Hi Ryan

I haven't tried this with KOH - I'm looking forward to someone giving it a try and reporting back.

Please be sure to protect yourself from the methanol fumes - they pack a punch at the beginning of this process and I'd hate for anyone to get harmed.

I made up a breathing tube described here, which I've found to work very well, for very little cost or effort, though some folks seem to prefer a more traditional mask solution.

Wishing you success!

Graham,

How about distilling the methanol out of the biodiesel (after glycerin has been drained) while the biodiesel is still in the appleseed. This way you could get a higher recovery rate on your methanol - while still keeping everything toxic and nasty in a safe, confined place. Once all the methanol is gone, THEN transfer the biodiesel over to your "Air Wash" rig...

Hi Dogma,

I may take this approach. But not just yet...

The reason I didn't do it in the processor is that the soaps become clumped and need to be 'harvested'
periodically. You really need access to it. Plus, I wanted to be able to see the whole process happening, to get an idea of how it works.

I took most of the bd out of the 'AirWash' rig (good name, thanks!) at the end of the process and was surprised to see this hard caked soap/glycerol deposit at the bottom.

There's around 8 litres biodiesel still in there, the tank was tilted towards me.

I just did another shake-em-up test with some of this little remaining BD and it still shows no visible sign of soap in the BD - The water remains crystal clear....



You can see how you need to be able to clean this stuff out before the next batch.

I will do another 80 litre batch this coming weekend, but will do a 5% prewash in the reactor and leave it to settle for a few hours before I move it on to the AirWash tank.

We'll see how that works out.

I've cleaned the crud out - got 1.5 litres of it. Consistency and colour of Apricot marmalade.

It is a mixture of soap and glycerine.

I dissolved some in water and found it to be very mildly alkaline.

I did a 'tongue test' on a pinhead sized dab of it - no sharp bite, so very little if any NaOH, slightly sweet, so some glycerine - mostly soapy.



It seems great for handwashing, behaves just like Swarfega industrial hand cleaner.

More to follow when the next prewashed batch is complete.

hmmm.

Okay... list of things i'd like to see tried:

(1) Separate the glycerine from the biodiesel.
(2) Begin to distill the methanol from the biodiesel (while still in the appleseed)
(3) Transfer mix over to Airwash(TM) rig
(4) Air wash it.

I totally agree with your assessment (especially since this is a "new" idea) that its nice to be able to to watch what is going on. But, i think that it should be possible to lower the methanol concentration considerably (thus reducing the hazard) before the airwash stage starts.

I'm thinking maybe distilling the methanol in 5 minute increments and then transfer to the Air wash rig via the sock filter.

Graham - the experts have told us that this method will result in a lower yield of biodiesel. How much loss do you think is realistic to espect? 1/2 percent? 1 percent? 10 percent?

Hi Dogma,

I'm happy to give it a try, but want to understand a bit more about the chemistry first.

Neutral, Tilly can you offer advice?

Can I safely heat the raw BD to drive off MeOH without risk of reverse reaction? (converting methyl esters back to glycerides)

Would 5% prewash reduce this risk?

If there is a risk of reversing the reaction, what would be the maximum threshold temperature before reverse reaction occurs, for both un-prewashed and prewashed BD?

I guess I'd need to control or limit my heat input to prevent localised hotspots. An electrical element may be unsuitable.

Perhaps I can use the secondary heating coil in the processor to take hot water, I can then control the water temperature accurately to ensure I dont exceed max. temperature input.

It would certainly be an added bonus to reclaim MeOH rather than vent it or wash it away.

As to yield, I'd like everything to be as efficient as possible.

There's only so much soap I can use in a month!

But surplus oil I can use for heating in winter, or for process heating. My diesel burning water heater is just asking to be used for this new process.

Thoughts?

As long as the vast majority of the glycerin is out of the mix, then there can't be a reverse reaction.

Hello Graham

If you remove the biodiesel from the glycerine layer the reaction will not be reversed..

Hello Producer
Triglycerides easily pass through an auto filter. Mono and di glycerides are a smaller molecule than the triglyceride.
I typically made my winter fuel on 15% methanol and summer fuel on 8-12% methanol.
Neutral advised this should give a nice healthy crop of mono and di glycerides.
I never experienced filter clogging using this fuel.

