodiesel and glycerine followed by a subsequent settling time,only shorter this time (1-2 hours) and a repeat of the glycerine draining procedure done. This is very time consuming, but failure to do it could result in glycerine mixing with the biodiesel in the wash tank and the result would be excessive soap formation making washing difficult.
The other method to work around the problem, and the simplest in our opinion, is to have the entire mixture be moved to a seperate, alternate settling tank before settling time has begun. This is what we do. This requires another tank that is insulated to make sure it retains heat and is closed to the air due to hot methanol being present and allows the reaction to complete. The design of the second settling tank should be like that of the wash tank, the StandPipe design. This allows the finished biodiesel to sit atop of the glycerine by-product and then be drained off first via the StandPipe and afterwards the glycerine is drained via the floor dain. This eliminates the possibility of the glycerine by-product getting mixed with the biodiesel and entering the wash cycle. We use a purge valve at the lowest point in the plumbing to remove the little bit of glyerine that has settled into it.

Once the reacted biodiesel has had time to settle out the glycerine, the process isn't quite done yet. There is still residual lye and methanol suspended in the biodiesel that needs to be removed so that it will not damage the vehicle's engine or components. The greater lubricating properties of B100 (pure) biodiesel has been reported to lengthen engine life and is much more environmentally friendly than is petroleum diesel fuel, so we want it free of contaminants.
So we must remove the residual impurities that could harm our engine, and we do this by washing it, literally, with water. Once again using Sean Park's StandPipe design for the wash tank we are now ready to transfer the settled biodiesel into the wash tank to be cleaned up.

This is an area that has brought about much controversy and heated debate as to which method of washing is best. Of course the best method is the one that works. The University of Idaho initiated the bubble wash method some years ago, which is where an aquarium air pump and air ring or other delivery sytem for tiny bubbles are used to gently wash the biodiesel by blowing air into the water at the bottom of the wash tank causing the air / water bubble to rise and burst once at the top and releasing the water droplet to fall back down through the biodiesel effectively washing it on the way down. Sometimes this method is combined with misting, where a gentle mist is first used that falls like a gentle rain over the biodiesel and slowly passes through it washing it as it goes. In subsequent washes the bubbling method is used as a compliment to misting.

Following our own advice about using a seperate settling tank we do neither. The method we have adopted is quicker, does the job, requires more force, but has less room for error in processing. A small industrial pump by ShurFlo is hooked up to draw water from a heated water source (21C / 70F) and then pumped up to an industrial stainless steel 120 degree full cone spray head held over the biodiesel and allowed to spray for about 2 minutes for the first wash.

The strength of the spray is controlled by two seperate ball valves that open to the sprayer at one end and an open tube at the other. By opening the sprayer's full and the open tube's partially it effectively controls the strength with which the water is hitting the surface of the biodiesel. Then the whole thing settles in a few hours and then drained followed by another wash with the sprayer is initiated only this time for about 5 minutes washing time. Then settled again another few hours and the third and usually final wash takes about 10 minutes or so of rigorous washing with the pump drawing water from the bottom of the wash tank and sending it back up to the top to be driven through the biodiesel by the spray head with the flow control valve closed. This causes complete saturation of the biodiesel. The industrial 120 degree full cone sprayer covers the entire surface of the biodiesel ensuring complete saturation of the biodiesel and a complete wash as a result. It then settles overnight before being sent to the drying tank. This method has not yet become popular amongst homebased biodiesel producers, however the method has proven itself quite effective. Where there are no soaps (glycerine contamination) or unreacted oil there can be no emulsions therefore a more vigrous wash will not have any negative effects.

The industrial MP-187 stainless steel 120 degree full cone spray head by Bete in action.

More wash photos : Washing up close - Full Cone Spray Head - Full Coverage Spray - Hard Wash

Once the biodiesel has been completely washed of residual impurities it can then be "dried" to be certain that no water remains in it. There are many ingenius methods of accomplishing this from time storage / settling to open air exposure, to pumping it through the air and letting it fall back into itself ect. The method we use is by circulation via a 1" clear water pump, the same kind used for the main reactor, with a fan blowing air over it and an industrial FF-316 stainless steel 145 degree flat fan spray nozzle.

The industrial 145 degree stainless steel flat fan spray nozzle in 3/4" by Bete does a great job.


Then after a couple hours of drying with the heat going to 40-45C and then another hour with the heat turned off to allow the biodiesel to cool it is run through two truck filters. At first we used a 10 micron Fleetguard FS1000 (Donaldson P55-1000) spin on filter and head and the second a Perma Cool 81074, 2 micron spin on filter which can be replaced with a CAT 1R-0750/49 filter canister, but since have changed to 2x2 micron CAT cans mounted on Fleetguard FS-1000 heads.It is not necessary to use PermaCool heads for the CAT cans as the Fleetguard heads work just fine. They are equally 1-1/14 thread.
Filtering is done using the trickle method, where the circulation continues while the main return line is slightly restricted (by closing the ball valve a bit) at the same time the hose leading to the twin filters is slightly opened (another ball valve). This allows for the biodiesel to slowly filter taking full advantage of the filters' micron ratings without putting unecessary back pressure on the pump.. This will continue until the pump will no longer draw the biodiesel.


The FAME (fatty acid methyl esters) come out bright and polished and ready for use direct in the fuel tank or into storage.