Many thanks Tilly

Do you know if there is a threshold temperature
below which reversing the reaction would never occur?

The reason I ask is that I expect there will still be some residual glycerine in the raw biodiesel.

Regards,

Hello Graham
I doubt that it depends on temperature

Hi Graham

This is a most interesting, possibly revolutionry technique - you are in serious danger of becoming the UK's answer to 'Girl Mark' - just watch you don't start dying your hair with phenolphthalein

Re the amount of glycerine, my last batch of 80 litres had the 5% water prewash and I drew off the glyc after about two hours. I then ran the pump for a couple of minutes to free up any residual glyc in the pipework then left it to settle overnight. Next morning I drew off one litre as carefully as possible and there was only a tablespoonful of glyc in it. The bulk then sat in the processor for a week before I had a chance to wash it when I carefully drew off another litre to see how much more glyc had dropped out. This time I got about half a cupfull but it was much runnier and lighter in colour than normal, suggesting that it contained a significant quantity of water. Based on these observations, it would suggest that no more than 70ml or so of glyc remains in the bulk (80 litre) after drawing off the glyc following the initial 2 hour settling period. For reference, my oil titrates at 0.7 - I used 5.7g NaOH/litre and 20% methanol by volume. I reacted at 55*C for 1.5 hours after addition of the methoxide had completed.

Cheers

Nick

quote:

Originally posted by GrahamLaming:

Hi Folks,

I believe the loss of methanol thru evaporation has rendered soap and glycerol insoluble in the biodiesel. Or, better put, has rendered the methanol-free biodiesel a poor solvent of soap and glycerol.

[I've deleted my mistaken theory that water was holding the soap - a chemist friend confirms that it is the methanol more so than the water]

I have also been giving some thought to a safe way of getting air through the raw fuel. How about an open head drum (the type with a lid that can be sealed with circular clip) so that you can get at it for cleaning? You then have a pipe going down to near the bottom of the drum to feed in the air and a second pipe fixed to the top of the lid which goes straight outside. No need to get anywhere near the methanol fumes.

If it turns out to be reproducible and reliable, this wonderful discovery of yours could be the most important developmnent in home brewing since the design of the Appleseed. Like many of the best ideas, it's so beautifully simple, you wonder why no-one thought of it before.

Cheers and thanks for sharing!

Nick

Limeys say reckon?
Sorry, everyone teases me bout the way I taulk. I have been following this thread closely, probably will make my first true batch, besides tests, with this method over the weekend. Hey, all that could happen is I ruin my pickup.
Cheers!

[Edited 20 June 06 - A closed system is desirable. My open system showed that you could achieve soap-free biodiesel by removing methanol, but it is not acceptable to use this process for general production. Too much methanol vapour released. Much better would be to distil the methanol from the batch in a closed system. Because of this, I've removed reference to the open vented system. - GL ]

When oil purity is good enough, I propose the BD is is pumped to a separate recirculating filter tank for final cleaning / storage.

This is because I notice the main cleaning tank has become coated in a soap gel. It seems to stay together pretty well, and doesn't readily mix into the BD, but it could be freed and released into the body of the fluid if disturbed, for example when changing the bag filter, or by back-flow in the pipework when you switch the pump off.

A separate tank would only need to deal with nominal amounts of soap, so would scrub up the BD quickly, with little residue to settle out or be swirled up.

Things are getting complex now! It would be very difficult to hold positive pressure in the room because of gaps between roof and walls, air bricks etc. You'd have to build a room within a room (I am familiar with clean room technology so I'm talking from experience). I wondered if you could perhaps eliminate most of the filtration.... suppose you bubble air through the raw fuel but just forget about the soap which comes up to the top. Once all the water and methanol were removed, would a short settlement period drop all of the soap to the bottom? Would then a single filtration to say 5 microns do?

Nick

I rigged up a sparkless vent fan in one wall of my processing area, and attached an "elephant trunk" to it made from a pair of Harbor Freight collapsible laundry hampers. So I can lower the end of it over the open top barrel I use to evaporate methanol. This will have to do until I get a still put together to recover the methanol.

Twenty4Seven

The soap may float.

Consider:
Specific Gravity (SG) of Biodiesel = 0.88
SG of Soap = 0.8
SG of Water = 1.0
SG of Glycerol = 1.26
SG of Methanol = 0.79

Pure soap will float in BD. Pure glycerol will sink in BD.

This procedure calls for the removal of both water and methanol. And most of the glycerol was already settled out and removed before starting the air bubbling/methanol evaporation.

If the soap produced is pure it will float. If large amounts of soap attach to very small residual amounts of glycerol the agglomerated particle may float. If the residual glycerol attaches to small amounts of soap, then this particle will sink.

So, most of the material to be removed will be "pure" soap floating on the surface.

I believe this has been Graham's experience.

I started another batch yesterday at 17:00 Friday 16 June.

It is now 09:30 Saturday 17 June and I have 80 litres of clean, dry biodiesel.

This is just over 1/2 a day from start to finish, starting with buckets of cold used oil.

I processed 80 litres at 60C, 5.5g NaOH/l 22% MeOH

I used 5% prewash, and the soap content of the BD was much lower than my last batch when I used no prewash.

The water content in the remaining BD seems to have been insignificant.

I settled for 1 hour after prewash, before draining.

I did not use a filter during the spraying process - it appears not to be necessary.

I drove off the methanol by evaporation, but this is not environmentally sound - it would seem to be be an environmentally good idea to condense the vapour, so we can re-use it in the next process. [ edit 20 June 2006 - removed the evaporation section because it releases too much methanol vapour into the environment and I now realise it should be either re-captured as distillate or processed in some other way - GL].

It is necessary to filter the final batch before storage, just to be sure there are no lumps of soap. But any soap in the BD truly is in lump form, I cannot detect any strands or wisps, so filtering should be straightforward after settling.

The bd shake-up test is again excellent, no evidence of water layer clouding.

Remarkable!
What arrangement did you use to draw the air through? Could you detect any evidence of biodiesel being carried out with the expelled air - as a fine mist say? Did you skim off any soap from the surface of did it all sink to the bottom? If you're only getting a litre or so of soap fall-out do you think you'd have to clean it out before drying the next batch or could you let it build up for a few batches before cleaning the barrel?

Cheers

Nick

Hi Nick,

The air is moved by the jet of biodiesel. There is no need to use a fan.

There was a very small amount of atomisation evident in the outlet pipe in the very beginning, I suspect caused by the methanol being given off rapidly, and by the initial high temperature of the liquid, but this disappeared after 1/2 an hour or so. I don't think this will be a problem.

I skimmed off a thick layer of soap this morning, just like the skin on a rice-pudding or custard. It was covering 2/3 of the surface area of the BD, with a clear area directly under the spray. The reason it floats is that it has mant entrapped air bubbles.

It came out in one piece, but there is a soap gel on the bottom of the tank too - I'm not sure how much, as I haven't emptied the tank completely.

I took 25 litres out, filtered it once thru a 10 micron bag, once thru a 5 micron bag and it is sparkling clear, soap free and dry.

The bags are now flowing slowly, so have caught a fair amount of soap.

The suspended soap tends to sink, not float. A settling period of an hour or so may be enough to drop the bulk out.

I have left the remaining BD in the tank settling, I will fill another 25 litre container after lunch, filter with new bags, and see how they perform after the settling period.

The soap drops rapidly, so there is clearly some glycerol binding to it. This seems to be an advantage. I have a 1 litre measurring flask, tall and narrow, which is ideal for viewing the soap settle.

Because I only settled after prewash for 1 hour, I'd like to do a longer settle next time, maybe 5 hours, to see how the soap behaves then. There must be an optimum settling time, beyond which no further advantage is gained.

I propose to close off this thread and start a fresh one with a more appropriate title (with a referring link back here, so the history is still easily available).

This one started off as a water reducing idea, and has progressed somewhat to a zero wash idea.
It has become confusing I suspect, with too much trialling info and not enough result info.

Graham, I have been following this thread from the beginning and think starting a new thread with an appropriate title is a fantastic idea; this way we can keep the newly discovered method clean from other stuff that has been discussed! Great job BTW.

I notice that in all your drawings that you are soley relying on the spray of BD to draw air (positive pressure) into the container/vessel as a way to drive the methanol gases out. Are you sure this is suffient? I think twenty4seven's (Nick's) idea of creating a closed system (with the use of a open head drum) for cleaning would resolve the possibilities of this problem if it exists at all. I know I don't want any health repercussions from methanol exposure.

Stuart